Prostate Radiotherapy for Metastatic Hormone-sensitive Prostate Cancer: A [44_TD$DIFF] STOPCAP Systematic Review and Meta-analysis

Background: Many trials are evaluating therapies for men with metastatic hormone- sensitive prostate cancer (mHSPC). Objective: To systematically review trials of prostate radiotherapy. Design, setting, and of hormone therapy, volume of disease, prostate radiotherapy dose, and toxicity. We sought overall results for survival, PFS, FFS, and biochemical progression according to our prespeci ﬁ ed de ﬁ nitions, as well as results for survival, PFS, and FFS by patient subgroups (age, varied by metastatic burden — a pattern consistent across trials and outcome measures, including survival ( < 5, (cid:5) 5; interaction HR = 1.47, 95% CI 1.11 – 1.94, p = 0.007). There was 7% improvement in 3-yr survival in men with fewer than ﬁ ve bone metastases. Conclusions:


1.
Introduction Randomised controlled trials have evaluated, or are currently evaluating, promising therapies for metastatic hormone-sensitive prostate cancer (mHSPC), including prostate radiotherapy [1,2]. Systematic reviews of these trial results can help determine effective treatments, but are usually planned after most trials have reported and focus on published results. Consequently, their design and conduct can be influenced by existing results, and they may not include enough data to produce reliable findings.
A new framework for adaptive meta-analysis (FAME) [5] defines review methods prospectively, prior to trial results being published. It also helps anticipate emerging trial results and identify the earliest opportunity for reliable meta-analysis [6,7]. Results of the key trials investigating prostate radiotherapy were due, which could provide sufficient evidence about its effects.

Patients and methods
We aimed to assess the effects of adding prostate radiotherapy to androgen deprivation therapy (ADT) in men with mHSPC. We prespecified methods in a protocol prior to the results of eligible trials being known (PROSPERO registration: CRD42018096108).

Treatment comparisons
Some eligible trials are assessing the effects of prostate radiotherapy in conjunction with other agents. To allow for the possibility of an interaction between these different treatments, we wanted to review the effects of prostate radiotherapy, via two comparisons.

Framework for adaptive meta-analysis
We applied key FAME principles to: (1) start the systematic review process whilst trials are ongoing or yet to report, (2) search comprehensively for all eligible trials, (3) liaise with trial teams to develop a detailed picture of these trials, (4) predict when sufficient results will be available for reliable meta-analysis, (5) conduct meta-analysis and interpret results taking account of any unavailable data, and (6) assess the value of updating.

Trial eligibility
Randomised controlled trials were eligible if they randomised men with mHSPC, starting or responding to first-line hormone therapy, and compared prostate radiotherapy plus ADT versus ADT. Trials including additional agents (eg, docetaxel, abiraterone) were also eligible, provided that the same additional agents were used in both treatment and control arms. Trials were ineligible if they included men who had stopped responding to first-line hormone therapy, those with castrate-refractory prostate cancer, or those in whom radiotherapy was administered to metastases.

Trial identification
We We assessed the risk of bias [10] of included trials based on sequence generation, allocation concealment, completeness of outcome data, and selective outcome reporting, using information obtained from trial protocols, manuscripts, or investigators.
However, as only the STAMPEDE results for 367 men randomised to receive docetaxel as part of the standard of care were anticipated in 2018, a definitive meta-analysis of comparison B is planned later.    assess statistical heterogeneity [15]. We aimed to summarise grade 1-5 acute bladder and bowel toxicities in the radiotherapy arm.

Measuring treatment effects
We also planned analyses of the effects of prostate radiotherapy on

Network meta-analysis of current therapies
Previously, we compared the relative effects of recent therapies combined with ADT in a network meta-analysis [18]. If the survival results of comparison A were deemed sufficiently reliable, we would include them in an updated network meta-analysis.

Characteristics of eligible trials
Our searches retrieved 19,830 unique records, and seven mHSPC trials that were potentially eligible for comparison A. Four trials were excluded: two because radiotherapy was administered to metastases as well as the prostate ( [19] and NCT02913859), one because men did not receive ADT (NCT02680587), and one because surgery or radiotherapy was allowed as local treatment (NCT01751438), leaving three eligible trials ( Supplementary Fig. 1).  (Table 1).
All men were classified as having newly diagnosed mHSPC and were receiving long-term ADT for the first time, mostly LHRH-based therapy (>99%). Across the two trials, men were aged similarly (HORRAD, median age 67 yr; STAMPEDE, median age 68 yr); most had World Health Organisation/ECOG performance status 0 (HORRAD, 84%; STAMPEDE, 71%) and a Gleason sum score of !8 (HORRAD, 66%; STAMPEDE, 79%). All men recruited to the HORRAD trial had bone metastases, while 89% of men in the STAMPEDE trial had bone metastases with (5%) or without (84%) visceral metastases. As the HORRAD trial did not collect data on nonbone metastases, it was not possible to use the CHAARTED [8] or LATITUDE [9] trial definitions of disease volume. However, the STAMPEDE team were able to reclassify patients according to the HORRAD definition [20] (low volume: Gleason sum score <9, fewer than five bone lesions, and PSA 142; the HORRAD median). Few men in the HORRAD trial (17%) had low-volume disease compared with around half of the men in the STAMPEDE trial (45%).
Based on randomisation sequence generation, allocation concealment, completeness of outcome data, and selective outcome reporting, both trials were judged to be at a low risk of bias (Supplementary Table 2).

Overall treatment effects
Survival results are based on all 2126 men (969 deaths) from HORRAD and STAMPEDE. Overall, there was no evidence that the addition of prostate radiotherapy to ADT improved survival (HR = 0.92, 95% confidence interval [CI] 0.81-1.04, p = 0.195; heterogeneity chi-square = 0.08, degree of freedom = 1, p = 0.78; Fig. 1). The PFS results based on all men (1305 events) also provided no clear evidence that, overall, prostate radiotherapy extended PFS (HR = 0.94, 95% CI 0.84-1.05, p = 0.238; Fig. 1). Although, in the HORRAD trial, biochemical progression was defined as the time between diagnosis and a PSA increase after the initiation of ADT of >50% of the lowest PSA value after the start of treatment (with a minimum of 1 ng/ml), and in the STAMPEDE trial as a rise above the lowest PSA within 24 wk of enrolment of 50% to at least 4 ng/ml, we considered them sufficiently compatible to combine. Based on all men and 1533 events, we observed a highly statistically significant benefit of prostate radiotherapy (HR = 0.74, 95% CI 0.67-0.82, p = 0.94 Â 10 À8 ; Fig. 1) in biochemical progression, which translates to an absolute improvement of 11 (7)(8)(9)(10)(11)(12)(13)(14)% at 3 yr from 25% to 36%. Toxicity results are not yet available for HORRAD. Based on the results collected from STAMPEDE, 4% of men who received prostate radiotherapy had severe acute bladder toxicity, and 1% had severe acute bowel toxicity (RTOG scale). Reported STAMPEDE results showed that 4% of men had severe late effects.
Based on shorter median follow-up (21.3 mo) and only 367 men, the STAMPEDE survival results for men planned for docetaxel (HR = 0.81, 95% CI 0.49-1.34, p = 0.379) were broadly similar to the results of comparison A.
[ ( F i g . _ 1 ) T D $ F I G ] Fig. 1 -Effect of adding prostate radiotherapy to ADT on (A) survival, (B) progression-free survival, (C) biochemical progression, and (D) failure-free survival in men with mHSPC. Each filled square denotes the HR for that trial comparison, with the horizontal lines showing the 95% confidence interval (CI). The size of the square is directly proportional to the amount of information contributed by a trial. The diamond represents a (fixed-effect) meta-analysis of the trial HRs, with the centre of this diamond indicating the HR and the extremities the 95% CI. ADT = androgen deprivation therapy; HR = hazard ratio; RT = radiotherapy.

Network meta-analysis of all current therapies
As the effect of prostate radiotherapy on survival was influenced by metastatic burden, this would need to be accounted for in the planned update of the network meta-analysis of recent therapies for mHSPC [18], and such methods are still in development [21]. It would also require the collection and analysis of individual participant data (IPD) from all trials.

Summary of results
Prostate radiotherapy did not clearly improve survival or PFS in unselected men with mHSPC. However, there was a clear difference in the effect [ 6 2 _ T D $ D I F F ] by metastatic burden on survival, with an absolute improvement of 7% in 3-yr survival in men who had four or fewer bone metastases.
There was no evidence that the effect of prostate radiotherapy on survival varied by other patient or disease characteristics. Prostate radiotherapy improved 3-yr biochemical progression and FFS by $10% in unselected men, but the size of effect varied by metastatic burden.

Strengths
Based on 90% of all men randomised to prostate radiotherapy plus ADT versus ADT, we have shown that the effect of prostate radiotherapy on survival varies by metastatic burden. Despite different recruitment periods, radiotherapy [ ( F i g . _ 2 ) T D $ F I G ] Fig. 2 -Effect of adding prostate radiotherapy to ADT on survival by patient age at randomisation, performance status, clinical T stage, and Gleason sum score. Each filled square denotes the HR for each subgroup of men defined by, age at randomisation, performance status, clinical T stage, and Gleason sum score within each trial, with the horizontal lines showing the 95% confidence interval (CI). The size of the square is directly proportional to the amount of information contributed by a subgroup. Each filled circle denotes the HR for the interaction between the effect of radiotherapy and these subgroups for each trial, with the horizontal lines showing the 95% CI. The size of each circle is directly proportional to the amount of information contributed by a trial. The open circle represents a (fixed-effect) meta-analysis of the interaction HRs, with the horizontal line showing the 95% CI. ADT = androgen deprivation therapy; HR = hazard ratio; RT = radiotherapy.
approaches, and proportions of men with low and high metastatic burdens, this pattern was remarkably and reassuringly consistent across trials and outcomes. As a prospectively designed FAME review, all methods were published (unless otherwise specified) before trial results were known. This includes the preplanned subgroup analyses by metastatic burden, albeit that we had to collapse subgroup categories. We were able to anticipate when the results of STAMPEDE and HORRAD were due, allowing us to align the review with publication of their results [11,12]. By obtaining unpublished trial results, we could harmonise outcome and subgroup definitions and conduct additional analyses. Hence, we have been able to provide a more timely, reliable, and thorough synthesis of the effects of prostate radiotherapy than is usually possible with summary results [5].

Limitations
Only two of the relevant trials are included, but the 234 eligible men from the [ 6 3 _ T D $ D I F F ] recently completed PEACE-1 trial (Table 1) represent just 10% of the total, and so its results are unlikely to materially affect our findings. While the STAMPEDE trial heavily influences the results, the HORRAD trial has longer follow-up and adds considerable weight to the analyses of all outcome measures (23-28%; Fig. 1), including the survival analysis by metastatic burden (25%; Fig. 3). Therefore, until internationally agreed, optimised definitions of the metastatic burden and the oligometastatic state are determined [22], the number of bone metastases alone could help identify groups of men who might benefit from prostate radiotherapy.

Context
Results from PEACE-1 in combination with the current results of STAMPEDE will provide the first substantive evidence of how prostate radiotherapy works in conjunction with docetaxel and/or abiraterone. A new trial (SWOG S1802, NCT03678025) of standard systemic therapy with or without definitive treatment (surgery or radiotherapy) may also contribute to this comparison, if it stratifies by definitive treatment. Three trials (TROMBONE [ISRCTN 15704862], g-RAMPP [NCT02454543], and SIMCAP [NCT03456843]) are investigating whether radical prostatectomy offers an alternative to radical radiotherapy in this setting, and two trials [19] (NCT[ 6 4 _ T D $ D I F F ] 02913859) and a new STAMPEDE arm are evaluating the effects of administering radiotherapy to metastatic sites as well as the prostate.

Implications
The collection of IPD from relevant trials could help determine which men with mHSPC benefit more or less from prostate radiotherapy and what the optimal definition of metastatic burden might be. A comprehensive repository [ ( F i g . _ 3 ) T D $ F I G ] of IPD from all modern mHSPC trials (STOPCAP M1 IPD repository) is being established with funding from MRC and Prostate Cancer UK to tackle these and other important clinical uncertainties (http://www.stopcapm1.org/). However, applying the review findings in settings where newer imaging techniques (eg, prostate-specific membrane antigen positron emission tomography) are available could be problematic, as men currently classed as having a low metastatic burden may be reclassified as having a greater number of metastases. Questions also remain regarding the timing and optimal dose of radiotherapy.

Conclusions
The addition of prostate radiotherapy to ADT should be considered for men with mHSPC who have four or fewer bone metastases.