Abstract PO-042: Next generation liquid biopsy for diagnosis and prognostication in HPV+OPSCC
Bibliographic record
Abstract
Abstract Introduction: Circulating tumor (ct)DNA approaches are revolutionizing cancer diagnostic and monitoring approaches. In HPV-associated oropharyngeal cancer (HPV+OPSCC), droplet digital PCR(ddPCR)-based detection of ctHPVDNA has gained significant attention but only demonstrates ~90% sensitivity at time of diagnosis and critically, performs poorly in the minimal residual disease setting, and is unable to provide additional prognostic information. For example, our group recently demonstrated the strong prognostic impact of single nucleotide polymorphisms (SNPs) in the HPV genome, which cannot be annotated by existing ddPCR approaches. To overcome these current limitations, we developed, optimized, validated, and applied a fully custom HPV whole genome hybrid-capture-based NGS assay termed HPVseek. Here, we evaluate the performance and capabilities of multiple blood-based tests, both as standalone assays and taken together in a machine learning approach, for HPV+OPSCC in a large patient cohort. Methods: Plasma samples were obtained from 150 HPV+OPSCC patients at time of presentation as part of a prospective observational study, along with 150 controls. From these, cell-free DNA was extracted and run on HPVseek, as well as on our existing multiplexed ddPCR assay. Plasma samples also underwent multiplexed HPV antibody profiling. Paired tumor samples from 25 patients additionally had gold-standard HPV whole genome sequencing performed for a head-to-head comparison of viral variant calling from plasma and tissue. HPV genotyping, lineage assignment, variant calling, fragmentomics, and antibody titers were performed for all samples by custom bioinformatic pipelines. A machine learning approach incorporating multiple features was applied to composite datasets to identify the most informative features for detection. Results: Sensitivity and specificity of HPVseek for detecting HPV+OPSCC were 99% and 100%, compared to 92% and 99% with ddPCR and 90% and 99% with E6 serology. HPV genotyping from blood was concordant with tissue in all positive samples, with 91% of patients having HPV16 (78% A1 sublineage), and the remaining patients with HPV 18, 33, 35, 45, 56, 66, 67, and 69, including patients with multiple genotypes detected in the blood. 100% genome coverage from blood was obtained from 91% of positive cases. 91% of HPV16 positive cases had complete coverage of prognostic HPV16 SNP regions, with 18 patients harboring at least one known high-risk SNPs. ctHPVDNA fragmentomics demonstrated a median fragment size of 141 bp, 26 bases shorter than that of non-cancer controls. Comparison of HPV variant calling from a gold-standard tissue assay and HPVseek demonstrated 95% concordance between high allele frequency variants. Conclusions: Here, we present the largest single-institution analysis of ctHPVDNA, utilizing a fully custom HPV whole genome NGS-based liquid biopsy, with matched HPV ddPCR, serology, and tissue WGS, demonstrating the powerful capabilities of a fully non-invasive test to both diagnose and prognosticate HPV+OPSCC at time of presentation. Citation Format: Brian Y. Zhao, Shun Hirayama, Saskia Naegele, Vasileios Efthymiou, Julia Mendel, Michael Lawrence, Viktor Adalsteinsson, Anthony J. Iafrate, Dipon Das, Daniel L. Faden. Next generation liquid biopsy for diagnosis and prognostication in HPV+OPSCC [abstract]. In: Proceedings of the AACR-AHNS Head and Neck Cancer Conference: Innovating through Basic, Clinical, and Translational Research; 2023 Jul 7-8; Montreal, QC, Canada. Philadelphia (PA): AACR; Clin Cancer Res 2023;29(18_Suppl):Abstract nr PO-042.
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 | 0.002 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
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".