Background
Phase III, randomised, double-blind, placebo-controlled, global trial; N=812. Previously untreated metastatic TNBC; 30.7% had PIK3CA/AKT1/PTEN tumour alterations (retrospective central testing). Capivasertib = pan-AKT inhibitor; trial built on the OS signal of the phase II PAKT study.
Results
Interventions and follow up: Arm A: Paclitaxel 80 mg/m2 (d1, wks 1–3 of 4-wk cycle) + capivasertib 400 mg BID (d2–5, wks 1–3)
Arm B: Paclitaxel + placebo
Primary endpoint: Dual — OS in overall population and OS in PIK3CA/AKT1/PTEN-altered
mFollow up: NR (final OS data cut-off 18 Mar 2024)
Results: OS (overall): 17.7 vs 18.0 mo; HR 0.92 (95% CI 0.78–1.08); P=.3239 — not met
OS (altered): 20.4 vs 20.4 mo; HR 1.05 (95% CI 0.77–1.43); P=.7602 — not met
PFS (overall): 5.6 vs 5.1 mo; HR 0.72 (95% CI 0.61–0.84)
PFS (altered): 7.5 vs 5.6 mo; HR 0.70 (95% CI 0.52–0.95)
Arm B: Paclitaxel + placebo
Primary endpoint: Dual — OS in overall population and OS in PIK3CA/AKT1/PTEN-altered
mFollow up: NR (final OS data cut-off 18 Mar 2024)
Results: OS (overall): 17.7 vs 18.0 mo; HR 0.92 (95% CI 0.78–1.08); P=.3239 — not met
OS (altered): 20.4 vs 20.4 mo; HR 1.05 (95% CI 0.77–1.43); P=.7602 — not met
PFS (overall): 5.6 vs 5.1 mo; HR 0.72 (95% CI 0.61–0.84)
PFS (altered): 7.5 vs 5.6 mo; HR 0.70 (95% CI 0.52–0.95)
Adverse events
Grade ≥3 diarrhoea: 12.7% vs 0.7%
Capivasertib discontinuation for AEs: 8.5% vs 4.9%
AEs leading to death: 4.2% of all patients
Capivasertib discontinuation for AEs: 8.5% vs 4.9%
AEs leading to death: 4.2% of all patients
Conclusions
Adding capivasertib to first-line paclitaxel did not improve OS in metastatic TNBC, either overall or in PIK3CA/AKT1/PTEN-altered tumours, despite a numerical PFS advantage. AKT inhibition is not moving forward in first-line TNBC.
Key Limitations
PFS/OS disconnect with an earlier PFS data cut-off (May 2022) than final OS (Mar 2024); post-progression therapy NR. Biomarker status determined retrospectively. Enrolled 2019–2022, largely predating the pembrolizumab-chemo 1L standard for PD-L1 CPS≥10 (KEYNOTE-355); comparator was chemo alone. Contradicts the phase II PAKT OS signal.
Clinical Context
Capivasertib remains FDA approved (CAPItello-291) for HR+/HER2− advanced breast cancer with PIK3CA/AKT1/PTEN alterations; no established role in TNBC. 1L mTNBC standards: pembrolizumab + chemo (PD-L1 CPS≥10), PARP inhibitors (gBRCA), and Dato-DXd for IO-ineligible patients (TROPION-Breast02).