Cliby, William A.

5028 Background: Edmonston vaccine strains of MV have significant antitumor activity against ovarian cancer xenografts (Peng et al, Cancer Res, 2002). Oncolytic selectivity is due to overexpression of the MV CD46 receptor in ovarian cancer cells. Furthermore, engineering the virus to express the marker peptide CEA (MV-CEA) allows real time monitoring of viral gene expression/propagation in vivo. Methods: We are conducting a Phase I trial of MV-CEA in patients (pts) with recurrent, taxol and platinum refractory, ovarian or primary peritoneal carcinoma. Eligible pts should have protective anti-MV antibody (ab) titers and normal CEA levels. A standard phase I cohorts of three statistical design is employed. MV-CEA is administered IP via a port-a-cath, q 4 wks for up to 6 cycles. Results: To date, 12 pts have received viral doses from 103 to 106 TCID50. No dose-limiting toxicity has been observed. Most common toxicities included grade 1–2 nausea (4 pts); grade 1–2 fatigue (4 pts); grade 1 fever (3 pts); grade 1 abdominal pain (2 pts). One pt had grade 2–3 arthralgias in cycles 4–6. There was no treatment induced immunosuppression as assessed by DTH, CD4, CD8, immunoglobulin and complement levels, no significant increase in the titers of anti-MV abs in serum or ascites in response to treatment, and no development of anti-CEA abs. Viral genomes were detected in the peripheral blood mononuclear cells of 2 asymptomatic pts by QRT-PCR. No CEA elevation in the serum was observed in the dose range tested, and there was no viral shedding in urine or mouth gargle specimens. Testing of pt tumor samples for CD46 by IHC demonstrated overexpression of the target receptor in 7/8 pts. Best objective response was stable disease (2 - 9 mo duration) in 5 pts, while 3 pts had significant decrease in CA125 levels, of 44%, 72%, and 78%, respectively. Conclusions: In this first human application of an engineered oncolytic MV strain, we have observed excellent tolerance after repeat IP administration of the virus. Despite the low doses employed so far, there is early evidence of biologic activity. Dose escalation continues to the target dose of 109 TCID50. No significant financial relationships to disclose.

INTRODUCTION: To compare two published risk stratification criteria (Milwaukee Model vs. Mayo Criteria) to predict lymphatic dissemination (LD) in endometrioid endometrial cancer (EC). METHODS: Patients who underwent surgery for EC between 1/1/2004-9/30/2013 and had endometrioid histology without stage IV disease were included and classified as either low-risk or at-risk for LD using two independent risk stratification systems. LD was defined as positive nodes at time of surgery or lymph node recurrence within 2 years of surgery after negative lymph node assessment (LNA) or when LNA was not performed. False positive (FP) and false negative (FN) rates for each risk model were calculated. RESULTS: Among 1103 patients, 81 (7.3%) had LD (72 positive LN and 9 LN recurrences). Most patients (90.2%) had stage I disease. Integration of the Milwaukee Model was associated with a low at-risk rate for LD (38.1%) but a relatively high FN rate (13.6%, 95% CI 7.0-23.0). Conversely, the traditional Mayo Criteria using a tumor diameter (TD) cutoff of ≤2 cm (vs >2 cm) had a higher at-risk rate for LD (69.5%) but a FN rate of 0.0% (95% CI, 0-4.5). Modifying Mayo Criteria using a TD cutoff of ≤3 cm (vs >3 cm) identified fewer women at-risk than using ≤2 cm cutoff (56.8% vs. 69.5%), but had a higher FN rate of 3.7% (95% CI, 0.8-10.4). CONCLUSIONS: The Milwaukee Model had the lowest at-risk rate of LD but an unacceptable FN rate. Modifying the Mayo Criteria by increasing the TD cutoff from the traditional ≤2 cm to ≤3 cm would spare an estimated 12.7% of patients LNA, but the accompanying FN rate is unacceptably high. The traditional Mayo Criteria for low-risk EC remains the most accurate in determining which patients LNA can be omitted.

Abstract Objectives: To investigate the antitumor potential of the widely used anticholesterol drug Atorvastatin (ATV) alone, in combination with Carboplatin (CBPT) and with the glycolytic inhibitor PFK158 (PFKi) in ovarian cancer (OC). Methods: MTT assay and the Chou-Talalay methodology were used to test anticancer activity in vitro; half maximal inhibitory concentration (IC50) and combination index (CI) were calculated. Cellular changes were studied with transmission electron microscopy (TEM). ATV was combined with PFKi as we previously showed that PFKi, in addition to glycolysis, inhibits lipid metabolism by targeting PLA2G3. PLA2G3 expression was studied by western blot given that downregulation in OVCAR8 using two shRNAs (sh33, sh35) induced autophagy, a prodeath condition in OC, and was associated with chemosensitization compared to nontargeted controls (NTC) (unpublished data). A random sample of 10 patient-derived OC xenografts (PDX) was studied by TEM. Results: OVCAR8 exhibited a dose dependent proliferation inhibition when treated with ATV (IC50 8μM). PFKi and CBPT IC50 were 4μM and 78μM. PFKi treatment combined with 4μM and 8μM of ATV resulted in ATV sensitizing cells to PFKi by shifting PFKi IC50 down to 2.7μM and 0.4μM, respectively. When ATV was combined in constant ratio with CBPT, CIs were >1.1 indicating antagonism. Combination of ATV 4μM with PFKi 2μM resulted in greater decrease in PLA2G3 and p62 and greater increase in cleaved PARP compared to PFKi alone suggesting higher levels of autophagy and apoptosis. ATV and combination treatment resulted in increased caveolin-1 levels, previously shown to be associated with apoptotic cell death. TEM analysis of ATV and PFKi treated cells showed greater number of lipid rich multilamellar bodies (MLB) compared to controls. PLA2G3 knockdowns exhibited higher MBL formation, lower p62, and when treated with statin and with PFKi, cleaved PARP was significantly higher compared to NTCs. TEM showed that PDXs clustered in low and high MLB groups suggesting a distinct novel phenotype. Conclusions: ATV could enhance PFKi antitumor activity by jointly targeting PLA2G3 leading to metabolic reprogramming of OC. Importantly, ATV antagonized CBPT. Additional studies are in progress to explore the mechanistic pathway of ATV and characterize the significance of MLB in OC as a prodeath phenotype and biomarker of response. Citation Format: Eleftheria Kalogera, Debarshi Roy, Jamie N. Bakkum-Gamez, William A. Cliby, Sean C. Dowdy, Viji Shridhar. Atorvastatin antagonizes carboplatin, synergizes with the glycolytic inhibitor PFK-158, and induces lipid laden multilamellar bodies in ovarian cancer [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 5821.

Abstract Background: High Grade Serous Ovarian Cancer (HGSOC) is a highly aggressive form of ovarian cancer accounting for up to 70% of all cases. The Cancer Genome Atlas (TCGA) has further defined 4 molecular subtypes within HGSOC: differentiated (DIF), immunoreactive (IMM), mesenchymal (MES), and proliferative (PRO). The MES subtype is associated with the worst prognosis and the least likely to be completely resectable when compared with other subtypes. A recent study suggested the existence of a 5-th, so-called anti-mesenchymal (ANM), subtype which is characterized by down-regulation of genes typically upregulated in the MES subtype. Importantly, previous molecular subtype studies have been performed using tumor samples consisting of both cancer cells and stroma. However, the origin of MES biomarkers in OC is still largely unknown and several studies suggest that many of the MES biomarkers could arise from cancer-associated stroma rather than cancer cells. Methods: Specimens from 20 patients (MES subtype, n=15; ANM subtype, n=5) diagnosed with HGSOC between 2010 and 2013 were obtained from Pathology Research Core in Mayo Clinic. A total of 60 FFPE blocks from 1 primary tumor and 2 metastatic sites of each patient were used to make the slides. Five MES-subtype biomarkers, including ACTA2, FAP, POSTN, COL5A1, and VCAN, were selected for immunohistochemistry (IHC) to determine the protein localization and intensities. Two researchers scored tumor sections for intensity as 0 (negative), 1 (weak), 2 (moderate), and 3 (strong), and for localization (stroma/tumor, cytoplasm/nuclear). Statistical analysis was performed using GraphPad Prism 7 software and statistical differences were evaluated using t-test. Results: All 5 proteins evaluated exclusively localized in stroma rather than cancer cells (p<0.001 for all 5 proteins). Only FAP showed significantly different staining intensity between MES and ANM subtype of OC patient samples (p= 0.0114). When comparing primary vs. metastatic sites, only COL5A1 showed significant difference (p=0.019). Conclusion: We report that 5 biomarkers traditionally associated with the MES subtype HGSOC arise from cancer-associated stroma instead of cancer cells. Considering the potential roles of these proteins in mediating HGSOC behavior, it may suggest “stroma targeting” as an alternative therapy to “cancer targeting” to improve OC patient outcomes. Although all 5 genes were reported as MES biomarkers in mRNA-based analysis, we found that FAP was the only protein differentially expressed between MES and ANM subtype suggesting it is a better characteristic biomarker for MES subtype. COL5A1 showed significance in staining between primary and metastatic cancer which may indicates the potential role of COL5A1 in metastasis in OC. Citation Format: Qing Zhang, Kimberly R. Kalli, Chen Wang, William Cliby. The mesenchymal subtype biomarkers arise from stroma rather than cancer cells in ovarian cancer [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 4491.

Transforming growth factor beta (TGF-β) signaling has pleiotropic functions regulating cancer initiation, development, and metastasis, and also plays important roles in the interaction between stromal and cancer cells, making the pathway a potential therapeutic target. LY2157299 monohydrate (LY), an inhibitor of TGF-β receptor I (TGFBRI), was examined for its ability to inhibit ovarian cancer (OC) growth both in high-grade serous ovarian cancer (HGSOC) cell lines and xenograft models. Immunohistochemistry, qRT-PCR, and Western blot were performed to study the effect of LY treatment on expression of cancer- and fibroblast-derived genes. Results showed that exposure to TGF-β1 induced phosphorylation of SMAD2 and SMAD3 in all tested OC cell lines, but this induction was suppressed by pretreatment with LY. LY alone inhibited the proliferation, migration, and invasion of HGSOC cells in vitro. TGF-β1-induced fibroblast activation was blocked by LY. LY also delayed tumor growth and suppressed ascites formation in vivo. In addition, independent of tumor inhibition, LY reduces ascites formation in vivo. Using OVCAR8 xenograft specimens we confirmed the inhibitory effect of LY on TGF-β signaling and tumor stromal expression of collagen type XI chain 1 (COL11A1) and versican (VCAN). These observations suggest a role for anti-TGF-β signaling-directed therapy in ovarian cancer.

Tumor-targeted fluorescence imaging for cancer diagnosis and treatment is an evolving field of research that is on the verge of clinical implementation. As each tumor has its unique biologic profile, selection of the most promising targets is essential. In this review, we focus on target finding in ovarian cancer, a disease in which fluorescence imaging may be of value in both adequate staging and in improving cytoreductive efforts, and as such may have a beneficial effect on prognosis. Thus far, tumor-targeted imaging for ovarian cancer has been applied only in animal models. For clinical implementation, the five most prominent targets were identified: folate receptor α, vascular endothelial growth factor, epidermal growth factor receptor, chemokine receptor 4, and matrix metalloproteinase. These targets were selected based on expression rates in ovarian cancer, availability of an antibody or substrate aimed at the target approved by the Food and Drug Administration, and the likelihood of translation to human use. The purpose of this review is to present requirements for intraoperative imaging and to discuss possible tumor-specific targets for ovarian cancer, prioritizing for targets with substrates ready for introduction into the clinic.

Abstract Background: Most ovarian cancer (OC) patients present with advanced disease and ultimately develop chemo-resistant relapses. Stroma plays a significant role in OC behavior including invasiveness, fibrosis and chemoresistance. TGF-β is important for crosstalk between stroma and cancer cells, and stromal activation in cancer has similarities to matrix remodeling in fibrosis. We have shown that TGF-β and fibrosis-inhibitors can suppress OC cell proliferation, migration and invasion. Such inhibitors are available for clinical use, and we have begun to test their efficacy in OC in vivo. Methods: Animal experiments were approved by Mayo Clinic IACUC. Intraperitoneal xenograft models were derived from OVCAR8 (5×106) cells (OV8-CDX) or 0.1 gm OC patient tumor tissue (PDX) in NSG mice. OV8-CDX and PDX PH003 (an aggressive, chemoresistant and ascites-prone model) (n=5-10 mice/arm) were treated with 2 agents and compared to untreated controls (C): 1) C vs. the fibrosis inhibitor Pirfenidone (P, 200 mg/kg, BID, provided by Genentech); 2) C vs. TGF-β receptor I inhibitor Galunisertib (G, 75 mg/kg, BID, provided by Eli Lily). Mice were pretreated for 2 days before xenografting and treated for 4 weeks with ultrasound (US) to measure tumor growth. At day (d) 28, we i) collected tumor (OV8-CDX); or ii) observed until progression (PH003, PH053). OV8-CDX tumors were used for target gene analysis. Results: P treatment delayed tumor growth in OV8-CDX relative to control (C): tumors established in 0/5 (P) vs. 1/5 (C), and 1/5 (P) vs. 4/5 (C) of mice at d14 and d28, respectively. The reduced tumor weight (g) at d28 was also confirmed (0.20±0.10 vs. 0.54±0.23, p=0.02). Similarly, P treatment in PH003 inhibited the tumor growth (tumor area (cm2) at d28: 0.74±0.43 vs. 1.04±0.51, p=0.28), ascites development (volume (ml): 1.14±0.54 vs. 3.37±0.62, p=0.05) and prolonged survival time (p=0.04). G treatment was also associated with delayed tumor establishment in OV8-CDX relative to C: tumors presented in 0/5 (G) vs. 1/5 (C), and 1/5 (G) vs. 4/5 (C) of mice at d14 and d28, respectively. Associated d28 tumor weight was significantly reduced (0.20±0.10 vs. 0.54±0.23, p=0.02) as well. For PH003, G treatment was associated with delayed tumor growth (tumor area at d28: 0.57±0.27 vs. 1.04±0.51, p=0.05), a marked reduction in ascites (volume: 1.25±0.56 vs. 3.37±0.62, p=0.05) and a prolonged survival time (p=0.04) We identified down-regulation of TGF-β signaling-related genes (e.g. TGFB1, TGFBR2) and fibrosis-related biomarkers (e.g. FAP, COL11A1) following drug treatment confirming inhibition of relevant pathways. Conclusion: We report that 2 currently available inhibitors targeting either fibrosis or TGF-β appear to have efficacy in vivo and delay tumor growth in OC PDX models. These results suggest promise for a strategy of targeting the stromal component of OC. Citation Format: Qing Zhang, Xiaonan Hou, Bradley J. Evans, John J. Weroha, William A. Cliby. Inhibitors of fibrosis and TGF-beta delay tumor xenograft growth and reduce ascites in ovarian cancer models [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 961. doi:10.1158/1538-7445.AM2017-961

Abstract Purpose: Here we assess whether molecular subtyping identifies biological features of tumors that correlate with survival and surgical outcomes of high-grade serous ovarian cancer (HGSOC). Experimental Design: Consensus clustering of pooled mRNA expression data from over 2,000 HGSOC cases was used to define molecular subtypes of HGSOCs. This de novo classification scheme was then applied to 381 Mayo Clinic HGSOC patients with detailed survival and surgical outcome information. Results: Five molecular subtypes of HGSOC were identified. In the pooled dataset, three subtypes were largely concordant with prior studies describing proliferative, mesenchymal, and immunoreactive tumors (concordance > 70%), and the group of tumors previously described as differentiated type was segregated into two new types, one of which (anti-mesenchymal) had downregulation of genes that were typically upregulated in the mesenchymal subtype. Molecular subtypes were significantly associated with overall survival (P < 0.001) and with rate of optimal surgical debulking (≤1 cm, P = 1.9E−4) in the pooled dataset. Among stage III-C or IV Mayo Clinic patients, molecular subtypes were also significantly associated with overall survival (P = 0.001), as well as rate of complete surgical debulking (no residual disease; 16% in mesenchymal tumors compared with >28% in other subtypes; P = 0.02). Conclusions: HGSOC tumors may be categorized into five molecular subtypes that associate with overall survival and the extent of residual disease following debulking surgery. Because mesenchymal tumors may have features that were associated with less favorable surgical outcome, molecular subtyping may have future utility in guiding neoadjuvant treatment decisions for women with HGSOC. Clin Cancer Res; 23(15); 4077–85. ©2017 AACR.