Long-Term Adverse Effects and Complications After Prostate Cancer Treatment

Joseph M Unger; Cathee Till; Catherine M Tangen; Dawn L Hershman; Phyllis J Goodman; Michael LeBlanc; William E Barlow; Riha Vaidya; Lori M Minasian; Howard L Parnes; Ian M Thompson, Jr

doi:10.1001/jamaoncol.2024.4397

. 2024 Nov 7;10(12):1654–1662. doi: 10.1001/jamaoncol.2024.4397

Long-Term Adverse Effects and Complications After Prostate Cancer Treatment

Joseph M Unger ^1,^✉, Cathee Till ¹, Catherine M Tangen ¹, Dawn L Hershman ², Phyllis J Goodman ¹, Michael LeBlanc ¹, William E Barlow ¹, Riha Vaidya ¹, Lori M Minasian ³, Howard L Parnes ³, Ian M Thompson Jr ⁴

¹SWOG Statistics and Data Management Center, Fred Hutchinson Cancer Center, Seattle, Washington

²Department of Medicine, Columbia University Medical Center, New York, New York

³Division of Cancer Prevention, US National Cancer Institute, Bethesda, Maryland

⁴CHRISTUS Santa Rosa Health System and The University of Texas Health Science Center at San Antonio

Accepted for Publication: July 21, 2024.

Published Online: November 7, 2024. doi:10.1001/jamaoncol.2024.4397

^✉

Corresponding Author: Joseph M. Unger, PhD, SWOG Statistics and Data Management Center, Fred Hutchinson Cancer Center, 1100 Fairview Ave N, M3-C102, PO Box 19024, Seattle, WA 98109-1024 (junger@fredhutch.org).

Author Contributions: Dr Unger and Ms Till had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Unger, Tangen, Hershman, Barlow, Minasian, Thompson.

Acquisition, analysis, or interpretation of data: Unger, Till, Tangen, Hershman, Goodman, LeBlanc, Vaidya, Parnes, Thompson.

Drafting of the manuscript: Unger, Hershman, Minasian, Thompson.

Critical review of the manuscript for important intellectual content: Unger, Till, Tangen, Hershman, Goodman, LeBlanc, Barlow, Vaidya, Parnes, Thompson.

Statistical analysis: Unger, Till, Tangen, LeBlanc.

Obtained funding: Unger, Tangen, Hershman, Thompson.

Administrative, technical, or material support: LeBlanc, Thompson.

Supervision: Unger, Tangen, Hershman, Goodman, Thompson.

Conflict of Interest Disclosures: Dr Unger reported consulting fees from AstraZeneca and Loxo/Lilly outside the submitted work. Dr Barlow reported grants from US National Cancer Institute during the conduct of the study. Dr Vaidya reported employment with Flatiron Health at the time of manuscript submission and review. No other disclosures were reported.

Funding/Support: This research was supported in part by the US National Institutes of Health, National Cancer Institute (grants CA189974, CA182883, and legacy grant CA37429); and in part by The Hope Foundation for Cancer Research in support of infrastructure, data analysis, and trial design within the SWOG Cancer Research Network, at the SWOG Statistics and Data Management Center.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the US National Institutes of Health. The content is solely the responsibility of the authors and does not necessarily represent the official views of the study sponsor.

Data Sharing Statement: See Supplement 2.

^✉

Corresponding author.

PMCID: PMC11544550 PMID: 39509091

Key Points

Question

What long-term complications of treatment for prostate cancer (PCA) are experienced by treated patients compared with a general population of older males?

Findings

This cohort study, including 3946 patients with PCA, used a novel data linkage between 2 large PCA prevention clinical trials and Medicare claims to determine PCA treatment was associated with substantially higher rates of complications 12 years after treatment. Moreover, participants treated with radiotherapy had a 3-fold increased risk of bladder cancer and a 100-fold increased risk of radiation cystitis and radiation proctitis.

Meaning

After accounting for age-related symptoms and diseases, PCA treatment was significantly associated with higher rates of complications 12 years after treatment, a finding that highlights the importance of patient counseling and provides a rationale for pursuing opportunities for cancer prevention.

Abstract

Importance

Due to the often indolent nature of prostate cancer (PCA), treatment decisions must weigh the risks and benefits of cancer control with those of treatment-associated morbidities.

Objective

To characterize long-term treatment-related adverse effects and complications in patients treated for PCA compared to a general population of older males.

Design, Setting, and Participants

This cohort study used a novel approach linking data from 2 large PCA prevention clinical trials (the Prostate Cancer Prevention Trial and the Selenium and Vitamin-E Cancer Prevention Trial) with Medicare claims records. This analysis included patients with PCA who had been treated with prostatectomy or radiotherapy compared with an untreated control group. Multivariable Cox regression was used, with a time-varying covariate for the occurrence of PCA treatment, adjusted for age, race, and year of time-at-risk initiation, and stratified by study and intervention arm. Data analyses were performed from September 21, 2022, to March 18, 2024.

Exposure

Prostatectomy and radiotherapy occurring after a PCA diagnosis, identified from trial data or Medicare claims records.

Main Outcomes and Measures

Ten potential PCA treatment-related complications identified from Medicare claims data.

Results

The study sample comprised 29 196 participants (mean [SD] age at time-at-risk initiation, 68.7 [4.8] years). Of these, 3946 participants had PCA, among whom 655 were treated with prostatectomy and 1056 with radiotherapy. The 12-year hazard risk of urinary or sexual complications was 7.23 times greater for those with prostatectomy (95% CI, 5.96-8.78; P < .001) and 2.76 times greater for radiotherapy (95% CI, 2.26-3.37; P < .001) compared to untreated participants. Moreover, among participants treated with radiotherapy, there was a nearly 3-fold greater hazard risk of bladder cancer than in the untreated (hazard ratio [HR], 2.78; 95% CI, 1.92-4.02; P < .001), as well as an approximately 100-fold increased hazard risk of radiation-specific outcomes including radiation cystitis (HR, 131.47; 95% CI, 52.48-329.35; P < .001) and radiation proctitis (HR, 87.91; 95% CI, 48.12-160.61; P < .001). The incidence per 1000 person-years of any 1 of the 10 treatment-related complications was 124.26 for prostatectomy, 62.15 for radiotherapy, and 23.61 for untreated participants.

Conclusions and Relevance

This cohort study found that, even after accounting for age-related symptoms and disease, PCA treatment was associated with higher rates of complications in the 12 years after treatment. Given the uncertain benefit of PCA treatment for most patients, these findings highlight the importance of patient counseling before PCA screening and treatment and provide a rationale for pursuing opportunities for cancer prevention.

This cohort study assesses the long-term adverse effects and complications of prostate cancer treatment in treated patients compared with a general population of older males.

Introduction

Prostate cancer (PCA) is ubiquitous, affecting more than half of male individuals by their eighth decade and 12.5% during their lifetime; by comparison, lifetime risk of PCA death is 2.5%.^1,2,3 The advent of prostate-specific antigen (PSA) screening to detect PCA raised the specter of cancer overdetection and overtreatment, confirmed by 2 screening trials.^4,5 In the US trial, with frequent population PSA testing (ie, PSA testing in the no-screening group), no reduction of PCA mortality was seen.⁴ By comparison, the European study found a 21% relative reduction in PCA mortality after 11 years, at which point 1055 men must be invited for screening and 37 cases of prostate cancer detected to prevent 1 PCA death.⁵

The principle of early PCA detection presumes cancer treatment, often with radiation therapy or surgery. Given that tumors often progress slowly, screening may prevent death years in the future whereas treatment complications often present soon after therapy. Treatment complications include urinary (eg, incontinence, obstruction), sexual (eg, loss of ejaculation, impotence), radiation cystitis and proctitis, and secondary cancers (ie, colorectal, bladder). It is the trade-off of benefit vs risk of treatment that informs patient decision-making. Low-grade tumors are increasingly managed with active surveillance, including periodic PSAs, magnetic resonance imaging, and prostate biopsies; treatment is initiated if a higher-risk tumor is detected. Although surveillance avoids initial treatment complications, 50% or more of patients are ultimately treated.

Informed patient decision-making should include a complete understanding of treatment adverse effects. Most previous reports of adverse effects are limited to selected domains (eg, urinary, sexual), present single or limited institutional outcomes, or do not provide a comparable control group. This study aimed to characterize long-term treatment-related adverse effects and complications after prostatectomy or radiation therapy in the follow-up of 2 large PCA prevention trials.^6,7

Methods

This cohort study was reviewed and approved by the Institutional Review Board of Cancer Research and Biostatistics (Seattle, Washington). Written informed consent was previously obtained from participants. Demographic information was self-reported. We followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cohort studies.

Data Sources

This cohort study used data from 2 large PCA prevention trials. The Prostate Cancer Prevention Trial (PCPT)^6,7 randomly assigned 18 880 participants from 1994 to 1997 to finasteride or placebo daily for 7 years. Participants were aged 55 years or older with normal digital rectal examination and prostate-specific antigen (PSA) of 3.0 ng/mL or less (to convert to µg/L, multiply by 1.0). The Selenium and Vitamin E Cancer Prevention Trial (SELECT)⁸ randomly assigned 34 887 participants from 2001 to 2004 to daily supplementation with selenium, vitamin E, both, or placebo. Included participants were 55 years or older (Black men were eligible at age 50 years) and had a normal digital rectal examination and PSA of 4.0 ng/mL or less.

Study Design and Participant Characteristics

Data Linkage to Medicare Claims

We linked the PCPT and SELECT trial records^6,7,8 with Medicare Parts A and B claims for each participant using their social security number, sex, and date of birth. Medicare physician supplier Part B, hospital outpatient, and hospital inpatient claims from 1999 to 2019 were used. Unless Medicare was received for disability, participants who were younger than 65 years at study entry were included if they aged into the Medicare claims cohort.

Definition of PCA Diagnosis

We used Medicare claims to augment the identification of PCA among trial participants.⁹ A Medicare-defined diagnosis of PCA was identified as any hospital claim or 2 or more physician or outpatient claims at least 30 days apart but less than 6 months apart (eTable 1 in Supplement 1).

Definitions of PCA Treatments, Adverse Effects, and Complications

PCA treatments (prostatectomy and radiation therapy; eTable 1 in Supplement 1) and 10 potential treatment-related functional outcomes and complications (ie, urethral stricture, placement of artificial urinary sphincter, placement of penile prosthesis, urinary incontinence, erectile dysfunction, radiation cystitis, radiation proctitis, bladder cancer, bladder cancer followed by cystectomy, and rectal cancer; eTable 2 in Supplement 1) were identified as any hospital claim or at least 2 physician or outpatient claims at least 30 days apart for prostatectomy, radiotherapy, or any of their potential complications. Participants may have received the given treatment alone or as their first of a series of treatments for PCA. Because independent events using claims could not be reliably delineated, only first events were assessed.

Definitions of Treated and Untreated Participants

This analysis included individuals with at least 12 months of continuous Medicare coverage at any time. All participants began time-at-risk after 6 months of continuous coverage to allow for sufficient time to identify and rule out preexisting conditions; thus, the analysis emphasized the occurrence of incident (new) events. This time included participant time without PCA and participant time with PCA that was untreated, and represents control time against which time after receipt of prostatectomy or radiation therapy was compared; men in this group were referred to as untreated participants. The inclusion of men with untreated PCA in control time was based on the presumption that complications were related to treatment (not PCA diagnosis), which aimed to improve the generalizability of the cohort. Twelve months of coverage time was required to ensure sufficient time to identify complications: 6 months to identify potentially preexisting complications before time-at-risk initiation and at least 6 months after time-at-risk initiation.

Treated participants were patients with PCA treatment any time after PCA diagnosis or less than 1 year before PCA diagnosis (to allow for late specification of the PCA diagnosis claim). These participants entered the treated group at the time of the first claim for PCA treatment. PCA treatment was defined as prostatectomy (either through Medicare claims or reported on-trial to diagnose PCA) or radiotherapy, through Medicare claims only. Patients with evidence of more than 1 treatment were included, with treatment time indexed by the first treatment received. PCA-treated patients were included if they had at least 6 months of continuous Medicare coverage after PCA treatment to ensure sufficient time to identify treatment-related complications.

Men diagnosed with PCA before enrolling in Medicare, with no evidence of PCA treatment while insured by Medicare, were removed to minimize the risk of misclassification due to potential unobserved treatments that would have designated them as treated participants. Participants must have had no Health Maintenance Organization (HMO) coverage concurrent to Medicare Parts A and B because Medicare claims for HMO-insured individuals are not available for research. Patients who had undergone a prostatectomy reported on-trial and used to diagnose PCA while under Medicare observation were considered for inclusion as prostatectomy-treated participants; otherwise, all participants who reported a prostatectomy or cystoprostatectomy during the trial were removed. Date of death was collected from trial records when available, or from Medicare records.

Statistical Analysis

Cox regression was used to calculate the risk of complications after PCA treatment. Confounding was controlled for through the inclusion of a cohort of male participants without PCA or PCA treatment, with additional covariate adjustment for age at time-at-risk initiation (<70 vs ≥70 years), self-reported race (Black or not Black), year of time-at-risk initiation (≤2005 or >2005), and stratified by study and treatment arm.¹⁰ Race was used as a covariate because Black men are more likely to be diagnosed with and die of prostate cancer.² The index time (time zero) was defined as day 180 of continuous Medicare coverage. Applying a time-varying covariate strategy, all participants entered as controls (ie, untreated) to reduce bias related to males being older at prostate cancer diagnosis and treatment. Patients with evidence of PCA treatment following a diagnosis entered the treated group on the treatment date. Time ended at the first date of a complication or censoring. In Cox regression, censor date for each treatment-complication analysis was the last date of Medicare coverage or first claim for a PCA treatment not currently being considered, including androgen deprivation therapy, or date of death as a competing risk.¹¹ If a complication event had occurred before PCA treatment, then only untreated control time (prior to the event) was included. Participants whose Medicare claims or trial data indicated a complication before index time were removed from that specific analysis.

Hazard ratios (HRs) and 95% CIs from Cox regression after 2, 6, and 12 years were generated by specifying a change point.¹² The 2-year time point was chosen to align with estimates derived in prior studies for purposes of comparison,^{13,14,15,16,17,18} and also represents the approximate time at which differences in hazard risk between groups was maximal. The 12-year estimate represents the point at which long-term estimates could reliably be derived in our dataset using Medicare claims, with a 6-year midpoint.

Standardized incidence (SI) for treated and untreated participants was defined as the number of first events per 1000 person-years of follow-up and was calculated at 2, 6, and 12 years. For each time point-specific SI calculation, events after the time point were censored and time-at-risk was truncated at the time point, separately for each participant’s treated and untreated time. Risk ratios (RRs) were calculated as the ratio of SI for treated (ie, prostatectomy, radiotherapy) to untreated participants.

Cumulative incidence estimates for any urinary or sexual complication and for any complication were derived as the number of new events divided by the total number of individuals at risk, estimated over time, with deaths treated as competing risks through censoring. Because finasteride affects urinary and sexual function, we assessed urinary and sexual complications excluding participants from the PCPT finasteride arm. We removed patients with PCA and no treatment to evaluate secondary cancers because cancer rates may differ in those with and without prior PCA.

Statistical tests were 2-tailed and P values < .05 were considered statistically significant. Data analyses were performed from September 21, 2022, to March 18, 2024, using SAS, version 9.4 (SAS Institute) and R, version 3.6.1 (The R Foundation for Statistical Computing).

Results

Cohort Description

Of the 51 961 total PCPT and/or SELECT participants, 29 196 (mean [SD] age at time-at-risk initiation, 68.7 [4.8] years; 2453 [8.4%] Black and 26 743 [91.6%] of other racial groups) had records that were linked to Medicare and Medicare coverage of 12 months or more (Figure 1). Of these, 3946 participants were diagnosed with PCA, among whom 655 and 1056 participants were treated with prostatectomy or radiotherapy, respectively; thus, 2235 participants with PCA did not receive treatment (Table). The median (IQR) follow-up duration for the entire cohort was 10.2 (6.3-15.8) years, totaling 312 882 person-years of follow-up. Of these, 6855 person-years occurred after prostatectomy with a median (IQR) of 10.5 (5.9-15.3) years, and 8949 person-years occurred after radiotherapy with a median (IQR) of 8.5 (4.1-12.3) years.

Figure 1. — DOB indicates date of birth; PCPT, Prostate Cancer Prevention Trial; SELECT, Selenium and Vitamin E Cancer Prevention Trial; and SSN, social security number.

Table. Characteristics of Participants With Prostate Cancer (PCA) and Complications After Prostatectomy and Radiation Therapy.

Characteristic	Participants, No. (%)
Characteristic	Total (PCPT and SELECT)	With PCA, treated and untreated	Prostatectomy treated^a	Radiotherapy treated^a	Excluded (PCPT and SELECT)
Participants, No.	29 196	3946	655	1056	22 765
Age at time-at-risk initiation, y^b
Median (IQR)	65.8 (65.4-70.9)	66.5 (65.4-70.8)	65.4 (65.4-67.7)	66.8 (65.4-70.7)	61.2 (57.0-67.0)
<70	20 770 (71.1)	2786 (70.6)	574 (87.6)	762 (72.2)	19 252 (84.6)
≥70	8426 (28.9)	1160 (29.4)	81 (12.4)	294 (27.8)	3513 (15.4)
Racial group
Black	2439 (8.4)	366 (9.3)	45 (6.9)	80 (7.6)	2866 (12.6)
All other	26 757 (91.6)	3580 (90.7)	610 (93.1)	976 (92.4)	19 899 (87.4)
Year of time-at-risk initiation
≤2005	19 877 (68.1)	3183 (80.7)	510 (77.9)	907 (85.9)	NA
>2005	9319 (31.9)	763 (19.3)	145 (22.1)	149 (14.1)	NA
PCPT (treated vs untreated)
Finasteride	5990 (50.3)	1009 (45.6)	162 (43.2)	321 (48.4)	2508 (48.6)
Placebo	5926 (49.7)	1203 (54.4)	213 (56.8)	342 (51.6)	2650 (51.4)
SELECT (treated vs untreated)
Selenium only	4335 (25.1)	468 (27.0)	86 (30.7)	98 (24.9)	4417 (25.1)
Selenium and vitamin E	4373 (25.3)	402 (23.2)	52 (18.6)	93 (23.7)	4329 (24.6)
Placebo	4291 (24.8)	416 (24.0)	68 (24.3)	91 (23.2)	4405 (25.0)
Vitamin E only	4281 (24.8)	448 (25.8)	74 (26.4)	111 (28.2)	4456 (25.3)
Gleason score, grade
High (≥7)	NA	512 (13.0)	105 (16.0)	160 (15.2)	NA
Low/intermediate (≤6)	NA	1176 (29.8)	256 (39.1)	393 (37.2)	NA
Missing data	NA	2258 (57.2)	294 (44.9)	503 (47.6)	NA

Open in a new tab

Abbreviations: NA, not applicable; PCPT, Prostate Cancer Prevention Trial; SELECT, Selenium and Vitamin E Cancer Prevention Trial.

^{^a}

After PCA diagnosis (first or only diagnosis).

^{^b}

For PCPT and SELECT participants not included in present analysis, age at trial baseline was used.

Prostatectomy Adverse Effects and Complications

The 12-year hazard risk of any complication was 6.57 times greater (95% CI, 5.39-8.01; P < .001) for prostatectomy compared to untreated participants, with an SI of 124.3 vs 23.6, respectively (Figure 2). The difference in hazard risk of any complications for prostatectomy was greatest early in the period, including 13.20 (95% CI, 8.70-20.04; P < .001) at 2-year, and decreased over time as untreated participants experienced aging-related events (eTable 3 in Supplement 1).

Figure 2. — Boxes show hazard ratios; horizontal lines indicate 95% CIs; and vertical line is the line of equal hazard (boxes to the right indicate greater hazard risk of a complication among treated participants and to the left, lower hazard risk). Incidence estimates as events per 1000 years.

At 12 years, there was a 7-fold increase in hazard risk of any urinary or sexual complication (HR, 7.23; 95% CI, 5.96-8.78; P < .001), with an SI of 124.07 for prostatectomy vs 22.08 for untreated participants (Figure 2). The 12-year hazard risk for urinary incontinence, erectile dysfunction, and urethral stricture were all 5-fold or greater for prostatectomy compared to untreated participants (Figure 2). For any urinary complication, the 2-, 6-, and 12-year SI (RRs) were 307.62 vs 28.22 (10.90), 162.28 vs 24.01 (6.76), and 124.07 vs 22.08 (5.60), respectively (eTables 3 and 4 in Supplement 1). Surgical implantation of penile prosthesis (SI = 4.40) and artificial urinary sphincter (SI = 4.76) were uncommon but were much more frequent after prostatectomy, with 12-year RRs of 10.20 and 109.59, respectively (Figure 3).

Figure 3. — For each category, the line of equal risk, indicating no difference between groups in risk, is shown. For urinary or sexual complications, the axis is interrupted to accommodate the large risk ratio. AUS indicates artificial urinary sphincter.

Radiotherapy Adverse Effects and Complications

The 12-year hazard risk of any complication was 3.04 times greater (95% CI, 2.51-3.67; P < .001) for radiotherapy-treated participants compared to those untreated, with SI = 62.15 for radiotherapy and 23.61 for untreated participants (Figure 2). The 12-year hazard risk of any urinary or sexual complication was significantly greater for the radiotherapy-treated compared to the untreated participants (HR, 2.76; 95% CI, 2.26-3.37; P < .001; Figure 2). For urethral stricture, the 12-year hazard risk was more than 6-fold greater for radiotherapy (HR, 6.49; 95% CI, 4.71-8.94; P < .001). The 12-year hazard risks of erectile dysfunction, urinary incontinence, and placement of penile prosthesis were more than 2-fold greater with radiation-treated compared to untreated participants; differences in hazard risks were higher early in follow-up, including 6-fold greater for erectile dysfunction at 2-year (HR, 5.98; 95% CI, 3.30-10.84; P < .001; eTable 3 in Supplement 1).

Radiation-specific outcomes were rare but almost entirely associated with radiotherapy, with 12-year SI of radiation cystitis of 4.70 for treated and 0.03 for untreated participants (corresponding HR, 131.47; 95% CI, 52.48-329.35; P < .001), and a 12-year SI of radiation proctitis of 5.88 for treated and 0.07 among untreated participants (HR, 87.91; 95% CI, 48.12-160.61; P < .001; Figure 2). RRs were 148.75 and 82.82, respectively (Figure 3).

Secondary Cancers

At 12 years, participants who had undergone radiotherapy had an almost 3-fold increase in hazard risk of bladder cancer than those untreated (HR, 2.78; 95% CI, 1.92-4.02; P < .001), including more than a 3-fold increase in hazard risk of bladder cancer followed by cystectomy (HR, 3.56; 95% CI, 1.40-9.02; P = .008; Figure 2; Figure 3). No differences in rectal cancer were observed with either treatment.

Additional Analyses

The cumulative incidence of any complication for prostatectomy at 12 years was 61.4% (Figure 4). Urinary and sexual complications were the primary complications after prostatectomy of the 10 complications we analyzed. In fact, the 12-year estimate of any urinary or sexual complication was almost the same as for any complication at 60.9%. In contrast, for radiotherapy, the 12-year cumulative incidence for any urinary or sexual complication was 36.1% and for any complication was 41.7%, illustrating the contributions of events in other categories (eg, secondary cancers, radiation cystitis, radiation proctitis). A sensitivity analysis including participants receiving a single treatment (eg, prostatectomy only) had similar findings (eTable 5 in Supplement 1). Results for urinary or sexual complications after prostatectomy and radiotherapy were similar when participants in the PCPT finasteride arm were removed (eTable 6 in Supplement 1). The comparison of prostatectomy- and radiotherapy-treated participants to only controls (without PCA) generated qualitatively similar findings for secondary cancers (eTable 7 in Supplement 1).

Discussion

Complications after PCA treatment are common and are associated with a myriad of harms^19,20; inadequate understanding of these risks may lead to patients regretting their treatment decisions.²¹ Accurate characterization of treatment risks is vital for informed decision-making, especially because complications can occur early after treatment whereas benefits may accrue years later. Using the linkage between the 2 large clinical trials and Medicare claims, we found that PCA treatment was associated with substantial risks of complications, decreases in function, and potentially life-threatening conditions (eg, bladder cancer).

Previous studies of PCA treatment complications used varied data sources (eTable 8 in Supplement 1).^{13,14,16,17,18,22,23,24,25,26,27,28,29,30,31,32,33} Some assessed single-institution reviews^25,30,32 or included only a single surgeon,^28,29 limiting generalizability. Others used registries or included men diagnosed 25 years ago^17,18; others evaluated only a single complication.¹⁶ Recent studies have used the SEER (Surveillance, Epidemiology, and End Results program) or linked SEER to Medicare data, only comparing active treatments.^22,24,26,27 Without an untreated control group, those studies likely overstated the risk of complications associated with PCA treatment. To account for this, some relied on patients with PCA who were assigned to active surveillance as controls.¹³ This approach is biased because patients not undergoing treatment may not do so because of a greater risk of complications.²²

Our novel approach assessed risk of complications and functional outcomes due to PCA treatment over the long term, which limited bias. The control population included participants without PCA and another group with untreated PCA, all prospectively enrolled and extensively characterized in 2 large trials. These features enabled a comparison of rates of age-dependent functional changes, such as erectile dysfunction, which increase with age.

Even after accounting for the underlying rate of conditions among controls, urinary and sexual complications were common. The 12-year hazard risk of urinary or sexual complications was more than 7-fold greater among participants who had undergone prostatectomy and nearly 3-fold greater for those treated with radiotherapy. Beyond urinary and sexual complications, due to its large sample size and long duration of follow-up, our study also permitted evaluation of conditions that are rare, understudied, and almost only observed after PCA treatment. By 12 years after prostatectomy, we found 4.76 cases of placement of an artificial urinary sphincter per 1000 years of follow-up, with virtually no sphincter operations among untreated participants. This observation is important because this procedure is most commonly performed for severe incontinence. At 12 years, penile prosthesis implantation, a procedure that is most commonly performed for medication-refractory erectile dysfunction, occurred 4.40 and 1.50 times for each 1000 years of follow-up in participants who had undergone prostatectomy or radiotherapy, respectively, compared to 0.43 times in untreated participants. Radiation cystitis and radiation proctitis, conditions with significant potential morbidity, occurred approximately 5 times per 1000 years of follow-up among participants who had undergone radiotherapy, with virtually no cases among those untreated.^34,35

Lastly, bladder cancer has been identified as a sequela of radiotherapy for PCA.^36,37,38 A recent meta-analysis found a 39% increased risk after radiotherapy.³⁹ Studies included in the meta-analysis primarily used patients with PCA treated with active surveillance or undergoing prostatectomy as controls, likely underestimating bladder cancer risk associated with radiotherapy. By comparison, we found nearly a 3-fold increased risk compared to untreated participants after 12 years. This is an important observation because bladder cancer after radiation is commonly of higher grade and potentially more lethal.^40,41 Accordingly, we found 3.5 times increase in hazard risk of bladder cancer followed by cystectomy for participants who underwent radiation.

Limitations

This study was limited because it did not account for multiple comparisons, although most differences would be statistically significant using any multiple adjustments approach. Also, claims data are subject to misclassification. Claims data also select sequelae sufficiently symptomatic to prompt a health care visit and diagnostic coding, which may focus on more severe sequelae, but may underreport complications (eg, erectile dysfunction) that are not reported to a physician. Cardiac outcomes were not included in the analysis due to the likelihood of differential referral to radiation or surgery based on preexisting cardiac conditions. We also did not differentiate among strategies of prostatectomy (ie, robotic-assisted surgery) or radiotherapy (eg, brachytherapy), which may have resulted in different patterns of complications. Furthermore, the study cohort comprised men enrolled to large, randomized prevention trials with inclusion and exclusion criteria that determined eligibility; this could limit generalizability of the incidence estimates. Finally, given the possibility of confounding by unknown factors, our study cannot attribute differences in the risks between treated and untreated participants to PCA treatment alone.

Conclusions

This cohort study relied on prospectively collected data from 2 large trials linked to a novel claims data resource to evaluate the impact of treatment for a PCA diagnosis on long-term adverse effects and complications. We found that, after accounting for baseline population rates, most patients with PCA undergoing treatment experience complications associated with worse quality of life and/or new health risks. The magnitude of these risks, compared with the relatively small benefit found by randomized clinical trials of PCA screening and treatment, should be explicitly reflected in national cancer screening and treatment guidelines and be integral to shared decision-making with patients before initiation of PSA screening, biopsy, or PCA treatment. No national organization currently includes quantitative information on PCA treatment-related risks and benefits (eTable 9 in Supplement 1). We recommend it be included to encourage truly informed decision-making.

Supplement 1.

eTable 1. ICD9, ICD10 and HCPCS codes for prostate cancer treatments

eTable 2. ICD9, ICD10 and HCPCS codes for complications of prostate cancer treatments

eTable 3. Hazard ratios at 2, 6, and 12 years for each complication

eTable 4. Incidence per 1,000 years and Risk Ratios for Each Complication over Time

eTable 5. Hazard ratios at 2, 6, and 12 years for each complication, removing participants with evidence of more than one treatment

eTable 6. Hazard ratios at 2, 6, and 12 years for Urinary and Sexual Complications Following Prostatectomy and Radiation Therapy with Participants on the PCPT finasteride arm removed

eTable 7. Hazard ratios at 2, 6, and 12 years for Secondary Cancers Following Prostatectomy and Radiation Therapy Compared to Non-Cancer Controls Only

eTable 8. Selected literature examining complications of prostate cancer (PC) treatment

eTable 9. U.S. National Guidelines and Patient Information Materials for Treatment of Localized Prostate Cancer; Focus on Information Related to Treatment Risk

eReferences

jamaoncol-e244397-s001.pdf^{(478KB, pdf)}

Supplement 2.

Data Sharing Statement

jamaoncol-e244397-s002.pdf^{(14.2KB, pdf)}

References

1.Bell KJ, Del Mar C, Wright G, Dickinson J, Glasziou P. Prevalence of incidental prostate cancer: a systematic review of autopsy studies. Int J Cancer. 2015;137(7):1749-1757. doi: 10.1002/ijc.29538 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Siegel RL, Miller KD, Fuchs HE, Jemal A. Cancer statistics, 2022. CA Cancer J Clin. 2022;72(1):7-33. doi: 10.3322/caac.21708 [DOI] [PubMed] [Google Scholar]
3.Grossman DC, Curry SJ, Owens DK, et al. ; US Preventive Services Task Force . Screening for prostate cancer: US Preventive Services Task Force Recommendation Statement. JAMA. 2018;319(18):1901-1913. doi: 10.1001/jama.2018.3710 [DOI] [PubMed] [Google Scholar]
4.Andriole GL, Crawford ED, Grubb RL III, et al. ; PLCO Project Team . Mortality results from a randomized prostate-cancer screening trial. N Engl J Med. 2009;360(13):1310-1319. doi: 10.1056/NEJMoa0810696 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Schröder FH, Hugosson J, Roobol MJ, et al. ; ERSPC Investigators . Prostate-cancer mortality at 11 years of follow-up. N Engl J Med. 2012;366(11):981-990. doi: 10.1056/NEJMoa1113135 [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Feigl P, Blumenstein B, Thompson I, et al. Design of the Prostate Cancer Prevention Trial (PCPT). Control Clin Trials. 1995;16(3):150-163. doi: 10.1016/0197-2456(94)00XXX-M [DOI] [PubMed] [Google Scholar]
7.Thompson IM, Goodman PJ, Tangen CM, et al. The influence of finasteride on the development of prostate cancer. N Engl J Med. 2003;349(3):215-224. doi: 10.1056/NEJMoa030660 [DOI] [PubMed] [Google Scholar]
8.Lippman SM, Klein EA, Goodman PJ, et al. Effect of selenium and vitamin E on risk of prostate cancer and other cancers: the Selenium and Vitamin E Cancer Prevention Trial (SELECT). JAMA. 2009;301(1):39-51. doi: 10.1001/jama.2008.864 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Unger JM, Hershman DL, Till C, et al. Using Medicare claims to examine long-term prostate cancer risk of finasteride in the Prostate Cancer Prevention Trial. J Natl Cancer Inst. 2018;110(11):1208-1215. doi: 10.1093/jnci/djy035 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Cox DR. Regression models and life-tables. J R Stat Soc B. 1972;34:187-220. doi: 10.1111/j.2517-6161.1972.tb00899.x [DOI] [Google Scholar]
11.Kalbfleisch JD, Prentice RL. The Statistical Analysis of Failure Time Data. John Wiley & Sons; 2011. [Google Scholar]
12.Liang KY, Self SG, Liu XH. The Cox proportional hazards model with change point: an epidemiologic application. Biometrics. 1990;46(3):783-793. doi: 10.2307/2532096 [DOI] [PubMed] [Google Scholar]
13.Barocas DA, Alvarez J, Resnick MJ, et al. Association between radiation therapy, surgery, or observation for localized prostate cancer and patient-reported outcomes after 3 years. JAMA. 2017;317(11):1126-1140. doi: 10.1001/jama.2017.1704 [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Benoit RM, Naslund MJ, Cohen JK. Complications after radical retropubic prostatectomy in the Medicare population. Urology. 2000;56(1):116-120. doi: 10.1016/S0090-4295(00)00539-2 [DOI] [PubMed] [Google Scholar]
15.Chen AB, D’Amico AV, Neville BA, Earle CC. Patient and treatment factors associated with complications after prostate brachytherapy. J Clin Oncol. 2006;24(33):5298-5304. doi: 10.1200/JCO.2006.07.9954 [DOI] [PubMed] [Google Scholar]
16.Elliott SP, Meng MV, Elkin EP, McAninch JW, Duchane J, Carroll PR; CaPSURE Investigators . Incidence of urethral stricture after primary treatment for prostate cancer: data From CaPSURE. J Urol. 2007;178(2):529-534. doi: 10.1016/j.juro.2007.03.126 [DOI] [PubMed] [Google Scholar]
17.Potosky AL, Legler J, Albertsen PC, et al. Health outcomes after prostatectomy or radiotherapy for prostate cancer: results from the Prostate Cancer Outcomes Study. J Natl Cancer Inst. 2000;92(19):1582-1592. doi: 10.1093/jnci/92.19.1582 [DOI] [PubMed] [Google Scholar]
18.Stanford JL, Feng Z, Hamilton AS, et al. Urinary and sexual function after radical prostatectomy for clinically localized prostate cancer: the Prostate Cancer Outcomes Study. JAMA. 2000;283(3):354-360. doi: 10.1001/jama.283.3.354 [DOI] [PubMed] [Google Scholar]
19.Aizer AA, Gu X, Chen MH, et al. Cost implications and complications of overtreatment of low-risk prostate cancer in the United States. J Natl Compr Canc Netw. 2015;13(1):61-68. doi: 10.6004/jnccn.2015.0009 [DOI] [PubMed] [Google Scholar]
20.Krupski TL, Foley KA, Baser O, Long S, Macarios D, Litwin MS. Health care cost associated with prostate cancer, androgen deprivation therapy and bone complications. J Urol. 2007;178(4 Pt 1):1423-1428. doi: 10.1016/j.juro.2007.05.135 [DOI] [PubMed] [Google Scholar]
21.Aronowitz R, Greene JA. Contingent knowledge and looping effects: a 66-year-old man with PSA-detected prostate cancer and regrets. N Engl J Med. 2019;381(12):1093-1096. doi: 10.1056/NEJMp1811521 [DOI] [PubMed] [Google Scholar]
22.Al Hussein Al Awamlh B, Wallis CJD, Penson DF, et al. Functional outcomes after localized prostate cancer treatment. JAMA. 2024;331(4):302-317. doi: 10.1001/jama.2023.26491 [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Alibhai SMH, Leach M, Tomlinson G, et al. 30-day mortality and major complications after radical prostatectomy: influence of age and comorbidity. J Natl Cancer Inst. 2005;97(20):1525-1532. doi: 10.1093/jnci/dji313 [DOI] [PubMed] [Google Scholar]
24.Bekelman JE, Mitra N, Efstathiou J, et al. Outcomes after intensity-modulated versus conformal radiotherapy in older men with nonmetastatic prostate cancer. Int J Radiat Oncol Biol Phys. 2011;81(4):e325-e334. doi: 10.1016/j.ijrobp.2011.02.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Gotto GT, Yunis LH, Vora K, Eastham JA, Scardino PT, Rabbani F. Impact of prior prostate radiation on complications after radical prostatectomy. J Urol. 2010;184(1):136-142. doi: 10.1016/j.juro.2010.03.031 [DOI] [PubMed] [Google Scholar]
26.Hu JC, Gu X, Lipsitz SR, et al. Comparative effectiveness of minimally invasive vs open radical prostatectomy. JAMA. 2009;302(14):1557-1564. doi: 10.1001/jama.2009.1451 [DOI] [PubMed] [Google Scholar]
27.Kim S, Shen S, Moore DF, et al. Late gastrointestinal toxicities following radiation therapy for prostate cancer. Eur Urol. 2011;60(5):908-916. doi: 10.1016/j.eururo.2011.05.052 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Kundu SD, Roehl KA, Eggener SE, Antenor JA, Han M, Catalona WJ. Potency, continence and complications in 3,477 consecutive radical retropubic prostatectomies. J Urol. 2004;172(6 Pt 1):2227-2231. doi: 10.1097/01.ju.0000145222.94455.73 [DOI] [PubMed] [Google Scholar]
29.Lepor H, Nieder AM, Ferrandino MN. Intraoperative and postoperative complications of radical retropubic prostatectomy in a consecutive series of 1,000 cases. J Urol. 2001;166(5):1729-1733. doi: 10.1016/S0022-5347(05)65662-5 [DOI] [PubMed] [Google Scholar]
30.Lerner SE, Blute ML, Lieber MM, et al. Morbidity of contemporary radical retropubic prostatectomy for localized prostate cancer. Oncology (Williston Park). 1995;9:379-382. [PubMed] [Google Scholar]
31.Resnick MJ, Koyama T, Fan KH, et al. Long-term functional outcomes after treatment for localized prostate cancer. N Engl J Med. 2013;368(5):436-445. doi: 10.1056/NEJMoa1209978 [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Schultheiss TE, Lee WR, Hunt MA, Hanlon AL, Peter RS, Hanks GE. Late GI and GU complications in the treatment of prostate cancer. Int J Radiat Oncol Biol Phys. 1997;37(1):3-11. doi: 10.1016/S0360-3016(96)00468-3 [DOI] [PubMed] [Google Scholar]
33.Sheets NC, Goldin GH, Meyer AM, et al. Intensity-modulated radiation therapy, proton therapy, or conformal radiation therapy and morbidity and disease control in localized prostate cancer. JAMA. 2012;307(15):1611-1620. doi: 10.1001/jama.2012.460 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Smit SG, Heyns CF. Management of radiation cystitis. Nat Rev Urol. 2010;7(4):206-214. doi: 10.1038/nrurol.2010.23 [DOI] [PubMed] [Google Scholar]
35.Do NL, Nagle D, Poylin VY. Radiation proctitis: current strategies in management. Gastroenterol Res Pract. 2011;2011:917941. doi: 10.1155/2011/917941 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Neugut AI, Ahsan H, Robinson E, Ennis RD. Bladder carcinoma and other second malignancies after radiotherapy for prostate carcinoma. Cancer. 1997;79(8):1600-1604. doi: 10.1002/(SICI)1097-0142(19970415)79:8<1600::AID-CNCR24>3.0.CO;2-0 [DOI] [PubMed] [Google Scholar]
37.Nieder AM, Porter MP, Soloway MS. Radiation therapy for prostate cancer increases subsequent risk of bladder and rectal cancer: a population based cohort study. J Urol. 2008;180(5):2005-2009. doi: 10.1016/j.juro.2008.07.038 [DOI] [PubMed] [Google Scholar]
38.Brenner DJ, Curtis RE, Hall EJ, Ron E. Second malignancies in prostate carcinoma patients after radiotherapy compared with surgery. Cancer. 2000;88(2):398-406. doi: 10.1002/(SICI)1097-0142(20000115)88:2<398::AID-CNCR22>3.0.CO;2-V [DOI] [PubMed] [Google Scholar]
39.Wallis CJ, Mahar AL, Choo R, et al. Second malignancies after radiotherapy for prostate cancer: systematic review and meta-analysis. BMJ. 2016;352:i851. doi: 10.1136/bmj.i851 [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Au S, Keyes M, Black P, Villamil CF, Tavassoli P. Clinical and pathological characteristics of bladder cancer in post brachytherapy patients. Pathol Res Pract. 2020;216(3):152822. doi: 10.1016/j.prp.2020.152822 [DOI] [PubMed] [Google Scholar]
41.Keehn A, Ludmir E, Taylor J, Rabbani F. Incidence of bladder cancer after radiation for prostate cancer as a function of time and radiation modality. World J Urol. 2017;35(5):713-720. doi: 10.1007/s00345-016-1934-z [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials