Abstract 2917: Utilizing artificial intelligence for the discovery of a novel CNS-penetrating ATR inhibitor
Bibliographic record
Abstract
Abstract Introduction: Ataxia telangiectasia and Rad3-related protein serine/threonine kinase (ATR) plays a crucial role in regulating DNA damage repair. ATR inhibitors have shown promise against solid tumors, not only as monotherapies, but also in combination with chemotherapy, radiotherapy, and immunotherapy with several clinical trials ongoing. The ATR inhibitors currently in clinical development have low CNS-penetration and are therefore sub-optimal for the treatment of brain tumors and brain metastases. It has been proposed that ATR inhibitors could be used as a monotherapy or could be used to potentiate the cytotoxicity of radiotherapy, one of the mainstay treatments for brain tumors. Development of novel, potent, CNS-penetrating ATR inhibitors could provide a new treatment alternative for patients with few options. Traditional drug discovery methods are time-consuming and costly, necessitating innovative approaches. Here, we describe the application of the Enki™ platform, an artificial intelligence approach, in the discovery of a novel ATR inhibitor for use against brain tumors. Methods: The Enki™ platform utilizes generative AI and deep machine learning to search a broad chemical space to identify hits. It uses variational autoencoders to optimize many properties simultaneously within the search space, including maximal potency against the primary target, selectivity, ADMET, and physicochemical properties. This platform was used to generate de novo molecules with optimized properties to penetrate the blood-brain barrier and specifically target ATR. Results: We will present preliminary results from the de novo molecule generation by the Enki™ platform. Validation of in vitro data, including ATR inhibition and selectivity, BBB permeability and metabolic stability, will be described. Conclusion: The Enki™ platform is being leveraged for the development of a CNS-penetrant ATR inhibitor, significantly accelerating the drug discovery and development process. Use of this system enables a broader exploration of the vast chemical space and refinement of drug properties, both shortening the drug development process and increasing specificity and efficacy of potential therapeutics. Citation Format: Sarah Truong, Beibei Zhai, Louise Ramos, Mona Marzban, Fariba Ghaidi, Peter Guzzo, Mehran Khodabandeh, Ahmad Issa, Seyed A. Saberali, Jason Rolfe, Sarah Omar, Marshall Drew-Brook, John Langlands, Dennis Brown, Jeffrey Bacha, Mads Daugaard. Utilizing artificial intelligence for the discovery of a novel CNS-penetrating ATR inhibitor [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 2917.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| 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 | 0.001 |
| 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".