Long-term Outcomes of Stereotactic Body Radiotherapy for Low-Risk and Intermediate-Risk Prostate Cancer

Importance Stereotactic body radiotherapy harnesses improvements in technology to allow the completion of a course of external beam radiotherapy treatment for prostate cancer in the span of 4 to 5 treatment sessions. Although mounting short-term data support this approach, long-term outcomes have been sparsely reported. Objective To assess long-term outcomes after stereotactic body radiotherapy for low-risk and intermediate-risk prostate cancer. Design, Setting, and Participants This cohort study analyzed individual patient data from 2142 men enrolled in 10 single-institution phase 2 trials and 2 multi-institutional phase 2 trials of stereotactic body radiotherapy for low-risk and intermediate-risk prostate cancer between January 1, 2000, and December 31, 2012. Statistical analysis was performed based on follow-up from January 1, 2013, to May 1, 2018. Main Outcomes and Measures The cumulative incidence of biochemical recurrence was estimated using a competing risk framework. Physician-scored genitourinary and gastrointestinal toxic event outcomes were defined per each individual study, generally by Radiation Therapy Oncology Group or Common Terminology Criteria for Adverse Events scoring systems. After central review, cumulative incidences of late grade 3 or higher toxic events were estimated using a Kaplan-Meier method. Results A total of 2142 men (mean [SD] age, 67.9 [9.5] years) were eligible for analysis, of whom 1185 (55.3%) had low-risk disease, 692 (32.3%) had favorable intermediate-risk disease, and 265 (12.4%) had unfavorable intermediate-risk disease. The median follow-up period was 6.9 years (interquartile range, 4.9-8.1 years). Seven-year cumulative rates of biochemical recurrence were 4.5% (95% CI, 3.2%-5.8%) for low-risk disease, 8.6% (95% CI, 6.2%-11.0%) for favorable intermediate-risk disease, 14.9% (95% CI, 9.5%-20.2%) for unfavorable intermediate-risk disease, and 10.2% (95% CI, 8.0%-12.5%) for all intermediate-risk disease. The crude incidence of acute grade 3 or higher genitourinary toxic events was 0.60% (n = 13) and of gastrointestinal toxic events was 0.09% (n = 2), and the 7-year cumulative incidence of late grade 3 or higher genitourinary toxic events was 2.4% (95% CI, 1.8%-3.2%) and of late grade 3 or higher gastrointestinal toxic events was 0.4% (95% CI, 0.2%-0.8%). Conclusions and Relevance In this study, stereotactic body radiotherapy for low-risk and intermediate-risk disease was associated with low rates of severe toxic events and high rates of biochemical control. These data suggest that stereotactic body radiotherapy is an appropriate definitive treatment modality for low-risk and intermediate-risk prostate cancer.


Introduction
Prostate cancer (PCa) is the leading cause of cancer treatment-related years lived with disability worldwide, reflecting the confluence of its high incidence, high cure rate, and treatment-associated morbidity. 1  Traditionally, definitive EBRT has been delivered in small fractions of 1.8 to 2.0 Gy spread across 8 to 9 weeks. Considerable preclinical and clinical data suggest that PCa specifically may exhibit an enhanced sensitivity to higher doses per fraction by virtue of a low α to β ratio (a proxy of radiosensitivity). 3 A significant implication is that hypofractionation-treating with higher doses per fraction-may allow, at the least, isoeffective oncologic results in a shorter time frame. Moderate hypofractionation (using fractions of 2.5-3.0 Gy) has been studied extensively, with 3 noninferiority randomized clinical trials demonstrating the efficacy and safety of this approach. [4][5][6] Stereotactic body radiotherapy (SBRT), or extreme hypofractionation, is a specific form of EBRT in which advanced radiotherapy techniques are used to deliver very large doses of radiation per fraction. Since its inception in 2000, several single-institution trials and 2 multi-institutional trials of SBRT for PCa have been reported, generally with median follow-up periods of 3 to 5 years. 7 A large, multi-institutional consortium report of 1100 patients with a median follow-up of 3 years presented 5-year biochemical relapse-free survival rates of 95% for patients with low-risk disease and 84% for patients with intermediate-risk disease. 8 Based on the overall favorable outcomes of these studies, the NCCN guidelines since 2014 have stated that SBRT "can be considered as an alternative to conventionally fractionated regimens at clinics with appropriate technology, physics, and clinical expertise." [2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17] Stereotactic body radiotherapy regimens have been associated with reduced cost 9,10 and less regret about undergoing treatment 11 compared with other radiotherapy regimens.
approving contribution of data to the coordinating data center (UCLA) and waiving the need for patient informed consent. Abstraction and reporting of data from the compiled studies is in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cohort studies.
Doses of SBRT ranged from 33.5 to 40.0 Gy in 4 to 5 fractions (with 1885 of 2142 patients [88.0%] receiving 5 fractions). Treatments were delivered on consecutive days, every other day, or once a week per individual protocol specifications. Stereotactic body radiotherapy was delivered with either a robotic arm-mounted linear accelerator (CyberKnife; Accuray Inc) or a gantry-mounted linear accelerator. Specific treatment planning and delivery information is presented in eTable 1 in the Supplement. Patients were followed up with clinical evaluations performed and prostate-specific antigen level measured every 3 to 6 months for the first 2 years followed by every 6 to 12 months for the next 3 years.

End Points
The cumulative incidence of biochemical recurrence (BCR) was the primary disease control end point, with BCR determined using the Phoenix definition of a prostate-specific antigen level of 2 ng/mL or more higher than the lowest post-SBRT value. 28 Secondary disease control end points included the cumulative incidence of distant metastases (DMs), BCR-free survival, and overall

Statistical Analysis
Statistical analysis was performed based on follow-up from January 1, 2013, through May 1, 2018. A competing risk framework was used to estimate the cumulative incidences of BCR and DMs, with death by any cause as a competing risk. 33 Kaplan-Meier analysis was performed to estimate BCR-free survival and overall survival, with time to event set using the final day of SBRT as the starting point.
This framework was also used to estimate the cumulative incidence of grade 3 or higher GU or GI toxic events. Univariate and multivariable Fine-Gray competing risk and Cox proportional hazards regression models were developed to assess the association between time to BCR and equivalent dose in 2-Gy fractions (EQD 2 ) (calculated assuming an α to β ratio of 1.5), age, clinical T stage, ln (initial prostate-specific antigen level), use of androgen deprivation therapy, and Gleason grade group. The Cox proportional hazards regression models were used as the hazard ratios derived from such an analysis are easily interpretable, whereas competing risk regressions were used as it was thought that a competing risk framework more accurately modeled the relevant incidence of BCR. The proportional hazard test based on scaled Schoenfeld residuals was used to examine proportional hazard assumption. Univariate and multivariable logistic regression approaches were used to evaluate the association between the toxic event outcomes and EQD 2 , daily vs every-other-day fractionation, and treatment platform. To account for residual ecological bias, presuming patients enrolled in the same institutional study were likely to be more similar to each other than to patients enrolled in other institutional studies, the trial in which patients were enrolled was also included as a stratification factor in both the Cox proportional hazards regression and Fine-Gray competing risk models and as a random effect in the logistic regression models. Analyses were completed using R, version 3.3.2. 34 All P values were from 2-tailed tests, and results were deemed statistically significant at P < .05.

Results
Patient and treatment characteristics are presented in Table 2  Crude rates and cumulative incidence estimates of acute grade 2 and grade 3 or higher GU and GI toxic events and late grade 2 and grade 3 or higher GU and GI toxic events are shown in Table 3;

JAMA Network Open | Oncology
narrative descriptions of grade 3 or higher toxic events are provided in eTable 6 in the Supplement.
One participating institution, Georgetown, did not provide data for toxic events of grade 2 or lower, as toxic events were not a tracked end point in that institutional study. Thirteen patients experienced acute grade 3 or higher GU toxic events (crude incidence, 0.60%), and 2 others experienced acute grade 3 or higher GI toxic events (crude incidence, 0.09%). The most common acute grade 3 or higher GU toxic event was urinary frequency, which accounted for 8 events (61.5%). Among these, 2 patients (25.0%) had a prior cystoscopy and 3 (37.5%) developed late grade 3 or higher urinary frequency. The CTCAE, version 4.0 scale does not provide scores higher than 2 for urinary frequency; thus, these events were originally scored via RTOG and CTCAE, version 3.0, representing urinary frequency of once or more per hour or necessitating catheter placement.
Forty-two patients experienced late grade 3 GU toxic events, with 3 patients experiencing 2 separate late grade 3 GU toxic events and 1 patient subsequently developing a grade 4 GU toxic event, for an estimated 7-year cumulative incidence of grade 3 or higher GU toxic events of 2.4% (95% CI, 1.8%-3.2%) ( Table 3). The median interval from SBRT to development of a grade 3 GU toxic event was 27 months (IQR, 18-61 months). Nine events (19.6%) occurred after 5 years of follow-up, and 2 events (4.3%) occurred after 6 years of follow-up. Of the 4 patients with grade 3 urinary frequency, 2 had undergone cystoscopy prior to SBRT and had experienced acute grade 3 urinary frequency. One patient with grade 3 hematuria had a papillary bladder tumor at 18 months. This tumor was ultimately thought to be the cause of the hematuria, but the time course was such that it was not deemed a secondary malignant neoplasm. The 1 grade 4 GU toxic event was an episode of hemorrhagic cystourethritis treated with multiple endoscopic treatments and bladder irrigation; this event occurred 1 month after dilation of a urethral stricture, which itself was detected 52 months after SBRT.
Six patients experienced late grade 3 GI toxic events and 1 patient developed a grade 4 GI toxic event, for an estimated 7-year cumulative incidence of grade 3 or higher GI toxic events of 0.4% (95% CI, 0.2%-0.8%) ( Table 3). The median interval to development of a late grade 3 or higher toxic event was 31 months (IQR, 24-38 months). All events occurred within 5 years of SBRT. Of the 6 patients with grade 3 GI toxic events, 1 had a history of ulcerative colitis and a known dysplastic polyp outside the radiation field in the midsigmoid colon. This patient had hematochezia 24 months after SBRT, which was ultimately attributed to a colonic adenocarcinoma arising in the polyp. The patient with a grade 4 toxic event had an anal fistula 9 months after SBRT. This fistula arose in the context of prior diverticulitis, and an attempted surgical repair failed to correct the fistula. As the patient had occasional (monthly) perianal discharge, he deferred definitive repair with colostomy.
On multivariable logistic regression, only acute composite RTOG and CTCAE grade 3 or higher toxic events (GI or GU) were associated with late composite GI or GU RTOG and CTCAE grade 3 or higher toxic events (odds ratio, 19.42; 95% CI, 5.14-73.42; P = .008), while EQD 2 , fractionation, and treatment platform were not (eTables 7 and 8 in the Supplement). Both fractionation and acute composite RTOG and CTCAE grade 2 or higher toxic events (GI or GU) were associated with late composite GI or GU RTOG and CTCAE grade 2 or higher toxic events, with an odds ratio of 0.38 (95% CI, 0.16-0.89; P = .03) for fractionation and an odds ratio of 3.15 (95% CI, 1.96-5.07; P = .006) for acute composite RTOG and CTCAE grade 2 or higher toxic events.

Discussion
In this individual patient data analysis of 2142 patients, SBRT was associated with a favorable longterm disease control and safety profile for low-risk and intermediate-risk PCa. Biochemical control was excellent, with 7-year BCR rates of less than 10% for low-risk and favorable intermediate-risk disease and just 15% for unfavorable intermediate-risk disease. Severe toxic events were rare in both the acute and late settings, with a 7-year cumulative incidence of late grade 3 or higher GU and GI toxic events of 2.4% and 0.4%. To our knowledge, these data represent the largest series of patients treated with SBRT for low-risk and intermediate-risk disease with long-term follow-up, and the cohort studied includes a substantial number of patients with follow-up extending beyond 9 years.
Our findings suggest that the major trepidation with SBRT-a risk of severe late toxic events-is not reported that significantly more patients treated with SBRT had claim codes indicative of treatmentrelated GU toxic events when compared with patients receiving intensity-modulated radiotherapy (IMRT), particularly with respect to diagnostic procedures for incontinence and obstruction. They also found significantly more claims indicative of diagnosis of or procedure to correct or investigate urethritis, urethral strictures, and obstruction among patients treated with SBRT. A potential limitation of such a population-based analysis, however, is that inferences are made upon amalgamating diagnostic and therapeutic intervention codes, leading to an overclassification of toxic events. A more recent analysis of claims in the MarketScan database found no significant differences in composite GU, GI, or erectile dysfunction toxic events between patients treated with SBRT or IMRT. 9 The investigators did report a significant increase in claims pertaining to obstruction or retention (SBRT, 21% vs IMRT, 15%) and urinary fistula (SBRT, 1% vs IMRT, 0.1%) at 2 years. However, only 43 patients receiving SBRT were eligible for these end points at 2 years compared with 619 patients receiving IMRT; thus, these findings must be viewed with extreme caution given the very small numbers of patients included. On the other hand, the present study includes 2142 patients with individual patient data, and no patients had a urinary fistula and only 1 patient had a GI-related fistula (crude incidence, 0.05%). The most robust factors associated with late toxic events appear to be development of acute toxic events and fractionation, with treatment every other day associated with lower incidences of grade 2 or higher toxic events. are thus better suited to conclude the true noninferiority of this approach. These trials include HYPO-RT-PC (ISRCTN45905321), PACE (NCT01584258), HEAT (NCT01794403), and NRG GU005 (NCT03367702). The preliminary results of the HYPO-RT-PC trial, which compared 78 Gy in 39 fractions with an extremely hypofractionated regimen of 42.7 Gy in 7 fractions, have been presented in abstract form. 42 With a median follow-up of nearly 5 years, the proportion of patients free of biochemical or clinical failure at 5 years was 83.8% with conventional fractionation and 83.7% with extreme hypofractionation, suggesting noninferiority. There was no difference in modified late grade 2 or higher RTOG toxic events at 4 years with regard to either urinary or bowel axes. Although it is nonrandomized, the current consortium study provides prospective data supporting the safety of SBRT at even longer follow-up than in the HYPO-RT-PC study and also presents outcomes after more typical SBRT regimens than the 7-fraction course used in that trial. Given the low-risk and intermediate-risk nature of the disease that patients in this study had, it is likely that more clinical events (DMs or PCa-specific death)

JAMA Network Open | Oncology
would be seen over time; however, as indicated in eTable 9 in the Supplement, the follow-up in the present study is among the longest for a series exploring radiotherapy modalities in this context.
Another limitation is that the present study includes only physician-scored toxic events, even though patient-reported outcomes may be more relevant, particularly for patients with low-risk and intermediate-risk PCa who have an excellent prognosis. Nonetheless, much of the aversion to SBRT is based on a preconceived risk of severe grade 3 or higher toxic events. These data suggest that this risk is minimal and commensurate with the risk after other, more widely accepted treatment modalities. Finally, the treatment protocols were heterogeneous, and there was no central review or quality assurance process.

Conclusions
We present prospective, multi-institutional data with long-term follow-up demonstrating that SBRT for low-risk and intermediate-risk PCa is associated with a favorable safety and disease control profile. No unexpected increase in late toxic events or compromise in disease control was identified with longer-term follow-up (ie, beyond 5 years) than reported in prior studies. Although randomized trials comparing conventional or moderate hypofractionation regimens and SBRT are under way, the favorable outcomes described herein strongly suggest that SBRT be considered a standard option for treating low-risk and intermediate-risk PCa.