Detection of mutation in cell-free DNA predicts poor outcomes in adult patients with langerhans cell histiocytosis

Abstract Background and aims: Langerhans cell histiocytosis (LCH) is a rare, heterogeneous histiocytic disease with recurrent MAPK pathway mutations. Cell-free DNA (cfDNA) is becoming a widely used prognostic and predictive biomarker in oncology. This study investigated the detection of the mutations in cfDNA at diagnosis and during follow-up in adult LCH cohort to clarify the associations between the clinical features and prognosis value. Methods: One hundred and twenty-seven adult patients (age ≥18 years) diagnosed with LCH with both biopsy and plasma samples available were included in this study. Lesion biopsy samples of enrolled patients were subjected to next-generation sequencing (NGS). The presence and level of genetic alternations of plasma cfDNA in the MAPK pathway were detected by QX200TM Droplet Digital PCR System (Bio-Rad). Results: The median age at diagnosis was 34 years (range 18-70). There were 83 males (65.4%) and 44 females (34.6%). One hundred and five patients (82.7%) were treatment-naïve, and 22 patients (17.3%) were refractory or relapsed. The patients were classified into three groups: the unifocal group (n=28, 22.0%), single system multifocal (SS-m) group (n=19, 15.0%), and MS group (n=80, 63.0%). Among the MS group, 22 patients (17.3%) had risk organ (RO, liver, spleen or bone marrow) involved. The baseline plasma cfDNA was analyzed in each patient, and 53 patients were detected with the same mutations as the tissue lesion, with a sensitivity of 41.7%. The positivity rates of BRAFV600E, BRAFindel, and MAP2K1 alternations were 38.0%, 64.5%, and 37.5%, respectively. The average variant allele frequency (VAF) of plasma cfDNA was 0.79% (range 0-15.9%). Our cohort had a higher frequency of detectable cfDNA in patients with MS disease (43/80, 53.8%) compared to unifocal (5/28, 17.9%) or SS-m disease (5/19, 26.3%) (p = 0.0004). A significant association was observed between ctDNA positivity frequency and disease status, with a substantially greater proportion of positive cases identified in refractory/relapsed patients (14/22, 63.6%) than in newly diagnosed individuals (39/105, 37.1%) (p = 0.031). Besides, positive cfDNA in plasma was associated pituitary involvement (73.5 vs 30.1%, p<0.001). In addition, we also found that baseline cfDNA positivity was associated with elevated levels of inflammatory cytokines, such as hypersensitive C-reactive protein (p=0.041), erythrocyte sedimentation rate (p=0.016), fibrinogen (p=0.038), and tumor necrosis factor-α (TNF-α, p=0.023). Ninety-seven patients underwent systemic therapy. Disease response was assessed per PET Response Criteria in Solid Tumors (PERCIST 1.0). The overall response rate (ORR) in patients with positive cfDNA at baseline was significantly lower than those with negative results (74.5% vs. 96.8%, p=0.0008). After a median follow-up of 18 (range 6-64) months, 19 patients (15.0%) experienced disease relapse. The estimated 3-year overall survival (OS) was 100% and the median event-free survival (EFS, with events defined as poor response to the treatment, or death from any cause) was not reached (estimated 3-year EFS was 71.3%). During follow-up, 36 of 53 patients (57.1%) had negative cfDNA in 6 months. Univariate prognostic analyses showed that patients who had positive cfDNA at baseline showed poorer outcomes (median EFS 49.7 months vs not reached, p=0.0005). Early cfDNA negativity predicts better prognosis (median EFS not reached vs 17.8 months, p<0.0001). Conclusion: Positive cfDNA at baseline was associated with pituitary involvement, multiple system involvement, refractory/relapsed status and hyperinflammatory state in adult patients with LCH, and was also an independent prognostic factor for poor treatment response and outcomes. Early cfDNA negativity after therapy predicts better prognosis.