Construction of a prognostic model for extensive-stage small cell lung cancer patients undergoing immune therapy in northernmost China and prediction of treatment efficacy based on response status at different time points
Junjie Dang,Gang Xu,Ge Guo,Huan Zhang,Lihua Shang
DOI: https://doi.org/10.1007/s00432-024-05767-6
2024-05-15
Abstract:Background and purpose: Recently, the emergence of immune checkpoint inhibitors has significantly improved the survival of patients with extensive-stage small cell lung cancer. However, not all patients can benefit from immunotherapy; therefore, there is an urgent need for precise predictive markers to screen the population for the benefit of immunotherapy. However, single markers have limited predictive accuracy, so a comprehensive predictive model is needed to better enable precision immunotherapy. The aim of this study was to establish a prognostic model for immunotherapy in ES-SCLC patients using basic clinical characteristics and peripheral hematological indices of the patients, which would provide a strategy for the clinical realization of precision immunotherapy and improve the prognosis of small cell lung cancer patients. Methods: This research retrospectively collected data from ES-SCLC patients treated with PD-1/PD-L1 inhibitors between March 1, 2019, and October 31, 2022, at Harbin Medical University Cancer Hospital. The study data was randomly split into training and validation sets in a 7:3 ratio. Variables associated with patients' overall survival were screened and modeled by univariate and multivariate Cox regression analyses. Models were presented visually via Nomogram plots. Model discrimination was evaluated by Harrell's C index, tROC, and tAUC. The calibration of the model was assessed by calibration curves. In addition, the clinical utility of the model was assessed using a DCA curve. After calculating the total risk score of patients in the training set, patients were stratified by risk using percentile partitioning. The Kaplan-Meier method was used to plot OS and PFS survival curves for different risk groups and response statuses at different milestone time points. Differences in survival time groups were compared using the chi-square test. Statistical analysis software included R 4.1.2 and SPSS 26. Results: This study included a total of 113 ES-SCLC patients who received immunotherapy, including 79 in the training set and 34 in the validation set. Six variables associated with poorer OS in patients were screened by Cox regression analysis: liver metastasis (P = 0.001), bone metastasis (P = 0.013), NLR < 2.14 (P = 0.005), LIPI assessed as poor (P < 0.001), PNI < 51.03 (P = 0.002), and LDH ≥ 146.5 (P = 0.037). A prognostic model for immunotherapy in ES-SCLC patients was constructed based on the above variables. The Harrell's C-index in the training and validation sets of the model was 0.85 (95% CI 0.76-0.93) and 0.88 (95% CI 0.76-0.99), respectively; the AUC values corresponding to 12, 18, and 24 months in the tROC curves of the training set were 0.745, 0.848, and 0.819 in the training set and 0.858, 0.904 and 0.828 in the validation set; the tAUC curves show that the overall tAUC is > 0.7 and does not fluctuate much over time in both the training and validation sets. The calibration plot demonstrated the good calibration of the model, and the DCA curve indicated that the model had practical clinical applications. Patients in the training set were categorized into low, intermediate, and high risk groups based on their predicted risk scores in the Nomogram graphs. In the training set, 52 patients (66%) died with a median OS of 15.0 months and a median PFS of 7.8 months. Compared with the high-risk group (median OS: 12.3 months), the median OS was significantly longer in the intermediate-risk group (median OS: 24.5 months, HR = 0.47, P = 0.038) and the low-risk group (median OS not reached, HR = 0.14, P = 0.007). And, the median PFS was also significantly prolonged in the intermediate-risk group (median PFS: 12.7 months, HR = 0.45, P = 0.026) and low-risk group (median PFS not reached, HR = 0.12, P = 0.004) compared with the high-risk group (median PFS: 6.2 months). Similar results were obtained in the validation set. In addition, we observed that in real-world ES-SCLC patients, at 6 weeks after immunotherapy, the median OS was significantly longer in responders than in non-responders (median OS: 19.5 months vs. 11.9 months, P = 0.033). Similar results were obtained at 12 weeks (median OS: 20.7 months vs 11.9 months, P = 0.044) and 20 weeks (median OS: 20.7 months vs 11.7 months, P = 0.015). Finally, we found that in the real world, ES-SCLC patients without liver metastasis (P = 0.002), bone metastasis (P = 0.001) and a total number of metastatic organs < 2 (P = 0.002) are more likely to become long-term survivors after receiving immunotherapy. Conclusion: This study constructed a new prognostic model based on basic patient clinical characteristics and peripheral blood indices, which can be a good predictor of the prognosis of immunotherapy in ES-SCLC patients; in the real world, the response status at milestone time points (6, 12, and 20 weeks) can be a good indicator of long-term survival in ES-SCLC patients receiving immunotherapy.