Results from a phase III trial presented today suggest that Biodesix's VeriStrat proteomics-based test may be able to predict which patients may benefit from treatment with Tarceva (erlotinib) for non-small cell lung cancer, Biodesix said.
At the Second European Lung Cancer Conference in Switzerland, David Carbone, professor of medicine and cancer biology at the Vanderbilt University School of Medicine, presented data that applied the VeriStrat test to a group of 441 patients and classified them either as "VeriStrat Good," meaning they are likely to benefit from erlotinib therapy; or "VeriStrat Poor," meaning they are likely to "derive little therapeutic benefit.".
The 441 patients come from a total of 729 patients who are involved in NCIC Clinical Trials Group study BR.21, a phase III multicenter trial of erlotinib versus placebo in previously treated patients with NSCLC.
The study, which was funded by Biodesix, found that VeriStrat Good patients had a median survival of 10.5 months compared to VeriStrat Poor patients who had a median survival of 4 months.
VeriStrat Good patients on placebo had a median survival of 6.6 months while VeriStrat Poor patients on placebo had a median survival of 3.1 months.
“The bottom line is that the proteomic test – comparing ‘good’ and ‘poor’ profiles – was strongly prognostic in both erlotinib and placebo arms,” Carbone, who sits on Biodesix's advisory board, said in a statement issued by the organizers of the conference.
Fluorescence in situ hybridization is a better overall predictor of benefit, he said, but the method requires "adequate biopsy tissue," which was available in the study in only 22 percent of patients.
"With the serum test, 99 percent of patients had a successful determination of proteomic status," Carbone said. "Thus, I think this test may be of potential value in identifying a subgroup of patients with a good prognosis and who are likely to have response to erlotinib; it may be of particular value for those in whom tumor tissue is inadequate or unavailable.”
The study also found that VeriStrat Good patients had a "significantly" higher tumor response rate, 9.8 percent, than VeriStrat Poor patients, 0.9 percent. VeriStrat status did not correlate with other biomarkers analyzed for the study, Biodesix said.
Continuing analysis is being conducted of VeriStrat in other patient subsets and in patients with known and unknown EGFR mutation status, the company added.
Based in Broomfield, Colo., Biodesix develops products to help clinicians make treatment decisions. It uses a proprietary MALDI mass spec-based technology called ProTS.
VeriStrat is serum-based test. After the serum is analyzed by MALDI, the resulting raw spectra are processed using Biodesix's proprietary technology to generate comparable data. The processed data are then analyzed using the VeriStrat classifier and classified as VeriStrat Good, VeriStrat Poor, or indeterminate based on replicates.
The test was launched in May 2009.
According to the American Cancer Society, between 80 percent and 90 percent of lung cancers are the non-small cell type, which would translate to between 175,550 and 197,500 cases of all estimated lung cancers diagnosed in 2009.