Ruiqi Song, Xinyue Li, Na Han, Dujiang Liu, Yanjun Xu
Within this lorlatinib-treated cohort, 1L lorlatinib was associated with more favorable lorlatinib-specific systemic, CNS, and survival outcomes than Post-2G lorlatinib. These findings are hypothesis-generating and do not constitute an all-comer comparison of complete first-line treatment strategies.
BACKGROUND: The phase III CROWN study established lorlatinib as a standard first-line therapy for advanced anaplastic lymphoma kinase (ALK)-positive non-small cell lung cancer (NSCLC). In routine practice, it may be used upfront or after a second-generation (2G) ALK tyrosine kinase inhibitor (TKI), but real-world lorlatinib-specific systemic and central nervous system (CNS) outcomes by timing remain inadequately defined.
OBJECTIVES: To compare lorlatinib-specific progression-free survival (PFS), tumor response, overall survival (OS), CNS-first progression, and safety between 1L lorlatinib and Post-2G lorlatinib.
DESIGN: Single-center, retrospective, real-world cohort study.
METHODS: Consecutive patients with ALK-positive advanced NSCLC who initiated lorlatinib at Zhejiang Cancer Hospital between January 2023 and August 2025 were included, with follow-up updated through June 2026. Patients were classified as first-line lorlatinib (1L lorlatinib) or post-2G ALK TKI lorlatinib (Post-2G lorlatinib). The primary endpoint was lorlatinib-specific PFS from lorlatinib initiation. CNS progression was estimated using competing-risk methods.
RESULTS: Among 112 patients, 55 received 1L lorlatinib and 57 received Post-2G lorlatinib. Median lorlatinib-specific PFS was not reached versus 13.0 months (hazard ratio [HR], 0.09; 95% confidence interval [CI], 0.04-0.21; P<0.001), and ORR was 92.7% versus 66.7% (P<0.001). CNS-first progression occurred in 0 versus 19 patients; the 36-month cumulative incidence was 0% versus 43.9% (Gray's test P<0.001). Median OS from lorlatinib initiation was not reached versus 28.8 months (HR, 0.22; 95% CI, 0.09-0.53; P<0.001). Lorlatinib-emergent TRAEs were mainly grade 1-2, and permanent discontinuation due to TRAEs was uncommon.
CONCLUSIONS: Within this lorlatinib-treated cohort, 1L lorlatinib was associated with more favorable lorlatinib-specific systemic, CNS, and survival outcomes than Post-2G lorlatinib. These findings are hypothesis-generating and do not constitute an all-comer comparison of complete first-line treatment strategies.