Role of Geriatric Assessment Scores as Predictors of Intensive Therapy Feasibility and Survival in Elderly Patients with Primary CNS Lymphoma

Background/Objectives: Short induction followed by high-dose chemotherapy and autologous stem cell transplantation (HCT-ASCT) is effective in newly diagnosed elderly patients with primary central nervous system lymphoma (PCNSL) but associated with significant toxicity. Geriatric assessments (GAs) may help to predict treatment risk and prognosis, yet no standardized GAs exist for PCNSL. Our aim was to evaluate the impact of GA on HCT-ASCT eligibility and survival. Methods: We analyzed 65 patients > 65 years treated in the MARiTA and MARTA studies. Treatment comprised 2 cycles of rituximab, HD-MTX and cytarabine followed by HCT-ASCT. GAs at diagnosis were analyzed for progression-free survival (PFS), overall survival (OS) and premature end of treatment (pEOT). Results: After median follow-up of 43 months, 12-month PFS/OS were 69.2% (95% CI 56.5–78.9%)/70.8% (58.1–80.2%) from diagnosis and 80.4% (66.6–88.9%)/84.3% (71.1–91.8%) from time of HCT-ASCT. ECOG PS ≥ 2, Lachs geriatric screening (Lachs) ≥30% and Cumulative Illness Rating Scale-Geriatric (CIRS-G) ≥6, ≥7 and ≥8, respectively, were significantly associated with pEOT in univariate analysis (UVA). In multivariate analysis (MVA), CIRS-G remained significant. A composite EBL score (ECOG PS ≥ 2, Barthel Index of Activities of Daily Living (Barthel) < 20, Lachs ≥ 30%) ≤1 predicted successful completion of HCT-ASCT in >90% of patients. ECOG PS ≥ 2 and Barthel < 20 were associated with decreased PFS and OS in UVA; ECOG PS ≥ 2 remained significant in MVA. Conclusions: This is the first study to link GA with treatment feasibility in elderly PCNSL patients undergoing intensive therapy. Our results will be validated in the PRIMA-CNS trial (EudraCT 2020-001181-10).