A model that estimates Oncotype DX recurrence score (RS) accurately predicted high- and low-risk RS categories in most early breast cancer cases in a retrospective analysis involving nearly 1,600 individuals.

Investigators looked at ODX RS and pathology reports (n=1,113) from 5 sites over an 8-year period ending in 2013. Ordering patterns at 1 site were analyzed using a model that considered treatment guidelines, immunohistochemistry markers, and Oncotype DX. The findings were validated in a separate group (n=472). Among the results:

• RS distribution was similar at all sites and correlated with clinical practice experience.
• Histopathologic markers alone determined risk category in more than half of cases.
• Applying the model at 1 site showed that the frequency of testing would not have changed overall.
• There would have been less testing of estimated clinically high- or low-risk cases, and more testing of intermediate-risk cases.
• The model correctly predicted risk category in slightly more than half of the validation group.