Deep 6, a Pasadena, CA-based startup, has raised $17 million in Series A funding led by Point72 Ventures along with GSR Ventures and strategic partners. A “clinical trials acceleration software” maker, Deep 6 will use the funds to meet growing demand for the company’s suite of AI-driven tools designed to accelerate and optimize clinical trials design, recruitment, and management.
Rationalizing Supply and Demand
With tens of thousands of new clinical trials registered with the FDA each year in America alone, sponsors and CROs have struggled for decades to recruit sufficient numbers of patients in a timely manner to clinically validate the efficacy of the treatment under study.
A Complex Process
It is ultimately very difficult to identify, match, and evaluate a patient against a trial’s complex recruitment criteria—the process consumes time and effort—not to mention complicated and prone to inefficiencies, requiring full-time FTEs dedicated to laboriously combing through patient records—one by one. With the explosive growth of complex, precision-based clinical trials—from gene therapy to cell therapy to immune-oncology investigational treatments, the quest for finding the right patient that can not only fit into a maze of requirements makes the sponsor, CRO and the principal investigator and coordinators jobs even more challenging.
Status Quo Challenges
These challenges manifest in the enrollment track record. The complexity-driven inefficiency leads to over 85% of trials that are delayed and many ultimately fail due to poor recruitment. And these delays and failure come at a dear cost—each day of delay potentially costs trial sponsors millions of dollars in lost revenue, not to mention the patient’s personal costs in being unable to find a possibly life-saving study as they often don’t know where to look.
Deep 6 uses AI to parse clinical trial qualification criteria and match against repositories of structured and unstructured patient health data at participating sites. Deep 6’s software then analyzes structured data, such as ICD-10 codes, and unstructured clinical data, including doctor’s notes, pathology reports, operating notes and other important medical data in free-text form that cannot be searched easily. By mobilizing natural language processing, the software can extract tens of thousands of new clinical data points—symptoms, diagnoses, treatments, genomics, lifestyle data, and more, thus transforming fragmented medical documents into unified patient graphs that contain all the information needed to match complex clinical trial criteria.
The Deep 6 value proposition can be summarized as follows: by integrating this next-generation natural language processing AI software with the electronic medical record or data warehouse of some type that contains all of the patient records, they are able to rapidly search and train the system to learn how to not only find better matches for trials but to do so in a more granular and precision manner.
Today, Cedars Sinai medical Center and Texas Medical Center are among several other large health systems across the U.S. that utilize the technology. Moreover, they touted in their recent press release that they have secured large numbers of life science companies, medical device companies and contract research organizations.
Founded in 2015 and based in Pasadena, CA, Deep 6 disrupts the clinical trial enrollment process by transforming the way researchers identify eligible patients. They position that they find more, better-matching patients for trials in minutes, not months. They employ between 40 and 50 team members and have raised $17 million. The founders include Brian Dolan, Doug Cassidy, Eric Gildenhuys, Paul Kasinski, and Wout Brusseelaers.
It would appear that Deep 6 is a potentially powerful tool that would need to integrate into underlying systems containing patient records. As a third-party provider processing health data, including patient data, their cloud-based service may fall under the HIPAA and HITECH Acts in terms of compliance requirements. Moreover, integration projects with electronic health record systems in major academic medical centers are not always simple and straightforward. Any project to do so requires a major business case, extensive leadership buy-in, and the right team for implementation.