Abstract P4-08-12: Comparative analyses of the prognostic value of oncotype and mammaprint using the National Cancer Database

INTRODUCTION: Majority of the approximately 40,610 deaths from metastatic breast cancer in the US each year occur in women with hormone receptor positive breast cancer who recur after treatment for early stage disease. Genomic analysis is increasingly used to personalize breast cancer treatment for women with early breast cancer resulting in AJCC 8 modification of TNM staging. The 70-gene Mammaprint was developed using both ER- and ER+ breast tissue samples, while the 21-gene Oncotype DX (ODX) assay was developed using only ER+ breast tissue. Previous studies found that the two genomic assays gave discordant testing results. OBJECTIVE: To compare the performance of Mammaprint and Oncotype DX in assigning prognosis in early stage hormone receptor positive breast cancer. METHOD: A retrospective cohort of women diagnosed with early stage, hormone receptor positive breast cancer who received ODX or Mammaprint was established using the National Cancer Data Base (NCDB), 2010-2014. Using the propensity score matching method, we defined two groups of patients with similar clinical and demographic characteristics; one group received ODX and another received Mammaprint. The groups were matched by clinicopathologic and demographic factors. We examined the association between ODX or Mammaprint and overall survival using Log-rank test and Cox models in the two groups separately. Hazard ratios (HR) and 95% confidence intervals (CI) were calculated as strength of association. The prognostic values were evaluated using c-index (i.e. area under ROC curve). RESULTS: Of 320,276 eligible patients with breast cancer, 41.5% received ODX and 1.3% received Mammaprint testing. The use of ODX increased from 34.3% in 2010 to 45.2% in 2014, while the use of Mammaprint increased from 0.5% in 2010 to 2.0% in 2014. After propensity score matching, we identified 3319 patients who received ODX and matched to 3319 patients who received Mammaprint. Compared to patients with a low risk Mammaprint score (n=1915), patients with a high risk Mammaprint score (n=1404) had 4.53-fold increased risk of dying (95% CI 2.79-7.36). The c-index for Mammaprint was 0.683. Relative to patients with a low ODX recurrence score (n=1927), the HR for intermediate ODX score was 1.23 (95% CI 0.76-1.98) and the HR for high ODX score was 3.62 (95% CI 2.21-5.94). The c-index for ODX was 0.601. In patients with ODX testing, 28.2% received chemotherapy. In patients with Mammaprint testing, 42.4% received chemotherapy. Based on MINDACT9s modification of Adjuvant!Online, 49.2% patients were assigned to the clinical high risk group, including 22.3% to the clinical high risk/genomic low risk (C-high/G-low) subgroup and 26.9% to the C-high/G-high subgroup. The percentage of patients receiving chemotherapy with C-low/G-low, C-low/G-high, C-high/G-low, and C-high/G-high were 4.3%, 70.4%, 32.2%, and 84.9%, respectively. CONCLUSION: The findings from our preliminary study suggest that Mammaprint may achieve better separation of high risk from low risk patients. However, it is possible that having more genes in multigene assays would better capture the heterogeneity of hormone receptor positive breast cancer and guide choice of optimal systemic therapy to reduce risk of metastases. Citation Format: Ibraheem AF, Olopade O, Huo D. Comparative analyses of the prognostic value of oncotype and mammaprint using the National Cancer Database [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 P4-08-12.