The experimental pan-AKT inhibitor capivasertib (AZD5363) helped stabilize or shrink tumor growth in most patients with metastatic AKT1-mutant tumors enrolled in the U.S.-wide NCI-MATCH trial, according to data presented at the EORTC-NCI-AACR Symposium on Molecular Targets and Cancer Therapeutics.

The experimental pan-AKT inhibitor capivasertib (AZD5363; AstraZeneca) helped stabilize or shrink tumor growth in most patients with metastatic AKT1-mutant tumors enrolled in the NCI-MATCH trial, a massive U.S.-wide basket study designed to test whether cancers can be treated based on gene abnormalities rather than tumor type. The findings dovetail with previous reports of capivasertib's clinical activity in women with breast and gynecologic cancers.

The NCI-MATCH results additionally show that the drug works against some less common tumors—and reveal safety concerns with capivasertib that can prompt people to stop taking the drug, especially those treated at community clinics.

“This is a promising agent,” says Kevin Kalinsky, MD, of the Columbia University Irving Medical Center in New York, NY, who presented the data at the EORTC-NCI-AACR Symposium on Molecular Targets and Cancer Therapeutics in Dublin, Ireland, on November 13. “Now the goal is to think about how to best develop this drug” for it to earn marketing approval for treating specific tumor types.

The capivasertib subprotocol of NCI-MATCH enrolled 35 patients with a missense mutation that causes an amino acid change at position 17 of AKT1. This rare AKT1 E17K mutation is found in about 1% of tumors, most often in those of the breast.

Among the 35 participants, the drug yielded partial responses in eight individuals—six of whom had HR-positive, HER2-negative breast cancer—and led to stable disease in 16 others.

“Overall,” says David Hyman, MD, of Memorial Sloan Kettering Cancer Center in New York, NY, “these data are highly consistent with our previous report of this agent in this setting.”

Last year, Hyman and his colleagues published the results of a capivasertib-focused basket study with comparable, although slightly more restrictive, enrollment criteria as the NCI-MATCH subprotocol. Among the 52 participants with AKT1 E17K mutations, about one quarter of patients experienced partial responses.

In that study, nearly all the responses occurred in patients with breast, endometrial, or cervical cancers, although one had non–small cell lung cancer. The only rare tumor that responded to therapy was an anal adenocarcinoma.

Adding to the list of responsive tumor types, Kalinsky's team documented a partial response in a woman with an endometrioid adenocarcinoma and another in a woman with a rare uterine leiomyosarcoma. Another patient with an oncocytic carcinoma of the parotid gland—a rare type of salivary gland tumor—experienced stable disease and has been on therapy for 21 months and counting.

“We really are seeing activity in some rare cancers,” Kalinsky says. “So as a field, we have to think about whether there's utility to move forward with a small cohort of patients with these specific rare tumors”—an extrapolation that would be impossible were it not for the pan-cancer nature of the basket study design.

The side effects documented in the two trials were similar, with the most common serious adverse events being hyperglycemia, diarrhea, and rash. However, the rate of toxicity-related treatment discontinuation was twice as high in the new study—25% versus 12%.

Kalinsky chalks up the difference to the nature of the clinical sites in NCI-MATCH, which, along with academic medical centers, included many community clinics where doctors typically have less experience managing the toxicities commonly associated with experimental agents targeted at the PI3K/AKT/mTOR pathway.

“This study gives a real-world experience,” Kalinsky says, “and it really highlights that we need to be very mindful” of these tolerability issues. –Elie Dolgin