Abstract B011: Dynamics of predicted tumor neoepitope burden in a pan-cancer solid tumor pediatric cohort
Bibliographic record
Abstract
Abstract Human leukocyte antigen (HLA) binding of tumor neoepitopes confers clinical value in certain adult malignancies. However, the prevalence of tumors that result in HLA binding of neoepitopes in pediatric malignancies is not as well-characterized. We set out to establish the feasibility of predicting neoepitope burden and the prevalence of predicted neoepitope across a previously established cohort of pediatric oncology patients. Additionally, because this analysis requires knowledge of each patient’s HLA haplotype for predicting binding of tumor peptides, we also set out to develop a novel algorithm to profile HLA haplotypes within the context of a larger whole genome (WGS) and RNAseq analysis pipeline. Finally, in this high-risk pediatric oncology cohort, we sought to determine the dynamics of predicted neoepitope burden at multiple time points in disease progression, including relapsed/refractory disease and metastatic disease. A previously established cohort of 147 high-risk pediatric oncology patients, including solid tumors CNS tumors, and leukemias/lymphomas was used for this analysis. This comprised patients with relapsed/refractory disease (66), and rare diagnoses (14). For these 147 patients, tumor/normal WGS (tumor ~60X; germline ~30X), as well as tumor RNAseq (polyA selected, ≥20 million reads) was performed. In addition to the initial timepoints, additional tumor samples were profiled for 27 patients, resulting in 179 total samples. A special focus on longitudinal analysis is devoted to osteosarcoma patients (12 with multiple timepoints). WGS and RNAseq were analyzed using a previously established pipeline. Somatic variants (SNVs), mutational burden, structural rearrangements (SVs), mutational signatures, and copy-number alterations (CNAs) were identified using WGS. As part of this new analysis, HLA class I haplotypes were identified from WGS integrated with RNAseq expression. Putative neoepitopes were predicted from expressed protein altering somatic variants using MHCFlurry. Importantly, HLA and neoepitope analysis was able to use intermediate data from the existing WGS/RNAseq analysis pipeline, resulting in significantly faster turnaround times. Our results demonstrate that our algorithm for determining HLA haplotypes by sampling already-mapped WGS and RNASeq performs with comparable accuracy to similar previously published methods that rely on unmapped data. In terms of predicted of tumor neoepitope burden, of the 171 samples with at least one protein altering variant, 166 are predicted to have at least one bound neoantigen (median=6). Of these, 133 samples are predicted to have at least one bound neoantigen that is clonal and expressed in RNA (median=3). Further characterization of the neoepitope burden of these tumors and the evolution of predicted neoepitope burden across multiple time points will be shared at the meeting. Citation Format: Charles W Macaulay, Marcus R Breese, E. Alejandro Sweet-Cordero. Dynamics of predicted tumor neoepitope burden in a pan-cancer solid tumor pediatric cohort [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Tumor Immunology and Immunotherapy; 2023 Oct 1-4; Toronto, Ontario, Canada. Philadelphia (PA): AACR; Cancer Immunol Res 2023;11(12 Suppl):Abstract nr B011.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.003 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| 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 teacher head, 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".