Abstract
Background. BRAF (V600E) mutation is the molecular hallmark of hairy cell leukemia (HCL), being detected in more than 95% of patients. Purine analogs are highly effective in HCL with a complete remission (CR) rate of 80-90% and a median duration of response of 8 years. CR is defined as resolution of peripheral cytopenias, splenomegaly and bone marrow infiltration (Grever MR et al, Blood, 2017). Few studies have explored minimal residual disease (MRD) after therapy with immunohistochemistry, flow cytometry or PCR-based methods. The impact of achieving a deep response with undetectable MRD is currently unclear.
Aim. The aims of this study were to assess MRD status by droplet digital PCR (ddPCR), and to evaluate whether the MRD status predicts long-term outcome in HCL patients treated with purine analogs frontline.
Methods. We collected data from a prospectively maintained database of patients with HCL, diagnosed and followed at Division of Hematology of IRCCS Policlinico San Matteo between 1989 and 2025. We analyzed only patients who had available bone marrow or peripheral blood samples before and after first-line therapy with purine analogs. BRAF (V600E) mutation status was assessed with droplet digital Polymerase Chain Reaction (ddPCR) with a deep sensitivity (Limit of Detection – LoD: 0.02% variant allele frequency – VAF). Overall survival (OS) and progression-free survival (PFS) were estimated with Kaplan-Meier method and comparison between curves was performed with log-rank test. Survival Receveir Operating Characteristic (ROC) analysis was applied to identify the most discriminant cut-off of VAF.
Results. BRAF (V600E) mutation was assessed with ddPCR before therapy in 37 patients with HCL candidate to frontline therapy with purine analogs. Two of 37 (5%) patients were BRAF (V600E) wild type by ddPCR and were excluded from the analysis. In one of them, we identified an alternative BRAF mutation (c.1457_1471del; p.N486_P490del; VAF 3.8%) with next generation sequencing (NGS). The median age at diagnosis of the study population was 54 years (IQR: 43-62), 28/35 (80%) were males, 22/35 (65%) had splenomegaly. The median blood cell counts before therapy were: hemoglobin level 11.7 g/dL (IQR: 10.5-13.5), absolute neutrophil count 0.85x109/L (IQR: 0.490-1.189), platelets 80x109/L (IQR: 59-104). The median LDH and beta2-microglobulin value were respectively 170 U/L (IQR: 151-221) and 1949 mcg/L (1840-2490). The median bone marrow infiltration before therapy was 80% (IQR: 60-90) and the median VAF for BRAF mutation was 10% (IQR: 0.5-30). As frontline therapy, 31/35 patients (89%) received a single course of cladribine 0.14 mg/kg/daily for 5 days subcutaneously, 4/35 (11%) received pentostatin every 15 days for 8 doses. At the end of therapy, 17 (49%) patients achieved a CR, 16 (46%) a partial response, 1 was not evaluable and 1 had stable disease. The median bone marrow infiltration after treatment was 20% (IQR: 0-30). BRAF (V600E) mutation was assessed after treatment in 31/35 patients (89%). Median VAF of BRAF mutation after treatment was 1.23% (IQR: 0.08-1.31). Five of 35 (14.3%) achieved MRD negativity after treatment and all of them were in CR. The median follow-up from diagnosis was 11 years (IQR: 2-13). The 10-year PFS and OS of the entire study population were respectively 66% (95% CI 43-82) and 90% (95% CI 66-97). None of the patients with a post-treatment VAF ≤0.028 (which was identified as the optimal cut-off by ROC analysis) experienced a relapse, whereas we observed 6 relapses among patients with a VAF >0.028, corresponding to a 10-year PFS of 100% and 65.1% respectively. The OS was not affected by MRD status at the end of treatment.
Conclusions. In this study including 35 HCL patients treated with purine analogs frontline, we found that 14% achieved an undetectable MRD. More importantly, we demonstrated that the clearance of BRAF mutation was associated with a better long-term outcome.
