Shallow Whole Genome Sequencing of Circulating Tumor DNA Is Predictive of Outcome in Diffuse Large B-Cell Lymphoma
Bibliographic record
Abstract
Introduction Patients with diffuse large B-cell lymphoma (DLBCL) have heterogeneous outcomes, which necessitates the discovery of novel prognostic biomarkers. Plasma from peripheral blood is a minimally invasive source of circulating tumor DNA (ctDNA). Prior studies using targeted DNA sequencing have revealed that pre-treatment ctDNA levels are predictive of outcome in DLBCL (Kurtz et al. 2018, etc.). However, such assays are expensive, and there remains a need for cost-effective methods to reliably estimate tumor burden in plasma. To this end, we evaluated the clinical utility of ctDNA in patients with newly diagnosed DLBCL using shallow whole genome sequencing (sWGS) - a low-cost method to estimate tumor burden without prior knowledge of tumor mutations. Methods sWGS (intended coverage 0.1-1X) was performed on plasma from peripheral blood samples of patients with newly diagnosed DLBCL at our institution. ichorCNA was used to estimate tumor fraction in plasma from sWGS (Adalsteinsson et al. 2017). FDG PET-CT scan was performed for disease burden assessment. Targeted DNA sequencing (hybridization capture) for recurrently mutated genes in lymphoma was performed on plasma samples; ctDNA concentration was expressed as log haploid genome equivalents per mL of plasma. The Fisher's exact test, Mann Whitney test and Pearson's correlation were used as appropriate to assess association between variables. Results We performed sWGS on pre-treatment plasma samples from 53 patients with newly diagnosed DLBCL (median age 62 years). The median follow-up time of the cohort was 2.6 years (range: 0.02 - 4.8 years). The majority (47/53, 89%) of patients were treated with curative-intent chemoimmunotherapy regimens including R-CHOP (n=42), R-CHOP/HD-MTX (n=4) and DA-EPOCH-R (n=1). FDG PET-CT data from time of diagnosis was available for 50 (94%) patients. High tumor fraction, as stratified by the median value (0.07, range: 0 - 0.61), in pre-treatment plasma was associated with established measures of tumor burden, including advanced-stage disease (Ann Arbor stage III/IV; 69% vs. 22%; P < 0.001), elevated serum lactate dehydrogenase (median 351 vs. 250 U/L; P < 0.001), increased tumor size (maximum diameter; median 7 vs. 4 cm; P < 0.001), and >1 extranodal site (38% vs. 7%; P = 0.009). Tumor fraction in pre-treatment plasma was correlated with total lesion glycolysis assessed by FDG PET-CT (R = 0.51, P < 0.001). Targeted DNA sequencing was performed in pre-treatment plasma samples for 50 (94%) patients, of which the majority (47/50, 94%) had detectable mutation(s) in plasma. Tumor fraction quantified using sWGS was concordant with ctDNA concentration as determined by targeted DNA sequencing (R = 0.75, P < 0.001). Among patients treated with curative-intent chemoimmunotherapy regimens, high tumor fraction in pre-treatment plasma, as determined by sWGS, was associated with inferior progression-free survival (PFS, 2-year PFS 53% vs. 94%, P < 0.001) and overall survival (OS, 2-year OS 60% vs. 97%, P = 0.003). In multivariable analysis, high cell-free tumor fraction remained a predictor for inferior PFS but not OS (PFS: P = 0.037, OS: P = 0.102), independent of International Prognostic Index scoring and total lesion glycolysis. Conclusions Tumor fraction in plasma quantified using sWGS is correlated with established measures of tumor burden in DLBCL and orthogonal measures of tumor burden in plasma. Importantly, high tumor fraction is associated with inferior outcomes in patients with newly diagnosed DLBCL. Our study highlights the potential clinical utility of sWGS performed on ctDNA as a minimally invasive assay for DLBCL risk assessment.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| 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".