Proteomic Test Stratifies Patients for Immunotherapies

Anti PD-1 inhibitors are the treatments of choice in second line non-small cell lung cancer (NSCLC), however they only benefit a fraction of patients, while existing biomarkers such as PD-L1 expression correlate with outcomes, but are not sufficient to guide therapy.

A serum-based test, BDX008, utilizing mass spectrometry was developed to assist in treatment selection for melanoma patients being considered for nivolumab. The possible role of BDX008 in NSCLC patients treated with nivolumab was evaluated and additionally, the role of the commercially available VeriStrat test, known to have prognostic and predictive roles in NSCLC2, was investigated in nivolumab-treated patients for the first time.


Scientists at the National Institute for Cancer Research (Genova, Italy) collaborating with those at Biodesix, Inc (Boulder, CO, USA) enrolled previously treated patients with advanced NSCLC (both squamous and non-squamous histology) in a single-institution translational study. Patients in the study were treated with nivolumab, an anti-PD-1 agent, and serum samples were evaluated using both an investigational test developed for use in melanoma, and with the Biodesix VeriStrat testing.

Spectra were collected from pre-treatment serum samples on matrix-assisted laser desorption/ionization (MALDI) time of flight mass spectrometer using standard VeriStrat acquisition procedures or “deep MALDI” approach for BDX008 testing. The BDX008 test generates a classification of positive (BDX008+, good outcomes) or negative (BDX008-, poor outcomes); VeriStrat classifies samples as VeriStrat Good (VS Good) or VeriStrat Poor (VS Poor).

The team reported that no correlations were found between objective response or disease control rates and BDX008 or VeriStrat tests by any of the response definition. Early death (ED), defined as death before the first radiological assessment, during 1-4 cycles, occurred in 10 patients (26.3%) classified as BDX008- versusone patient (4.6%) classified as BDX008+. By VeriStrat, ED occurred in two patients (5.4%) classified as VS Good and nine patients (39.1%) classified as VS Poor.

David Brunel, Chief Executive Officer of Biodesix, said, “Our testing methodology, which measures hundreds of proteins in a single assay and then utilizes the machine learning intelligence of our Diagnostic Cortex is uniquely suited to characterizing different immune states and tumor immune escape strategies over time, with the goal of enabling physicians to provide advanced precision medicine.” The study was presented at the 17th World Conference on Lung Cancer, hosted by the International Association for the Study of Lung Cancer held December 4-7, 2016, in Vienna, Austria.

Related Links:
National Institute for Cancer Research
Biodesix