Circulating Cell-Free DNA in Classical Hodgkin Lymphoma in Children, Adolescents and Young Adults. Results from the Holy Study, a French Ancillary Study of Euronet PHL-C2 Protocol

Background: Circulating cell-free tumor DNA (ctDNA) has shown interesting results in adult classical Hodgkin lymphoma (cHL). However, its applicability in children, adolescents, and young adults (CAYA) with cHL remains undefined. This prospective national study aimed to elucidate the potential role of ctDNA in CAYA patients with cHL. Methods: This prospective trial, conducted in France between December 2019 and January 2023, recruited CAYA patients (≤ 25 years old) newly diagnosed with cHL from 31 centers from the SFCE (Société Française des Cancers de l'Enfant et de l'Adolescent). Patients were treated according to the Euronet PHL-C2 trial (EudraCT: 2012-004053-88). An 18-gene amplicon-based NGS (Next Generation Sequencing) targeted panel was designed, encompassing the most frequently mutated genes in cHL (TNAIP3, NFKBIE, SOCS1, PTPN1, STAT6, B2M, ITPKB, GNA13, XPO1, ARID1A, TP53, CD70, SPEN, BTG1, CIITA, HIST1H1E, IGLL5, CD36). ctDNA evaluations were performed at diagnosis, after 2 cycles of chemotherapy (OEPA), and in case of relapse. Results: A total of 278 CAYA patients with cHL with a median follow-up of 14 months were included in the study. The median age at diagnosis was 15 years (range 2-22), with 45% of male. At diagnosis, 38% had a bulky mass (≥200 ml), 48% presented with B-symptoms, 89% had sclero-nodular histology, and 16% had an EBV association. Furthermore, 48% of the patients were treated as advanced stages (TL-3). Using the 18-gene NGS panel, 2309 variants were detected in 239 out of 278 patients (86%). The most frequently mutated genes were SOCS1 (68%), IGLL5 (44%), B2M (44%), CIITA (42%), TNFAIP3 (37%), NFKBIE, and STAT6 (32%) with a mean variant allele frequency (VAF) of 5.7% (range 3.7%-8.5%) per gene. Level of ctDNA at diagnosis was strongly correlated with the presence of B-symptoms, sedimentation rate and Ann Arbor stages. Patients with no detectable ctDNA at diagnosis (39/278, 14%) exhibited distinct clinical and biological parameters compared to patients with detectable ctDNA (239/278, 86%). These patients were younger (p < 0.01), had fewer bulky masses (p < 0.01), fewer B-symptoms (p = 0.02), higher EBV association (p < 0.01), more mixed cellularity histology (p < 0.01), and excellent outcomes (no relapse among these 39 patients). Among patients with detectable ctDNA at diagnosis (86%) median variant number per patient was 9 (range 1-54) with mean VAF per patient of 4.8% (0.13 - 22%). ctDNA became undetectable at C2 in 93% of the cases. Detectable ctDNA at C2 (n=14) was a significant prognostic marker of relapse, and combined with PET CT evaluation at C2, detectable ctDNA identify patients with a very high risk of relapse. TP53 mutations at diagnosis (18/278, 7%) were strongly associated with inadequate response (IR) evaluated by PET-CT (Deauville Score ≥4 at early response assessment, ERA); 70% of patients with TP53 mutations had an IR at ERA (vs 31%, p <0.001). In a multivariate Cox model analysis including ERA by PET-CT, IGLL5 mutations were associated with a significantly higher risk of relapse; Hazard Ratio (HR); 3.2, 95%CI(1.1, 9.1). Conclusion: To our knowledge, we report the largest cohort of CAYAs with cHL included in a clinical trial and analysed by ctDNA serial sequencing. Using an 18-gene NGS panel, we detected at least one mutation in 86% of diagnostic samples in CAYA patients with cHL. Variant detection in ctDNA, including TP53/IGLL5 mutations could refines therapeutic stratification and when combined with PET-CT in CAYA cHL.