Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2014 Sep 30;5(18):8750-64.
doi: 10.18632/oncotarget.2373.

Using Heterogeneity of the Patient-Derived Xenograft Model to Identify the Chemoresistant Population in Ovarian Cancer

Affiliations
Free PMC article

Using Heterogeneity of the Patient-Derived Xenograft Model to Identify the Chemoresistant Population in Ovarian Cancer

Zachary C Dobbin et al. Oncotarget. .
Free PMC article

Abstract

A cornerstone of preclinical cancer research has been the use of clonal cell lines. However, this resource has underperformed in its ability to effectively identify novel therapeutics and evaluate the heterogeneity in a patient's tumor. The patient-derived xenograft (PDX) model retains the heterogeneity of patient tumors, allowing a means to not only examine efficacy of a therapy, but also basic tenets of cancer biology in response to treatment. Herein we describe the development and characterization of an ovarian-PDX model in order to study the development of chemoresistance. We demonstrate that PDX tumors are not simply composed of tumor-initiating cells, but recapitulate the original tumor's heterogeneity, oncogene expression profiles, and clinical response to chemotherapy. Combined carboplatin/paclitaxel treatment of PDX tumors enriches the cancer stem cell populations, but persistent tumors are not entirely composed of these populations. RNA-Seq analysis of six pair of treated PDX tumors compared to untreated tumors demonstrates a consistently contrasting genetic profile after therapy, suggesting similar, but few, pathways are mediating chemoresistance. Pathways and genes identified by this methodology represent novel approaches to targeting the chemoresistant population in ovarian cancer.

Conflict of interest statement

Conflict of Interest Statement

The authors have no conflict of interests to disclose.

Figures

Figure 1
Figure 1. Take rates of different sites for implantation and maintenance PDX histology
(A) Tumors were implanted subcutaneously (SQ), in the mammary fat pad (MFP), intraperitoneal (IP), or sub-renal capsule. The success of implantation was similar comparing SQ to MFP, however more PDX lines were established from SQ implant due to number of implants. IP and SRC implants are not effective for establishing a PDX line. (B)Representative pictures of implanted tumors at either SQ, MFP, IP, or SRC. (C) After implantation, tumor volume decreased to an undetectable size then re-grew after a dormancy period. This implicate the small population of tumorigenic cells survive and re-capitulate the tumor after implantation. Representative growth chart showed of 4 different PDX lines after implantation. (D) Histology of the original tumor is maintained throughout subsequent generations. Patient 127 had a histology of papillary serous adenocarcinoma that has been maintained for 6 generations in the corresponding PDX.
Figure 2
Figure 2. Establishment of the PDX line does not enrich for the tumorigenic cell population and human stroma is replaced in the implanted PDX
(A) Representative staining for ALDH1A1, CD133, and CD44 on the patient sample and untreated PDX. (B) Quantification of change in expression of ALDH1A1, CD133, and CD44 between the patient sample and the untreated PDX. Only CD44 had a significant decrease in expression (p-value <0.05). ALDH1A1 and CD133 had no significant change in expression. (C) Human HLA expression in patient and untreated PDX tumors, demonstrating replacement of human stroma with murine cells.
Figure 3
Figure 3. Cancer drug targets are maintained in the PDX line and the PDX response to treatment correlates to the patient's response to primary chemotherapy
(A) The SABiociences RT2 qPCR array for cancer drug targets was run on the patient's tumor and their matched untreated PDX tumor. Differences in relative gene expression for each target was calculated and the 2ΔCt value was determined. Most of the 84 cancer drug target genes had similar expression in the PDX and the original patient sample. 5 gene were down-regulated in the PDX sample, though all 5 are related to VEGF and PDGF signaling (circled in grey). (B) The SABiosciences RT2 qPCR array for cancer drug targets was run on matched subcutaneous PDX tumors and intraperitoneal PDX tumors. Differences in relative gene expression for each target was calculated and the 2ΔCt value was determined. All 84 cancer drug target genes showed a strong correlation between the IP and SQ PDX tumors(C) PDX lines were treated with combination carboplatin and paclitaxel IP weekly. The percent change in tumor volume at 30 days was compared to the patient's response to primary therapy. PDX lines with the greatest decrease in volume significantly correlated to patients with a complete response to therapy (p=0.0009) (D) Classifying reduction in tumor volume by outcome of tumor reductive surgery (optimal debulking vs suboptimal) shows a trend towards PDX with the greatest reduction in volume correlating to optimal debulking for the patient (p-value = NS).
Figure 4
Figure 4. Chemotherapy treatment reduces proliferation and enriches the PDX for cancer stem cells
Tumor cell proliferation was quantified using the Ki67 marker on original patient samples, untreated PDX samples, and chemotherapy treated PDX samples. Change in cancer stem cell marker expression was analyzed after chemotherapy treatment. (A) Representative IHC of Ki67 staining in the patient sample, untreated PDX, and treated PDX. (B) On average, proliferation decreases with chemotherapy treatment in all PDX lines tested. There is no significant change in proliferation between the patient and the untreated PDX. (C) Proliferation rates for each treated and matched untreated pair show that the majority of tumors have a reduced proliferation rate after chemotherapy treatment (D) Representative IHC of CSC markers ALDH1A1, CD133, and CD44 of PDX treated with carboplatin and paclitaxel for 4 weeks. (E) In the treated PDX, expression of ALDH1A1 and CD133 are significantly increased (p-value = 0.0023 and p-value = 0.011 respectively).
Figure 5
Figure 5. RNAseq comparing the treated PDX lines to the untreated PDX lines
Principal component analysis of genes expression in the treated and untreated PDX tumors. While matched treated and untreated PDX tumors clustered together, most treated PDX tumors had change of expression in the same direction indicating a small subset of genes responding to chemotherapy.

Similar articles

See all similar articles

Cited by 47 articles

See all "Cited by" articles

References

    1. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2013. CA: A Cancer Journal for Clinicians. 2013;63(1):11–30. - PubMed
    1. Romero I, Bast RC. Minireview: Human Ovarian Cancer: Biology, Current Management, and Paths to Personalizing Therapy. Endocrinology. 2012;153(4):1593–1602. - PMC - PubMed
    1. Vaughan S, Coward JI, Bast RC, Berchuck A, Berek JS, Brenton JD, Coukos G, Crum CC, Drapkin R, Etemadmoghadam D, Friedlander M, Gabra H, Kaye SB, Lord CJ, Lengyel E, Levine DA, et al. Rethinking ovarian cancer: recommendations for improving outcomes. Nat Rev Cancer. 2011;11(10):719–725. - PMC - PubMed
    1. Bast RC, Hennessy B, Mills GB. The biology of ovarian cancer: new opportunities for translation. Nat Rev Cancer. 2009;9(6):415–428. - PMC - PubMed
    1. Domcke S, Sinha R, Levine DA, Sander C, Schultz N. Evaluating cell lines as tumour models by comparison of genomic profiles. Nat Commun. 2013:4. - PMC - PubMed

Publication types

MeSH terms

Feedback