Our approach

Optimizing clinical development

Cutting-edge machine learning to de-risk and accelerate clinical trials


Clinical trials are slow and expensive

Challenge 1
Slow and expensive trials because researchers must increase recruitment due to patient heterogeneity to adequately power their trial
Challenge 2
The non-data-driven definition of inclusion criteria and selection of covariates make it hard to detect treatment signal from the noise
Challenge 3
Researchers lack early efficacy information on the asset to inform phase transition decisions

Our solution: Optimized Development Engine

Better trials mean better drugs, sooner

We optimize each phase of clinical development by applying novel machine learning methodologies.
Phase I/II
Phase II/III
Phase III
AI external control arms
We provide early estimates of efficacy for single-arm phase I/II clinical trials.
Increased probability of phase III trial success through informed phase transition decisions.
Inclusion criteria models
We develop biomarker models to inform trial recruitment.
Increased probability of phase II/III trial success through better defined patient populations.
Data driven covariate adjustment
We develop prognostic biomarker models to identify key covariates linked to outcome from external data analysis.
Increased probability of phase III trial success through increased statistical power and broader inclusion criteria.
Get in touch to learn more

AI drug development optimization

Why Owkin?

Regulatory savy
Engaged with regulators

Continued conversations with EMA and FDA including an EMA letter of support for our data-driven covariate adjustment approach and a CE-marked AI diagnostic tool: MSIntuit CRC.

Robust external data access
Robust, up-to-date, multimodal external data

Access to continuously enriched, up-to-date, multimodal data representing the latest standard of care and capturing disease heterogeneity and the tumor microenvironment.

Leading AI team
Leading AI and biostatistics team

Renowned AI expertise with # published papers in leading journals, two strategic alliances with leading biopharma companies, and AI digital pathology diagnostics.

Get in touch
“Expanding eligibility criteria and recruiting more diverse patient populations is becoming more of a necessity from the viewpoint of regulators. Regulators are very much, I believe, willing to consider these analytical methods to either maintain power or increase power and reduce background noise.”
Sean Khozin
CEO, CancerLinQ LLC (Ex FDA Oncology Center of Excellence)

We respond to the rapidly changing regulatory landscape

New guidelines

New guidelines from regulators encourage use of real-word evidence. FDA (2021) and EMA (2015) published guidelines on the use of RWE in regulatory submissions. Both regulators outlined that covariate adjustment improves the efficiency of analysis and produces stronger and more precise evidence if the covariates are prognostic.

EMA letter of support

The European Medicines Agency (EMA) has issued Owkin with a letter of support for our proposed statistical adjustment on deep learning prognosis covariates obtained from histological slides. Our method uses the predictions of two deep-learning models, MesoNet and HCCnet, as prognostic biomarkers for the adjustment of efficacy analysis on the overall survival of life-prolonging drugs in randomized phase II and phase III clinical trials.