Tech News

Clinical trials are better, faster, cheaper with big data

“One of the most difficult parts of my job is getting patients into studies,” says Nicholas Bories, a senior physician at Lawrence Celsion, New Jersey, who develops liver and ovarian cancer and next-generation chemotherapy and immunotherapy agents of various kinds. brain tumors. Borys estimates that less than 10% of cancer patients are registered in clinical trials. “If we could get that up to 20% or 30%, we could probably have several cancers now conquered.”

Clinical trials test new drugs, devices, and procedures to determine if they are safe and effective before they are approved for general use. But the path from study design to homologation is long, winding, and expensive. Today, researchers use artificial intelligence and advanced data analysis to speed up the process, reduce costs, and make it easier for those who need effective treatment. And they’re using a little-used but fast-growing resource: data on patients from past trials

Building external controls

Clinical trials typically involve at least two groups or “arms”: a test or experimental arm that receives the treatment being investigated and a control arm that does not. A control arm may receive no treatment, placebo, or current standard of care for the disease being treated, depending on the type of treatment being studied and the study protocol. It’s easy to see the recruitment problem of researchers studying therapies for cancer and other deadly diseases: life-threatening patients now need help. While they may be willing to take a risk with a new treatment, “the last thing they want is to be randomly in control of an arm,” Bories says. Combine this reluctance with the need to recruit patients with relatively rare diseases — for example, a type of breast cancer with a specific genetic marker — and enough time to hire people for months or even years. Nine out of 10 clinical trials worldwide – not only for cancer but for all types of conditions – cannot hire enough people within the target deadlines. Some rehearsals fail due to a lack of participants.

What if the researchers did not have to hire a control team and could offer experimental treatment to all those who participated in the study? Celsion is exploring such an approach with its New York-based Medidata, which provides management software and data for more than half of the world’s clinical trials, offering it to leading pharmaceutical and medical device companies and academic medical centers. Acquired by French software company Dassault Systèmes in 2019, Medidata has amassed a huge “big data” resource: detailed information on more than 23,000 trials and nearly 7 million patients delayed by about 10 years.

The idea is to reuse patient data in past trials to create “external control weapons”. These groups perform the same function as conventional control arms, but can be used in environments where it is difficult to hire a control group: for very rare diseases, for example, or for situations such as cancer, which are life-threatening. They can also be used effectively for “one-arm” trials where a control group is not practiced: for example, to measure the effectiveness of an established device or surgical procedure. Perhaps the most valuable immediate use is to perform rapid preliminary trials to assess whether a treatment is worthwhile until a full clinical trial is completed.

Medidata uses artificial intelligence to source its database and create a proprietary version of an external control weapon that found patients who had acted as controls in pre-treatment trials under certain conditions. “We can carefully select these historical patients and link the current experimental arm with historical test data,” says Arnaub Chatterjee, vice president of products, Acorn AI Measures. (Acorn AI is a division of Medidata data and analytics.) Trials and patients are consistent with research goals — so-called endpoints, such as reduced mortality or how long patients are without cancer — and other aspects of the disease. study designs, such as the type of data collected at the beginning and end of the study.

In creating an external control arm, “we do everything we can to mimic an ideal randomized controlled trial,” says Ruthie Davik, vice president of data science, at Acorn AI Measures. The first step in researching the database is to look for possible candidates in the control arms using the main eligibility criteria of the study: for example, the type of cancer, the key characteristics of the disease and the level of progress and the patient’s first time. they are being treated. It’s basically the same process used to select control patients in a standard clinical trial, except for data recorded at the beginning of a past trial, rather than the current one, which Davi says is used to determine feasibility. “We are finding historical patients who would qualify for trial if they existed today.”

Download full report.

This content was created by Insights, a custom content from the MIT Technology Review. It was not written in the editorial board of the MIT Technology Review.

Source link

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button