Long-Term Outcomes and Genetic Predictors of Response to Metastasis-Directed Therapy Versus Observation in Oligometastatic Prostate Cancer: Analysis of STOMP and ORIOLE Trials

1The use of metastasis-directed therapy (MDT) is rapidly increasing in the setting of oligometastasis. STOMP and ORIOLE, the only two prospective trials of stereotactic ablative radiation versus observation in metachronous oligometastatic castration-sensitive prostate cancer (omCSPC), demonstrated that MDT, as compared with observation, prolong androgen deprivation–free survival and progression-free survival (PFS). Although MDT appears to be effective in omCSPC, little is known regarding the utility of biomarkers to guide treatment for these patients., Thus, the goal of this study was to report long-term outcomes of STOMP and ORIOLE and assess the ability of genomics to stratify treatment response after MDT.

2Comprehensive details regarding STOMP and ORIOLE have been reported previously., Both were prospective phase II trials enrolling individuals with omCSPC, defined as ≤ three metastases, with random assignment to observation or MDT. Active systemic therapies were not allowed with MDT. Both had institutional review board approval, and all participants provided informed consent.

3Next-generation sequencing was performed on primary prostate tumor or blood from patients enrolled. A high-risk mutational signature was defined as pathogenic somatic mutations within ATM, BRCA1/2, Rb1, and TP53 on the basis of their strong association with prostate cancer outcomes.- Pathogenic mutations were defined by commercial tests and the publicly available COSMIC tumor variant database.

4The primary end point of interest was PFS as defined previously. Additional end points included radiographic progression-free survival (rPFS) defined as development of new nodal lesions, intrapelvic or distant, bone, or visceral lesions or death. Time-to-event analysis was performed to detect differences in end points of interest using the Kaplan-Meier method, stratified by treatment (MDT v observation) or high-risk mutational status. All analyses were conducted using R version 4.1.1.

5One hundred and sixteen patients in total were included for analysis—62 from STOMP and 54 patients from ORIOLE. The CONSORT diagram is shown in Figure 1. Baseline characteristics were well balanced between groups (Table 1). The median follow-up was 52.5 months (range, 5.8-92.0 months).

6PFS was prolonged with MDT in both trials (Data Supplement, online only). The median PFS for the pooled cohort was 11.9 months (95% CI, 8.0 to 18.3) with MDT compared with 5.9 months (95% CI, 3.2 to 7.1) with observation. This corresponded with a pooled hazard ratio (HR) of 0.44 (95% CI, 0.29 to 0.66; P value <.001, Fig 2). The pooled HR for rPFS, time to castration-resistant prostate cancer, and overall survival did not differ between treatment groups (Fig 2 and Data Supplement).

7A total of 103 patients (89%) had tissue available for sequencing, and 70 patients (60%) had tissue that was successfully subjected to somatic next-generation sequencing (Fig 1). Clinical characteristics of these 70 patients are given in the Data Supplement and are similar to the entire cohort. In the entire population, the median PFS in those without a high-risk mutation was 11.9 months (95% CI, 7.0 to 16.3) compared with 5.9 months (95% CI, 5.8 to 11.1) in those with a high-risk mutation (HR, 0.57; 95% CI, 0.32 to 1.03; P =.06, Data Supplement). In those without a high-risk mutation, the median rPFS was 22.6 months (95% CI, 18.1 to 36) compared with 10.0 months (95% CI, 5.9 to 17.1) in those with a high-risk mutation (HR, 0.38; 95% CI, 0.20 to 0.17; P <.01, Data Supplement).

8We then stratified patients by both treatment arms and separately on the basis of high-risk mutational status to assess differential magnitude of benefit of MDT. Both those with and without a high-risk mutation benefited from MDT; however, a potential larger magnitude of benefit was experienced in those with a high-risk mutation. Tumors harboring a high-risk mutation treated with MDT experienced a median PFS of 7.5 months (95% CI, 5.9 to not reached [NR]) compared with a PFS of 2.8 months (95% CI, 2 to NR) with observation (HR, 0.05; 95% CI, 0.01 to 0.28; P <.01, Fig 3A). In tumors without a high-risk mutation, the median PFS with MDT was 13.4 months (95% CI, 7.0 to 36) compared with 7.0 months (95% CI, 4.0 to 15.4) with observation (HR, 0.42; 95% CI, 0.23 to 0.77; P =.01, Fig 3B) with a p-interaction of 0.12 (Data Supplement). Differences in rPFS were not seen (high-risk mutation: HR, 0.83; P =.74; no high-risk mutation: HR, 0.82, P =.58; P interaction:.40).

9Within the MDT cohort alone (Fig 3C), the PFS was 13.4 months (95% CI, 7.0 to 36.0) without a high-risk mutation, compared with 7.5 months (95% CI, 5.9 to NR) with a high-risk mutation (HR, 0.53; 95% CI, 0.25 to 1.11; P =.09). The median rPFS after MDT was 25.3 months (95% CI, 17.0 to NR) without a high-risk mutation, compared with 8.0 months (95% CI, 5.9 to NR) with a high-risk mutation (HR, 0.43; 95% CI, 0.20 to 0.95; P =.04; Fig 3D).

10MDT is rapidly emerging as a therapy in omCSPC, and this study presents long-term outcomes and genomic predictors of response to MDT in omCSPC. We report that with long-term follow-up, STOMP and ORIOLE MDT remains associated with improved PFS. Of note, the PFS beyond four years was 15%-20% with MDT regardless of mutation status, and thus, a sizable proportion of patients will experience durable response to therapy. Although more follow-up is needed, the encouraging PFS report here suggests that in appropriately selected patients, MDT without systemic therapy might be a reasonable option upfront in well-informed patients wishing to avoid side effects of androgen deprivation. However, future trials, which are planned or underway, will more rigorously study this question.

11In the quest for treatment personalization in omCSPC,, genetic biomarkers are likely to play a critical role.,,- Within our cohort, those treated with MDT without a high-risk mutation experienced the best outcomes (median PFS 13.4 months), whereas observation in those with a high-risk mutation experienced the poorest outcomes (median PFS 2.8 months). This suggests that individuals with omCSPC without a high-risk mutation might initially be treated with MDT alone and conversely highlights the need for novel treatment paradigms in those with a high-risk mutation. Importantly, although, those both with and without a high-risk mutation appeared to benefit from MDT, thus suggesting that this therapy should be offered to most, if not all, omCSPC. Ongoing trials combining systemic therapy (DART trial: ClinicalTrials.gov identifier: NCT04641078) or radiopharmaceuticals (RAVENS trial: ClinicalTrials.gov identifier: NCT04037358) might help define novel paradigms and hopefully further elucidate the role of genetic biomarkers within this population.

12There are several limitations to this report. First, the genomic analysis did not have an a priori end point and was based on small sample size. Thus, prospective validation is needed. Second, differing imaging modalities were used (conventional in ORIOLE and choline in STOMP) and with the introduction of PSMA, how we define omCSPC might change in the future. Nevertheless, these data provide a framework to investigate such questions in the future.

13In conclusion, long-term outcomes of STOMP and ORIOLE demonstrate sustained benefit to MDT over observation in omCSPC. Genomic alterations appear to have prognostic value in this patient population, suggesting that biomarkers should be evaluated in future studies to optimize patient selection.