First Blood: Pre-Therapy BRAFV600E+ PBMC Identifies a Novel Clinical Risk Category for Patients with LCH

Abstract

Purpose: Langerhans Cell Histiocytosis (LCH) is a myeloproliferative neoplasia driven by somatic mutations in the mitogen-activated protein kinases pathway (MAPK). LCH has historically been clinically staged as “high risk” (e.g. liver, spleen and/or bone marrow involvement, HR) or “low risk” (LR) due relative to risk of death in earlier trials. However, clinical staging is complicated by different sensitivities of imaging and, more recently, molecular analyses. We previously identified that BRAFV600E allele in peripheral blood mononuclear cells (PBMC) in patients with HR LCH, which supported a model of LCH arising from hematopoietic myeloid precursors. To determine the clinical significance of somatic lesion BRAFV600E mutations as well as ability to detect pre-therapy BRAFV600E+ PBMC at diagnosis, we analyzed an extended cohort of LCH patients with longitudinal clinical evaluations. Methods: Genomic DNA was isolated from 366 lesion biopsies and the PBMCs of 294 patients with active LCH before initial therapy. The percentage of PBMC with the BRAFV600E allele in patients was determined by quantitative polymerase chain reaction. Lesion genotype and ability to detect pre-therapy BRAFV600E in PBMC were correlated with clinical features including i) clinical risk status, ii) extent of disease (single lesion vs multifocal; single system vs multisystem), iii) response to therapy, v) and time to relapse were analyzed using a chi-square test. Fluorescence-activated cell sorting (FACS), DNA amplification and subsequent qPCR was performed on 15 pretherapy BRAFV600E+ blood samples to explore the relationship between different myeloid lineages (CD3, CD19, CD56, CD14, CD16, CD11c) and extent of LCH disease. Survival proportions were calculated using the Kaplan-Meier (product limit) method and a log-rank (Mantel-Cox) test was used to calculate the significance of the difference between survival curves. Results: Median age was 3.01 years (range 3.6 days to 21.5 years), 58.5% of the patients were male (41.5% female), and 15.4% had HR disease (84.6% LR). Among patients with LR disease, 33.0% had multisystem, 23.1% had multi-focal single system, and 44.0% had single lesions. Lesion genotypes were 51.9% BRAFV600E and 49.1% not BRAFV600E. BRAFV600E in LCH lesions was associated with lower age at diagnosis and higher prevalence of HR disease. However, after a median follow-up time of 1.6 years (range 1 month to 8.5 years) there was no difference in EFS between BRAFV600E and alternative mutations (p=0.35). Notably, for the entire cohort, presence of BRAFV600E in diagnostic PBMC was highly associated with younger patients, clinical HR disease and risk of developing LCH-associated neurodegeneration (LCH-ND). For patients with BRAFV600E+ lesions, 32.5% had detectable BRAF V600E in pre-therapy PBMC: 100% of HR, 52.6% of multisystem LR, 18.4% of multi-focal single system LR and 0% of single lesion. Further, after a median follow-up time of 2.0 years (range 1 month to 9,1 years) clinical LR PBMC BRAFV600E+ represented a group at higher risk of treatment failure compared to BRAFV600E-neg LR (p=0.01). BRAFV600E alleles were present in PBMC at diagnosis in all of the sorted myeloid populations, with consistent enrichment in myeloid dendritic cells and monocytes. There was no significant correlation between BRAFV600E allele burden (percentage of BRAFV600E+ PBMC) and clinical risk, and no significant differences in the distribution of allele burden within lineages. Conclusion: Ability to detect pretherapy BRAFV600E+ represents an important clinical biomarker for patients with LCH. Results from this study support the potential for pre-therapy PBMC BRAFV600E to predict clinical risks, inform pathogenic mechanisms in LCH, and identify LR PBMC BRAFV600E+ as a novel clinical risk group. We hypothesize that persistence of hematopoietic clones and circulating myeloid precursors may underlie the increased risks of treatment failure and development of LCH-ND in patients previously presumed to have “low risk” disease.