Indirect Comparison of PROpel and TALAPRO‐2 Trials Using the Shiny Method: A Comparative Analysis in Metastatic Castration‐Resistant Prostate Cancer

The role of poly (ADP-ribose) polymerase (PARP) inhibitors in metastatic castration-resistant prostate cancer (mCRPC) has drawn increasing attention, particularly in patients harboring homologous recombination repair (HRR) gene alterations [1]. Co-inhibition of androgen receptor and PARP may confer synergistic therapeutic benefits, potentially broadening the efficacy of PARP inhibitors. The phase 3 PROpel and TALAPRO-2 trials demonstrated that the combination of PARP inhibitors with androgen receptor signaling inhibitors (ARSIs) significantly improved radiographic progression-free survival (rPFS) compared with ARSI alone as first-line treatment for patients with mCRPC, irrespective of HRR gene alteration status [2, 3]. However, no direct comparative trials have been conducted between the olaparib–abiraterone combination therapy in the PROpel trial and the talazoparib–enzalutamide combination therapy in the TALAPRO-2 trial. Therefore, in this study, we performed a comparative analysis of these two ARSI–PARP inhibitor combination therapies by extracting rPFS data from the PROpel and TALAPRO-2 trials using the Shiny method, which uses Kaplan–Meier (KM) curves for visualization of temporal survival patterns, and reconstructed individual patient data (rIPD) [4, 5]. Although the overall patient characteristics and adverse events in the TALAPRO-2 and PROpel trials were generally comparable (Table S1), the TALAPRO-2 cohort had a lower proportion of patients with HRR mutations (21% vs. 28%) and BRCA1 (7% vs. 11.8%) and BRCA2 (8% vs. 9.6%) mutations compared with the PROpel cohort. The definition of rPFS was identical in both trials, using RECIST v1.1 and PCWG3 criteria. In the PROpel trial, follow-up assessments were conducted every 8 weeks until week 24 and then every 12 weeks, whereas in the TALAPRO-2 trial, they were conducted every 8 weeks until week 25 and then every 12 weeks. We constructed rIPD from the published KM curves of each trial using the Shiny method with the "IPD from KM" package [6]. The validity of the reconstructed KM curves was confirmed by comparison with the original data. Upon comparing the rIPD data, in the overall population, the TALAPRO-2 talazoparib plus enzalutamide group demonstrated longer rPFS than the PROpel olaparib plus abiraterone group (HR = 1.26, 95% CI: 1.00–1.58, p = 0.049). In the comparison of placebo groups, the TALAPRO-2 enzalutamide group also outperformed the PROpel abiraterone group (HR = 1.30, 95% CI: 1.06–1.58, p = 0.011) (Figure 1A). The use of different ARSIs in TALAPRO-2 and PROpel presents a significant issue in interpreting the results. The existence of significant differences in rPFS between the placebo groups may limit the validity of indirect comparisons for the analysis results in the overall population. In contrast, subgroup analyses of HRR-positive and HRR-negative populations showed no significant difference in rPFS between the two treatment groups (HRR-positive: HR = 1.07, 95% CI: 0.68–1.68, p = 0.777, Figure 1B; HRR-negative: HR = 1.24, 95% CI: 0.90–1.70, p = 0.184, Figure 1C), and no significant differences were observed between placebo groups. These results suggest that the differences in the analysis of the overall population may reflect fundamental differences between the study populations of the TALAPRO-2 cohort and the PROpel cohort, particularly the lower proportion of patients with HRR and BRCA mutations in TALAPRO-2. Talazoparib has been shown to have significantly higher trapping potency compared with other PARP inhibitors [7]. Although clear comparisons are difficult in this study due to differences in baseline patient characteristics, future research on the antitumor effects of talazoparib and other PARP inhibitors is anticipated. This study has several limitations. First, survival data were reconstructed from published KM curves using the Shiny method, yielding approximated individual patient data without the ability to perform multivariate analyses. Consequently, adjustments for baseline differences or detailed subgroup analyses were not feasible. Second, in the PROpel study, HRR mutation testing was predefined, whereas the TALAPRO-2 study included patients with unknown HRR mutation status. These differences in study design should be taken into account when interpreting and comparing the results. Although these findings require careful interpretation as they are based on indirect comparisons, this analysis suggests that rPFS results for talazoparib and olaparib in combination with ARSI are generally comparable when stratified by HRR status. Although definitive conclusions will require direct comparison trials or matched individual patient data analyses in mCRPC populations, this analysis provides important interim insights into combination treatment strategies with ARSI and PARP inhibitors. Yosuke Yasuda: data curation, conceptualization, writing – original draft. Zijie Wang: data curation, writing – original draft. Soichiro Yoshida: supervision, writing – review and editing, methodology, conceptualization. Wei Chen: writing – review and editing, methodology. Hajime Tanaka: writing – review and editing. Yasuhisa Fujii: supervision, writing – review and editing. Approval of the research protocol by an Institutional Review Board: The current study is an indirect comparison of previously reported randomized clinical trials. No formal ethics approval was required in this case. Informed Consent: N/A. Registry and the Registration No. of the study/trial: N/A. Animal Studies: N/A. Dr. Yasuhisa Fujii and Dr. Hajime Tanaka are editorial board members of the International Journal of Urology and co-authors of this article. To minimize bias, they were excluded from all editorial decision-making related to the acceptance of this article for publication. Table S1: Comparison of Baseline Patient Characteristics and Safety Profiles Between the PROpel and TALAPRO-2 Trials Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.