0022-5347/05/1742-0646/0 THE JOURNAL OF UROLOGY® Copyright © 2005 by AMERICAN UROLOGICAL ASSOCIATION

Vol. 174, 646 – 650, August 2005 Printed in U.S.A.

DOI: 10.1097/01.ju.0000165342.85300.14

TREATMENT OF ERECTILE DYSFUNCTION FOLLOWING THERAPY FOR CLINICALLY LOCALIZED PROSTATE CANCER: PATIENT REPORTED USE AND OUTCOMES FROM THE SURVEILLANCE, EPIDEMIOLOGY, AND END RESULTS PROSTATE CANCER OUTCOMES STUDY ROBERT A. STEPHENSON,* MOTOMI MORI, YI-CHING HSIEH, TOMASZ M. BEER, JANET L. STANFORD, FRANK D. GILLILAND, RICHARD M. HOFFMAN AND ARNOLD L. POTOSKY From the Division of Urology, University of Utah School of Medicine (RAS), Salt Lake City, Utah, Biostatistics Shared Resource, Cancer Institute (MM, Y-CH) and Division of Hematology and Medical Oncology (TMB), Oregon Health and Science University, Portland Oregon, Department of Epidemiology, Fred Hutchinson Cancer Research Center, University of Washington (JLS), Seattle, Washington, Department of Preventive Medicine, Keck School of Medicine, University of Southern California (FDG), Los Angeles, California, Medicine Service, New Mexico Veterans Affairs Health Care System (RMH), Albuquerque, New Mexico, and Division of Cancer Control and Population Sciences, National Cancer Institute (ALP), Bethesda, Maryland

ABSTRACT

Purpose: Erectile dysfunction (ED) persists for years following curative therapies for clinically localized prostate cancer. We report use and treatment outcomes in a 5-year interval in a population based cohort from the Surveillance, Epidemiology, and End Results Prostate Cancer Outcomes Study. Materials and Methods: A sample of 1,977 men with localized prostate cancer who received external beam radiation therapy or radical prostatectomy in 1994 to 1995 were surveyed for 5 outcome measures of ED treatment, namely treatment, perceived helpfulness, erectile sufficiency, sexual activity frequency and erection maintenance. Subjects were surveyed 6, 12, 24 and 60 months after prostate cancer diagnosis. Results: Overall 50.5% of men ever used ED treatment. The use of ED treatments increased during the study course. Subject age, regular sexual partner and baseline sexual activity were factors positively associated with ED treatments. While it was used uncommonly (1.9%), a penile prosthesis was perceived as the most helpful ED treatment (helped a lot in 52% of respondents). Sildenafil helped a lot in 12% of respondents. Erectile fullness, erection maintenance and sexual activity frequency were modestly improved in men using ED treatment compared with those in men not using ED treatment. Conclusions: Approximately half of the patients in this population based cohort of men used ED treatment during the 5 years following prostate cancer diagnosis. Men using ED treatments had modest improvement in sexual function compared with men that in who did not receive ED treatment at 60 months. More effective treatments for ED following local therapy for prostate cancer are needed. KEY WORDS: penis, prostate, impotence, prostatic neoplasms, treatment outcome

Several recent studies have described patient reported quality of life outcomes following treatment for localized prostate cancer.1⫺4 These studies demonstrate that overall quality of life is uniformly good after treatment with radical prostatectomy (RP) or external beam radiation therapy (RT) for localized prostate cancer.1⫺3 However, sexual function and sexual satisfaction decrease significantly in patients subsequent to treatment with RP or RT.4⫺9 While published studies have documented improved sexual function with erectile dysfunction (ED) treatment following localized prostate cancer treatment, there has been little information on the population wide prevalence of ED treatment or on the comparative results achieved with the various forms of ED treatment. We used data on a population based, randomly sampled group of men treated in diverse clinical

settings to examine the prevalence and results of treatment for ED following treatment for clinically localized prostate cancer.10 MATERIALS AND METHODS

Prostate Cancer Outcomes Study (PCOS). In 1994 the National Cancer Institute initiated the PCOS to obtain longitudinal, population based estimates of health outcomes in men diagnosed with prostate cancer using self-administered questionnaires and abstracted medical records. The study was based on men diagnosed with prostate cancer who were randomly selected from the regions of 6 Surveillance, Epidemiology, and End Results cancer registries (Connecticut, Utah, New Mexico, metropolitan Atlanta, Los Angeles County and Seattle, King County) from October 1, 1994 through October 31, 1995. A total of 3,533 men were initially enrolled from a sample of 5,672 and they were surveyed 6, 12, 24 and 60 months after diagnosis. The survey instrument included general and disease specific measures of health related quality of life. The rationale, objectives and methods have been previously reported in more detail.10 In the current study we focused on ED treatment, includ-

Submitted for publication December 2, 2004. Supported by Contracts NO1-PC-67000, NO1-PC-67005, NO1-PC67006, NO1-PC-67007, NO1-PC-67009 and NO1-PC-67010 from the National Cancer Institute, National Institutes of Health, Bethesda, Maryland to participating SEER institutions. * Correspondence: Division of Urology, 50 North Medical Dr., Room 3B420, Salt Lake City, Utah 84132 (telephone: 801-581-4705; FAX: 801-585-2891; e-mail: [email protected]). 646

ERECTILE DYSFUNCTION TREATMENT OUTCOMES AFTER PROSTATE CANCER THERAPY

ing the type and use of ED treatment (vacuum erection devices, penile injections, psychosexual counseling, penile prosthesis, nonsildenafil medications and sildenafil), the perceived helpfulness of ED treatments and sexual function measures (erectile fullness, erection maintenance and sexual frequency activity). Study population. A total of 3,486 men completed a 6 and/or 12-month survey and underwent RP or RT within year 1 of diagnosis in PCOS. We restricted our analysis to patients diagnosed with clinically localized disease (2,176) who completed a 6-month survey (1,977). Of these study patients 1,753 (89%) completed a 12-month survey, 1,643 (83%) completed a 24-month survey, 1,462 (74%) completed a 60-month survey and 1,282 (65%) completed all 4 surveys. Definition of outcome variables. We examined ED treatment use, perceived helpfulness, erectile fullness, erection maintenance and sexual activity frequency. The use of ED treatment was defined as a series of binary variables indicating the use of each of 5 treatments (ie vacuum erection device, penile injection, psychological/sexual counseling, penile prosthesis and nonsildenafil medication) on the 6, 12, 24 and 60-month surveys. Sildenafil became available as a treatment for ED in 1997 and it was included as an ED treatment option in the 60-month survey. Use of treatments for ED was based on the question, “In the past 6 months, did you try any of the following to help with problems with sexual function?” This means that the survey could not distinguish between concurrent and serial use of ED treatments. The perceived helpfulness of each ED treatment was based on the question, “How much did it help?” (ie helped a lot vs helped somewhat vs helped not at all). Sexual function in the last month was assessed using the 3 questions, “Have you had any erections firm enough for sexual intercourse? ” (full vs partial or none), “How often have you engaged in sexual activity? ” (any vs none at all) and “How much difficulty did you have keeping an erect penis during sexual activity?” (some/little/or no difficulty vs no erections/a lot of difficulty). Statistical methods. All analyses were performed using SUDAAN.11 We specified stratified random sampling with replacement and the Horvitz-Thompson weight, which is the inverse of the sampling fraction for each stratum, defined by patient age, race and study area, to obtain unbiased estimates of proportion and regression parameters. All estimates presented were weighted according to the sampling fraction. Simple cross-tabulations were used to estimate proportions for demographic and clinical characteristics, use of ED treatment with time and perceived helpfulness. Logistic regression analysis was performed to evaluate the association between baseline sexual characteristics (regular sexual partner, interest in sex, sexual activity, change in impotency status and perception of a problem) and ED treatment at 60 months, adjusted for patient cancer treatment and demographic characteristics. Logistic regression analysis was performed to evaluate the effects of ED treatment on erectile sufficiency, sexual activity frequency and erection maintenance at 60 months. Predicted percents were obtained using the method of predicted margins.12 RESULTS

Table 1 shows the distribution of selected clinical and sociodemographic characteristics of the study cohort, that is 1,977 men who completed the 6 and/or 12-month surveys. Of this patient cohort 68% men underwent RP and 32% received RT. Overall 50.5% of the men used some form of treatment for ED during the 60 months following the prostate cancer diagnosis (table 2). Prior to the introduction of sildenafil (6, 12 and 24-month surveys) the most commonly used treatment was vacuum erection devices, followed by penile injections. Penile prostheses were the least used form of ED treatment.

647

TABLE 1. Select demographic characteristics Level Treatment: RP RT Age group: Younger than 55 55–59 60–64 65–69 70–74 Older than 74 Registry: Seattle Connecticut New Mexico Utah Atlanta Los Angeles Race: Hispanic Black White Education: Advanced College graduate Some college High school or less Marital status: Yes No/unknown Based on 1,977 patients and adjusted for sampling fractions.

% 68 32 9 13 23 26 21 8 5 23 10 11 16 35 10 13 77 22 16 44 18 83 17

Use of ED treatment increased steadily during the study duration with the largest increase at 60 months after the introduction of sildenafil. At 60 months 16.7% of respondents used sildenafil alone and 20.9% used sildenafil and other treatments for ED. Of those using sildenafil and other ED treatments at 60 months 33.1% also used a vacuum erection device, 22.6% used penile injections and 33.3% used more than 1 additional ED treatment. More than half of the men using vacuum erection devices and penile injections at 24 months changed to sildenafil alone or sildenafil plus an additional ED treatment, generally the same ED treatment that they were using at 24 months. We next examined which factors were associated with treatment use for ED 60 months after diagnosis (table 3). After adjustment for multiple sociodemographic and health status variables (race, education and cancer treatment) younger age at diagnosis, having a sexual partner prior to prostate cancer and higher sexual activity frequency prior to prostate cancer were statistically significantly associated with reported ED treatment use 60 months after diagnosis. We examined cases in which only 1 ED treatment was used to assess the perceived helpfulness of individual ED therapies. Perceived helpfulness at 60 months varied by ED treatment type (see figure). The proportion of users who believed that the treatment helped a lot was highest for penile prostheses (52%) and lowest for psychosexual counseling (7%). Notably only 12% of men using sildenafil reported that the treatment helped a lot. The level of helpfulness was stable for all nonsildenafil treatments from 6 to 60 months except for penile injections, which decreased from 43% (helped a lot) to 30%. The proportion of users at 60 months who believed that treatment helped a lot or helped somewhat was 71% for penile prostheses, 71% for vacuum erection devices, 69% for penile injection, 61% for nonsildenafil medication, 47% for sildenafil and 40% for psychosexual counseling. Respondents who reported using more than 1 ED treatment reported less helpfulness than respondents who used only 1 ED treatment (data not shown). We then examined the association of individual ED therapies with sexual function in patients who used only a single ED therapy (table 4). At 60 months full erections were reported in less than half of patients using any ED treatments.

648

ERECTILE DYSFUNCTION TREATMENT OUTCOMES AFTER PROSTATE CANCER THERAPY TABLE 2. ED treatment during 60 months following prostate cancer diagnosis and before RP or RT % Cell Entries ⫾ SE At Any Time

6 Mos

12 Mos

No. pts 1,977 1,977 Any ED treatment 50.5 ⫾ 1.30 14.0 ⫾ 0.87 Vacuum erection device only 16.5 ⫾ 0.94 4.5 ⫾ 0.57 Penile injection only 11.1 ⫾ 0.79 2.7 ⫾ 0.37 Nonsildenafil medication only 5.0 ⫾ 0.54 1.7 ⫾ 0.32 Psychosexual counseling only 4.5 ⫾ 0.55 1.5 ⫾ 0.29 Penile prosthesis only 1.9 ⫾ 0.34 0.6 ⫾ 0.19 Sildenafil only — — Sildenafil ⫹ others — — Other multiple treatments 11.6 ⫾ 0.80 2.9 ⫾ 0.44 No ED treatment 49.5 ⫾ 1.30 86.0 ⫾ 0.87 Question did not distinguish between concurrent and serial ED treatments during 6-month than 1 form of therapy (percent adjusted for sampling fractions).

24 Mos

60 Mos

1,753 1,642 25.4 ⫾ 1.19 32.7 ⫾ 1.31 9.8 ⫾ 0.82 11.9 ⫾ 0.89 6.6 ⫾ 0.68 7.4 ⫾ 0.74 1.6 ⫾ 0.32 2.1 ⫾ 0.39 2.0 ⫾ 0.44 1.9 ⫾ 0.42 1.1 ⫾ 0.30 1.3 ⫾ 0.32 — — — — 4.3 ⫾ 0.55 8.0 ⫾ 0.74 74.6 ⫾ 1.19 67.3 ⫾ 1.31 interval preceding each questionnaire in men

1,462 50.8 ⫾ 1.53 5.7 ⫾ 0.69 2.0 ⫾ 0.43 0.8 ⫾ 0.25 0.8 ⫾ 0.25 1.6 ⫾ 0.35 16.7 ⫾ 1.18 20.9 ⫾ 1.21 2.3 ⫾ 0.47 49.2 ⫾ 1.53 who used more

TABLE 3. Multivariate analysis of characteristics associated with ED treatments 60 months after localized prostate cancer diagnosis Level

OR

95% CI

p Value

Treatment: RP 1.19 0.81–1.73 0.38 RT 1.00 Age group: Younger than 60 3.05 2.04–4.55 ⬍0.0001 60–64 2.46 1.60–3.78 65–69 1.22 0.81–1.84 70 or Older 1.00 Race: Hispanic 0.87 0.58–1.29 0.31 NonHispanic black 1.23 0.85–1.78 NonHispanic white 1.00 Education: Advanced 1.19 0.73–1.93 0.85 College graduate 1.22 0.72–2.07 Some college 1.19 0.79–1.80 Less high school 1.00 Baseline regular sexual partner: Yes 2.14 1.32–3.48 0.0021 No/unknown 1.00 Baseline sex interest: A lot 1.78 0.82–3.86 0.34 Somewhat/only little 1.64 0.80–3.40 Not at all 1.00 Baseline sexual activity: At least 1/wk 2.91 1.65–5.13 0.0003 At least 1/mo 1.79 1.04–3.07 Not at all 1.00 Baseline-6-mo impotency status change: Potent to impotent 1.35 0.97–1.88 0.07 No change 1.00 Predicted percent of each covariate was obtained using predicted margin method and it was adjusted for all other table covariates.

Predicted % ⫾ SE 53 ⫾ 2 49 ⫾ 3 64 ⫾ 3 60 ⫾ 3 44 ⫾ 3 40 ⫾ 3 48 ⫾ 4 56 ⫾ 4 51 ⫾ 2 52 ⫾ 3 53 ⫾ 4 52 ⫾ 2 48 ⫾ 4 53 ⫾ 2 37 ⫾ 5 53 ⫾ 3 51 ⫾ 2 41 ⫾ 8 59 ⫾ 3 48 ⫾ 3 36 ⫾ 5 54 ⫾ 2 48 ⫾ 3

erection maintenance. Adjustments for potential clinical and sociodemographic confounders in multivariate models did not materially alter these results, and so we report only unadjusted percents (table 4). For purposes of comparison table 4 also shows recalled baseline sexual function prior to the diagnosis of prostate cancer. DISCUSSION

Perceived helpfulness of ED treatments at 60 months in men using only 1 ED treatment.

Full erections were reported by 42% of men using penile prostheses, 39% using penile injections and 39% using sildenafil alone. Men using penile prostheses reported the greatest success in maintaining erections (67%). Sexual activity frequency was similarly high in men using vacuum erection devices, penile injection, penile prostheses, sildenafil alone and sildenafil with other ED treatments (67% to 73%). Men who used no ED treatment reported low sexual success, as reflected by erectile ability, sexual activity frequency and

In a prior report we noted substantial decreases in sexual potency in patients treated with RT and RP (61.5% and 79.6% at 2 years, respectively).1 Others have reported similar effects of RP and RT on sexual function.2⫺9 Many men consider that sexual dysfunction is a substantial problem following treatment for prostate cancer with RP and RT.1 Those experiencing the largest amount of dysfunction relating to sexual function also have decrements in overall quality of life.13 The substantial decrease in sexual function and general health related quality of life observed in these studies suggests the need for effective treatment for ED in many patients following RT and RP. It is notable that half of the patients in the current study did not receive ED treatment in the 5 years following the

ERECTILE DYSFUNCTION TREATMENT OUTCOMES AFTER PROSTATE CANCER THERAPY TABLE 4. Predicted effect of each treatment 60 months after prostate cancer diagnosis Predicted % ⫾ SE Treatment

Erectile ability

Sexual Activity Frequency

Erection Maintenance

Vacuum erection device 21 ⫾ 5 68 ⫾ 6 24 ⫾ 5 only Penile injection only 39 ⫾ 12 72 ⫾ 9 42 ⫾ 11 Nonsildenafil (yohimbine, 25 ⫾ 12 44 ⫾ 14 31 ⫾ 13 etc) medication only Psychosexual counseling 16 ⫾ 9 40 ⫾ 14 11 ⫾ 8 only Penile prosthesis only 42 ⫾ 11 73 ⫾ 11 67 ⫾ 10 Sildenafil only 39 ⫾ 4 69 ⫾ 4 41 ⫾ 4 Sildenafil ⫹ others 26 ⫾ 3 67 ⫾ 3 28 ⫾ 3 Other multiple treatments 13 ⫾ 6 43 ⫾ 10 29 ⫾ 9 No ED treatment 23 ⫾ 2 36 ⫾ 2 24 ⫾ 2 Overall baseline 74 ⫾ 1.2 80 ⫾ 1.1 75 ⫾ 1.2 Values were obtained by the method of predicted margin method, representing percents unadjusted for other covariates and confounders since adjustment did little to change predicted margin magnitude (erectile ability—full vs partial or none, frequency—any vs not at all and maintenance— some/little/no difficulty vs no erection/a lot of difficulty).

diagnosis. Since those who did not receive ED therapy had generally low sexual function (table 4), adequate sexual function does not appear to account for low levels of ED treatment use in these men. It is possible that a combination of patient reluctance and failure of physicians to offer therapy accounts for the large number of men who did not receive ED therapy. In the first 2 years following diagnosis and prior to the introduction of sildenafil only about a third of men used some form of ED treatment. This proportion increased to half of patients at 5 years. Most of the increase at 5 years was due to sildenafil use. Because sildenafil is a convenient oral treatment, it has become the preferred initial treatment for erectile dysfunction.14 However, the drug is not generally effective in men with complete loss of erectile function.15 Furthermore, sildenafil appears to be associated with decreasing efficacy with time, it is not uniformly effective in men with milder forms of ED and it is associated with side effects that induce some men to discontinue its use.16 Our results indicate that sildenafil and other newer, related agents have become the most widely used form of treatment for ED following prostate cancer treatment but sildenafil was not considered to be as helpful by users as penile prostheses, vacuum erection devices or penile injection therapy. Based on these findings it appears that these other methods to treat ED will continue to have an important role in many patients.17 Of non-oral ED treatments it is notable that penile prostheses were associated with similar or better sexual function and perception of helpfulness and yet penile prostheses were the least used form of ED therapy. The substantially lower use despite similar or better outcomes of penile prostheses may be related to the greater risk, expense, inconvenience and discomfort associated with surgically placed penile prostheses compared with those of other, less invasive forms of ED treatment.18 Our results demonstrate that the effectiveness of currently available ED treatments is at best modest. Similar to the findings of Schover et al,18 our results indicate substantial room for improvement in the use, effectiveness and acceptability of therapy for ED following treatment for localized prostate cancer. To our knowledge the current study is the first to describe the prevalence and effectiveness of ED treatment in a population based cohort of men following treatment for localized prostate cancer. The major strengths of this series are derived from its large sample size and the use of a population based cohort. Population based data should be less affected

649

by the selection bias seen in institutional series and, therefore, they should be more reflective of outcomes in the population at large. An important potential limitation of this nonrandomized study is related to the introduction of sildenafil late in the study course. Patients who were highly satisfied with pre-sildenafil sexual function may have been less likely to try sildenafil. As a result, the sildenafil cohort may have been enriched for individuals with poorer sexual function and a lower probability of response to sildenafil, thereby, leading to falsely low sexual function outcomes in sildenafil treated patients. Other limitations are related to the possibility of recall bias of baseline sexual function because it was recalled by patients 6 months after diagnosis at the time of the 6-month questionnaire. However, a validation study of men in PCOS assessing 6-month recall did not reveal any large, systematic biases in the recall of pretreatment sexual function.19 Using a longer period of recall others have noted that men may overestimate sexual function before therapy.20 Bias may also be introduced due to differences between those who responded to the questionnaire and nonrespondents, especially when there is loss to followup at each survey. Nonrespondents to the questionnaire may have had poorer functional status, leading to underestimation of functional loss due to exclusion of the poorer outcomes of nonrespondents from the analysis. CONCLUSIONS

The use and effectiveness of ED treatment was limited in this population based cohort subsequent to definitive local therapy for localized prostate cancer. After it became available the convenient oral medication sildenafil was the most commonly chosen treatment for ED. Penile prostheses, penile injections and vacuum erection devices were more effective treatments for ED than sildenafil. However, these more intrusive ED treatments were only modestly effective. Our results indicate that there is substantial room for improvement in the use, helpfulness and functional sexual outcome of ED treatments in men following therapy for prostate cancer. REFERENCES

1. Potosky, A. L., Legler, J., Albertsen, P. C., Stanford, J. L., Gilliland, F. D., Hamilton, A. S. et al: Health outcomes after prostatectomy and radiotherapy for prostate cancer: results from the Prostate Cancer Outcomes Study. J Natl Cancer Inst, 92: 1582, 2000 2. Litwin, M. S., Flanders, S. C., Pasta, D. J., Stoddard, M. L., Lubeck, D. P. and Henning, J. M.: Sexual function and bother after radical prostatectomy or radiation for prostate cancer: multivariate quality-of-life analysis from CaPSURE. Cancer of the Prostate Strategic Urologic Research Endeavor. Urology, 54: 503, 1999 3. Lilleby, W., Fossa, S. D., Waehre, H. R. and Olsen, D. R.: Longterm morbidity and quality of life in patients with localized prostate cancer undergoing definitive radiotherapy or radical prostatectomy. Int J Radiat Oncol Biol Phys, 43: 735, 1999 4. Stanford, J. L., Feng, Z., Hamilton, A. S., Gilliland, F. D., Stephenson, R. A., Eley, J. W. et al: Urinary and sexual function after radical prostatectomy for clinically localized prostate cancer: the Prostate Cancer Outcomes Study. JAMA, 283: 354, 2000 5. Fowler, F. J., Jr., Barry, M. J., Lu-Yao, G., Roman, A., Wasson, J. and Wennberg, J. E.: Patient-reported complications and follow-up treatment following radical prostatectomy: The National Medicare Experience: 1988 –1990 (updated June 1993). Urology, 42: 622, 1993 6. Talcott, J. A., Rieker, P., Propert, K. J., Clark, J. A., Wishnow, K. I., Loughlin, K. R. et al: Patient-reported impotence and incontinence after nerve-sparing radical prostatectomy. J Natl Cancer Inst, 89: 1117, 1997 7. Hamilton, A. S., Stanford, J. L., Gilliland, F. D., Albertsen, P. C., Stephenson, R. A., Hoffman, R. M. et al: Health outcomes after

650

8.

9.

10.

11. 12. 13.

14.

15. 16.

ERECTILE DYSFUNCTION TREATMENT OUTCOMES AFTER PROSTATE CANCER THERAPY

external-beam radiation therapy for clinically localized prostate cancer: results from the Prostate Cancer Outcomes Study. J Clin Oncol, 19: 2517, 2001 Fowler, F. J., Jr., Barry, M. J., Lu-Yao, G. L., Wasson, J. H. and Bin, L.: Outcomes of external-beam radiation therapy for prostate cancer: a study of Medicare beneficiaries in three Surveillance, Epidemiology, and End Results areas. J Clin Oncol, 14: 2258, 1996 Talcott, J. A., Rieker, P., Clark, J. A., Propert, K. J., Weeks, J. C., Beard, C. J. et al: Patient-reported symptoms after primary therapy for early prostate cancer: results of a prospective cohort study. J Clin Oncol, 16: 275, 1998 Potosky, A. L., Harlan, L. C., Stanford, J. L., Gilliland, F. D., Hamilton, A. S., Albertsen, P. C. et al: Prostate cancer practice patterns and quality of life: the Prostate Cancer Outcomes Study. J Natl Cancer Inst, 91: 1719, 1999 Shah, B. V., Barnwell, B. G. and Bieler, G. S.: SUDAAN User’s Manual, Release 7.5. Research Triangle Park: Research Triangle Institute, 1997 Graubard, B. I. and Korn, E. L.: Predictive margins with survey data. Biometrics, 55: 652, 1999 Penson, D. F., Feng, Z., Kuniyuki, A., McClerran, D., Albertsen, P. C., Deapen, D. et al: General quality of life 2 years following treatment for prostate cancer: what influences outcomes? Results from the Prostate Cancer Outcomes Study. J Clin Oncol, 21: 1147, 2003 Goldstein, I., Lue, T. F., Padma-Nathan, H., Rosen, R. C., Steers, W. D. and Wicker, P. A.: Oral sildenafil in the treatment of erectile dysfunction. Sildenafil Study Group. N Engl J Med, 338: 1397, 1998 McMahon, C. G.: High dose sildenafil citrate as a salvage therapy for severe erectile dysfunction. Int J Impot Res, 14: 533, 2002 El-Galley, R., Rutland, H., Talic, R., Keane, T. and Clark, H.: Long-term efficacy of sildenafil and tachyphylaxis effect. J Urol, 166: 927, 2001

17. Chen, J., Mabjeesh, N. J. and Greenstein, A.: Sildenafil versus the vacuum erection device: patient preference. J Urol, 166: 1779, 2001 18. Schover, L. R., Fouladi, R. T., Warneke, C. L., Neese, L., Klein, E. A., Zippe, C. et al: The use of treatments for erectile dysfunction among survivors of prostate carcinoma. Cancer, 95: 2391, 2002 19. Legler, J., Potosky, A. L., Gilliland, F. D., Eley, J. W. and Stanford, J. L.: Validation study of retrospective recall of disease-targeted function: results from the Prostate Cancer Outcomes Study. Med Care, 38: 847, 2000 20. Litwin, M. S. and McGuigan, K. A.: Accuracy of recall in healthrelated quality-of-life assessment among men treated for prostate cancer. J Clin Oncol, 17: 2882, 1999 EDITORIAL COMMENT This is an important report because it highlights the modest improvements in sexual function noted with the use of various interventions in men treated for prostate cancer. The study is strengthened by its large sample size and use of a population based cohort. The domain of sexual function is clearly the one most affected by prostate cancer treatment. It is also the one most affected by age and comorbidity independent of prostate cancer. Although many men with prostate cancer have significant limitations in this domain before treatment, those who are the most sexually active are most at risk (i.e. those who have the most to lose, lose the most). Although I fully agree with the authors that better interventions are necessary; more immediate gains may be made by improved pre-treatment counseling, active surveillance rather than immediate treatment in some men, better case selection, and improved surgical and radiation technique. Peter R. Carroll Department of Urology University of California-San Francisco San Francisco, California