Nov. 30, 2022 – It is anticipated that the application of artificial intelligence will hasten the completion of clinical trials and reduce their costs while simultaneously improving their overall effectiveness. The creation of “synthetic control arms,” which make use of data to construct “simulants” or computer-generated “patients” in a trial, is one component of this overall strategy.
Researchers are able to speed up the process of participant recruitment and reduce the number of real people they need to enrol.
According to the opinions of the experts, patients and pharmaceutical corporations both stand to benefit. An advantage for people, for instance, is that the standard-of-care or placebo treatment is given to the simulants. This means that everyone who participates in the study ultimately receives the experimental treatment. Artificial intelligence and machine learning are two methods that can help pharmaceutical companies determine which of their potential medicine candidates offer the most promise.
“Machine learning has primarily been effective so far at optimising efficiency. This does not mean that it has been effective at developing a better medicine; rather, it has been effective at optimising the efficiency of screening. According to Angeli Moeller, PhD, head of data and integrations creating insights at drugmaker Roche in Berlin and vice chair of the Alliance for Artificial Intelligence in Healthcare board, “AI exploits the learnings from the past to make drug research more productive and more efficient.”
“I’ll give you an example. It’s possible that you have a thousand different small compounds in front of you, and you want to determine which one of those molecules will bind to a receptor that’s associated with a disease. You won’t have to go through the trouble of screening thousands of applicants if you use AI. Perhaps you could screen only one hundred people,” she suggests.
Participants in a “Synthetic” Clinical Trial
The first clinical trials to use data-created matches for patients rather than control individuals matched for age, sex, or other attributes have already begun. These clinical trials will be conducted in place of traditional clinical trials. For instance, Imunon Inc., a biotechnology company that develops next-generation chemotherapy and immunotherapy, used a synthetic control arm in its phase 1B trial of an agent added to pre-surgical chemotherapy for ovarian cancer. This trial was for the purpose of determining whether or not the addition of the agent improved patient outcomes.
The researchers concluded that it would be beneficial to continue assessing the new medication in a phase 2 trial after seeing the results of this preliminary study.
According to Sastry Chilukuri, co-CEO of Medidata, the business that donated the data for the Phase 1B trial as well as the founder and president of Acorn AI, using a synthetic control arm is “very cool.”
He says, “What we have here is the first FDA and EMA approval of a synthetic control arm. This means that you’re replacing the entire control arm by using synthetic control patients, and these are patients that you pull out of historic clinical trial data.” “What we have here is the first FDA and EMA approval of a synthetic control arm.”
A Tide of AI-Enhanced Research on the Horizon?
It is anticipated that the use of AI will become increasingly prevalent in research. To this day, the majority of research into AI-driven medication development has been on neurology and oncology. According to a news and analysis piece published in the magazine Nature in the month of March 2022, the start in these disciplines is “probably attributable to the substantial unmet medical need and several well-characterized targets.”
It was hypothesised that this application of AI is merely the beginning of “a coming wave.”
According to a review paper that was published in the journal Nature Medicine in September, “There is a rising interest in the employment of synthetic control approaches,” which can be defined as the use of external data to construct controls.
According to the report, the Food and Drug Administration (FDA) had already granted approval to a medicine in 2017 for a form of Batten disease, a rare juvenile neurologic illness, on the basis of a research with historical control “participants.”
According to Chilukuri, one area of oncology where the use of a synthetic control arm could potentially make a difference is in glioblastoma research. Patients generally drop out of trials because they want the experimental treatment and don’t want to remain in the standard-of-care control group, he says. This brain cancer is incredibly difficult to cure, and patients typically drop out of trials because of this. Additionally, “given the life expectancy, it is very difficult to complete a trial.” [Citation needed]
Chilukuri believes that the use of a synthetic control arm could hasten the pace of research and increase the likelihood of successfully completing a glioblastoma study. And the participants in the experiment are the ones who receive treatment.
Still Early Days
In research, AI could potentially help reduce the number of “non-responders.”
Clinical trials are “really difficult, they’re time-consuming, and they’re extremely expensive,” according to Naheed Kurji, president and CEO of Cyclica Inc, a data-driven drug discovery company based in Toronto. Kurji also chairs the board of the Alliance for Artificial Intelligence in Healthcare and serves as chair of the board.
“Companies are working very hard to discover more efficient ways to integrate AI to clinical trials so that they can acquire findings faster at a lower cost while also improving the quality of those outcomes.”
There are many clinical trials that are unsuccessful, and it’s not because the molecule being tested is ineffective… however since the patients who participated in the experiment included a significant number of individuals who did not respond to the treatment. They simply remove the response data, explains Kurji.
Chilukuri adds, “You’ve probably heard a lot of people talk about how we are going to make more progress in the coming decade than we did in the previous century.” Chilukuri is referring to predictions that have been made by a number of different people. “And this is simply due to the availability of high-resolution data that enables you to grasp what is happening on an individual level,” the author says.
He forecasts that as a result, an explosion will occur in the field of precision medicine.
In some respects, artificial intelligence in clinical research is still in its infancy. “There is a lot more work to be done,” adds Kurji, “but I think you can point to a lot of examples and a lot of firms that have made some really huge steps.”
