Ultra-deep whole-genome sequencing of cell-free DNA recovers most bone marrow-derived mutations in newly diagnosed multiple myeloma
Bibliographic record
Abstract
Abstract Introduction Cell-free DNA whole-genome sequencing (cfWGS) offers a non-invasive alternative to molecular profiling of multiple myeloma (MM), potentially reducing reliance on invasive bone marrow (BM) biopsies. Standard-depth (~30–40×) cfWGS reliably estimates tumor fraction and detects broad copy number alterations, but often fails to capture somatic single-nucleotide variants (SNVs) present at low variant allele frequencies (VAFs) due to tumor subclonality or low tumor shedding. Recent studies in solid tumors suggest that pushing cfWGS to ultra-deep coverage (≥100×) can rescue low-frequency somatic mutations, yet this has not been explored in MM. With sequencing costs declining, ultra-deep cfWGS has potential as a comprehensive tool for MM genomic profiling directly from peripheral blood (PB) cfDNA. To determine whether ultra-deep cfWGS can match the mutational yield of BM WGS and capture actionable variants, we compared standard-depth (40×) Illumina cfWGS with 150× cfWGS (targeted 200×) generated using the Ultima Genomics platform. Methods Paired BM and PB cfDNA samples with matched buffy coat germline controls were collected at diagnosis from 11 transplant-eligible patients enrolled in the Multiple Myeloma Molecular Monitoring (M4) study (n=8) and IMMAGINE study (n=3). DNA from BM CD138+ selected cells underwent 30-80× WGS on Illumina NovaSeq 6000 with somatic variants called versus a matched normal using MuTect2. Matched cfDNA libraries were prepared from a median of 20ng (range 20-83ng) of cfDNA extracted from 10-20mL of plasma for Illumina, and from 50ng (range 22-150ng) for Ultima. Libraries were sequenced to 40× on Illumina (aligned with BWA-MEM/GATK; variants force-called using MRDetect) and 150× (targeted 200×) on Ultima Solaris (prepared with PPM-Seq; variants force-called via Ultima's ‘Single Read SNV’ pipeline). Tumor fractions were estimated using ichorCNA. Results At diagnosis (median age 57, range 41–75 years; 6 male, 5 female), 4 patients were high risk, 2 standard, and 5 unknown. Subtypes were 4 IgG, 5 IgA, 1 light-chain only, and 1 unknown; ISS stages: I (n=3), II (n=6), III (n=1), and unknown (n=1). BM WGS identified a median of 3,087 somatic SNVs (range 467–4,059). Matched cfDNA tumor fraction was 7.6% (range 4.5-33.7%). Median deduplicated genome-wide coverage was 49× (range 25–57×) for Illumina and 155× (range 116–320×) for Ultima cfWGS, with median alignment rates of 98.4% and 99.0%, respectively. Duplication rates were higher for Ultima (median 20.2%, range 13.3–21.3%) than Illumina (median 8.5%, range 5.5–17.4%), consistent with increased sequencing depth. While this showed diminishing returns on efficiency at higher depths, coverage was sufficient to recover low-VAF mutations. Ultra-deep 150× cfWGS recovered significantly more somatic SNVs compared to 40× cfWGS (median 2,571 vs. 1,861; median increase of 21%, range -2–87%; paired t-test, p < 0.01). Compared to BM tumor DNA, 150× cfWGS recovered a median of 88% of somatic mutations (range, 67–97%) versus 66% (range, 36–91%) at 40×. Notably, 150× cfWGS detected 69% (11/16) of OncoKB-annotated oncogenic or likely oncogenic variants (including NRAS p.Q61R, KRAS p.A146V, and TP53 p.M237I), vs. 31% at 40×. Mutation recovery gains with ultra-deep sequencing were inversely correlated with cfDNA tumor fraction (Spearman’s ρ = –0.83; p < 0.01), with the greatest improvement (median 73%, range 33–87%) in cases with low (≤6%) cfDNA tumor fraction. Among variants present at ≥7.5% VAF in Ultima data, 4 of 5 actionable mutations were also detected at 40×, indicating strong concordance for high-burden alterations. Seven additional actionable mutations were uniquely identified at a 150× coverage, predominantly at low allele frequencies (≤6%, median VAF 3.8%, range 1.7–13.2%). Conclusions Ultra-deep cfWGS at 150× using Ultima sequencing recovered over twice as many actionable somatic mutations compared to 40× Illumina, despite lower sequencing costs in this pilot cohort. This tumor-informed analysis demonstrated that somatic mutations identified by Illumina BM WGS could be reliably recovered from cfDNA sequenced on the Ultima platform. These findings support the feasibility of ultra-deep cfWGS for comprehensive genomic profiling in MM and motivate further evaluation of cfDNA-based analyses as less invasive alternatives to BM biopsies. Future work will validate these findings in larger cohorts and extend analyses to other variant types, including translocations.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".