Assessment of prognostic value and development of predictive model for prolonged lymphopenia in patients with glioblastoma following chemoradiotherapy.

Abstract

Lymphopenia during chemoradiotherapy (CRT) for glioblastoma has been shown to be a poor prognostic factor. However, the relationship between prolonged lymphopenia (PL) after CRT and prognosis remains unclear. This study aimed to explore the relationship between PL and glioblastoma prognosis and develop a predictive model for PL risk. We analyzed 87 patients with primary glioblastoma who underwent postoperative CRT with 60 Gy in 30 fractions of radiotherapy and temozolomide. PL was defined as grade 2 or higher lymphopenia 1 month after the completion of CRT. We conducted survival analysis, identified risk factors for PL, and developed a predictive model for PL risk. Of the 87 patients, 41 developed PL, and progression-free survival (PFS) was significantly shorter in the PL group (median 8.0 months vs 15.4 months, P = 0.003). However, there was no significant difference in overall survival between the two groups. PL was also a significant factor for shorter PFS in multivariable analysis (P = 0.040). Brain V20Gy (percentage of brain volume receiving ≥20 Gy), gross total resection (GTR), and preoperative Karnofsky performance status (KPS) were identified as significant risk factors for PL. The predictive model showed that the risk of PL could be predicted by brain V20Gy, sex, age, GTR and preoperative KPS. PL was identified as a PFS shortening factor. Our model suggests that reducing irradiated brain volume may help prevent PL and could potentially improve glioblastoma prognosis by preserving cancer immunity.