Background: A dichotomous index combining two gene expression assays, HOXB13:IL17BR (H:I) and molecular grade index (MGI), was developed to assess risk of recurrence in breast cancer patients. The study objective was to demonstrate the prognostic utility of the combined index in early-stage breast cancer.
Methods: In a blinded retrospective analysis of 588 ER-positive tamoxifen-treated and untreated breast cancer patients from the randomised prospective Stockholm trial, H:I and MGI were measured using real-time RT-PCR. Association with patient outcome was evaluated by Kaplan-Meier analysis and Cox proportional hazard regression. A continuous risk index was developed using Cox modelling.
Results: The dichotomous H:I+MGI was significantly associated with distant recurrence and breast cancer death. The >50% of tamoxifen-treated patients categorised as low-risk had <3% 10-year distant recurrence risk. A continuous risk model (Breast Cancer Index (BCI)) was developed with the tamoxifen-treated group and the prognostic performance tested in the untreated group was 53% of patients categorised as low risk with an 8.3% 10-year distant recurrence risk.
Conclusion: Retrospective analysis of this randomised, prospective trial cohort validated the prognostic utility of H:I+MGI and was used to develop and test a continuous risk model that enables prediction of distant recurrence risk at the patient level.