Abstract RF4-01: Central Nervous System Outcomes from the Phase III PATINA Trial (AFT-38)

Abstract Background Central nervous system (CNS) metastases are a major clinical concern in patients with HER2-positive metastatic breast cancer (MBC). The PATINA phase III study showed improved PFS with palbociclib added to maintenance anti-HER2/endocrine therapy (44.3 vs. 29.1 months; hazard ratio 0.75; 95% CI 0.59-0.96; 1-sided P=0.0109). Preclinical and early clinical data suggest palbociclib may have CNS activity. We conducted a pre-specified secondary analysis to evaluate the incidence and timing of CNS progression. Methods CNS progression was defined as the emergence of new intracranial lesions per investigator assessment. Baseline CNS imaging was not mandated but recommended if symptomatic. Competing risks analyses evaluated time from registration to first CNS progression or death (CNS PFS), treating non-CNS progression as a competing event. All randomized patients were included in the primary analysis, regardless of known baseline CNS involvement. Patients with baseline CNS metastases were not censored if progression involved the CNS. A secondary analysis excluded patients with CNS metastases present at baseline. Cumulative incidence functions were estimated and compared using 2-sided Gray’s test. Results Data cutoff for this analysis was on October 15, 2024. 518 participants were enrolled between June 2017 and July 2021, 261 to the palbociclib arm and 257 to the control arm. CNS metastases at baseline were present in 11 patients in the palbociclib arm and 9 patients in the control arm. Among all randomized patients, CNS progression occurred in 35 of 261 (13.4%; 95% CI: 9.5-18.2) in the palbociclib arm and 50 of 257 (19.5%; 95% CI: 14.8-24.8) in the control arm. When excluding patients with CNS metastases at baseline, CNS progression was observed in 32 of 250 (12.8%; 95% CI: 8.9-17.6) in the palbociclib arm and 47 of 248 (19.0%; 95% CI: 14.3-24.4) in the control arm. The median follow-up for patients who were alive and progression-free was 53.5 months. At 36 months, the cumulative incidence of CNS PFS was 13.4% vs. 19.9% for palbociclib versus control (Gray’s test p=0.0386). Importantly, when the analysis was restricted to pts without evidence of CNS disease at baseline (Table 1), —the CNS benefit with palbociclib was maintained (13.0% vs. 19.2% at 36 months, Gray’s test p = 0.0378). Conclusion The addition of palbociclib to anti-HER2 and endocrine therapy was associated with a lower incidence of CNS metastases. Among patients without known baseline CNS involvement, the absolute reduction in CNS progression at 36 months was 6.2% in favor of palbociclib. This benefit was observed at 12 months and sustained over 3 years (Table 1). These findings suggest that palbociclib may help delay or prevent CNS involvement in HR+/HER2+ MBC, warranting confirmation in future studies. Citation Format: O. Metzger, S. Mandrekar, T. Dockter, L. Gianni, E. Ciruelos, A. De Michele, J. Gligorov, E. Lim, S. Loibl, X. Gonzàlez Farré, F. Lynce, J. Lanzillotti, L. Carey, S. Goel, A. Partridge, E. Winer. Central Nervous System Outcomes from the Phase III PATINA Trial (AFT-38) [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr RF4-01.