Lorlatinib-Associated Dyslipidemia in ALK-Positive NSCLC: Atherosclerotic Risk and the Unresolved Question of Venous Thromboembolism.
Srirangadhamu Gopu Sooraj S, Hasan Shadali Abdul Khader Abul A, Patel Anjali Pareshbhai AP, Nadeem Ahmed A et al.
Lorlatinib, a brain-penetrant third-generation anaplastic lymphoma kinase (ALK) tyrosine kinase inhibitor, has markedly improved outcomes in ALK-positive non-small cell lung cancer (NSCLC). In the CROWN trial, median progression-free survival had not been reached with first-line lorlatinib after 7 years of follow-up, compared with 9.1 months for crizotinib, with 7-year progression-free survival of 55% versus 3%. Median treatment duration was 62.6 months versus 9.6 months. This duration extends cumulative exposure to lorlatinib's characteristic metabolic toxicity: hypercholesterolemia and hypertriglyceridemia. At 7 years, these occurred in 73% and 71% of patients receiving lorlatinib, with grade 3 or 4 events in 22% and 26%. Onset is rapid, at a median of about 2 weeks. At 5 years, any-grade elevations had persisted for a median of 36 to 38 months; 71% of events were managed with lipid-lowering agents. ALK-positive NSCLC carries a high risk of venous thromboembolism across observational cohorts and a meta-analysis, more so for venous than arterial events. Laboratory evidence suggests the EML4-ALK fusion raises tissue factor through the pERK1/2-AP-1 pathway. At 7 years, aggregate cardiovascular adverse events were similar between treatment groups, and venous events were not more frequent with lorlatinib; small event counts and differing treatment exposure limit comparison of individual events. Statin choice is limited because lorlatinib is a moderate cytochrome P450 3A inducer; rosuvastatin and pravastatin are preferred. We review lorlatinib-associated dyslipidemia and its management and separate two claims: prompt, guideline-concordant lipid management is supported by established atherosclerotic prevention principles, whereas any effect on venous thromboembolism remains untested.