Abstract
AbstractBackgroundTraditional Phase 1 trials often provide important drug development insights, which can be limited by ethical issues, costs, and lengthy timelines. Advancements in AI based simulations offer a potential avenue for mitigating these challenges. The present study used the aiHumanoid platform, specifically the upgraded DeepNEU database v8.1, to design and conduct a virtual Phase 1 trial, to assess the comparative efficacy and toxicity of standard chemotherapy with gemcitabine plus Taxol vs a combination of standard chemotherapy plus COTI-2 for treating Pancreatic Adenocarcinoma (PAC).MethodsApplying the updated DeepNEU database of 7267 genotypic and phenotypic concepts linked through 67491 relationships, the study used aiHumanoid simulations to predict outcomes from 30 virtual patients. Data from the standard chemotherapy arm and the standard treatment plus COTI2 arm were analyzed at 25%, 50%, 75%, and 100% of maximal dose. Estimates of efficacy and potential toxicities were based on a combination of the paired 2 tailed T test and Cohen’s d values as a true estimate of treatment effects.ResultsThe novel combined treatment regimen, especially at 100% dosage, showed medium to large treatment effects on the entire Pancreatic Adenocarcinoma disease profile. Notably, a significant decrease was observed in all disease profile components, bolstered by p-values less than 8.68E-5 and Cohen’s d values >=0.335. While evidence hinted at an increased bone marrow toxicity in the novel treatment arm, no individual organoid toxicity exceeded one standard deviation above predicted values. Importantly, COTI-2 treatment demonstrated a dose-dependent increase in p53 levels, significant at p < 0.006.ConclusionThis aiHumanoid’s virtual Phase 1 trial emphasizes the potential of computational simulations in the drug development process. Our findings indicate a promising treatment pathway combining COTI-2 with standard chemotherapy for Pancreatic Adenocarcinoma. Ongoing development and validation of the aiHumanoid based virtual Phase 1 clinical trial methodology is warranted.
Publisher
Cold Spring Harbor Laboratory
Reference43 articles.
1. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries CA Cancer;J Clin,2018
2. American Cancer Society. Cancer Facts & Figures 2022. Atlanta: American Cancer Society; 2022. https://www.cancer.org/research/cancer-facts-statistics/all-cancer-facts-figures/cancer-facts-figures-2022.html
3. Whole genomes redefine the mutational landscape of pancreatic cancer
4. Subtypes of pancreatic ductal adenocarcinoma and their differing responses to therapy
5. Preclinical models of pancreatic ductal adenocarcinoma