A Spark Of Good News For Precision Oncology In Breast Cancer

Click here to view original web page at www.forbes.com

medicine in hand ( Getty

Last June, I almost wrote about a drug called AZD5363. It’s an AKT inhibitor, the sort of chemical we studied to see if it killed cancer cells in the lab, years ago. I was intrigued by an ASCO abstract: in an AstraZeneca-sponsored trial of metastatic triple-negative breast cancer this pill significantly extended patients’ overall survival from 12.6 to 19.1 months, vs. placebo, when given with standard paclitaxel chemotherapy.

This potentially big finding—that an AKT inhibitor extends survival in patients with metastatic breast cancer—got lost amid news about giving less treatment for early-stage disease. It turns out, results from the LOTUS trial of a similar drug, Genentech’s AKT inhibitor ipatasertib, were presented too; there was a trend for prolongation of overall survival.

Details of results in planned subset analyses—of patients with triple negative breast cancer and alterations in the PIK3CA, AKT1 or PTEN genes—have not been published for either AKT inhibitor. In June, Dr. Peter Schmid reported that in the PAKT trial of AZD5363 (capivasertib), survival and overall response rates were better against tumors with relevant gene variants. “In patients without pathway alterations, we did not see a statistically significant benefit,” he said.

Now, a new report about AZD5363 leads me back to this novel kind of cancer medicine and the broader topic of precision oncology. At the EORTC-NCI-AACR Symposium on Molecular Targets and Cancer Therapeutics in Dublin, Dr. Kevin Kalinsky, an oncologist at New York’s Columbia University, presented findings for AstraZeneca’s AKT inhibitor, capivasertib (AZD5363), in the NCI-MATCH trial.

At the NCI, MATCH stands for Molecular Analysis for Therapy Choice. Today this basket trial includes 18 open arms, or baskets, for advanced cancer with particular kinds of abnormalities, such as mutations, skips, fusions and amplifications, involving specific genes. With one exception, the NCI MATCH study pairs advanced cancer patients with single drugs. For instance, patients with PI3KCA mutations might get copanlisib, a PI3K inhibitor that’s been approved for treating some forms of lymphoma. You can find more details on open and closed MATCH baskets here.

As reported this week at the ECCO Plenary Session 1, 35 patients with E17K mutations of the AKT1 gene were prescribed AZD5363 in the MATCH trial. 23% of those (8) had measurable, partial responses to this oral AKT inhibitor, assessed by tumor shrinkage. Another 46% (16) experienced stable disease. Over half of the participants experienced a half year (6 months) of progression-free survival. Toxicity was not trivial: nearly half of patients had diarrhea; many were tired; a third reported nausea; one in four developed a rash; high blood sugar and abnormal liver function occurred, each not infrequently; some patients developed kidney problems.

That sounds not-great. But keep in mind, this was a heavily pre-treated group of advanced cancer patients; most received 3 or more kinds of prior treatment for metastatic disease before starting the experimental AKT inhibitor. Half of the patients (18) had metastatic breast cancer; other tumors included endometrial and ovarian cancers. The study’s primary endpoint is the objective response rate; progression-free survival is a secondary endpoint.

In a press release, Kalinsky emphasized that capivasertib is a pill, i.e. it is taken by mouth. Convenient! He spoke to the need for identifying patients who might benefit from this kind of drug, and future trials. “Although this mutation occurs in several different types of cancer, overall it is rare.” Researchers found the AKT mutation in 1.3% of patients (70 of 5548) tested in the NCI-MATCH trial, he said.

One might wonder, as I do, what would happen if the experimental drug, in this case an AKT inhibitor, were given sooner—before patients received and “failed” three or more agents, before their prognosis became so harsh. Perhaps a lower dose would lead to fewer side effects. And, based on previous findings for AstraZeneca’s AZD5363 and Genentech’s ipatasertib, which extended survival when taken along with paclitaxel (a taxane, related to the “backbone” chemo, Abraxane, given with atezolizumab in a recent study), there’s data, besides reason, to support combining new drugs, possibly at lower doses, with other kinds of anti-cancer medicines.

Never mind the skeptics. We know that with metastatic triple-negative breast cancer, for instance, median survival remains under 3 years. For advanced pancreatic cancer, most lung cancers, liver cancer, sarcoma and other types, some patients are desperate to try new medicines that might extend their lives. A search of ClinicalTrials.gov today reveals 12 active studies recruiting patients to test AZD5363 in a range of tumors, some in combinations with paclitaxel or PARP inhibitors. For ipatasertib, ClinicalTrials.gov lists 7 actively recruiting trials for solid tumors, with various combinations including the immune checkpoint inhibitor, atezolizumab.

What’s clear from the new NCI MATCH trial basket result is that some metastatic cancer patients with AKT mutations benefited from AZD5363; even in late-stage cases, the drug shrank 1 in 4 tumors, but it wasn’t easy to take. What’s unknown from the previous report of this drug’s activity, with paclitaxel, is if patients with advanced triple-negative breast cancer respond to AKT inhibitors regardless of whether they’ve got the PIK3CA, AKT1 or PTEN variants that were assessed in those trials, because overall survival was extended, more generally.

We’re well past the point of thinking that Precision Oncology will be so straightforward as giving an AKT inhibitor to all patients with AKT gene abnormality. Perhaps there’s a better biomarker, such as a more specific gene variant or signaling state, possibly along another cell pathway such as death inhibition, that determines responses to these drugs. Unfortunately, pharmaceutical companies have little incentive to find and publicize data that only a subset of patients benefit, unless it’s a slam-dunk kind of result. Meanwhile, I don’t dismiss the potential clinical value of AKT inhibitors and other novel agents—and especially pills, which are much less of a hassle for patients, if insurance will pay—such as PARP drugs, PI3Kinase inhibitors (about which very encouraging new data were presented last month at ESMO), CDK inhibitors, and other in-the-pipeline medicines, based on results of single agents given to very late-stage patients in early NCI-MATCH baskets.

It’ll take years to figure out how (in what doses), in what combinations, and who are the cancer patients who will live longer and with better quality-of-life upon receiving any of these agents. I don’t like to call cancer trials “games,” so I won’t call this the first inning. But it’s a start.


Related posts

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.