www.redjournal.org Clinical Investigation Comparative Effectiveness Research in Localized Prostate Cancer: A 10-Year Follow-up Cohort Study Olatz Garin, PhD,*,y,z José Francisco Suárez, MD,x Ferran Guedea, PhD,k Àngels Pont, MSc,*,y Yolanda Pardo, PhD,*,y,{ Alai Goñi, MD,# Alfonso Mariño, MD,** Asunción Hervás, PhD,yy Ismael Herruzo, PhD,zz Patricia Cabrera, PhD,xx Gemma Sancho, PhD,kk Javier Ponce de León, PhD,{{ Vı́ctor Macı́as, PhD,## Cristina Gutierrez, PhD,k Manel Castells, MD,x and Montse Ferrer, PhD,*,y,{ Multicentric Spanish Group of Clinically Localized Prostate Cancer *Health Services Research Group, Hospital del Mar Medical Research Institute, Barcelona, Spain; y CIBER en Epidemiologı´a y Salud Pública, CIBERESP,Spain; zPompeu Fabra University, Barcelona, Spain; xHospital Universitari de Bellvitge, L’Hospitalet de Llobregat, Spain; kInstitut Català d’Oncologia, L’Hospitalet de Llobregat, Spain; {Universitat Autònoma de Barcelona, Bellaterra, Spain; #Institutu Onkologikoa, San Sebastián, Spain; **Servicio de Oncologı´a Radioterápica, Centro Oncológico de Galicia, A Coruña, Spain; yyServicio de Oncologı´a Radioterápica, Hospital Ramón y Cajal, Madrid, Spain; zzServicio de Oncologı´a Radioterápica, Hospital Regional Carlos Haya, Málaga, Spain; xxServicio de Oncologı´a Radioterápica, Hospital Virgen del Rocı´o, Sevilla, Spain; kkServicio de Oncologı´a Radioterápica, Hospital Santa Creu i Sant Pau, Barcelona, Spain; {{Servicio de Urologı´a, Fundació Puigvert, Barcelona, Spain; and ##Hospital Universitario y Politecnico La Fe, Valencia, Spain Received Aug 12, 2020. Accepted for publication Dec 22, 2020. Purpose: Long-term comparative effectiveness research on localized prostate cancer treatments is scarce, and evidence is lacking especially for brachytherapy. The aim of this study was to assess the long-term impact of the side effects of radical prostatectomy, brachytherapy, and external radiation therapy on patients with localized prostate cancer at 10 years, using propensity score analyses. Corresponding author: Montse Ferrer, PhD; E-mail: mferrer@imim.es Data sharing statement: Data are available for bona fide researchers This study was supported by grants from the Instituto de Salud Carlos who request it from the authors. III FEDER: Fondo Europeo de Desarrollo Regional (grant numbers PI02/ Supplementary material for this article can be found at https://doi.org/ 0668, PI08/90090, and PI13/00412); Generalitat de Catalunya, Spain 10.1016/j.ijrobp.2020.12.032. (grant numbers 086/24/2000 and 2017 SGR 452); and Ministerio de AcknowledgmentsdThe authors acknowledge the collaboration and Ciencia e Innovación, Spain (grant number PTA2011-04891). The funders contribution of other members of the Multicentric Spanish Group of of the study had no role in the study design and conduct, data collection, Clinically Localized Prostate Cancer. The authors also thank the patients management, analysis, or interpretation; they also had no role in the and families who made this study possible and the clinical study teams preparation, writing, reviewing, and submission of this manuscript. who participated in the study. Moreover, the group thanks Aurea Martin for Disclosures: The authors declare no potential conflicts of interest. her support in English editing, proofreading, and preparing this manuscript Study identifier at ClinicalTrials.gov is NCT01492751. for submission. Int J Radiation Oncol Biol Phys, Vol. 110, No. 3, pp. 718e726, 2021 0360-3016/Ó 2021 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/). https://doi.org/10.1016/j.ijrobp.2020.12.032 Volume 110  Number 3  2021 Localized prostate cancer: 10-year follow-up 719 Methods and Materials: This was a prospective observational study of a cohort of men who received a diagnosis of clinically localized prostate cancer (clinical stage T1 or T2, low and intermediate risk group) and were treated with radical prostatec- tomy (n Z 139), brachytherapy (n Z 317), or external radiation therapy (n Z 194). Treatment decisions were jointly made by patients and physicians. Patient-reported outcome (PRO) evaluation included the Expanded Prostate Cancer Index Com- posite and Short Form-36, administered centrally by telephone interviews before and annually after treatment. The Expanded Prostate Cancer Index Composite covers urinary, bowel, sexual, and hormonal domains. To assess PRO changes over time, while accounting for correlation among repeated measures, generalized estimating equation models adjusted by propensity scores were constructed. Results: The PRO completion rate at 10 years was 85.8%. Generalized estimating equation models showed that the pattern of radical prostatectomy side effects, with substantial urinary incontinence and sexual dysfunction, remained until 10 years after treatment (standard deviation [SD], e1.1 and e1.3, respectively). Brachytherapy produced late deterioration in urinary conti- nence (SD, e0.4) and sexual function (SD, e0.9) that appeared midterm, but the differences from radical prostatectomy re- mained statistically significant at 10 years (P < .001 after adjusting by propensity score). External radiation therapy showed similar results to brachytherapy, but with bowel bother (SD, e0.3). Conclusions: Although late deterioration in radiation therapy groups attenuated differences from radical prostatectomy, rele- vant PRO differences still remained after 10 years. Our findings support that brachytherapy is the treatment option that causes the least impact on PROs; it is therefore an alternative to be considered when making evidence-based decisions on localized prostate cancer treatment. Ó 2021 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY- NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Introduction limitations: a small sample size (4210 or 612 patients in the brachytherapy group at the last follow-up), a very low response rate (13%),11 or only covering bowel impact.12 Prostate cancer is the most frequently diagnosed cancer Therefore, evidence on brachytherapy remains scarce and among men in the United States and Europe.1 Currently, controversial in the long term. most patients with prostate cancer are diagnosed in local- The aim of this study was to assess the long-term impact ized stages2 and will probably be long-term survivors.3 Randomized controlled trials of curative intention of the side effects of radical prostatectomy, brachytherapy, and external radiation therapy on patients with localized treatments for localized prostate cancer are mainly prostate cancer 10 years after treatment, using propensity restricted to the Prostate Testing for Cancer and Treatment score analyses. (ProtecT) trial.4,5 This study showed similar very high rates of survival at 10 years of follow-up4 for radical prostatec- tomy, external radiation therapy, and active monitoring; Methods and Materials however, treatments differed in side-effect patterns as measured by patient-reported outcomes (PROs).5 These This was a prospective observational study (ClinicalTrials. results have been confirmed by an analysis of the trial re- gov Identifier: NCT01492751) of a cohort of men who sults according to the treatment received and including received a diagnosis of clinically localized prostate cancer patients who refused the randomization.6 Long-term PRO and were treated with radical prostatectomy (n Z 193), follow-ups have been also reported by the Cancer of the brachytherapy (n Z 317), or external radiation therapy (n Prostate Strategic Urologic Research Endeavor (CaP- Z 194). Study details have been described elsewhere.13-15 SURE)7 trial and Prostate Cancer Outcomes Study,8 but Briefly, eligible patients were those with clinical stage T1 CaPSURE is the only study to evaluate brachytherapy. or T2, low and intermediate D’Amico risk group disease,16 A systematic review of PROs in clinically localized treated at one of the participating centers, and without prostate cancer treatments9 identified 6 studies evaluating previous prostate transurethral resection. Decisions brachytherapy with follow-ups of 5 to 6 years, and their regarding treatment were made jointly by patients and combined results indicated a reduced adverse effect profile. physicians. The study was approved by the ethics review In contrast, 5 and 10 years after treatment, patients under- boards of the participating hospitals, and written informed going brachytherapy in the CaPSURE study7 presented a consent was obtained from patients per the 2000 revision of similar risk of sexual deterioration compared with those the Helsinki Declaration. who received radical prostatectomy, a lower risk of urinary The Expanded Prostate Cancer Index Composite function worsening, and a higher risk of bowel dysfunction. (EPIC)17,18 and Short Form-36 (SF-36)19,20 were adminis- Since that review and up to the present, no other study on tered centrally by telephone interviews before treatment brachytherapy has shown relevant evidence on this treat- and during follow-up at 1, 3, 6, and 12 months after ment. Three articles have published data at 12 to 14,10 8,11 treatment during the first year and annually thereafter. EPIC and 8 to 912 years of follow-up, but they presented relevant covers urinary, bowel, sexual, and hormonal domains, with 720 Garin et al. International Journal of Radiation Oncology  Biology  Physics 50 items and scores ranging from 0 to 100.17 The SF-36 with patients who underwent radical prostatectomy pre- (version 2) generates physical and mental component senting the greatest deterioration. The use of absorbent pads summaries, scored as recommended by developers and (Fig. 1B) was null before treatment and increased 10 years standardized to have a mean of 50 and standard deviation after to 34.3% in the prostatectomy group, 6.9% in the (SD) of 10 in the general US population.19 Higher scores brachytherapy group, and 7.8% in the external radiation indicate better results in both instruments. In addition to the therapy group. Although the urinary irritativeeobstructive EPIC scores, we selected the key EPIC items already score was quite stable during follow-up (Fig. 1C), the identified in the ProtecT study5 to aid in the interpretation radiation therapy groups presented statistically lower of clinical relevance. Responses to these key items were all (worse) means than radical prostatectomy. Figure 1D shows dichotomized to show the percentage of men reporting the rate of men reporting pain or burning with urination, problems, except for the erection firmness item (percentage which was quite low during the whole follow-up (<7% for of men reporting sexual potency). all groups). The sample size was calculated to detect between-group All scores and selected key items from the EPIC sexual differences of 5 points on the urinary irritativeeobstructive domain showed statistically significant different results score of the EPIC questionnaire.13 over 10 years of follow-up across the treatment groups (Fig. 2). The sexual function score decreased (deteriorated) Statistical analysis with radical prostatectomy and external radiation therapy, but patients treated with brachytherapy showed stability Summary statistics and 95% confidence intervals (CIs) are during the first 5 years after treatment and deterioration reported by treatment group. To account for treatment se- thereafter (Fig. 2A). Before surgery, 58.1% of men reported lection bias, previously estimated propensity scores,15 erections firm enough for intercourse, but this rate fell to which had already been demonstrated to balance the dis- 9.4% after 1 year (Fig. 2B). In the brachytherapy group, tribution of possible confounders among treatment groups, this rate remained stable at around 40% from before were incorporated into multivariate analyses. treatment to the fifth year and declined thereafter. To assess PRO changes over time, while accounting for Figure 3 shows that only the key item about loose stools correlation among repeated measures, separate generalized (Fig. 3B) and the bowel summary score (Fig. 3G) presented estimating equation (GEE) models were constructed for statistically significant differences among treatments over each EPIC score or key item (included as dependent vari- time (P Z .027 and P Z .032), both indicating worse results ables). Time was included in the model as a categorical in the external radiation therapy group. Calculating the variable, and interactions between treatment and time were P-value of trend differences for fecal incontinence (Fig. 3C) considered to test differences on trends among the treat- and bloody stools (Fig. 3D) was not possible, because in ment groups. The GEE models were constructed with the some evaluations no men reported these problems. use of SAS/STAT software, version 9.4. Figure 4 shows the changes at 10 years obtained by GEE models constructed to assess treatment impact on PROs after adjusting by propensity scores. The GEE models with Results all the coefficients and P-values are available in the Table E1. Units of change on the left of the figure are EPIC At baseline, the treatment groups presented statistically scores and those on the right SDs (change: moderate Z 0.5 significant differences in age (means from 64.2 years in SD and large Z 1 SD, indicated by gray lines).21 On radical prostatectomy to 70.1 years in external radiation average, 10 years after radical prostatectomy, patients therapy; P < .001), prostate-specific antigen level (means scored 29.9 points less than before treatment (Fig. 4A). from 7.0 ng/mL in brachytherapy to 8.1 ng/mL in external This change of 30 points is greater than 1 SD (large dete- radiation therapy; P < .001), and Gleason score (means rioration). Patients treated with brachytherapy fared þ17.4 from 5.5 in brachytherapy to 6.3 in radical prostatectomy; points (95% CI, 11.0-23.9) better (positive sign) than those P < .001). After propensity score adjustment,15 the differ- who underwent radical prostatectomy (reference group) ences in baseline clinical characteristics among treatment and, similarly, those who underwent external radiation groups disappeared. therapy. Urinary continence deterioration in the brachy- The flow chart in Figure E1 shows that at 10 years after therapy and external radiation therapy groups has an near treatment, 149 of 704 participants had died, 74 missed the 0.5 SD (moderate deterioration). interview, 5 were lost to follow-up, and 476 completed the Figure 4B shows that both radiation therapy groups PRO questionnaires. The completion rate at 10 years was presented worse results than radical prostatectomy on the 85.8% (median, 10.24 years; interquartile range, 10.28-10.14). urinary irritativeeobstructive score (e3.9 points for Figure 1 shows the annual results of the EPIC urinary brachytherapy; e5.5 points for external radiation therapy). domain with unadjusted means or percentages with their Figure 4C shows substantial worsening of the EPIC sexual 95% CI and P-values adjusted by propensity scores. Uri- function score at 10 years in the 3 treatment groups, but nary incontinence (Fig. 1A) presented statistically signifi- patients treated with brachytherapy fared þ10.8 points cant timeetrend differences among the treatment groups, (95% CI, 4.8-16.8) better than those in the radical Volume 110  Number 3  2021 Localized prostate cancer: 10-year follow-up 721 A EPIC Urinary Incontinence Score C EPIC Urinary Irritative/Obstructive Score 100 100 Better Better 75 75 Mean score Mean score P < .001 50 50 P < .001 Worse Worse 25 25 0 0 Pre- 1 2 3 4 5 6 7 8 9 10 Pre- 1 2 3 4 5 6 7 8 9 10 treatment Years since treatment treatment Years since treatment B EPIC Item: ≥ 1 pad per Day D EPIC Item: pain or burning on urination Men using ≥ 1 Pads per day (%) 100 100 Men reporting problem (%) Worse Worse P = .031 P = .569 75 75 50 50 Better Better 25 25 0 0 Pre- 1 2 3 4 5 6 7 8 9 10 Pre- 1 2 3 4 5 6 7 8 9 10 treatment Years since treatment Years since treatment E EPIC Urinary Summary Score 100 Better 75 ——■—— Radical Prostatectomy Mean score P = .141 50 — ▲ — Brachytherapy ——●—— External Beam Radiation Worse 25 0 Pre- 1 2 3 4 5 6 7 8 9 10 Years since treatment Fig. 1. Ten years of annual results in the urinary domain. Effects of treatments on the urinary domain measured with the Expanded Prostate Cancer Index Composite (EPIC) instrument. EPIC urinary scores comprise several urinary items and are formed by linear transformation of raw scores, ranging from 0 to 100 (from worse to better). (A) EPIC score for urinary incontinence. (B) Results for one of the items included in that score: Percentage of men who used 1 absorbent pads per day for urinary incontinence. (C) EPIC urinary irritative-obstructive score. (D) Percentages for men who reported moderate-to- severe pain or burning on urination. (E) EPIC urinary summary score. P-values show the strength of evidence for a difference in mean response over 10 years of follow-up across the 3 groups, adjusted by propensity score. I bars represent unadjusted 95% confidence intervals. prostatectomy group. Finally, worsening was not observed in sexual function and moderate urinary incontinence that in the bowel function domain (Fig. 4E), and the small appears at midterm (5 and 7 years after treatment). External worsening presented in bowel bother (<0.5 SD) did not radiation therapy showed results similar to brachytherapy, statistically differ among treatments (Fig. 4F). but with bowel bother. Although late deterioration in the radiation therapy groups attenuated their differences from radical prostatectomy, they remained relevant at 10 years. Discussion The main limitation of this study is its observational design. The main concern regarding this design is treatment Our findings show that the main traits of radical- selection bias where, for example, brachytherapy is pref- prostatectomy side-effect patterns, with large urinary incon- erentially prescribed to patients with lower tumor risk and tinence and sexual dysfunction, remain until 10 years after surgery to younger patients.13-15 In our cohort, the pro- treatment. Brachytherapy produced substantial deterioration pensity scores balanced treatment selection bias.15 Results 722 Garin et al. International Journal of Radiation Oncology  Biology  Physics A EPIC Sexual Function Score C EPIC Sexual Bother Score 100 100 Better Better 75 75 Mean score Mean score P < .001 P = .002 50 50 Worse Worse 25 25 0 0 Pre- 1 2 3 4 5 6 7 8 9 10 Pre- 1 2 3 4 5 6 7 8 9 10 treatment Years since treatment Years since treatment B EPIC Item: Erection Firmness D EPIC Item: Problem with Erectile Dysfunction Men reporting sexual potency (%) 100 100 Men reporting problem (%) Worse Better P < .001 75 75 P < .001 50 50 Better Worse 25 25 0 0 Pre- 1 2 3 4 5 6 7 8 9 10 Pre- 1 2 3 4 5 6 7 8 9 10 treatment Years since treatment treatment Years since treatment E EPIC Sexual Summary Score 100 Better 75 Mean score P < .001 Radical Prostatectomy 50 Brachytherapy Worse 25 External Beam Radiation 0 Pre- 1 2 3 4 5 6 7 8 9 10 treatment Years since treatment Fig. 2. Ten years of annual results in the sexual domain. Effects of treatments on the sexual domain measured with the Expanded Prostate Cancer Index Composite (EPIC) instrument. EPIC sexual scores comprise several sexual items and are formed by linear transformation of raw scores, ranging from 0 to 100 (from worse to better). (A) EPIC sexual function score. (B) Results for one of the items included in this score: Percentage of men reporting erections firm enough for intercourse. (C) EPIC sexual bother score. (D) Percentages for men who reported a moderate-to-severe problem with erectile dysfunction. (E) EPIC sexual summary score. P-values show the strength of evidence for a difference in mean response over 10 years of follow-up across the 3 groups, adjusted by propensity score. I bars represent unadjusted 95% confidence intervals. adjusted by propensity scores were consistent with those and the consistency of interpretation based on the magni- from traditional models adjusted by baseline clinical vari- tude of change with previously established minimally ables (Table E2), likely allowing an estimation of the true important differences for EPIC scores.24 This is the first treatment effects.22 Second, the treatments were applied >1 study to provide robust evidence on the long-term outcomes decade ago; since then, diagnostic techniques and treat- of brachytherapy because only 85 of 684 patients treated ments for prostate cancer have evolved. with this therapy in the CaPSURE registry were evaluated The strengths of the study include the collection of data 10 years after treatment.7 Of the 317 patients treated with per the present recommendations of the International brachytherapy in our cohort, 69 died and 217 completed the Consortium for Health Outcomes Measurement for local- 10-year evaluation. ized prostate cancer, which were developed in 2015,23 the CaPSURE applied the University of California, Los use of a propensity score adjustment for comparative Angeles, Prostate Cancer Index to measure clinically effectiveness research,22 regular follow-up with high rates meaningful declines in PROs (defined as 0.5 SD) after of response (median, 83.3%; interquartile range, 85.7-74.0), treatment.7 To facilitate the comparison with CaPSURE, we Volume 110  Number 3  2021 Localized prostate cancer: 10-year follow-up 723 A EPIC Bowel Function Score E EPIC Bowel Bother Score 100 100 Better Better 75 75 Mean score Mean score P = .074 P = .073 50 50 Worse Worse 25 25 0 0 Pre- 1 2 3 4 5 6 7 8 9 10 Pre- 1 2 3 4 5 6 7 8 9 10 treatment Years since treatment Years since treatment B EPIC Item: Loose Stools F EPIC Item: Bowel Habits Men reporting ≥ half the time (%) Men reporting problem (%) 100 100 Worse Worse 75 P = .027 75 P = .794 50 50 Better Better 25 25 0 0 Pre- 1 2 3 4 5 6 7 8 9 10 Pre- 1 2 3 4 5 6 7 8 9 10 treatment Years since treatment Years since treatment C EPIC Item: Fecal Incontinence G EPIC Bowel Summary Score Men reporting ≥ one per week (%) 100 100 Worse Better 75 75 Mean score P = .032 50 50 Better Worse 25 25 0 0 Pre- 1 2 3 4 5 6 7 8 9 10 Pre- 1 2 3 4 5 6 7 8 9 10 treatment Years since treatment treatment Years since treatment D EPIC Item: Bloody Stools Men reporting ≥ half the time (%) 100 Worse 75 Radical Prostatectomy 50 Brachytherapy Better 25 External Beam Radiation 0 Pre- 1 2 3 4 5 6 7 8 9 10 treatment Years since treatment Fig. 3. Ten years of annual results in the bowel domain. Effects of the treatments on the bowel domain measured with the Expanded Prostate Cancer Index Composite (EPIC) instrument. EPIC bowel scores comprise several bowel items and are formed by linear transformation of raw scores, ranging from 0 to 100 (from worse to better). (A) EPIC score for bowel function. (B-D) Results for items included in that score as percentages of men who reported having (B) loose stools half the time or more, (C) fecal incontinence at least once per week, and (D) bloody stools half the time or more. (E) EPIC bowel bother score. (F) Percentages for men who reported a moderate-to-severe negative effect on bowel habits. (G) EPIC bowel summary score. P-values show the strength of evidence for a difference in mean response over 10 years of follow-up across the 3 groups, adjusted by propensity score. I bars represent unadjusted 95% confidence intervals. 724 Garin et al. International Journal of Radiation Oncology  Biology  Physics Radical Prostatectomy Brachytherapy External Radiotherapy A EPIC Urinary Incontinence Score B EPIC Urinary Irritative/Obstructive Score +10 +10 0 0 Reference -3.9 -5.5 (-7.6, -0.2) (-9.4, -1.7) -10 -10 .5 SD .5 SD +17.4 +18.0 1 SD -20 -20 (11.0, 23.9) (11.2, 24.8) -30 1SD -30 Reference -40 -40 -50 -50 C EPIC Sexual Function Score D EPIC Sexual Bother Score +10 +50 0 +40 1 SD -10 +30 .5 SD -20 +20 +7.3 +13.4 .5 SD Reference (-1.6, 16.2) (3.9, 22.9) -30 +10.8 1SD +10 (4.8, 16.8) +1.1 -40 Reference (-5.1, 7.4) 0 -50 -10 E EPIC Bowel Function Score F EPIC Bowel Bother Score +10 +10 +1.4 -0.8 +2.2 0 (-0.6, 3.4) (-3.3, 1.7) 0 (-0.4, 4.8) Reference -1.1 .5 SD Reference (-4.5, 2.3) .5 SD -10 1 SD -10 1 SD -20 -20 -30 -30 -40 -40 -50 -50 Fig. 4. Impact on patient-reported outcomes 10 years after treatment. On the left side of the figures, the units of change are EPIC scores and on the right side, they are standard deviations (moderate Z 0.5 SD; large Z 1 SD, indicated by gray lines). Mean changes in Expanded Prostate Cancer Index Composite instrument scores from pretreatment to 10 years of follow-up, obtained by generalized estimating equation models, are adjusted by propensity scores. Each bar is complemented with the difference compared with radical prostatectomy (mean and 95% confidence interval). similarly calculated odds ratios adjusted by propensity scores group were also similar to those reported by CaPSURE, but (Table E3) with radical prostatectomy as the reference group. not in the brachytherapy group. Because EPIC was derived from the University of California, The probability of experiencing a meaningful decline Los Angeles, Prostate Cancer Index, mainly by transforming after brachytherapy compared with radical prostatectomy at the urinary domain, only the sexual and bowel domains are 10 years after treatment was higher among the CaPSURE comparable. The results for sexual function were similar for patients than in our cohort, with odds ratios of 3.3 (95% CI, the radiation therapy groups, and differences were not sta- 1.4-7.4)7 versus 2.3 (95% CI, 0.8-6.7) in bowel function and tistically significant at 10 years in our cohort, nor in the 2.3 (95% CI, 1.0-5.2)7 versus 1.5 (95% CI, 0.5-4.0) in bowel CaPSURE study, with an odds ratio of 0.8 (95% CI, 0.3-1.7) bother. Therefore, discrepancies compared with CaPSURE and 0.9 (95% CI, 0.3-2.6)7 in brachytherapy and 1.3 (95% CI, were mainly restricted to the better results obtained in our 0.5-3.2) and 0.7 (95% CI, 0.2-3.4)7 in external radiation study for brachytherapy in the bowel domain. Although therapy. Our bowel results in the external radiation therapy CaPSURE is a nationwide prostate cancer registry, it is Volume 110  Number 3  2021 Localized prostate cancer: 10-year follow-up 725 important to highlight the low number of patients in the prostate cancer 10 years after treatment compared with brachytherapy group with results at 10 years7 ranged from 76 radical prostatectomy and external radiation therapy. Our to 85 according to the domain evaluated. Follow-up losses findings support that, despite some late deterioration in could have produced some bias in CaPSURE, which may urinary continence and sexual function around 5 years after explain the differences when compared with our study. For treatment, brachytherapy is the treatment option causing example, completion rate could be higher among patients the least impact on PROs. These results provide patients, with worse PRO results, because patients without problems clinicians, and health care planners with clear information may tend to disregard clinical encounters. to make evidence-based decisions and facilitate shared In addition, patients who underwent brachytherapy in clinical decision-making, taking into account patients’ our cohort presented a similar sexual pattern to results re- perspective. Although no single treatment can be consid- ported by Keyes et al25 in a large cohort of almost 3000 ered the preferred strategy for managing all patients, our patients followed for a median of 3.5 years and a maximum study supports brachytherapy as a possible alternative for of 14 years, when focusing on patients with full potency patients with localized prostate cancer seeking attempted before treatment and without neoadjuvant androgen depri- curative treatment while avoiding bowel effects and post- vation therapy: a gradual and continuous erectile function poning the risk for urinary incontinence and sexual decline. In our study, the percentage of men reporting dysfunction. sexual potency in this selected subsample went from 64% at 1 year after treatment to 53% after 4 years and 34% after 7 years. In the study by Keyes et al, the most significant drop in erectile function among patients with full potency before References treatment was also during the first months: approximately 65% reported full potency 1 year after treatment, approxi- 1. Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. mately 55% at 4 years, and 40% at 7 years after treat- Global cancer statistics 2018: GLOBOCAN estimates of incidence ment.25 This similarity is especially noteworthy because the and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2018;68:394-424. erectile function measurement in both studies has a 2. Shao YH, Demissie K, Shih W, et al. Contemporary risk profile of different nature. Ours is an item response in a PRO (EPIC), prostate cancer in the United States. J Natl Cancer Inst 2009;101: and Keyes et al assessed erectile function through a 1280-1283. physician-reported 3-point scale.25 3. Chou R, Croswell JM, Dana T, et al. Screening for prostate cancer: A Our long-term findings in PROs confirm those reported review of the evidence for the U.S. Preventive Services Task Force. Ann Intern Med 2011;155:762-771. by the ProtecT trial at 6 years5,6 regarding the distinct 4. Hamdy FC, Donovan JL, Lane JA, et al. 10-year outcomes after impact of localized prostate-cancer treatment side effects, monitoring, surgery, or radiotherapy for localized prostate cancer. N but similar general health. Figure E2 and Table E1 show the Engl J Med 2016;375:1415-1424. stability of the SF-36 summaries until the sixth year of 5. Donovan JL, Hamdy FC, Lane JA, et al. Patient-reported outcomes follow-up. Both studies show a very similar negative effect after monitoring, surgery, or radiotherapy for prostate cancer. N Engl J Med 2016;375:1425-1437. of prostatectomy on sexual domain (ie, only 14.7% of men 6. Neal DE, Metcalfe C, Donovan JL, et al. Ten-year mortality, disease reported sexual potency at year 6 in our study and 15% in progression, and treatment-related side effects in men with localised the ProtecT trial).6 Regarding radical prostatectomy’s prostate cancer from the ProtecT randomised controlled trial according impact on urinary continence, the percentage of patients to treatment received. Eur Urol 2020;77:320-330. who required the use of pads was lower in the ProtecT trial 7. Punnen S, Cowan JE, Chan JM, Carroll PR, Cooperberg MR. Long- term health-related quality of life after primary treatment for localized than in our cohort (21%6 vs 32% by year 6). In both studies, prostate cancer: Results from the CaPSURE registry. Eur Urol 2015; external radiation therapy is related to bowel bother and 68:600-608. function deterioration at 6 years after treatment (ie, 5.9% of 8. Resnick MJ, Koyama T, Fan KH, et al. Long-term functional out- men reporting bloody stools more than half the time in the comes after treatment for localized prostate cancer. N Engl J Med ProtecT trial6 and 3.8% in our study). 2013;368:436-445. 9. Ávila M, Patel L, López S, et al. Patient-reported outcomes after Of note, the ProtecT trial’s inclusion criteria were treatment for clinically localized prostate cancer: A systematic review limited to patients diagnosed by screening. As a conse- and meta-analysis. Cancer Treat Rev 2018;66:23-44. quence, the baseline tumor characteristics of patients in the 10. Jang JW, Drumm MR, Efstathiou JA, et al. Long-term quality of life ProtecT trial and our cohort are very different, including after definitive treatment for prostate cancer: Patient-reported out- median prostate-specific antigen levels of 4.8 versus 8.1 comes in the second posttreatment decade. Cancer Med 2017;6:1827- 1836. ng/mL, 78% versus 43.3% of patients with Gleason score 11. Ling DC, Chen KS, Benoit RM, et al. Long-term patient-reported 6, and 79% versus 58.2% with clinical T1 stage.4 rectal bleeding and bowel-related quality of life after Cs-131 pros- tate brachytherapy. Int J Radiat Oncol Biol Phys 2019;104:622-630. 12. Frank SJ, Pugh TJ, Blanchard P, et al. Prospective phase 2 trial of Conclusions permanent seed implantation prostate brachytherapy for intermediate- risk localized prostate cancer: Efficacy, toxicity, and quality of life outcomes. Int J Radiat Oncol Biol Phys 2018;100:374-382. Novel long-term results are provided on the impact of 13. Ferrer M, Suarez JF, Guedea F, et al. Health-related quality of life 2 brachytherapy adverse effects in patients with localized years after treatment with radical prostatectomy, prostate 726 Garin et al. International Journal of Radiation Oncology  Biology  Physics brachytherapy, or external beam radiotherapy in patients with clini- 19. Ware JE, Kosinski M, Dewey JE. How to Score. Version 2 of the SF-36 cally localized prostate cancer. Int J Radiat Oncol Biol Phys 2008;72: Health Survey (Standard & Acute Forms). Lincoln RI: Quality Metric 421-432. Incorporated; 2000;1-231. 14. Pardo Y, Guedea F, Aguilo F, et al. Quality-of-life impact of primary 20. Alonso J, Prieto L, Anto JM. La versión española del “SF-36 Health treatments for localized prostate cancer in patients without hormonal Survey” (Cuestionario de Salud SF-36): Un instrumento para la treatment. J Clin Oncol 2010;28:4687-4696. medida de los resultados clı́nicos. Med Clin (Barc) 1995;104:771-776. 15. Ferrer M, Guedea F, Suarez JF, et al. Quality of life impact of treat- 21. Cohen J. Statistical power analysis for the behavioral sciences. 2nd ments for localized prostate cancer: cohort study with a 5 year follow- Edition. Hillsdale, NJ: Lawrence Elrbaum Associates; 1988;1-559. up. Radiother Oncol 2013;108:306-313. 22. Desai RJ, Franklin JM. Alternative approaches for confounding 16. D’Amico AV, Whittington R, Malkowicz SB, et al. Biochemical adjustment in observational studies using weighting based on the outcome after radical prostatectomy, external beam radiation therapy, propensity score: A primer for practitioners. BMJ 2019;367:l5657. or interstitial radiation therapy for clinically localized prostate cancer. 23. Martin NE, Massey L, Stowell C, et al. Defining a standard set of JAMA 1998;280:969-974. patient-centered outcomes for men with localized prostate cancer. Eur 17. Wei JT, Dunn RL, Litwin MS, Sandler HM, Sanda MG. Devel- Urol 2015;67:460-467. opment and validation of the expanded prostate cancer index 24. Umbehr MH, Bachmann LM, Poyet C, et al. The German version of composite (EPIC) for comprehensive assessment of health-related the Expanded Prostate Cancer Index Composite (EPIC): Translation, quality of life in men with prostate cancer. Urology 2000;56: validation and minimal important difference estimation. Health Qual 899-905. Life Outcomes 2018;16:36. 18. Ferrer M, Garin O, Pera J, et al. Evaluation of the quality of life of 25. Keyes M, Pickles T, Crook J, et al. Effect of aging and long-term patients with localized prostate cancer: Validation of the Spanish erectile function after iodine-125 prostate brachytherapy. Brachy- version of the EPIC. Med Clin (Barc) 2009;132:128-135. therapy 2015;14:334-341.