Abstract CT025: Impact of the EML4-ALK fusion variant on the efficacy of lorlatinib in patients (pts) with ALK-positive advanced non-small cell lung cancer (NSCLC)

Abstract

Abstract BACKGROUND: Lorlatinib is a selective, potent, brain-penetrant, 3rd-generation (gen) ALK/ROS1 tyrosine kinase inhibitor (TKI) preclinically active against most known ALK resistance mutations. In pts with ALK+ advanced NSCLC with disease progression following 2nd-gen ALK TKIs, lorlatinib has shown robust clinical activity and we found that tumor genotyping for ALK mutations may identify such pts more likely to respond to lorlatinib treatment. Prior analyses have suggested that EML4-ALK variant type may influence ALK TKI treatment benefit. To further identify molecular correlates of response, we performed an exploratory subgroup analysis by EML4-ALK variant type and ALK resistance mutations in pts previously treated with 2nd-gen ALK TKIs and who received the recommended Ph 2 dose of lorlatinib (100 mg once daily). METHODS: Baseline plasma samples were collected from pts with ALK+ NSCLC with ≥1 prior 2nd-gen ALK TKI enrolled in the ongoing registrational Ph 2 study (NCT01970865). Circulating free DNA (cfDNA) was analyzed using Guardant360 (Guardant Health, Inc., CA, USA) to determine EML4-ALK variant and ALK kinase domain mutations. Objective response rate (ORR) and duration of response (DOR), by independent central review, were evaluated according to EML4-ALK variant type and ALK resistance mutation status. RESULTS: ALK fusions were detectable in 64 (41.0%) of 156 pt plasma samples. EML4-ALK variants 1, 2, and 3 were detected at frequency of 17.3%, 2.6% and 15.4%, respectively. Other EML4-ALK variants (including the variant types 4, 5, 7, and 8) were also detected in 3.2% of pts, as well as some other less frequent fusion partners (e.g., KIF5B) in 2.6% of pts. Based on cfDNA, ALK resistance mutations were detected in 40 pts; of whom, 6 pts harbored EML4-ALK variant 1, 1 pt had variant 2 and 18 had variant 3. The G1202R/Del mutation was detected in 23 pts samples; of which, 15 (65.2%) also harbored EML4-ALK variant 3. ORR was 33.3% (95% CI 16.5-54.0), 75.0% (95% CI 19.4-99.4) and 45.8% (95% CI 25.6-67.2) for variants 1, 2 and 3, respectively, while median DOR was similar for pts with variant 1 or 3 (both 6.9 months). Median DOR was not reached for variant 2. Of note, no responses were observed in the pts with other types of ALK rearrangements detected. Finally, ALK fusions were not detected in the cfDNA of 92 pts (59.0%), including 12 pts who had detectable ALK mutations. Among these 92 pts, ORR was 39.1% (95% CI 29.1-49.9) and median DOR was 7.1 months (95% C 5.5-not reached). Confirmation of these results in the tumor tissue is ongoing. CONCLUSION: In this heavily pretreated group of ALK+ NSCLC pts, the presence of an ALK resistance mutation might enrich for EML4-ALK variants 1 and 3. Lorlatinib exhibited antitumor activity irrespective of EML4-ALK variant and across a variety of ALK resistance mutations. Citation Format: Todd M. Bauer, Jean-François Martini, Benjamin Besse, Chia-Chi Lin, Ross A. Soo, Gregory J. Riely, Sai-Hong Ignatius Ou, Francesca Toffalorio, Antonello Abbattista, Holger Thurm, D. Ross Camidge, Steven Kao, Rita Chiari, Shirish Gadgeel, Enriqueta Felip, Alice T. Shaw, Benjamin J. Solomon. Impact of the EML4-ALK fusion variant on the efficacy of lorlatinib in patients (pts) with ALK-positive advanced non-small cell lung cancer (NSCLC) [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr CT025.