Methotrexate (MTX) Polyglutamates Test | Early RA Prediction Algorithm

Early Rheumatoid Arthritis (RA) Prediction Algorithm - An important new tool for determining whether patients presenting with early arthritis symptoms are likely to develop RA

The Early RA Prediction Algorithm is a personalized medicine laboratory service that we anticipate launching in late 2008, which can be used to help determine the likelihood of patients developing RA when they present with some signs and symptoms of arthritis, but fail to meet ACR diagnostic criteria for RA.

The proprietary algorithm produces a prediction score using both laboratory and clinical data. The laboratory measures consist of C-reactive protein levels (CRP), rheumatoid factor (RF), and the presence of antibodies to cyclic citrullinated peptide antibodies (anti-CCP). The clinical measures include patient age, gender, distribution of joints involved, morning stiffness, and number of tender and swollen joints. The algorithm produces a prediction score ranging between 0 and 14; a higher score indicates a greater risk of developing RA. If the Early RA score is ≤6.0, then the patient is unlikely to develop RA (negative predictive value of 91%); if the score is ≥8.0, then the patient has a high likelihood of developing RA (positive predictive value of 84%). The predictive accuracy of the algorithm is excellent, with AUC values for the prediction rule, the prediction model after cross validation, and the external validation cohort of 0.89, 0.87, and 0.97, respectively.

The exceptional discriminant ability and overall accuracy of the Early RA Prediction Algorithm enable the treating physician to determine an appropriate course of care sooner in the disease process, based on the individual patient’s likelihood of progressing to RA, with the goal of preventing long term disability and joint damage.

Market Background – Undifferentiated Arthritis and Rheumatoid Arthritis

The term “arthritis” encompasses more than 100 rheumatologic disorders and autoimmune diseases. Cypress’ initial personalized medicine laboratory services relate to the diagnosis, prognosis and treatment monitoring of patients with RA.

A diagnosis of RA is typically based on a patient’s medical history, physical exam, lab tests and X-rays. Blood tests, such as anti-CCP, RF, erythrocyte sedimentation rate (ESR) and CRP, are used to help diagnose RA, but thought leaders agree that none of these markers is independently conclusive. For example, only about 60-70% of all people with RA have an elevated rheumatoid factor, and not all people with an elevated rheumatoid factor have RA.2 Further, rheumatoid factor testing can produce false negative results if performed early in the course of disease.

Diagnosis is particularly complicated in early, nonspecific rheumatic disease, because early RA has symptoms similar to other forms of arthritis. In a study of patients who present with recent-onset arthritis, about 37% were categorized as having Undifferentiated Arthritis (UA).2 UA is typically defined as any arthritis that has the potential for a persistent course, without fulfilling the classification criteria for a specific rheumatic disorder.

While 40-50% of UA patients have symptoms that will spontaneously resolve, about 33% develop RA within a year of appearance of symptoms. Early and aggressive treatment of RA with disease-modifying anti-rheumatic drugs (DMARDs) can prevent long term disability and joint damage in patients. A recent study also showed that early treatment of UA with methotrexate slows progression to RA as well as progression of joint damage. There is considerable benefit to identifying patients with UA who are likely to develop RA, so that where appropriate, disease modifying treatments can be provided, rather than just palliative care, and unnecessary exposure to potential adverse events from improper therapies can be avoided.

Clinical Validation

The Early RA Prediction Algorithm was derived from a cohort of more than 1900 patients with recent-onset arthritis, 570 of whom had an arthritis that could not be classified and were documented as undifferentiated arthritis. The algorithm has been further validated in the Netherlands, Germany and United Kingdom, in a total of 343 patients.

Process for Testing

The laboratory testing will be performed in our CLIA certified laboratory with a serum sample, and the ordering physician provides the clinical measures via a secure web portal or on a written test requisition, at the preference of the physician. An Early RA Prediction Algorithm score is calculated and results are sent to the physician electronically or by fax.

Intellectual Property

Cypress Bioscience holds a worldwide, exclusive license to the Early RA Prediction Algorithm technology, which includes US, European and PCT patent applications.