Construction of brain metastasis prediction model in limited stage small cell lung cancer patients without prophylactic cranial irradiation

Introduction Small cell lung cancer (SCLC) is a highly aggressive lung cancer variant known for its elevated risk of brain metastases (BM). While earlier meta‐analyses supported the use of prophylactic cranial irradiation (PCI) to reduce BM incidence and enhance overall survival, modern MRI capabilities raise questions about PCI's universal benefit for limited‐stage SCLC (LS‐SCLC) patients. As a response, we have created a predictive model for BM, aiming to identify low‐risk individuals who may not require PCI. Methods A total of 194 LS‐SCLC patients without PCI treated between 2009 and 2021 were included. We conducted both univariate and multivariate analyses to pinpoint the factors associated with the development of BM. A nomogram for predicting the 2‐ and 3‐year probabilities of BM was then constructed. Results Univariate and multivariate analyses revealed several significant independent risk factors for the development of BM. These factors include TNM stage, the number of chemotherapy (ChT) cycles, Ki‐67 expression level, pretreatment serum lactate dehydrogenase (LDH) levels, and haemoglobin (HGB) levels. These findings underscore their respective roles as independent predictors of BM. Based on the results of the final multivariable analysis, a nomogram model was created. In the training cohort, the nomogram yielded an area under the receiver operating characteristic curve (AUC) of 0.870 at 2 years and 0.828 at 3 years. In the validation cohort, the AUC values were 0.897 at 2 years and 0.789 at 3 years. The calibration curve demonstrated good agreement between the predicted and observed probabilities of BM. Conclusions A novel nomogram has been developed to forecast the likelihood of BM in patients diagnosed with LS‐SCLC. This tool holds the potential to assist healthcare professionals in formulating more informed and tailored treatment plans. A novel nomogram was constructed for predicting the risk of BM in LS‐SCLC patients, which could help clinicians make better treatment strategies.