Exploratory ctDNA analyses for the EVOKE-1 study in metastatic non-small cell lung cancer (mNSCLC).

Abstract

8522 Background: In NSCLC, ctDNA analysis complements assessment of clinical efficacy and identifies molecular alterations that may be prognostic or predictive of treatment. EVOKE-1 compared sacituzumab govitecan (SG) vs docetaxel in mNSCLC that had progressed on platinum-based and anti–PD-(L)1 therapy. Although statistical significance was not met, SG showed numerical improvement in overall survival (OS) vs docetaxel. Here, we assessed the value of ctDNA as a biomarker in available samples from EVOKE-01. Methods: The biomarker evaluable population (BEP) comprised 497 patients (pts), representing 82% of the ITT population. Cell-free DNA was extracted from blood collected at baseline and cycle 2 day 1 (C2D1). Samples were analyzed using the Guardant Infinity assay, a tumor-agnostic platform that measures ctDNA levels and gene variants from a comprehensive gene panel. Results: ctDNA was detected at baseline in 449 pts (90.3% of BEP), and higher ctDNA was a negative prognostic for OS, regardless of treatment. mOS was 12.7 vs 10.0 mos (HR 1.58, 95% CI: 1.11–2.26) for < vs ≥ median ctDNA level with SG and 10.8 vs 7.2 mos (HR 1.78, 95% CI: 1.28–2.46) with docetaxel. ctDNA was undetectable in 48 pts who had a longer survival (mOS NR in either arm) than those with detected ctDNA. At C2D1, median ctDNA reduction was 59% and 75% with SG and docetaxel ( P =.33); ≥50% reduction was achieved in 103 (44%) vs 121 pts (51%), respectively. Changes in ctDNA levels at C2D1 were prognostic, with pts achieving ≥50% reduction having a longer OS than those with <50% reduction of ctDNA. Actionable genomic alterations identified included KRAS , EGFR , ALK , ROS , ERBB2 , MET , and NTRK alterations. As expected from the required local/central testing for mutations at study entry, only a small number of pts with EGFR/ALK alterations were identified. KRAS mutations had a negative prognostic effect, whereas the group of pts with EGFR/ALK/ROS alterations was too small for conclusive results. Analysis of mutations ( TP53 , KEAP1 , STK11 ) potentially contributing to anti–PD-(L)1 resistance showed that between SG and docetaxel, 167 (68%) and 171 pts (68%) had TP53 mutations and 184 (75%) and 194 pts (77%) had ≥1 of these 3 mutations. The frequency of these mutations was similar across pts with PD/SD and those with CR/PR as best response to last prior PD-(L)1 therapy. Harboring TP53 mutations was a negative prognostic factor in both arms: mOS was 11.3 mos vs NA (HR 1.67, 95% CI: 1.11–2.54) with TP53 mutation vs wildtype with SG and 9.2 vs 13.9 mos (HR 1.71, 95% CI: 1.18–2.47), respectively, with docetaxel. KEAP1/STK11 mutations were also negative prognostic markers. Conclusions: This analysis did not identify differences in ctDNA clearance with SG vs docetaxel. Regardless of treatment and in line with previous reports, high ctDNA was a negative prognostic marker for OS in mNSCLC. Furthermore, alterations of KRAS , TP53 , and KEAP1/STK11 represent negative prognostic factors. Clinical trial information: NCT05089734 .