Patient-derived xenograft (PDX) models and transplantable genetically engineered mouse models faithfully recapitulate human disease and are preferred models for basic and translational breast cancer research. Here, a method is described to orthotopically transplant breast tumor fragments into the mammary fat pad to study tumor biology and evaluate drug response.
Preclinical models that faithfully recapitulate tumor heterogeneity and therapeutic response are critical for translational breast cancer research. Immortalized cell lines are easy to grow and genetically modify to study molecular mechanisms, yet the selective pressure from cell culture often leads to genetic and epigenetic alterations over time. Patient-derived xenograft (PDX) models faithfully recapitulate the heterogeneity and drug response of human breast tumors. PDX models exhibit a relatively short latency after orthotopic transplantation that facilitates the investigation of breast tumor biology and drug response. The transplantable genetically engineered mouse models allow the study of breast tumor immunity. The current protocol describes the method to orthotopically transplant breast tumor fragments into the mammary fat pad followed by drug treatments. These preclinical models provide valuable approaches to investigate breast tumor biology, drug response, biomarker discovery and mechanisms of drug resistance.
Most breast cancer deaths can be ascribed to recurrent disease that is resistant to conventional therapies1,2. The inter- and intra-tumor heterogeneity of breast cancers contribute to therapy resistance. Moreover, tumor heterogeneity can impinge on accurate prognosis and challenge disease management3,4. Identification of predictive biomarkers of response will significantly improve clinical outcomes of patients with breast cancer. Even though most breast cancer types are immunologically 'cold' tumors that are likely unresponsive to immunotherapy, immune checkpoint inhibitors have shown promise in clinical trials2,5. For example, a phase III trial showed improved disease-free survival (DFS) and preliminary evidence that atezolizumab (monoclonal antibody against PD-L1) combined with nab-paclitaxel may provide an overall survival benefit as compared with nab-paclitaxel alone in tumors with ≥1% PD-L1 staining6. Development of therapies that sensitize breast tumors to immunotherapy will revolutionize treatment regimens.
Preclinical models that faithfully recapitulate human breast cancer heterogeneity and drug response are critical to study tumor biology and identify potential biomarkers for targeted therapy. Immortalized cell lines are widely used for breast cancer research since these cell lines are easy to grow and genetically modify to study molecular mechanisms. However, due to the selective pressure from long term cell culture in vitro, genetic drift may occur over time and breast cancer cell lines may carry cell line-specific genomic alterations that are distinct from aberrations in primary breast tumors7,8,9.
Patient-derived xenograft (PDX) tumor chunks are able to recapitulate the heterogeneity of human disease, and are histologically and immunohistochemically similar to the tumor of origin10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29. Importantly, PDX models are phenotypically stable across multiple transplantations as evidenced by histology, transcriptome, proteome and genomic analysis10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29. PDX models show treatment responses comparable to those observed clinically10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29. PDX models for estrogen receptor positive (ER+), progesterone receptor positive (PR+), epidermal growth factor 2 positive (ERBB2+, HER2+) and triple negative breast cancer (TNBC) PDX models have been established, and provide an excellent platform to test endocrine-, chemo- and targeted therapies. However, one main caveat of PDX models at present is the lack of a functional immune system in the mouse.
The genetically engineered mouse models (GEMM), such as Trp53 homozygous null, cMyc, Wnt1, PyMT, or Her2 overexpression models, allow the study of spontaneous tumor initiation, progression and metastasis in the context of an intact immune system. However, the tumor latency is long, which makes it difficult to conduct preclinical trials with multiple arms30,31. However, GEMM can be transplanted to syngeneic hosts to generate sufficient numbers of tumors that closely recapitulate human tumors32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55. For example, the mammary epithelium from a p53-null BALB/c mouse was transplanted into the cleared fat pads of syngeneic wild-type recipient mice to form primary tumors, which can be further transplanted into syngeneic hosts56,57. The p53-null tumors recapitulated different subtypes of human tumors.
The combination of PDX models and transplantable GEMM provides valuable preclinical tools to investigate breast tumor biology, drug response and anti-tumor immunity. In the current protocol, a method of orthotopic transplantation of PDX and GEMM tumor fragments into the mouse mammary fat pad is described. These models are amenable for serial passages and usually retain a stable phenotype. To mitigate the risk of genetic drift or loss of heterogeneity across passages over time, multiple tissue fragments are cryopreserved at each passage for subsequent transplantation in the event that biological or morphological changes are observed over time29,58.
All protocols using animals have been reviewed and approved by the Institutional Animal Care and Use Committee (IACUC). The tumor fragments, around 1−2 mm3 in size, are from viably frozen stock obtained from the Patient-Derived Xenograft and Advanced In Vivo Models Core at Baylor College of Medicine.
1. Preparation of cryopreserved mammary tumor fragments for transplantation
2. Collection and preparation of fresh mammary tumor for transplantation
3. Prepare animal for surgery
4. Transplantation of tumor fragments into the fourth (inguinal) mammary fat pad
5. Monitoring tumor growth in response to drug treatment
NOTE: Palpable tumors of established transplantable PDXs and p53-null tumors takes between 2 weeks and 8 weeks to develop after surgery, depending on intrinsic tumor growth rates.
Figure 1 shows the equipment (Figure 1A) and key procedures (Figure 1B) of orthotopic transplantation. Figure 2 shows characterization of a transplanted PDX tumor (MC1). Tumor fragments (1 mm3) of MC1 model were transplanted into the #4 fat pad of SCID/Beige mice. One month later, the average tumor size reached around 350 mm3. Tumor volume was monitored twice a week for one month. Normally we obtain palpable tumors for various PDX or GEMM in around 2 weeks to 8 weeks with 1 mm3 tumor fragments transplanted (Figure 2A). Tumor samples can be collected for morphology and signaling analysis (Figure 2B−D). H&E staining was performed to analyze the pathology (Figure 2B). IHC was used to monitor markers for specific cell lineage (keratin 19 (K19), epithelial cell, Figure 2C), cell status (Ki67, proliferation, Figure 2D) or signaling molecule of interest.
Figure 1: Schematic showing the surgery technique. (A) Surgical equipment required for the orthotopic transplantation. (B) Representative image showing the exposure of mammary fat pad for tumor trunk transplantation. Please click here to view a larger version of this figure.
Figure 2: Characterization of the transplanted tumors. (A) Representative kinetics of tumor growth measured by a caliper. Tumor volume (mm3) = W2 x L/2 (W = width and L = length). (B) H&E staining showing pathology of MC1 PDX. IHC showing epithelial marker keratin 19 (C) and proliferation maker Ki67 (D) in MC1 PDX. Scale bar = 20 µm, magnification = 40x. Please click here to view a larger version of this figure.
To reduce variations in tumor growth across animals, it is critical to cut the tumor tissue into 1 mm3 fragments for transplantation. Models that grow soft tissue are harder to work with and the tumor fragments need to be cut slightly larger (1−2 mm3). When placing the tissue into the mammary fat pad pocket take care not to split the tissue into multiple pieces as this will result in multiple small tumors or oddly shaped tumors.
In addition, use fresh tumor for transplanting animals that will be used for drug treatment studies. Implanting tissue from cryopreservation will yield a more variable take rate and slightly slower growth kinetics. Once tumors grow from the cryopreserved material, the second transplant generation will yield the typical take rate and growth kinetics for that model. Moreover, try to use tumors with no or mild necrosis for transplantation. For most models this will be a size range of 400−600 mm3. If an obvious necrotic core is observed, use tissue from the outer layer of the tumor for transplantation and do not use the necrotic areas. It is important to keep the tumor tissue on ice and to protect from drying.
To reduce variability among tumor chunks from GEMM that may have been derived from the periphery or tumor core with different microenvironments, an alternative method is to prepare a primary cell suspension and transplant approximately 5,000−30,000 cells depending upon the tumor model into the mammary fat pad. The limited collagenase digestion is carried out with 1 mg/mL type I collagenase in DMEM/F12 without any additives for 2 h at 37 °C rotating at 125 rpm. Mammary tumor cells can be enriched by 3 short centrifugation steps. Briefly, transfer the cell suspension to a 15 mL conical tube and centrifuge at 450 x g for 7 s. Aspirate the supernatant and resuspend the pellet in 10 mL of 1x Dulbecco's phosphate buffered saline (DPBS). Repeat the pulse centrifugation for two more times. This will help randomize differences between chunks.
Transplantation of normal mammary epithelium will not regenerate a morphologically normal and functional ductal tree in the presence of endogenous mouse epithelium. It is necessary to remove the endogenous mouse epithelium (clearing) for the normal epithelial transplant to grow62. However, neoplastic tissue is able to grow even in the presence of intact endogenous mouse epithelium. Yet this does not necessarily mean no such inhibitory signals exist. Mammary fat pad clearing is necessary for certain experimental protocols to prohibit the interaction of endogenous mouse mammary epithelium with the engrafted material. In addition, the endogenous epithelium may complicate some downstream analysis such as genome, transcriptome and proteome analysis.
The PDX model and transplantable GEMM can faithfully recapitulate the heterogeneity of clinical subtypes and the response to drug therapy of human breast cancer. Importantly, these models are easy to transplant and maintain a stable phenotype during a limited number of serial passages. Tumor growth can be easily measured with calipers. One caveat of the PDX model and transplantable GEMM is that these models do not recapitulate early steps of tumor initiation. Also, PDX models lack the interaction of the tumor with a functional immune system. These preclinical models provide a valuable system to study breast cancer biology and evaluate drug response. Combining drug response with the genomic and proteomic information for each tumor model will facilitate the identification of biomarkers for response prediction and treatment resistance mechanisms. These types of data may lead to novel targeted therapies that could be used alone and in combination with chemotherapy or immunotherapy to improve patient outcomes.
The authors have nothing to disclose.
This work was supported by the National Institutes of Health (R37CA228304 and R01HL146642 to Xi Chen, CA148761 to Jeffrey M. Rosen), US Department of Defense (W81XWH-19-1-0524 to Xi Chen, W81XWH-19-1-0035 to Xiangdong Lv), American Cancer Society (RSG-18-181-01-TBE to Xi Chen) and Cancer Prevention and Research Institute of Texas (RR150009 CPRIT Scholar in Cancer Research Award to Xi Chen), the Patient-Derived Xenograft and Advanced In Vivo Models Core at Baylor College of Medicine (funding from RP170691 CPRIT Core Facility Award and NCI-CA125123 P30 Cancer Center Support Grant).
1 mg/mL Buprenorphine-SR | ZooPharm (via BCM veterinarians) | Sterile | |
26G syringe | BD | 148232E | Sterile |
Betadine Scrub | Fisher | 19-027132 | |
Cotton Swabs | VWR International Laboratory | 89031-272 | Sterile |
DMEM | Fisher | MT 10-013-CM | Sterile |
Electric shaver | Oster | 78005-050 | |
Glass beads sterilizer (Germinator) | Roboz Surgical Store | DS-401 | |
Lubricant ophthalmic ointment | Akorn Animal Health | 17478-062-35 | |
Micro Dissecting Forceps; Serrated, Angular (regular forceps) | Roboz Surgical Store | RS-5139 | Sterile |
Micro Dissecting Spring Scissors (fat pad cutter) | Roboz Surgical Store | RS-5658BT | Sterile |
Micro Forceps (tissue placing forceps) | Roboz Surgical Store | RS-5069 | Sterile |
Petri Dish | Fisher | 08-757- 100D | Sterile |
Sterile drape | Sai Infusion Technology | PSS-SD1 | Sterile |
Surgery scissors | Roboz Surgical Store | RS-5960 | Sterile |
Tissue Forceps (claw forceps) | Roboz Surgical Store | RS-5158 | Sterile |
Wound clip applier | BD Autoclip Wound System | 01-804 | Sterile |
Wound clip remover | BD Autoclip Wound System | 01-804-15 | Sterile |
Wound clips | BD Autoclip Wound System | 01-804-5 | Sterile |