Capivasertib plus abiraterone in PTEN-deficient metastatic hormone-sensitive prostate cancer: exploratory analysis of the proportional hazards assumption

We read with great interest the article by Fizazi et al. reporting the CAPItello-281 trial, which demonstrated a statistically significant improvement in radiographic progression-free survival with capivasertib plus abiraterone in patients with PTEN-deficient metastatic hormone-sensitive prostate cancer (1). In the post hoc analysis of patients with 100% PTEN deficiency, the hazard ratio (HR) for overall survival (OS) was 0.77 [95% confidence interval (CI): 0.51–1.14], which did not reach statistical significance.

While the Cox proportional hazards (PH) model is the standard approach in oncology trials, its validity relies on the PH assumption. Visual inspection of the Kaplan-Meier curves for OS in the 100% PTEN-deficient subgroup (Fig. 3B in the original article) suggested potential late divergence, which prompted us to explore whether the PH assumption might be questioned in this dataset.

Using the Guyot algorithm (2), we reconstructed individual patient data (r-IPD) from the published Kaplan-Meier curves and number-at-risk tables for the 100% PTEN-deficient population (capivasertib: n=169, 43 events; placebo: n=162, 55 events). The reconstructed data showed good agreement with the published results (r-IPD HR =0.77, 95% CI: 0.52–1.13 vs. published HR =0.77, 95% CI: 0.51–1.14). We then performed exploratory PH testing using the Schoenfeld residual test with log-transformation and time-varying coefficient analysis.

Our exploratory analysis suggested possible departure from PH: the Schoenfeld test with log-transformation (P=0.037) and time-varying coefficient analysis (P=0.049) both indicated potential non-proportionality. The Schoenfeld residual plot demonstrated a downward trend in β(t) over time (Figure 1A), which may suggest that the treatment effect increases with longer follow-up. The log-log plot showed early crossing of curves (Figure 1B), consistent with potential early non-PH.

Figure 1 Exploratory assessment of the proportional hazards assumption for OS in the 100% PTEN-deficient population using r-IPD. (A) Schoenfeld residual plot showing scaled Schoenfeld residuals [β(t)] for the treatment arm over time (months). The solid line represents the smoothed estimate, with dashed lines indicating the 95% CI. The horizontal red dashed line indicates β(t) =0 (no treatment effect). A downward trend suggests a decreasing HR over time. (B) Log-log survival plot showing log[−log(survival)] vs. log(time). Parallel curves would indicate proportional hazards; crossing or diverging curves suggest potential violation. Abi, abiraterone; β(t), time-varying coefficient; Capi, capivasertib; CI, confidence interval; HR, hazard ratio; OS, overall survival; Pbo, placebo; r-IPD, reconstructed individual patient data.

We acknowledge important limitations of our analysis. First, r-IPD reconstruction, while following established methodology, cannot perfectly replicate the original data. Second, the 100% PTEN subgroup was a post hoc analysis with relatively immature OS data (26.4% maturity). Third, multiple testing without adjustment may inflate false-positive rates. Therefore, these findings should be considered hypothesis-generating rather than definitive.

Nevertheless, if non-PH are present, the conventional HR alone may not fully capture the treatment effect, and alternative measures such as restricted mean survival time (RMST) could provide complementary insights (3). We would like to raise a modest suggestion that future analyses of this trial, particularly with longer follow-up, might consider incorporating RMST alongside conventional Cox regression to provide a more comprehensive assessment of treatment benefit.

Acknowledgments

During the preparation of this work, the authors used Claude (Anthropic) to assist with manuscript formatting and language editing. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

Footnote

Provenance and Peer Review: This article was a standard submission to the journal. The article did not undergo external peer review.

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tau.amegroups.com/article/view/10.21037/tau-2026-1-0104/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Fizazi K, Clarke NW, De Santis M, et al. Capivasertib plus abiraterone in PTEN-deficient metastatic hormone-sensitive prostate cancer: CAPItello-281 phase III study. Ann Oncol 2026;37:53-68. [Crossref] [PubMed]
2. Guyot P, Ades AE, Ouwens MJ, et al. Enhanced secondary analysis of survival data: reconstructing the data from published Kaplan-Meier survival curves. BMC Med Res Methodol 2012;12:9. [Crossref] [PubMed]
3. Royston P, Parmar MK. Restricted mean survival time: an alternative to the hazard ratio for the design and analysis of randomized trials with a time-to-event outcome. BMC Med Res Methodol 2013;13:152. [Crossref] [PubMed]