260P - M-bioscore: a practical tool for predicting outcomes among patients with previously untreated metastatic breast cancer.

Date 11 September 2017
Event ESMO 2017 Congress
Session Poster display session
Topics Breast Cancer, Metastatic
Breast Cancer
Presenter Omar Abdel-Rahman
Citation Annals of Oncology (2017) 28 (suppl_5): v74-v108. 10.1093/annonc/mdx365
Authors O. Abdel-Rahman
  • Clinical Oncology, Ain Shams University Faculty of Medicine, 11566 - Cairo/EG

Abstract

Background

Two prognostic models “bioscore” and “Neo-bioscore” were recently published and validated to help predict the outcomes of patients with non-metastatic breast cancer treated with either upfront surgery or upfront neoadjuvant chemotherapy. A comparable model for metastatic disease is yet to be developed. The current study thus sought to propose and validate a third model “M-bioscore” to help predict the outcomes of treatment-naïve patients with metastatic breast cancer.

Methods

Through SEER*Stat program, surveillance, epidemiology and end results (SEER) database (2010-2013) was accessed. The resulting cohort was equally split into two halves: training set (to guide model development) and validation set (to test the model prediction). Multivariate analysis for the candidate prognostic factors (extent of metastases, estrogen receptor (ER), progesterone receptor (PR), HER2 neu and nuclear grade) was conducted through a Cox proportional model. M-bioscore was then calculated for each patient. Cancer-specific survival analyses according to M-bioscore were conducted through Kaplan-Meier analysis/log-rank testing.

Results

A total of 6655 patients with previously untreated metastatic breast cancer and complete data were identified in the period from 2010-2013. The following factors were associated with better cancer-specific survival in multivariate analysis in the training set (isolated distant nodal metastases, ER positivity, PR positivity, HER 2 neu positivity and lower nuclear grade) (P 

Conclusions

M-bioscore is a novel, easy and reliable tool for predicting the outcomes of patients with previously untreated metastatic breast cancer. Further external validation within the context of other population-based cohorts is recommended.

Clinical trial identification

Legal entity responsible for the study

Omar Abdel-Rahman

Funding

None

Disclosure

All authors have declared no conflicts of interest.