Skip to main content
Premium Trial:

Request an Annual Quote

Study Describes Design, Validation of Foundation Medicine's Blood-Based TMB Test

NEW YORK (GenomeWeb) – Researchers from Genentech, Foundation Medicine, UC Davis, and other medical centers, have published a report on the development and early validation of Foundation's planned blood-based tumor mutational burden test.

Appearing today in Nature Medicine, the study describes the development of the test and its characteristics, and its retrospective validation in two cohorts. Investigators demonstrated, by applying the assay to samples from two clinical trials, that blood-based TMB (bTMB) could reproducibly identify lung cancer patients who respond to immunotherapy treatment with Roche/Genentech's atezolizumab (Tecentriq).

Foundation Medicine announced its intentions late last year to develop a blood-based version of its existing tissue TMB test as a companion diagnostic to atezolizumab in first-line treatment of non-small cell lung cancer patients.

Investigators had previously presented some of the data from the newly published study at the European Society for Medical Oncology Congress in Madrid last September. And Foundation said at the time that the bTMB test was being further studied in two ongoing clinical trials — B-F1RST and BFAST.

More recently, the US Food and Drug Administration granted a Breakthrough Device designation for an expanded version of Foundation Medicine's existing FoundationACT liquid biopsy test to include assessment of microsatellite instability (MSI) and bTMB.

And interim data from BFIRST presented at the annual meeting of the American Society for Clinical Oncology in June confirmed that the assay shows utility in identifying patients likely to respond to the drug.

As described by authors of the Nature Medicine study today, Foundation's bTMB assay uses the same hybridization-capture methodology as Foundation's FDA-approved FoundationOne CDx NGS assay, and targets 1.1 Mb of genomic coding sequence. The analysis requires a minimum amount of ctDNA in the blood — an allele frequency of at least 1 percent — for optimal performance.

Investigators reported on the retrospective application of the test to more than 1,000 samples from patients with advanced NSCLC who participated in Genentech's Phase II POPLAR and Phase III OAK clinical trials. The POPLAR trial samples were used first, to identify blood-based TMB thresholds that reflect the discriminatory ability of tissue-based TMB.

The positive predictive agreement for different cutoff points ranged from about 86 percent to 100 percent, and negative predictive agreements were spread between 82 percent and 100 percent. Overall, investigators calculated that the assay's performance was optimized at three different cut-points — bTMB of ten or more, 16 or more, and 20 or more mutations — and so they selected these to test further.

Based on results in the POPLAR cohort, the investigators narrowed down to the 16-mutation cutoff point for analysis in the OAK study. According to the authors, OAK study patients with at least 16 total mutations as calculated by the bTMB assay had significantly improved progression-free survival when treated with atezolizumab versus docetaxel chemotherapy with a hazard ratio of 0.65.

In addition, patients' bTMB results did not appear to correlate with PD-L1 expression levels, suggesting that the test provides independent predictive information that can't be determined using PD-L1.

Although the authors wrote that it did not affect the results of the current study, the fact that the assay as it stands only counts SNVs and not indels means that it could conceivably miss the presence of a high TMB in a patient whose tumor has a large number of indels but relatively few SNVs.

Based on this, the group said that future versions of the assay will incorporate indels into the bTMB-calling algorithm.

As Foundation Medicine seeks to advance its own test, several other companies are also working on their own blood-based assays for mutational burden, or load.

Like Foundation, Personal Genome Diagnostics has also received a Breakthrough Device designation from the FDA for its Elio Plasma Resolve test.

And Guardant Health said in February that the FDA had granted its Guardant360 liquid biopsy assay Expedited Access Pathway designation

Guardant published a study last month, also in Nature Medicine, which demonstrated that its Guardant360 could predict responsiveness to pembrolizumab (Merck's Keytruda) in a group of 61 patients with advanced gastric cancer.

PGDx, meanwhile, has been working to consolidate IP regarding both tissue- and blood-based TMB, with a move this February to exclusively license various methods from Memorial Sloan Kettering.