Skip to main content
Premium Trial:

Request an Annual Quote

NCI Launches Platform Trial to Test Biomarker-Informed Combination Therapies


NEW YORK – The National Cancer Institute has launched the ComboMatch platform trial, in which researchers will test the efficacy and safety of treatment combinations that inhibit genomic aberrations driving their tumors and stave off resistance to targeted therapies.

NCI will deploy ComboMATCH in collaboration with five US National Clinical Trials Network groups: Alliance for Clinical Trials in Oncology, Children's Oncology Group, the Eastern Cooperative Oncology Group and American College of Radiology Imaging Network (ECOG-ACRIN) Cancer Research Group, NRG Oncology, and Southwest Oncology Group Cancer Research Network. Researchers hope to enroll nearly 3,000 patients into the study, a successor to NCI-MATCH, in which investigators tested the activity of targeted monotherapies matched to patients' cancer biomarkers. In NCI-MATCH — one of the first large-scale platform trials that popularized the precision medicine research framework in the field — only a minority of treatment arms met the primary endpoint and achieved an objective response rate of 16 percent to single-agent targeted therapies.

In ComboMATCH, researchers are exploring whether patients will fare better on biomarker-informed combination treatment approaches. "This is a multi-pronged effort to test the strategy that using one drug aimed at the genetic driver of a cancer combined with another drug that attacks a major cause of resistance to the first drug can result in better patient outcomes," Jeff Moscow, branch chief of the NCI's Cancer Therapy Evaluation Program, said in a press briefing. "Therefore, ComboMATCH treatment trials can be seen as a benefit of decades-long NCI funding of basic research into the causes of drug resistance to targeted therapies."

Researchers have begun enrolling patients into three Phase II sub-studies within the platform trial. In the EAY191-N2 trial, NRG Oncology investigators will test the hormone therapy fulvestrant with Pfizer's MEK inhibitor Mektovi (binimetinib) against fulvestrant alone in patients with hormone receptor-positive, HER2-positive breast cancer bearing NF1-disrupting mutations. Researchers will enroll patients who are fulvestrant-naïve and those that have received it previously.

In the EYA191-N4 trial, also run by NRG Oncology, investigators will study a combination of AstraZeneca's MEK1/2 inhibitor Koselugo (selumetinib) and AstraZeneca and Merck's PARP inhibitor Lynparza (olaparib) in patients with gynecological cancers bearing alterations in RAS pathway genes including KRAS, NRAS, HRAS, BRAF, MEK1, and MEK2, or inactivating NF1 mutations. Patients with low-grade serous and other ovarian cancers, as well as endometrioid and other endometrial cancers, will be randomized to receive Koselugo-Lynparza or just Koselugo.

Then, in EAY191-S3, operated by SWOG Cancer Research Network, researchers will evaluate paclitaxel chemotherapy with Genentech's AKT inhibitor ipatasertib in patients with advanced or metastatic non-breast solid tumors who have progressed on taxane therapy and have cancers harboring mutations in AKT1, AKT2, and AKT3.

ECOG-ACRIN will start a fourth sub-study, EAY191-E4, "in the next week or two," James Ford, a medical oncologist at Stanford Medicine and a lead investigator for ComboMATCH said. In EAY191-E4, researchers will test a combination of Novartis' tyrosine kinase inhibitor Tasigna (nilotinib) and paclitaxel in patients with solid tumors previously treated with taxanes. Tasigna is typically used to treat leukemia.

By the end of July, the NCI is hoping to activate nine more Phase II sub-studies that will include combination strategies with Amgen's KRAS G12C inhibitor Lumakras (sotorasib) and EGFR-targeted monoclonal antibody Vectibix (panitumumab), Novartis' PI3K inhibitor Piqray (alpelisib), and Day One Biopharmaceuticals' investigational pan-RAF kinase inhibitor tovorafenib (DAY101). According to Moscow, 11 companies so far have agreed to contribute their marketed and investigational therapies for the project through collaborative research agreements.

Researchers decide which combination treatments to test in each ComboMATCH sub-study based on in vivo evidence from at least two tumor models that the agents will have synergistic antitumor activity and together are likely to impart a meaningful clinical benefit in patients. In other words, there must be evidence suggesting that the combination treatment won't merely slow the rate at which patients' tumors are growing but must "truly show evidence of cell killing or tumor killing or shrinkage in mouse models," said Ford.

"The selection of the most promising drug combinations to test against a genetic driver is much more complex than the selection of a single agent developed against that target," Moscow acknowledged. "This required unprecedented collaboration by physician scientists from all five of the NCTN groups to develop common standards for the levels of preclinical and clinical evidence to select the optimal combination therapies for ComboMATCH treatment trials."

In ComboMATCH, researchers decided not to include immunotherapies because there's no consensus on using in vivo models to gain insights about how well they'll work in patients, and because the NCI is conducting a separate initiative, iMATCH, to test immuno-oncology approaches.

There's always a risk of increasing toxicities when combining drugs. Before a combination approach can proceed to a Phase II efficacy sub-study, it must also have clinical evidence of safety and tolerability. For combinations lacking such evidence, a Phase I study will be conducted.

Ford said that while each sub-study may contain several cohorts, there should be at least one cohort that is histology agnostic. And unlike NCI-MATCH, which was comprised entirely of single-arm studies, ComboMATCH will include both randomized and single-arm study designs.

To qualify for ComboMATCH, patients with locally advanced or metastatic solid tumors must have progressed on at least one prior standard systemic therapy or have cancers that have no standard treatment options. Adults and children interested in joining the trial will be initially enrolled in the ComboMATCH registration trial and undergo genomic sequencing at one of 33 diagnostic laboratories in the NCI's lab network for the project. Over 100 sites have opened a ComboMATCH registration trial thus far and 200 more have started the process, Moscow said.

Trial sponsors aren't covering the cost of genomic testing, however, and will bill patients' health insurance, Ford said. Each patient will also have an on-study biopsy for whole-exome sequencing and RNAseq performed centrally by the NCI. The patient's clinical and genomic testing information will be entered into MATCHbox, the NCI's computational platform, for trial matching. Trial recommendations produced by MATCHbox will be based on tumor histology, detected genetic variants, allele frequency, strength of evidence on drug combinations, and the number of patients accrued in the trial.

In an interview, Ford said he does not anticipate ComboMATCH will have difficulty accruing patients into the various sub-studies. NCI-MATCH had overwhelming interest from patients when it first launched and ended up recruiting 6,000 participants across the country.

Ford attributes this success to the number of trial sites that were available for NCI-MATCH. "It was really driven by community sites opening across the country," he said. "And since then, genomic testing has become much more standard and more accessible."

Even as the first sub-studies are getting off the ground, Moscow said the NCI will begin reviewing data next month for another wave of studies to add. "What you see here, we hope, is just the beginning of the types of novel [therapy combinations] we hope ComboMATCH will be able to deliver to our patients."

Like NCI-MATCH, ComboMATCH is a signal finding study, Ford said. As such, he hopes that for drug combinations that show promise, drugmakers and investigators will try to further confirm that signal of benefit in other clinical trials. "For many of these combinations, this is really the first big test," he said.