The NCI Molecular Analysis for Therapy Choice trial has demonstrated the feasibility of sequencing tumor DNA and matching patients to targeted therapies according to their cancer's mutation profile. However, response rates of those assigned to a therapy were generally low.

The largest precision oncology trial to date has demonstrated the feasibility of sequencing tumor DNA and matching patients to targeted therapies according to their cancer's mutation profile (J Clin Oncol 2020;38:3883–94). The NCI Molecular Analysis for Therapy Choice (NCI-MATCH) trial also revealed therapies with promising activity in genetically defined cancer subtypes.

However, the genomically guided approach is far from a panacea: A minority of the 6,000-plus study participants experienced a clinical benefit, and many tumors with presumed genetic vulnerabilities proved insensitive to targeted inhibition.

“It is a little bit of a blow” to precision oncology, says Douglas Johnson, MD, a study investigator at the Vanderbilt University Medical Center in Nashville, TN. “Based on the available therapies at the moment, the vast majority of cancer patients don't have some alteration that can be identified, targeted, and treated to produce a dramatic response.”

graphic

The NCI MATCH trial demonstrated the feasibility of sequencing their DNA and prescribing a targeted therapy based on their cancer's mutational profile.

Nonetheless, the study provides important insight, says investigator Christos Vaklavas, MD, of the University of Utah's Huntsman Cancer Institute in Salt Lake City. “Not all aberrations that we thought were actionable proved to be so,” Vaklavas says. “We needed to do this study” to find that out.

NCI-MATCH launched in 2015 to investigate whether assigning patients to targeted therapies based on tumor genomics rather than histology would improve outcomes for patients with no standard drug options. On that front, experts say, the study came up a little short.

Although 38% of the patients whose tumors were analyzed had “actionable” molecular alterations, only 18% had access to a relevant therapy through one of 30 single-arm subprotocols. Among those who did, only a fraction benefited: Thus far, overall response rates ranged from 2% to 38%, with most on the lower end.

The response rates were likely deflated, however, by criteria excluding patients with cancer subtypes for which a drug's efficacy, or lack thereof, was already known. In clinical practice, genetics-guided decision-making would include those patients, increasing response rates. What's more, study author Alice Chen, MD, of the NCI, notes that “as more targets are found, more targeted options become available,” and thus more patients are likely to receive suitable therapies.

A successor study called NCI-ComboMATCH is planned to assess whether combinations of targeted drugs, or monotherapies directed at tumors with multiple mutations, can boost response rates.

A few therapies tested in NCI-MATCH did yield a large number of responses—which, according to Edward Mills, PhD, speaks to the trial's “tremendous success” in identifying cancer–drug pairings amenable to biomarker-driven development. “It represents the future of how clinical trials should be done,” says Mills, of the clinical research services company Cytel, who was not involved in the study.

Take, for example, the arm of NCI-MATCH that assigned patients with BRAFV600-mutated tumors to receive dabrafenib (Tafinlar; Novartis) plus trametinib (Mekinist; Novartis). Before the trial, oncologists knew the regimen worked in melanoma, thyroid cancer, and lung cancer, but did not work in colorectal cancer. Patients with those cancers were not included in the analysis.

NCI-MATCH, along with a parallel effort sponsored by Novartis, revealed that rare subtypes of BRAFV600-mutated biliary tract, ovarian, and brain cancers are likely amenable to the drug strategy as well (J Clin Oncol 2020;38:3895–904; Lancet Oncol 2020;21:1234–43). “It's promising and really worth further investigation,” says April Salama, MD, of the Duke Cancer Institute in Durham, NC, who led the BRAFV600 arm of NCI-MATCH.

With its finding that 38% of patients have targetable mutations, NCI-MATCH also sets a “benchmark of actionability,” says principal investigator Keith Flaherty, MD, of Massachusetts General Hospital in Boston, MA. This, he asserts, “is important for the purpose of communicating the utility of next-generation sequencing in clinical practice.” –Elie Dolgin

For more news on cancer research, visit Cancer Discovery online at http://cancerdiscovery.aacrjournals.org/CDNews.