Assessment of the prognostic value of metastatic free survival as overall survival surrogate in breast cancer female patients
Fagr M. Aboumadawy1 , Ahmed Gaber Gewil2, Inas Mohamed Elbadry1, Rasha el sakka2, Nahla Ahmed Gamal Eldin3
1. Cancer Management and Research Department, Medical Research Institute, Alexandria University, 2. Clinical Oncology Department, Faculty of Medicine, Alexandria University, 3. Community Medicine Department, Faculty of Medicine, Alexandria University
Various surrogate endpoints for overall survival (OS) have been recently used by FDA for drug approval in miscellaneous cancer types. This aims to facilitate the establishment of clinical trials, propose beneficial treatment options after shorter duration of assessment and provides new clinical prognostic tools. Metastasis-Free Survival (MFS) has been proven to be a Strong Surrogate for Overall Survival in Localized Prostate Cancer. However, this surrogacy has not been studied in breast cancer or any other solid tumor.
Evaluate the surrogacy correlation between Metastatic Free Interval (MFI) and OS in breast cancer patients, the relationship between MFI and clinicopathological tumor characteristics and to compare MFI in breast cancer subtypes.
Data was retrieved in a retrospective approach from December 2009 to 2019 to identify patients with initial diagnosis of stage I-III breast cancer who developed subsequent distant metastasis. MFS was measured from the date of diagnosis of initial breast cancer to the date of first evidence of distant metastases confirmed by imaging or histologic evidence. MFI is categorized into three categories (short:<3years, intermediate:3-5 years, long: >5years).
Retrospective data from 332 patients were collected from 2009 to 2019. 56% of patients (186 patients) had a short MFI of less than 3 years while 19.9% of patients (66 patients) had a long MFI of more than 5 years and 24.1% of patients (80 patients) had intermediate MFI of 3 to 5 years. The 10 years overall survival rate was 46.5%, 54.8%, and 79.1% for patients with short MFI, intermediate MFI and long MFI respectively (P-value : <0.001). MFI was strongly correlated to OS with a significant Pearson coefficient 0.855 (p value <0.001). The correlation was strongest for Long MFI with a significant pearson coefficient 0.429 (P value <0.001). Among 332 patients with known disease subtypes, Shorter MFI was associated with the triple-negative subtype and HER2 enriched subtype while longer MFI was associated with the hormone receptor-positive, HER2 negative subtype. The results of the multivariate analysis of MFS showed that after adjusting for confounders by Cox regression, Age, Tumor size (pT), Lymph Node Status (pN), Tumor subtype and Administration of Targeted Therapy were independent predictors for MFS.
MFI is strongly correlated to OS in breast cancer in addition to its relation to the tumor clinicopathological characteristics, which provides the potential validity of its usage as a surrogate endpoint for OS in clinical trials assessing novel adjuvant therapies in treatment paradigm.