A 107-gene Nanostring assay effectively characterizes complex multiomic gastric cancer molecular classification in a translational patient-derived organoid model.
Bibliographic record
Abstract
4049 Background: Multi-omics profiling of gastric cancer (GC) has produced numerous molecular classification systems. However, widespread clinical implementation and testing of molecular subtypes are currently limited. Here, we develop, validate and implement a custom Nanostring assay capable of allocating GC molecular subtypes to clinical specimens in a translational patient-derived organoid model. Methods: Using publicly available whole-transcriptome data, machine learning models were developed to predict GC molecular subtypes from 376 Cancer Genome Atlas (TCGA) and 1797 Tumour Microenvironment Score (TME) patients. Models were generated using feature selection with 10-fold nested cross-validation. GC biopsies from 10 local patients were preserved in paraffin (tumour) and established as an organoid culture (organoid). Gene expression was measured using Nanostring. The allocation of molecular subtypes was internally and externally validated using gold-standard reference features in public databases comprising 2202 GC patients and 10 tumour-organoid pairs, respectively. We evaluated the concordance of tumour-organoid molecular subtypes and explored the correlation between subtype scores and in-vitro chemotherapy response. Results: Classification models for TCGA (57 genes) and TME (50 genes) predicted subtypes with an accuracy ± standard deviation of 89.46% ± 0.04 and 89.33% ± 0.02, respectively. Subtype assignment of microsatellite instability (MSI) in reference to capillary electrophoresis was found to have 99.3% [95% CI 97.4-99.9, n = 277] internal and 100% [95% CI 83.2-100, n = 20] external accuracy. In reference to Epstein-Barr Virus (EBV) in-situ hybridization, EBV type internal and external accuracy was 98.7% [95% CI 97.4-99.5, n = 552] and 100% [95% CI 83.2-100, n = 20], respectively. TCGA Genomically Stable (GS) scores followed a previously characterized enrichment of diffuse-type histology compared to intestinal-type in internal and external cohorts (Dunn’s Test, p < 0.0001 and p < 0.05, n = 1471 and n = 15, respectively). Statistically similar subtype scores (Paired Wilcoxon, p > 0.05) were found for tumour-organoid pairs. Discordance occurred in three tumour-organoid pairs. In-vitro Drug Sensitivity Score was not statistically efficacious in any molecular subtype, but Pearson correlation identified increasing efficacy with increasing EBV and MSI scores. Conclusions: Patient-derived organoids generally recapitulate the molecular subtype of parent tumours; however, in specific cases, subtype discordance occurs. Although additional external validation is required, our 107 gene assay effectively captures multi-omics classification systems in GC and allows future inquiry into the prognostic and therapeutic implications of these molecular subtypes.
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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| 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) | 0.001 | 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 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".