Neurological complications of cancer chemotherapy Jerzy Hildebranda,b
Purpose of review To update central and peripheral nervous system neurological manifestations caused by anticancer agents. Recent findings Mostly unpredictable encephalopathy continues to be sporadically reported even in patients treated systemically with conventional chemotherapy doses. Recently, capecitabine, a 5-fluorouracil prodrug, has been added to the list. Magnetic resonance diffusion-weighted and fluidattenuated inversion-recovery imaging are useful in demonstrating chemotherapy-induced central nervous system lesions. The pathogenesis of these lesions is often poorly understood, and is probably multifactorial. A recent observation indicates that genetic polymorphism for methionine is a potent risk factor for methtrexate-induced central nervous system toxicity. Chronic peripheral neuropathy still represents a major limiting factor in a series of chemotherapeutic drugs, and the neuroprotective effect of several older and newer agents is either deceptive or insufficiently proven. In addition to chronic neuropathy, oxaliplatine causes a unique acute syndrome which may respond to calcium plus magnesium infusion. Summary Neurotoxicity remains a major limitation of many drugs used in cancer patients. Their list grows steadily, and magnetic resonance imaging makes easier the recognition of central nervous system toxicity. Synthesis and thorough clinical testing of neuroprotective molecules remain a major challenge. Keywords anti-cancer agents, neurotoxicity, preventive treatment Curr Opin Oncol 18:321–324. ß 2006 Lippincott Williams & Wilkins. a Institut Jules Bordet, Brussels, Belgium and bL’Hoˆpital de la Salpeˆtie`re, Paris, France
Correspondence to J. Hildebrand MD, Consultant Neurologist, Institut Jules Bordet, Rue He´ger-Bordet 1 Brussels, Belgium Tel: +1 32 68 445294; e-mail:
[email protected] Current Opinion in Oncology 2006, 18:321–324 Abbreviations CNS PNS
central nervous system peripheral nervous system
ß 2006 Lippincott Williams & Wilkins 1040-8746
Introduction Were it not for the blood–brain barrier, central nervous system (CNS) toxicity would be the main factor limiting anticancer chemotherapy. Despite this barrier, CNS toxicity, mainly encephalopathy with or without seizures, occurs occasionally even when conventional doses are used. CNS toxicity incidence increases markedly when the blood–brain barrier is either overwhelmed (high systemic dose, intracarotid infusion) or bypassed (intrathecal administration). Peripheral nervous system (PNS) toxicity is more common because proximal and distal extremities of the peripheral nerves are not protected by a blood–brainlike barrier, and peripheral neuropathy remains a major limiting factor for the administration of conventional doses of several agents.
Central nervous system toxicity Encephalopathy, with or without seizures, is uncommon when conventional drug doses are administrated systemically, but ifosfamide encephalopathy is an exception, occurring in 10–30% of treated patients [1]. The clinical manifestations include alteration of consciousness, asterixis, extrapyramidal signs, cranial nerve palsies, psychotic behaviour, hallucinations and seizures. These manifestations usually clear within 3–4 days of therapy cessation. Ifosfamide encephalopathy is attributed to metabolites such as thialysine ketamine and chloracetaldehyde, which cross the blood–brain barrier and inhibit mitochondrial respiration. Methylene blue is believed to prevent ifosfamide encephalopathy by acting on super oxide formation. Use of methylene blue, proposed over 10 years ago [2], is still based on case reports or small patient series [3,4,5], and some authors consider that prophylactic use of this agent may not be justified as even severe manifestations of encephalopathy can be rapidly and completely resolved [6]. Prospective and controlled studies are still needed to determine the optimal use of methylene blue in ifosfamide-induced encephalopathy. Sporadic, mostly unpredictable, encephalopathy after systemic administration of conventional-dose chemotherapy has been reported for virtually every anticancer drug [7]. The pathogenesis is complex and often poorly understood; it may involve metabolic changes such as hyponatremia or hypomagnesemia, renal and liver dysfunction, fever and direct CNS drug toxicity. The rarity and the pathogenic complexity of these encephalopathies are 321
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
322 Supportive care
illustrated by Steeghs et al. [8] for cisplatin-induced encephalopathy, with seizures observed in three patients. Another, recently reported example of occasional CNS toxicity is a multifocal leukoencephalopathy induced by capecitabine, a 5-fluorouracil (5FU) prodrug [9,10]. Capecitabine encephalopathy differs from 5FU encephalopathy by earlier onset (within a few days), and its symptoms resolve after a few days. Brain magnetic resonance imaging (MRI) shows increased signals on diffusionweighted, fluid-attenuated inversion-recovery (FLAIR) and T2 sequences, primarily affecting the corpus callosum. CNS toxicity of anticancer drugs is more frequent when the drugs are administrated intrathecally. Acute and chronic neurotoxicity of methotrexate has been studied the most. Transient aseptic meningitis, which resolves spontaneously within a few days, is the most common presentation of acute methotrexate toxicity. Other clinical manifestations of parenchymal lesions, including pseudobulbar palsy [11] or hemiperesis [12] have been reported. Early methotrexate toxicity may cause MRI changes detectable by diffusion-weighted imaging [12]. Chronic (late-delayed) leukoencephalopathy is a well known side effect of methotrexate. Its incidence is dose related and is increased by concomitant whole brain irradiation. A recent study by Linnebank et al. [13] indicates that genetic polymorphisms for methionine metabolism (required for myelination) are a potential risk factor for late methotrexate CNS toxicity. The ischemic lesion attributed by Steeghs et al. [8] to cisplatin in one of their patients and the ‘stroke-like’ presentation of transient encephalopathy ascribed to methotrexate by Ku¨ker et al. [12] have led us to consider the issue of cerebrovascular diseases caused by chemotherapy. This association has been reviewed by El Amrani et al. [14]. Several drugs or drug combinations are potential causes of stroke. Yet, the pathogenic role of some anticancer agents remains uncertain. The issue is complicated by at least two factors. First, both cerebrovascular diseases and cancer are common age-related diseases and may be associated by chance. Second, the diagnosis of cerebrovascular diseases is often based on acute clinical presentation, and cases such as that reported by Ku¨ker et al. [12] illustrate that direct drug toxicity may mimic a stroke. Corticosteroids and antiepileptic drugs (AEDs) are common adjuvant therapies used in neurooncology. Their use is more complicated in cancer patients than in the general population. First, both groups of drugs may interfere with chemotherapy activity, through P450 enzyme induction [15]. Second, they may mimic brain tumour progression through several neurological side effects including mental abnormalities such as memory
and attention disorders [16], which are probably underestimated [17]. Newer AEDs are increasingly used in cancer patients due to the lack of liver enzyme induction and possibly a better tolerance. Usually reversible cognitive and psychiatric adverse events have been observed with topiramate [18,19] and levetiracetam [20], among others. These observations, which have been made in a general population of epileptic patients, most probably apply to brain tumour patients.
Peripheral nervous system toxicity Chemotherapy is the main cause of peripheral neuropathy in cancer patients. Symptoms and signs of peripheral neuropathy may occur after administration of numerous anticancer drugs. The main culprits, however, are vinca alkaloids, which produce an axonal dying back neuropathy, platinum derivatives (cisplatin and oxaliplatin), which cause sensory neuronopathy, and taxanes (paclitaxel, docetaxel), which cause a sensory-motor neuropathy. For all these drugs, the severity of chronic peripheral neuropathy is related to total dose and to shorter interval of drug administration. Patients with preexisting neuropathy are at particular risk. Symptoms and signs of vinca alkaloid and taxane neuropathy predominate in the lower limbs, and this has been attributed to nerve length. The effect of a patient’s height, however, remains unproven [21]. In a number of patients, symptoms and signs of PNS toxicity progress over several weeks to a few months after drug withdrawal. This off-therapy worsening has been best documented for vincristine and cisplatin, and recently confirmed for vincristine [22]. In addition to chronic sensory neuronopathy, oxaliplatin causes a unique acute syndrome in up to 90% of patients, characterized by cold hypersensitivity, jaw and eye tightness, hand, feet and perioral paresthesias, cramps and pseudolaryngospasm. The symptoms start 30–60 min after drug infusion and vanish within a few days. Unlike chronic neuropathies, which are caused by structural lesions, acute oxaliplatin-induced neurotoxicity is attributed to peripheral nerve hyperexcitability, possibly similar to acquired myotonia. On the basis of this hypothesis, Gamelin et al. [23] treated the manifestations of early oxaliplatin neurotoxicity with prophylactic calcium and magnesium infusion. The results are promising, but the study is retrospective and the findings need to be confirmed. The mechanism causing chronic oxaliplatin-induced neuropathy has not been established, but Krishnan et al. [24] suggest that it may involve sodium channel dysfunction.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Neurological complications of chemotherapy Hildebrand 323
Conclusion Many agents have been used in an attempt to prevent chronic chemotherapy-induced neuropathy. Several products that showed a potent protective activity in vitro or in animal models did not gain routine clinical use [25], their activity being either too weak or nonreproducible. This list includes glutamic acid (tested in vincristine-induced neuropathy), Org-2766 (see [22], tested in cisplatin and vincristine-induced neuropathy), amifostine (tested in cisplatin and paclitaxel-induced neuropathy), and glutathione (tested in cisplatin-induced neuropathy). In a recent study, a significant neuroprotective effect of amifostine has been observed in patients treated with paclitaxel and carboplatin, yet the authors do not recommend its routine use [26]. Neurotropic factors have shown promising activity in experimental models, but their clinical use is limited by the difficulty with drug administration, side effects and the theoretical fear that they may affect tumour growth. At least two substances, however, deserve further investigation. A favourable effect of vitamin E 300 mg twice daily was observed in a randomized controlled trial including 31 patients treated with cisplatin or paclitaxel [27]: neurotoxicity was observed in 73% of control and in 25% of treated patients (P ¼ 0.019). BNP-7787 is a disulfide agent which is converted to mesna in various tissues. Unlike mesna, BNP-7787 does not interfere with antitumour activity of platinum derivatives [28]. So far, however, its protective effect against paclitaxel or platinum-derivative neurotoxicity is based on animal studies alone.
References and recommended reading Papers of particular interest, published within the annual period of review, have been highlighted as: of special interest of outstanding interest Additional references related to this topic can also be found in the Current World Literature section in this issue (pp. 397–398). 1
Merimsky O, Inbar M, Reider-Groswasser I, et al. Ifosfamide-related acute encephalopathy: clinical and radiological aspects. Eur J Cancer 1991; 27:1188–1189.
2
Ku¨pfer A, Aeschlimann C, Wermuth B, Cerny T. Prophylaxis and reversal of ifosfamide with methylene blue. Lancet 1996; 343:763–764.
3
Pelgrims J, De Vos F, Van den Brande J, et al. Methylene blue in the treatment of ifosfamide encephalopathy: report of 12 cases and review of the literature. Br J Cancer 2000; 82:291–293.
4
Turner AR, Duong CD, Good DJ. Methylene blue for the treatment and prophylaxis of ifosfamide-induced encephalopathy. Clin Oncol 2003; 15: 435–439.
Kasper B, Harter C, Meissner J, et al. Prophylactic treatment of known ifosfamide-induced encephalopathy for chemotherapy with high-dose ifosfamide? Support Care Cancer 2004; 12:205–207. This short communication describes a woman with Ewing sarcoma treated with ifosfamide 4 g/m2/day for 3 consecutive days; no encephalopathy occurred.
5
Raj AB, Berthone SL, Jaffe N. Methylene blue encephalopathy reversal of ifosfamide-related encephalopathy. J Pediatr Hematol Oncol 2004; 26:116. In this 10-year-old boy treated with ifosfamide 3.5 g/m2 for 5 consecutive days, pronounced drug-induced encephalopathy was promptly and completely reversed by methylene blue 50 mg every 4 h.
6
7
Hildebrand J, Brada M. Altered consciousness. Oxford: Oxford University Press; 2001. pp. 10–12.
8
Steeghs N, de Jongh FE, Sillevis Smitt PAE, van den Bent MJ. Cisplatininduced encephalopathy and seizures. Anti-Cancer Drugs 2003; 14: 443–446.
9 Neimann B, Rochlitz C, Herrmann R, Pless M. Toxic encepahalopathy induced by capecitabine. Oncology 2004; 66:331–335. To the best of the authors’ and our knowledge, this is the first case report of capecitabine-induced encephalopathy. 10 Videnovic A, Semenov I, Chua-Adajar R, et al. Capecitabine-induced multifocal leukoencephalopathy: a report of five cases [abstract]. Neurology 2005; 65:1792–1794. 11 Kinirons P, Fortune A, Enright H, Murphy RPJ. Acute pseudobulbar palsy due to methotrexate with rapid response to intravenous immunoglobulin. J Neurol 2005; 252:1401–1403. The authors describe a 17-year-old boy with acute leukaemia who presented with pseudobulbar palsy after 5 weekly intrathecal injections of methotrexate 12.5 mg. The interesting point of the observation is the prompt response to intravenous immunoglobulin. The observation is original, but needs confirmation as spontaneous recovery cannot be ruled out. 12 Ku¨ker W, Bader P, Herrlinger U, et al. Transient encephalopathy after intrathecal methotrexate chemotherapy: diffusion-weighted MRI. J Neurooncol 2005; 73:47–49. The authors describe a 16-year-old boy with relapsing acute lymphoblastic leukaemia who developed an acute transient hemiparesis 7 days after intrathecal administration of methotrexate 12 mg. FLAIR and T1-weighted contrast-enhanced MRI was normal, but diffusion-weighted images showed bilateral asymmetric white matter changes with residual changes on T2-weighted sequence. 13 Linnebank M, Pels H, Kleczar N, et al. MTX-induced white matter changes are associated with polymorphisms of methionine metabolism. Neurology 2005; 64:912–913. In 42 patients with primary CNS lymphoma treated with systemic and intraventricular methotrexate, the presence of at least one of three functional polymorphisms influencing methionine synthase activity was found to be a significant (P ¼ 0.001) risk factor for white matter changes. 14 El Amrani M, Aidi S, Amarenco P. [Cerebral ischemic events and anti-cancer therapy]. Rev Neurol (Paris) 2003; 159:371–380. 15 Oberndorfer S, Piribauer M, Marosi Ch, et al. P450 enzyme inducing and non-inducing antiepileptics in glioblastoma patients with standard chemotherapy. J Neurooncol 2005; 72:255–260. Glioblastoma patients treated with chemotherapy and mostly valproic acid (which does not induce P450) had a longer survival and a higher toxicity than those who received carbamazepine (a P450 inducer). These results are limited by the retrospective nature of the study, but are vital if they can be replicated prospectively. 16 Uttner I, Mu¨ller S, Zinser C, et al. Reversible impaired memory by pulsed methylprednisolone in patients with MS. Neurology 2005; 64:1971– 1973. Thirty patients treated with methylprednisolone 500 or 2000 mg/day for 5 days showed marked changes in declarative memory retrieval, which were reversible within 60 days. The observation was made in multiple sclerosis patients but is probably applicable to cancer patients. 17 Sacks O, Shulman M. Steroid dementia: an overlooked diagnosis. Neurology 2005; 64:707–709. 18 Mula M, Trimble MR, Lhatoo SD, Sanders JWAS. Topiramate and psychiatric adverse events in patients with epilepsy. Epilepsia 2003; 44:659–663. 19 Kanner AM, Wuu J, Faught E, et al. A past history may be a risk factor for topiramate-related psychiatric and cognitive adverse events. Epilepsy Behav 2003; 4:548–552. 20 Mula M, Trimble MR, Sander JWAS. Psychiatric adverse events in patients with epilepsy and learning disabilities taking levetiracetam. Seizure 2004; 13:55–57. From 517 patients started on levetiracetam, 118 had learning disabilities, and 15 (12.7%) experienced psychiatric disorders attributed to the drug. Aggressive behaviour represented the main psychiatric complication. It was not related to titration schedule. 21 Penshaw H, Beamon K, Longmate J, et al. The effect of height on paclitaxel nerve damage. J Neurooncol 2005; 74:207–210. Nerve conductions and sensory tests were performed in 21 breast cancer patients prior to and 20–40 days after the administration of paclitaxel 725 mg/m2. Changes correlated with height in peroneal but not in sural sensory nerve. The authors concluded that a larger number of patients and longer follow-up period would be required to determine height effect on paclitaxel neurotoxicity. 22 Verstappen CCP, Koeppen S, Heimans JJ, et al. Dose-related vincristine induced peripheral neuropathy with unexpected off-therapy worsening. Neurology 2005; 64:4055–4061. Off-therapy worsening of neurological signs and symptoms occurred in 30% and 24%, respectively, of 114 lymphoma patients with vincristine-induced neuropathy.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
324 Supportive care 23 Gamelin L, Boisdron-Celle M, Delva R, et al. Prevention of oxaliplatin-related neuro-toxicity by calcium and magnesium infusions: a retrospective study of 161 patients receiving oxaliplatin combined with 5-fluorouracil and leucovorin for advanced colo-rectal cancer. Clinical Cancer Res 2004; 10:4055– 4061. Preventive effect of calcium gluconate 1 g and magnesium sulfate 1 g infused before oxaliplatin administration (85, 100 or 130 mg/m2 plus 5-fluorouracil) every 3 weeks was tested in 161 patients with advanced colorectal cancer. Ninety-six received Ca/Mg infusion, and 65 did not. The two groups were comparable with respect to age, gender and performance status. A significant effect (P ¼ 0.001 or higher) on early neurotoxicity symptoms was observed. In addition, late neuropathy, evaluated by National Cancer Institute grading, was significantly (P ¼ 0.003) reduced. 24 Krishnan AV, Goldstein D, Friedlander M, Kiernan MC. Oxaliplatin-induced neurotoxicity and the development of neuropathy. Muscle Nerve 2005; 32:51–60. The neurophysiological investigation of nerve excitability was performed in 16 patients 2–8 weeks after the completion of oxaliplatin therapy, thus outside the period of expected acute hyperexcitability. 25 Hildebrand J. Neurotoxicity of cancer therapy and its prevention. Supportive Care Oncol 2005; 50:1–6.
26 Hilpert F, Sta¨hle A, Tome´ O, et al. Neuroprotection with amifostine in the first line treatment of advanced ovarian cancer with carboplatin/paclitaxel-based chemotherapy: a double-blind, placebo-controlled, randomized phase II study from the Arbeitsgemein-schaft Gyna¨kologische Onkologie (AGO) Ovarian Cancer Study Group. Support Care Cancer 2005; 13:797–805. In this study 71 women with ovarian cancer were treated with first-line combination of paclitaxel 175 mg/m2 plus carboplatin AUC 5 with or without epirubicin 60 mg/ m2 for 3 weeks. Thirty-seven patients were randomized to receive amifostine infusion (740 mg/m2, 30 min before paclitaxel administration), and a significant neuroprotective effect was observed for two-point discrimination, vibration perception and disappearance and depression of tendon reflex. Systemic toxicity, however, was higher in the amifostine group and the quality of life was not improved. 27 Argyriou AA, Chroni E, Koutras A, et al. Vitamin E for prophylaxis against chemotherapy-induced neuropathy. Neurology 2005; 64:26–31. The results of this study are encouraging but limited by the relatively small number (n ¼ 31) of clinically heterogeneous patients and the use of two drugs with different neurotoxicity mechanisms. They need to be replicated in a larger and more homogeneous group of patients. 28 Verschraagen M, Kedde MA, Hausheer FH, van der Vijgh WJF. The clinical reactivity of BNP-7787 and its metabolite mesna with the cytostatic agent cisplatin: comparison with the nucleophiles thiosulphate, DDTC, glutathione and its disulfide GSSG. Cancer Chemother Pharmacol 2003; 51:499–504.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
