Rakovina Therapeutics has presented compelling preclinical data on AI-discovered ATR/mTOR inhibitors with central nervous system (CNS) penetration at the 2025 Society for Neuro-Oncology (SNO) Annual Meeting in Honolulu, Hawaii. The findings, highlighted in a poster titled “Discovery and development of a novel CNS-penetrating ATR inhibitor: Dual inhibition of ATR and mTOR in PTEN-deficient tumors,” showcase the company’s use of Variational AI’s Enki™ generative AI platform to design novel dual inhibitors targeting two key cancer-driving pathways—ATR and mTOR—specifically engineered to cross the blood–brain barrier. This represents a first-of-its-kind approach, as no prior therapeutic has combined ATR and mTOR inhibition in a single molecule with demonstrated CNS penetration. The lead compounds were designed to address a critical limitation of current ATR inhibitors, which largely fail to reach brain tumors due to poor CNS distribution. The AI-generated virtual library included 138 predicted compounds, with 43 selected for synthesis and evaluation. Multiple molecules achieved over 50% ATR inhibition at 1 µM and demonstrated enzymatic potency equal to or greater than leading ATR inhibitors such as ceralasertib, tuvusertib, and elimusertib, while maintaining selectivity within the PIKK family of kinases. Preclinical testing in mice showed favorable pharmacokinetic profiles following a single 5 mg/kg intraperitoneal dose, including good tolerability, metabolic stability in human liver microsomes, and measurable brain exposure—key indicators for advancing into brain tumor models. PTEN deficiency is a common driver in aggressive cancers with high rates of CNS metastasis, including non-small cell lung cancer, breast cancer, ovarian cancer, melanoma, and prostate cancer. Loss of PTEN removes a critical brake on the PI3K/AKT/mTOR pathway, promoting tumor growth, genomic instability, and resistance to therapy. In these cancers, mTOR often acts as an adaptive escape route when ATR is inhibited, undermining the effectiveness of single-agent ATR therapies. Dual inhibition offers a promising strategy to overcome this resistance. The data presented at SNO underscore the potential of Rakovina’s AI-driven approach to tackle cancers with significant unmet medical needs. Prof. Mads Daugaard, President and Chief Scientific Officer, emphasized that the ability of generative AI to propose such precise, multi-targeted molecules marks a transformative shift in drug discovery. Jeffrey Bacha, Executive Chairman, noted the strong reception of the data and highlighted the synergy between Variational AI’s Enki™ platform and the translational expertise at the Vancouver Prostate Centre, enabling rapid development of differentiated DNA-damage response (DDR)-targeted therapies. Rakovina Therapeutics, based in Vancouver, British Columbia, is advancing AI-powered cancer therapies using its proprietary Deep-Docking™ and Enki™ platforms. The company aims to progress one or more candidates into clinical trials through partnerships with pharmaceutical firms. Further details are available at www.rakovinatherapeutics.com. Forward-looking statements in this release include expectations regarding the company’s business plan, drug development timeline, and strategic goals. These statements involve risks and uncertainties, including those related to biopharmaceutical development, regulatory approval, market conditions, and competition. Actual results may differ materially. The company undertakes no obligation to update these statements beyond required disclosures. More information on risks can be found in Rakovina’s filings on SEDAR+. For inquiries, contact Michelle Seltenrich, Director of Corporate Development, at [email protected] or 778-773-5432.