Trial Agents

Trial Agents — EU clinical-trial operational telemetry derived from CTIS authorisations, structured as a trial-by-service-provider-by-function matrix. For each authorised trial the data resolves: whether a CRO or CRO-like provider is engaged, which provider group (IQVIA/Q² Solutions, PPD, Clario/ERT/Bioclinica, ICON, Syneos, Medidata, Parexel), the outsourced functions present across a nine-category service taxonomy (central lab/bioanalysis/PK, imaging/ECG/central reading, patient recruitment/retention/reimbursement, data management/statistics, eClinical/IRT/randomisation, supplies/logistics,…

Trial Agents offers (Alternative, Fundamental, Supply Chain) — Demonstrated slice: 247 neurology Phase III CTIS trials, 1,087 country participations, 152 CRO-supported trials resolved across a nine-function taxonomy. Catalogue breadth: the reports library spans many indication and phase combinations (plaque psoriasis, gastric cancer, hypertrophic cardiomyopathy, ALS, IgA nephropathy, metastatic colorectal cancer, topical ophthalmic studies, hidradenitis suppurativa, C3 glomerulopathy, Stargardt disease, mitochondrial myopathy and more), so the underlying resolved-trial base plausibly runs to thousands of EU trials against a total CTIS universe in the tens of thousands.

CRO and clinical-supplier business development: win-likelihood and territory modelling from the demonstrated rule that CRO reliance becomes dominant above 6+ countries or 31+ sites, plus indication-level addressable workload (multiple sclerosis largest; myasthenia gravis and migraine highest adoption); sponsor sourcing and budgeting using which functions are actually outsourced rather than assumed (central lab 84.9%, imaging/central reading 74.3% versus monitoring only 42.1%); competitive market-share tracking across seven provider groups by indication and phase; healthcare-equity research on CRO volumes and mix as a forward indicator for listed providers; and supervised learning on trial-level outsourced-function prediction, where the retained non-CRO cohort gives genuine negative examples rather than an all-positive dataset