Savolitinib (Savo) combined with osimertinib (osi) versus chemotherapy (chemo) in EGFR-mutant (EGFRm) and MET -amplification ( MET amp) advanced NSCLC after disease progression (PD) on EGFR tyrosine kinase inhibitor (TKI): Results from a randomized phase 3 SACHI study.

LBA8505 Background: Savo, a highly selective MET -TKI, combined with osi, may overcome acquired MET -driven resistance in EGFRm advanced NSCLC after PD on EGFR-TKIs. Here we report primary results of the prespecified interim analysis (IA) in SACHI study, comparing efficacy and safety of savo + osi with chemo in this disease setting. Methods: In this randomized, open-label, phase 3 study, 250 EGFRm and MET amp advanced NSCLC patients (pts) post PD on first-line EGFR-TKI were planned ( MET copy number ≥5 or MET /CEP7 ratio of ≥2.0 by FISH for pts with prior 1 st /2 nd generation [G] EGFR-TKI; MET copy number ≥10 for pts with prior 3 rd G EGFR-TKI); T790M negative after PD on 1 st /2 nd G EGFR-TKI was required. Eligible pts were randomly assigned (1:1) to receive savo 400 or 600 mg QD (for body weight of < 50, or ≥ 50 kg respectively) + osi 80 mg QD, or chemo (pemetrexed + carboplatin/cisplatin), stratified by brain metastases, prior use of 3G EGFR-TKI, and type of EGFR mutations. Crossover to savo + osi after IRC-PD was permitted for chemo group. The primary endpoint, PFS by investigator (INV) per RECIST 1.1, was hierarchically tested via a stratified log-rank test in 3G EGFR-TKI treatment-naïve set firstly, then in ITT set. This is a prespecified IA conducted via an independent data monitoring committee to assess efficacy superiority or sample size re-estimation. Results: From 15 Oct 2021 to 30 Aug 2024 (DCO for IA), 211 pts were randomized to receive savo + osi or chemo (n=106 vs 105). Baseline characteristics were well balanced. mPFS by INV was significantly longer with savo + osi vs chemo in both 3G EGFR-TKI treatment-naïve set and ITT set ( p < 0.0001 in both sets), which met prespecified IA efficacy boundary ( p <0.0099 and 0.0228 in 2 sets, respectively); in 3G EGFR-TKI treated pts, mPFS was also significantly prolonged with savo + osi (6.9m vs 3.0m, HR=0.32, p < 0.0001). IRC-assessed PFS benefits were consistent (table). OS was immature at this DCO. Grade ≥3 TEAE occurred in 56.6% vs 57.3% of pts with savo + osi vs chemo; savo + osi had lower rates of hematologic events than chemo. Conclusion: Savo + osi significantly improved PFS versus chemo in MET amp NSCLC post EGFR-TKI, and the combination was safe and well tolerated. Savo + osi is a potential new treatment option for this genomically defined population. Clinical trial information: NCT05015608 . ITT set Savo + osiN=106 ChemoN=105 Hazard ratio/Odds ratio Two sided -p value mPFS (95% CI) (INV), m 8.2 (6.9, 11.2) 4.5 (3.0, 5.4) 0.34 <0.0001 mPFS (95% CI) (IRC), m 7.2 (5.7, 11.1) 4.2 (4.0, 5.7) 0.40 < 0.0001 ORR (95% CI) (IRC), % 63.2 (53.3, 72.4) 36.2 (27.0, 46.1) 3.05 < 0.0001 mDoR (95% CI) (IRC), m 9.7 (5.8, 12.4) 4.3 (2.8, 5.1) NA NA mOS (95%CI), m* 22.9 (16.8, NE) 17.7 (14.9, 26.3) 0.84 0.4191 *52.4% of pts in chemo group were crossover to receive savo + osi or other MET Inhibitors.