BACKGROUND

Radiation-induced lymphopenia (RIL) has been demonstrated in types of solid tumors. This study aimed to assess the association between dosimetric and clinical variables and RIL in patients with brain metastases after brain radiotherapy (RT).

METHODS

The craniofacial bones of 140 patients were retrospectively delineated, and absolute lymphocyte counts (ALC) were collected both prior to RT (pre-RT) and 1 month after completing RT (post-RT). Linear regression analyses were conducted to identify associations between dosimetric variables of the craniofacial bones (V5-V40), mean dose, clinical parameters, and post-RT ALC. Logistic regression analysis was used to evaluate independent predictors of RIL (ALC < 1000 cells/µL). The relationships between dosimetric and clinical variables and overall survival (OS) were analyzed using a Cox regression model.

RESULTS

After completing RT, ALC decreased in both the whole brain RT (WBRT) and focal RT subgroups ( < 0.05). Linear regression analysis suggested that pre-RT ALC, gross tumor volume (GTV), and the V5 of the craniofacial bones were independent risk factors for the decrease in post-RT ALC. Logistic regression analysis demonstrated that lower pre-RT ALC (OR: 31.969, 95%CI 8.44 to 121.092, : 0.000) and larger GTV (OR: 2.438, 95%CI 1.037 to 7.819, : 0.011) were associated with the development of RIL. Notably, no dosimetric variable of the craniofacial bones correlated with RIL. Cox analysis confirmed that pre-RT ALC and V5 of craniofacial bones were associated with OS. Additionally, receiver-operating characteristic (ROC) curve analysis indicated a predictive cutoff value of 702.5 cm³ for GTV in relation to RIL ( < 0.001, AUC = 0.704).

CONCLUSIONS

The consideration of modifiable factors, such as reducing V5 of craniofacial bones, may help preserve lymphocytes and maintain immunologic function, potentially affecting clinical outcomes, especially with a larger GTV.