Abstract 1685: Deep learning approaches to deciphering intra-tumoural heterogeneity in glioblastoma
Bibliographic record
Abstract
Abstract BACKGROUND: Emerging evidence strongly implicates intra-tumoral heterogeneous biology in treatment resistance and disease progression across many cancer types. Thus, there is a need for workflows capable of systematically resolving and targeting distinct tumor subpopulations1. Using glioblastoma (GBM) as a prototype, I have aimed to leverage the computational power of Artificial Intelligence (AI) and deep learning to develop an autonomous workflow for the objective definition of biologically distinct tumor subpopulations2. OBJECTIVES: I hypothesize that AI may be leveraged as a tool to resolve spatial heterogeneity, by identifying tumoral subpopulations with unique molecular profiles and therapeutic targets. To highlight the need for routine analysis of tumor heterogeneity, I will address if: METHODS: I apply our developed image clustering workflows to quantify AI-defined subregions within a clinical cohort of 10 GBM patient tumors2,3. Laser capture microdissection and mass spectrometry-based proteomics are leveraged to address if AI-defined subregions show intra-tumoral molecular variation. Further, existing pharmacogenomic databases are utilized to carry out drug sensitivity and transcriptional clustering to define AI-defined region-specific therapeutic sensitivities and resistances across my clinical GBM cohort4. RESULTS: Preliminary data shows that region to region heterogeneity can be found in IDH wild-type GBM using our unbiased omics approach, in addition to predicting different pharmacogenomic sensitivities. CONCLUSIONS: This project aims to develop the first AI-driven tool to guide the routine and systematic molecular analysis of spatial morphogenomic heterogeneity. Further, this tool may have the potential to provide novel approaches for personalized care by selecting drug combinations that target a larger fraction of a tumor’s true biology. 1. Patel, A. P. et al. Single-cell RNA-seq highlights intratumoral heterogeneity in primary glioblastoma. Science (80-. ). 344, 1396-1401 (2014). 2. Faust, K. et al. Unsupervised Resolution of Histomorphologic Heterogeneity in Renal Cell Carcinoma Using a Brain Tumor-Educated Neural Network. JCO Clin. Cancer Informatics 811-821 (2020) doi:10.1200/cci.20.00035. 3. Roohi, A., Faust, K., Djuric, U. & Diamandis, P. Unsupervised Machine Learning in Pathology: The Next Frontier. Surgical Pathology Clinics vol. 13 349-358 (2020). 4. Barretina, J. et al. The Cancer Cell Line Encyclopedia enables predictive modeling of anticancer drug sensitivity. Nat. 2012 4837391 483, 603-607 (2012). Citation Format: Anglin J. Dent, Kevin Faust, Brian Lam, Alberto J. Leon, Queenie Tsang, Phedias Diamandis. Deep learning approaches to deciphering intra-tumoural heterogeneity in glioblastoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1685.
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".