Thoracic Oncology Research Updates

A Markov model with >100,000 Monte Carlo simulations compared upfront stereotactic radiosurgery (SRS) versus hippocampal-avoidant whole-brain radiotherapy (HA-WBRT) for brain metastases from small cell lung cancer. SRS was more costly and more effective in the majority of simulations (base-case ICER $2,126.88); MRI and hospice costs, the utility decrement from HA-WBRT–related neurocognitive decline, and transition probabilities for control, recurrence, and death had the largest influence on the incremental cost-effectiveness ratio.

In a propensity‑matched real‑world cohort of adults with newly diagnosed metastatic EGFR‑mutant NSCLC (777 per arm), first‑line osimertinib yielded longer median overall survival than second‑generation EGFR TKIs (median 53.4 vs 33.2 months; HR 0.618). Sequential use of second‑generation TKIs followed by osimertinib achieved similar overall survival but with higher toxicity, whereas first‑line osimertinib reduced hospitalizations, ICU admissions, and severe infections and was associated with substantial health care savings despite higher drug costs; patients with brain metastases derived greater survival benefit with osimertinib (HR 0.563).

In a Viewpoint addressing advanced NSCLC, the authors critique a JAMA Network Open analysis of Flatiron Health data that suggested pembrolizumab rechallenge after chemo‑immunotherapy progression may confer a survival benefit. They argue that retrospective RWD analyses used for cost‑effectiveness modeling can be biased if unadjusted for key patient variables (eg, PD‑L1, tumor burden, performance status), which may inflate ICER/QALY estimates and mislead reimbursement decisions, and they call for greater analytic transparency and pre‑registered sensitivity analyses.