Related Experiment Video
Updated: May 12, 2026
Predictive Immune Modeling of Solid Tumors
Published on: February 25, 2020
A population-based nomogram for prognostic assessment in advanced lung cancer following progression with immune
Tianwei Luo1,2, Huiyu Li3, Ao Chen3
1Department of Oncology, Nanjing Drum Tower Hospital, Nanjing Drum Tower Hospital Clinical College, Nanjing University of Chinese Medicine, Nanjing, China.
Background:
The complexity of immunoresistance patterns in advanced lung cancer (LC) and the uniqueness of its efficacy response (e.g., pseudoprogression and hyperprogression) make it difficult to standardize treatment regimens after progression. This study aimed to analyze the independent factors after progression of immunotherapy in LC patients and to construct and validate a visual evaluation tool for assisting clinical prediction of prognosis.
Methods:
A total of 245 LC patients who progressed after receiving single or combination immune checkpoint inhibitors (ICIs) therapy in Jinling Hospital, Affiliated Hospital of Medical School, Nanjing University between January 2019 and February 2024 were enrolled for retrospective analysis. The nomogram was created based on the least absolute shrinkage and selection operator (LASSO) and multivariate Cox regression analysis to predict overall survival (OS) and progression-free survival (PFS) in R software. The time-dependent area under the receiver operating characteristic (ROC) curves (AUCs), calibration curves, and decision curve analysis (DCA) were used to assess the discrimination, accuracy, and clinical utility of the prediction models.
Results:
Four independent factors significantly associated with OS were utilized to create a nomogram to predict OS: platelet-to-lymphocyte ratio (PLR) level, smoking history, immune combination regimen, and liver metastasis factors. Three variables significantly associated with PFS were incorporated into the development of a nomogram for predicting PFS: PLR level, smoking history, and tumor stage. The results of the calibration curves showed that the predictive probabilities of the nomogram for OS and PFS were consistent with the observational data, with the C-indexes of the nomogram for predicting OS and PFS being 0.643 [95% confidence interval (CI): 0.592-0.695] and 0.588 (95% CI: 0.541-0.636), respectively. The AUCs for predicting the 12-month, 24-month, and 36-month OS and PFS were 0.681 and 0.763, 0.795 and 0.637, 0.649 and 0.694, respectively. The DCA curves indicated that the nomogram for OS had good net benefits. Compared to the high-risk group, the OS and PFS were significantly prolonged in the low-risk group.
Conclusions:
The novel nomogram for predicting the prognosis of advanced LC patients after progression with ICIs provides a scientific basis and an important reference for the development of individualized treatment strategies. Our study revealed that PLR level, smoking history, and liver metastasis were significant prognostic indicators. In the future, specific treatment options that could prolong patients' OS and PFS need to be further explored.
