Abstract A035 Characterization of circulating tumor DNA in a pediatric oncology cohort and implementation into the SickKids Cancer Sequencing (KiCS) precision oncology program
Bibliographic record
Abstract
Abstract Introduction Sensitive methods to monitor disease and identify novel therapeutic targets are key to improving survival for children with hard-to-cure cancers. Circulating tumor DNA (ctDNA) analysis is a promising, non-invasive technique that may improve outcomes. SickKids Cancer Sequencing (KiCS) is a precision oncology program that performs next-generation sequencing on samples from pediatric patients (pts) with rare or hard-to-cure tumors. To date, KiCS has enrolled >800 pts, collecting paired tumor/blood samples and clinical data. Here we describe the implementation of a plasma biobank initiative and preliminary results from our first cohorts. Methods Serial blood samples were collected from pts at scheduled timepoints, at time of treatment change, or if recurrence was suspected. Plasma and matched cell pellets were separated and banked. Extracted cell free DNA (cfDNA) that met adequate QC metrics by Qubit and Bioanalyzer was subjected to low-pass whole genome sequencing (LP-WGS) (1-2X coverage) and/or deep WGS (60-80X) and/or custom-designed target exome panel (cancer panel, CP, 1000X) sequencing. The bioinformatics analysis pipeline included single nucleotide variant (SNV) calling with Mutect2/Sage and filtering using a custom script; copy number variant (CNV) calling with ichorCNA/ PURPLE (for WGS) and CNVkit (for CP). ctDNA data were compared to available tumor and germline data from the same patient. Results Over 400 plasma samples from 167 pts have been banked. In our initial pilot, cfDNA was successfully extracted from 8 plasma samples (pts with neuroblastoma (n=4), rhabdomyosarcoma (n=3), GIST (n=1)). The concentration of cfDNA was comparable to published adult cohorts (average 28.4 ng/mL, median 7.53 ng/mL) though total yield was decreased due to lower collection volumes. cfDNA fragment size was within expected ranges for all samples. From deep WGS, tumor informed SNVs were called in 3/8 samples (estimated ctDNA fraction: 1.83% - 83%). Comparable CNV profiles were identified in 6/8 samples. Three additional samples, with known tumor CNV profiles, were submitted for LP-WGS and CP, yielding comparable profiles and tumour fractions, confirming that LP-WGS is an appropriate and sufficient tool to screen for tumor fraction. We next focused on baseline samples from pts with solid tumors and active disease (n=24). With an optimized extraction protocol, we obtained higher cfDNA concentrations (average=67.9 ng/mL; median=17.7 ng/mL). Samples with >35 ng cfDNA available were selected for LP-WGS (n=15). LP-WGS will be used to screen for tumour fraction of cfDNA samples. Samples with relevant tumour fraction will then be subjected to deep WGS and CP. Conclusion We implemented a robust blood collection and cfDNA sequencing workflow within the KiCS program. We successfully isolated cfDNA from minimal plasma volumes and identified tumor-informed SNVs and CNVs in ctDNA. Ongoing work focuses on optimizing our workflow and sequencing pipelines, with the aim to provide reliable clinically useful results with limited amounts of ctDNA. Citation Format: Sarah Cohen-Gogo, Taegi Choi, Ryan Ripsman, Matt Hudson, Sandy Fong, Reem Khan, Rosemarie E. Venier, Tristan Charlinski, Anita Villani, David Malkin, Adam Shlien. Characterization of circulating tumor DNA in a pediatric oncology cohort and implementation into the SickKids Cancer Sequencing (KiCS) precision oncology program [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Advances in Pediatric Cancer Research; 2024 Sep 5-8; Toronto, Ontario, Canada. Philadelphia (PA): AACR; Cancer Res 2024;84(17 Suppl):Abstract nr A035.
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) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.005 | 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".