Capivasertib fits the bill for HR+, HER2– ABC

Capivasertib, when added to fulvestrant, prolonged progression-free survival (PFS) by 4.1 months in an analysis of Chinese patients with hormone receptor–positive (HR+), HER2-negative (HER2–) advanced breast cancer (ABC) whose disease had progressed on aromatase inhibitor, with or without a cyclin-dependent kinase 4 and 6 (CDK4/6) inhibitor in the phase III CAPItello-291 trial.

Outcomes of the global study had been published previously. [N Engl J Med 2023;388:2058-2070] One of the investigators, Dr Xichun Hu from Shanghai Cancer Center, Fudan University, Shanghai, China, reported findings from the Chinese cohort at ESMO Asia 2023 in Singapore.

“Investigator-assessed PFS in the Chinese cohort increased from 2.8 months to 6.9 months with the addition of capivasertib to fulvestrant, for an adjusted hazard ratio [adjHR] of 0.51,” he told oncologists during his presentation. 

Looking at patients with AKT pathway–altered (PIK3CA, AKT1, or PTEN) tumours, the benefit was even greater. “PFS increased from 1.9 months to 5.7 months with capivasertib–fulvestrant [adjHR, 0.41, 95 percent CI, 0.19–0.85],” he pointed out.

In patients with nonaltered (including unknown) tumours, PFS was 9.2 months with capivasertib–fulvestrant vs 3.6 months with fulvestrant (HR, 0.56; 95 percent CI, 0.34–0.94).

“Many second-line endocrine therapy [ET] options are available for HR+, HER2– ABC, but all have disappointingly short PFS,” commented study discussant Professor Lee Soo Chin, head and senior consultant, Department of Haematology-Oncology, National University Cancer Institute, Singapore at ESMO Asia 2023. “More effective ET options are needed, particularly with biomarker enrichment. I should say capivasertib fits the bill.”

The Chinese cohort (n=134) from the CAPItello-291 trial comprised 118 patients from Mainland China and 16 from Taiwan. Seventy-one were randomly assigned to capivasertib–fulvestrant and 63 to placebo–fulvestrant. Thirty-four percent of the patients had PIK3CA, AKT1, or PTEN-altered tumours. 

“Consistent and meaningful benefit was observed with capivasertib across clinically relevant subgroups,” Hu said. “The same benefit extends to patients with visceral metastases and those previously treated with a CDK4/6 inhibitor and/or chemotherapy in the advanced setting.”

Similar to the global study, the benefit-risk profile of capivasertib–fulvestrant combination appeared favourable to the Chinese cohort. No new safety concerns were identified.

Outcomes from the global study

In the global CAPItello-291 trial, capivasertib–fulvestrant significantly improved the dual primary endpoints of PFS (assessed according to RECIST, version 1.1) in the overall population and among patients with PIK3CA, AKT1, PTEN-altered tumours compared with placebo–fulvestrant. [N Engl J Med 2023;388:2058-2070]

The median PFS in the overall population (n=708, unselected for AKT pathway alteration) was 7.2 months with capivasertib–fulvestrant vs 3.6 months with placebo–fulvestrant (adjHR, 0.60; 95 percent CI, 0.51–0.71; p<0.001).

The PFS benefit was consistent in the AKT pathway-altered group, which was about 41 percent of patients (7.3 vs 3.1 months; adjHR, 0.50; 95 percent CI, 0.38–0.65; p<0.001).

In those with AKT pathway-nonaltered group (including those with unknown results on NGS), the hazard ratio was 0.70 (95 percent CI, 0.56–0.88), suggesting that the benefit was primarily driven by the AKT pathway-altered group.

Capivasertib now US FDA-approved

This led to the US FDA’s approval of capivasertib, in combination with fulvestrant, specifically for locally-advanced or metastatic HR+, HER2– BC harbouring PIK3CA, AKT1, PTEN alterations, as detected by an FDA-approved test, on November 16, 2023.

The agency also approved FoundationOne CDx assay as a companion diagnostic device to identify patients suitable for capivasertib therapy.

Chinese vs global population – differences

“Chinese patients made up 19 percent of the population in CAPItello-291 trial,” said Lee following Hu’s presentation. “Seventy-one percent were Asians.”

“Looking at the patient characteristics, the Chinese cohort had lower median body weight than the global population [60 kg vs 60-65 kg]. They had less prior exposure to CDK4/6 inhibitor [37 percent vs 69 percent], which might reflect an access issue, and a lower prevalence of AKT pathway-altered tumours [34 percent vs 41 percent],” added Lee.

In addition, the Chinese had poorer prognostic features than the global population. They were younger, with higher proportion of patients with poorer performance status (51 percent vs 34 percent, ECOG 1-2), had more visceral disease (75 percent vs 68 percent), and prior chemotherapy (30 percent vs 18 percent).

Consistent efficacy

In terms of efficacy, there was a consistent PFS benefit with capivasertib in the Chinese cohort. “Again, this benefit was more significant in patients with pathway-altered tumours,” said Lee. “Interestingly, the hazard ratio for the Chinese cohort [HR, 0.41] appeared better than for the global population [0.50].”

Median PFS was numerically shorter in the Chinese cohort than in the global population, both in the experimental and control arms. This, according to Lee, “could be due to the poorer prognostic features in the Chinese subgroup.”

An exploratory analysis of the pathway-nonaltered group (including unknown population) in the Chinese cohort also showed a benefit to capivasertib, consistent with outcomes in the global study, she continued.

“Toxicities and tolerability of capivasertib among the Chinese were generally comparable to the global population, except for any adverse events leading to dose interruption, which was about 6 percent higher among the Chinese cohort [40.8 vs 34.9 percent],” Lee added.

Capivasertib likely more tolerable

Permanent discontinuation due to any adverse events was also similar at 11 and 13 percent in the Chinese and global population, respectively. “This was fairly manageable and actually lower than the permanent discontinuation rate due to adverse events reported for everolimus in the BOLERO2 trial [19 percent; grade ≥3 stomatitis, 8 percent] and alpelisib in the SOLAR1 trial [25 percent; grade ≥3 hyperglycaemia 33 percent], suggesting that capivasertib is likely to be more tolerable than everolimus or alpelisib,” he added.

There was no significant difference in grade ≥3 toxicities between groups. “But what was intriguing was that any-grade hyperglycaemia, anaemia, and rash were much higher in the Chinese population,” Lee said.

“Hyperglycaemia and diarrhoea were predominantly grade 1 events, but there must be some grade 2 toxicities that are not always trivial,” she pointed out. “This we need to take note of when treating Chinese patients.”

Lee considered the substantial sample size in this ethnically homogeneous population as the main strength of the study. “This allows for a direct comparison with the non-Asian population. Reassuringly, it also confirms the efficacy and manageable toxicity profile of capivasertib in Chinese patients.”