Syneos and Haystack leverage AI to speed up clinical trials

Syneos-and-Haystack-leverage-AI-to-speed-up-clinical-trials.jpg
© Getty Images (Getty Images)

The partners announce that they will combine to use artificial intelligence to improve the identification and enrolment of patients for clinical trials.

Syneos Health and Haystack Health state that their combined capabilities will lead to more effective and efficient means of identifying and matching clinical trial opportunities for patients. The two partners have already been collaborating directly with sites and sponsors over a period of several months.

The companies outlined that the partnership will deliver research tools and workflow efficiencies that will result in improved timelines and increase patient recruitment.

Initially, the data solution tools will be available for Syneos Health sites in the US, with a focus on oncology and immunology. Later this year, the partners plan to expand this to include additional therapeutic areas and sites across the globe.

In terms of what technological solutions will be offered, Haystack is a Roivant Health portfolio company that offers Haystack Enroll as its flagship program that uses AI-assisted chart review to validate the inclusion or exclusion of certain patients from trials. The platform also uses natural language processing (NLP) analysis to generate patient lists from unstructured medical records.

According to Haystack, the core enabling technologies provided by its AI are the lossless de-identification of the unstructured medical records, including free text, scans, and images. In addition, the clinical NLP can read patient records longitudinally and assess eligibility for trials.

As a result, Haystack suggest its technology can reduce time and labor required to enroll patients, enable the central management of enrollment across studies, increase access to future study opportunities, and provide better care to patients.

“At Haystack, our goal is to reduce the screen failure rate through the use of clinical AI and pre-screening services. Screen failure is extremely costly across the clinical research chain, in terms of both time and money. For overburdened and busy research sites, additional time spent on screening patients who never actually enrol into a trial could have been much more effectively spent on patient care. Clearly, better pre-screening methodologies are urgently needed to reduce screen failure rates, enabling faster clinical development and improved patient care,” said Steve Whitehurst, CEO, Haystack Health.

The pilot efforts that the companies had begun prior to the announcement had allowed them to identify eligible patients for specific trials in ‘near-real time’, and were able to focus recruitment on patients less likely to fail trial screeners – thereby avoiding the issues mentioned by Whitehurst.