Abstract P4-04-05: Spatial heterogeneity revealed by genomic profiles comparing 60 matched primary breast tumors and metastatic lymph nodes

Abstract

Abstract Background: Genomic profiling in primary breast tumors has been extensively performed and reveals the heterogenous complexity in mutational landscape of the disease. However, little is known about differential mutation spectrums in synchronous breast tumors and matched metastasis in axillary lymph nodes. To characterize the spatial heterogeneity and identify potential biological mechanisms involved in the lymph node metastasis, we determined the genomic mutational profiles in synchronous primary lesion (PL) and metastatic lymph nodes (MLN) by using ultra-deep targeted sequencing. Methods: Targeted deep sequencing was performed using a panel including 520 cancer-related genes and spanning 1.6MB of human genome. The mutational profiles were compared between the matched PL and MLN samples from 60 of treatment-naïve patients with invasive breast cancer and axillary lymph node metastasis. KEGG enrichment analysis was further performed between specific and shared mutations in either PL or MLN samples. The relationship between spatial heterogeneity and clinical characteristics was also explored. Results: The cohort had a median age of 45 (ranging from 28-67), with a majority (86.7%) of them diagnosed with infiltrating ductal carcinoma. In this 60-paired cohort, 961 genomic aberrations were identified in 242 genes, including 405 single nucleotide variants (SNVs), 66 insertions or deletions (INDELs), 482 copy-number amplifications (CNAs), and 8 translocations. Although 584 (60.8%) events were shared in PL and MLN samples, 226 (23.5%) mutations spanning 51 genes and 151 (15.7%) mutations involving 46 genes were specific PL and MLN samples, respectively. In addition, 7 of patients (11.7%), all of whom are hormonal receptor (HR) positive, harbored completely similar mutation spectrum between PL and MLN. Twenty-one patients (35%) had PL specific mutations, but had no MLN specific mutations. In contrast, 6 patients (10%) only had MLN specific mutations and the remaining 26 patients (43.3%) had both PL and MLN specific mutations. Interestingly, the Ki67-based proliferation index in 7 of patients with completely similar mutation spectrum were significantly lower than other patients with differential mutational landscapes between PL and MLN (p=0.019). Furthermore, KEGG pathway enrichment analysis revealed that deregulation in PI3K/AKT and Ras signaling pathway were enriched in both PL and MLN samples. More importantly, we found that aberrant activation of Proteoglycans pathway and JAK-STAT signaling, as well as aberrant HIF-1 pathway, was specifically occurred in the MLN samples, suggesting crucial roles for these signaling pathways in the involvement of lymph node metastasis. Conclusions: Our study revealed genomic heterogeneity between primary tumors and lymph nodes, and identified mutations as well as pathways that are potentially relevant to lymph node metastasis, emphasizing that such spatial heterogeneity may contribute to the evolution of the disease and result in differential responses of subsequent treatments. This study was supported by funding from National Natural Science Foundation of China (Grant No. 81602645) and Guangdong Provincial Natural Science Foundation (Grant No. 2016A030313768). Citation Format: Liao N, Zhang G, Wang Y, Guo L, Cao L, Ren C, Wen L, Li K, Jia M, Chuai S, Chen X. Spatial heterogeneity revealed by genomic profiles comparing 60 matched primary breast tumors and metastatic lymph nodes [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 P4-04-05.