Original ResearchNeoadjuvant buparlisib plus trastuzumab and paclitaxel for women with HER2+ primary breast cancer: A randomised, double-blind, placebo-controlled phase II trial (NeoPHOEBE)

- Buparlisib was evaluated in HER2+ primary breast cancer in PIK3CA mutated and wild type patients.
- Study was terminated early due to toxicity, and hence underpowered for the primary end-point.
- Trend towards a higher ORR was observed in the ER+/HER2+ subgroup.
- Increase in TILs in the first 15 days was significantly associated with increased pCR rates.
- More specific PI3K alpha inhibitors are currently being evaluated in the clinical setting.

AimThe Neoadjuvant PI3K inhibition in HER2 OverExpressing Breast cancEr (NeoPHOEBE) trial evaluated the efficacy and safety of buparlisib, a pan-phosphatidylinositol 3-kinase (PI3K) inhibitor, plus trastuzumab and paclitaxel as neoadjuvant treatment for human epidermal growth factor receptor-2 positive (HER2+) breast cancer.MethodsNeoPHOEBE was a neoadjuvant, phase II, randomised, double-blind study. Women with HER2+ breast cancer were randomised within two independent cohorts by PIK3CA mutation status and, in each cohort stratified by oestrogen receptor (ER) status to receive buparlisib or placebo plus trastuzumab (first 6 weeks) followed by buparlisib or placebo with trastuzumab and paclitaxel. Primary end-point was pathological complete response (pCR) rate; key secondary end-point was objective response rate (ORR) at 6 weeks. Exploratory end-points were evaluation of Ki67 levels and change in tumour infiltrating lymphocytes (TILs) in intermediate biopsies at day 15.ResultsRecruitment was suspended mainly due to liver toxicity after enrolment of 50 of the planned 256 patients. In each arm (buparlisib n = 25; placebo n = 25) 21 patients (84%) had wild type PIK3CA and 4 patients (16%) had mutant PIK3CA. Overall, pCR rate was similar between buparlisib and placebo arms (32.0% versus 40%; one-sided P = 0.811). A trend towards higher ORR (68.8% versus 33.3%; P = 0.053) and a significant decrease in Ki67 (75% versus 26.7%; P = 0.021) was observed in buparlisib versus placebo arm in the ER+ subgroup (Pinteraction = 0.03).ConclusionsAddition of the pan-PI3K inhibitor buparlisib to taxane-trastuzumab-based therapy in HER2+ early breast cancer was not feasible. However, the higher ORR and Ki67 reduction in the ER+, HER2+ subgroup indicates a potential role for PI3K-targeted therapy in this setting and may warrant further investigation with better-tolerated second-generation PI3K inhibitors.Trial registration identifierNCT01816594.