Abstract 1075: Piloting NTRK fusion-specific oncogenicity guidelines - lessons learned
Bibliographic record
Abstract
Abstract Gene fusions involving neurotrophic receptor tyrosine kinase genes (NTRK1, NTRK2, & NTRK3) are well-established oncogenic drivers and serve as critical diagnostic and therapeutic markers in cancer. Interpreting their clinical significance is a high priority given FDA approval of TRK inhibitors; however, this remains challenging due to the rapid pace of fusion discovery, the diversity of fusion partners and tumor types, inconsistent and incomplete reporting of fusion data elements, and the lack of standardized fusion-specific classification guidelines. The Clinical Genome Resource (ClinGen) NTRK Fusions Somatic Cancer Variant Curation Expert Panel (SC-VCEP) is addressing these challenges and creating a publicly available resource of high-quality clinically significant NTRK fusion classifications in the Clinical Interpretation of Variants in Cancer (CIViC; civicdb.org) knowledgebase to support patient care. Our NTRK fusion-specific oncogenicity guidelines (approved April 2022) classify NTRK fusions as Oncogenic, Likely Oncogenic, Fusion of Unknown Significance (FUS), or Benign based on four categories: Fusion Gene Structure (orientation/breakpoints/reading frame), Cancer Association (number of unique cases), Clinical Validity (response to targeted inhibitors), and Functional Status (pathway activation or expression). Piloting our guidelines on a range of common to rare NTRK fusions found in cancers resulted in 12 Oncogenic Assertions (6 Oncogenic, 1 Likely Oncogenic, 2 FUS, 3 Benign), 5 Diagnostic Assertions covering 3 cancers, and 10 Predictive Assertions supporting sensitivity to FDA approved TRK inhibitors, larotrectinib or entrectinib. This pilot introduced several modifications to our oncogenicity guidance including: 1) reducing case numbers required to reach cancer association or clinical validity due to the rarity of reported NTRK-positive tumors; 2) further clarifying NTRK fusion structure, e.g., both gene partners must contribute to the fusion protein sequence; 3) requiring fusions to be reported in the published literature as databases may lack vetting; 4) expanding our NTRK-associated tumor list as more tumors are screened for the presence of NTRK fusions; 5) strengthening Benign support by requiring the confirmed lack of both RNA and protein fusion expression. This process was a valuable step in enhancing the consistency and clarity of the ClinGen NTRK SC-VCEP fusion-specific oncogenicity guidelines to aid their widespread adoption. Future efforts will focus on the Oncogenic classification of the over 80 known NTRK fusions found in cancers and their distribution to the public through the CIViC knowledgebase. Citation Format: Jason Saliba, Shivani Golem, Arpad Danos, Laura B. Corson, Elan Hahn, Morteza Seifi, Emma G. Sullivan, Jan Clement A. Santiago, Valentina Nardi, Theodore W. Laetsch, Marilyn M. Li, Obi L. Griffith, Malachi Griffith, Gordana Raca, Larissa V. Furtado, Alanna J. Church, Angshumoy Roy. Piloting NTRK fusion-specific oncogenicity guidelines - lessons learned [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 1075.
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 | 0.087 | 0.211 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.004 | 0.003 |
| Science and technology studies | 0.002 | 0.002 |
| Scholarly communication | 0.008 | 0.010 |
| Open science | 0.007 | 0.006 |
| Research integrity | 0.008 | 0.012 |
| Insufficient payload (model declined to judge) | 0.018 | 0.012 |
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".