Abstract
PURPOSE: The VISION trial of tepotinib, a selective MET inhibitor, enrolled patients with non-small cell lung cancer and prospectively detected MET exon 14 (METex14) skipping in liquid biopsies (LBx) and/or tissue biopsies (TBx). We evaluated patient characteristics and outcomes according to METex14 positivity in LBx (LBx-positive) or TBx (TBx-positive). EXPERIMENTAL DESIGN: METex14 was centrally assessed by next-generation sequencing of ctDNA from LBx (Guardant360/ArcherMET) and/or RNA from TBx (Oncomine Focus/ArcherMET) or, in Japan only, local TBx PCR. Parallel LBx/TBx testing was recommended but not mandatory. Eligibility required LBx-positive or TBx-positive status. ctDNA burden was analyzed in patients with baseline Guardant360 data. RESULTS: METex14 was detected in 469 of 7,937 prescreened/screened patients, 313 of whom were enrolled (TBx-positive, n = 208; LBx-positive, n = 178). LBx-positive patients had higher radiographic tumor burden than TBx-positive patients, including higher median sum of target lesion diameters per RECIST v1.1 (67.1 vs. 55.2 mm) and more patients with ≥3 target lesions (27.5% vs. 18.8%). In 180 TBx-positive patients with matching LBx results, objective response rates were slightly higher in TBx-positive/LBx-positive patients, but TBx-positive/LBx-negative patients had longer duration of response, progression-free survival, and overall survival. In ctDNA analysis (n = 165), detectable baseline ctDNA burden was associated with shorter progression-free survival and overall survival. CONCLUSIONS: Tepotinib had robust, durable activity in TBx-positive/LBx-negative and TBx-positive/LBx-positive patients. Although LBx is a complementary method to TBx for detecting METex14, it may preferentially select patients with higher tumor burden and poorer prognosis. Undetectable METex14 in baseline ctDNA (due to low ctDNA shedding) may define more favorable treatment outcomes.