{"entity":{"id":"idea-bio1-in-silico-trials-dose","kind":"idea","name":"In silico trials to choose the dose before the first patient","aka":[],"tldr":"Simulating thousands of virtual patients on a computer can suggest which dose and schedule to test, so fewer real patients receive doses that are too high or too low.","summary":"Quantitative systems pharmacology and mechanistic tumour growth models, calibrated on prior trial data, can simulate exposure-response across virtual populations. Regulators already accept model-informed drug development for paediatric extrapolation and some dosing decisions. Project Optimus requires dose optimisation; simulation could narrow the candidate schedules before the randomised dose-comparison stage, saving patients and time.","asOf":"2026-09-08","links":[{"label":"Bottleneck evidence (Lab models that fail to predict what happens in patients): Wong, Siah & Lo, Estimation of clinical trial success rates (Biostatistics 2019)","url":"https://doi.org/10.1093/biostatistics/kxx069"}],"tags":[],"related":[],"cancers":[],"sections":[],"technologies":["ai-drug-design"],"targets":[],"drugs":[],"companies":["insilico-medicine"],"institutions":[],"pathways":[],"terms":[],"trials":[],"people":[],"bottlenecks":["b-preclinical-models","b-dose-optimisation"],"keyPapers":["paper-wong-biostatistics"],"journals":[],"dependsOn":[],"notes":[],"hypothesis":"For agents where a model-informed schedule was proposed prospectively, the simulation-selected dose matches the eventually recommended phase 2 dose more often than the traditional maximum tolerated dose approach.","rationale":"Most oncology drugs were historically dosed too high; the exposure-response and toxicity data needed for simulation usually exist by end of phase 1 but are analysed informally.","test":"Retrospective blinded simulation of 20 agents with known optimised doses, then prospective use in three phase 1 programmes with the model prediction locked before dose expansion.","maturity":"speculative","actor":"regulator","cost":"small","horizonYears":4},"route":"/ideas/idea-bio1-in-silico-trials-dose/","neighbours":{"technology":[{"id":"ai-drug-design","kind":"technology","name":"AI-driven drug & target discovery","route":"/technologies/ai-drug-design/"}],"company":[{"id":"insilico-medicine","kind":"company","name":"Insilico Medicine","route":"/companies/insilico-medicine/"}],"bottleneck":[{"id":"b-preclinical-models","kind":"bottleneck","name":"Lab models that fail to predict what happens in patients","route":"/bottlenecks/b-preclinical-models/"},{"id":"b-dose-optimisation","kind":"bottleneck","name":"Wrong doses","route":"/bottlenecks/b-dose-optimisation/"}],"paper":[{"id":"paper-wong-biostatistics","kind":"paper","name":"Estimation of clinical trial success rates and related parameters","route":"/key-papers/paper-wong-biostatistics/"}],"roadmap":[{"id":"ai-oncology-roadmap","kind":"roadmap","name":"AI in oncology roadmap: pattern readers → foundation models → agents in the workflow","route":"/roadmaps/ai-oncology-roadmap/"},{"id":"drug-discovery-roadmap","kind":"roadmap","name":"Drug discovery roadmap: screening in mice → maps of dependency → designing in silico","route":"/roadmaps/drug-discovery-roadmap/"}]}}