Diagnosis, Management, and Evaluation of Chemotherapy-Induced Peripheral Neuropathy Frederick H. Hausheera, Richard L. Schilskyb, Stacey Baina, Elmer J. Berghorna, and Frank Liebermanc Peripheral neuropathy induced by cancer chemotherapy represents a large unmet need for patients due to the absence of treatment that can prevent or mitigate this common clinical problem. Chemotherapy-induced peripheral neuropathy (CIPN) diagnosis and management is further compounded by the lack of reliable and standardized means to diagnose and monitor patients who are at risk for, or who are symptomatic from, this complication of treatment. The pathogenesis and pathophysiology of CIPN are not fully elucidated, but there is increasing evidence of damage or interference with tubulin function. The diagnosis of CIPN may present a diagnostic dilemma due to the large number of potential toxic etiologies and conditions, which may mimic some of the clinical features; the diagnosis must be approached with care in such patients. The incidence and severity of CIPN is commonly under-reported by physicians as compared with patients. The development of new and reliable methods for the assessment of CIPN as well as safe and effective treatments to prevent this complication of treatment would represent important medical advancements for cancer patients. Semin Oncol 33:15-49 © 2006 Elsevier Inc. All rights reserved.

C

hemotherapy-induced peripheral neuropathy (CIPN) involving sensory and motor nerve damage or dysfunction is a common and serious clinical problem that affects many patients receiving cancer treatment. This condition may pose challenges for the clinician to diagnose and manage, particularly in patients with coexisting conditions or disorders that involve the peripheral nervous system. Many chemotherapeutic agents used today are associated with the development of serious and dose-limiting CIPN that can adversely affect the administration of planned therapy and can impair quality of life by interference with the patients’ activities of daily living. The most important clinical objective in the evaluation of patients with CIPN is to determine their level of functional impairment involving activities of daily living. These findings are used to make medical decisions to continue, modify, delay, or stop treatment. The drugs most commonly reported to cause CIPN include taxanes, platinum agents, vinca alkaloids, thalidomide, and bortezomib (Table 1).

aBioNumerik

Pharmaceuticals, Inc, San Antonio, TX. Sciences Division, University of Chicago, Chicago, IL. cDepartment of Neurology, University of Pittsburgh, Pittsburgh, PA. Address correspondence to Frederick H. Hausheer, MD, FACP, BioNumerik Pharmaceuticals, Inc, Suite 1250, 8122 Datapoint Dr, San Antonio, TX 78229. E-mail: [email protected] bBiological

0093-7754/06/$-see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1053/j.seminoncol.2005.12.010

The diagnosis of CIPN must be approached with care, particularly in patients with potential coexisting or prior conditions that involve or predispose to peripheral neuropathy, and in these patients, diagnostic efforts should be directed at elucidation and differentiation of CIPN from other potential causes of peripheral neuropathy (Table 2). Currently, there is no standard treatment for the prevention, mitigation, or management of CIPN. Consequently, the development of this toxicity can result in treatment delays, dose modifications and, in severe cases, discontinuation of treatment. Such modification, interruption, or cessation in patient therapy may partially or completely alleviate symptoms of CIPN, but such maneuvers may adversely affect patient outcome, especially if the patient’s malignancy is responding to treatment. The development of safe and effective interventions to prevent or mitigate CIPN would represent an important medical advancement for cancer patients. Emerging evidence suggests that the incidence of CIPN is substantially under-reported in clinical trials due to important limitations in the available grading scales that are commonly used to assess CIPN. The challenges in diagnosing and assessing the extent of functional impairment in a reliable and reproducible manner is a paramount consideration for the clinician in practice for medical decision-making as well as in clinical trials involving the prospective evaluation of 15

Chemotherapy-induced peripheral neuropathy

16

Table 1 Agents Reported to Cause Peripheral Neuropathy: Classification According to Predominant Clinical and Pathologic Findings Sensory

Motor

Sensory and Motor

Demyelinating

Almitrine Atorvastatin Bortezomib Cadmium Carboplatin Chloramphenicol Cisplatin Dioxin Didanosine Ethambutol Ethionamide Etoposide* Flecanide Gemcitabine Glutethimide Hydralazine Ifosfamide* Interferon-␣* Isoniazid Lamuvidine Lead Leflunomide Metronidazole Misonidazole Nitrous oxide Oxaliplatin Phenytoin Procarbazine Propafenone Pyridoxine Stavudine Suramini Thalidomide Zalcitabine

␤-bungarotoxin

Acrylamide Alcohol (ethanol) Allopurinol Allyl chloride Arsenic Ara-C, Ara-A, Ara-G* Cadmium Captopril Carbon disulfide Chlorphenoxyl Ciguatoxin Colchicine Cyanide Dapsone Disulfiram Docetaxel DMAPN Epothilonesi Sulfasalaizine Tacrolimus Enalopril Ergots Ethylene oxide Hexamethylmelamine Indomethacin Lithium Methyl bromide Nitrofurantoin Organophosphates Penicillamine Paralytic shellfish poisoning Podophyllin PCBs Saxitoxin Spanish toxic oil Paclitaxel (all formulations) Tetrodotoxin Thallium Trichloroethylene TOCP Vacor (PNU) Vincristine Vinblastine Vinorelbine Vindesinei

Buckthorn Chloroquine Diphtheria FK506 (tacrolimus) Hexachlorophene Muzolimine Perhexiline Procainamide Tellurium Zimeldine

Botulism Gangliosides Latrotoxin Lead Mercury Misoprostol Tetanus Tick paralysis

Demyelinating and Axonopathy Amiodarone Ethylene glycol 1,1=-Ethylidinebis [tryptophan] Gold n-Hexane Methyl n-butyl ketone Naⴙ cyanate Suramini Toluene

NOTE. Bold ⴝ chemotherapeutic or radiation therapy agents. *Chemotherapy agents with uncommon or rarely reported association. iInvestigational.

neurotoxic chemotherapy or interventions aimed at the prevention or mitigation of CIPN. The clinical manifestations of CIPN are subjective and predominantly manifest as pure sensory symptoms, and they are most commonly reported as progressive distal symmetrically distributed symptoms of numbness, tingling, pins and needles, burning, decreased or altered sensation, or increased

sensitivity that may be painful in the feet and hands.1-7 The primary clinical objective in assessing patients is to determine the presence and severity of CIPN-associated symptoms that result in interference with activities of daily living, because this finding is critical for treatment decisions. Symptoms of motor weakness due to CIPN are less commonly reported, and when present, are observed in patients with more persis-

F.H. Hausheer et al

17

Table 2 Differential Diagnosis of Disorders Involving Peripheral Neuropathy I. Acquired Abnormal metabolic conditions Diabetes mellitus Neuropathy secondary to renal disease Hypothyroidism Primary biliary cirrhosis Vitamin deficiency states (deficiencies of vitamin B1, B6, pantothenic acid, ␣-tocopherol or B12) Excessive doses of pyridoxine (B6) Primary (and familial) amyloidosis Acromegaly Immune-mediated Myasthenia gravis Guillain-Barré syndrome Chronic inflammatory demyelinating polyneuropathy (CIDP) Vasculitis (polyarteritis nodosa; Churg-Strauss syndrome) Systemic vasculitis associated with connective tissue diseases: rheumatoid arthritis; lupus erythematosus; Sjögren’s syndrome Monoclonal antibody neuropathy: Waldenström’s macroglobulinemia, myeloma Plexitis-cervical and lumbosacral Multifocal motor neuropathy Infectious Herpes zoster (sensory) Cytomegalovirus (motor) HIV Sarcoid Lyme disease Mycobacterium leprae Camphylobacter jejuni Polio (motor) Hepatitis B or C (vasculitic) Cancer-related Eaton-Lambert myasthenic syndrome Lymphoma, carcinoma-related Paraneoplastic sensory neuropathy Horner’s syndrome Paraneoplastic motor neuropathy (rare) Drug or toxins (Table 1) Chemotherapy-induced Other medications Alcohol (ethanol) Heavy metal and industrial toxins Shellfish, marine, arthropod toxins and venoms Mechanical/compressive/physical Radiculopathy Mononeuropathy Ionizing radiation Unknown etiology Cryptogenic sensory and sensorimotor neuropathy Amyotrophic lateral sclerosis II. Hereditary Charcot-Marie-Tooth disease Riley-Day syndrome Familial amyotrophic lateral sclorosis X-linked spinobulbar muscular atrophy Gowers distal myopathy Hereditary motor sensory neuropathy Hereditary neuropathy with predisposition to pressure palsies Familial brachial plexopathy Familial amyloidosis Acute porphyria Other peripheral neuropathies (rare) Fabry’s disease, metachromatic leukodystrophy, adrenoleukodystropy, Refsum’s disease

18 tent and severe sensory findings.1 Isolated motor weakness with the complete absence of sensory involvement due to CIPN has not been reported, and if such findings are observed in a patient, consideration should be given to other conditions that may produce pure motor weakness including steroid myopathy (proximal), Eaton-Lambert myasthenic syndrome, diabetic neuropathy, cachexia with decreased activity level, paraneoplastic motor neuropathy, and unmasked Charcot-Marie-Tooth (CMT) disease. If the patient has coexisting diabetes, it can be quite difficult to differentiate the onset or progression of diabetic neuropathy from CIPN, which may be mitigated in some instances if the physician has carefully evaluated and recorded the patient’s baseline neurologic findings and symptoms prior to initiation of the neurotoxic chemotherapy. Diabetic neuropathy can be asymmetrical or symmetrical, focal or diffuse, or manifest as mononeuritis multiplex in its involvement and has many different clinical forms. The most common form of diabetic neuropathy, distal symmetrical polyneuropathic form, has clinical symptoms similar to CIPN. The onset of CIPN is usually gradually progressive, but some patients have rapid onset following administration of neurotoxic chemotherapy. CIPN must be differentiated from the symmetrical, distal neurosensory manifestations that are associated with paraneoplastic sensory neuropathy and diabetic neuropathy or toxic/metabolic neuropathies. This differentiation should be based on history and comparison to baseline findings and the time course of new neurosensory findings, recognizing that asymmetric, focal, or proximal involvement, or complete loss of sensation are indicative of other etiologies. The most commonly observed clinical findings of CIPN are symmetrical progressive onset of the following sensory symptoms and findings in a stocking-glove distribution: paresthesias, hyperesthesias, hypoesthesias, and dysesthesias, which more commonly appear earlier and with more pronounced symptoms in the toes and feet, with later involvement of the fingers and hands. Concurrent loss of deep tendon reflexes (loss of distal usually earlier than proximal) in the affected extremities with sensory deficits is an important diagnostic sign associated with greater neurosensory damage. Sensory findings, including diminished or absent proprioception, vibration, touch, two-point discrimination, sharp/ dull discrimination, temperature, and touch/pain are typically diminished in the stocking-glove distribution in symptomatic patients. The most important and challenging diagnostic assessment in patients with CIPN is to determine the degree of functional impairment in the patient’s activities of daily living since these findings alone determine the need for modification, interruption, or discontinuation of the neurotoxic treatment. When sensory symptoms of CIPN are severe, the patient commonly expresses symptoms of functional impairment of important activities of daily living such as being unable to walk, button clothing, drive, type, or write.8 There can be challenges in eliciting such information and findings in some patients, which may be facilitated by interviewing a family member or friend who regularly observes

Chemotherapy-induced peripheral neuropathy the patient’s activities. It is important to consider that the subjective nature of the patient symptoms and the degree of impairment by CIPN are analogous to patient symptoms of nausea, pain, and depression. Accordingly, a clinical approach must be taken that enables reliable, convenient, and non-invasive assessment of CIPN for the day-to-day management of the patient, as well as to apply such methods in clinical trials of interventions for the prevention or mitigation of CIPN. Many therapies have been evaluated experimentally and in clinical trials as potential neuroprotective agents for CIPN including amifostine, glutathione, glutamine/glutamate, calcium/magnesium infusions, neurotrophic factors, gabapentin, and carbamazepine. Some promising results have been reported from phase I and phase II trials with these interventions, but no definitive conclusions can be drawn regarding these approaches because all of these studies are underpowered for their endpoints or employ nonrandomized and/or uncontrolled trial designs. None of these empiric CIPN prevention or management therapies has become a standard of care or has otherwise documented evidence of benefit in the prevention, mitigation, or treatment of CIPN. Additionally, many of these empiric therapies are associated with adverse effects that may limit their utility in patients, and it is presently unknown if there is any concurrent interference with the antitumor activity of chemotherapy by these approaches. This article is aimed at a comprehensive review and evaluation of the common and serious clinical problem of CIPN, its pathogenesis and pathophysiology, the relationship to neurotoxic chemotherapeutic agents, and the clinical challenges in the assessment and management of this important condition. We also discuss potential neuroprotective approaches and new developments currently undergoing evaluation in clinical trials for the prevention or mitigation of CIPN and describe the requirements for the design of prospective adequate and well-controlled trials for the evaluation of neuroprotective agents. This discussion will also include the evaluation of new cytotoxic agents that are reported to be potentially less neurotoxic relative to others in the same class.

Pathogenesis and Pathophysiology of CIPN The underlying pathophysiologic mechanism(s) for the development of CIPN has not been fully elucidated. A variety of proposed mechanism(s) are reported in the literature.9-37 A complete mechanistic understanding of CIPN has not been achieved. However, it is apparent that CIPN has a high degree of similarity in the pattern and spectrum of clinical manifestations caused by different chemotherapeutic agents (eg, vinca alkaloids, platinum agents, thalidomide, bortezomib, and taxanes), which includes the length-dependent, symmetrical stocking-glove distribution with predominantly sensory symptoms noted by the patient. CIPN generally arises as a consequence of the disruption of axoplasmic microtubulemediated transport, distal axonal (Wallerian) degeneration,

F.H. Hausheer et al and direct damage to the sensory nerve cell bodies of the dorsal root ganglia (DRG).3,9-14,16-18,22-31,35,38-43 Demyelination (diffuse or segmental) secondary to chemotherapy is an uncommon finding (but occasionally observed with cisplatin, and reported for suramin, an investigational agent), and when observed, it is typically a secondary and isolated finding relative to the extent of axonal and DRG pathological findings. It is important to recognize that CIPN commonly follows the administration of chemotherapeutic agents that cannot appreciably distribute across the blood-brain barrier (eg, taxanes, platinum agents, vinca alkaloids, thalidomide, and bortezomib).3,16,43 One of the important anatomical targets for neurotoxic chemotherapeutic agents is the DRG and the afferent and efferent axons, which are located outside of the CNS.9 In contrast to the CNS, the DRG and peripheral axons lack an efficient neurovascular barrier, which allows the facile diffusion of large molecular weight compounds in the interstitium surrounding the DRG and along the axon filaments. This absence of a vascular barrier may play an important role in CIPN development, by allowing the DRG and the peripheral axons to be exposed to the unimpeded distribution of neurotoxic agents from the plasma. In addition to capillary fenestrations in the vascular supply to the DRG and axons, there are open junctions between adjoining endothelial cells of the peripheral epineural blood vessels that contain and allow proteins and drugs to pass readily from the blood into the extracellular space. Protein-bound and unbound forms of drugs can gain access to the epineurium and can readily diffuse along the nerve fascicles, a process that is facilitated by positive hydrostatic capillary pressure. The endoneural fascicles also lack lymphatics, which prevent removal of toxic substances in the endoneural fluid. Autopsy and experimental studies have shown high concentrations of platinum in the DRG and high concentrations of taxanes in the peripheral axons as compared to significantly lower levels of these substances in the brain and spinal cord.9-14,22-31,41 Cisplatin and carboplatin neurotoxicity have been reported to be the result of direct toxic damage to the DRG in animal models.13,44 Investigators have shown that cisplatin also substantially interferes with and disrupts axonal microtubule assembly, vesicular axonal transport, and inhibits neurite outgrowth from neurons.14,31,45-52 DRG neurons exposed to cisplatin in vitro demonstrated nuclear condensation, cell shrinkage, and fragmentation with maintenance of plasma membrane and intracellular organelle integrity. These studies demonstrated that cisplatin-treated DRG neurons continued to enter the cell cycle before any detectable morphological measure of cell death was observed as well.9-14 Cisplatin has been observed to directly cause axonal degeneration, thereby interfering with the supply of nutrients to distal axons mediated by fast anterograde axoplasmic flow; tubulin and microtubulin function are critical for normal axoplasmic transport. In rat models, carboplatin induces a highly similar pattern of histopathological (DRG and axonopathy with evidence of Wallerian degeneration) and neurosensory damage that appears to be indistinguishable from cisplatin, suggesting that these drugs act on a common intracellular target.11,26,29,41,44

19 Neurotoxicity from the taxanes and vinca alkaloids is also reported to involve the disruption of microtubule structure and function in peripheral nerve axonal filaments.23,24,28 A common feature of the underlying neuropathophysiology of CIPN for taxanes, cisplatin, carboplatin, and vinca alkaloids appears to involve disruption and dysfunction of axoplasmic microtubules, and an increasing number of experimental findings point to the disruption of tubulin function by neurotoxic medications as a potential mechanism. Tubulin serves a critical function in normal anterograde (fast and slow) and retrograde axoplasmic transport of peripheral axons. The role of tubulin and the associated axoplasmic motor proteins (eg, kinesin and actin) in the function of peripheral nerves poses some challenges for mechanistic studies involving chemotherapeutic and neuroprotective agents, but there has been an increasing level of progress in the development of mammalian peripheral nerve models that may allow further elucidation of the pathogenesis and pathophysiological mechanisms underlying CIPN.31-34,53 The clinical diagnostic hallmark of dose-limiting CIPN is a distal symmetrical sensory neuropathy, which may be accompanied in some patients with disproportionately diminished motor symptoms, but more commonly motor symptoms are completely absent.3-5 The relative predominance of neurosensory symptoms with partial or complete sparing of motor function observed in CIPN has not been characterized on a pathophysiologic or pharmacologic basis. None of the reported studies of CIPN have advanced a mechanistic rationale for the predominance of the neurosensory component. We postulate the following neuropathologic and neuropharmacologic mechanisms that may explain the predominance of the neurosensory relative to the neuromotor clinical presentation and manifestations of CIPN: (1) the most commonly associated neurotoxic cancer chemotherapeutic agents associated with CIPN cannot cross the blood-brain barrier in appreciable concentrations, which largely excludes a central neurotoxic process for the pathogenesis of CIPN; (2) the nerve cell bodies of lower motor neurons are located in the anterior gray horn of the spinal cord, and this region of motor nerve cell body distribution is highly protected from neurotoxic drug exposure by the blood-brain barrier, leaving only the efferent portion of peripheral motor axons with exposure to neurotoxic chemotherapy; (3) the nerve cell bodies and axons of the neurosensory neurons for touch, position, temperature, pressure, vibration, and tendon stretch receptors are all located outside of the CNS in the DRG, and thus all anatomic components (nerve cell bodies and axons) of the peripheral sensory nerves can be fully exposed to neurotoxic chemotherapy without any pharmacologic barrier; (4) the axons of motor neurons are more heavily myelinated (type A-␣, the largest diameter fibers), which, in addition to potentially different protein composition, may also provide greater tolerance and/or sparing of the degree of drug-induced motor axon damage relative to the damage to smaller diameter myelinated (eg, type A-␤ -␥, and - ␦ and type B and C) sensory fibers; and (5) with neurotoxic damage to the peripheral neurosensory neurons and axons, the relative sparing of neuromotor function could be explained in part on the basis of

20 Table 3 Symmetrical Neuropathic Disorders With Purely or Predominantly Sensory Symptoms Cancer chemotherapy Other medications Paraneoplastic sensory neuropathy Sjögren’s syndrome Idiopathic sensory neuronopathy Vitamin deficiency or excess (pyridoxine) HIV-related sensory neuronopathy Toxins and venoms

an intact upper motor neuron function, as well as a protected lower motor neuron cell body in the spinal gray matter. It is also important to consider that the CIPN does not appear in a diffuse or dermatomal or focal distribution, and the characteristic length-dependent distal symptoms and signs of CIPN are associated with damage/dysfunction involving only the longest peripheral nerve fibers. This physical pattern of involvement suggests that the peripheral neurons with the greatest length may have the greatest proportional surface area for drug toxicity, as well as containing the greatest amounts of tubulin, kinesin, and actin and other neural proteins that may result in greater susceptibility to the acute or cumulative neurotoxicity that is observed with CIPN.

Clinical Diagnostic Features of Chemotherapy-Induced Peripheral Neuropathy The onset of CIPN is, in general, gradually progressive, but some patients have rapid onset following administration of neurotoxic chemotherapy. CIPN must be differentiated from the symmetrical distal neurosensory manifestations that are associated with paraneoplastic sensory neuropathy and diabetic neuropathy or toxic/metabolic neuropathies. This differentiation is based on history and comparison to baseline findings and the time course of new neurosensory findings, recognizing that asymmetric, focal or proximal involvement, or complete loss of sensation are indicative of other etiologies (Tables 2, 3, and 4). The essential diagnostic features of CIPN include: (1) distal length-dependency for the longest peripheral nerve segments (eg, glove and stocking distribution); (2) symmetrical distribution of involvement; (3) temporal onset following administration of neurotoxic chemotherapy with (a) progressive onset with temporary resolution or progressive worsening of sensory symptoms (axonopathy) or (b) rapid onset

Chemotherapy-induced peripheral neuropathy after chemotherapy administration of sensory symptoms (neuronopathy); (4) signs and symptoms of neurosensory dysfunction, which may include paresthesias, dysesthesias, hypesthesias, hyperesthesias, and hypoesthesias, and pain; and (5) relative sparing of motor function with mild to moderate motor weakness that may be accompanied by myoatrophy in the same distal distribution as the sensory findings.3-5,15-18,21,35,38,42-43,54-71 An essential diagnostic feature of CIPN is the disproportionate degree of sensory symptoms reported in relationship to motor symptoms and findings. The acute form of oxaliplatin-related neuropathy has unique clinical features from these findings and will be discussed later. The clinical evaluation of CIPN is focused on the location and distribution of the patient’s symptoms and the temporal relationship to chemotherapy initiation, and the presence, character, and severity of sensory symptoms with particular attention to interference with activities of daily living. Localized areflexia (distal with progression to proximal involvement) in the affected extremities may be present as a helpful diagnostic sign of more advanced CIPN. Complete symmetrical loss of distal deep tendon reflexes (eg, Achilles or brachioradialis) of the affected extremities can be confirmed by having the patient perform reinforcement maneuvers by isometric voluntary muscle contraction in the upper or lower extremities in the opposite anatomic area of interest. The absence of reflexes with reinforcement testing is a reliable confirmatory sign of the true absence of deep tendon reflexes in the affected extremities. Clinical examination of symptomatic patients commonly demonstrates impairment of touch/ pain, two-point discrimination, sharp/dull, proprioception, temperature, and vibration in a symmetrical stocking-glove distribution that corresponds to anatomic area(s) involving the patient’s clinical symptoms of CIPN. Many patients with CIPN will demonstrate an abnormal Romberg test and may also have ataxia. The Romberg test may be abnormal due to labyrinthine, cerebellar as well as dorsal column dysfunction or damage, and weakness. Ataxia is a nonspecific sign that may be observed in patients with CIPN. However, if present with neurosensory findings and other symptoms of CIPN, ataxia can be an important sign of more severe functional impairment. Prior to initiating neurotoxic chemotherapy, consideration of coexisting diabetes, nutritional deficiency (eg, atrophic gastritis, pyridoxine, or alpha-tocopherol deficiency), or pyridoxine intoxication, past history of neuropathy, prior neurotoxic chemotherapy treatment, and family history of hereditary neuropathies should be ascertained. It is also important

Table 4 Symmetrical Neuropathic Disorders With Purely or Predominantly Motor Symptoms Motor neuron disease Multifocal motor neuropathy Guillain-Barré syndrome* Chronic inflammatory demyelinating neuropathy* Eaton-Lambert myasthenic syndrome Amyotrophic lateral sclerosis *Usually accompanied by sensory signs and symptoms.

Infections, toxins, and venoms (Table 1) Acute porphyria* Charcot-Marie-Tooth disease* Hereditary motor neuropathy Myasthenia gravis

F.H. Hausheer et al to determine if there are prior or co-existing medical conditions (eg, diabetes, HIV, alcoholism, paraneoplastic neuropathy) or medications that may confound or exacerbate the severity of CIPN (Table 2). Many common medications, including metronidazole, misonidazole, colchicine, sulfasalazine, nitrofurantoin, tacrolimus, nucleoside analogs (zalcitabine, didanosine, stavudine, and lamivudine), hydralazine, phenytoin, isoniazid, perhexiline, and disulfiram have all been reported to produce peripheral neurotoxicity, and they may confound the management of patients who are subsequently treated with neurotoxic chemotherapy (Table 1). Prior to the administration of potentially neurotoxic chemotherapy, the treating physician should perform and document the patient’s baseline clinical neurologic examination, recording any sensory and motor abnormalities that may be present. In patients with pre-existing neurologic conditions or disorders, evaluation by a neurologist may be invaluable in the future management of these patients. In the evaluation of patients prior to the administration of potentially neurotoxic chemotherapy, particularly with vincristine, it is important to evaluate for the presence of CMT disease, usually an autosomal dominant hereditary motor and sensory disorder that affects the nerves and muscles of the distal extremities. Patients with type I CMT disease present in middle childhood with footdrop and slowly progressive distal muscle atrophy, producing “stork legs” due to peroneal atrophy. Intrinsic muscle wasting in the hands has a later onset. These patients may have other degenerative diseases (eg, Friedrich’s ataxia) as well. These patients have a stocking-glove distribution of impaired vibration, pain, and temperature sensation with absent deep tendon reflexes. They have relatively normal nerve conduction velocities but have low amplitude evoked potentials. Biopsy demonstrates Wallerian degeneration of the distal axons. These patients exhibit increased sensitivity to neurotoxic chemotherapy, particularly reported with vincristine, which if used at all, must be given with great caution so as to avoid the development of severe and potentially irreversible CIPN that can mimic Guillain-Barré syndrome in these patients.

Diagnostic Classification of CIPN CIPN is most commonly characterized as a distal axonopathy, less commonly as a neuronopathy, and may simultaneously manifest with both forms in some patients. Distal axonopathy is an abnormality in peripheral nerve function that affects axonal function and transport, resulting in a retrograde Wallerian type of degeneration of the terminal regions of sensory axons and to a lesser extent, motor axons, which propagate proximally toward the DRG. Neuronopathy is the result of direct toxic damage involving neuronal cell bodies in the DRG. In neuronopathies the DRG cells are seriously damaged or die, and the axons degenerate in an anterograde manner, resulting in axons that fail to regenerate.4,9-14,30,40,42,57,60 The clinical features of CIPN neuronopathy are character-

21 ized by the rapid onset of a length-dependent symmetrical distal sensory neuropathy in a stocking-glove distribution that temporally follows neurotoxic chemotherapy administration (Table 5). The clinical manifestations simultaneously affect the upper and lower extremities, as well as cranial nerves, and are accompanied by the rapid loss of all deep tendon reflexes.4,17-18,21,30,39,40,54,57,60,62-63 Important clinical features of CIPN neuronopathy include the temporal onset of more severe and persistent sensory symptoms that are accompanied by symmetrical and complete loss of deep tendon reflexes in the extremities. These findings occur shortly (within hours to several days) following one or two treatments, usually with higher doses of neurotoxic chemotherapy. The sensory symptoms and findings of CIPN neuronopathy are persistent and are not progressive or waxing and waning, from their onset. The patient’s sensory symptoms are usually permanent, which may be due to the loss of nerve cell bodies in the DRG leading to anterograde Wallerian degeneration of the peripheral axons. Motor weakness and muscular atrophy may be found in the affected extremities, but CIPN neuronopathy generally spares motor function disproportionately relative to the sensory abnormalities.3-5 Taxanes and platinum therapy have been reported in association with the development of CIPN neuronopathy, particularly with higher doses or when combined therapy is administered.3,15,21,40,54,60,66,67 Distal axonopathy is the most common clinical presentation of CIPN.3,5,15-19,35,38-40,42-43,55-58,61,63,64,68,72 The clinical features of CIPN axonopathy include progressive onset following the administration of neurotoxic chemotherapy, usually following multiple cycles of treatment, characterized by a distal symmetrical length-dependent sensory neuropathy that manifests as paresthesia, dysesthesia, burning, pain, or numbness that more commonly initially involves the lower extremities, followed by the upper extremities in a stockingglove distribution. The neurosensory symptoms can progress and can eventually be accompanied by the progressive symmetrical loss of distal deep tendon reflexes. The sensory symptoms and interference with function may wax and wane between treatments, but they commonly progress with additional treatment and may eventually become persistent between treatment cycles or with the passage of time, even after treatment is interrupted or discontinued. The waxing and waning of neurosensory symptoms is likely due to partial or temporary recovery from disrupted microtubule function and/or Wallerian regeneration of the terminal axons. The finding of motor weakness and atrophy in the affected extremities is variable and is usually mild or moderate. Symptoms of distal weakness are less commonly reported by patients in proportion to the sensory symptoms that may be sufficiently severe to interfere with their activities of daily living. Distal symmetrical weakness may be detected in patients with CIPN axonopathy and may be accompanied by signs of myoatrophy. The important diagnostic features of associated motor weakness that are commonly observed with CIPN distal axonopathy include: (1) motor findings not perceived or reported by the patient at all, but they may be observed on clinical examination by the physician; (2) distal

Chemotherapy-induced peripheral neuropathy

22 Table 5 Diagnostic Approach to Chemotherapy-Induced Peripheral Neuropathy

1. System involvement: nature of symptoms *Sensory: paresthesias, dysesthesias, hypoesthesia, burning, pain Motor: weakness, atrophy, gait, activities (specific) *Combined sensory and motor Autonomic: diaphoresis, postural weakness, anhydrosis, orthostatic 2. Distribution of symptoms *Symmetrical Asymmetrical, focal, dermatomal *Distal: stocking glove Proximal Combined proximal and distal 3. Presence or *absence of upper motor neuron involvement Sensory deficit absent *Sensory deficit present 4. Temporal onset and duration of symptoms Acute (hours to days) Persistent Waxing and waning Temporal relationship of preceding events, recent/prior medication/toxin/venom/infection 5. Medication history Review medications: non–oncology-related and oncology treatment–related Prior neurotoxic chemotherapy Medication start/stop, duration, establish temporal relationship to symptoms 6. Evidence of acquired or hereditary neuropathy Diabetes, renal disease, hypothyroidism Prior history of neuropathy, alcohol HIV status Amyloid, sarcoid, vasculitis Family history of neuropathy Skeletal deformities 7. Degree of symptom interference with activities of daily living *Ambulation, use of hands, dressing, eating, driving, sleeping, climbing stairs, others *Do these symptoms interfere with your ability to perform daily activities? *If yes, describe the level of interference: “Sometimes”, “most of the time”, or “all of the time” Elucidate activities patient cannot perform without discomfort, interference, or complete disability 8. Neurologic examination (distal symmetrical stocking-glove) abnormalities in CIPN *Neurosensory: light touch, pin, dull/sharp, vibration/proprioception Neuromotor: signs of atrophy, extensor/flexor muscle strength, grip, gait *Absent/decreased reflexes: distal symmetrical; reinforcement maneuvers *Symptoms and/or signs that are present in CIPN; other findings may be present or suggest other etiologies.

weakness or decreased muscle mass, if present, is mild to moderate and is not typically interfering with functional activity; (3) the sensory abnormalities and resulting interference with functional activities that the patient is experiencing are far more disturbing to the patient than the motor findings; and (4) it is not uncommon for patients to grow increasingly concerned with the progressively worsening or persistent nature of their symptoms. The crescendo pattern and severity of neurosensory and neuromotor symptoms and potential interference with function may worsen with the continuation of treatment and may progress after complete cessation of treatment with agents such as cisplatin, carboplatin, oxaliplatin, thalidomide, or taxanes.73-76 Pure motor neuropathy, manifested as isolated weakness and/or atrophy with complete absence of sensory symptoms or signs of impairment, has not been reported for CIPN. The presence of such symptoms in a cancer patient would suggest

other etiologies, including paraneoplastic motor neuropathy or myopathy (rare), steroid (proximal involvement is more common) or drug-induced myopathy, myasthenia gravis, Eaton-Lambert myasthenic syndrome and poor nutritional status, toxic neuropathy, or cachexia and debilitation (Table 5). The insidious onset of symmetrical, distal sensory loss with proximal muscle weakness (especially in the hips and thighs), which is characterized by increasing strength with repeated contraction (facilitation) and may be accompanied by autonomic disturbances and occasional ocular involvement, are the key diagnostic features of the Eaton-Lambert myasthenic syndrome. The Eaton-Lambert myasthenic syndrome is most commonly reported in patients with small cell lung cancer and is less frequently reported in patients with non–small cell lung cancer (NSCLC), lymphoma, malignant thymoma, and carcinomas of the breast, stomach, colon, prostate, bladder, kidney, or gall bladder. In patients treated

F.H. Hausheer et al with taxanes and/or vinca alkaloids there may be myalgias that are severe and persistent enough to lead to disuse and some degree of muscular atrophy, which may coexist with CIPN in patients. Progressive, symmetrical, distal length-dependent axonopathy is the most commonly reported clinical presentation of CIPN for all classes of neurotoxic chemotherapy, including platinum, vinca alkaloid, and taxane therapy. The clinical features of thalidomide and bortezomib are also consistent with CIPN axonopathy, and in some cases, particularly with higher doses of these agents, are consistent with CIPN neuronopathy. Some patients may initially develop symptoms and signs of CIPN axonopathy and then may suddenly develop symptoms and signs of CIPN neuronopathy. If the degree of axonal damage is mild, the terminal axon may regenerate, thereby allowing partial or complete restoration of sensory function, which may be manifested by partial or complete clinical recovery between longer treatment cycles, treatment delays, or treatment cessation. Terminal axonal degeneration and axonal microtubule disruption are the most common pathological processes observed in CIPN and have been reported in association with exposure to vinca alkaloids, platinum agents, taxanes, thalidomide, suramin, and others. Peripheral myelin degeneration (diffuse or segmental) has been reported in association with other agents such as perhexiline, hexanes, and amiodarone. However, demyelination is far less commonly reported in CIPN, and the damage to myelin appears to be a secondary finding as compared to damage to the terminal axon or disruption of the axonal microtubulin architecture. It is well recognized that taxanes and epothilones induce abnormal tubulin polymerization by preventing tubulin depolymerization, whereas the vinca alkaloids depolymerize tubulin. Monohydrated platinum has been demonstrated to directly disrupt the ability of tubulin to polymerize, which is most likely due to denaturation of tubulin by the formation of platinum-tubulin adducts involving the accessible cysteine residues on tubulin.31,77-78 These findings suggest that tubulin may be a primary target in the pathogenesis and pathophysiology of CIPN. Tubulin is a ubiquitous cellular protein that is highly abundant in most cells and has many structural and transport functions, and plays a critical role in normal DRG and axonal physiologic functions. Because of the high degree of similarity in the clinical manifestations and symptoms of CIPN by different drugs, a common underlying mechanism is suggested, involving a direct toxic effect of taxanes, platinum agents, vinca alkaloids, and other neurotoxic chemotherapeutic agents that may be mediated largely by toxic interactions with neuronal tubulin.

The Common and Serious Clinical Problem of Chemotherapy-Induced Peripheral Neuropathy Many of the most commonly used chemotherapeutic agents are associated with dose-limiting CIPN, including, but not

23 limited to, taxanes (docetaxel and all formulations/derivatives of paclitaxel), platinum agents (cisplatin, carboplatin, and oxaliplatin), and vinca alkaloids (vincristine, and less commonly vinblastine, vindesine, and vinorelbine).69,73,79-82 More recently, the use of thalidomide and bortezomib has been reported to have an association with the development of CIPN, which can be dose-limiting.74,83 The incidence and prevalence of CIPN in patients undergoing chemotherapy as well as the proportion of patients with persistence or resolution of CIPN following treatment are not known, but the occurrence of CIPN is widely reported in various controlled and uncontrolled studies in association with the administration of various chemotherapeutic agents.3,17,18,43,61,71,72,84-86 In addition, the proportion of patients who experience treatment delays or discontinuation secondary to CIPN is rarely reported in clinical studies, and there is a lack of well-controlled clinical data regarding the time to onset, duration, and resolution or persistence of CIPN in the medical literature. More recent studies of CIPN symptoms that compare patient-based reporting methods to physician-based assessments consistently demonstrate that physicians under-report the severity and incidence of CIPN symptoms, compared to patients.87-93 These findings suggest that CIPN is an under-reported and under-recognized medical problem, possibly due to several factors including the lack of: (1) reliable and standardized diagnostic methods for CIPN; (2) methods that allow reliable quantifiable medical decision making and reporting; and (3) approved safe and effective medical treatment to prevent or mitigate the development of CIPN. Most of the earlier clinical studies for drugs such as cisplatin, carboplatin, oxaliplatin, paclitaxel (including the paclitaxel-containing formulations Abraxane [American Pharmaceutical Partners, Schaumburg, IL], Tocosol [Sonus Pharmaceutical, Bothell, WA], and Xyotax [Cell Therapeutics, Seattle, WA]), epothilones, and docetaxel report substantially lower incidences of more severe CIPN, in contrast to later studies involving larger patient populations. This apparent under-reporting of CIPN in earlier stages of drug development may be due in part to the absence of sufficiently sensitive and reliable diagnostic instruments for the assessment and reporting of CIPN. Another contributing factor may include the under-assessment and the high degree of variability of the reported incidence and severity of CIPN associated with the currently available diagnostic and reporting methods used in these trials. In addition, comparison of CIPN incidence rates and outcomes becomes difficult across studies when different assessment methods are employed. The importance of CIPN has increased with the advent and increased utilization of newer neurotoxic medicines, and supportive care interventions to prevent or mitigate toxicities of other organs (eg, neutropenia), along with the increased utilization of dose-dense and weekly treatment regimens and drug combinations involving more than one neurotoxic compound. The incidence and severity of CIPN depends on the dose, schedule, and duration of neurotoxic agent(s) administered to the patient, as well as any coexisting or pre-existing conditions that may predispose the patient to an increased

24 incidence of more severe forms of CIPN. The use of multimodal neurotoxic therapy involving the combination of agents such as taxane and platinum agents, which has resulted in improved patient survival, has also resulted in a greater reported incidence of CIPN.70,71,85 In addition, the use of weekly taxane therapy in patients with metastatic breast cancer was recently reported to result in a significantly higher objective tumor response rate, time-to-progression, and overall survival as compared with the standard every-3week regimen.94 The use of dose-dense and weekly taxane regimens is accompanied by an increased incidence and prevalence of CIPN, which further underscores the need for diagnostic methods for management and reporting as well as a safe and effective intervention.95-109

Clinical Features of CIPN Associated With Commonly Used Oncology Therapeutic Agents Platinum Agents (cisplatin, carboplatin, oxaliplatin) All platinum-containing chemotherapeutic agents are associated with the development of treatment-related CIPN that is dose-dependent, cumulative, and predominantly sensory in nature. Platinum agents are most commonly associated with the development of axonopathy, but platinum-induced neuronopathy may be observed, particularly with higher doses of platinum therapy. Cisplatin Cisplatin is a widely used chemotherapeutic agent and has antitumor activity for a wide cross-section of cancers. Along with its potent anticancer properties, cisplatin is also associated with significant toxicities, which can limit the clinical utility and present certain risks to patients undergoing treatment. The reported incidence of CIPN for single agent cisplatin ranges from 49% up to 100% depending on the dose, schedule, and combination of other drugs employed.6,19,40,54,60,67,79,110 The development of cisplatin neuropathy is dose-related and generally noted after cumulative doses in excess of 300 mg/m2.40,60 Significant factors in the development of cisplatin neuropathy include dose, treatment duration, and schedule of administration. Cisplatin-induced neuropathy presents predominantly as sensory neuropathy that is manifested as paresthesias and numbness in a symmetrical stocking and glove distribution, which may be accompanied by progressively reduced or absent reflexes in the affected extremities. Although less common, cisplatin neuropathy can include motor weakness, as well as other central and autonomic neuropathic symptoms and findings, including Lhermitte’s sign, seizures, dorsal column myelopathy, bilateral jaw pain, and autonomic neuropathy.6,40,111 Cisplatin neuropathy generally occurs progressively after several treatment cycles (axonopathy), but some patients report a rapid onset of neuropathy symptoms (neuronopathy)

Chemotherapy-induced peripheral neuropathy after the administration of a single dose of cisplatin, particularly with higher doses (eg, 100 mg/m2 or more).7 Symptoms and signs of cisplatin neuropathy usually occur during treatment. However, it is important to recognize that the cisplatin-induced neuropathy may occur up to several weeks after the last administered dose of cisplatin and the patient’s symptoms may continue to progress even after cisplatin therapy has been discontinued. The CIPN symptoms associated with cisplatin may be irreversible and may be more frequent and severe when the drug is used in combination with other neurotoxic drugs (eg, cisplatin or carboplatin with paclitaxel). Carboplatin Carboplatin is associated with the development of CIPN that is clinically indistinguishable from cisplatin. The carboplatin label reports an incidence of CIPN ranging from 13% to 42%.69 Carboplatin is reported to be less neurotoxic that cisplatin, but several controlled studies comparing carboplatin and cisplatin suggest that the incidence and severity of CIPN may be similar for these two agents.71,85 Carboplatin neuropathy is experimentally and clinically indistinguishable from cisplatin neuropathy.44 Oxaliplatin Oxaliplatin is structurally different from both cisplatin and carboplatin and has proven to be effective in the treatment of patients with advanced colorectal cancer. The most common and dose-limiting toxicity reported for oxaliplatin administration is a persistent sensory peripheral neuropathy, which has been reported in association with all oxaliplatin dose levels and schedules.112 The development of dose-limiting oxaliplatin neuropathy is primarily related to the total cumulative oxaliplatin dose. Oxaliplatin-induced neuropathy has been characterized into two distinct clinical forms: (1) an acute and usually transient and predominantly sensory disturbance in 85% to 95% of patients, which may be aggravated or precipitated by cold exposure, and has a rapid onset of hours to days following treatment and which may regress between treatment cycles but frequently recurs with further treatment; and (2) a persistent sensory CIPN that is gradually progressive in onset and is related to the cumulative oxaliplatin dose and characterized by sensory paresthesias, dysesthesias, and hypoesthesias that can interfere with daily activities and is reported to occur in up to approximately 16% to 21% of patients. Neurologic symptoms and abnormalities are clinically observed in up to 97% of patients receiving oxaliplatin-based treatment.111,113,131 The acute form of oxaliplatin neurotoxicity presents as paresthesia, dysesthesia, and hypoesthesia in hands, feet, perioral area, or throat. Pharyngolaryngeal dysesthesia is reported by a small percentage of patients and is characterized by subjective sensations of dysphagia or dyspnea and/or jaw spasm, which is unaccompanied by laryngospasm, bronchospasm, or stridor. Pharyngolaryngeal dysesthesias and myotonia (eg, myotonic inability to release grip from objects, similar to hyperkalemic periodic paralysis) can be very disturbing to patients and can be precipitated or exacerbated by exposure to cold ambient temperature, drink-

F.H. Hausheer et al ing cold liquids, or handling cold objects. The cumulative persistent oxaliplatin-induced neurotoxicity is similar to other forms of platinum- and taxane-induced neuropathy commonly manifested as symmetrical paresthesias, dysesthesias, and/or hypoesthesias in the distal extremities. Such deficits can interfere with Activities of Daily Living (ADLs). The persistent form of oxaliplatin neuropathy generally occurs more than 14 days following dosing and can become persistent between treatment cycles with progression of the severity of symptoms.111,113,115,120,121,123 One hypothesis suggests that the acute CIPN secondary to oxaliplatin may be related to the rapid chelation of calcium by oxalate released from oxaliplatin with consequent effects on sodium channels in neural membrane and synapses,29,120 while the chronic or persistent deficits may reflect the welldocumented neurotoxic effects of the gradual accumulation of platinum in dorsal root ganglion cells. Although the underlying pathophysiologic mechanisms may differ, a linkage between the acute and persistent deficits is suggested by the anecdotal reports that the persistent, dose-limiting form of oxaliplatin neuropathy may be prevented or delayed by adequate treatment of the acute form of oxaliplatin neuropathy. This hypothesis should be confirmed in controlled clinical trials. Dysesthesias that are precipitated with cold air, cold objects, cold drinks, or ice occur in approximately 68% of patients with 1% to 2% of patients reporting pharyngolaryngeal dysesthesia. Paresthesias without pain occur in 65% of patients and paresthesias with pain occur in approximately 11% of patients. Approximately 13% to 28% of patients receiving oxaliplatin doses ranging from 85 mg/m2 to 130 mg/m2 experienced severe neurosensory adverse effects with functional impairment.111,113,131 In a recent study of oxaliplatin neurotoxicity, there was a reported 13% to 16% incidence of severe persistent neurotoxicity (grade 3) in patients receiving the FOLFOX 4 and FOLFOX 6 regimens (FOL ⫽ fluorouracil, F ⫽ folinic acid (leucovorin), OX ⫽ oxaliplatin). The reported median time to the onset of oxaliplatin neuropathy was 23 weeks, and the median time to recovery was 12 weeks, with the reported probability of recovery of 82% by 26 weeks and 88% after 34 weeks.111,113,115,131 It is important to note that, similar to cisplatin, some patients develop delayed onset or worsening of the persistent form of oxaliplatin-induced neuropathy, even after treatment is discontinued.111,113,131 Severe (grade 3) oxaliplatin neuropathy is reported to be reversible, with partial to complete disappearance of symptoms. However, oxaliplatin-induced neuropathy is a considerable problem because of its frequent occurrence, the severity of symptoms, and the functional impairment that can adversely affect the patient’s quality of life for a prolonged period of time. The persistent form of oxaliplatin-induced neuropathy commonly results in treatment delays, and in severe cases, discontinuation of treatment. Persistent sensory CIPN is the dose-limiting toxicity of oxaliplatin.115,131 Neurologic deficits are dominated by sensory impairment of peripheral neural function, including

25 pronounced dysesthesias and paresthesias. Sensory symptoms demonstrate a clear length-dependent, distal-to-proximal gradient and usually progress in both intensity and duration as a function of cumulative dose. Eventually, neurologic signs and symptoms will persist between treatment cycles. It is clinically important to note that the onset of tumor response with oxaliplatin-based therapy usually occurs before a cumulative oxaliplatin dose of 700 mg/m2 has been reached. The dose-limiting neurotoxicity, in contrast, develops somewhat later. It is estimated that approximately 10% to 15% of patients have a moderate neuropathy after a cumulative dose of 780 to 850 mg/m2.115,117,122,131 This fact allows for a clinical decision to continue or stop oxaliplatin in view of the observed efficacy of treatment, but in patients who are responding to treatment the development of doselimiting CIPN can pose a major challenge to the physician and patient to continue treatment. The persistent form of oxaliplatin-induced neuropathy is reported generally reversible with a median time to recovery from grade 3 toxicity of 13 weeks.115,118,131 Such treatment delays may adversely affect patients who have demonstrated a partial response to treatment by the interruption of treatment.

Taxanes (paclitaxel, docetaxel, Cremophor EL-free, and nanoparticle albumin formulation of paclitaxel) The taxanes are widely used and have demonstrated activity against various types of malignancies. They are used both as single agents and in combination with other chemotherapeutic agents. Paclitaxel Paclitaxel is currently approved for use in patients with metastatic carcinoma of the ovary after failure of first-line therapy and in patients with metastatic breast cancer or AIDS-related Kaposi’s sarcoma. The standard approved doses and schedule of paclitaxel in these patient populations are 135 or 175 mg/m2 administered over 3 hours once every 3 weeks. However, administration of paclitaxel using a weekly schedule is an increasingly widespread practice. Although neutropenia historically has been the dose-limiting toxicity for paclitaxel in the past, and hypersensitivity reactions have been frequent, both toxicities can be prevented or mitigated with the administration of concomitant supportive medications.6,7,133 Paclitaxel-induced neuropathy, manifested primarily by peripheral sensory neuropathy with relative sparing of motor function, is a common side effect that can be dose-limiting.4-5,21,58,65-66,134 Paclitaxel neuropathy is doseand schedule-dependent and cumulative in nature, and the dose appears to be the most significant risk factor for CIPN. Paclitaxel-induced neuropathy has been reported for all doses and schedules.15 Various doses and schedules for paclitaxel have been extensively explored in an effort to maximize antitumor activity and minimize associated toxicities. Despite these efforts, the optimal dose and schedule for the taxanes is not known. Some recent phase I and II studies have been conducted to evaluate the use of weekly paclitaxel in patients with com-

26 mon types of solid tumors.98,100-109,135 These studies demonstrated promising results in terms of objective tumor response rates ranging from 20% to 78%. Although these studies were not large multicenter, randomized, placebocontrolled designs, the results suggested that paclitaxel given on a weekly schedule was potentially more effective than standard paclitaxel-containing regimens given every 3 weeks and was generally well tolerated with respect to hypersensitivity reactions and myelosuppression. The most commonly reported dose-limiting toxicity for the earlier weekly paclitaxel regimens was CIPN, with the overall reported incidence of CIPN ranging from 59% to 78%.95-98 Paclitaxel-induced neuropathy is reportedly more pronounced when the drug is administered via a short infusion schedule (1 hour v 3 hours v 24 hours). Recently, a randomized open label study of 1-hour versus 3-hour infusions of weekly paclitaxel (100 mg/m2) administration in patients with solid tumors was conducted for the purpose of evaluating the relative neurotoxicity between the two different paclitaxel infusion schedules.136 In this study, 92 of the 121 enrolled patients were evaluable for CIPN, and there was no significant difference in the incidence of CIPN between the two paclitaxel infusion schedules. It is notable that after 12 weeks of treatment, 75% of patients had developed CIPN, based on a standardized neurologic evaluation. The number of patients in each treatment group who had 25% dose reductions in paclitaxel due to the development of CIPN was similar (1-hour group, 10 patients; 3-hour group, 9 patients), and the number of patients experiencing treatment delays longer than 14 days was also similar (1-hour group, 7 patients; 3-hour group, 5 patients). Twelve patients experienced treatment delays exceeding 14 days. Three patients on the 1-hour regimen discontinued treatment due to CIPN whereas none of the patients on the 3-hour regimen discontinued therapy secondary to CIPN; this result was not statistically different. The findings from this small study suggest that the duration of the paclitaxel infusion by itself does not appear to have a significant effect on the incidence or severity of paclitaxel-induced neuropathy, but this study is underpowered to detect differences in the incidence of CIPN of less than 50%. It appears likely that there is a greater incidence of CIPN with paclitaxel administration using weekly and dosedense schedules (eg, every 2 weeks) as compared to administration every 3 or every 4 weeks. Weekly paclitaxel administration schedules may allow less time for neurologic recovery and may lead to a higher incidence of CIPN due to greater accumulation of paclitaxel in the peripheral nerves and greater disruption of peripheral axonal transport. However, this adverse outcome may be offset by greater efficacy of the weekly paclitaxel regimen. This was observed in a phase I study of weekly paclitaxel conducted by Seidman et al in patients with metastatic breast cancer where 9.5% of patients receiving ⱕ100 mg/m2 weekly and 56% of patients receiving 110 to 120 mg/m2 weekly experienced grade 3 neuropathy.96 The overall response rate was 53% (10% complete responses and 43% partial responses). Although in this study a median of 12 weekly paclitaxel infusions elapsed before the occurrence of neurotox-

Chemotherapy-induced peripheral neuropathy icity, paclitaxel neuropathy can also occur after a single dose.58 A large multicenter, randomized, phase III study was conducted by Cancer and Leukemia Group B (CALGB) to investigate the effectiveness of weekly paclitaxel compared with conventional administration every 3 weeks (CALGB study 9840).94 Patients were stratified according to the line of therapy and HER2 status. Approximately 580 patients were randomized to four possible treatment arms and received either weekly paclitaxel at 80 mg/m2 or paclitaxel every 3 weeks at 175 mg/m2 with or without trastuzumab, based on HER2 status. The initial study design included administration of weekly paclitaxel at 100 mg/m2 for the first 6 weeks followed by weekly paclitaxel at 80 mg/m2 thereafter, but due to the high incidence of severe sensory peripheral neuropathy (30% of patients with grade 3) the regimen was revised to include paclitaxel 80 mg/m2 weekly throughout the trial. The majority of patients (approximately 80%) were receiving paclitaxel as first-line therapy, and the treatment arms were well balanced for baseline characteristics (eg, performance status, extent of disease). The tumor response rate was statistically significantly higher in the group receiving weekly paclitaxel (40% v 28%, P ⬍.017). In addition, the weekly regimen produced a longer time to progression (9 months v 5 months, P ⬍.0008). Although the study was not powered to detect a difference between treatment groups for overall survival, there appeared to be a trend toward improved patient outcome with weekly paclitaxel (24 months v 16 months, P ⫽ .17). The toxicity profile for weekly paclitaxel administration compared with conventional every 3 weeks was significantly lower for myelosuppression (5% v 15%, P ⬍.013), but significantly higher for neurotoxicity (23% v 12%, P ⫽ .001). This study clearly demonstrates that weekly paclitaxel is an effective regimen with advantages in terms of both efficacy and safety. The development and approval of a chemoprotective agent that can effectively prevent or mitigate severe neurotoxicity associated with weekly paclitaxel may allow further optimization of this new regimen. Paclitaxel-related neuropathy is most commonly manifested as a sensory neuropathy that may be accompanied by motor neuropathy in some instances, but both axonopathy and neuronopathy clinical patterns have also been reported.4,5,58,66,137 All formulations of paclitaxel (Taxol [Bristol-Myers Squibb, Princeton, NJ], Abraxane, Xyotax, and Tocosol) have been reported to produce CIPN, which may be dose-limiting.138,139 Similar to platinum-induced CIPN, paclitaxel neuropathy manifests as a symmetrical progressive, length-dependent stocking-glove distribution of paresthesias, numbness, tingling, burning pain, dysesthesias, and decreased vibration and proprioception, which in more advanced stages is accompanied by the symmetrical loss of deep tendon reflexes with initial involvement of distal reflexes and later involving proximal reflexes. In patients with preexisting neuropathies and/or receiving high doses of paclitaxel, paclitaxel neuropathy can include motor and autonomic dysfunction and acute encephalopathy, resulting in coma and death.4,66,133,134 The reversibility of

F.H. Hausheer et al paclitaxel neuropathy has not been characterized in a large study, but it is reported to be partially reversible.4,7,66,133 Cremophor EL: Level of Evidence for a Causal Relationship to CIPN Some investigators have suggested that Cremophor EL (polyoxyethylated castor oil), a vehicle used to formulate paclitaxel, is responsible for the CIPN associated with paclitaxel administration.140-142 The level of medical and scientific evidence supporting the claim that Cremophor EL is directly involved in the pathogenesis of CIPN is very limited based on the following considerations: (1) the molecular weight of Cremophor EL is approximately 3,000, which generally limits tissue absorption and diffusion of this excipient, and the observed volume of distribution of Cremophor EL in humans and animals is very low, suggesting that the actual tissue distribution of intravenously administered Cremophor EL is limited due to the substantial inability of Cremophor EL to egress from the central vascular compartment142-144; (2) in vitro studies of Cremophor EL demonstrate that the vehicle is neurotoxic to cultured mammalian dorsal root ganglion and neurons, but this finding appears to be inconsistent with all reported in vivo clinical studies and human studies of Cremophor EL, which may be due to the absence of an intact vasa nervorum component to the in vitro dorsal root ganglion or neurite growth models, acting as an anatomic and physiologic barrier to the diffusion of Cremophor EL in vivo142-144; (3) the level of evidence that the taxane (paclitaxel and docetaxel) itself is highly neurotoxic appears to be compelling given the fact that paclitaxel has been observed to undergo rapid uptake and accumulation in mammalian dorsal root ganglion neurons in culture and that microtubule disruption in axons by paclitaxel has been observed; (4) none the available toxicology studies in animals demonstrate evidence of neurotoxicity for intravenously administered Cremophor EL; (5) the fact that other taxanes formulated in Cremophor ELfree solutions (including docetaxel, Abraxane, Xyotax, and liposomal paclitaxel) all produce severe dose-limiting CIPN with the same clinical manifestations and a similar incidence to Cremophor EL-formulated paclitaxel and some of these agents appear to have at least equivalent, and possibly more neurotoxicity, because higher doses of paclitaxel can be administered.16,35,38,54,138,139 All the evidence supports the fact that the taxane moiety itself is a neurotoxin in vivo, whereas Cremophor EL is not. Docetaxel Docetaxel, a semisynthetic analog of paclitaxel, reportedly has a toxicity profile different from paclitaxel. Some studies report that docetaxel is associated with minimal or mild neurotoxicity, yet some patients experience more severe neuropathy with cumulative docetaxel doses over 400 mg/ m2.16,35,38,54,59 Hilkens et al found moderate to severe neuropathy in 27% of patients who received more than 600 mg/m2 of docetaxel.38 Docetaxel-induced neuropathy is clinically indistinguishable from paclitaxel-induced neuropathy, and manifests as a symmetrical progressive, length-dependent sensory neuropathy often accompanied by paresthesia, pain, numbness, loss of dexterity, unsteady gait, sensory loss,

27 motor weakness, and loss of the distal deep tendon reflexes.16,35,54 Vibration and light touch are the most affected types of sensation. A higher incidence of docetaxel-induced neuropathy has been noted in schedules exceeding 36 mg/ m2/wk.6,16,35,54 In metastatic breast cancer studies, dose levels between 75 and 100 mg/m2 result in 5% to 14% grade 3 or 4 peripheral neuropathy.99 In phase II trials the incidence of docetaxel neurologic toxicity ranged from 6% to 59% for doses ranging between 75 to 100 mg/m2.54,99,104 The label reports the overall incidence of docetaxel-induced neuropathy ranging from 20% to 58%. Docetaxel-induced neuropathy is reported to be reversible in 95% of patients, although CIPN has been reported predominantly by the use of National Cancer Institute Common Toxicity Criteria (NCICTC) or similar methods, which have limitations that will be discussed below. Abraxane (nanoparticle albumin formulation of paclitaxel) Abraxane is a recently approved, albumin-stabilized nanoparticle formulation of paclitaxel designed to overcome poor water solubility and the hypersensitivity reactions associated with paclitaxel. Originally, there were reports that Abraxane was less neurotoxic than conventional paclitaxel formulated with Cremophor EL. In a controlled randomized phase III study, Abraxane in a dose of 260 mg/m2 administered as a 30-minute infusion every 3 weeks was associated with a significantly higher incidence of severe CIPN compared to paclitaxel formulated with Cremophor EL administered at 175 mg/m2 in a 3-hour infusion every 3 weeks. In this study, 10% of the patients treated with Abraxane experienced severe CIPN compared to 2% of the patients (P ⬍.001) who received paclitaxel formulated with Cremophor EL.145 Patients who experience CIPN with Abraxane have clinical symptoms and findings that are indistinguishable from CIPN due to administration of Cremophor EL-formulated paclitaxel. In patients who experienced NCI-CTC grade 3 sensory neuropathy from Abraxane, the reported median time to resolution was 22 days, with subsequent Abraxane dose reduction to 220 mg/m2, which allowed continued treatment at this level.145 The high incidence of severe CIPN was partially offset by the significantly decreased incidence of hypersensitivity reactions and myelosuppression. Cisplatin Versus Carboplatin and Paclitaxel Versus Docetaxel: Which Is Less Neurotoxic and What Is the Level of Medical Evidence? It is commonly reported that carboplatin is similar to cisplatin in terms of antitumor efficacy but that carboplatin administration is associated with a significantly lower incidence of CIPN. In making this type of claim, it is important to consider the dose level and schedule of administration for the agents used in the studies, as well as whether or not a randomized, controlled study design was employed to allow valid comparisons. Recent randomized controlled studies demonstrate that carboplatin may be similar or less efficacious and equally neurotoxic to cisplatin. A study conducted by Rosell et al85 demonstrated a statistically significant survival improvement in favor of cisplatin

28 versus carboplatin, given in combination with paclitaxel, and similar incidence of CIPN (58% cisplatin; 59% carboplatin). In this study both treatment groups received 200 mg/m2 of paclitaxel and either carboplatin AUC ⫻6 or cisplatin 80 mg/m2. Another randomized phase III study compared paclitaxel at 135 mg/m2 (24-hour infusion) plus cisplatin 75 mg/m2 vs. paclitaxel at 175 mg/m2 (3-hour infusion) plus carboplatin AUC ⫻7.5 in patients with optimally resected stage III ovarian cancer.71 This trial was designed to assess the non-inferiority of recurrence-free survival for the carboplatin regimen compared to the cisplatin regimen in 792 evaluable patients. Grade 2 to 4 neurologic toxicity was observed with similar frequency in both groups: 31% in the cisplatin arm and 28% in the carboplatin arm. There were no significant differences between the two treatment groups in progression-free or overall survival. The incidence of severe leucopenia and granulocytopenia was significantly higher in the cisplatin treatment group, whereas the incidence of thrombocytopenia and the incidence of grade 1 or 2 pain were significantly higher in the carboplatin treatment group. Assuming this study employed previously evaluated standard combinations of paclitaxel and cisplatin and paclitaxel and carboplatin, it would be difficult to conclude that carboplatin is actually less neurotoxic than cisplatin, particularly since the difference in the paclitaxel dose levels in the two treatment groups is 23%. Several studies have been conducted to investigate differences in efficacy and safety between docetaxel and paclitaxel in patients with advanced ovarian cancer and advanced NSCLC. In general, it appears that progression-free and overall survival are similar between the two compounds, while the reported safety profile is different in that docetaxel is associated with more frequent myelosuppression and paclitaxel is associated with more frequent neuropathy. A phase III trial (the Scottish Randomized Trial in Ovarian Cancer [SCOTROC]) involving 1,077 patients was conducted to evaluate potential differences in efficacy and safety between docetaxel and paclitaxel administered in combination with carboplatin in patients with advanced ovarian cancer. Patients were randomized to receive one of two treatments every 3 weeks for six cycles: (1) 75 mg/m2 docetaxel administered over 1 hour in combination with carboplatin; or (2) 175 mg/m2 paclitaxel administered over 3 hours in combination with carboplatin.146-148 No apparent statistical differences were noted in the efficacy parameters (eg, objective tumor response, progression-free survival, or overall survival), but differences were observed with respect to the toxicity profiles. The differences in the overall incidence of all grades of sensory and motor neuropathy for paclitaxel and carboplatin and docetaxel and carboplatin were highly statistically significant in this study. In the paclitaxel-carboplatin group, an increased incidence of sensory neuropathy and motor neuropathy was noted (sensory: 30% v 11%; motor: 8% v 3%). This very large randomized, controlled study strongly supports the observation that docetaxel is substantially less neurotoxic than paclitaxel. A randomized study of paclitaxel (135 mg/m2) and cisplatin (75 mg/m2), gemcitabine (1,000 mg/m2) and cisplatin

Chemotherapy-induced peripheral neuropathy (100 mg/m2), docetaxel (75 mg/m2) and cisplatin (75 mg/ m2), and paclitaxel (225 mg/m2) and carboplatin (AUC ⫻6) in 1,207 patients with advanced NSCLC demonstrated no significant differences between any of the four treatment groups in Eastern Cooperative Oncology Group (ECOG) grade 3 CIPN.149 The incidence of grade 3 CIPN was the greatest for the patients treated with paclitaxel plus carboplatin (10%), while the reported incidence of grade 3 CIPN was identical for patients treated with paclitaxel and cisplatin versus those treated with docetaxel and cisplatin (5% each). A prospective phase III trial of adjuvant docetaxel or paclitaxel, with or without preceding doxorubicin and cyclophosphamide treatment (four treatment groups), in 1,200 patients with early-stage breast cancer is currently ongoing in Japan. The results of this study will be important in terms of efficacy outcomes as well as the ability to make a direct comparison regarding the relative incidence of medically important CIPN associated with docetaxel and paclitaxel administration based on a controlled trial using standard dosages of the two taxanes in the adjuvant setting.

Vinca alkaloids (vincristine, vinblastine, vinorelbine, vindesine) Vincristine commonly produces sensory CIPN, but autonomic neuropathy and demyelination are also reported. The development of severe motor neurotoxicity accompanied by neurosensory impairment due to vincristine is rare,17 most likely because treatment is discontinued before severe motor weakness develops. The dose level and cumulative dose are the most significant risk factors for the development of severe vincristine-induced neuropathy. The peak incidence occurs 2 to 3 weeks following injection of vincristine. Recovery generally occurs 1 to 3 months after treatment cessation or after withholding drug or reducing the dosage, but CIPN symptoms may persist or worsen following vincristine therapy. Common symptoms experienced by patients on vincristine include symmetrical length-dependent paresthesias, pain in the hands and feet, muscle cramps, numbness and tingling in the hands and feet, loss of deep tendon reflexes, postural hypotension, urogenital dysfunction, and distally accentuated hyperesthesia.17 Sensory symptoms, most commonly reported as tingling paresthesias in the fingertips and the toes occur in virtually all patients treated with vincristine, although clinically detectable sensory loss is often present in early stages. Loss of ankle stretch reflexes is an early and almost universal sign, and with continued therapy all reflexes may diminish or disappear. Weakness in the form of a lengthdependent symmetrical, progressive distal axonopathy is the most common clinical presentation. When vincristine axonopathic weakness is mild, patients lose the ability to walk on their heels and lose strength in wrist extensors. More severely affected patients manifest footdrop and have a foot slapping gait. Motor weakness from vincristine can become severe enough to render the patient immobile, and treatment should be discontinued well in advance of the development of such marked weakness. Vincristine neurotoxicity generally appears after cumulative vincristine doses of 6 to 8 mg.150

F.H. Hausheer et al A considerable increase in the incidence and severity of vincristine-induced neuropathy occurs with total cumulative doses above 15 to 20 mg, and severe neurotoxicity occurs at cumulative doses greater than 30 mg. In patients with preexisting hereditary neuropathy (particularly CMT disease), there are reports of severe, vincristine-induced paralysis at total doses lower than 30 mg. This toxicity can be severe and may mimic Guillain-Barré ascending paralysis. The toxicity of vincristine is believed to occur through its binding to tubulin and its subsequent disruption of microtubule polymerization, which causes mitosis to arrest in metaphase.18 This tubulin disruption causes primary axonal degeneration through interfering with axonal transport and secretory function. In addition to vincristine, three other vinca alkaloids are used clinically: vinblastine, vindesine, and vinorelbine. The primary dose-limiting toxicity of vinblastine and vinorelbine is neutropenia. While vinblastine and vinorelbine are all reported to cause CIPN that is clinically similar to that of vincristine, the incidence appears to be lower than for vincristine.151

Thalidomide Thalidomide is an oral immunomodulatory agent used in the treatment of patients with multiple myeloma. The mechanisms of action of thalidomide are not fully characterized, but they may involve the suppression of tumor necrosis factoralpha production and down-modulation of selected cell surface adhesion molecules involved in leukocyte migration. Thalidomide-induced CIPN is a common and potentially severe side effect of therapy that may be permanent. Thalidomide causes distal axonal degeneration without demyelination. The incidence of thalidomide CIPN in larger studies is approximately 30%, and it is often reported as a dose-limiting outcome. Thalidomide-induced neuropathy generally occurs following chronic use over a period of months, but there are reports of thalidomide-induced neuropathy that have occurred after relatively short periods of therapy. Unlike other chemotherapeutic agents, there is no clear correlation between the cumulative dose of thalidomide and the development of CIPN. In addition, CIPN symptoms may occur some time after thalidomide treatment has been stopped, and these symptoms may resolve slowly or continue to persist. The clinical signs and symptoms of thalidomide-induced neuropathy include numbness and tingling, which can be painful and occur in a symmetrical distribution in the hands and feet. Patients should be periodically monitored for signs of CIPN during thalidomide treatment and should be counseled and questioned specifically for CIPN symptoms. If symptoms of thalidomide-induced neuropathy develop, it is recommended that thalidomide treatment be immediately discontinued to limit further damage, if clinically indicated. Thalidomide treatment may be resumed if the symptoms and findings of CIPN have resolved to baseline status. Other antitumor agents known to produce CIPN (eg, platinum, taxanes, and vincristine) should be used with caution in patients

29 receiving thalidomide treatment. The mechanism of thalidomide neuropathy is not known.76 An analysis of long term toxicity of thalidomide in 40 patients with multiple myeloma who received salvage therapy with thalidomide (200 to 400 mg/d) with or without dexamethasone (40 mg for 4 days every 4 weeks) for more than 12 months revealed that CIPN was the most troublesome and frequent toxic effect observed, with an incidence of 75%. Among the 30 patients with thalidomide-induced neuropathy, six patients had grade 1 (15%), 13 patients had grade 2 (32.5%), which required dose reductions of thalidomide, and 11 patients had grade 3 (27.5%), which required interruption of treatment despite treatment response. Quantitative sensory testing (QST) demonstrated a predominantly sensory neuropathy with minor motor involvement. The development of CIPN was not related to the thalidomide dose level, but it was clearly related to the duration of treatment in this study. These investigators reported that the benefits of thalidomide were offset by the frequency and severity of thalidomide neuropathy, particularly in responding patients, and that close neurologic monitoring should be considered mandatory for the management of patients treated with this agent.152 The development of CIPN was reported to be the most frequent important dose-limiting toxicity in 59 patients with multiple myeloma who were treated with thalidomide 100 to 400 mg/d plus melphalan 0.2 mg/kg/d for 4 days every 28 days. CIPN developed in 39% of the patients, and a median thalidomide dose of more than 150 mg/d was significantly associated with a higher frequency of CIPN and actuarial risk of CIPN. There was no benefit in terms of greater response at thalidomide dose levels of greater than 150 mg/d.153

Bortezomib Bortezomib is an intravenous medication also used for the treatment of multiple myeloma. The reported mechanism of action involves the reversible inhibition of the 26S proteasome in mammalian cells, which is a large protein complex that degrades ubiquitinated proteins. Inhibition of the 26S proteasome prevents the targeted proteolysis in cells, which affects multiple signaling cascades in the cell and can lead to cytotoxicity. In a large randomized study bortezomib administration was associated with the development of CIPN in approximately 37% of patients, with 14% experiencing grade 3 neuropathy.83 This toxicity is predominantly sensory in nature, although cases of accompanying motor involvement have been reported. Patients with preexisting CIPN may experience a worsening of peripheral neuropathy with the administration of bortezomib. The most commonly reported symptoms of bortezomib-induced CIPN include burning sensation, hyperesthesias, hypoesthesias, paresthesias, and discomfort from neuropathic pain. Dose reductions for bortezomib-induced neuropathy are reported to result in improvement or resolution of symptoms in the majority of patients, but the time course of this toxicity has not yet been fully characterized. The development of bortezomib-induced neuropathy was the most commonly reported adverse event

30 leading to discontinuation of treatment in 331 patients, and it occurred in 8% of the patients.83

Procarbazine, Cytarabine, Etoposide, and Alfa-Interferon Procarbazine is associated with the development of CIPN in 10% to 20% of treated patients. Procarbazine-induced neuropathy characteristically is manifested by neurosensory features that are similar to those produced by other chemotherapeutic agents. Cytarabine may cause CIPN, particularly with higher dose levels (eg, 1 g/m2) and in conjunction with the administration of other neurotoxic agents such as fludarabine.154 One case of fatal peripheral neuropathy was reported following administration of cytarabine and fludarabine therapy.155 Cytarabineinduced neuropathy is characterized by a symmetrical, length-dependent sensory and motor involvement, which may be permanent. Peripheral neuropathy associated with cytarabine has been characterized in both upper and lower extremities as muscle weakness, gait disturbances, walking difficulties, paresthesias, numbness, hypoalgesia, hypoesthesia, and myalgia.154 The mechanism of this CIPN is not known. Etoposide is occasionally reported to cause peripheral neuropathy (incidence: 1% to 2%). Patients with preexisting neuropathy or patients receiving concomitant neurotoxic chemotherapy (eg, vincristine) may be at greater risk for this toxicity.156 Alfa-interferon is associated with a very low incidence of peripheral neuropathy (⬍1%).157

Clinical Challenges in the Diagnosis and Quantification of Chemotherapy-Induced Peripheral Neurotoxicity for Clinical Trials The symptoms of CIPN are largely subjective in nature, and therefore the diagnosis and quantification of this clinical problem can be challenging for many reasons. Given the information known about CIPN, one may believe that a physician would be able to identify and diagnose the more severe forms of CIPN (grades 3 and 4) more consistently than less severe forms of CIPN (grades 1 and 2). However, several studies reported in the literature contradict this tenet and demonstrate not only disagreement among two different clinicians, but also increasing disagreement between physicians and patients with increasing symptom severity.89,91-92,158 The diagnosis and grading of CIPN is not always straightforward, despite the fact that there are several instruments available for this purpose, because there is no universally accepted or standardized method currently used for this assessment and many of the known grading scales used in clinical studies contain serious limitations with respect to the identification and clinical grading of CIPN. These considerations are not surprising given the fact there is currently no

Chemotherapy-induced peripheral neuropathy approved treatment or intervention available for the prevention or mitigation of this common and serious toxicity. To date, a successful standardized clinical diagnostic approach has not been defined and implemented, nor has there been approval of a compound for the prevention or mitigation of CIPN based on adequate and well-controlled clinical trials. A more reliable method for the identification and quantification of CIPN is needed in order to accurately evaluate potential neuroprotective agents. To obtain a new drug approval, regulatory agencies, such as the Food and Drug Administration (FDA), require substantial evidence of effectiveness generally from two adequate and well-controlled clinical trials,159-161 as well as demonstration of patient safety. Specifically, the assessment of treatment response (efficacy) must be well-defined and quantified by reliable diagnostic methodologies.

Selection, Standardization, and Determination of Reliability of CIPN Evaluation Methods The use of patient-reported outcomes or health-related quality of life for the evaluation of a new drug must involve the same rigorous scientific methods and principles in terms of study design and analysis as traditional end points of survival and objective response rates.162 When selecting and applying a reliable method for the assessment of CIPN in clinical trials, several factors must be taken into account in addition to the considerations described above: (1) the symptoms (numbness, paresthesias, dysesthesias, and pain) and severity of CIPN are subjective in nature, which is clinically analogous to the symptoms of depression, pain, or nausea (ie, there is no laboratory or physical examination test that can grade their severity, other than history and observation); (2) a primary neuroprotection end point in a clinical study must be clinically meaningful and quantified by a reliable methodology; and (3) the diagnostic method employed must be sufficiently sensitive, specific, and able to reliably detect the presence of CIPN, as well as the detection of increased or decreased grades of CIPN. The use of diagnostic methods for CIPN assessment should be practical and convenient for patients and health care providers and should not require any invasive procedures or large amounts of time and resources to perform. The development and validation effort for a medical assessment instrument for use in the quantification of subjective symptoms is a comprehensive and long-term process.163,164 We formulate and propose several important design and analysis parameters that should be evaluated during the development of a CIPN assessment instrument including the following. (1) sensitivity—the ability to detect the presence or absence of, and reliably distinguish, medically important grades of CIPN in patients and thereby enable determination of the overall incidence of CIPN and subsets of CIPN grades in defined patient populations;

F.H. Hausheer et al (2) compliance—the method should be used without omissions or patient refusal in a high proportion (ⱖ80%) of evaluations. (3) specificity—the ability to detect CIPN reliably versus other neurologic disorders in defined patient populations, unless such patients are excluded due to medical risks (eg, diabetes). It is important to recognize the improbability that any assessment method will be 100% specific for CIPN, because there are many other conditions or toxicities that may mimic some or all of the symptoms and signs of CIPN; (4) reliability—the ability of the assessment method when administered to the same patient at separate but contemporaneous time points (intra-patient variability), and when used by different examiners (inter-examiner variability), to produce the same reported result. Reliability is further determined by the inter-trial and multi-institutional reproducibility and concordance of the method being applied to populations of similarly treated patients; (5) responsiveness–the ability of the method to detect a true increase or decrease in the level of patient symptoms as a function of dose level, the number of treatments, duration of treatment, or following the completion of treatment; (6) the method should demonstrate correlation of the severity grade of CIPN with cumulative neurotoxic exposure (if appropriate) and by correlation of severity grades of CIPN with medical outcomes, particularly with modification, delays, or cessation of treatment; (7) the method should identify the presence or absence of symptoms of potential functional impairment of the patient’s activities of daily living, and when symptoms are present, the method should further identify the specific type of activities of daily living that are impaired or interfered with; and (8) the method should have clear demarcation for medical decisions to modify dose, withhold or discontinue neurotoxic treatment based on the level of symptomatic functional impairment, and should identify the difference between such clinically important neurotoxicity as an endpoint versus less severe neurotoxicity. The foregoing are difficult standards to achieve, and presently there are no reliable diagnostic methods to assess CIPN that have been advanced into medical practice and incorporate all of these parameters. For the purpose of this review, we will apply the foregoing parameters as definitions for assessing the reliability of diagnostic, management, and reporting methods for CIPN. The ability to diagnose and assess the severity of CIPN accurately and predictably regarding symptoms that interfere with patients’ activities of daily living, as differentiated from less severe forms of CIPN, is a key primary objective for the future medical management of patients with cancer, as well as for reporting or as a primary endpoint for CIPN from controlled studies. In practice the evaluation and manage-

31 ment decisions of patients with CIPN are made by the treating physician, who must decide if the CIPN is severe enough to modify, delay, or discontinue neurotoxic treatment, all of which may affect outcome of therapy. There is currently no gold standard for the assessment of CIPN. One major challenge with the accurate assessment and quantification of CIPN is that various scales and methods are available for use without a universal or standardized approach that is consistently or precisely implemented. Several types of instruments used for the assessment of CIPN include physician-based instruments such as the National Cancer Institute Common Toxicity Criteria for Adverse Events (NCICTCAE), World Health Organization (WHO), ECOG, and Ajani criteria, and quantitative sensory tests (QST) such as vibration perception threshold (VPT), nerve conduction velocity tests (NCV), nerve biopsy, or electrophysiologic measurements such as electromyography (EMG), which have usually been applied in combination with neurologic evaluations. There are many studies that report the use of objective assessments of CIPN including neurologic evaluation, QST, electrophysiological testing, and nerve biopsy in an attempt to diagnose and objectively quantify CIPN. The limitations of these approaches include the lack of standardization, poor correlation of objective findings with patient reported symptoms and severity, the additional time and resources to perform the testing, and the fact that some (eg, NCV testing) are invasive and painful to patients. Another important consideration for the assessment of CIPN is that it may be unreasonable to expect an objective method or proxy could reliably diagnose or quantify the severity of a predominantly subjective medical condition. The clinical assessment of pain, depression, and nausea, which are analogous to the subjective nature of CIPN, are most commonly and reliably assessed by patient-based questionnaires and scales that have been demonstrated to be reliable and convenient for patients and their providers.165,166 We will review these approaches, because they are illustrative of the fact that there appears to be no clinical advantage, earlier detection, or ability to achieve greater reliability in using objective diagnostic methods, proxies, or physician-based methods to assess CIPN.55-57,64,65,88,89,91,92,158

Quantitative Sensory Tests and Electrophysiologic Measurements in Combination With Neurologic Evaluations Many studies reported in the literature employed the combined use of clinical neurologic evaluation and QST such as VPT or electrophysiologic measurements (EPM) such as NCV, nerve biopsy, or EMG in the assessment of CIPN.15,42,55-57,64,72,137,167-171 Inconsistent results have been reported from studies with these types of tests. However, in general, poor correlation between the measures and clinical symptoms and severity of CIPN were documented. The diagnostic value of these types of tests for CIPN has not yet been proven. These methods also do not permit earlier diagnosis of CIPN as compared to patient reporting or physician examinations.55,56,64,172 It is not an unforeseen observation that

32 such objective measurements of peripheral nerve function do not appear to correlate with the clinical manifestations of CIPN, given the fact that the symptoms of CIPN are subjective. Although used occasionally in the assessment of diabetic neuropathy,173,174 QST and EPM are not routinely used to make treatment decisions for patients with CIPN, and none of these assessments would meet the criteria we have proposed for diagnosis, management, and reliability in this patient population, either in practice or as endpoint assessments of CIPN. In addition, no adequate and well-controlled studies have been presented using these measurements as the primary endpoint for the assessment of CIPN. None of the studies involving QST, biopsy, and EPM have been subjected to adequate well-controlled reliability determination as described above in the evaluation of CIPN. Many of the studies reported in the literature are comprised of small sample sizes and are underpowered for the end points or employed uncontrolled study designs. There are many studies that report high intra-patient variability and the inability to reproduce reliable results with the available instruments. In general, none of these studies have demonstrated that QST and EPM reliably diagnose clinical symptoms of CIPN, and more important, none of these studies provide any description of the reliable identification of patients with clinically important functional impairment secondary to CIPN for the purpose of medical decision making. One important aspect of EPM such as NCV assessments is poor patient compliance due to a high incidence of discomfort during the procedure. It is not uncommon for patients to withhold consent to repeat NCV measurements, which further limits the utility of this approach. In addition, the major limitation of NCV testing is that it measures velocity and amplitude in the largest diameter and fastest conducting nerve fibers and therefore does not provide reliable data on the small-sized fibers. A possible more important consideration is the fact NCV and nerve biopsy are commonly applied to anatomic regions that are proximal to the anatomic stocking-glove distribution of CIPN, which may partially explain the apparent poor correlation and diagnostic utility of these methods. It has also been reported that NCV testing reflects the status of the best surviving nerve fibers, and the NCV may remain normal if even a few fibers are unaffected by a disease process. Thus, a normal NCV test result can still occur despite extensive nerve damage. The most sensitive neurologic signs of CIPN are the presence of impaired vibration sensation, proprioception, as well as impaired epicritic (two point discrimination) touch. VPT testing is a direct measurement of the objective impairment in these sensory parameters. However, the electronic instruments used to assess VPT are not widely available, and decreased vibration perception is not 100% specific for CIPN (eg, dorsal column disease). Objective measures of assessing CIPN, including NCV measurements, EMG, nerve biopsy, and detailed neurologic examinations, have not demonstrated sufficient clinical reliability in the assessment of CIPN and are not routinely used to make clinical decisions in the management of patients. The use of NCV, EMG, nerve biopsy, and other neurophysiologic

Chemotherapy-induced peripheral neuropathy evaluations may be helpful if there is another neurologic disorder (eg, diabetes) that may be coexistent or another potential diagnosis (eg, B1, B12, or pyridoxine deficiency, pyridoxine intoxication, or other etiologies). The diagnosis of CIPN should be made in part by careful exclusion of other potential etiologies. Berger et al conducted a study in 14 cancer patients who had received at least one to seven courses of paclitaxel and cisplatin therapy (the cumulative doses for each drug ranged from 175 mg/m2 to 1,225 mg/m2 and 100 mg/m2 to 700 mg/m2, respectively).55 Patients underwent neurologic examinations conducted by one physician, which included the use of a standardized evaluation of sensory symptoms, vibration sense (with a tuning fork), strength, and deep tendon reflexes. In addition, sensory potentials were obtained from patients who underwent electrophysiologic measurements (NCV) of the peroneal nerve. The patient distal latencies (msec), conduction velocities (m/s), and amplitudes (mV) were measured in this study. Of the 14 patients, 12 (86%) experienced sensory neurotoxicity, predominantly numbness or paresthesias in a stocking and glove distribution.55 The authors reported that the use of EPMs did not result in an earlier detection of neurosensory symptoms of CIPN as compared to the clinical neurologic examination alone. Forsyth et al conducted a study in 37 patients with metastatic breast cancer who had failed prior chemotherapy.64 Patients received either 200 mg/m2 or 250 mg/m2 of paclitaxel every 3 weeks given in a 24-hour infusion until disease progression. The patients completed neurologic examinations and underwent QST during treatment. The average number of treatment cycles received was 7.3, with an average duration of 20.1 weeks. A total of 34 patients underwent the QST measurements where VPT and temperature threshold (TT) testing were performed, and the results were compared to normal healthy controls. A total of 31 patients (84%) developed symptoms of CIPN after an average of 1.7 treatment cycles (average cumulative paclitaxel dose of 371.5 mg/m2). Although abnormal results were noted in the VPT and TT measurements, CIPN identified and quantified by QST was found to be less sensitive than the clinical examination and patient history. No significant correlation was observed between the cumulative dose of paclitaxel and changes in any of the QST parameters.64 du Bois et al conducted a study to evaluate paclitaxelinduced CIPN as a function of cumulative dose, treatment duration, and infusion schedule.56 A total of 38 cancer patients receiving either a 3-hour or 24-hour paclitaxel infusion at 135 mg/m2 or 175 mg/m2 were included in the study and were compared to a control group of 140 healthy female volunteers. A total of 227 chemotherapy treatments were administered with an average of 6 treatments per patient. A patient questionnaire and QST measurements were performed including VPT. Approximately 76% of the patients in the paclitaxel treatment group experienced symptoms and signs of CIPN. A disproportionate lack of sensitivity in VPT measurements was observed in the paclitaxel treated patients in that 71% reported symptoms of paresthesias and numbness as assessed by

F.H. Hausheer et al patient questionnaire, while only 48% these patients had an increase in VPT, representing a 23% under-assessment of CIPN by VPT. These results indicate that VPT methods appear to under-assess paclitaxel-induced CIPN substantially. In addition, no significant correlation was observed between cumulative paclitaxel dose and impaired VPT measurements. The authors emphasized the need for a more reliable, sensitive, and accurate methodology for the assessment and quantification of CIPN.56 Chaudhry et al reported clinical and electrophysiologic results from a single institution in 32 patients who underwent treatment with paclitaxel and cisplatin in combination, with paclitaxel doses of 135 mg/m2 up to 350 mg/m2 and cisplatin doses of 75 mg/m2 up to 100 mg/m2.42 These investigators employed a total neuropathy scoring system that included physician-based assessment of the quantification of taxane-platinum CIPN by sensory symptoms, pin sensibility, strength, tendon reflexes, and vibratory threshold by QST, and by electrophysiologic testing including the measurement of sural and peroneal amplitude. Thirty of the 32 patients underwent neurologic evaluations before receiving the chemotherapy. Nine (30%) did not have follow-up evaluations and were excluded from the analysis. No explanation was given for this high percentage of patients failing to complete follow-up. During treatment, 16 (76%) of 21 patients complained of paresthesias or numbness in the feet, and 12 (57%) developed similar symptoms in the fingers. A patient with diabetes developed disabling symptoms after a single treatment dose. Twenty (95%) patients had at least one abnormal finding on examination, with absent or reduced ankle reflexes being the most frequent (19 patients; 90%).42 VPTs were elevated in the toes of 17 (81%) patients, and distal loss of pinprick sensation was observed in 11 (52%) patients. The severity of the sensory loss was reported to correlate with the cumulative paclitaxel dose. In 18 (86%) patients, motor examination showed mild distal weakness of the toe extensors. Nerve conduction testing and sensory VPT testing in these patients demonstrated significant reduction in nerve function from baseline.42 The investigators concluded that approximately 95% of the evaluable patients in their study developed a sensory, motor, axonal, length-dependent, and symmetrical peripheral neuropathy that was observed when higher doses of paclitaxel (up to 350 mg/m2) were administered in combination with cisplatin (75 or 100 mg/m2).42 Most of the patients’ symptoms were sensory in nature, but physical examination and EPM demonstrated that the paclitaxelcisplatin CIPN commonly had sensory and motor components. Although the severity of the CIPN reported in this study was clearly related to the cumulative dose of paclitaxel, the study also suggested that the paclitaxel dose given each course was an important factor, particularly in patients who received a total dose of paclitaxel ⱖ300 mg/m2 at each cycle. In summary, the medical literature demonstrates that QST, nerve biopsy, and EPM measurements by themselves cannot reliably assess the severity of CIPN. Therefore,

33 these methods should not solely be used for diagnosis and treatment decisions. Unfortunately, the most important clinical objective appears to be unaddressed by all of these studies in that none employ a method that reliably determines when a patient’s level of functional impairment would justify dose modification, treatment interruption, or discontinuance.15,24,55-57,64,72,87,90-92,137 In addition, because the medical diagnostic value of QST and EPM has not been established, the costs of these methods when used in the assessment of CIPN are currently not reimbursed by US healthcare companies (eg, Aetna) or the Centers for Medicare and Medicaid Services (CMS).175

Physician-Based Methods of Assessing Chemotherapy-Induced Peripheral Neuropathy There are many physician-based assessment methods for CIPN published in the medical literature, that are often used either alone or in combination with QST and EPM. It is important to review and evaluate these physician-based methods to achieve a clear perspective on the development and actual level of medical and scientific evidence as to their reliability and utility in the diagnostic assessment and reporting of CIPN in clinical practice as well their suitability for use in controlled clinical trials. The most widely recognized physician-based approaches to assess CIPN include the NCI-CTCAE, ECOG, WHO, and Ajani criteria (Table 6). These types of assessments require both patient cooperation and physician skill to obtain the essential diagnostic information regarding CIPN. In general, physician-based grading systems typically range from grade 0 (indicating normal findings) to grade 4 or 5 (indicating most severe symptoms or findings or death, respectively). The clinical differences between grade 1 and grade 4 CIPN are conceptually straightforward, but the reliability in assessing all grades of CIPN, in particular grades 2 and 3, appears to be far more challenging due to the subjective nature of CIPN and the potential ambiguity of interpretations of the grading definitions. The most widely used physician-based grading scale is the NCI-CTCAE, which will be discussed in detail. Studies in the literature report that physician-based assessments of subjective symptoms such as numbness and tingling are associated with a high proportion of variability between examiners as well as variability between the physician score and the reported patient score. This variability increases as the observed severity of CIPN symptoms increases. Under-assessment of the incidence and severity of CIPN by the physician are also noted.91,92 Postma et al158 conducted an important clinical study to evaluate the inter-examiner and inter-test reliability between several widely used physician-based CIPN assessment methods. They included a prospective analysis of the differences in the severity of the CIPN grade assigned to patients by two different neurologists in the same institution who had independently and contemporaneously performed examinations using the different physician-based CIPN assessment meth-

Chemotherapy-induced peripheral neuropathy

34 Table 6 Physician-based Neuropathy Grading Scales Scale

Grade 1

Grade 2

NCI-CTCAE (version 3.0) Neuropathy: sensory Asymptomatic: loss Sensory alteration or paresthesia of deep tendon (including reflexes or tingling), paresthesia interfering with (including function but not tingling) but not interfering with interfering with ADL function Neuropathy: motor Asymptomatic, Symptomatic weakness of weakness exam/testing only interfering with function but not interfering with ADL ECOG Sensory Mild paresthesias; Mild or moderate loss of deep objective tendon reflexes sensory loss; moderate paresthesias Motor Subjective Mild objective weakness; no weakness objective findings without significant impairment of function Ajani Sensory Paresthesia, Mild objective decreased deep abnormality, tendon reflexes absence of deep tendon reflexes, mild to moderate functional abnormality WHO Severe Paresthesias and/ paresthesias or, decreased and/or mild deep tendon weakness reflexes

Grade 3

Grade 4

Sensory alteration Disabling or paresthesia interfering with ADL

Grade 5 Death

Weakness interfering with ADL: bracing or assistance to walk indicated

Life-threatening: disabling (eg, paralysis)

Death

Severe objective sensory loss of paresthesias that interfere with function Objective weakness with impairment of function

—

—

Paralysis

Severe paresthesia, moderate objective abnormality, severe functional abnormality

Complete sensory loss, loss of function

Intolerable paresthesias and/or marked motor loss

Paralysis

—

Abbreviation: ADL, activities of daily living.

ods.158 The study included patients who received neurotoxic chemotherapy and who had symptoms and signs of CIPN. A standardized approach for the neurologic examination was specifically not employed in an effort to mimic the clinical practice setting. The neurologist grading of CIPN in each patient was made by the following assessment methods: (1) National Cancer Institute of Canada-Common Toxicity Criteria (NCIC-CTC; adapted from the NCI-CTC); (2) WHO; (3) ECOG; and (4) Ajani criteria. The two neurologists were blinded to each other’s independent neurologic assessment until the conclusion of the study and completion of data analysis. The concordance between the two different neurologists was evaluated and expressed as percentage agreement (or disagreement) for each grade of severity (grade 0 to grade 4) of each CIPN assessment scale. A total of 37 patients and

148 observations (four ratings per patient) were included in the analysis. The majority of patients were women with ovarian cancer who had previously received treatment with paclitaxel and cisplatin. There was disagreement between the two neurologists on at least one of the four different grading scales in 80% of the patient evaluations (complete agreement on all grades of all scales was noted in only 20% of the patients). The percentage of overall interobserver agreement on all grades of CIPN using the four grading scales ranged from 46% to 84% (NCICCTC ⫽ 46%, Ajani ⫽ 57%, ECOG ⫽ 76%, and WHO ⫽ 84%). A statistically significant difference (P ⬍.01) in mean scores (overall grades) for both the NCIC-CTC and Ajani grading systems was noted between the two examining neurologists.158

F.H. Hausheer et al In this study a comparison of the different grading scales showed that the overall (grades 0 to 4) interobserver agreement was lowest using the NCIC-CTC grading scale at 45.9%, while the exact agreement on severe (grade 3) neuropathy using the NCIC-CTC grading scale was only 42% (58% disagreement). The proportion of agreement between neurologists using the ECOG criteria for grade 3 CIPN was 40% and was 0% for the WHO and Ajani criteria. Although grade 3 CIPN was most often reported with the use of the NCIC-CTC grading scale, it is notable that two different neurologists disagreed 58% of the time on the diagnosis of grade 3 CIPN that is considered severe enough to interfere with patient function.158 Importantly, this study demonstrates that the evaluation criteria and scoring for physician-based scales are not interpreted the same way by different examiners, and therefore a high variability in the scoring of CIPN is possible. The authors state that the disagreement in grading CIPN resulted from differences in the interpretation of patient signs and symptoms related to “interference with function.” This study also noted that “If one is interested in incorporating the patient’s own subjective experience of daily living functional impairment in such evaluations, it is clear that the existing grading symptoms are insufficient.”158 This study highlights the importance of developing a reliable assessment method for CIPN and further that findings of CIPN reported using different medical assessment methods should be interpreted with caution because there is a great deal of variation in the CIPN scores among methods. Although physician-based instruments are widely used in the assessment of CIPN for the purpose of treatment decisions, there are several important limitations that must be considered and discussed when evaluating this approach for use as the primary efficacy end point in a clinical trial with a potential neuroprotective agent. The most commonly used CIPN assessment scales are the physician-based NCI-CTCAE grading system that involves patient evaluation and does not contain standardized or defined evaluation parameters for CIPN. The NCI-CTCAE is a physician-based grading system that includes criteria and definitions for quantifying and grading CIPN (both neurosensory and neuromotor components), but it currently contains no clear demarcation of clinically important neuropathy, which may in part contribute to the high inter-examiner variability reported with the use of this grading scale. The NCI-CTCAE grading scale has been modified over time and although currently references no interference (grade 2) versus interference (grade 3), no specific ADLs are defined. This grading scale comprises a sensory and motor assessment that utilizes a 5-point scale ranging from grade 1 (findings are normal) to grade 5 (death) as shown in Table 6. Although the NCI-CTCAE system is widely used, there are major considerations that limit its utility and reliability and those of other similar grading systems in assessment of CIPN. The evaluation parameters for each NCI-CTCAE CIPN grade are not clearly defined, including specific neurologic examination parameters. Particularly, the “loss of deep tendon reflexes” in the definition of grade 1 sensory neuropathy is not

35 specified and does not describe the anatomic locations for such reflexes which must be absent to qualify for grade 1 or higher grades of CIPN (the problem is not the location but whether this means loss of just any or all DTRs to qualify as grade 1). The authors have found highly variable responses when physicians are asked which reflexes must be absent and what sensory or motor findings or levels of functional impairment must be present in grade 1 or higher grades using the NCI-CTCAE. Some respondents report only absence of Achilles reflexes, while others suggest absence of all major reflexes as an appropriate test. The description of “sensory alteration” in the definition of grades 2 and 3 sensory neuropathy does not categorically define whether the degree of symptoms is an increase or decrease in sensory functioning of the affected locations, or how it is determined (eg, by history or by a standardized neurologic examination). The diagnostic provision of “interfering with function, but not interfering with activities of daily living” appears to be ambiguous and may be interpreted inconsistently by patients and physicians. Additionally, although the interference with activities of daily living is stated, the specific activities and level of function that are compromised are neither defined nor captured as part of this assessment instrument. There are no sensory or motor testing criteria (eg, pin-prick, pain, vibration, proprioception, or distal muscle strength testing; with particular attention to the glove and stocking distribution) specified or utilized in the NCI-CTCAE. These limitations clearly are important in the consideration of applying the NCI-CTCAE and similar grading scales in clinical decision making and in clinical trials to assess CIPN as a primary efficacy or toxicity endpoint. Clinical trial endpoints must have clear definitions for each grade, and to be useful and interpretable they must reliably identify and demarcate a clinically important toxicity level from clinically allowable toxicity levels. In addition, it is important to recognize that the reliability of NCI-CTCAE and other similar physician-based grading scales has not been determined by a prospective study. In using the NCI-CTCAE system for CIPN the physician must elicit, assess, and record the patient’s symptoms and signs and determine their severity and degree of functional impairment as well as other supportive findings. In practice, many physicians would agree that the presence of NCICTCAE grade 3 CIPN would be justification for treatment delay, dose modification, or treatment cessation. However, the previously described variability between physicians, and even neurologists, in determining the NCI-CTCAE numerical value would make the justification for treatment delay, modification, or cessation ambiguous at best. In addition, acute onset oxaliplatin CIPN cannot be reliably assessed or captured with the use of NCI-CTCAE, because no provisions for the acute neurotoxicity symptoms associated with oxaliplatin have been devised (eg, pharyngolaryngeal dysesthesias or myotonia provoked or exacerbated by cold). Based on the above-mentioned limitations of the NCI-CTCAE, it remains difficult to quantify the incidence and severity of CIPN accurately in cancer patients, and it is especially challenging to

Chemotherapy-induced peripheral neuropathy

36 compare the incidence and severity across different clinical studies. Due to the absence of means to assess acute oxaliplatininduced neurotoxicity, a specific oxaliplatin neurotoxicity scale was developed (Table 7) in an attempt to quantify the incidence and severity of neuropathy associated with this agent, including the unique features of the acute form of its toxicity as well as its cumulative and more persistent form.115,116,120 Similar limitations as described above for the NCI-CTCAE scale also apply to the oxaliplatin specific scale and there are additional considerations: (1) the concept that grade 1 (regression) or grade 2 (regression) (persistence) of dysesthesias and paresthesias is not linked to functional impairment or interference with activities of daily living, and may be incorrect from a patient perspective; (2) functional impairment is not clearly defined and could be interpreted as ranging from any impairment, such as difficulty fastening a small clip, to interference with activities of daily living. In the event that the physician fails to elicit the presence, persistence, or progression, and the severity and symptoms regarding functional impairment, this instrument may miss potentially important CIPN due to oxaliplatin.

Patient Based Methods of Assessing Chemotherapy-Induced Peripheral Neuropathy Based on these and other considerations, we developed a new patient-based instrument for CIPN associated with cisplatin, carboplatin, and taxane therapy, as well as a modified version of this questionnaire to include the important clinical and physical attributes affected by the development of acute oxaliplatin neuropathy for assessment and quantification of CIPN. Recently it has been reported that the Functional Assessment of Cancer Therapy/Gynecologic Oncology Group-Neurotoxicity (FACT/GOG-Ntx) and the FACT-Taxane patientbased instruments have been studied in patients with CIPN.176-178 These patient questionnaires are more discerning but contain questions that are not specific for the assessment of CIPN (eg, I have trouble hearing, I get ringing in my ears). Most notably there is not a score level that demarcates the presence or absence of severe symptoms involving impairment of prespecified activities of daily living that would aid the physician in medical decision making for dose modification, treatment delay, or discontinuation. These questionnaires are not designed to capture the clinical features associated with the acute form of oxaliplatin CIPN.

In an effort to address these unmet needs, we developed a new patient-based questionnaire called the Patient Neurotoxicity Questionnaire (PNQtaxane/cisplatin/carboplatin) for taxanes and cisplatin or carboplatin, and a modified version of this instrument for oxaliplatin (PNQoxaliplatin) (Figs 1 and 2). The PNQ instrument is a simple self-administered patient questionnaire that was designed and developed at BioNumerik Pharmaceuticals, Inc (Frederick Hausheer and Elmer Berghorn), facilitated by extensive discussions with FDA staff and other physicians and nurses familiar with CIPN, for the specific purpose of determining the incidence of clinically significant CIPN functional impairment in registration trials conducted to evaluate the potential efficacy of chemoprotective agents. Specifically, the PNQ was developed for use as the primary neuroprotection end point for pivotal studies evaluating the effectiveness of BNP7787, an investigational neuroprotective agent for the prevention of CIPN caused by taxane and platinum therapy. We were also concerned with overcoming the high degree of variability, under-assessment, and under-reporting of CIPN, discussed previously and with the failure of other instruments to adequately assess the degree of functional impairment of daily living activities in patients at risk for development of CIPN. It is well recognized that patient-reported symptoms and functional impairment due to CIPN are predominantly subjective and involve functional impairment of daily activities associated with the use of the feet and/or hands. Other available patient questionnaires (eg, FACT/GOGNtx and FACT-Taxane) do not define the demarcation or scoring of medically important CIPN functional impairment of activities of daily living that can be ascertained by this instrument. These questionnaires report an overall numerical score that is the sum of several subscores, which include neurosensory, neuromotor, and autonomic symptoms. This summation of subscores leads to a total score that does not demarcate the absence or presence of clinically important CIPN. In addition, no medical interpretation of functional impairment of activities of daily living is provided by this scoring system. A larger numerical score from this instrument suggests greater neurotoxicity or decreased quality of life, or both. The FACT/GOG-NTX is a useful instrument that can be complimentary to other approaches. These considerations as well as others provided the stimulus to develop a new patient-based instrument for CIPN assessment that was aimed at determining and capturing the presence or absence of functional impairment secondary to CIPN in the form of defined activities of daily living, and to create a simple questionnaire that had

Table 7 Oxaliplatin-Specific Neurotoxicity Grading Scale Used in Clinical Trials Grading Scale

Grade 1

Grade 2

Grade 3

Grade 4

Oxaliplatin-specific neurotoxicity scale

Dysesthesias or paresthesias that completely regressed before the next cycle of therapy

Dysesthesias or paresthesias persisting between courses of therapy

Dysesthesias or paresthesias causing functional impairment

Not applicable

F.H. Hausheer et al

37

Patient Neurotoxicity Questionnaire (PNQ) © Taxanes, Cisplatin and Carboplatin Item 1.

A

B

I have no numbness, pain or tingling in my hands or feet.

I have mild tingling, pain or numbness in my hands or feet. This does not interfere with my activities of daily living.

C

D*

I have moderate tingling, pain or numbness in my hands or feet. This does not interfere with my activities of daily living.

E*

I have moderate to severe tingling, pain or numbness in my hands or feet. This interferes with my activities of daily living.

I have severe tingling, pain or numbness in my hands or feet. It completely prevents me from doing most activities of daily living.

Item 2.

A

B

I have no weakness in my arms or legs

I have a mild weakness in my arms or legs. This does not interfere with my activities of daily living.

C

D*

I have moderate weakness in my arms or legs. This does not interfere of my activities of daily living.

E*

I have moderate to severe weakness in my arms or legs. This interferes with my activities of daily living.

I have severe weakness in my arms or legs. It completely prevents me from doing most activities of daily living.

* Please indicate by placing an X in the box or writing in the space provided which activity or activities have been interfered with as a result of therapy. My ability to: Button clothes

Open doors

Fasten buckles

Write

Sew

Use a knife

Put in or remove contact lenses

Sleep

Walk

Work

Use a fork

Dial or use telephone

Climb stairs

Put on jewelry

Tie shoes

Use a spoon

Operate a remote control

Type on a keyboard

Knit

Drive

Other eating utensils, etc

Perform activities of importance to me, specify: _______________________________________

Figure 1 Patient Neurotoxicity Questionnaire: taxanes, cisplatin, and carboplatin.

fewer questions, thereby allowing greater compliance and reduced omissions in reporting and convenience for patients and providers. The primary impetus for designing the PNQ was the noted inability of the NCI-CTC and similar physicianbased grading scales to reliably assess and quantify CIPN and in particular distinguish the presence or absence of medically important symptoms. Further, when CIPN symptoms are present, it is important to be able to identify whether or not the patient is experiencing interference with activities of daily living, which are specifically defined in the instrument. The PNQ is comprised of specific questions designed to obtain specific information directly

from the patient regarding the incidence and severity of CIPN, using a clearly defined breakpoint between no interference versus interference with defined activities of daily living. This demarcation is specifically between PNQ Grades C and D, which corresponds to the absence (grade C or less) or presence (grade D and higher) of neurosensory and neuromotor symptoms that either do or do not interfere with activities of daily living. Another purpose in developing the PNQ was based on many published reports that document the fact subjective symptoms of numbness, pain, paresthesias, and weakness are commonly underreported by physicians and further that a high percentage (25% to 50%) of physicians report lower degrees of sever-

Chemotherapy-induced peripheral neuropathy

38

Patient Neurotoxicity Questionnaire (PNQ) © Oxaliplatin Item 1.

A

B

I have no numbness, pain, burning, tingling or change in my sense of touch in my hands/fingers, or feet/toes or mouth area

I have mild numbness, burning, pain, tingling or change in my sense of touch in my hands/fingers, or feet/toes or mouth area. This does not interfere with my activities of daily living.

C

D*

I have moderate burning, numbness, pain, tingling or change in my sense of touch in my hands/fingers, or feet/toes or mouth area. This does not interfere with my activities of daily living.

I have moderate to severe burning, numbness, pain, tingling or change in my sense of touch in my hands/fingers, or feet/toes or mouth area. This interferes with my activities of daily living.

E* I have severe numbness, pain, tingling or change in my sense of touch in my hands/fingers, or feet/toes or mouth area. It completely prevents me from doing most activities of daily living.

Item 2.

A

B

I have no difficulty in swallowing, breathing, drinking or chewing food, or muscle spasms in my mouth/jaws, hands/fingers or feet/toes.

I have a mild difficulty in swallowing, breathing, drinking or chewing food, or muscle spasms in my mouth/jaws, hands/fingers or feet/toes. This does not interfere with my activities of daily living.

C

D*

I have moderate difficulty in swallowing, breathing, drinking or chewing food, or muscle spasms in my mouth/jaws, hands/fingers or feet/toes. This does not interfere of my activities of daily living.

I have moderate to severe difficulty in swallowing, breathing drinking or chewing food, or muscle spasms in my mouth/jaws, hands/fingers or feet/toes. This interferes with my activities of daily living.

E* I have severe difficulty in swallowing, breathing, drinking or chewing food, or muscle spasms in my mouth/jaws, hands/fingers or feet/toes. It completely prevents me from doing most activities of daily living.

* Please indicate by placing an X in the box or writing in the space provided which activity or activities have been interfered with as a result of therapy. My ability to: Button clothes

Zippers

Fasten buckles

Write

Sew

Use a knife

Put in or remove contact lenses

Sleep

Walk

Work

Use a fork

Dial or use telephone

Climb stairs

Put on jewelry

Tie shoes

Use a spoon

Operate a remote control

Type on a keyboard

Knit

Drive

Swallowing

Use other eating utensils

Eating/chewing

Drinking liquids

Shortness of breath

Open doors

Other eating utensils, etc

Work or perform activities of importance to me, specify: _______________________________________

Figure 2 Patient Neurotoxicity Questionnaire: oxaliplatin.

ity of these patient symptoms relative to obtaining this information directly from the patients themselves through a patient questionnaire.87-92,179,180 The PNQ has been recently validated in a 300-patient

study conducted in Japan involving patients receiving adjuvant paclitaxel or docetaxel, with or without doxorubicin or cyclophosphamide therapy.93 This large prospective, controlled study in a defined patient population

F.H. Hausheer et al receiving neurotoxic and non-neurotoxic treatment demonstrated that the PNQ is a reliable, sensitive, and responsive instrument in the diagnosis and grading of CIPN with greater sensitivity than the FACT/GOG-Ntx and NCICTC. The level of reporting compliance by patients and physicians was very high (⬎95%) in this study. A substantially lower incidence of more severe forms of CIPN was reported by physicians based on the NCI-CTC and the administration of the patient questionnaire to the physician independently of the patient. This finding strongly supports the fact that CIPN appears to be substantially under-reported by physicians. The results of this PNQ validation study and the comparative performance of the PNQ, in quantifying the incidence and severity of docetaxel and paclitaxel CIPN in patients, and other important findings will be reported in detail elsewhere. The FDA has accepted and supported the use of the PNQ as the primary end point to assess the incidence and severity of CIPN in phase II and phase III trials in the United States, and this instrument is currently in use by some of the US cooperative groups. The PNQ has also been used in a total of four ongoing or completed registration trials of an investigational neuroprotective agent as a primary end point assessment. European and Japanese health regulatory agencies have also accepted the PNQ, and it is being used as the primary neuroprotection end point assessment in phase III clinical trials in other countries. The results achieved thus far indicate this instrument will reliably capture patient’s symptoms and the severity of functional impairment in a format that is appropriate for a neuroprotection end point. In addition, this instrument, a convenient and time-saving questionnaire, may assist the treating physician in the management of patients with CIPN. When used as a primary end point in clinical trials, the PNQ sensory score (item 1) is given the greatest weight in the analysis, because neurosensory symptoms are the diagnostic hallmark for CIPN and the primary symptoms linked to functional impairment in patients with CIPN.

Prevention and Treatment of Chemotherapy-Induced Peripheral Neurotoxicity The development of CIPN may interfere with activities of daily living, causing patients to have difficulty dressing, eating, and walking and leading to a decrease in physical independence. The costs associated with the development of CIPN have been studied on a limited basis. It has been reported that an event of CIPN results in lost work and costs for home care that amount to approximately $4,908 per event.181,182 Accordingly, the prevention of these common and serious complications of chemotherapy will also have substantial economic impact. Several agents have been evaluated for potential use as chemoprotective agents, but none have proven effective for the prevention of CIPN.183 There are several medically im-

39 portant properties of an ideal neuroprotective agent that should be considered during the evaluation of potential neuroprotective or less neurotoxic drug candidates including: (1) the ability to prevent or mitigate CIPN associated with chemotherapy; (2) absence of interference with the antitumor activity of cytotoxic agents (prevention of neuropathy while also decreasing chemotherapy efficacy is not clinically acceptable); and (3) a lack of significant additional toxicity associated with administration of the neuroprotective agent, thereby avoiding the substitution of CIPN prevention by another adverse effect.184

Nerve Growth Factor and Analogs One of the agents undergoing investigation for possible antineuropathy activity is nerve growth factor (NGF). It has been postulated that NGF can effectively prevent cisplatininduced neuropathy. Hol and Bär19 found that NGF is effective in protecting a DRG model from cisplatin-induced toxicity. Others have reported that NGF reverses the down regulation of several putative neurotransmitters in sensory neurons, including substance P and calcitonin gene-related peptide (CGRP), and that NGF can prevent the up regulation of vasoactive intestinal polypeptide (VIP), galanin, and neuropeptide Y (NPY). Gill found that NGF prevents cisplatininduced apoptosis in PC12 cells and DRG neurons, but not colon cancer cells. When PC12 cells are exposed to NGF, it halts cell division and causes differentiation into a neuronlike cell. PC12 expresses high (gp140TrkA) and low (p75) affinity NGF receptors, but it does not express other neurotrophic receptors. Gill’s findings suggest that NGF may prevent cisplatin-induced CIPN through the gp140TrkA receptor and initiation of downstream signal cascade elements involved in cell survival or differentiation. In cancer cells without a high affinity for NGF receptors, chemotherapy-induced apoptosis was not prevented. In contrast to earlier reports, Windebank reported that excess NGF had no effect on cisplatin inhibition of neurite outgrowth in embryonic DRG cultures.185-187 A randomized prospective study was conducted to evaluate the potential safety and efficacy of subcutaneously administered rhNGF in 1,019 patients with diabetic neuropathy.188 Patients were randomized to receive either recombinant human (rh)NGF or placebo three times weekly for 12 months. Changes from baseline neuropathy score were assessed and compared between the two treatment groups via quantitative sensory tests, NCV studies, and the Neuropathy Impairment Score (NIS) and Neuropathy Symptoms and Change (NSC), as well as a monofilament test. The study demonstrated no significant benefit of rhNGF compared to placebo in terms of the primary and secondary efficacy end points. The ACTH analog, Org 2766,189-192 which is in the same family as NGF, is one of the neurotrophic factors studied for prevention of CIPN. NGF and Org 2766 are substances that promote survival of neurons exposed to toxic chemotherapy. Org 2766 has been observed to be effective against cisplatin-

Chemotherapy-induced peripheral neuropathy

40 induced neuropathy in animal experiments and in cases of impaired sensory nerve conduction velocity. van der Hoop’s study reported that Org 2766 prevents cisplatin-induced neuropathy. However, Roberts et al reported that van der Hoop’s study had important limitations, noting that the latter’s study treatment was not continued long enough to detect CIPN development. Roberts conducted long-term follow-up on 18 of the 55 women reported in the van der Hoop et al study. Hovestadt found cisplatin-induced CIPN actually to worsen for up to 4 months after treatment had been discontinued. Roberts’ study confirmed this post-treatment finding; this study demonstrated a lack of benefit from Org 2766. The latter study found that cisplatin-related CIPN continued to worsen for a minimum of 3 months after the drug therapy was completed.22,37,173,184,187,188,193

Amifostine Amifostine was approved by the FDA in 1996 for use in reducing the cumulative renal toxicity associated with repeated administration of cisplatin-based therapies in patients with advanced ovarian or NSCLC. Amifostine has also been evaluated for potential use in the prevention of other important chemotherapy-induced toxicities including cisplatinand paclitaxel-induced peripheral neuropathy. Inconsistent findings regarding the potential efficacy of amifostine for prevention of CIPN have been reported in studies to date. Variability in the interpretation of clinical results with amifostine led to the development of a clinical practice guideline issued by the American Society of Clinical Oncology describing the recommended uses for this agent in patients not participating in clinical trials.194 In general, the studies have demonstrated no benefit of amifostine in the prevention of CIPN. Promising results were reported from early phase I and II studies, but no multicenter, randomized, placebo-controlled phase III studies have been conducted to confirm these results. Gelmon et al conducted a multicenter, randomized, open label study in 40 patients with metastatic breast cancer to evaluate the potential efficacy of amifostine in preventing or reducing peripheral neuropathy induced by high-dose paclitaxel (250 mg/m2) given every 3 weeks.195 All patients were administered questionnaires and were evaluated using a standard neurologic evaluation as well as objective assessments (power and vibration sense) to assess changes from baseline in symptoms of CIPN. Generally, no differences in reported toxicity were observed between the two treatment arms, except nausea was reported more often in the amifostine arm (P ⫽ .031). Neuropathy was reported in 100% of patients in both treatment arms. There were no differences observed in any of the neuropathy assessments in patients treated with amifostine, demonstrating that in this study, amifostine provided no benefit in preventing or reducing paclitaxel-induced peripheral neurotoxicity. A phase II study in patients with various gynecologic cancers was conducted to evaluate the potential efficacy of amifostine in preventing or mitigating clinically important neurotoxicity associated with the administration of paclitaxel (175 mg/m2) and cisplatin (75 mg/m2) given every 3 weeks.

Patients had a baseline neurologic evaluation as well as subsequent evaluations consisting of the NCI-CTC, FACT/GOGNtx neurotoxicity questionnaire, and VPT testing to assess symptoms and limitations associated with the development of peripheral neuropathy. A total of 27 patients were assessable for response and toxicity. The study was closed early because the prospectively defined threshold for neuropathy events was exceeded when 4 patients developed NCI-CTC grade 2 to 4 CIPN. The level of observed efficacy of amifostine in this trial was insufficient to warrant a Phase III study. It was also observed that VPT measurements were less sensitive for detecting the development of CIPN relative to the patient questionnaire.196 Patients with advanced breast cancer receiving high-dose paclitaxel (725 mg/m2 over 24 hours) in combination with doxorubicin (165 mg/m2 over 96 hours) and cyclophosphamide (100 mg/kg over 2 hours) were studied on two autologous peripheral blood stem cell transplant protocols, one with and the other without amifostine (740 mg/m2) administered over 10 minutes before and 12 hours after initiation of the paclitaxel infusion. Patients were evaluated pretreatment and 20 to 40 days later with neurologic evaluations, a composite peripheral neuropathy score, peroneal and sural nerve conduction studies, and QST. All patients developed neurosensory impairment, and there was no significant benefit of amifostine in preventing or mitigating CIPN induced by high-dose paclitaxel in this study.15 A single-center, randomized placebo-controlled study was conducted by Leong et al to evaluate the use of amifostine in the reduction of CIPN. A total of 60 patients with advanced NSCLC were randomized to receive either amifostine or placebo in addition to paclitaxel 175 mg/m2 and carboplatin AUC ⫻6, followed by thoracic radiotherapy with concurrent weekly paclitaxel 60 mg/m2. Patients were evaluated clinically by a physician and by objective assessment, including NCV measurement pre- and post-treatment. This study also found no significant reduction in paclitaxel- or carboplatininduced CIPN by the administration of amifostine.197 Another study was conducted in 44 patients with metastatic breast cancer treated with cisplatin (120 mg/m2 over 30 minutes) preceded by the administration of amifostine (910 mg/m2 over 15 minutes); all patients received saline hydration and mannitol diuresis. Patients were treated every 3 weeks with this regimen. Neurologic toxicity was observed in 52% of patients, and there was also no evidence of amifostine neuroprotective efficacy in this study.198 Makino et al also conducted a taxane-related study and found that amifostine did not have any impact in preventing or mitigating docetaxel-induced CIPN.199

Glutathione The administration of reduced glutathione (GSH) in sulfhydryl or ester forms has been reported to be beneficial in the prevention of platinum-induced CIPN in experimental models and limited clinical trials.200-218 Although there are reports of randomized double-blind, placebo-controlled trials of GSH for the prevention of cisplatin- and oxaliplatin-induced

F.H. Hausheer et al CIPN, these studies were underpowered for the reliable assessment of safety and efficacy end points.201 A single randomized, double-blind, placebo-controlled trial investigated the neuroprotective effects of intravenously administered GSH for the prevention of oxaliplatin-induced CIPN. After the eighth cycle of a regimen containing oxaliplatin at 100 mg/m2/cycle, nine of 21 assessable patients in the GSH treatment group developed CIPN versus 15 of 19 in the placebo arm, suggesting neuroprotective efficacy of GSH. One of the limitations of this study was the use of the NCICTC to assess CIPN, which (as discussed previously) may have large inter-examiner variability in the assessment of CIPN. The objective tumor response rates and progressionfree survival rates were similar in the two treatment groups. However, the study included only a small sample size (n ⫽ 52) and therefore was substantially underpowered for this end point, as well as for the assessment of potential tumor protection by GSH. GSH is not an approved medicament in the US or in most European countries, which limits its utilization in current practice.

Glutamine and Glutamate The administration of glutamine to prevent CIPN has been reported in nonclinical studies and in a study of patients treated with high-dose paclitaxel.219,220 In a nonrandomized, open-label study, Vahdat et al reported the results of glutamate administration 10 g orally three times daily for 4 days, which was initiated 24 hours after administration of paclitaxel 825 mg/m2 over 24 hours.219,220 The first cohort of patients (n ⫽ 33) did not receive glutamate, while the second cohort (n ⫽ 12) received glutamate. The investigators reported significant reductions in the patient reported symptoms of CIPN, but there was no significant difference in the reduction in numbness and paresthesias of the fingers and toes, nor in the neurophysiologic evaluations of these patients. This study has important limitations for interpretation due to the nonrandomized, unblinded design, the absence of placebo control, and the small sample size, which is substantially underpowered for the assessment of CIPN. The safety and efficacy of glutamate administration for the prevention and mitigation of CIPN needs to be tested using a doubleblind, placebo-controlled method and should be adequately powered to assess neuroprotection, through the use of a patient-based instrument, and tumor protection.221 The administration of glutamate, which is considered to be an excitotoxic neurotransmitter in the CNS, has been studied for the prevention of paclitaxel-, cisplatin-, and vincristineinduced CIPN.219,220 Early investigation of glutamate in rat models demonstrated evidence of neuroprotective efficacy without tumor protection.222,223 The mechanisms of action of glutamate for neuroprotection are not known.

Gabapentin Gabapentin has been evaluated in phase III trials for use in the management of pain and other symptoms associated with CIPN. A randomized, double-blind, placebo-controlled phase III trial of gabapentin was conducted in 115 patients

41 with CIPN to evaluate its use for improvement in pain intensity and sensory neuropathy.224 A total of 57 patients were randomized to receive gabapentin for 6 weeks followed by a cross-over to placebo for 6 weeks. The co-primary end points used for the study were the average daily pain intensity score (0 to 10 where 0 ⫽ no pain and 10 ⫽ worst pain) and ECOG grading scale for sensory neuropathy (grade 0 to grade 3). The results of this study demonstrated that gabapentin did not significantly improve either pain intensity or sensory neuropathy. A nonrandomized phase II study was conducted by Mitchell et al to evaluate the use of gabapentin as a neuroprotectant against oxaliplatin-induced peripheral neurotoxicity in patients with advanced colorectal cancer.225 A total of 81 patients were enrolled (40 patients on FOLFOX alone and 41 patients on FOLFOX in combination with gabapentin). Objective tumor responses were observed in 48% of patients. No statistically significant differences were noted between treatment arms in the incidence and severity of neurotoxicity (FOLFOX: grade 2 or 3, 51%; FOLFOX ⫹ gabapentin: grade 2 or 3, 56%). Based on these findings, it appears that gabapentin does not improve or reduce the incidence or severity of oxaliplatin-induced peripheral neuropathy.

Oxaliplatin/Carbamazepine Patients with acute sensory symptoms from oxaliplatin display little or no axonal degeneration, suggesting a specific effect of this agent on sensory neurons and motor neurons that is not observed with other agents. The acute form of oxaliplatin-induced CIPN has features that are reported to be similar to a “channelopathy,” whereby there is dysfunction of the ion channels in the neurosensory or neuromuscular cell membranes. Recent data indicate that oxaliplatin may act on the voltage-gated sodium channel to increase the excitability of sensory neurons, an action that is blocked by carbamazepine.226 This hypothesis appears to be either incomplete or inconsistent due to the apparent lack of clinical efficacy of carbamazepine administration to prevent or treat acute oxaliplatin-induced CIPN and suggests other mechanism(s) may be involved. Incubation of rat DRG neurons with oxaliplatin increased the amplitude and duration of compound action potentials and lengthened the refractory period of peripheral nerves, suggesting an interaction with voltage-gated sodium channels. Carbamazepine was observed to antagonize the effect of oxaliplatin, and these investigators suggested that carbamazepine might be used to reduce the neurotoxic side effects of oxaliplatin based on this observation. Conflicting results have been reported with the use of carbamazepine for the management of oxaliplatin neuropathy. In a study conducted by Wilson et al, patients received 130 mg/m2 oxaliplatin given over 2 hours in combination with capecitabine oral therapy.227 Patients were evaluated for neurotoxicity symptoms using a neurologic examination that included EMG and nerve conduction studies. Carbamazepine was administered to 12 patients. Based on the clinical and electromyographic findings, no apparent benefit was notable

42 in terms of prevention or mitigation of oxaliplatin-induced neurotoxicity. Another study conducted by Lersch et al demonstrated promising results of carbamazepine in patients with colorectal cancer who received oxaliplatin 85 mg/m2 in combination with fluorouracil/leucovorin weekly for 6 weeks.228 No patient experienced moderate to severe neuropathy after cumulative doses were achieved that are known to cause grade 2 to 4 neuropathy. Randomized controlled studies with adequate statistical power are needed to fully evaluate the safety and efficacy of carbamazepine for the prevention of oxaliplatin CIPN.

Calcium/Magnesium (Ca/Mg) Infusions for Acute Oxaliplatin Neuropathy Ca/Mg infusions have been studied for the management of the acute form of oxaliplatin-induced peripheral neuropathy. An open label, non-randomized study of 101 patients with advanced colorectal cancer tested the efficacy of Ca/Mg infusions involving three different oxaliplatin-based protocols.120 A total of 63 patients in this study received calcium gluconate 1 g and magnesium sulfate 1 g infusions before and after oxaliplatin administration, and the remaining 38 patients did not receive these infusions. The results indicated that the Ca/Mg-treated patients achieved a higher cumulative oxaliplatin dose (910 mg/m2 v 650 mg/m2), experienced fewer treatment discontinuations for neurotoxicity (5% v 56%), had a lower incidence of oxaliplatin-induced CIPN of any grade (27% v 75%), a lower incidence of laryngopharyngeal dysesthesia (1.6% v 26%), less grade 3 neurotoxicity (5% v 26%), and a greater likelihood of treatment durations greater than 9 months (15% v 9%) compared to the untreated patient population. The overall antitumor efficacy of the regimens did not appear to be adversely affected, and patients were able to stay on therapy for a longer period of time, thus potentially deriving prolonged benefit from oxaliplatinbased therapy. Larger randomized, controlled clinical trials of Ca/Mg infusions for the prevention and mitigation of oxaliplatin-CIPN, particularly the acute form, are currently in progress.

Tavocept (disodium 2,2=-dithiobisethane sulfonate, BNP7787, dimesna) BNP7787 is an investigational new drug that was designed to prevent and mitigate common and clinically important toxicities associated with taxane- and platinum-type chemotherapeutic agents including CIPN. The mechanism(s) of action, safety, effectiveness, and potential for tumor protection of this agent have been evaluated extensively using in vitro and in vivo models. The investigational agent appears to be chemically and mechanistically distinct from other sulfur-containing drugs that have been studied for the purpose of preventing or reducing toxicities associated with platinum type chemotherapy.229

Chemotherapy-induced peripheral neuropathy Results from preclinical studies conducted in rat models show that under the applied experimental conditions, this agent exerts a protective effect against the neurotoxicity induced by multiple administrations of cisplatin or paclitaxel.230,231 Studies conducted in human tumor xenografts demonstrated that the agent does not interfere with the efficacy of chemotherapy.232 Three phase I studies were conducted with BNP7787 to identify the most appropriate dose for future studies as well as to characterize the safety and potential efficacy of the investigational agent for use in the prevention or mitigation of CIPN.233-235 The treatment regimen employed in these phase I studies included paclitaxel and/or cisplatin (paclitaxel 175 mg/m2 ⫾ cisplatin 75 mg/m2 every 3 weeks). Results from these clinical studies demonstrated that the number of patients experiencing neurotoxicity appeared less than expected based on findings from other studies reported in the literature using similar treatment regimens. No treatment-related grade 3 or 4 CIPN was reported in any patient enrolled on the phase I studies with BNP7787, including patients treated with as many as nine cycles of chemotherapy. Major objective tumor responses were observed, but these were all uncontrolled study designs. These initial results appear promising, but confirmation from adequately powered multicenter, randomized placebo-controlled trials are required to support the fact that the agent is safe and efficacious for the prevention of CIPN. This investigational agent is currently undergoing clinical development worldwide for the prevention and mitigation of clinically important CIPN associated with taxane and platinum agents in several randomized, multicenter, double-blind, placebo-controlled phase III trials in patients with metastatic breast cancer and advanced NSCLC.

Design and Endpoint Considerations for Clinical Trials to Establish Safety and Efficacy of Neuroprotective Agents: Neuroprotection or Reduced Neurotoxicity There are several important considerations for the design and conduct of prospective clinical trials to evaluate the safety and efficacy of neuroprotective agents as well as for chemotherapeutic agents in the same class (eg, a new formulation of the same active ingredient) that are purported to have a lower incidence or prevalence of clinically important CIPN. A fundamentally important trial design objective is to demonstrate convincing evidence of neuroprotective efficacy or attenuated neurotoxicity of a treatment regimen. This design objective requires well-defined control and treatment groups that are representative of the standard of care for treatment. The primary endpoint for the assessment of neuroprotective efficacy must be medically justified and clearly defined. We have previously discussed the important limitations in employing physician-based and objective QST/EPM assessments that appear to be less reliable in contrast to patient-based assessments in assessing primary neuroprotective efficacy. The end point demonstration of neuroprotective ef-

F.H. Hausheer et al ficacy must involve the analysis of prevention or mitigation of medically important CIPN as demonstrated by a significant reduction in the proportion of patients in the treatment group who experience CIPN that interferes with defined activities of daily living as compared to the control population. A multicenter, randomized, double-blind, placebo-controlled trial with two prospectively defined primary end points (neuroprotection and tumor protection assessment) is the most straightforward approach that can provide the requisite level of evidence of safety and efficacy of the intervention or neurotoxicity attenuation. The primary neuroprotection or neurotoxicity reduction endpoint should rely on a patient-based assessment since CIPN is subjective and requires direct assessment of interference with defined activities of daily living. As we have described herein, it is well documented that physician-based instruments and objective assessments such as QST and EPM, consistently under-assess the incidence and severity of medically important CIPN. The primary neuroprotective end point should compare the overall incidence of medically important CIPN in the treatment group and control group, unless there is reason to believe that the efficacy of the neuroprotective agent is subject to tachyphylaxis or will delay, not prevent, the onset of medically important CIPN. If a neuroprotective agent or neurotoxic therapy is believed to act by delaying the onset of CIPN, rather than preventing its development, the primary neuroprotection efficacy end point should involve a defined time to onset of a prespecified medically important level of CIPN that is assessed by a patientbased instrument. The statistical methodology for the analysis of the neuroprotection end point should employ a superiority design and analysis based on a two-sided test and significance level of .05, and the sample size of the population should have adequate statistical power to detect clinically important differences in the incidence of CIPN. Secondary end points to assess potential neuroprotection may include supportive measurements such as physician-based instruments, quality of life questionnaires, and time to onset of neuropathy, duration of neuropathy, and the assessment of the incidence of treatment delays or discontinuations due to CIPN as an efficacy end point. The primary tumor protection endpoint should rely on the use of objective tumor response rate using standardized methods (eg, Response Evaluation Criteria in Solid Tumors [RECIST] criteria), because this measurement is a direct patient outcome that is solely due to the effect of chemotherapy (eg, objective tumor responses do not occur in the absence of chemotherapy). Therefore, potential interference with this objective outcome can be readily quantified at an earlier stage relative to other possible tumor protection end points. In some patient populations objective tumor response rates are not as reliable, and in these patient groups the use of progression-free survival or time-to-progression should be used as a primary tumor protection end point, relying on standardized methods and time points for disease assessment. Objective tumor response rate is the most specific indicator of tumor protection compared with time-to-event measurements such as time-to-progression or survival, and as an ethical and

43 safety consideration, this end point would also provide earlier evidence of the presence or absence of tumor protection in the course of treatment in a clinical trial involving a chemoprotective agent. The statistical methodology for assessment of the tumor protection end point should be comprised of a non-inferiority design and analysis with a two-sided significance level of .05 and have adequate statistical power to detect prespecified and clinically important reductions in antitumor activity. It is important to recognize that non-inferiority designs require pre-specification of the non-inferiority margin and that to meet the criteria of non-inferiority where the lower bound of the 95% confidence interval for response or time-to-event in the active treatment group must lie wholly to the right of the margin. The defined margin is determined by the observed objective tumor response or time-to-event in the control population. Non-inferiority designs are sometimes misunderstood to mean that it is acceptable for the average value of the treatment group to lie anywhere to the right of the margin, neglecting the value of the 95% confidence interval. For registration trials, the non-inferiority tumor protection assessment margin to be used in a trial should be discussed and settled with regulatory agencies prior to the initiation of the clinical trial, because the sample size increases or decreases as a function of this parameter.

Closing Remarks Chemotherapy-induced peripheral neuropathy is a common and serious problem that can present diagnostic and therapeutic challenges and, most important, can materially interfere with the patient’s quality of life and the administration of their cancer treatment. Although there is no currently approved safe and effective treatment for CIPN, current and future efforts to develop and implement improved diagnostic and therapeutic measures hold the promise of benefit for the many patients who experience this important clinical problem.20,25,27,36,114,119,124-130,132

Acknowledgment The authors would like to thank Drs Michael C. Perry, Raymond B. Weiss, Vojo Vukovic, and Francis Ruvuna for their review and valuable discussions, Stephen Koehler for assistance in compilation of research, and Erika Ramirez in technical assistance in editing and formatting of the manuscript.

References 1. Docetaxel. American Hospital Formulary Service (AHFS). Bethesda, MD, American Society of Health-System Pharmacists (ASHP), 2002, p 966 2. Cisplatin. American Hospital Formulary Service (AHFS). Bethesda, MD, American Society of Health-System Pharmacists (ASHP), 2002, p 927 3. Hilkens PH, van den Bent MJ: Chemotherapy-induced peripheral neuropathy. J Peripher Nerv Syst 2:350-361, 1997 4. Rowinsky EK, Chaudhry V, Cornblath DR, et al: Neurotoxicity of Taxol. JNCI Monogr 15:107-115, 1993 5. Lipton RB, Apfel SC, Dutcher JP, et al: Taxol produces a predominantly sensory neuropathy. Neurology 39:368-373, 1989 6. Weiss RB: Miscellaneous toxicities, in DeVita VT Jr, Hellman S, Rosen-

Chemotherapy-induced peripheral neuropathy

44

7.

8.

9.

10.

11.

12.

13.

14.

15.

16. 17. 18. 19. 20.

21. 22.

23.

24.

25.

26.

27.

28.

29.

30.

berg SA (eds): Cancer: Principles and Practice of Oncology, vol 2 (ed 7). Philadelphia, PA, Lippincott, 2005, pp 2603-2607 Paclitaxel. American Hospital Formulary Service (AHFS). Bethesda, MD, American Society of Health-System Pharmacists (ASHP), 2002, p 1096 Oxaliplatin. American Hospital Formulary Service (AHFS). Bethesda, MD, American Society of Health-System Pharmacists (ASHP), 2002, p 1093 McKeage MJ, Hsu T, Screnci D, et al: Nucleolar damage correlates with neurotoxicity induced by different platinum drugs. Br J Cancer 85: 1219-1225, 2001 Thompson SW, Davis LE, Kornfeld M, et al: Cisplatin neuropathy. Clinical, electrophysiologic, morphologic, and toxicologic studies. Cancer 54:1269-1275, 1984 Gregg RW, Molepo JM, Monpetit VJ, et al: Cisplatin neurotoxicity: the relationship between dosage, time, and platinum concentration in neurologic tissues, and morphologic evidence of toxicity. J Clin Oncol 10:795-803, 1992 Screnci D, Er HM, Hambley TW, et al: Stereoselective peripheral sensory neurotoxicity of diaminocyclohexane platinum enantiomers related to ormaplatin and oxaliplatin. Br J Cancer 76:502-510, 1997 Verdu E, Vilches JJ, Rodriguez FJ, et al: Physiological and immunohistochemical characterization of cisplatin-induced neuropathy in mice. Muscle Nerve 22:329-340, 1999 Gill JS, Windebank AJ: Cisplatin-induced apoptosis in rat root ganglion neurons is associated with attempted entry into the cell cycle. J Clin Invest 101:2842-2850, 1998 Openshaw H., Beamon K, Synold TW, et al: Neurophysiological study of peripheral neuropathy after high-dose paclitaxel. Clin Cancer Res 10:461-467, 2004 New PZ, Jackson CE, Rinaldi D, et al: Peripheral neuropathy secondary to docetaxel (Taxotere). Neurology 46:108-111, 1996 Quasthoff S, Hartung HP: Chemotherapy-induced peripheral neuropathy. J Neurol 249:9-17, 2002 Peltier AC, Russell JW: Recent advances in drug-induced neuropathies. Curr Opin Neurol 15:633-638, 2002 Hol EM, Bär PR: Cisplatin neuropathy. Neurology 45:596, 1995 Müller, LJ: Morphological and electrophysiological study of the effects of cisplain and ORG 2766 on rat spinal ganglion neurons. Cancer Res 50:2437-2442, 1990 Sahenk Z, Barohn R, New P, et al: Taxol neuropathy. Arch Neurol 51:726-729, 1994 Barajon I, Bersani M, Quartu M, et al: Neuropeptides and morphological changes in cisplatin-induced dorsal root ganglion neuronopathy. Exp Neurol 138:93-104, 1996 Cavaletti G, Cavalletti E, Oggioni N, et al: Distribution of paclitaxel within the nervous system of the rat after repeated intravenous administration. Neurotoxicology 21:389-393, 2000 Cavaletti G, Tredici G, Braga M, et al: Experimental peripheral neuropathy induced in adult rats by repeated intraperitoneal administration of taxol. Exp Neurol 133:64-72, 1995 Cavaletti G, Tredici G, Marmiroli P, et al: Morphometric study of the sensory neuron and peripheral nerve changes induced by chronic cisplatin (DDP) administration in rats. Acta Neuropathol (Berl) 84: 364-371, 1992 Holmes J, Stanko J, Varchenko M, et al: Comparative neurotoxicity of oxaliplatin, cisplatin, and ormaplatin in a Wistar rat model. Toxicol Sci 46:342-351, 1998 Kociba RJ, Sleight SD: Acute toxicologic and pathologic effects of cis-diamminedichloroplatinum (NSC-119875) in the male rat. Cancer Chemother Rep 55:1-8, 1971 Masurovsky EB, Peterson ER, Crain SM, et al: Morphological alterations in dorsal root ganglion neurons and supporting cells of organotypic mouse spinal cord-ganglion cultures exposed to taxol. Neuroscience 10:491-509, 1983 Screnci D, McKeage MJ, Galettis P, et al: Relationships between hydrophobicity, reactivity, accumulation and peripheral nerve toxicity of a series of platinum drugs. Br J Cancer 82:966-972, 2000 Krarup-Hansen A, Fugleholm K, Helweg-Larsen S, et al: Examination

31.

32. 33.

34. 35. 36.

37.

38.

39. 40. 41.

42.

43. 44. 45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

of distal involvement in cisplatin-induced neuropathy in man. An electrophysiological and histological study with particular reference to touch receptor function. Brain 116:1017-1041, 1993 Tulub AA, Stefanov VE: Cisplatin stops tubulin assembly into microtubules. A new insight into the mechanism of antitumor activity of platinum complexes. Int J Biol Macromol 28:191-198, 2001 Hollenbeck PJ: Kinesin delivers: Identifying receptors for motor proteins. J Cell Biol 152:F25-F28, 2001 Heins S, Song YH, Wille H, et al: Effect of MAP2, MAP2c, and tau on kinesin-dependent microtubule motility. J Cell Sci Suppl 14:121-124, 1991 Pfister KK: Cytoplasmic dynein and microtubule transport in the axon: The action connection. Mol Neurobiol 20:81-91, 1999 Fazio R, Quattrini A, Bolognesi A, et al: Docetaxel neuropathy: A distal axonopathy. Acta Neuropathol (Berl). 98:651-653, 1999 Xu Y, Sladky JT, Brown MJ: Dose-dependent expression of neuronopathy after experimental pyridoxine intoxication. Neurology 39: 1077-1083, 1989 Peterson ER, Crain SM: Nerve growth factor attenuates neurotoxic effects of Taxol on spinal cord-ganglion explants from fetal mice. Science 217:377-379, 1982 Hilkens PH, Verweij J, Vecht CJ, et al: Clinical characteristics of severe peripheral neuropathy induced by docetaxel (Taxotere). Ann Oncol 8:187-190, 1997 Argov Z, Mastaglia FL: Drug-induced peripheral neuropathies. Br Med J 1:663-666, 1979 Cersosimo RJ: Cisplatin neurotoxicity. Cancer Treat Rev 16:195-211, 1989 Screnci D, McKeage MJ: Platinum neurotoxicity: Clinical profiles, experimental models and neuroprotective approaches. J Inorg Biochem 77:105-110, 1999 Chaudhry V, Rowinsky EK, Sartorius SE, et al: Peripheral neuropathy from taxol and cisplatin combination chemotherapy: clinical and electrophysiological studies. Ann Neurol 35:304-311, 1994 Corbo M, Balmaceda C: Peripheral neuropathy in cancer patients. Cancer Invest 19:369-382, 2001 Cavaletti G, Fabbrica D, Minoia C, et al: Carboplatin toxic effects on the peripheral nervous system of the rat. Ann Oncol 9:443-447, 1998 Russell JW, Windebank AJ, McNiven MA, et al: Effect of cisplatin and ACTH4-9 on neural transport in cisplatin induced neurotoxicity. Brain Res 676:258-267, 1995 Mandys V, Viale M, Vrana J, et al: Neurotoxic effect of cisplatin and the cisplatin-procaine complex DPR studied in organotypic cultures of chick embryonic dorsal root ganglia. Anticancer Drugs 9:659-663, 1998 Geldof AA: Nerve-growth-factor-dependent neurite outgrowth assay; a research model for chemotherapy-induced neuropathy. J Cancer Res Clin Oncol 121:657-660, 1995 Jirsova K, Mandys V. Differences in the inhibition of neuritic outgrowth in organotypic cultures of rat foetal dorsal root ganglia treated with cisplatin and carboplatin: A comparative study. Folia Histochem Cytobiol 35:215-219, 1997 Konings PN, Makkink WK, van Delft AM, et al: Reversal by NGF of cytostatic drug-induced reduction of neurite outgrowth in rat dorsal root ganglia in vitro. Brain Res 640:195-204, 1994 Konings PN, Philipsen RL, van den Broek JH, et al: Morphometric analysis of cisplatin-induced neurite outgrowth in N1E-115 neuroblastoma cells. Neurosci Lett 178:115-118, 1994 Peyrot V, Briand C, Crevat A, et al: Action of hydrolyzed cisplatin and some analogs on microtubule protein polymerization in vitro. Cancer Treat Rep 67:641-646, 1983 Peyrot V, Briand C, Momburg R, et al: In vitro mechanism study of microtubule assembly inhibition by cis-dichlorodiammine-platinum(II). Biochem Pharmacol 35:371-375, 1986 Goldstein LS, Yang Z: Microtubule-based transport systems in neurons: The roles of kinesins and dyneins. Annu Rev Neurosci 23:39-71, 2000 Tankanow RM: Docetaxel: A taxoid for the treatment of metastatic breast cancer. Am J Health Syst Pharm 55:1777-1791, 1998

F.H. Hausheer et al 55. Berger T, Malayeri R, Doppelbauer A, et al: Neurological monitoring of neurotoxicity induced by paclitaxel/cisplatin chemotherapy. Eur J Cancer 33:1393-1399, 1997 56. du Bois A, Schlaich M, Luck HJ, et al: Evaluation of neurotoxicity induced by paclitaxel second-line chemotherapy. Support Care Cancer 7:354-361, 1999 57. Chaudhry V, Chaudhry M, Crawford TO, et al: Toxic neuropathy in patients with pre-existing neuropathy. Neurology 60:337-340, 2003 58. Rowinsky EK, Cazenave LA, Donehower RC: Taxol: A novel investigational antimicrotubule agent. J Natl Cancer Inst 82:1247-1259, 1990 59. Katsumata N: Docetaxel: An alternative taxane in ovarian cancer. Br J Cancer 89:S9-S15, 2003 (suppl 3) 60. Mollman JE, Glover DJ, Hogan WM, et al: Cisplatin neuropathy. Risk factors, prognosis, and protection by WR-2721. Cancer 61:21922195, 1988 61. Anastasia P, Hay JW: Chemotherapy-induced neuropathy: Results of an oncology nurse survey. Proc Am Soc Clin Oncol 21:200b, 2002 (abstr 2618) 62. Cavaletti G, Bogliun G, Crespi V, et al: Neurotoxicity and ototoxicity of cisplatin plus paclitaxel in comparison to cisplatin plus cyclophosphamide in patients with epithelial ovarian cancer. J Clin Oncol 15: 199-206, 1997 63. Cavaletti G, Bogliun G, Marzorati L, et al: Peripheral neurotoxicity of Taxol in patients previously treated with cisplatin. Cancer 75:11411150, 1995 64. Forsyth PA, Balmaceda C, Peterson K, et al: Prospective study of paclitaxel-induced peripheral neuropathy with quantitative sensory testing. J Neurooncol 35:47-53, 1997 65. van Gerven JM, Moll JW, van den Bent MJ, et al: Paclitaxel (Taxol) induces cumulative mild neurotoxicity. Eur J Cancer 30A:1074-1077, 1994 66. Rowinsky EK, Eisenhauer EA, Chaudhry V, et al: Clinical toxicities encountered with paclitaxel (Taxol). Semin Oncol 20:1-15, 1993 (suppl 3) 67. Ostchega Y, Donohoe M, Fox N: High dose cisplatin-related peripheral neuropathy. Cancer Nurs 11:23-32, 1988 68. Schaumburg HH, Spencer PS: Toxic neuropathies. Neurology 29: 429-431, 1979 69. Paraplatin [package insert]. Princeton, NJ: Bristol-Meyers Squibb Company; 2004 70. du Bois A, Luck HJ, Meier W, et al: Carboplatin/paclitaxel versus cisplatin/paclitaxel as first-line chemotherapy in advanced ovarian cancer: An interim analysis of a randomized phase III trial of the Arbeitsgemeinschaft Gynakologische Onkologie Ovarian Cancer Study Group. Semin Oncol 24:S15-44-S15-52, 1997 (suppl 15) 71. Ozols RF, Bundy BN, Greer BE, et al: Phase III trial of carboplatin and paclitaxel compared with cisplatin and paclitaxel in patients with optimally resected stage III ovarian cancer: a Gynecologic Oncology Group study. J Clin Oncol 21:3194-3200, 2003 72. Cornblath DR, Chaudhry V, Carter K, et al: Total neuropathy score: Validation and reliability study. Neurology 53:1660-1664, 1999 73. Screnci D, McKeage MJ: Platinum neurotoxicity: Clinical profiles, experimental models and neuroprotective approaches. J Inorg Biochem 77:105-110, 1999 74. Petersen PM, Hansen SW: The course of long-term toxicity in patients treated with cisplatin-based chemotherapy for non-seminomatous germ-cell cancer. Ann Oncol 10:1475-1483, 1999 75. Eloxatin [package insert]. New York, NY, Sanofi-Synthelabo Inc, 2002 76. Thalomide [revised package insert]. Warren, NJ, Calgene Corp, 1998 77. Hausheer F, Kochat H, Zhao A, et al: BNP7787: A novel neuroprotective agent in taxane and platinum regimens does not interfere with chemotherapeutic effects. Proc Am Assoc Cancer Res 42:370, 2001 (abstr 1990) 78. Hausheer F, Kochat H, Reddy D, et al: BNP7787: A Novel chemoprotecting agent for platinum and taxane toxicity. Proc Am Assoc Cancer Res 41:769-770, 2000 (abstr 4890)

45 79. Platinol-AQ [package insert]. Princeton, NJ, Bristol-Meyers Squibb Co, 1999 80. Taxol [package insert]. Princeton, NJ, Bristol-Meyers Squibb Co, 2003 81. Taxotere [package insert]. Bridgewater, NY, Aventis Pharmaceuticals, Inc, 2005 82. Vincristine sulfate injection [package insert]. Irvine, CA, GensiaSicor Pharmaceuticals, 1999 83. Velcade [package insert]. Cambridge, MA, Millennium Pharmaceuticals, Inc, 2005 84. Wasserheit C, Frazein A, Oratz R, et al: Phase II trial of paclitaxel and cisplatin in women with advanced breast cancer: Active regimen with limiting neurotoxicity. J Clin Oncol 14:1993-1999, 1996 85. Rosell R, Gatzemeier U, Betticher DC, et al: Phase III randomised trial comparing paclitaxel/carboplatin with paclitaxel/cisplatin in patients with advanced non-small-cell lung cancer: A cooperative multinational trial. Ann Oncol 13:1539-1549, 2002 86. Connelly E, Markman M, Kennedy A, et al: Paclitaxel delivered as a 3-hr infusion with cisplatin in patients with gynecologic cancers: Unexpected incidence of neurotoxicity. Gynecol Oncol 62:166-168, 1996 87. Wul FL, Brites C, Quevedo S, et al: Difference in the evaluation of symptoms between patients and their doctors in an oncologic hospital. Proc Am Soc Clin Oncol 21:264b, 2002 (abstr 2874) 88. Hopwood P, Stephens RJ, Machin D: Approaches to the analysis of quality of life data: experiences gained from a Medical Research Council Lung Cancer Working Party Palliative Chemotherapy Trial. Qual Life Res 3:339-352, 1994 89. Slevin ML, Plant H, Lynch D, et al: Who should measure quality of life, the doctor or the patient? Br J Cancer 57:109-112, 1988 90. Varricchio CG, Bass E, Novotny P, et al: Common toxicity criteria versus symptom specific measures: How congruent are the data? Proc Am Soc Clin Oncol 21:270b, 2002 (abstr 2897) 91. Savage C, Pater J, Tu D, et al: He said/she said: How well do common toxicity criteria and quality of life agree on patient symptoms? Proc Am Soc Cln Oncol 2002 (abstr 1540) 92. Stephens RJ, Hopwood P, Girling DJ, et al: Randomized trials with quality of life end points: Are doctors’ ratings of patients’ physical symptoms interchangeable with patients’ self-ratings? Qual Life Res 6:225-236, 1997 93. Shimozuma K, Ohashi Y, Takeuchi A, et al: Validation of the patient neurotoxicity questionnaire (PNQ) during taxane chemotherapy in a phase III randomized trial of breast cancer: N-SAS BC 02. San Antonio Breast Cancer Symposium, San Antonio, TX, December 2004 (abstr 6037) 94. Seidman AD, Berry D, Cirrincione C, et al: CALGB 9840: Phase III study of weekly paclitaxel via 1-hour (h) infusion versus standard 3h infusion every third week in the treatment of metastatic breast cancer, with trastuzumab for HER2 positive MBC and randomized for T in HER2 normal MBC. Proc Am Soc Clin Oncol 22:6s, 2004 (abstr 512) 95. Sikov W, Akerley W, Strenger R, et al: Weekly high-dose paclitaxel demonstrates significant activity in advanced breast cancer. Proc Am Soc Clin Oncol 17:112a, 1998 (abstr 432) 96. Seidman AD, Hudis CA, Albanel J, et al: Dose-dense therapy with weekly 1-hour paclitaxel infusions in the treatment of metastatic breast cancer. J Clin Oncol 16:3353-3361, 1998 97. Perez EA, Voegel CL, Irwin DH, et al: Multicenter phase II trial of weekly paclitaxel in women with metastatic breast cancer. J Clin Oncol 19:4216-4223, 2001 98. Fountzilas G, Tsavdaridis D, Kalogera-Fountzila A, et al: Weekly paclitaxel as first-line chemotherapy and trastuzumab in patients with advanced breast cancer; a Hellenic Cooperative Oncology Group phase II study. Ann Oncol 12:1545-1551, 2001 99. Burstein HJ, Manola J, Younger J, et al: Docetaxel administered on a weekly basis for metastatic breast cancer. J Clin Oncol 18:1212-1219, 2000 100. Akerley W, Herndon JE, Egorin MJ, et al: CALGB 9731: Phase II trial of weekly paclitaxel for advanced non-small cell lung cancer (NSCLC). Proc Am Soc Clin Oncol 18:462a, 1999 (abstr 1783)

Chemotherapy-induced peripheral neuropathy

46 101. Akerley W, Herndon JE, Egorin MJ, et al: Weekly, high-dose paclitaxel in advanced lung cancer carcinoma: A phase II study with pharmacokinetics by the Cancer and Leukemia Group B (CALGB). Cancer 97:2480-2486, 2003 102. Akerley W, Glantz M, Choy H, et al: Phase I trial of weekly paclitaxel in advanced lung cancer. J Clin Oncol 16:153-158, 1998 103. Akerley W III: Paclitaxel in advanced non-small cell lung cancer: An alternative high-dose weekly schedule. Chest 117:152S-155S, 2000 (suppl 1) 104. Fidias P, Supko JG, Martins R, et al: A phase II study of weekly paclitaxel in elderly patients with advanced non-small cell lung cancer. Clin Cancer Res 7:3942-3949, 2001 105. Klaassen U, Wilke H, Strumberg D, et al: Phase I study with a weekly 1 h infusion of paclitaxel in heavily pretreated patients with metastatic breast and ovarian cancer. Eur J Cancer 32A:547-549, 1996 106. Fornier M, Seidman AD, Esteva FJ, et al: Weekly (W) herceptin (H) ⫹ 1 hour taxol (T): phase II study in HER2 overexpressing (H2⫹) and non-overexpressing (H2-) metastatic breast cancer (MBC). Proc Am Soc Clin Oncol 1999 (abstr 482) 107. Seidman AD, Fornier MN, Esteva FJ, et al: Weekly trastuzumab and paclitaxel therapy for metastatic breast cancer with analysis of efficacy by HER2 immunophenotype and gene amplification. J Clin Oncol 19:2587-2595, 2001 108. Breier S, Lebedinsky C, Pelayes L, et al: Phase I/II weekly paclitaxel (P) 80 mg/m2 in pretreated patients (pts) with breast (BC) and ovarian cancer. Proc Am Soc Clin Oncol 16:163a, 1997 (abstr 568) 109. Chang L, Hui L, Boros R, et al: Phase I study of weekly one-hour paclitaxel treatment in advanced malignant diseases. Proc Am Soc Clin Oncol 16:232a, 1997 (abstr 817) 110. Cisplatin, in Physician’s Desk Reference. Montvale, NJ, Thompson PDR, 2003, p 859 111. Becouarn Y, Ychou M, Ducreux M, et al: Phase II trial of oxaliplatin as first-line chemotherapy in metastatic colorectal cancer patients. Digestive Group of French Federation of Cancer Centers. J Clin Oncol 16:2739-2744, 1998 112. Grothey A. Oxaliplatin-safety profile: Neurotoxicity. Semin Oncol 30:5-13, 2003 (suppl 15) 113. Gilles-Amar V, Garcia ML, Sebille A, et al: Evolution of severe sensory neuropathy with oxaliplatin combined to the bimonthly 48h leucovorin (LV) and 5-fluorouracil (5FU) regimens (FOLFOX) in metastatic colorectal cancer. Proc Am Soc Clin Oncol 18, 1999 (abstr 944) 114. Maindrault-Goebel F, Louvet C, Andre T, et al: Oxaliplatin added to the simplified bimonthly leucovorin and 5-Fluorouracil regimen as second-line therapy for metastatic colorectal cancer (FOLFOX6). Eur J Cancer 35:1338-1342, 1999 115. Cassidy J, Misset JL: Oxaliplatin-related side effects: Characteristics and management. Semin Oncol 29:11-20, 2002 (suppl 15) 116. Armand JP, Boige V, Raymond E, et al: Oxaliplatin in colorectal cancer: An overview. Semin Oncol 27:96-104, 2000 (suppl 10) 117. Cvitkovic E, Bekradda M: Oxaliplatin: A new therapeutic option in colorectal cancer. Semin Oncol 26:647-662, 1999 118. de Gramont A, Figer M, Seymour M, et al: Leucovorin and fluorouracil with or without oxaliplatin as first-line treatment in advanced colorectal cancer. J Clin Oncol 18:2938-2947, 2000 119. Falcone A, Masi G, Allegrini G, et al: Biweekly chemotherapy with oxaliplatin, irinotecan, infusional fluorouracil, and leucovorin: A pilot study in patients with metastatic colorectal cancer. J Clin Oncol 20: 4006-4014, 2002 120. Gamelin E, Gamelin L, Bossi L, et al: Clinical aspects and molecular basis of oxaliplatin neurotoxicity: Current management and development of preventive measures. Semin Oncol 29:21-33, 2002 (suppl 15) 121. Giacchetti S, Perpoint B, Zidani R: Phase III multicenter randomized trial of oxaliplatin added to chronomodulated fluorouracil-leucovorin as first-line treatment of metastatic colorectal cancer. J Clin Oncol 18:136-147, 2000 122. Gilles-Amar V, Garcia ML, Sebille A: Evolution of severe sensory neuropathy with oxaliplatin combined to the bimonthly 48h leucovorn

123.

124.

125.

126.

127.

128.

129. 130. 131. 132.

133. 134. 135.

136.

137.

138. 139.

140.

141.

142.

143.

144.

145. 146.

and 5-fluorouracil regimens with mestatic colorectal cancer. Proc Am Soc Clin Oncol 18:246a, 1999 (abstr 944) Maindrault-Goebel F, de Gramont A, Louvet C, et al: High-dose intensity oxaliplatin added to the simplified bimonthly leucovorin and 5-fluorouracil regimen as second-line therapy for metastatic colorectal cancer (FOLFOX 7). Eur J Cancer 37:1000-1005, 2001 Maindrault-Goebel F, de Gramont A, Louvet C, et al. Bi-monthly oxaliplatin with leucovorin and 5-fluorouracil in pretreated metastatic colorectal cancer (FOLFOX6). Proc Am Soc Clin Oncol 17:273a, 1998 (abstr 1049) Maindrault-Goebel F, de Gramont A, Louvet C, et al: Evaluation of oxaliplatin dose intensity in bimonthly leucovorin and 48-hour 5-fluorouracil continuous infusion regimens (FOLFOX) in pretreated metastatic colorectal cancer. Oncology Multidisciplinary Research Group (GERCOR). Ann Oncol 11:1477-1483, 2000 Jin M, Chen Q, Cheng FQ, et al: Oxaliplatin (OXA) in combination with LV5FU2 in Chinese patients with advanced gastric cancer (AGC). Proc Am Soc Clin Oncol 21:140a, 2002 (abstr 558) Ravaioli A, Marangolo M, Pasquini E, et al: Bolus fluorouracil and leucovorin with oxaliplatin as first-line treatment in metastatic colorectal cancer. J Clin Oncol 20:2545-2550, 2002 Schmoll HJ:. The role of oxaliplatin in the treatment of advanced metastatic colorectal cancer: Prospects and future directions. Semin Oncol 29:34-39, 2002 (suppl 15) Topham C, Adam R: Oncosurgery: a new reality in metastatic colorectal carcinoma. Semin Oncol 29:3-10, 2002 (suppl 15) Oxaliplatin, in Physician’s Desk Reference. Montvale, NJ, Thompson PDR, 2003, p 2999 Grothey A, Schmoll HJ: New chemotherapy approaches in colorectal cancer. Curr Opin Oncol 13:275-286, 2001 Louvet C, Andre T, Tigaud JM, et al: Phase II study of oxaliplatin, fluorouracil, and folinic acid in locally advanced or metastatic gastric cancer patients. J Clin Oncol 20:4543-4548, 2002 Paclitaxel Drug Package Insert. Princeton, NJ, Bristol-Myers Squibb, March 2003 Rowinsky EK, Donehower RC: Paclitaxel (Taxol). N Engl J Med 332: 1004-1014, 1995 Sikov W, Akerley W, Strenger R, et al: Weekly high-dose paclitaxel demonstrates significant activity in advanced breast cancer. Proc Am Soc Clin Oncol 17:112, 1998 (abstr 432) Mielke S, Mross K, Gerds TA, et al: Comparative neurotoxicity of weekly non-break paclitaxel infusions over 1 versus 3 h. Anticancer Drugs 14:785-792, 2003 Cavaletti G, Bogliun G, Marzorati L, et al: Grading of chemotherapyinduced peripheral neurotoxicity using the Total Neuropathy Scale. Neurology 61:1297-1300, 2003 Abraxane [package insert]. Schaumburg, IL, American Pharmaceutical Partners, Inc, 2005 Skubitz KM, Verschraegen CF, Daud A, et al: A phase I study of paclitaxel poliglumex (PPX) in combination with cisplatin (CIS) for patients (pts) with advanced cancer. J Clin Oncol 23:157s, 2005 (suppl 1, abstr 2089) ten Tije AJ, Verweij J, Loos WJ, et al: Pharmacological effects of formulation vehicles: Implications for cancer chemotherapy. Clin Pharmacokinet 42:665-685, 2003 Henningsson A, Karlsson MO, Vigano L, et al: Mechanism-based pharmacokinetic model for paclitaxel. J Clin Oncol 19:4065-4073, 2001 Henningsson A, Sparreboom A, Sandstrom M, et al: Population pharmacokinetic modelling of unbound and total plasma concentrations of paclitaxel in cancer patients. Eur J Cancer 39:1105-1114, 2003 Sparreboom A, van Zuylen L, Brouwer E, et al: Cremophor EL-mediated alteration of paclitaxel distribution in human blood: Clinical pharmacokinetic implications. Cancer Res 59:1454-1457, 1999 Sparreboom A, Loos WJ, Verweij J, et al: Quantitation of Cremophor EL in human plasma samples using a colorimetric dye-binding microassay. Anal Biochem 255:171-175, 1998 MacNeil JS: ABI 007. Drugs Res Dev 5:155-159, 2004 Vasey PA: Survival and longer-term toxicity results of the SCOTROC

F.H. Hausheer et al

147.

148. 149.

150.

151. 152.

153.

154.

155.

156.

157. 158.

159.

160.

161. 162.

163.

164.

165.

166.

167. 168.

169.

study: Docetaxel-carboplatin (DC) vs. paclitaxel-carboplatin (PC) in epithelial ovarian cancer (EOC). Proc Am Soc Clin Oncol 21:202a, 2002 (abstr 804) Guastalla JP 3rd, Dieras V. The taxanes: Toxicity and quality of life considerations in advanced ovarian cancer. Br J Cancer 89:S16-S22, 2003 (suppl 3) Vasey PA: Role of docetaxel in the treatment of newly diagnosed advanced ovarian cancer. J Clin Oncol 21:136-144, 2003 (suppl) Schiller JH, Harrington D, Belani CP, et al: Comparison of four chemotherapy regimens for advanced non-small-cell lung cancer. N Engl J Med 346:92-98, 2002 Trobaugh-Lotrario AD, Smith AA, Odom LF: Vincristine neurotoxicity in the presence of hereditary neuropathy. Med Pediatr Oncol 40: 39-43, 2003 Vinorelbine Tartrate Package Insert. Irvine, CA, Gensia Sicor Pharmaceuticals, Inc, 2002 Tosi P, Zamagni E, Cellini C, et al: Neurological toxicity of long-term (⬎ 1 yr) thalidomide therapy in patients with multiple myeloma. Eur J Haematol 74:212-216, 2005 Offidani M, Corvatti L, Marconi M, et al: Common and rare sideeffects of low-dose thalidomide in multiple myeloma: focus on the dose-minimizing peripheral neuropathy. Eur J Haematol 72:403-409, 2004. Cytarabine. American Hospital Formulary Service (AHFS). Bethesda, MD, American Society of Health-System Pharmacists (ASHP), 2002, p 965 Osborne WL, Holyoake TL, McQuaker IG, et al: Fatal peripheral neuropathy following FLA chemotherapy. Clin Lab Haematol 26:295296, 2004 Etoposide. American Hospital Formulary Service (AHFS). Bethesda, MD, American Society of Health-System Pharmacists (ASHP), 2002, p 1001 Roferon. Drug Package Insert. Nutley, NJ, Hoffman-LaRoche, September 2003 Postma TJ, Heimans JJ, Muller MJ, et al: Pitfalls in grading severity of chemotherapy-induced peripheral neuropathy. Ann Oncol 9:739744, 1998 Guidance for the Industry Providing Evidence of Effectiveness for Human Drug and Biology Products. Washington, DC, FDA, US Department of Health and Human Services, 1998 Guidance for the Industry E9 Statistical Principles for Clinical Trials. Washington, DC, FDA, US Department of Health and Human Services, 1998 United States Code of Federal Regulations, Title 21, Part 314 Leidy NK, Revicki DA, Geneste B: Recommendations for evaluating the validity of quality of life claims for labeling and promotion. Value Health 2:113-127, 1999 Chassany O, Sagnier P, Marquis P, et al: Patient-reported outcomes: The example of health related quality of life a European guidance document for the improved integration of health related quality of life assessment in the drug regulatory process. Drug Inf J 36:209-238, 2002 Lohr KN, Aaronson NK, Alonso J, et al: Evaluating quality-of-life and health status instruments: Development of scientific review criteria. Clin Ther 18:979-992, 1996 Galer BS, Jensen MP: Development and preliminary validation of a pain measure specific to neuropathic pain: The neuropathic pain scale. Neurology 48:332-338, 1997 Au E, Loprinzi CL, Dhodapkar M, et al: Regular use of a verbal pain scale improves the understanding of oncology inpatient pain intensity. J Clin Oncol 12:2751-2755, 1994 Dellon AL. The vibrometer. Plast Reconstr Surg 71:427-431, 1983 Shy ME, Frohman EM, So YT, et al: Quantitative sensory testing: Report of the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology. Neurology 60:898-904, 2003 Elderson A, Gerritsen van der Hoop R, Haanstra W, et al: Vibration perception and thermoperception as quantitative measurements in

47

170. 171.

172.

173.

174.

175.

176.

177.

178.

179.

180.

181.

182.

183.

184.

185.

186.

187. 188.

189.

190.

the monitoring of cisplatin induced neurotoxicity. J Neurol Sci 93:167-174, 1989 Kamon N: Quantitative measurement of vibratory perception threshold using a new vibrometer TM-31A. J Occup Med 36:989-996, 1994 Oshita F, Saijo N, Shinkai T, et al: Correlation between total dose of cisplatin and vibratory perception threshold in chemotherapy induced neuropathy of cancer patients. Cancer J 5:165-169, 1992 Mayerhofer K, Bodner-Adler B, Bodner K, et al: Paclitaxel/carboplatin as first-line chemotherapy in advanced ovarian cancer: Efficacy and adverse effects with special consideration of peripheral neurotoxicity. Anticancer Res 20:4047-4050, 2000 Apfel SC, Kessler JA, Adornato BT, et al: Recombinant human nerve growth factor in the treatment of diabetic polyneuropathy. NGF Study Group. Neurology 51:695-702, 1998 Van Deursen RW, Sanchez MM, Derr JA, et al: Vibration perception threshold testing in patients with diabetic neuropathy: Ceiling effects and reliability. Diabet Med 18:469-475, 2001 Tunis S, Shuren J, Fernandez MJ: Medicare coverage Policy ⬃NCDs, electrodiagnostic sensory nerve conduction threshold. Administrative File CAG-00106N 2002 Calhoun E, Fishman D, Roland P, et al: Validity and selective sensitivity of the FACT/GOG-Ntx. Proc Am Soc Clin Oncol 19:446a, 2000 (abstr 1751) Calhoun EA, Welshman EE, Chang CH, et al: Psychometric evaluation of the functional assessment of cancer therapy/gynecologic oncology group-neurotoxicity (Fact/GOG-Ntx) questionnaire for patients receiving systemic chemotherapy. Int J Gynecol Cancer 13:741748, 2003 Cella D, Peterman A, Hudgens S, et al: Measuring the side effects of taxane therapy in oncology: The Functional Assessment of Cancer Therapy-Taxane (FACT-taxane). Cancer 98:822-831, 2003 Epstein AM, Hall JA, Tognetti J, et al: Using proxies to evaluate quality of life. Can they provide valid information about patients’ health status and satisfaction with medical care? Med Care 27:S91-S98, 1989 (suppl) Grossman SA, Sheidler VR, Swedeen K, et al: Correlation of patient and caregiver ratings of cancer pain. J Pain Symptom Manage 6:53-57, 1991 Calhoun EA, Fishman DA, Roland PY, et al: Total cost of chemotherapy-induced hematologic and neurologic toxicity. Proc Am Soc Clin Oncol 18a:1606, 1999 (abstr 886) Calhoun EA, Chang CH, Welshman EE, et al: Evaluating the total costs of chemotherapy-induced toxicity: results from a pilot study with ovarian cancer patients. Oncologist 6:441-445, 2001 Gandara DR, Perez EA, Weibe V, et al: Cisplatin chemoprotection and rescue: pharmacologic modulation of toxicity. Semin Oncol 18:4955, 1991 (suppl 3) Hausheer FH, Kochat H, Parker AR, et al: New approaches to drug discovery and development: A mechanism-based approach to pharmaceutical research and its application to BNP7787, a novel chemoprotective agent. Cancer Chemother Pharmacol 52:S3-S15, 2003 (suppl) Windebank AJ, Smith AG, Russell JW: The effect of nerve growth factor, ciliary neurotrophic factor, and ACTH analogs on cisplatin neurotoxicity in vitro. Neurology 44:488-494, 1994 Hayakaawa K, Itoh T, Niwa H, et al: NGF prevention of neurotoxicity induced by cisplatin, vincristine and Taxol depends on toxicity of each drug and NGF treatment schedule: In vitro study of adult rat sympathetic ganglion explants. Brain Res 794:313-319, 1998 Apfel SC, Arezzo JC, Lipson L, et al: Nerve growth factor prevents experimental cisplatin neuropathy. Ann Neurol 31:76-80, 1992 Apfel SC, Schwartz S, Bruce T, et al: Efficacy and safety of recombinant human nerve growth factor in patients with diabetic polyneuropathy. JAMA 284: 2215-2221, 2000 Müller LJ: Morphological and electrophysiological study of the effects of cisplatin and ORG 2766 on rat spinal ganglion neurons. Cancer Res 50:2437-2442, 1990 Roberts JA, et al: A randomized, multicenter, double-blind, placebocontrolled, dose-finding study of ORG 2766 in the prevention or

Chemotherapy-induced peripheral neuropathy

48

191.

192.

193.

194.

195.

196.

197.

198.

199. 200.

201.

202.

203.

204.

205.

206. 207.

208.

209. 210.

211.

delay of cisplatin-induced neuropathies in women with ovarian cancer. Gynecol Oncol 67:172-177, 1997 van der Hoop RG, Vecht CJ, van der Burg ME, et al: An ACTH- (4-9) analog prevents cisplatin-induced neuropathy in ovarian cancer patients. N Engl J Med 322:89-94, 1990 Hovestadt A, van der Burg ME, Verbiest HB, et al: The course of neuropathy after cessation of cisplatin treatment, combined with Org 2766 or placebo. J Neurol 239:143-146, 1992 Apfel SC: Managing the neurotoxicity of paclitaxel (Taxol) and docetaxel (Taxotere) with neurotrophic factors. Cancer Invest 18:564573, 2000 Hensley ML, Schuchter LM, Lindley C, et al: American Society of Clinical Oncology clinical practice guidelines for the use of chemotherapy and radiotherapy protectants. J Clin Oncol 17:3333-3355, 1999 Gelmon K, Eisenhauer E, Bryce C, et al: Randomized phase II study of high-dose paclitaxel with or without amifostine in patients with metastatic breast cancer. J Clin Oncol 17:3038-3047, 1999 Moore DH, Donnelly J, McGuire WP, et al: Limited access trial using amifostine for protection against cisplatin- and paclitaxel-induced neurotoxicity: A phase II study of the Gynecologic Oncology Group. J Clin Oncol 21:4207-4213; 2003 Leong SS, Tan EH, Fong KW, et al: Randomized double-blind trial of combined modality treatment with or without amifostine in unresectable stage III non-small-cell lung cancer. J Clin Oncol 21:1767-1774, 2003 Gradishar WJ, Stephenson P, Glover DJ, et al: A phase II trial of cisplatin plus WR-2721 (amifostine) for metastatic breast carcinoma: An Eastern Cooperative Group study (E8188). Cancer 92:2517-2522, 2001 Makino H: Treatment and care of neurotoxicity from taxane anticancer agents. Br J Cancer 11:100-104, 2004 Cascinu S, Graziano F, Del Ferro E, et al: Expression of p53 protein and resistance to preoperative chemotherapy in locally advanced gastric carcinoma. Cancer 83:1917-1922, 1998 Anderson ME, Naganuma A, Meister A: Protection against cisplatin toxicity by administration of glutathione ester. FASEB J 4:3251-3255, 1990 Leone R, Fracasso ME, Soresi E, et al: Influence of glutathione administration on the disposition of free and total platinum in patients after administration of cisplatin. Cancer Chemother Pharmacol 29:385390, 1992 Zunino F, Pratesi G, Micheloni A, et al: Protective effect of reduced glutathione against cisplatin-induced renal and systemic toxicity and its influence on the therapeutic activity of the antitumor drug. Chem Biol Interact 70:89-101, 1989 Aebi S, Assereto R, Lauterburg BH: High-dose intravenous glutathione in man. Pharmacokinetics and effects on cyst(e)ine in plasma and urine. Eur J Clin Invest 21:103-110, 1991 Hamers FP, Brakkee JH, Cavalletti E, et al. Reduced glutathione protects against cisplatin-induced neurotoxicity in rats. Cancer Res 53: 544-549, 1993 Tognella S: Pharmacological interventions to reduce platinum-induced toxicity. Cancer Treat Rev 17:139-142, 1990 Oriana S, Bohm S, Spatti G, et al: A preliminary clinical experience with reduced glutathione as protector against cisplatin-toxicity. Tumori 73:337-340, 1987 Pirovano C, Balzarini A, Bohm S, et al. Peripheral neurotoxicity following high-dose cisplatin with glutathione: clinical and neurophysiological assessment. Tumori 78:253-257, 1992 Bernstein EF, Pass HA, Glass J: Modulation of cisplatin toxicity by glutathione. Int J Oncol 7:353-358, 1995 Bogliun G, Marzorati L, Marzola M, et al: Neurotoxicity of cisplatin ⫹/- reduced glutathione in the first-line treatment of advanced ovarian cancer. Int J Gynecol Cancer 6:415-419, 1996 Bohm S, Battista Spatti G, Di Re F, et al: A feasibility study of cisplatin administration with low-volume hydration and glutathione protection in the treatment of ovarian carcinoma. Anticancer Res 11:16131616, 1991

212. Cascinu S, Labianca R, Alessandroni P, et al: Intensive weekly chemotherapy for advanced gastric cancer using fluorouracil, cisplatin, epidoxorubicin, 6S-leucovorin, glutathione, and filgrastim: A report from the Italian Group for the Study of Digestive Tract Cancer. J Clin Oncol 15:3313-3319, 1997 213. Cascinu S, Catalano V, Cordella L, et al: Neuroprotective effect of reduced glutathione on oxaliplatin-based chemotherapy in advanced colorectal cancer: A randomized, double-blind, placebo-controlled trial. J Clin Oncol 20:3478-3483, 2002 214. Cozzaglio L, Doci R, Colella G, et al: A feasibility study of high-dose cisplatin and 5-fluorouracil with glutathione protection in the treatment of advanced colorectal cancer. Tumori 76:590-594, 1990 215. Romero FJ, Roma J: Glutathione and protein kinase C in peripheral nervous tissue. Methods Enzymol 252:146-153, 1995 216. Parnis FX, Coleman RE, Harper PG, et al: A randomised double-blind placebo controlled clinical trial assessing the tolerability and efficacy of glutathione as an adjuvant to escalating doses of cisplatin in the treatment of advanced ovarian cancer. Eur J Cancer 31A:1721, 1995 217. Smyth JF, Bowman A, Perren T, et al: Glutathione reduces the toxicity and improves quality of life of women diagnosed with ovarian cancer treated with cisplatin: Results of a double-blind, randomised trial. Ann Oncol 8:569-573, 1997 218. Tedeschi M, De Cesare A, Oriana S, et al: The role of glutathione in combination with cisplatin in the treatment of ovarian cancer. Cancer Treat Rev 18:253-259, 1991 219. Vahdat L, Papadopoulos K, Lange D, et al: Reduction of paclitaxelinduced peripheral neuropathy with glutamine. Clin Cancer Res 7:1192-1197, 2001 220. Vahdat L: Reduction of paclitaxel induced peripheral neuropathy with glutamine. Chemotherapy Foundation Symposium XVIII Innovative Cancer Therapy for Tomorrow, New York, NY, November 8-11, 2000 (abstr 1590) 221. Vardimon L. Neuroprotection by glutamine synthetase. Isr Med Assoc J 2:46-51, 2000 (suppl) 222. Boyle FM, Wheeler HR, Shenfield GM: Glutamate ameliorates experimental vincristine neuropathy. J Pharmacol Exp Ther 279:410-415, 1996 223. Boyle FM, Wheeler HR, Shenfield GM: Amelioration of experimental cisplatin and paclitaxel neuropathy with glutamate. J Neurooncol 41: 107-116, 1999 224. Wong GY, Michalak JC, Sloan JA, et al: A phase III double blinded, placebo controlled, randomized trial of gabapentin in patients with chemotherapy-induced peripheral neuropathy: A North Central Cancer Treatment Group study. J Clin Oncol 23:729s, 2005 (abstr 8001) 225. Mitchell P, Goldstein D, Michael M, et al: White addition of gabapentin to a modified FOLFOX regimen does not reduce neurotoxicity in patients with advanced colorectal cancer. J Clin Oncol 23:266s, 2005 (abstr 3581) 226. Adelsberger H, Quasthoff S, Grosskreutz J, et al. The chemotherapeutic oxaliplatin alters voltage-gated Na⫹ channel kinetics on rat sensory neurons. Eur J Pharmacol 406:25-32, 2000 227. Wilson, RH Lehky T, Thomas RR, et al: Acute oxaliplatin-induced peripheral nerve hyperexcitability. J Clin Oncol 20:1767-1774, 2002 228. Lersch C, Schmelz R, Eckel F, et al: Prevention of oxaliplatin-induced peripheral sensory neuropathy by carbamazepine in patients with advanced colorectal cancer. Clin Colorectal Cancer 2:54-58, 2002 229. Hausheer FH, Kanter P, Cao S, et al: Modulation of platinum induced toxicities and therapeutic index: Mechanistic insights and first and second-generation protecting agents. Semin Oncol 25:584-599, 1998 230. Hausheer F, Cavaletti G, Tredici G, et al: Oral and intravenous BNP7787 protects against platinum neurotoxicity without in vitro or in vivo tumor protection. Proc Am Assoc Cancer Res 40:398, 1999 (abstr 2633) 231. Cavalletti E, Cavaletti G, Tredici G, et al: Oral and intravenous BNP7787 protects against paclitaxel-mediated neurotoxicity in Wistar rats. Proc Am Assoc Cancer Res 40:398, 1999 (abstr 2632) 232. Boven E, Verschraagen M, Hulscher TM, et al: BNP7787, a novel protector against platinum-related toxicities, does not affect the effi-

F.H. Hausheer et al cacy of cisplatin or carboplatin in human tumour xenografts. Eur J Cancer 38:1148-1156, 2002 233. Schilsky R, Schwartz G, Berghorn E, et al: Phase I trial of escalating doses of BNP7787 in patients receiving cisplatin (CDDP) and paclitaxel (Tax). Proc Am Soc Clin Oncol 18:169a, 1999 (abstr 647) 234. Boven E, Westerman M, van Groeningen CJ, et al: Phase I and pharmacokinetic study of the novel chemoprotector BNP7787 in combi-

49 nation with cisplatin and attempt to eliminate the hydration schedule. Br J Cancer 92:1636-1643, 2005 235. Takeda K, Negoro S, Matsui K, et al: Phase I Safety and pharmacokinetic trial of BNP7787 in patients receiving cisplatin (CDDP) and paclitaxel (PTX) for advanced non-small cell lung cancer (NSCLC). An Osaka Phase I Study Group trial. Proc Am Soc Clin Oncol 19:114a, 2002 (abstr 453)