Liquid-biopsy mutation landscape and its concordance with skin biopsies in cutaneous T-cell lymphoma

Abstract

Abstract Introduction: Liquid biopsy (LBx) is widely utilized for clinical evaluation and therapeutic monitoring in solid and hematologic malignancies (Talotta et al. Frontiers in Oncology 2023; Huang et al. Molecular Cancer 2024). Analyzing circulating cell-free DNA (cfDNA) and RNA (cfRNA) through LBx can offer minimally invasive insights into tumor burden and disease dissemination (Huang et al. Molecular Cancer 2024). Given that diagnosing CTCL is challenging both clinically and histologically, employing liquid biopsy as a diagnostic tool could enable the detection of disease, potentially reducing reliance on skin biopsies. LBx application in CTCL is complicated by overlapping somatic mutations found in clonal hematopoiesis of indeterminate potential (CHIP) and other hematologic malignancies (Marnell et al. Journal of Molecular and Cellular Cardiology 2021). CHIP-associated mutations in genes like TET2 and DNMT3A closely resemble those in myeloid and T-cell neoplasms, complicating LBx interpretation (Kusne et al. Leukemia Research 2022). In CTCL, mutations commonly affect pathways critical for T-cell receptor signaling, cytokine activation, cell-cycle regulation, apoptosis, and epigenetic regulation. Notably, epigenetic regulators (TET2, DNMT3A, ASXL1, ARID1A, histone-modifying enzymes) are commonly mutated in CTCL and myeloid neoplasms (Choi et al. Nature Genetics 2015). In our study, we analyzed cfDNA and cfRNA from patients with CTCL, comparing LBx, skin biopsies (SBx), and paired LBx-SBx samples to characterize the mutations present in LBx associated with CTCL. Our goal is to enable accurate interpretation of LBx data and supports its potential role in identification in CTCL. Methods: Using hybrid capture and next generation sequencing (NGS), we performed targeted sequencing of DNA (302 gene) and RNA (1600 genes) extracted from SBx, and cfDNA and cfRNA extracted from peripheral blood plasma. B and T cell clonality in cfRNA was confirmed by evaluating heavy and light chain immunoglobulin and TCRs, respectively. For B and T cell repertoire analysis, sequences were mapped and assembled into clonotypes using MiXCR software. Results: In the SBx cohort (n=200), 263 unique genetic mutations were identified. The most frequently mutated genes included DNMT3A (25.3%), KMT2C (16.4%), TET2 (15.1%), TP53 (15.1%), KMT2D (13.0%), NOTCH1 (13.0%), STAT3 (9.59%), NOTCH3 (7.53%), ATM (6.85%), and CHEK2 (6.85%). These mutations are predominantly involved in epigenetic regulation, DNA repair, cell-cycle control, and key signaling pathways essential for T-cell proliferation and survival. In the LBx cohort (n=41), 66 unique genetic mutations were identified. The 10 most frequent mutations were found in DNMT3A (38.7%), KMT2C (29.0%), TET2 (19.4%), PPM1D (12.9%), ARID1A (9.68%), EPHA5 (9.68%), JAK2 (9.68%), AKT1 (6.45%), ASXL1 (6.45%), and ATM (6.45%). Notably, there was overlap between LBx and SBx, with 4 out of the top 10 mutated genes (TET2, DNMT3A, KMT2C, and ATM) detected in both sample types. Clonotype-naive TCR clonality analysis of cfRNA in 32 LBx samples identified clonality in 6 samples (19%). These clonally expanded samples exhibited a distinct mutation profile, with notable involvement of genes such as DNMT3A, TET2, EZH2, JAK2, BRAF, and NOTCH3, likely reflecting more tumor cfDNA and cfRNA in circulation. Non-clonal samples generally lacked these specific mutation signatures or presented them at markedly lower frequencies. Higher sensitivity in demonstrating TCR clonality is expected if testing was clonotype-informed. Among 16 paired LBx and SBx samples, mutations were detectable in 81% of LBx and 75% of SBx samples. Within 12 pairs where mutations were present, 92% showed concordance in at least one mutation. A total of 63 mutations were identified, with 34 variants (54%) being concordant across LBx and Bx. High concordance was observed in mutations of key genes involved in epigenetic regulation (DNMT3A, TET2), NOTCH/T-cell receptor signaling (NOTCH2, NOTCH3, BTK), and the PI3K–mTOR signaling pathway (MTOR, BRAF), representing 72% of the shared variants. Conclusions: These results indicate that LBx has potential clinical utility as a non-invasive tool for identifying key driver mutations in CTCL, thus aiding diagnosis and possibly reducing reliance on skin biopsies. Additional studies and clinical trials are needed to evaluate its prognostic value.