Background: Previous studies have pointed out discrepancies among multiparameter tests at the individual patient level, often attributing these to methodological limitations in in silico analyses. It was anticipated that direct comparisons of diagnostic assays would demonstrate strong agreement. The OPTIMA prelim study aimed to evaluate the consistency in risk stratification and subtype classification across various multiparameter tests when applied to the same group of patients.
Methods: The study included 313 women diagnosed with early breast cancer who were randomly assigned to either standard treatment (chemotherapy and endocrine therapy) or treatment guided by test results (chemotherapy only if the Oncotype DX recurrence score was greater than 25). Risk stratification was performed using Oncotype DX, Prosigna (PAM50), MammaPrint, MammaTyper, NexCourse Breast (IHC4-AQUA), and conventional IHC4 (IHC4). Subtype classification was determined by Blueprint, MammaTyper, and Prosigna.
Results: Oncotype DX classified a larger proportion of tumors as low risk (82.1%, 95% CI = 77.8% to 86.4%) compared to the low/intermediate risk classifications by Prosigna (65.5%, 95% CI = 60.1% to 70.9%), IHC4 (72.0%, 95% CI = 66.5% to 77.5%), MammaPrint (61.4%, 95% CI = 55.9% to 66.9%), and NexCourse Breast (61.6%, 95% CI = 55.8% to 67.4%). Notably, the agreement between the five tests was only moderate when results were categorized into high versus low/intermediate risk. Only 119 tumors (39.4%) received a consistent risk classification (either low/intermediate or high) across all tests, while 183 tumors (60.6%) were assigned to different risk categories by different tests. However, a subset of 94 tumors (31.1%) showed agreement among four out of the five tests. In terms of subtype classification, all three tests (Blueprint, MammaTyper, and Prosigna) categorized between 59.5% and 62.4% of tumors as luminal A subtype. However, only 121 tumors (40.1%) were consistently classified as luminal A by all three tests, and a mere 58 tumors (19.2%) were uniformly classified as nonluminal A. Discordant subtyping was observed in 123 tumors (40.7%).
Conclusions: Current evidence comparing the prognostic value of different multiparameter tests suggests that these tests offer broadly comparable risk information for populations of women with estrogen receptor (ER)-positive breast cancer. Despite this general agreement at the population level, individual patient risk categorization and subtype information can vary significantly depending on the test used.
