Artificial intelligence (AI) is increasingly being employed to improve the operational efficiency of clinical trials, accelerate drug discovery and minimise costs.
AI can be used to unlock information from a diverse range of data sources, from medical records and disease registries to scientific papers, and support faster and more meaningful patient recruitment.
AI has the potential to streamline the operational processes of clinical trials by allowing sponsors and contract research organisations (CROs) to track patient health from their homes, monitor responses to treatment and help keep on top of patient engagement.
AI can also predict which patients will benefit from treatment and those that could be at risk and could even circumvent the need for placebo-controlled arms and animal models.
Clinical Trials Arena profiles some of the most exciting AI start-ups aiming to improve various aspects of clinical trials.
Owkin
Owkin is a pioneering French AI start-up founded in 2016 by Thomas Clozel and Gilles Wainrib and headquartered in New York.
Based on a company mission to make clinical research more collaborative, the AI platform connects data scientists, clinicians, academic researchers and pharmaceutical companies to unlock medical insights from siloed, multimodal datasets and help pharma companies to discover new drugs and biomarkers, optimise clinical trials, and quickly identify patient populations of interest.
Owkin harnesses federated learning to train and develop its machine learning models to increase clinical trial efficiency.
The company has a catalogue of over 70 life models built using both public data and its own and interpretable AI, which enables further research and the identification of new biological targets.
The company is working with its biopharmaceutical partners to use their insights for clinical trials by identifying high-risk patients who might respond best to an experimental drug.
As part of a long-running Series A funding round, the startup closed $18m from Mubadala Capital and Bpifrance in June 2020, bringing the total amount raised in the round to $70m.
Unlearn.AI
Unlearn.AI is a San Francisco startup founded in 2017 by former Pfizer principal scientist Charles Fisher. The firm claims to be the first to create ‘digital twins’ to populate intelligent control arms in clinical studies.
In the context of clinical research, Unlearn explains, a digital twin is a “longitudinal, computationally generated clinical record that describes what would have happened if a specific patient received a placebo”. Digital twins integrate data from real patients with the aim of replacing them in placebo control groups. By removing the need for placebo-controlled cohorts, they could reduce the number of participants needed for clinical trials.
According to the firm, digital twins can also increase study power by decreasing variability, as well as powering secondary endpoints and exploratory analyses and providing additional evidence to understand how each patient responded to an experimental treatment in a trial.
Currently, the company is focusing its DiGenesis machine learning platform on Alzheimer’s disease and multiple sclerosis – diseases with few or no effective treatment options.
Unlearn announced in April 2020 that it had raised $12m in a Series A funding round.
Deep Lens
Founded in 2017 by Dave Billiter, TJ Bowen and Simon Arkell, Deep Lens is a startup harnessing AI to match the right patient with the right trial at the right time.
Hailing from Ohio, Billiter joined the state’s Nationwide Children’s Hospital in 2004, where he oversaw the creation of an AI-driven oncology studies suite that has now been used by clinicians at institutions all over the world to diagnose cancer. It became the foundation of Deep Lens’s free toolset – Virtual Imaging for Pathology Education and Research (VIPER.)
Now the firm is harnessing its network of partner institutions and a swarm of cloud-based data to identify eligible patients at the time of their cancer diagnosis to accelerate enrolment in clinical trials.
In April 2019, Deep Lens raised $14m in Series A funding led by Northpond Ventures, with participation from existing investors Rev1 Ventures, Sierra Ventures, and Tamarind-Hill Partners bringing the startup’s total raised to $17.5m.
In April 2021, Oregon Oncology Specialists signed a strategic collaboration to integrate Deep Lens’s AI-based clinical trial screening platform to automatically identify qualified patients to clinical trials.
AiCure
AiCure is a New York-based AI and advanced data analytics company that uses AI to understand how patients will respond to treatment and provides real-time monitoring of patient dosing and behaviour, with the aim of facilitating smaller, faster trials.
AiCure also improves the predictability of study timelines, reduces costs and accelerates timelines through remote patient engagement and assessments, including measuring digital biomarkers.
AiCure was founded by Adam Hanina and Laura Shafner in 2010 and funded by the National Institutes of Health (NIH) and other institutional investors. The firm has more than 65 issued patents and works with global clients in over 30 countries.
In April 2020 the firm appointed former Medidata founder Ed Ikeguchi as CEO to advance AiCure’s commitment to improve health “through an understanding of the science behind human responses to illness and treatment”.
AiCure’s new patient engagement application Patient Connect enables sites to monitor and analyse patient dosing behaviour for new routes of drug delivery beyond pills. Through AI-powered video monitoring, the app can also check whether patients are taking medicines correctly.
AiCure’s latest funding round was in 2019, raising $24.5m in Series C funding, bringing the total raised up to $52.8m.
VeriSIM Life
San Francisco biotech startup Verisim Life was founded in 2017 by veterinarian Jo Varshney, who also holds a PhD in comparative oncology and genomics. The firm has created a platform that enables AI-driven bio-simulations that de-risk drug research and development decisions by predicting the clinical value of investigational drugs before they have started human trials. The company’s BIOiSIM platform for small molecules, large molecules and viruses is designed to eliminate the need to test drugs on animals.
The AI-powered biosimulation models can quickly predict how a drug will interact with an animal’s biological system, allowing drug developers to accelerate the pre-clinical phase and move swiftly on to human trials.
Like Unlearn.AI, Verisim has said it plans to create digital twins at some point in the future.
In 2019, the company raised $5.2m in seed funding to expand its reach, bringing the total amount raised to $6.4m.
Source: clinicaltrialsarena.com
