CHICAGO – The VeriStrat test is useful in guiding second-line treatment decisions in non–small-cell lung cancer patients with unknown or wild-type epidermal growth factor receptor status, according to results from the prospective phase III PROSE trial.
Patients classified as VeriStrat "good" had the same median overall survival of 10.92 months with chemotherapy and of 10.95 months with erlotinib (Tarceva) (hazard ratio 1.06; P = .714).
VeriStrat "poor" patients, however, derived a significant survival benefit from chemotherapy, with a median survival of 6.38 months vs. 2.98 months with erlotinib (HR 1.72; P = .022).
The interaction between treatment outcome and VeriStrat classification was statistically significant (HR 1.85; P = .031), Dr. Vanessa Gregorc reported at the annual meeting of the American Society of Clinical Oncology.
Erlotinib, an antiepidermal growth factor receptor (EGFR) tyrosine kinase inhibitor (TKI), is frequently used in patients who harbor an EGFR mutation, but about 90% of non-Asian patients are EGFR wild-type and 35%-45% are never tested for EGFR status.
Patrice Wendling/IMNG Medical Media
Dr. Vanessa Gregorc
VeriStrat is a serum protein test that uses mass spectrometry technique – Matrix-Assisted Laser Desorption/Ionization-Time of Flight Mass Spectrometry (MALDI-TOF MS) – and proprietary algorithms to analyze proteins in a patient’s serum, according to the manufacturer, Biodesix.
The VeriStrat test is already known to be of value in identifying good versus poor prognosis in NSCLC patients. Initial retrospective validation studies showed that pretreatment classification with VeriStrat correlated with overall survival after treatment with EGFR-TKIs, but not with chemotherapy. Additional retrospective studies have shown the test to be both predictive and prognostic, explained Dr. Gregorc of the San Raffaele Scientific Institute, Milan.
There have been some doubts about whether the VeriStrat signature was only prognostic, but now "... we know VeriStrat is also predictive," said Dr. Luis Paz-Ares of Virgen Del Rocio University Hospital, Seville, Spain, who was invited to discuss the study.
Possibly more important, in a subanalysis restricted to only EGFR wild-type patients, the hazard ratio was even higher (1.94) and the VeriStrat treatment interaction was significant (P = .024), he added.
PROSE study
PROSE (Randomized Proteomic Stratified Phase III Study of Second Line Erlotinib versus Chemotherapy in Patients with Inoperable Non–Small-Cell Lung Cancer) involved 285 patients at 14 centers across Italy who had stage IIIb or IV inoperable NSCLC treated with one previous line of platinum-based chemotherapy and no EGFR-TKIs. Patients were stratified by ECOG performance status, smoking status, and blinded pretreatment VeriStrat classification, and randomly assigned within 5 working days of their blood draw to receive erlotinib or chemotherapy with pemetrexed or docetaxel at standard doses.
Among the 263 patients evaluable for the primary analysis, 129 received chemotherapy, of which 88 were VeriStrat classified as good and 41 as poor, and 134 received erlotinib, of which 96 were VeriStrat classified as good and 38 as poor.
EGFR mutation status analysis was performed in 190 of the 263 evaluable patients and indicated that 91 of 97 (94%) chemotherapy patients tested and 85 of 93 (91%) erlotinib patients tested were EGFR wild-type.
Median overall survival was 9.0 months with chemotherapy and 7.7 months with erlotinib (HR 1.14; P = .313), Dr. Gregorc reported.
In a survival analysis adjusted for several potential confounding factors, ECOG performance status 2 vs. 0-1 (HR 2.55; P less than .001) and VeriStrat poor vs. good classification (HR 1.89; P = .002) were prognostic for outcome independent of treatment. VeriStrat was predictive of a significant differential treatment benefit between chemotherapy and erlotinib (HR 1.98; P = .022).
During a discussion of the study, Dr. Tony Mok, lead investigator for the pivotal IPASS (Iressa Pan Asian Study), said he agreed that VeriStrat was predictive in patients with EGFR-negative status, but wasn’t sure about this conclusion for patients with unknown mutation status "because the proportion with EGFR-mutation positive likely had higher predictive power than what the VeriStrat can offer."
Dr. Gregorc said the sample sizes aren’t representative of what they’d planned to show, but that additional analyses are supportive of the findings.
"Our suggestion is, if you can, do the analysis with tissue," she said. "If you can’t or the patients have EGFR mutations, this is what you have to drive your clinical decision."
Another attendee described PROSE as an important trial, but said it "feels like the cake is only half baked" without data on progression-free survival and response rates. Dr. Gregorc responded that they didn’t have sufficient time to complete a thorough progression-free survival analysis before the meeting, and that these data will be presented at the European Society for Medical Oncology meeting this fall.