Phenotypic and functional characteristics of patient derived murine xenograft models

e14694

Background: Generation of patient-derived models for breast cancer has been difficult. In other histologies, success was higher in refractory disease. With the goal to develop a protean resource for testing novel compounds and combinations for treatment refractory breast cancer, we establish patient-derived murine xenograft models (PDX) from tumors clinically unresponsive to at least anthracyclin, platin and taxane. Here we report phenotypic and drug testing data of our established breast cancer models. Methods: On immunodeficient NOG mice, PDX were engrafted from breast cancer tissue samples. The tissues were obtained from patients with disease progression after chemotherapy with three to four drugs. The established PDX were characterized by immunohistochemistry and the PDX and primary tumor are currently undergoing exome and transcriptome sequencing. The PDX were tested for response to all standard chemotherapy agents. The drug testing included platin, taxane, anthracycline, 5-FU, Everolimus, Eribulin and depending on the subtype tamoxifen, olaparib, and a CDK4/6 Inhibitor. Results: Since May 2017 30 breast cancer samples have been processed under rapid and stringent conditions. Currently five models, three TNBC and two hormone receptor positive PDX, have been fully engrafted, phenotypically compared and drug tested. The immunohistochemistry (estrogen/progesterone/androgen/Her2 receptor, Ki-67, CK5/6) of original tumor and PDX were correlative. All patients had shown clinical resistance to platin, anthracyclines, and taxanes in the neoadjuvant or palliative setting. Concordant with the clinical resistance, the PDX models showed only limited or transient sensitivity to single agents. Six more PDX, three TNBC, two HR+ and one Her2+, are currently in various states of engraftment and testing. Conclusions: With our stringent approach, we successfully generated PDX models in 15%, which may rise to 30% with the ongoing models. Phenotype between patient tumors and PDX was consistent. The minor differences in responsiveness to chemotherapy may be due to differences in stromal factors. In summary, the PDX of refractory tumors are a versatile resource for preclinical studies of novel treatment approaches.

Abstract P6-03-06: Androgen supplementation in patient derived xenografts in androgen receptor positive breast cancer to increase engraftment and growth rate

Background:

The European fund for regional development (EFRE) supported Precision Oncology and Personalized Therapy Prediction (POP) Project is establishing preclinical models to further the development of personalized therapy options. In the subgroup breast cancer the current goal is to increase the growth and engraftment rates of breast cancer patient derived xenografts (PDX) models.

Methods:

Breast cancer patients of the Department of Gynecology with Breast Center Charité Universitätsmedizin Berlin, Germany are recruited since May 2017. In total 29 tissue samples were collected and included so far.

Treatment naive and treatment refractory patients, tripple negative breast cancer (TNBC), hormone receptor positive (HR+) and Her2 postivie tumors, primary disease, recurrence or metastasis are sampled. Fresh tumor tissue is extracted via surgery or biopsy. The materials are then implanted into female immunodeficient NOG mice. To establish PDX models for HR+ breast cancer the mice received estrogen supplementation.

To increase engraftment and growth rates androgen receptor (AR) testing and subsequently androgen replacement was started since April 2018.

Up to date, 6 new samples have been collected. One HR+ and two TNBC samples tested also positive for AR. These samples are currently in passage 0 (p0) and are now supplied with androgens to increase engraftment and growth rate. One already established AR+ TNBC PDX is being regrown with androgen supplementation to compare growth rates.

Results:

Out of the initial 23 tissue samples ten (six HR+ and four TNBC) have been able to be engrafted into PDX mice.

The TNBC PDX models are one in p1, one in p2, one in p3 and one is being tested with systemic therapy. Engraftment time in p1 were between 19 and 97 days. Growth time to passagable size between 21 and 112 days.

The HR+ PDX models are four in p1 and two in p2. Engraftment time in p1 was between 26 and 123 days. Growth time to passagable size has been achieved in 2 HR+ PDX within 17 to 48 days.

The engraftment/growth rates and times of the androgen supplemented PDX models will be presented.

Conclusion:

Breast cancer growths in humans slowly and this is also the case in the PDX models. To achieve faster growth and higher engraftment rates androgen supplementation in AR+ breast cancer might be an additional enhancive factor.

Citation Format: Kiver VII, Wulf-Goldenberg A, Jurmeister PS, Schweiger C, Gorea O, Hoffmann J, Denkert C, Keilholz U, Liedtke C, Blohmer J-U. Androgen supplementation in patient derived xenografts in androgen receptor positive breast cancer to increase engraftment and growth rate [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 P6-03-06.

Abstract 5013: Rapid generation of phenomic and functional profiles of patient-derived 3D cell culture models for identification of treatment vulnerabilities of breast cancer: Early results of the EFRE-PoP project

Targeted treatment for breast cancer subsets currently relies on the occurrence of estrogen, progesterone and Her2/neu receptors. For triple negative breast cancer (TNBC) there is no identified targeted therapy. The mutational landscape for breast cancer subsets has been characterized, but drug development has been limited due to the lack of appropriate preclinical models. Development of patient-derived xenograft models (PDX) has been difficult with take rates of below 30%.To establish a series of patient-derived 3D (PD3D) cell culture models as a versatile resource for ex vivo drug sensitivity screens as well as secondary establishment of PDX. We obtained breast cancer specimens either by biopsy or surgical resection. Upon arrival tumor tissue was minced and enzymatically digested. After subsequent filtering, respective tissue size fractions were seeded as Matrix-droplets into 24-well plates and incubated under standard conditions. Growth of PD3Ds was monitored daily by microscopy. PD3Ds were splitted when their diameter reached appropriate size. Organoids were than fixed and embedded. FFPE sections of donor-tissues and derived PD3D models were than used for IHC and inspected by a pathologist. In parallel, mutational profiling of snap-frozen tumor tissue and cell cultures was performed. All models were subjected to a semi-automated multi-drug response assay in a 384-well format to assess individual compound sensitivities.We established a scalable workflow for culturing and screening of PD3D cell cultures from limited quantities of breast cancer tissue from true-cut and vacuum biopsies as well as surgical specimen. Tissue fragments of ca. 3mm in diameter were sufficient to successfully establish PD3D cell cultures at a high yield. The time needed to expand PD3D cell cultures to obtain a sufficient amount of cells for subsequent molecular analyses varies from 2-6 weeks depending on amount and quality of samples as well as grade and stage of the donor tumor tissue. At the time, we can report a rate of culture establishment rate of 75% (11 models from 15 patient samples) for various clinical relevant subgroups, including TNBC, HR+, Her 2 pos. BC and one DCIS sample. The immunohistology and mutational profiles of these samples are currently being confirmed. The established workflow for culturing PD3D cell cultures offers high yield rates within a potentially clinically relevant time-frame for HTP cytotoxicity screenings. We are encouraged that this method is suitable for various molecular subtypes of breast cancer. Once adequately validated in co-clinical trials, PD3D models would make for an intriguing tool in supporting clinical decision-making for individual patients. Particularly patients with breast cancer refractory to standard of care compounds could benefit from this approach.

Citation Format: Verena I. Kiver, Guido Gambara, Olga Gorea, Jens-Uwe Blohmer, Philipp Jurmeister, Carsten Denkert, Alessandra Silvestri, Caroline M. Schweiger, Maxine Silvestrov, Ulrich Keilholz, Christian R. Regenbrecht. Rapid generation of phenomic and functional profiles of patient-derived 3D cell culture models for identification of treatment vulnerabilities of breast cancer: Early results of the EFRE-PoP project [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 5013.