HUS: Technology combines suitable patients and drug research

Help for analyzing the increasing amount of data

Perttu Koskenvesa is part of a team led by Kimmo Porkka, which is developing a clinical Trial Matching application. The development project automates a previously manual process in which suitable medicine trials are sought for patients with a blood disease, and patients who meet the research criteria are sought for ongoing clinical trials.

“We use the data stored in patient information systems in a way that benefits the treatment and eliminates laborious steps in manual data processing and aggregation,” says Koskenvesa.

The amount of patient-specific data has increased significantly over the last 10–15 years. The term acute leukemia includes a number of different forms of the disease, and the treatment of each can vary greatly. The molecular genetic differences of the disease and the drugs available for them shape the treatment path.

Combining ongoing research with suitable patients requires information on current research about the number of research participants and who can participate. The data is also intended to assess different studies that can be carried out, in other words, whether suitable patients can be found.

"Utilizing all the information stored in the patient information system is almost impossible if the analysis of the data depends only on human activity," says Koskenvesa.

Precise guidance for clinical trials

The Trial Matching application, which supports the research of acute leukemia and speeds up diagnostics, utilizes data from HUS's data pool, i.e. Finland's largest healthcare data repository.

In Trial Matching, patient data needed to diagnose leukemia and plan treatment are extracted from the HUS data pool and collected as a separate data repository in accordance with the research permit. Patients are retrieved from the data mass for clinical trials using predefined study criteria.

The tool's key features are the creation of patient cohorts and the development of reports to support research activities. The clinician makes the final decision on patients to be included in the study.

Trial Matching can also be used to find out the best medication proposal for a patient from the options already available. The application can take into account all patient-specific findings and possible differences in treatment planning.

"Prescribing drug therapy to patients with a molecule abnormality that predicts a good response in cancer cells. In this way, targeting can be used to get more effective treatment for them. Using the same medicine for a person who doesn't have this change is practically useless and would not produce anything other than side effects," says Koskenvesa.

Inclusion and exclusion criteria sieve patients and studies

The development process of Trial Matching is based on the clinical research routines of Helsinki University Hospital. So far, the research criteria have arrived as PDF files for professionals who make the decisions concerning treatments, and combining data on patients with the criteria has been manual.

During the construction phase of the application, the criteria for clinical trials are stored in a structured format, which allows for their efficient processing, the development of artificial intelligence, and the real-time updating of the criteria.

"Data is obtained from research plans, i.e. protocols that define inclusion and exclusion criteria. In the future, we will be able to respond better to preliminary surveys that assess the possibility of participating in planned research," says Koskenvesa.

The aim is to have a shared information model in Finland, which also facilitates national research cooperation and information sharing. The data is also converted to an international standard format (OMOP data model), making the solution scalable at the national and global levels.

The first phase of the development project will be completed at the end of 2021.

Discover more about our data and analytics solutions!