Diversity index as a novel prognostic factor in breast cancer

Abstract

Background: Intratumoral heterogeneity is an important factor leading to tumor progression and therapeutic resistance. However, application of intratumoral heterogeneity as a prognostic or predictive marker remains limited due to difficulties associated with its assessment. To investigate the value of Shannon diversity index in measuring genetic heterogeneity using gene copy number variations, we performed fluorescence in situ hybridization (FISH) for c-MYC and correlated the Shannon index for c-MYC copy number variation with clinicopathologic features and patient survival in breast cancer. Methods: Using 283 cases of invasive breast cancer, we performed FISH for c-MYC and calculated the Shannon index using its copy number variation. Then we took 308 cases of invasive breast cancer from a different institution and performed the same procedure for validation. To confirm our results using another gene, we performed FGFR1 FISH in the test set for comparative analysis. For comparison of primary and metastatic breast cancers, 40 patients with invasive breast cancer who subsequently developed recurrence or metastasis were included. Results: Shannon index for c-MYC copy number variation correlated with average c-MYC copy number and was higher in tumors with c-MYC amplification and those with c-MYC genetic or regional heterogeneity. High Shannon index for c-MYC copy number variation was associated with adverse histologic features including high histologic grade, lymphovascular invasion, p53 overexpression, high Ki-67 proliferation index, and negative hormone receptor status. High Shannon index was associated with decreased disease-free survival in the whole group, in the subgroup excluding c-MYC amplified cases, and in hormone receptor-positive subgroup in the test set and validation set. We investigated the prognostic value of Shannon index using another gene, FGFR1 and observed that high Shannon index for FGFR1 gene copy number variation was an independent adverse prognostic factor. Comparing primary and metastatic breast cancers, we observed a biological association between high Shannon indices in both primary and metastatic tumors and decreased overall survival. Conclusion: This study revealed significant associations between high diversity index and adverse pathologic parameters as well as decreased survival in breast cancer suggesting that Shannon index can be used as a biomarker for tumor progression in breast cancer patients.