Abstract A009: AI-driven metabolomics and synthetic drug design for targeting CNS metastases
Bibliographic record
Abstract
Abstract Background: CNS metastases affect ∼30% of patients with advanced-stage cancers, however treatment is largely limited by the blood-brain barrier (BBB), which restricts drug penetration. Based on previous preclinical models, tumor cells adapt to the brain’s microenvironment through metabolic shifts, such as 2-3-fold increases in glutamine metabolism. Metabolomic profiling has identified biomarkers, including lactate, associated with tumor progression. Additionally, synthetic inhibitors that target metabolic enzymes, such as glutaminase, function by BBB permeation, emphasizing the optimization for tailored drug synthesis that maintains potency and pharmacokinetic properties. This study aims to analyze MRI image datasets from preclinical CNS metastasis models to identify biomarkers and propose optimized inhibitors to enhance drug efficacy. Methodology: The data was collected from Stanford University’s BrainMetShare, and it includes 156 pre- and post- contrast brain MRI images in patients with ≥1 CNS metastases. The primary tumors were lung (n=99), melanoma (n=7), breast (n=33), gastrointestinal (n=5), genitourinary (n=7), and miscellaneous (n=5). An in-house segmentation process via Matlab was used to segment the MRI images by slices/pixels and the tumor surfaces for each sample were reconstructed using the “isosurface” and gaussian kernel functions. The size, volume, and location of the tumors were calculated to identify morphological variables that can detect metabolite biomarkers. Results: The initial MRI segmentation showed that the tumor ranges were between 2.0 cm^3 to 4.3 cm^3 (mean: 3.3 +/- 0.7 cm^3). A vast majority of the metastases were found in the cerebellum and brainstem with more advanced tumors being correlated with increased glutamine metabolism. This data will be used to determine effective concentrations for four inhibitors that previous studies have found to have promising BBB permeability: CB-839, BPTES, EPCG derivatives that inhibit LDHA, and 6-Diazo-5-oxo-L-norleucine (DON). Discussion/Conclusion: Ongoing AI-driven analysis will be done to further evaluate the effectiveness of the inhibitors as plausible therapeutic approaches. Citation Format: Rajvi Babaria, Satviki Shankaran, Prarthana Daswani, Tylaah George, Kamau Pierre. AI-driven metabolomics and synthetic drug design for targeting CNS metastases [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Artificial Intelligence and Machine Learning; 2025 Jul 10-12; Montreal, QC, Canada. Philadelphia (PA): AACR; Clin Cancer Res 2025;31(13_Suppl):Abstract nr A009.
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 | 0.000 |
| 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 | 0.001 |
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".