Comparative risk of secondary central nervous system metastases in metastatic non-small cell lung cancer: A real-world propensity-matched analysis of pembrolizumab vs atezolizumab monotherapy conducted on TriNetX

Background: Bone and bone-marrow metastases significantly worsen morbidity in metastatic non-small cell lung cancer (mNSCLC). Comparative real-world data assessing skeletal metastatic progression between pembrolizumab and atezolizumab monotherapy remain limited. Methods: We performed a retrospective cohort analysis using the TriNetX US Collaborative Network (69 HCOs). Stage IV mNSCLC patients treated with first-line pembrolizumab or atezolizumab monotherapy were identified. Propensity score matching (1:1) produced balanced cohorts (n=7,768 each). Outcomes were assessed over 365 days following treatment initiation. The primary endpoint was new-onset secondary bone/bone-marrow metastases (ICD-10 C79.51, C79.52, C79.5). Patients with prior skeletal metastases were excluded from outcome analyses. Results: Following exclusions, 5,277 atezolizumab and 5,780 pembrolizumab patients were analyzed. Atezolizumab was associated with a significantly higher incidence of bone/bone-marrow metastases compared with pembrolizumab (16.5% vs 10.8%; RR 1.53; 95% CI 1.39-1.68; p<0.001). Kaplan-Meier analysis demonstrated reduced metastasis-free survival with atezolizumab (79.5% vs 86.8%, Log-Rank p<0.001). Hazard of metastatic progression was significantly higher with atezolizumab (HR 1.62; 95% CI 1.46-1.80). Atezolizumab also demonstrated a greater mean number of metastatic instances (0.93 vs 0.55; p<0.001). Conclusions: In this large real-world matched cohort, atezolizumab monotherapy was associated with a significantly greater 1-year risk of secondary bone or bone-marrow metastases compared with pembrolizumab. These findings suggest clinically meaningful differences in skeletal metastatic progression between ICIs and warrant further prospective and mechanistic investigation.