Abstract P3-08-23: Prosigna assay for treatment decisions in early breast cancer: A single center, decision impact study

Background: Therapeutic decisions in early breast cancer (EBC) are based on clinical and pathological features, which are subject to intra- and inter-observer variability. Hence, in the era of precision medicine, there is growing need for predictive biomarkers. The Prosigna assay utilizes Prediction Analysis of Microarray, a test based on the analysis of 50 intrinsic subtype-linked gene clusters. This single center decision impact study aimed to evaluate the effect of Prosigna test results on physicians' adjuvant treatment choices.

Methods: Between September 2017 and February 2018, FFPE tumor samples from 53 newly diagnosed, postmenopausal, hormone receptor-positive, HER2-negative EBC (T1-T2; pN0-N1a) patients were analyzed. Pre-test clinical judgments and Prosigna test results were compared.

Results: Mean age was 59 (42-77). Invasive ductal carcinoma (79.2%), grade 2 (52.8%) and T1c-N0 tumors (43.4%) represented the majority. Before the Prosigna test, 65.4% of the patients were classified as luminal A and 34.6% as luminal B. Of the pre-test risk groups, 40.4% were low-risk, 40.4% were intermediate risk and 19.2% were high risk. Prosigna assay grouped 50% of patients as luminal A, 44.2% as luminal B, 3.8% as basal type and 1.9% as HER2-expressing. Post-test ROR score-based groups were distributed as 25% low-risk, 40.4% intermediate risk and 34.6% high risk. There was a statistically significant correlation between clinically defined and molecularly assessed intrinsic BC subtypes (kappa:0.334, p=0.007). Similarly, pre-test and post-test recurrence risk groups were correlated (kappa:0.397, p=0.001). Before the Prosigna test, endocrine treatment was physicians' primary choice in 20 patients (39.2%), chemotherapy was recommended to 31 patients (60.8%). Overall, the Prosigna assay led to a change in choice of treatment for one patient (2%). There was 40.4% discordance between pre- and post-test recurrence risk groups. In addition, intrinsic subtypes were 34.6% discordant, which is largely driven by the reclassification of pre-test luminal A tumors into Prosigna luminal B group.

Table 1.Impact of Prosigna Results on Final Treatment Decision Prosigna low risk N=13 (%24.5)Prosigna intermediate risk N=21 (%39.6)Prosigna high risk N=18 (%34)Total N=51 (%100)Treatment choice before Prosigna    CT+HT0121729 (56.9)HT only127120 (39.2)CT offered, not accepted by the patient0202 (3.9)Treatment choice after Prosigna    CT+HT0121830 (58.8)HT only127019 (37.3)CT offered, not accepted by the patient0202 (3.9)Change in treatment choice    HT to CT001100 (0)CT to HT0000 (0)CT: Chemotherapy HT: Hormonotherapy

Conclusions: Although conventional risk assessment methods are relatively inexpensive with shorter turnaround times, their accuracy for risk assessment and value for risk reduction are suboptimal. According to our results, Prosigna assay was found as a relevant tool for clinical decision-making process. In cases where there is a discrepancy between the clinical assessment results and the Prosigna assay, tumor boards may guide treatment recommendations. Long term follow-up of these patients will elucidate the potential benefits of using multigene molecular tests as biomarkers for EBC treatment.

Citation Format: Esin E, Oksuzoglu BO, Markoc F, Bilgetekin I, Yildiz F, Guntekin S, Yukruk F, Atalay R. Prosigna assay for treatment decisions in early breast cancer: A single center, decision impact study [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P3-08-23.