Chemotherapy is still the mainstay treatment for triple-negative breast cancer (TNBC) and patient outcomes remain poor. PI3K–AKT signalling is often dysregulated in TNBC; now, new data from the randomized, double-blind, phase II PAKT trial support the addition of AKT inhibition to chemotherapy.

PAKT involved 140 women with untreated metastatic TNBC who were randomly assigned (1:1) to receive paclitaxel plus either the pan-AKT inhibitor capivasertib or placebo. Intermittent dosing of capivasertib (4 days on, 3 days off) was chosen over continuous daily dosing on the basis of data from phase I trials. The median progression-free survival (mPFS) duration was 5.9 months with capivasertib versus 4.2 months with placebo (1-sided P = 0.06, meeting the predefined significance level of 1-sided P = 0.10). Only small, nonsignificant increases in the objective response rate (ORR, 34.8% versus 28.8%), clinical benefit rate (CBR, 41.4 versus 37.1%) and median duration of response (mDoR, 7.6 months versus 7.3 months) were observed. However, median overall survival (mOS) was prolonged by 6.5 months with capivasertib (19.1 months versus 12.6 months; 2-sided P = 0.04).

The benefits of capivasertib seem to be limited mostly to patients with genetic alterations in PIK3CA, AKT1 or PTEN (n = 28). In this subgroup, the ORR, CBR, mDoR, mPFS and mOS were 35.3%, 52.9%, 13.3 months, 9.3 months and not reached, respectively, with capivasertib versus 18.2%, 27.3%, 3.5 months, 3.7 months and 10.4 months with placebo.

The grade 3–4 adverse event rate was higher with capivasertib (54.4% versus 25.7%). The excess toxicities were mainly limited to the known class effects of AKT inhibitors (such as diarrhoea, rash and hyperglycaemia) and were manageable.

These results are similar to those observed with ipatasertib, a different AKT inhibitor, in the LOTUS trial, thus further validating the approach of adding an AKT inhibitor to chemotherapy for TNBC.