Epithelial ovarian cancer (EOC) is the leading cause of gynecologic cancer death. Despite initial responses to intervention, up to 80% of patient tumors recur and require additional treatment. Retrospective clinical analysis of OC patients indicates antibiotic use during chemotherapy treatment is associated with poor overall survival. Here, we assessed whether antibiotic (ABX) treatment would impact growth of EOC and sensitivity to cisplatin. Immunocompetent or immunocompromised mice were given untreated control or ABX-containing (metronidazole, ampicillin, vancomycin, and neomycin) water prior to intraperitoneal injection with EOC cells, and cisplatin therapy was administered biweekly until endpoint. Tumor-bearing ABX-treated mice exhibited accelerated tumor growth and resistance to cisplatin therapy compared with control treatment. ABX treatment led to reduced apoptosis, increased DNA damage repair, and enhanced angiogenesis in cisplatin-treated tumors, and tumors from ABX-treated mice contained a higher frequency of cisplatin-augmented cancer stem cells than control mice. Stool analysis indicated non-resistant gut microbial species were disrupted by ABX treatment. Cecal transplants of microbiota derived from control-treated mice was sufficient to ameliorate chemoresistance and prolong survival of ABX-treated mice, indicative of a gut-derived tumor suppressor. Metabolomics analyses identified circulating gut-derived metabolites that were altered by ABX treatment and restored by recolonization, providing candidate metabolites that mediate the crosstalk between the gut microbiome and ovarian cancer. Collectively, these findings indicate that an intact microbiome functions as a tumor suppressor in EOC, and perturbation of the gut microbiota with ABX treatment promotes tumor growth and suppresses cisplatin sensitivity.

BackgroundOvarian cancer is the leading cause of gynecologic cancer death in the United States despite effective first-line systemic chemotherapy. Cancer stem cells (CSCs) retain the ability to self-renew and proliferate and may be a means of harboring disease that evades standard treatment strategies. We previously performed a high-throughput screen to assess differential protein expression in ovarian CSCs compared to non-CSCs and observed that Thy-1 was more highly expressed in CSCs. Our primary aim was to validate Thy-1 (CD90) as a cancer stem cell (CSC) marker in epithelial ovarian cancer (EOC), correlate with clinical outcomes, and assess as a potential therapeutic target.ResultsKaplan Meier (KM) Plotter data were correlated with survival outcomes. Quantitative real-time PCR, flow cytometry, and immunoblots assessed RNA and protein expression. Limiting dilution assays assessed self-renewal capacity and proliferation assays assessed proliferative capacity. RNA in-situ hybridization was performed on patient specimens to assess feasibility. Thy-1 (CD90) is more highly expressed in ovarian CSCs than non-CSCs, in EOC compared to benign ovarian epithelium (P < 0.001), and is highest in serous EOC (P < 0.05). Serous ovarian cancers with high Thy-1 expression have poorer outcomes (median PFS 15.8 vs. 18.3 months, P = 0 < 0.001; median OS 40.1 v. 45.8 months, P = 0.036). Endometrioid ovarian cancers with high Thy-1 have poorer PFS, but no difference in OS (upper quartile PFS 34 v. 11 months, P = 0.013; quartile OS not reached, P = 0.69). In vitro, Thy-1 expression is higher in CSCs versus non-CSCs. EOC cells with high Thy-1 expression demonstrate increased proliferation and self-renewal. Thy-1 knockdown in EOC cells decreases proliferative capacity and self-renewal capacity, and knockdown is associated with decreased expression of stem cell transcription factors NANOG and SOX2. RNA in situ hybridization is feasible in ovarian cancer tissue specimens.ConclusionsThy-1 is a marker of ovarian CSCs. Increased expression of Thy-1 in EOC predicts poor prognosis and is associated with increased proliferative and self-renewal capacity. Thy-1 knockdown decreases proliferative and self-renewal capacity, and represents a potential therapeutic target.

ObjectiveTo evaluate intraperitoneal (IP) tumor engraftment, metastasis and growth in a pre-clinical murine epithelial ovarian cancer (EOC) model using both transabdominal ultrasound (TAUS) and bioluminescence in vivo imaging system (IVIS). MethodsTen female C57Bl/6J mice at six weeks of age were included in this study. Five mice underwent IP injection of 5x10 6 ID8-luc cells (+ D-luciferin) and the remaining five mice underwent IP injection of ID8-VEGF cells. Monitoring of tumor growth and ascites was performed weekly starting at seven days post-injection until study endpoint. ID8-luc mice were monitored using both TAUS and IVIS, and ID8-VEGF mice underwent TAUS monitoring only. Individual tumor implant dimension and total tumor volume were calculated. Average luminescent intensity was calculated and reported per mouse abdomen. Tumor detection was confirmed by gross evaluation and histopathology. All data are presented as mean +/-standard deviation. ResultsOverall, tumors were successfully detected in all ten mice using TAUS and IVIS, and tumor detection correlated with terminal endpoint histology/ H&E staining. For TAUS, the smallest confirmed tumor measurements were at seven days post-injection with mean long axis of 2.23mm and mean tumor volume of 4.17mm 3 . However, IVIS imaging was able to detect tumor growth at 14 days post-injection. Ascites formation was detected in mice at 21 days post-injection.

AbstractObjectiveTo evaluate intraperitoneal (IP) tumor engraftment, metastasis and growth in a pre-clinical murine epithelial ovarian cancer (EOC) model using both transabdominal ultrasound (TAUS) and bioluminescence in vivo imaging system (IVIS).MethodsTen female C57Bl/6J mice at six weeks of age were included in this study. Five mice underwent IP injection of 5×106 ID8-luc cells (+ D-luciferin) and the remaining five mice underwent IP injection of ID8-VEGF cells. Monitoring of tumor growth and ascites was performed weekly starting at seven days post-injection until study endpoint. ID8-luc mice were monitored using both TAUS and IVIS, and ID8-VEGF mice underwent TAUS monitoring only. Individual tumor implant dimension and total tumor volume were calculated. Average luminescent intensity was calculated and reported per mouse abdomen. Tumor detection was confirmed by gross evaluation and histopathology. All data are presented as mean +/- standard deviation.ResultsOverall, tumors were successfully detected in all ten mice using TAUS and IVIS, and tumor detection correlated with terminal endpoint histology/ H&E staining. For TAUS, the smallest confirmed tumor measurements were at seven days post-injection with mean long axis of 2.23mm and mean tumor volume of 4.17mm3. However, IVIS imaging was able to detect tumor growth at 14 days post-injection.ConclusionsTAUS is highly discriminatory for monitoring EOC in pre-clinical murine model, allowing for detection of tumor dimension as small as 2 mm and as early as seven days post-injection compared to IVUS. In addition, TAUS provides relevant information for ascites development and detection of multiple small metastatic tumor implants. TAUS provides an accurate and reliable method to detect and monitor IP EOC growth in mouse xenografts.