Case Studies

Mining Clinical Trial Data

Learn How a Biopharmaceutical Company Used Frontier® to Mine Clinical Data

Goal

Use CrushTM to develop and test new treatments for a variety of illnesses that cause severe pain, discomfort and even disability in millions of people worldwide.

Business

Amarillo Biosciences, Inc. (ABI), a leading biopharmaceutical company

Industry

Pharmaceutical Research

Challenge

ABI develops and tests possible new treatments for a variety of illnesses, focusing on treatments that reduce symptoms without causing debilitating side effects. ABI is a pioneer in the research of low-dose orally-administered interferon alpha. Interferon alpha occurs naturally in the human body and was being explored as a treatment for a variety of illnesses including fibromyalgia syndrome (FMS). High-dose injectable interferon alpha is FDA approved for treatment of certain diseases, but often comes at a high cost and with severe side effects. ABI conducted Phase II clinical trials of interferon alpha for treatment of FMS. When faced with analyzing the results of these trials, ABI turned to Parabon for help.

Solution

A deep, statistical analysis of the clinical trial data using Parabon's Crush, run on the Frontier grid platform, was necessary. The ABI researchers examined 69 billion possible models using Crush. Traditional statistical techniques that sacrifice thoroughness for speed, such as step-wise regression, did not detect the same response differences that Crush found. Crush painstakingly searched all possible combinations of clinical factors for explanatory models that were statistically significant, and very significant and useful results were discovered. This analysis, which was completed in three days on Frontier, would have required more than two years to complete on a single PC.

Results

At the time of this trial, there was no FDA approved therapy for FMS, so most patients in the United States are treated only for their symptoms with pain and anti-depression medication. Patients characteristically suffer from pain, stiffness and depression, often severe enough to be disabling. Through this in-depth statistical analysis of the Phase II clinical trial data, ABI discovered an important difference between the responses of patients who had an injury (a car accident, for example) that contributed to the development of FMS and those patients with no history of prior injury. They determined that, with patients whose FMS was not attributed to injury, the debilitating side effects involved in the treatment of FMS would be deterred if treatment was switched from high-dose interferon alpha to much lower doses, administered in ways that closely mimic the way it occurs naturally. Patients who did not report a past injury as a causative factor in their illness experienced significantly greater improvement in their stiffness with interferon alpha treatment. Roughly one-third of FMS patients had an injury as a causative factor in their history. The remaining 3-5 million FMS patients in the US today should benefit significantly from this new treatment.