Saving your clinical trial in a fortnight using Phesi’s AI platform

We are often approached by clients who are struggling with troubled clinical trials, most often due to slow site enrollment. When this happens, it’s common for trial sponsors and CROs to think the issue is caused by a pool of under-performing investigator sites. Such thinking naturally leads to a presumption of needing more and better sites to save the trial. But often this presumption is wrong.

To illustrate, the following chart shows site activation curves from two comparable trials. Each trial enrolled a similar number of patients each month and activated roughly the same number of sites by Month 9. The difference between the trials is that the sites in Trial A were activated faster and therefore had been enrolling for much longer on average: the average enrollment time for Trial A was 5.4 months (58% x 9 months), versus 2.7 months (29% x 9 months) for Trial B. Trial A therefore enrolled twice as many patients as Trial B, which is problematic for the sponsor of the latter trial, who was considering opening more sites at roughly the same rate – potentially increasing costs, reducing focus, and likely limiting enrollment further.

img

The above case exemplifies how we can use Phesi’s vast database and AI-driven analytics capabilities to predict, diagnose, and mitigate common issues that plague troubled trials, thereby getting trials back on track within just a few weeks.

Sometimes, we are asked to execute a trial that we know will be operationally difficult by virtue of its protocol design. Here, Phesi’s AI platform can be massively helpful in the planning or execution stages of a trial. For example, we recently analyzed data [over just a couple of days] from 5,415 patients with non-small cell lung cancer (NSCLC) who were enrolled in 35 trials with similar protocol designs. The analysis took just a couple of days. Notably, we found that only 13% of patients had an ECOG performance status score of 2 (PS 2). NSCLC trials that included only ECOG PS 2 patients had a median enrollment cycle time of 1,445 days, compared to 618 days for “regular” NSCLC trials (i.e., those with looser PS inclusion criteria). Therefore, if we designed a trial that only included ECOG PS 2 patients, we know it would be very difficult to execute on the grounds of patient availability alone.

img

Clinical trials with difficult designs can, with our data-driven AI services, still be successfully executed. Phesi helps teams communicate the difficulties associated with protocol design, site and country selection, and site activation among all stakeholders, quantitatively and objectively. We allow clients to see potential problems before they become actual problems, and to know what issues to manage; a much-needed ”targeted treatment solution” for trial design and execution.

Phesi can conduct a holistic and integrated analysis of any trial in the design or execution stage, and can apply those insights to provide a multi-faceted, actionable recommendation to get a troubled trial back on track in weeks. The issues we find are typically small, but significant and are interconnected. But we are there every step of the way, and once a trial is fixed, the sponsor can rest easier knowing the clinical development program will finish on time and often under budget.

Phesi: AI in clinical development delivered!

Privacy Preferences
When you visit our website, it may store information through your browser from specific services, usually in form of cookies. Here you can change your privacy preferences. Please note that blocking some types of cookies may impact your experience on our website and the services we offer.