Gregory D. Lyng

Bony changes in the jaws of rats treated with zoledronic acid and dexamethasone before dental extractions mimic bisphosphonate-related osteonecrosis in cancer patients

Osteonecrosis of the jaw is associated with aminobisphosphonate use in patients treated with intravenous doses for the prevention of bony metastases. A more complete understanding of the natural history of bisphosphonate-related osteonecrosis of the jaws (BRONJ), factors associated with risk, and its pathobiology has been limited by the availability of human material and the absence of clinical predictability. We now describe an animal model, developed in female Sprague-Dawley rats, in which we replicate many of the clinical, radiographic, and histologic features described in humans. Animals treated with a sequence of zoledronic acid (ZA) and dexamethosone (DX) over a one to three week period developed BRONJ-like changes following extraction of mandibular or maxillary molars. Whereas the extraction sites of control animals underwent predictable healing with rapid epithelialization, animals treated with ZA/DX demonstrated clinical and histological evidence of ulceration overlying areas of necrotic bone. In contrast to images from control animals, radiographs from animals treated with ZA/DX demonstrated poor definition of the alveolar ridge with mixed radiodensity. Modest increases in the extent of the inflammatory infiltrate were seen fourteen days after extraction in ZA-only treated animals compared to control or ZA/DX-treated rats. However, by post-extraction day 28, no differences were observed. Tissue vascularity was most pronounced in ZA-only treated animals compared to ZA/DX or control specimens. Apoptosis of epithelial cells was not observed in any experimental groups, and no evidence of Actinomyces was observed as determined by Periodic Acid Schiff (PAS) staining. The administration of ZA/DX preceding dental extractions in rats therefore results in the development of bony and soft tissue changes that are similar to those noted humans who develop BRONJ, and may provide a useful model for study of its pathogenesis, as well as strategies for its prevention and treatment.

Abstract 2683: The gastrointestinal microbiome and its composition are critical for antitumor efficacy of immune checkpoint inhibition by anti-PD-L1

The intestinal microbiome has become increasingly appreciated as a significant mediator of systemic antitumor immunity/response in both naive and treatment contexts. In naive contexts, an intact intestinal microbiome has been demonstrated to enhance tumorigenesis, and its composition to mediate primary tumor growth kinetics. In the context of cancer treatment, antibiotic depletion of the intestinal microbiota has been reported to inhibit the efficacy of cyclophosphamide and that of the immune checkpoint inhibitor αCTLA4. Compositional modulation of the intestinal microbiota has been found to be sufficient to enhance the antitumor efficacy of αPD-L1. Here, we assessed the relative importance of the intestinal microbiota in mediating αPD-L1 antitumor efficacy in a B16.F10.SIY murine model of melanoma, by performing parallel efficacy studies in C57BL/6 germ-free (Taconic) or specific pathogen free (Taconic or Jackson) mice. We observed that αPD-L1 treatment provided significant antitumor efficacy of in Taconic mice carrying an intact microbiome; however, this efficacy was abolished in germ-free Taconic mice. Furthermore, we observed that tumors of Jackson mice carrying an intact but compositionally different microbiome did not respond to αPD-L1 treatment. Phenotyping of local tumor and systemic immune responses, as well as characterization of the intestinal microbiome in responder vs nonresponder animals provided mechanistic insights. Taken together, these observations suggest that rational modulation of the microbiome may enhance response to immune checkpoint inhibition, and indicate that the gastrointestinal microbiome and its composition are critical for the antitumor efficacy of αPD-L1. Citation Format: Benjamin G. Cuiffo, Caitlin S. Parello, Chelsea Ritchie, Katie Pedrick, Alexandra Kury, Catarina Costa, Brett Van Dam, Jonathan Jung, Gregory D. Lyng, Stephen T. Sonis. The gastrointestinal microbiome and its composition are critical for antitumor efficacy of immune checkpoint inhibition by anti-PD-L1 [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 2683. doi:10.1158/1538-7445.AM2017-2683

Abstract 633: Gene-expression characterization of bioluminescent PDX in orthotopic and ectopic models of pancreatic cancer and metastatic triple-negative breast cancer

Background: Patient derived xenografts (PDX) are regarded as the gold-standard for translational preclinical cancer research due to their preservation of tumor heterogeneity and propagation in vivo. Similarly, grafting human tumor cells of a particular tissue origin into the corresponding orthotopic context enhances translational accuracy; with recapitulation of tumor::microenvironment biology, as well as more accurate tumor progression kinetics and metastasis, a s well as for systemic delivery of therapies as compared to traditional subcutaneous xenografts. Bioluminescent luciferase-reporters into traditional cancer cell lines has allowed for in-life imaging of tumor growth kinetics in real-time, within obfuscated internal organs, improving experimental precision. We have reported on the development and validation of bioluminescent pancreatic cancer and breast cancer PDX in clinically recapitulative orthotopic models. Here, we characterize these bioluminescent PDX by gene-expression profiling; comparing orthotopic and ectopically propagated bioluminescent PDX to their non-transduced parental PDX counterparts. Methods: Pancreatic (PANx-005-Luc) or metastatic breast (HBCx-14-Luc) PDXs stably transduced with lentiviral luciferase were dissociated and implanted orthotopically into pancreas or mammary fat pad respectively or subcutaneously into the flank of NOD scid gamma (NSG) mice. Likewise, the parental reporter-free PANx-005 or HBCx-14 PDX were implanted subcutaneously into NSG mice. For luciferase-expressing PDX implanted orthotopically, tumor growth was monitored in-life (Xenogen IVIS® Lumina Series III instrument (IVIS)). For the PANx-005-Luc, orthotopic tumors were harvested at a mean tumor radiance (TR) of 2.53 × 106 + 1.63 × 106 ph/s, while ectopic luciferase-transduced and parental tumors were harvested at a mean tumor volume of 530 + 337mm3. For the HBCx-14-Luc, orthotopic tumors were harvested individually when TR reached 6.5 × 109 ph/s while ectopic luciferase-transduced or parental tumors were harvested when tumor volumes reached 1500 + 362.5mm3. Tumor samples were fixed in 10% neutral buffered formalin and embedded in paraffin-blocks for histological analysis. RNA was extracted from all tumors and full-coverage human genome gene-expression profiling was performed. Major differences observed in microarray data was validated via quantitative biochemical assays. Results: Stable expression of lentiviral luciferase did not result in significant gene-expression changes to PDX lines, while site of implantation altered some gene-expression characteristics of the PDX. Conclusions: The introduction of an integrating stable bioluminescent reporter does not significantly alter the gene-expression characteristics of PANx-005 or HBCx-14 PDX. Site of implantation plays a role in gene-expression of PDX tumor models. Citation Format: Benjamin G. Cuiffo, Olivier Deas, Ingrid Rankin, Gregory D. Lyng, Stephano Cairo, Katie Pedrick, Alexandra Kury, Enora Le Ven, Jean-Gabriel Judde, Stephen T. Sonis. Gene-expression characterization of bioluminescent PDX in orthotopic and ectopic models of pancreatic cancer and metastatic triple-negative breast cancer. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 633.

Abstract A9: Bioluminescent orthotopic PDX models of primary pancreatic cancer and residual/metastatic breast cancer predict efficacy of standard of care and experimental treatments

Background: Only 5% of investigational anti-cancer agents are ultimately clinically successful. This may be attributed in part to the historical preclinical use of human cell lines that poorly recapitulate tumor heterogeneity in imperfect xenograft models. To model critical influences of tissue environment on tumor behavior and therapeutic response, heterogeneous patient-derived xenografts (PDX) can be assayed following implantation at orthotopic sites in rodents. Challenges to this approach include inaccessibility of certain organs or inability to track residual or metastatic disease. To address these problems, we utilized pancreatic (PANx-005) or metastatic breast (HBCx-14) PDXs transduced with stable bioluminescent reporters in efficacy studies of standard of care (SOC) and experimental treatments in clinically recapitulative models. Methods: Freshly excised PDXs were transduced with lentiviral vectors stably expressing luciferase, and implanted orthotopically into NOD scid gamma (NSG) mice. For our primary pancreatic cancer model, the pancreas was surgically exposed, and ∼2.5 × 106 PANx-005-Luc cells were inoculated directly. Post-op, tumor growth was monitored in-life (Xenogen IVIS® Lumina Series III instrument (IVIS)). Mice were randomized into treatment groups when mean tumor radiance (TR) reached 3.0 × 106 photons/sec. Control animals received no treatment; SOC groups received either focused fractionated radiation (12 Gy as 3×4 Gy fractions, Q5D (SRT)) or gemcitabine (75 mg/kg, 2QWx4, i.p.), and the experimental group received (+)-JQ-1 BET bromodomain inhibitor (50 mg/kg, QD, i.p.). For the residual/metastatic breast cancer model, ∼1.5 × 106 HBCx-14-Luc were injected directly into the 4th inguinal mammary fat pad. Tumor growth and response were monitored throughout the study by IVIS. When individual TR reached 7.5 × 109 photons/sec, the primary tumor was resected and the animal enrolled into a treatment group to be treated for residual/metastatic disease. Control animals received no treatment; SOC groups received either SRT, capacetabine (540 mg/kg QDx5/ 1week rest, p.o.) or docetaxel (20 mg/kg, Q3Wx2, i.p.); while experimental animals received (+)-JQ-1 (50 mg/kg, QD, i.p.). Results: Tumor seeding approached 100% in both models. Growth kinetics resembled clinical indications, including rapid growth and lung metastases for the HBCx-14 tumors, and slower growth of PANCx-005 tumors. Efficacies of individual treatments recapitulated clinical responses: HBCx-14 tumors were high responders to capecitabine and (+)-JQ-1, and low-responders to docetaxel, while PANx-005 tumors displayed high response to (+)-JQ-1 and lower response to gemcitabine. Conclusions: The translational predictivity of preclinical cancer models is enhanced by the use of human PDXs that preserve tumor heterogeneity, are assayed at orthotopic sites, and are utilized in models that recapitulate clinical situations (e.g. treatment of residual disease vs primary tumor). The use of stable bioluminescent reporters in such assays greatly enhances precision in monitoring tumor growth and treatment response. Currently available preclinical oncology models are more recapitulative, precise and predictive than ever before, and appear poised to engender improved translational success. Citation Format: Benjamin G. Cuiffo, Olivier Deas, Gregory D. Lyng, Stephano Cairo, Enora Le Ven, Jean-Gabriel Judde, Stephen T. Sonis. Bioluminescent orthotopic PDX models of primary pancreatic cancer and residual/metastatic breast cancer predict efficacy of standard of care and experimental treatments. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2015 Nov 5-9; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2015;14(12 Suppl 2):Abstract nr A9.

Animal Models of Toxicities Caused by Anti-Neoplastic Therapy

Radiation and chemotherapy induce a wide range of acute and chronic toxicities. Not only are these associated with poor health outcomes but they also limit patients’ ability to tolerate treatment and incur significant increases in resource use and cost. Universally, they impair patients’ quality of life (QoL). In addition to hematological complications such as anemia, thrombocytopenia, and neutropenia, cancer patients are also at risk for a wide range of non-hematological taxicities. These may occure during or soon after cancer treatment (acute toxicities), or they may not develop until well after the completion of treatment (# 100 days, late toxicities) and become chronic and linger for years after the patient’s disease is controlled. The overall incidence of some form of treatment toxicity is almost 100%. Toxicities include those that are tissue-specific such as mucosal injury of some or all of the parts of the gastrointestinal tract (mucositis), cutaneous damage (dermatitis), salivary gland dysfunction, and venous thrombosis. Alteratively, patients may develop more systemic forms of toxicity that result in conditions such as fatigue, depression, cognitive impairment, and cachexia.