Abstract 2498: miFISH single cell analysis as a potential tool to stratify the progression risk in patients with Atypical Ductal Hyperplasia (ADH)

Abstract

Abstract We previously studied synchronous ductal carcinomas in situ (DCIS) and invasive ductal carcinomas (IDC) using our novel approach of multiplex interphase Fluorescence in situ Hybridization (miFISH) that allows simultaneous enumeration of copy numbers of up to 20 gene loci per single cell, providing new insights into tumor clonality and intratumor heterogeneity (ITH). A high degree of chromosomal instability was present in DCIS, and we frequently observed a direct clonal evolution from DCIS to IDC. We now ask whether this degree of instability is also present in even earlier potential precursor lesions like atypical ductal hyperplasias (ADH). We are therefore conducting a retrospective study analyzing ADH lesions from patients without synchronous DCIS or IDC (pure ADH) and from patients with ADH adjacent to higher-grade lesions. Using our miFISH assay, we simultaneously assess copy number changes of six oncogenes (COX2 (1q), PIK3CA (3q), MET (7q), MYC (8q), CCND1 (11q), HER2 (17q), ZNF217 (20q)) and six tumor suppressor genes (FHIT (3p), DBC2 (8p), RB1 (13q), CDH1 (16q), TP53 (17q), NF2 (22q), typical for breast tumorigenesis. So far, we have analyzed fifteen pure ADH and five ADH with synchronous DCIS or IDC. The miFISH analysis showed that the majority of all pure ADH displayed a stable diploid genome without any copy number changes observed for the genes tested while only 3/15 (20%) showed changes, namely (i) a COX2 and HER2 gain in a diploid genome, (ii) a CDH1and MET loss in a tetraploid genome and (iii) a COX2 gain in a diploid genome. In the group of ADH with adjacent higher-grade lesions (n=5), all DCIS and IDC lesions analyzed displayed breast cancer specific copy number changes corroborating our previous results. Of note, two of the ADH lesions adjacent to these aberrant DCIS or IDC lesions had normal diploid genomes. Of the remaining three synchronous ADH lesions, one showed a complex gain and loss pattern similar to its adjacent DCIS, the second one was characterized by a simple COX2 gain which was maintained in the adjacent DCIS in addition to a DBC2 loss and MYC gain. The third ADH displayed a tetraploid clone without any gains or losses, while the adjacent IDC revealed a complex gain and loss pattern in its tetraploid genome. Our preliminary data indicate that the majority of pure ADH does not harbor copy number changes. However, there is a small subset of pure ADH revealing gain and loss patterns typical for DCIS and IDC often involving COX2 gain indicating it as an early event. Of note, ADH lesions adjacent to DCIS or IDC tend to display similar changes as their adjacent higher-grade lesions suggesting them as direct precursor lesions. We are currently analyzing more ADH samples and plan to add material from an ADH patient cohort with long-term follow-up to explore whether miFISH can stratify ADH patients into different risk groups. Citation Format: Kiara Whitaker, Jausheng Tzeng, Daniela Hirsch, Irianna Torres, Steven Brower, Gert Auer, Miguel Sanchez, Thomas Ried, Kerstin Heselmeyer-Haddad. miFISH single cell analysis as a potential tool to stratify the progression risk in patients with Atypical Ductal Hyperplasia (ADH) [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 2498.