Association of Driver Oncogenic Alterations with SUVmax, Preoperative Serum Calcium, and Smoking Status in Surgically...

Background: Non-small-cell lung cancer (NSCLC) is driven by distinct oncogenic alterations with important therapeutic and prognostic implications. Noninvasive biomarkers that predict molecular status in surgically resectable disease may aid in their management. We investigated the association of preoperative primary-tumor SUVmax on PET/CT, smoking history, and corrected serum calcium levels with driver oncogenic alterations and PD-L1 expression in surgically resected NSCLC. Methods: We retrospectively studied 170 patients with surgically resected NSCLC at a single tertiary center. Resected tumors were assessed for EGFR, KRAS, and BRAF mutations, ALK and ROS1 rearrangements, and PD-L1 expression. Associations between molecular status, PD-L1 expression, and clinicometabolic parameters were evaluated using univariate analyses and multivariable regression models. Results: A driver alteration was detected in 51.2% of tumors, and 30% of evaluable cases showed high PD-L1 expression (≥50%). Corrected serum calcium was positively correlated with SUVmax and emerged as the strongest independent predictor retained in the final linear regression model, with pack-years also contributing independently. Most molecular subgroups did not show significant differences in SUVmax. EGFR-mutated tumors showed a trend toward a lower SUVmax compared with EGFR wild-type tumors, although this did not reach statistical significance. Smoking history was not significantly associated with PD-L1 expression, and pack-years did not differ significantly across the molecular groups examined. Conclusions: In this cohort of surgically resected NSCLC, preoperative corrected serum calcium and smoking exposure were more closely associated with tumor metabolic activity than with specific molecular alterations. These findings suggest that simple clinical and biochemical parameters may provide complementary information, although their utility for discriminating individual molecular subgroups appears limited.