Unbiased gene selection, based on patient outcomes

With the completion of the Human Genome Project in 2001, which identified approximately 25,000 genes in the human genome, rapid developments in cancer research and discovery have ocurred. On the forefront of cancer translational research, Agendia’s founders, Dr. Rene Bernards and Dr. Laura van’t Veer of the Netherlands Cancer Institute in Amsterdam sought to utilize this information to help improve the lives of those who have been affected by breast cancer.

Drs. Bernards and van’t Veer hypothesized that since breast cancer is a genetic, heterogeneous disease, gene expression should be different in aggressive breast tumors that develop recurrences following surgery from those that are less aggressive and do not recur or spread throughout the body.  To identify a novel and independent predictor of breast cancer recurrence, they utilized DNA microarray technology to assess the activity of 25,000 genes, rather than pre-selecting a few genes based on literature and known information.  They allowed tumor biology to uncover which genes were most predictive of known patient outcomes.  This “unbiased” approach helps ensure that as even as breast cancer biology evolves, the genes selected and assay approach stays relevant.  In addition to interrogating the full human genome, the researchers evaluated breast tumors from untreated patients.  Unlike other approaches, this key differentiator allows physicians to confidently know their patient’s risk of recurrence, without any treatment bias or assumptions, therefore patients do not have to “earn” their recurrence result.

[Click image to enlarge]

Using the approaches above, Drs. Bernards and van’t Veer identified a group of 70 genes that were most predictive of breast cancer recurrence. This ground-breaking work was published in the prestigious scientific journals Nature and New England Journal of Medicine in 2002. This 70-gene diagnostic test was called MammaPrint®.

Agendia’s molecular diagnostic portfolio is developed utilizing this same approach. By analyzing all genes in the human genome and using known patient outcomes, we let the cancer biology tell us which genes are most important and predictive for an assay.