Abstract Background Most breast cancers (~85%) are of no special histologic subtype (NST), and the most common special subtype is invasive lobular cancer (ILC). ILC accounts for 10-15% of all breast cancers, and there will be ~40,000 new cases in 2022 in the US alone. If considered an “independent” cancer type, ILC is the 6th most common cancer in women. The pathognomonic feature of ILC is loss of E-cadherin (CDH1). The resulting lack of adherens junctions causes the unique single-file growth pattern of discohesive ILC cells, which decreases the ability for detection by mammography, in turn resulting in late detection and hence larger tumors. Although ILCs show better prognostic factors than NST, patients with ILC have worse long-term outcome, which is not well understood. Additionally, ILC has historically been understudied, which is in part due to lack of appropriate research models. For example, the Cancer Cell Line Encyclopedia (CCLE) contains 54 NST cell lines but only 2 ILC cell lines, and only a limited number of patient-derived xenograft (PDX) models are evident in the published literature. There is a critical need for additional in vitro and in vivo models to study ILC biology, as well as to test targeted therapies. ILC PDX and patient derived xenograft organoids (PDXO) are particularly valuable tools to enable target validation and assess drug treatment response. Methods To identify and validate new ILC PDX models, we used Champions Oncology’s Lumin Bioinformatics to screen Champions’ collection of breast cancer models (n=126) for PDX harboring CDH1 mutation and/or low E-cadherin expression. We performed histological analysis on selected PDX models including H&E staining, and immunohistochemical assessment of E-cadherin, P120, estrogen receptor, progesterone receptor, HER2 and Ki67. Models with 2+ HER2 staining were assessed by FISH. All staining was interpreted by a certified breast pathologist. PDX tumor tissue was further used to develop PDXO models. Results Using Champions Oncology’s Lumin tool, we identified 10 putative ILC PDX models based upon CDH1 mutation, low E-cadherin mRNA expression, or clinical annotation of ILC (Table 1). Of the 10 PDX models analyzed, two cases were clinically annotated as ILC, while the remainder were classified as ‘carcinoma’ (n=5) or as NST (n=3). Histologic analysis revealed loss of E-cadherin and cytoplasmic P120 (lobular pattern) in 8/10 models assessed, and pathologic assessment confirmed these as having a lobular histology. IHC analysis classified these PDXs as 5 TNBC and 5 ER+ tumors, with none showing amplification of HER2 by FISH. All tumors demonstrated high (35%, n=1) or very high (>55%, n=9) Ki67 proliferation marker levels. We further developed PDXOs from one PDX model as proof of concept, and the resulting organoid demonstrated classic ‘grape-like’ ILC morphology. Conclusion Our study demonstrates how existing PDX banks with in-depth multi-omic and pathology analyses can be interrogated to identify models of unique histological and molecular subtypes of breast cancer. Of the PDX models selected from Champions Oncology’s breast cancer cohort, 7 models were classified as ILC either through re-classification from NST/carcinoma to ILC or confirmation of ILC histology. In addition, 1 PDX was re-classified as mixed type. Some of these models are ER+/HER- and thus have the classic molecular features of ILC. Our collaborative omics guided approach allows for reclassification of PDX models to increase available research models for unique breast cancer subtypes such as ILC which in turn will enhance translational research in unique histological subtypes of breast cancer. Citation Format: Jennifer M. Atkinson, Megan Yates, Daniel D. Brown, Jagmohan Hooda, Rohit Bhargava, Paolo Schiavini, Marianna Zipeto, Steffi Oesterreich, ADRIAN V. LEE. Combining multiomics and histological assessment to identify patient derived xenograft models of invasive lobular carcinoma [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P3-08-10.
Read full abstract
Mar 13, 2023
S Benbachir + 4
Open Access
