Real-world comparative safety of first-line lorlatinib versus alectinib in metastatic ALK-rearranged non-small cell lung cancer: A propensity score-matched analysis.
Shankar Biswas, Yashasvi Srivastava, Rahul Falodia, Ayman Hamadttu283
Background:
Lorlatinib and alectinib are both NCCN-preferred first-line therapies for ALK-positive metastatic NSCLC, yet no head-to-head randomized trials exist. Lorlatinib carries a distinct safety profile including hyperlipidemia and CNS effects, while alectinib is associated with hepatotoxicity and GI toxicities. Real-world comparative safety data are needed to inform treatment selection.
Methods:
We conducted a retrospective propensity score-matched (PSM) cohort study using the TriNetX Global Collaborative Network (171 healthcare organizations). Adults with metastatic NSCLC (ICD-10: C34 + C78/C79) receiving first-line lorlatinib (Cohort A) or alectinib (Cohort B) were identified. PSM was performed on age, sex, race/ethnicity, brain metastases (C79.31), Charlson comorbidity components, and baseline labs (albumin, hemoglobin, creatinine, LDH). Safety outcomes were assessed using risk analysis (odds ratios) and Kaplan-Meier survival analysis (hazard ratios) with outcomes occurring prior to the index event excluded.
Results:
After PSM, 844 patients were matched per cohort. Median follow-up was 350 days (lorlatinib) vs 697 days (alectinib). Baseline characteristics were well-balanced (all standardized differences < 0.1). Key safety findings are in the Table. Notably, the divergence between risk-based and time-to-event analyses for several outcomes (cognitive disorders, mood disorders, bradycardia) reflects the substantial follow-up imbalance, with alectinib patients having nearly double the observation time.
Conclusions:
In this large real-world PSM analysis, lorlatinib was associated with significantly higher hypercholesterolemia risk but lower cumulative rates of mood disorders, bradycardia, constipation, and weight gain compared with alectinib. CNS-related effects and time to new intracranial metastases were comparable between the two agents. These findings suggest distinct but manageable safety trade-offs and support individualized first-line ALK-TKI selection. Sensitivity analyses are warranted to account for the observed follow-up asymmetry.