Skip to main content
Premium Trial:

Request an Annual Quote

NCI ComboMATCH Trial Gets Off the Ground After Addressing Complex Design Challenges


NEW YORK – Since the National Cancer Institute launched the ComboMATCH trial, study organizers have had to overcome numerous challenges associated with implementing the complex multi-study platform trial but are now beginning to enroll patients with specific tumor biomarkers into combination treatment cohorts.

At a webcast workshop at the annual meeting of the American Association for Cancer Research on Saturday, James Doroshow, director of the division of cancer treatment and diagnosis at the NCI, discussed some of the difficulties researchers have been dealing with getting ComboMATCH off the ground.

The NCI launched the platform trial in mid-2023, in collaboration with the Alliance for Clinical Trials in Oncology, Children's Oncology Group, ECOG-ACRIN Cancer Research Group, NRG Oncology, and the SWOG Cancer Research Network. Within ComboMATCH, these groups want to explore the safety and efficacy of biomarker-informed combination treatments in patients. The trial is a successor to NCI-MATCH, in which researchers tested the activity of targeted monotherapies matched to patients' cancer biomarkers.

Just over 100 patients have now enrolled in ComboMATCH, but getting the trial underway has been complicated, acknowledged Doroshow. "We thought erroneously that we would just be able to build on NCI-MATCH and this would be a piece of cake," he said. "It has not been."

The NCI-MATCH ultimately enrolled over 6,000 patients in 39 single-arm studies in less than three years. Upon seeing the initial readouts from some of these studies, there was consensus among research organizations involved that they should test the activity of evidence-based biomarker-informed targeted treatment combinations next, reasoning that simultaneously hindering multiple molecular drivers of a tumor is likely to benefit patients more than single agents.

Since developing the ComboMATCH concept five years ago, Doroshow recounted the myriad challenges NCI has had to overcome, including organizing molecular testing, gaining access to targeted drugs included in combination approaches, and finding patients with rare tumor mutations. "It's the most complicated [clinical trial development] process I've ever been involved in," Doroshow said.

Like NCI-MATCH, ComoboMATCH is a signal-seeking rather than a registrational study. Researchers rely on preclinical in vivo data to identify the additive or synergistic effects of drug combinations that target molecular aberrations driving patients' tumors and could potentially prolong tumor regressions. The trial design can facilitate single-arm and randomized Phase II studies, as well as histology- and disease-specific cohorts.

Patients are accrued via a screening cohort study that's led by ECOG-ACRIN and jointly governed by the National Clinical Trials Network and NCI.

To support ComboMATCH, investigators developed a series of algorithms called ComboMATCHbox to assign patients into the appropriate trial based on next-generation sequencing results. Separately, they also developed an infrastructure, molecular diagnostics net (MDNet), for circulating tumor DNA (ctDNA) assessments before, during, and after treatment.

Doroshow emphasized that NCI has set a high bar for the level of preclinical evidence a combination treatment strategy must have in order to be included in the study.

For example, in a patient-derived xenograft model of a uterine leiomyosarcoma treated with AbbVie's investigational PARP inhibitor veliparib or the alkylating agent temozolamide, neither agent was particularly active. However, treatment with a combination of the two drugs prolonged stable disease and significantly improved overall survival in the model. "This is the kind of data we're looking for," Doroshow said.

Although researchers often "get excited" by preclinical data in animal models showing activity for two agents in combination, Doroshow cautioned that until the data are carefully parsed, one can't be sure that the outcomes were related to one, the other, or both drugs together. When there is clinical evidence supporting a drug combination in a particular tumor type, researchers may test the regimen in a ComboMATCH study in another kind of tumor driven by the same genomic markers.

"Just as important as efficacy, we're looking for tolerability, because many times these drugs are not so far out from their first-in-human [study] or they're in Phase I," Doroshow added. That often means there's not much data yet regarding how well patients tolerate the combination regimens or how the drugs interact with each other. All of these factors mean that for some studies in the trial, researchers may not be able to jump into a Phase II study but will have to first conduct a Phase I run-in study to test safety.

After establishing how patients would be enrolled and how the drug combinations would be chosen within ComboMATCH, investigators had to figure out how to obtain the agents they wanted to study. Researchers always attempt to obtain the best agents based on the available evidence, Doroshow said, but "we can't always get those drugs."

Since combination regimens are the focus of the trial, that meant ComboMATCH organizers had to ink agreements with at least two different drug companies for each study. Doroshow said that required the NCI to develop specialized contract language to protect the intellectual property of all companies. However, even with such contract language, Doroshow noted it can be a major effort to get trials up and running in line with timelines at pharmaceutical companies. 

Beyond that, Doroshow said, "it would be impossible to even consider trials of this nature if the NCI did not have a mechanism for holding all of the different [investigational new drug applications] for all of these combinations. And it's not infrequent that we're expected to develop an investigational device exemption as well."

The next-generation sequencing assays being used to test patients' tissue samples in ComboMATCH are being performed by nearly 40 commercial and academic CLIA-certified laboratories, but the majority of these tests are not approved by the US Food and Drug Administration. As such, the agency requires a quality control monitoring plan for testing done withing the ComboMATCH lab network.

In addition to having tissue-based molecular profiling via NGS testing, patients in ComboMATCH will also receive whole-exome sequencing, RNAseq, and ctDNA analysis performed by the NCI's Molecular Characterization Laboratory and MD Anderson Cancer Center. As trials are completed, Doroshow said clinical data will be entered into the NCI's Cancer Research Data Commons.

Although getting ComboMATCH off the ground has been a challenge, Doroshow said trials are already underway. Some studies are in the process of being designed and evaluated, while others are open and accruing "very well."  Certain studies are focused on tumor types with very rare mutations, which Doroshow said will take longer to accrue. 

"We view this as a very high-risk, but high-reward opportunity because it may provide us with some notion about how we should be doing targeted combination therapy going forward," Doroshow said.