Prognostic Model for De Novo and Recurrent Metastatic Breast Cancer

Abstract

PURPOSE Metastatic breast cancer (MBC) has a heterogeneous clinical course. We sought to develop a prognostic model for overall survival (OS) that incorporated contemporary tumor and clinical factors for estimating individual prognosis. METHODS We identified patients with MBC from our institution diagnosed between 1998 and 2017. We developed OS prognostic models by Cox regression using demographic, tumor, and treatment variables. We assessed model predictive accuracy and estimated annual OS probabilities. We evaluated model discrimination and prediction calibration using an external validation data set from the National Comprehensive Cancer Network. RESULTS We identified 10,655 patients. A model using age at diagnosis, race or ethnicity, hormone receptor and human epidermal growth factor receptor 2 subtype, de novo versus recurrent MBC categorized by metastasis-free interval, Karnofsky performance status, organ involvement, frontline biotherapy, frontline hormone therapy, and the interaction between variables significantly improved predictive accuracy (C-index, 0.731; 95% CI, 0.724 to 0.739) compared with a model with only hormone receptor and human epidermal growth factor receptor 2 status (C-index, 0.617; 95% CI, 0.609 to 0.626). The extended Cox regression model consisting of six independent models, for < 3, 3-14, 14-20, 20-33, 33-61, and ≥ 61 months, estimated up to 5 years of annual OS probabilities. The selected multifactor model had good discriminative ability but suboptimal calibration in the group of 2,334 National Comprehensive Cancer Network patients. A recalibration model that replaced the baseline survival function with the average of those from the training and validation data improved predictions across both data sets. CONCLUSION We have generated and validated a robust prognostic OS model for MBC. This model can be used in clinical decision making and stratification in clinical trials.

Demographic and Clinical Features of Patients with Metastatic Breast Cancer: A Retrospective Multicenter Registry Study of the Turkish Oncology Group

“…Nevertheless, the high incidence of dnMBC (43.9%) in Cohort II exceeds the previously reported ratios of de novo presentation, ranging between 28 and 30% among all MBC patients [ 16 , 17 ]. We also observed a higher proportion of de novo presentation among the entire HER2+ subgroup (48.7%), which is in line with existing data reporting that 37.5–49.8% of HER2+ MBC present with de novo disease [ 13 , 18 , 19 , 20 , 21 ].…”

“…A consistent observation over the analyzed period was the poor survival in the TN subgroup, which has been determined as an independent prognostic factor on overall mortality in our cohort, as well as many others [ 8 , 20 , 26 , 27 , 41 ].…”

Systematic Bias in Medical Algorithms: To Include or Not Include Discriminatory Demographic Information?

“…Although the authors report that they included race or ethnicity as covariates in the prognostic models, they did neither discuss possible implications of doing so nor did they evaluate how including these data may affect algorithmic fairness. 1 In this letter, we would like to discuss some of the potential pitfalls of this approach and provide a brief overview of how bias against less-privileged groups can be revealed and avoided.…”

“…We read the recent article by Barcenas et al 1 who developed a prognostic model for estimating individual prognosis for women with metastatic breast cancer with great interest. We congratulate the authors for providing a robust prognostic model for metastatic breast cancer that may better inform clinical treatment decisions.…”

Reply to A. Pfob and C. Sidey-Gibbons

“…Drs Pfob and Sidey-Gibbons expressed concerns about two sources of potential bias 1 in our proposed prognostic model for overall survival (OS) in patients with metastatic breast cancer (MBC). 2 The first is in regard to the interpretation of the prediction of OS for under-represented race or ethnic groups, specifically for Black patients. Their concern suggests that the less favorable OS found for Black patients may imply differences in the cancer treatments received and/or in responses to such treatments.…”