The Importance of Using an Untreated Patient Population for Assay Development

Recent studies have shown that approximately 49% of patients do not comply with their hormone therapy, reinforcing the importance of unbiased assay development and validation on untreated patient cohorts.

Developing and validating an assay on an untreated population provides information reflective of tumor biology rather than response to a given therapy. Therefore, the test results are valid regardless of treatment selection as the assay is based on biology-driven outcomes, unperturbed by treatment.  Consider a population of breast cancer patients consisting of a mixture of “low” and “high” risk patients represented by green and red figures, respectively. Using an unbiased gene profile, those patients are categorized by their true risk of recurrence, regardless of treatment selection and compliance. (Figure 1)

Conversely, if an assay is developed and validated on a treated breast cancer patient population, one assumes the benefit and risk reduction associated with a therapeutic intervention, such as Tamoxifen (or aromatase inhibitor) treatment for 5 years. Thus, the clinician must question what the patient’s true risk is, if they are not compliant with their hormonal therapy.  The patient classified as “low risk” that does not fully comply with hormonal therapy, may actually be a “high risk” patient. Profiling patients with an assay validated on a treated population introduces a bias making therapy compliance a prerequisite to the outcome predicted by the gene expression profile. (Figure 2)

MammaPrint: Developed and Validated on an Untreated Patient Population

As reported in Nature, MammaPrint was developed on 10 year outcome data from an untreated breast cancer patient population, thereby ensuring the validity of the results regardless of the ultimate treatment regimen selected.   In the independent TRANSBIG, multi-center study on 302 untreated patients published in the Journal of the National Cancer Institute, MammaPrint demonstrated it could accurately identify patients at low risk of recurrence or at high risk of recurrence. This ability to identify an untreated patient’s true risk of developing distant metastases  is critical in terms of reliably stratifying a woman into a low risk or high risk of recurrence group, independant of treatment.