Sacha J. Howell, BMBS, MSc, PhD
Adding the Akt inhibitor capivasertib to fulvestrant led to a more than doubling of progression-free survival (PFS) compared with fulvestrant alone in patients with endocrine-resistant estrogen receptor (ER)-positive advanced breast cancer.
In the placebo-controlled phase II FAKTION trial, patients who relapsed or progressed on aromatase inhibitor (AI) therapy who were randomized to capivasertib plus fulvestrant had a median PFS of 10.3 months (95% CI, 5.0-13.2) compared with 4.8 months (95% CI, 3.1-7.7) in those randomized to fulvestrant plus placebo (unadjusted HR, 0.58; 95% CI, 0.39-0.84; 2 sided P
The results were announced at the 2019 ASCO Annual Meeting by study investigator Sacha J. Howell, BMBS, MSc, PhD, HCA Healthcare, UK.
“Tumor PI3K/Akt/PTEN pathway activation, defined as PIK3CA exon 9/20 hotspot mutation or PTEN null by immunohistochemistry, does not appear to affect sensitivity to capivasertib in ER-positive metastatic breast cancer,” he said.
The PI3K/Akt/PTEN pathway is activated in about half of ER-positive metastatic breast cancers. “This is predominantly due to PIK3CA
activating mutations but also loss or activation of PTEN
, or AKT1
activating mutation,” said Howell. In the preclinical setting, pathway activation leads to ligand-independent activation of ER “and conversely pathway inhibition also leads to a compensatory increase in ER-dependent transcription.” Thus, there is a strong rationale for simultaneous targeting of the PI3K/Akt/PTEN and ER pathways, he said.
Capivasertib is a potent and selective inhibitor of Akt 1-3 isoforms. Synergistic activity with fulvestrant was observed preclinically in both endocrine-sensitive and resistant models2,3
and modest clinical activity has been seen as monotherapy in tumors with Akt1
mutations but not in tumors with PIK3CA
-mutated ER-positive metastatic breast cancer.4
No effect of capivasertib was found in the BEECH study when used in combination with paclitaxel in PIK3CA
-mutant ER-positive advanced breast cancer5
“but importantly, co-treatment with endocrine therapy was not permitted in that trial,” Howell said.
In FAKTION, 140 postmenopausal women with ER-positive/HER2-negative metastatic or unresectable locally advanced breast cancer recruited from 21 sites in the UK were randomized 1:1 to fulvestrant (500 mg on days 1 and 15 of cycle 1 and day 1 only of subsequent 28-day cycles) plus either capivasertib (n = 69; 400 mg twice daily) or placebo (n = 71), 4 days on/3 days off, until disease progression, unacceptable toxicity, or withdrawal of consent.
Eligible patients had to have progressed on an AI for metastatic/locally advanced breast cancer or relapsed on adjuvant AI. Patients were allowed a maximum of 1 line of prior chemotherapy and up to 3 lines of endocrine therapy for metastatic breast cancer. Prior fulvestrant or any inhibitor of the PI3K/Akt/mTOR pathway was prohibited.
Baseline characteristics were well balanced between the 2 arms, with a median age of 61 years (range, 40-82) in the placebo arm and 62 years (range, 42-81) in the capivasertib arm. Measurable disease was present in about 70% in each arm and visceral disease in about two-thirds. Some 63% in the placebo arm and 57% in the capivasertib arm had 1 prior line of endocrine therapy for metastatic disease, and 28% and 29%, respectively, had ≥2 prior lines. About one-fourth had prior chemotherapy for metastatic disease. Some 34% in the placebo arm and 39% in the capivasertib arm had PIK3CA
exon 9/20 mutations.
When PFS was analyzed by PI3K/Akt/PTEN pathway activation status, the median PFS advantage to capivasertib was similar in the 59 patients with pathway activation (HR, 0.59; 2-sided P
=.064) compared with the 81 without pathway activation (HR, 0.56; 2-sided P
= .035). Additional next-generation sequencing analyses are ongoing to more comprehensively characterize the tumors, Howell said.
Object response was determined in patients with measurable disease, which included 49 patients in the capivasertib arm and 50 patients in the placebo arm. The objective response rate was 41% in the capivasertib group versus 12% in the placebo group, for an odds ratio for response of 5.17 (95% CI, 1.83-14.62; 2-sided P
= .002). All responses in both arms were partial responses.
The clinical benefit rate was 55% and 36%, respectively (OR, 2.17; 95% CI, 0.97-4.87; 2-sided P
= .061). The median duration of response was 7.1 months (range, 3.8-9.9) in patients randomized to capivasertib compared with 5.0 months (range, 2.8-7.3) in the placebo arm.
The median overall survival was 26.0 months in the capivasertib arm versus 20.0 months in the placebo arm (HR, 0.59; 95% CI, 0.34-1.05; P
=.071). Data maturity, however, is only 37%, Howell emphasized.