Next-generation sequencing (NGS) of DNA isolated from nail specimens and run in parallel with tumor NGS accurately and efficiently detects germline predisposition variants in myeloid malignancies

Abstract

Abstract Background: Germline predisposition occurs in patients of all ages with myeloid malignancies and is important to identify due to implications for family members and donor selection for stem cell transplant. The gold standard method of germline testing for patients with myeloid malignancies is to analyze DNA extracted from cultured skin fibroblasts. However, widespread adoption has been limited by the need for a skin punch biopsy, high costs, and insurance coverage. Historically, fingernail clippings have not been considered a preferred source of DNA for germline testing due to concerns about extraction of adequate DNA and the risk of peripheral blood contamination. Our laboratory has developed a method to extract DNA and perform NGS on nail clippings in parallel with routine NGS testing that is ordered on tumor bone marrow samples. To investigate whether this accurately detects germline genetic variants in patients with myeloid malignancies, we conducted a quality improvement study to compare results from paired nail and skin fibroblast specimens. Methods: Paired fingernail and skin fibroblast germline testing was ordered on consecutive patients with newly diagnosed AML or MDS treated at the University of Colorado Hospital from April 2024 through April 2025 who met NCCN recommendations for consideration of germline testing based on age < 50, personal or family history, or clinical suspicion for germline predisposition. Skin fibroblast culture and testing was performed at GeneDx using a custom gene panel: RUNX1, ANKRD26, CEBPA, DDX41, ETV6, GATA2, MBD4, MECOM, SAMD9, SAMD9L, TERC, TERT, RTEL1, ATG2B, and GSKIP. Fingernail clippings were sent to our on-campus molecular laboratory in parallel with diagnostic bone marrow testing. NGS was performed using a custom, 602 gene myeloid malignancy panel. Target enrichment is performed via a hybridization-capture-based method followed by sequencing on an Illumina NovaSeq6000 instrument. A custom in-house bioinformatics pipeline is applied to both specimens, and a unique DNA fingerprint is generated to ensure fingernails and bone marrows are from the same patients. Variant calls, quality, and allelic frequencies are assessed in both specimens simultaneously and somatic, heterozygous, and homozygous germline variants are confirmed. Data regarding pathogenic and likely pathogenic (P/LP) variants that were detected by each method was recorded. Results: Paired germline testing using DNA isolated from skin fibroblasts and nail clippings was ordered on 41 consecutive adults with newly diagnosed AML or MDS. One patient who declined both methods of germline testing was excluded. All patients met NCCN recommendations for consideration of germline testing due to age < 50 (n=23), family history of hematologic malignancy (n=14), or based on personal history (n=3). Skin punch biopsies were collected from 37/40 (92.5%) patients (2 patients declined, 1 specimen was not collected for an unknown reason). Nail specimens were collected from 34/40 (85%) patients. Adequate DNA for testing was isolated in 37/37 (100%) of skin fibroblast samples and 33/34 (97%) nail specimens. For the one nail specimen in which adequate DNA could not be extracted, a repeat nail specimen was collected which was adequate for testing. Of the 31 patients who had paired fingernail and skin fibroblast testing, germline P/LP (tier 1 or tier 2) variants were identified in 4/31 patients (12.9%). This included 3 P/LP variants in DDX41 (c.571G>A, p.A191T; c.653 G>A, p.G218D; c.3 G>A; p.Met1) and 1 P/LP variant in RUNX1 (c.611G>A, p.R204Q). Results were 100% concordant between the fingernail and skin fibroblast specimens. There were no instances of false positive results in the nail clippings (false positive rate 0%). The median turn-around time (TAT) was significantly faster for nail specimens than skin fibroblast specimens (median TAT 16 versus 31 days, P<0.001). Conclusions: NGS on DNA isolated from nail specimens accurately and efficiently detects germline P/LP variants in patients with AML and MDS when compared to gold standard skin fibroblast testing. In addition, adequate DNA for germline testing can be isolated from nail specimens. Our data suggests that the risk that peripheral blood contamination of nail specimens will lead to somatic variants being erroneously classified as germline is minimal when the NGS testing of DNA extracted from nail specimens is paired with tumor NGS testing from bone marrow aspirate samples.