Send to printer »

GEN News Highlights : Sep 15, 2010

CompanDx Launches Biomarker Panel for Predicting Time to Breast Cancer Metastasis

Firm claims 31 gene signature can predict prospective Kaplan Meier survival curve.

U.K.-based biomarker discovery services firm Compandia has changed its name to CompanDx and launched a 31 gene breast cancer metastasis prediction panel it calls Time to an Event. Developed using the company’s Risk Distiller bioinformatics tool, the Time to an Event biomarker panel can accurately predict time to metastasis following initial surgery and biopsy, CompanDx claims. The firm says it is now applying for funding to further validate the panel in the clinic. It already has collaborations in place with leading clinical centers, and industry support has been secured with a major diagnostics company.

Time to Event represents “a world first of a totally new type of diagnostic paradigm with potential to significantly alter clinical management,” claims Andy Sutton, CompanDx CEO. The 31 gene panel was derived using three public datasets from different clinics. A fourth dataset was used for validation. The firm says comparison with these existing datasets confirmed that across patient cohorts the actual versus signature-predicted time to metastases had a correlation of 0.86.

Moreover, CompanDx claims, the panel allows prediction of a prospective Kaplan Meier curve that is capable of accurately tracking the actual survival curve. Using the combined clinical data, the gene panel predicted a median time to event of 3.7 years, whereas the actual time calculated was 3.5 years, and there were no significant differences between the actual and CompanDx curves, the firm maintains.

CompanDx’ name change reflects its strategy of building an in-house pipeline of biomarker-based products and services that it will commercialize alongside the biomarker discovery services already offered to the pharma, biopharma, and diagnostic industries. Current service offerings span patient stratification in the form of molecular classification of disease subtypes, predictive biomarker signatures for drug response, and prognostic biomarker panels for prediction of outcome during clinical trials. The new in-house products will exploit CompanDx expertise in molecular biology, protein chemistry, and mass spectrometry, together with its  Biomarker Distiller, Pathways Distiller, and Risk Distiller platforms for interrogating complex datasets to identify predictive and/or prognostic signatures.