Can Nutritional and Inflammatory Markers Be Combined to Develop an Accurate Survival Model for Surgical Decision-making in Patients With Lung Cancer Bone Metastases?

Background Targeted and immunotherapies have improved the survivorship of patients with lung cancer and bone metastases. However, most existing models were developed during the chemotherapy era and do not accurately reflect survival outcomes in the current therapeutic context, leading to limited clinical applicability. Additionally, a wide range of inflammatory and nutritional markers has been identified as useful for cancer survival assessment. The most effective selection or combination of these markers for evaluating survival in patients with lung cancer and bone metastases, especially within the framework of modern multimodal treatments, has not yet been systematically investigated. Questions/purposes (1) What combination of inflammatory and nutritional markers can better evaluate survival in patients with lung cancer and bone metastases? (2) Can an accurate lung cancer–specific bone metastasis model be constructed by integrating the above marker combination with the latest advancements in targeted and immunotherapies to guide clinical decision-making? Methods Between January 1, 2019, and June 1, 2024, we treated 319 patients with bone metastases from lung cancer at Guangdong Provincial People’s Hospital, a tertiary academic medical center in Guangzhou, PR China. We considered patients with severe pain, pathological fracture, skeletal instability, or spinal cord/nerve compression as potentially eligible. Of the potentially eligible patients, 7% (23 of 319) were excluded due to repeated admissions, 4% (14 of 319) declined surgical intervention, 0.6% (2 of 319) died within 1 week postoperatively, 0.6% (2 of 319) were lost before the minimum follow-up period, and 0.3% (1 of 319) had incomplete datasets, leaving 87% (277 of 319) of patients for analysis. Of those, 277 underwent surgical procedures, including open spinal decompression (44% [123 of 277]), tumor resection with replacement or internal fixation (25% [68 of 277]), and minimally invasive procedures (31% [86 of 277]). The mean age was 61 ± 11 years, and 63% (174 of 277) of patients were male. The most common histological subtype was adenocarcinoma (82% [228 of 277]), followed by squamous cell carcinoma (14% [38 of 277]). Bone metastases primarily involved the spine (65% [179 of 277]), limbs (22% [62 of 277]), and pelvis (10% [28 of 277]). Candidate variables were selected using stepwise regression based on the minimum Akaike information criterion, and their associations with survival were assessed using Cox regression and restricted cubic splines. A nomogram was developed and validated through calibration curves, a decision curve analysis, and internal validation via bootstrap resampling. Discrimination was assessed through time-dependent receiver operating characteristic (time-ROC) analysis to calculate the area under the curve (AUC). Results This study identified the systemic inflammation response index (systemic immune-inflammation index [SIRI], neutrophil count × monocyte count/lymphocyte count [all in 10 9 cells/L]) and prognostic nutritional index (PNI) (albumin [g/L] + 5 × lymphocyte count [10 9 cells/L]) as associated factors. Specifically, female sex (HR 0.68 [95% confidence interval (CI) 0.48 to 0.96; p = 0.03), fewer bone metastases (< 3 versus ≥ 3, HR 0.62 [95% CI 0.43 to 0.91]; p = 0.01), higher Eastern Cooperative Oncology Group (ECOG) score (0 to 2 versus 3 to 4, HR 0.58 [95% CI 0.41 to 0.83]; p = 0.003), higher PNI (HR 0.96 [95% CI 0.94 to 0.99]; p = 0.02), and first-line treatment (targeted or immunotherapy versus chemotherapy, HR 0.59 [95% CI 0.42 to 0.83]; p = 0.002) were independently associated with improved survival, whereas higher SIRI (HR 1.04 [95% CI 1.01 to 1.08]; p = 0.01) indicated decreased survival. The bootstrap calibrated AUCs were 0.81 (95% CI 0.76 to 0.89), 0.83 (95% CI 0.77 to 0.88), and 0.82 (95% CI 0.77 to 0.88), for 3-, 6-, and 12-month survival, respectively. Calibration curves illustrated good agreement between estimated and observed survival rates, and the decision curve analysis validated the clinical applicability across diverse risk thresholds. The final models were developed into an online application, which is available at https://lbp-apps.shinyapps.io/lbp-app. Conclusion As inflammatory and nutritional markers, PNI and SIRI exhibit great value in assessing survival time for patients with lung cancer and bone metastases. The addition of immunotherapy and targeted therapy has notably improved survival outcomes. By incorporating variables such as sex, the number of bone metastases, ECOG performance status, and first-line treatment, the model provides a reliable tool for guiding surgical decision-making. Level of Evidence Level III, prognostic study.