Oops, you're using an old version of your browser so some of the features on this page may not be displaying properly.

MINIMAL Requirements: Google Chrome 24+Mozilla Firefox 20+Internet Explorer 11Opera 15–18Apple Safari 7SeaMonkey 2.15-2.23

Identification of Actionable Molecular Alterations in NCI-MATCH

Feasibility of complex, genomically guided clinical trial in oncology
02 Nov 2020
Personalised medicine

Centralised molecular screening performed in the NCI-MATCH trial on freshly procured tumour biopsy specimens led to discovery of actionable molecular alterations in 38% of patients. Furthermore, 18% of patients were assigned to actively enrolling treatment arms. The investigators team led by Keith T. Flaherty of the Massachusetts General Hospital Cancer Center in Boston, MA, US reported on 13 October 2020 in the Journal of Clinical Oncology that performance of next-generation sequencing (NGS) in biopsy specimens from patients with relapsed, refractory advanced cancers permits triaging of nearly one-fifth of patients to evidence-based investigational therapy.

The National Cancer Institute–Molecular Analysis for Therapy Choice (NCI-MATCH) trial was the first trial performed at national scale in the United States that incorporated centralised diagnostic testing and geographically distributed clinical investigation of dozens of parallel treatment options. It was designed and co-administered by the Division of Cancer Treatment and Diagnosis, NCI, and the ECOG-ACRIN Cancer Research Group with the participation of the NCI National Clinical Trials Network and NCI Community Oncology Research Program.  

Patients with a tumour molecular alteration addressed by a targeted treatment lacking established efficacy in that tumour type were assigned to 1 of 30 treatments in parallel, single-arm, phase II subprotocols. Treatments included in NCI-MATCH were single agents or combinations, approved by US Food and Drug Administration or investigational, and required to have a recommended phase II dose, as well as a molecular alteration that might predict response on the basis of preclinical or clinical data, and at least provisional evidence of clinical activity.

Tumour biopsy specimens from 5,954 patients with refractory cancers were analyzed centrally with NGS and selected immunohistochemistry in a master screening protocol. The assignment rate to treatment arms was assessed. Molecular alterations in seven tumours profiled in both, NCI-MATCH and The Cancer Genome Atlas (TCGA) of primary tumours were compared.

Molecular profiling was successful in 93.0% of specimens. Molecular alterations for assignment to an NCI-MATCH subprotocol were present in 37.6% of patients. When molecular, prior treatment, and specific cancer exclusions were accounted for, the match rate was 26.4%. Lack of subprotocol availability, because the subprotocol had reached accrual or because of the limit on certain histologic types was reached, led to a treatment assignment rate of 17.8% (985 of 5,540).

Assignment rates for non-small cell lung cancer, colorectal, breast, and prostate cancer were 17.4%, 13.7%, 17.8%, and 23.0%. Assignment rates >25% were found in patients with central nervous system malignancy (37.2%), urothelial cancer (36.0%), cholangiocarcinoma pancreaticobiliary (25.9%), cervical cancer (28.4%), gastro-oesophageal cancer (27.8%), melanoma (26.3%), uterine cancer (26.2%) and anal cancer (25.5%). However, assignment rates were low in patients with pancreatic cancer (5.8%), small cell lung cancer (5.1%) and lymphoma (5.0%).

In total, 70% of assigned patients received treatment on a subprotocol. Furthermore, 11 of 30 subprotocols reached the accrual goal of at least 31 eligible patients. 

Deleterious or activating mutations in TP53 (47.4%), KRAS (21.2%), and APC (12.4%) were commonly observed. The most prevalent co-occurring mutations were KRAS and TP53 in 12.1% of tumours. The most frequently observed actionable alterations were in PIK3CA (11.8%) and PTEN (6.3%). All other actionable alterations were observed in ≤3%. Patients with single eligible alterations and without other actionable or non-actionable alterations (34.1%) were hypothesised to be the most likely to be responsive to the assigned targeted therapy.

Of patients with the most commonly identified actionable alterations, 37.6% (723 of 2,083) were excluded from treatment due to co-occurring mutations known to confer resistance. Of patients assigned to NCI-MATCH subprotocols, 54.0% of patients with actionable alterations also had co-occurring mutations in tumour-suppressor genes that have been implicated in therapeutic resistance.

Multiple actionable or resistance-conferring tumour mutations were seen in 11.9% and 71.3% of specimens, respectively. Known resistance mutations to targeted therapies were numerically more frequent in NCI-MATCH than TCGA tumours, but not markedly.

The study team hypothesised that the genetic complexity of the NCI-MATCH cohort, with a median of three lines of prior, mostly cytotoxic chemotherapy, would be far greater than that seen in the largely untreated primary tumours of TCGA. However, they found that the frequencies of the 10 genes most commonly harbouring mutations in the seven tumour types compared across both cohorts were broadly similar.

The authors concluded that their findings support the feasibility and efficiency of NGS to triage patients to investigational therapy, provided that a sufficiently large pool of agents is made available. Co-occurring resistance mutations were common. Therefore, the molecular landscape of the patient population with relapsed, refractory advanced cancer strongly endorses a shift to investigation of combination targeted therapy regimens in such genetically complex tumours, notably those harbouring multiple actionable molecular alterations.

This work was supported by multiple NCI grants.


Flaherty KT, Gray RJ, Chen AP, et al. Molecular Landscape and Actionable Alterations in a Genomically Guided Cancer Clinical Trial: National Cancer Institute Molecular Analysis for Therapy Choice (NCI-MATCH). JCO; Published online 13 October 2020. DOI: 10.1200/JCO.19.03010.

This site uses cookies. Some of these cookies are essential, while others help us improve your experience by providing insights into how the site is being used.

For more detailed information on the cookies we use, please check our Privacy Policy.

Customise settings