Precision Medicine Oncology NCI-MATCH trial

Largest-Ever Precision Medicine Oncology Trial Ready for Launch
Anita T. Shaffer @Shaffer1
Published Online: Monday, June 1, 2015

A landmark clinical trial that will channel patients into treatment arms based on molecular abnormalities rather than cancer types aims to test the efficacy of more than 20 drugs simultaneously in an ambitious National Cancer Institute (NCI) plan to further propel oncology drug discovery into the precision medicine era.

Starting in July, the NCI-MATCH trial will seek to recruit 1000 adults 18 years of age or older with progressive advanced solid tumors and lymphomas that are either refractory to standard therapy or for which there is no standard therapy. Participants will be assigned to small phase II trials based on molecular tumor profiling of specimens from biopsies conducted at the time of study entry.

The trial is “a critical and leading part” of the nation’s precision medicine initiative, Clifford A. Hudis, MD, FACP, said during a 2015 ASCO Annual Meeting press briefing where NCI-MATCH and other innovative research projects were detailed.

“In oncology, we’ve embraced this idea for years,” said Hudis, a past president of ASCO who is a breast cancer specialist at Memorial Sloan Kettering Cancer Center. “The initiatives that we’re discussing today reflect not a new initiative but an expansion of an ongoing dream that we have been pursuing.”

“This is the largest and most rigorous precision oncology trial that’s every been attempted,” said James H. Doroshow, MD, the NCI’s deputy director.

The NCI has made an internal commitment to fully fund the study independent of the discussions now under way in Congress over the $215 million appropriation that the Obama administration has proposed for the precision medicine initiatives in cancer and other diseases, according to Doroshow.

Doroshow said it probably would cost $30 million to $40 million for the first stages of NCI-MATCH and that the budget could expand by 15% to 20% as more drugs are added to the list of agents tested and additional substudies are conducted. The NCI will pay for biopsies and laboratory sequencing tests, officials indicated.

The project will launch with an initial list of 10 substudies in which both previously approved drugs and investigational agents will be evaluated. Barbara A. Conley, MD, NCI study co-chair, said plans call for the trial to ramp up to more than 20 treatment arms within months of its launch.

First Batch of Drugs Identified

Here is the list of drugs that Conley identified for the first batch of studies and the molecular targets with which they are paired:
Crizotinib—Separate studies in ALK rearrangements and ROS1 translocations
Dabrafenib and trametinib—BRAF V600E or V600K mutations
Trametinib—BRAF fusions or non-V600E, non-V600K BRAF mutations
Afatinib—Separate studies in EGFR and HER2 activating mutations
AZD9291—EGFR T790M and rare EGFR activating mutations
T-DM1—HER2 amplifications
VS-6063—NF2 loss
Sunitinib—cKIT mutations
The FDA has approved six of the drugs on the list: crizotinib (Xalkori), dabrafenib (Tafinlar), trametinib (Mekinist), afatinib (Gilotrif), T-DM1 (Kadcyla), and sunitinib (Sutent).

AZD9291, a third-generation EGFR inhibitor, is being evaluated under the FDA’s breakthrough therapy program for patients with non–small cell lung cancer (NSCLC) whose tumors harbor the T790M resistance mutation. VS-6063, which also is called defactinib, is a small-molecule FAK inhibitor in phase I/II testing in mesothelioma, NSCLC, and ovarian cancer.

In order to enroll enough patients with mutations that allow a treatment match, organizers anticipate that 3000 patients will have to be screened. The goal is to enroll approximately 30 patients in each study.

An essential facet of the trial is the need to employ accurate assays to identify patients with the appropriate molecular features of their tumor, Conley said. Organizers have set up a network expected to provide molecular profiling results within 14 days or less, she said.

The genomic testing will be performed with the Ion Torrent Personal Genome Machine System’s custom panel of 143 genes, which in turn harbor more than 4000 variants.

For every trial, the primary endpoint will be overall response. Secondary endpoints include 6-month progression-free survival, time to progression, toxicity, and biomarker status.

The NCI-MATCH trial marks the next step in the agency’s efforts to harness the promise of precision medicine in oncology, starting with The Cancer Genome Atlas project to characterize genetic abnormalities in a range of cancer types, Doroshow said.

Doroshow described the planning and ultimately the conduct of NCI-MATCH as a national effort. “It has taken an absolute village to build this trial,” he said. “Hundreds of people supported the launch of the trial so far. Ultimately, it will take thousands of investigators to execute this study.”

The ECOG-ACRIN Cancer Research Group, which was formed three years ago through the merger of two oncology research organizations, is partnering with the NCI to plan and carry out the study at 2400 sites nationwide.

One of the army of investigators who will be involved in NCI-MATCH is Juneko Grilley-Olson, MD, an assistant professor at UNC Lineberger Comprehensive Cancer Center in North Carolina who specializes in thoracic and bone and soft tissue oncology.

She will help lead a substudy involving the investigational PI3K inhibitor GDC-0032, also called taselisib, which is expected to be studied in the second wave of trials that start as NCI-MATCH expands. Patients whose tumors harbor a PIK3CA mutation without a KRAS mutation and without PTEN loss are candidates for the study, Grilley-Olson said in an interview.

Grilley-Olson noted that NCI-MATCH organizers are hoping that at least 25% of the patients who enroll in studies have rare cancers.

“Those are tumors that often don’t have dedicated trials,” she said. “In the PI3 kinase arm we would be looking to enrich it for rarer tumors that have not been as extensively studied. With tumors such as breast cancer or lung cancer we probably wouldn’t learn as much additional information in a 30-patient cohort because they have been studied in trials with hundreds and hundreds of people [in those cancers].”