Student Research Year Highlight: Ryan Rilinger

The Cleveland Clinic Lerner of College of Medicine of Case Western Reserve University (CCLCM) is a five-year program dedicated to developing the next generation of physician investigators. During their first year, students are assigned dedicated physician and research advisors to help ensure they reach their educational goals. In their fourth year, students work with a mentor to develop a master’s-level thesis in basic science, translational medicine, clinical medicine or health systems. When the students graduate, they each receive an MD with Special Qualification in Biomedical Research from Case Western Reserve University.

The research in which students are involved is critical and timely, and often results in presentations at national conferences and publication in scholarly journals. The example below briefly describes a student’s research project involving a risk prediction model for earlier diagnosis of leptomeningeal metastasis:

"For my research year, I am completing a clinical research project, ‘Machine learning for prediction of leptomeningeal metastasis,’ with Andrew Dhawan, MD, DPhil, within the Rose Ella Burkhardt Brain Tumor and Neuro-Oncology Center at Cleveland Clinic. Leptomeningeal metastasis (LMM), the spread of any cancer type into the cerebrospinal fluid (CSF), carries a very poor prognosis and is difficult to diagnose. As a result, many patients do not receive targeted treatment for LMM until late in their disease course. If LMM could be diagnosed earlier, treatment may be more effective, with improved clinical outcomes. My project sought to build a robust LMM risk prediction model to facilitate earlier diagnosis.

“I began this project with a large database of patients undergoing CSF analysis, first working to identify clinical and demographic variables associated with subsequent development of LMM. I then used several machine learning modeling techniques to combine these variables, using deidentified codes for each patient, into a single prediction model with higher sensitivity and specificity for LMM diagnosis than current diagnostic methods (imaging and CSF analysis). By using a multi-stage design within the overall model, we are now able to both predict likelihood of current disease presence and identify patients at highest risk for subsequent LMM development.

“In addition to outperforming current diagnostic approaches, our model will be one of the first prediction tools for LMM to require only routinely available demographic and clinical variables rather than specialty laboratory tests, making it accessible to neurologists and oncologists in a variety of practice settings. We envision this model will enable clinicians to identify high-risk patients earlier in their disease course — or even prior to onset of LMM — thereby enabling prompt treatment and extending survival."

– Ryan Rilinger (’27)