Accounting for the Unique Molecular Landscape of Pediatric Malignancies Improves Target-Agent Pair Identification for Pediatric Precision Oncology
Bibliographic record
Abstract
Abstract Precision medicine has significant potential to improve therapy options for patients who have exhausted treatment options for their relapsed or refractory cancers. Next-generation sequencing has enabled the detection of genetic abnormalities in individual tumours that bestow sensitivity to selective agents targeting these dysregulated cellular pathways. Several programs have recently launched that utilize sequencing technology to identify patient mutations and match the patients to clinical trials for agents that target the dysregulted pathway. The Pediatric MATCH trial initiated by the National Cancer Institute (NCI) and Children's Oncology Group (COG) exemplifies this approach, but also emphasizes the numerous obstacles involved in molecularly-guided therapies The target and agent prioritization strategy for the Pediatric MATCH trial was recently outlined. It employs an amplicon-based targeted sequencing panel (OCAV3) that was originally developed to capture informative variants for adult tumours, with the current version 3 including a set of pediatric-specific variants. However, the landscape of genomic alterations across pediatric malignancies differs from adult malignancies, which translate to distinct oncogenic drivers. The fundamental differences in the genomic landscapes observed in adult and pediatric malignancies must therefore be considered in the optimization of target identification for precision oncology. We hypothesized that a pediatric-focused targeted sequencing panel (OCCRA) will be more informative than an adult-focused panel (OCAV3) for pediatric malignancies. To compare target-agent pairs identified by OCAV3 and OCCRA, we performed a retrospective analysis of 28 childhood tumour samples, 10 B-ALL, 4 T-ALL, 6 neuroblastomas, and 8 additional solid tumours. For 12 of these tumours, whole genome sequencing (WGS) was performed on matched samples and all sequencing results were filtered for pediatric cancer driver genes. For the 12 samples tested by all three sequencing modalities, we found that the greatest discordance occurred in samples with low (<25%) tumour content. In these samples, however, all three modalities gave highly concordant detection of pediatric driver mutations, indicating the robustness of amplicon-based sequencing. Following the prioritization strategies outlined by the Pediatric MATCH committee, at least one target-agent pair was detected for 16 of 28 samples sequenced by the OCAV3 panel with 23 total agents identified due to multiple variants per sample. Using the OCCRA panel, however, we detected at least one target-agent pair for 19 of 28 samples with 25 total agents. Importantly, a fusion was detected by OCCRA in 3 samples, whereas OCAV3 was unable to detect these fusions due to panel content. Of particular note was the detection of homozygous loss in CDKN2A, which was the most commonly identified target, assigned seven times for OCAV3 and 11 times for OCCRA. Although genomic loss in CDKN2A or CDKN2B is not validated by the manufacturer, orthologous verification through clinical reports and WGS indicates that the OCCRA panel called nine of nine verifiable losses in these genes across the 28 samples. Variants were also detected within targetable pathways currently not included in the Pediatric MATCH trial, such as the JAK/STAT or tyrosine kinase pathways; inclusion of inhibitors targeting these pathways increased the total number of agents to 25 for OCAV3 and 29 for OCCRA. In addition, use of the OCCRA panel enabled detection of potentially targetable cancer driver mutations, including gains in RUNX1, ABL2, or ERBB2 and a SMARCA4 SNV, in several samples for which no established target-agent pairing was assigned. Overall, amplicon-based sequencing was highly reproducible across independent panels, robust to low tumour content, and enabled cost-effective target-agent pairing with short 2-3 day turnaround times. Furthermore the pediatric cancer-specific OCCRA panel improved the identification rate of target-agent pairs and resulted in more frequent matching with multiple agents. Disclosures No relevant conflicts of interest to declare.
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 | 0.003 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 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".