Autologous hematopoietic cell transplantation versus whole‐brain radiotherapy consolidation in primary central nervous system lymphoma: A systematic review and meta‐analysis

Abstract The management of newly diagnosed primary central nervous system lymphoma (PCNSL) includes administration of high‐dose methotrexate based regimens followed by consolidation therapy to minimize the risk of relapse. However, the best consolidation strategy (autologous hematopoietic cell transplant [auto‐HCT] vs. whole‐brain radiotherapy [WBRT]) is controversial. Hence, we performed a systematic review and meta‐analysis of all randomized controlled trials that compared auto‐HCT versus WBRT consolidation for patients with PCNSL after first‐line treatment.The primary outcome was overall survival (OS), while the secondary outcomes included progression‐free survival (PFS), response rates (overall response rate [ORR] and complete remission [CR]), relapse rate, treatment‐related mortality (TRM), and neuropsychological adverse events. We performed a pooled analysis of the single‐arm studies that incorporated auto‐HCT or WBRT consolidation and evaluated neurocognitive outcomes. Only two studies met the inclusion criteria ( n = 240). There was no significant difference in OS (HR = 1.50; 95% CI = 0.95–2.36), PFS (HR = 0.99; 95% CI = 0.44–2.22), ORR (RR = 1.48; 95% CI = 0.90–2.44), CR rate (RR = 1.21; 95% CI = 0.90–1.63), relapse rate (RR = 0.46; 95% CI = 0.05–4.28), and TRM (RR = 5.67; 95% CI = 1.01–31.91). The neuropsychological tests to assess neurocognitive domains were different and inconsistently reported in the two studies and therefore we were unable to perform a meta‐analysis but provide a descriptive assessment. Both the studies showed a significant decline in the attention/executive function (based on the trail making test A and trail making test B) in those receiving WBRT compared to auto‐HCT. We found 9 single‐arm phase II studies that reported data on outcomes associated with either auto‐HCT (5 studies) or WBRT (4 studies) consolidation. Of these, two studies ( n = 43) reported data on neurocognitive decline following auto‐HCT consolidation. Pooled proportion of patients with neurocognitive decline in these studies was 6% (95% CI, 0%–17%) for those receiving auto‐HCT and there was no heterogeneity between studies ( I 2 = 0%). Three studies ( n = 122) reported data on neurocognitive decline following WBRT consolidation. Pooled proportion of patients with neurocognitive decline in these studies was 43% (95% CI, 11%–78%) for those receiving WBRT and there was high heterogeneity between studies ( I 2 = 94%). There was significant heterogeneity between subgroups ( p = 0.035). The outcomes were not significantly different in patients with PCNSL receiving auto‐HCT or WBRT consolidation therapies, however, there is a higher degree of neurocognitive decline associated with WBRT compared to auto‐HCT consolidation. The decision to choose a consolidation strategy needs to be individualized based on age, frailty, and co‐morbidities.