Faster, better, cheaper... how often have we heard these words, in the context of any process along the long path of drug development? There are a myriad of solutions that can help at different stages, enabling more comprehensive target assessment, more rapid lead optimization, and so on. One of the most expensive parts of the drug development process is clinical trials, with bottlenecks including access to knowledge for site selection, patient populations, principal investigators and key opinion leaders.
Researchers naturally look to utilize information from current and past trials but manually extracting the relevant information can be resource-intensive, repetitive and, therefore, prone to errors. Time is money, so reducing costs and errors is critical.
One of our customers, Merck, use Linguamatics I2E for text analytics over public domain clinical trial data, to improve clinical trial site selection.
One example of the benefits of text analytics is a site selection project for Merck Experimental Medicine division (EMS). They needed to locate a clinical trial site that would be able to conduct gastric bypass trials with the ability to measure gut peptides before and after surgery. The ideal trial site needed to fit many different characteristics - over a dozen - which would be hugely time-consuming to find using the public domain search interface to ClinicalTrials.gov.
Using I2E to index ClinicalTrials.gov enabled Merck to run specific queries both to extract “gastric bypass” terms from intervention fields, as well as biomarker, gut peptide and other key metrics from unstructured Clinical Trial record text. In order to get some idea of scientific rank for the investigators, the author and/or the trial ID were searched in Medline and the results integrated into the clinical trial results output. This process enabled them to create a structured summary table for analysis and ranking, facilitating systematic and rapid review of the possible sites.
Merck said, Data collected from this search yielded three ideal sites for this trial, one of which was previously not known to the Experimental Medicine group.
Site selection characteristics for Merck Experimental Medicine division, showing the depth and variety of precise needs to find an appropriate clinical trial site.
