Case Study: Assess MPS II Diagnosis and Treatment Referral Pathways and Identify Potential Patients via Predictive Modeling

The Situation

Mucopolysaccharidosis type II (MPS II), or Hunter syndrome, is a rare disease caused by a deficiency in the lysosomal enzyme iduronate-2-sulfatase (I2S). This enzyme is crucial for the degradation of glycosaminoglycans (GAGs). Early diagnosis and symptoms management are critical to prevent irreversible neurological and organ damage and improve patient outcomes.

A pharmaceutical company developing a new therapy for MPS II sought to understand the journey of current patients to drive referrals to specific regional treatment centers where its therapy would be offered. The goal was to identify both the experts providing the initial diagnosis and/or treatment and the adept providers referring patients to these experts, facilitating the adoption of its therapy.

The Solution

Identifying MPS II patients is challenging due to the inconsistent use of specific ICD-10 codes and the variability of symptoms among affected individuals. Accurately singling out potential MPS II patients using extensive medical claims data is also difficult. Additionally, mapping the pathways leading patients to the company’s treatment centers is complex due to the often-incomplete information in claims data, requiring both technical and clinical expertise.

EVERSANA first identified the correct patients’ cohort treated with ELAPRASE® specifically indicated for MPS II and then found patients in the database who share the same features with the Elaprase-treated patients but had not received Elaprase.

In addition, the EVERSANA team identified the distinct phases of the patient’s journey. Historical data from open claims was analyzed to map referrals to key events along the treatment pathway – from first clinical suspicion to initial diagnosis to the first treatment – to algorithmically identify the most influential providers. EVERSANA created a machine learning model to identify key predictors of this rare disease and deployed it across millions of patients in the claims database, identifying strong candidates for genetic screening.

The Results

The results of this analysis revealed that median time to treatment for patients retained by the same provider from first diagnosis to first treatment was three months faster than those referred out. EVERSANA identified top providers and high-value treatment centers for retaining patients for diagnosis and treatment across six distinct regions, providing the client with top targets for its novel therapy.

The top features predicted by the model aligned with those reported in Human Phenotype Ontology, with metabolic disease literature and with results expected by expert clinicians. The model found an additional 11,000 patients who were recommended for confirmatory diagnosis with genetic testing and had not previously been diagnosed or treated with Elaprase.