Main Session
Sep
28
PQA 02 - Lung Cancer/Thoracic Malignancies, Patient Reported Outcomes/QoL/Survivorship, Pediatric Cancer
2367 - Nutrition-Related Hematological and Body Composition Biomarkers in Stage III Unresectable NSCLC Treated with Immunochemoradiotherapy: A Retrospective Cohort Study
Presenter(s)
Xia Gan, MBBS - Hunan Cancer Hospital, Changsha, Hunan
X. Gan1, J. Fang1, L. Zhang2, W. Xie1, X. Sun1, Y. Li1, X. Zhang1, Q. Liu1, Y. Liu1, Z. Liu1, Y. Tang1, C. Zhou1, H. Liu1, and H. Wang3; 1Hunan Cancer Hospital & the Affiliated Cancer Hospital of Xiangya School of Medicine, Central South University, Changsha, Hunan, China, 2Hunan Cancer Hospital & the Affiliated Cancer Hospital of Xiangya School of Medicine, Central South University, Changsha, China, 3Hunan Cancer Hospital and The Affiliated Cancer Hospital of Xiangya School of Medicine, Central South University, Changsha, China
Purpose/Objective(s): To evaluate the prognostic value of nutrition-related hematological and body composition indicators in stage III unresectable non-small cell lung cancer (NSCLC) patients treated with immunotherapy combined with definitive chemoradiotherapy. Materials/Methods: This single-center retrospective study analyzed 124 patients treated between January 2020 and September 2022. Pretreatment hematological parameters (hemoglobin, albumin, platelets) and CT-based body composition metrics (muscle and adipose indices) were assessed. Prognostic factors were identified using Cox regression, and optimal cutoffs were determined via ROC analysis. A composite scoring system integrating key predictors was developed. Results: Univariate Cox regression identified seven potential prognostic indicators, including body mass index (BMI), hemoglobin-to-red blood cell distribution width ratio (HRR), platelet count (PLT), subcutaneous adipose tissue (SAT), skeletal muscle gauge (SMG), visceral-to-subcutaneous adipose tissue ratio (VSR), total adipose tissue index (TATI), all P < 0.2. Pearson correlation analysis confirmed no strong collinearity among these variables (|r| < 0.8). These variables were subsequently incorporated into a multivariate Cox proportional hazards model adjusted for clinical confounders, including sex, age, smoking history, family history, histology (squamous vs. non-squamous), TNM stage (AJCC 8th edition), Charlson Comorbidity Index (CCI), immunotherapy sequence (induction, consolidation or induction plus consolidation), and immune checkpoint inhibitor type. Four independent prognostic factors were established: PLT (=234.5 × 10?/L,HR = 3.21, 95% CI: 1.29–7.98, P = 0.012), VSR (=1.09,HR = 4.41, 95% CI: 1.67–11.65, P = 0.003), HRR (=1.2, HR = 0.31, 95% CI: 0.1–0.95, P = 0.04), TATI (=69.12cm2/m2, HR = 0.3, 95% CI: 0.11–0.85, P = 0.023). A scoring system was developed by assigning 1 point for each favorable prognostic criterion met (PLT <234.5 × 10?/L, VSR <1.09, HRR =1.2, TATI =69.12cm2/m2), with total scores ranging from 0 to 4. Initial analysis demonstrated a significant inverse association between cumulative score and progression-free survival (PFS, HR = 0.48, 95% CI: 0.33–0.70, P < 0.001), though no OS difference was observed (P = 0.067). To mitigate type II error risk due to small sample sizes in extreme score groups (score 0: n = 4,score 4: n = 3), we optimized risk stratification: High-risk group(scores 0–1,n =55), Intermediate-risk group(score 2,n =47), Low-risk group(scores 3–4,n = 22). Post-optimization analysis revealed significantly improved PFS in the low-risk group (HR = 0.61, 95% CI: 0.47–0.8, P < 0.001). The model exhibited moderate predictive accuracy, with an AUC of 0.70 for PFS stratification. Conclusion: Multidimensional indicators are critical for guiding personalized therapy. Integrating hematological and body composition biomarkers (PLT, HRR, VSR, TATI) effectively stratifies PFS outcomes in stage III unresectable NSCLC.