03.13 Effects of an Artificial Placenta on Development and Brain Injury in Premature Lambs

Posted on January 18, 2016

N. L. Werner1, M. Coughlin1, C. J. Perez-Torres4, H. Parmar5, R. Shellhaas2, J. Barks6, J. R. Garbow3, G. B. Mychaliska1 1University Of Michigan,Pediatric Surgery,Ann Arbor, MI, USA 2University Of Michigan,Pediatric Neurology,Ann Arbor, MI, USA 3Washington University,Radiology,St. Louis, MO, USA 4Purdue University,Radiological Health Sciences,West Lafayette, IN, USA 5University Of Michigan,Radiology, Division Of Neuroradiology,Ann Arbor, MI, USA 6University Of Michigan,Pediatric Neonatolgy,Ann Arbor, MI, USA

Introduction: The effect of an extracorporeal artificial placenta (AP) on the developing brain is unknown. The aim of this study was to evaluate brain structure after AP support using magnetic resonance imaging (MRI).

Methods: Institutional guidelines for animal research were followed. AP lambs (n=5), with estimated gestational age (EGA) of 120±2 days (term=145 days), were cannulated for veno-venous ECMO on placental support and delivered. AP support with heparin was maintained for 7±0.5 days before euthanasia at EGA~127 days. Control(C) lambs, with an EGA of 120±2 days (n=3) or 127±2 days (n=2), were euthanized immediately after delivery. All brains were flushed with formalin for fixation and then characterized by MRI. A 25-direction diffusion sequence with an additional b=0 image (no diffusion weighting) was used to calculate fractional anisotropy (FA) and apparent diffusion coefficient (ADC) to qualitatively evaluate white-matter integrity. The b=0 image, plus a T1 sequence, were used to measure brain volume and hull, and to evaluate for hemorrhage.

Results: The AP brains had no evidence of hemorrhage or white-matter injury (Figure A, B). Brain volume appeared to increase with gestational age (C-120 day 1.77E4 mm3 v. C-127 day 1.79E4 mm3). AP brains were on average smaller than either control (1.55E4 mm3). The hull increased with gestational age (C-120 day 1.92E4 mm3 v. C-127 day 2.08E4 mm3). The volume/hull ratio, which is surrogate marker of cerebral folding, showed AP brains with a ratio between the 120 and 127 day controls (C-120 days 86% v. AP 89% v. C-127 days 92%). The ventricles of AP brains were mildly enlarged compared to controls (Figure C, D).

Conclusion: The artificial placenta provided long-term extracorporeal support with no evidence of brain hemorrhage or white-matter injury. The data suggest some ongoing brain development during AP support. Compared to controls, the AP brains were smaller and had larger ventricles; the clinical significance of these findings is unknown.

03.09 IGF-1 intraplacental gene transfer restores serum bioavailability of IGF-1 in growth restricted mice

Posted on January 18, 2016

S. M. Deeney1, K. N. Powers1, B. Dodson1, K. W. Liechty1, A. Marwan1, T. M. Crombleholme1 1Laboratory For Fetal And Regenerative Biology,Department Of Surgery, School Of Medicine, University Of Colorado Denver – Anschutz Medical Campus; Division Of Pediatric General Thoracic And Fetal Surgery, Colorado Children’s Hospital,Aurora, CO, USA

Introduction:

Intrauterine growth restriction (IUGR) is characterized by decreased serum bioavailability of IGF-1 both clinically and across a wide range of animal models. Our laboratory recently described a novel surgical mouse model of intrauterine growth restriction (IUGR) and rescued the IUGR phenotype using intraplacental adenovirus mediated gene transfer of human insulin-like growth factor-1 (Ad-IGF-1). Mouse birthweight, growth trajectory, glucose tolerance and blood pressure correct to control levels. The mechanism by which fetal reprogramming occurs has yet to be fully elucidated. We hypothesize that Ad-IGF-1 intraplacental gene transfer restores the bioavailability of serum levels of IGF-1 in IUGR pups.

Methods:
In time-mated e18 C57/BL6J dams, one of two naturally occurring mesenteric uterine artery branches were ligated in one pup (excluding the pups at the uterine ends), inducing IUGR (n=3). Control pups were those with dual arteries on the opposite uterine horn as well as sham-operated pups (n=5). A subset of IUGR pups (n=3) received 5 ul intraplacental injections of 1×10^8 pfu adenovirus expressing human IGF-1 after artery ligation. Serum samples were obtained by decapitation of the pups following delivery via hysterotomy at e20. Serum and placenta homogenate levels of mouse IGF-1 and IGFBP-3 were measured in duplicate by ELISA. Bioavailability of IGF-1 was defined as the molar ratio of IGF-1 to IGFBP-3. Statistical analysis was by Student’s T-test.

Results:
Serum levels of bioavailable IGF-1 were significantly reduced in IUGR pups compared to controls (P=0.01) and were restored with Ad-IGF-1 gene transfer (p>0.05). Placental levels of bioavailable IGF-1 were not different in all groups (p>0.05).

Conclusion:
Intraplacental Ad-IGF-1 gene transfer restores the serum levels of bioavailable IGF-1, while not changing these levels in the placenta itself. This is the first demonstration of how Ad-IGF-1 gene transfer restores pup serum bioavailable IGF-1 levels in IUGR mice, suggesting a possible mechanism by which IGF-1 induces fetal reprogramming of IUGR mice.

03.10 Ileal CCL3 and IL-12 Expression in Inflammatory Events Associated with Necrotizing Enterocolitis

Posted on January 18, 2016

K. Connolly1, P. J. Matheson2, J. A. Shepherd2, J. W. Smith2, R. N. Garrison2, C. D. Downard2 1University Of Louisville,School Of Medicine,Louisville, KY, USA 2University Of Louisville,Department Of Surgery,Louisville, KY, USA

Introduction: Macrophages clear bacteria from the gut in the early stages of necrotizing enterocolitis (NEC), a potentially lethal inflammatory bowel disease of premature infants. Chemokine (C-C) ligand 3 (CCL3 or macrophage inflammatory protein-1a, MIP-1a) and Interleukin-12 (IL-12) are key chemokines/cytokines in this process. CCL3, a chemokine, recruits immune cells such as monocytes and neutrophils to the site of injury and induces the release of other pro-inflammatory cytokines including IL-1, IL-6, and tumor necrosis factor-a (TNF-a). IL-12, a pro-inflammatory cytokine, recruits and activates immune cells including T cells and NK cells.. We hypothesized that CCL3 and IL-12 expression in the ileum might be increased in the context of NEC-associated inflammation.

Methods: Sprague-Dawley rats were randomized to groups by litter. CONTROLS were delivered vaginally and dam-fed. NEC groups were delivered by C-section 12 hours prematurely, formula fed, exposed to intermittent cold and hypoxia, and given a single oral dose of lipopolysaccharide. Ileum samples were obtained at 0, 12, 24, 48, 72 and 96 hours of life and western blots were performed with antibodies against CCL3, IL-12, and b-actin for normalization by individual animal. Statistical analysis was performed using 2-way ANOVA and a priori P<0.05.

Results: Ileal CCL3 and IL-12 protein expression (see Figure) in NEC was below Control levels at 24 hours, but increased at 72 hours versus Controls (*P<0.05). The change in ileal protein expression occurred between the 24 and 48 hour time points for both mediators.

Conclusion: These data support our prior studies that demonstrated a similar pattern of CCL3 and IL-12 expression in the serum of NEC rats versus Control. Since macrophage function plays a key role in the early development of NEC, down-regulation of IL-12 in the setting of increased CCL3 might represent deranged or inhibited macrophage function in the NEC disease process.

03.11 Chronic Hypoxia Promotes Hypoxia-Inducible Factor-Independent Growth Response in Neuroblastoma

Posted on January 18, 2016

A. L. Alvarez1, B. T. Craig1, E. J. Rellinger1, J. Qiao1, D. H. Chung1 1Vanderbilt University Medical Center,Pediatric Surgery,Nashville, TN, USA

Introduction: Neuroblastoma is responsible for nearly 15% of all pediatric cancer-related deaths and remains a difficult pediatric solid tumor to cure. Acute tumor hypoxia in neuroblastoma is known to promote dedifferentiation and drug resistance in part by stabilizing hypoxia-inducible factors (HIF-1α and HIF-2α ) which function as key drivers of glycolysis, and has been well characterized. Chronic tumor cell hypoxia is also likely to exist in aggressive solid tumors such as neuroblastoma, yet its exact role in tumorigenicity remains poorly understood. We hypothesize that chronic hypoxia is a microenvironmental stress stimulus that plays a critical role in driving neuroblastoma progression. The aim of this study was to establish an in vitro model of chronic hypoxia to assess neuroblastoma cell growth.

Methods: Human neuroblastoma cell lines, BE(2)-C (MYCN-amplified) and SK-N-SH (MYCN-single copy), were cultured in 1% ambient oxygen for 1, 3, 7, or 10 d. Post-hypoxia cell viability was assessed over a time course (0 to 96 h) using CCK-8 assay. Soft agar colony formation after 2 weeks incubation in 0.4%/0.8% agarose gel was used to assess anchorage-independent growth. SDS-PAGE was used to assess protein expression. Two-tailed unpaired Student’s t-test was used for statistical analysis and a p value of <0.05 was considered significant.

Results: Ten days of 1% O2 increased the proliferative capacity of both cell lines compared to controls cultured at ambient O2 (p <0.01 for each individual comparison). Specifically, MYCN–single copy SK-N-SH cells demonstrated a two-fold increase in proliferation following 10 d exposure to 1% O2 compared to controls cultured in 21% O2. A more modest increase (1.5-fold) was observed in the MYCN-amplified BE(2)-C cells. Anchorage-independent growth was also observed in both BE(2)-C and SK-N-SH cells cultured for 10 d at 1% O2. The well-established mediators of the acute hypoxic response HIF-1α and HIF-2α were both upregulated by 1 d as expected, validating the hypoxia culture chamber. Interestingly, expression levels of both factors returned to basal levels by 10 d of continuous hypoxic exposure. Hexokinase-1 and lactate dehydrogenase-A, two downstream markers of glycolytic flux, were similarly decreased at 10 d of continuous exposure to 1% O2.

Conclusion: Neuroblastoma is a pediatric solid tumor with significant intra-tumoral heterogeneity featuring areas of necrosis and hemorrhage that may implicate chronic hypoxia as a key microenvironmental regulator of tumor progression. Our model shows that prolonged continuous exposure to subnormal oxygen levels induces a phenotype that enhances both cellular proliferation and anchorage independent growth. This behavioral change appears to be both a HIF- and glycolysis-independent phenomenon.

03.06 An Alternative DNA Repair Pathway Enhances Neuroblastoma Chemoresistance in Hypoxic Microenvironments

Posted on January 18, 2016

L. Gaston1, E. Corwin1, D. Bashllari1, B. Cummings1, A. Shiang1, X. Topalli1, V. Castle1, E. Newman1 1University Of Michigan,Pediatrics, Obstetrics And Gynecology, And Pediatric Surgery,Ann Arbor, MI, USA

Introduction: Neuroblastoma (Nb) is a neoplasm of neural crest origin that accounts for 8% of childhood malignancies and 15% of cancer deaths in this same population. This discrepancy may be attributed to high-risk patients presenting with widely disseminated disease, of which only 25% survive despite cytotoxic therapies. Metastatic Nb cells display increased chemoresistance and a hypoxic environment has been shown to dedifferentiate Nb cells to an immature neuroblastic state with rapid growth and cell survival advantage. Our preliminary data demonstrates that hypoxia induces expression of DNA Ligase III (lig3), a mediator of an alternate nonhomologus end-joining (altNHEJ) DNA repair program in Nb cells. Additional work from our laboratory has shown increased expression of this efficient yet error prone repair pathway in tumorigenic Nb cell lines and in tumors of patients with poor survival outcomes. Given this, we hypothesized that induction of altNHEJ confers a survival advantage to Nb cells in hypoxic microenvironments and to standard chemotherapeutic agents.

Methods: S-type Nb cells (SHEP) were divided into three treatment groups and cultured for 1) 72h at 21% O2, 2) 72h at 1% O2, and 3) 72h at 1% O2 followed by 24h reperfusion. Cells were fixed and analyzed by immunocytochemistry (ICC) for gH2AX and lig3 with DAPI mounting solution. Images were generated with fluorescent microscopy at 40x. Quantitative analysis was carried out with ImageJ software, expressed as mean number of foci/nucleus±SD. SHEP and IMR32 Nb cells were then cultured under the above conditions with doxorubicin, a standard cytotoxic utilized in Nb treatment (0, 1, 5, and 10µg/mL). Cell viability was determined at 72h by trypan blue exclusion. Lig3 expression was analyzed by ICC as described above. All experiments were repeated in triplicate and statistical analysis was performed with ANOVA.

Results: In hypoxic conditions, Nb cells acquired more DNA damage compared to normoxic controls (0.080±0.014 vs. 0.025±0.021 gH2AX positive foci/nucleus, p<.05) with decreased overall survival (38%±4 vs 78%± 12, p<.01). Hypoxic Nb cells had relative resistance to increasing doses of doxorubicin compared to normoxic controls at 1, 5, and 10µg/mL( p<.05). Lig3 expression increased in surviving hypoxic, doxorubicin-treated Nb cells in a dose-dependent manner (p<.0001).

Conclusion: Lig3, a major DNA repair factor in altNHEJ is activated in hypoxic conditions and confers Nb cell survival advantage to Doxorubicin. Lig3 or its interacting altNHEJ partners may prove promising therapeutic targets for Nb patients with high-risk disease.

03.07 The Artificial Placenta Promotes Type 2 Pneumocyte Maturation in Fetal Lambs

Posted on January 18, 2016

M. A. Coughlin1, N. L. Werner1, K. Marchetti1, L. Pierce1, T. Major1, G. B. Mychaliska1 1University Of Michigan,ECLS Laboratory,Ann Arbor, MI, USA

Introduction: Recreating the intrauterine environment with an extracorporeal artificial placenta (AP) to treat extreme prematurity is a promising strategy. The aim of this study was to assess lung development during AP support.

Methods: Institutional guidelines for animal research were followed. AP lambs of estimated gestational ages 116-120 days (term=145 days) were cannulated for VV-ECLS (internal jugular vein drainage and umbilical vein reinfusion) on placental support and delivered. They were intubated and the endotracheal tube filled with fluid and clamped (n=3) or held at a constant pressure of 8cmH2O (n=2). Tissue control (TC) animals were delivered and immediately euthanized. These were divided into early gestational age (115-122 days, n=4) and later gestational age (123-129 days, n=3). A bronchoalveolar lavage was obtained during necropsy on all animals and spun down to remove the solid cell component. The remaining fluid was used to obtain ELISA assays for surfactant related proteins B and C (SPB/SPC).

Results: The AP animals survived an average of 7 days. Levels of SPB in early vs. late tissue controls were 37.3±2 vs. 44.6±1.4 pg/ml (p=0.002). Levels of SPC in early vs. late tissue controls were 12.7±2 vs. 17±5.6 pg/ml (p=0.07). The average concentration of SPB in the AP lambs vs. later tissue controls was 41.1±5.5 vs. 44.6±1.4 pg/ml (p=0.24). The average concentration of SPC in the AP lambs vs. later tissue controls (Figure) was 22.4±1.8 vs. 17±5.6 pg/ml (p=0.02).

Conclusion: There was a significant difference seen in the levels of SPB and a trend toward increased levels of SPC with increasing ages of tissue controls. SPB levels are comparable between controls and AP animals. SPC levels were significantly higher in AP animals compared to controls. These data suggest that the artificial placenta promotes type 2 pneumocyte maturity.

03.08 Silencing of Intersectin 1 Decreases Neuroblastoma Tumor Growth in an Orthotopic Mouse Model

Posted on January 18, 2016

J. C. Harris1, A. Russo3, E. Herrero3, J. P. O’Bryan3, B. Chiu2 1Rush University Medical Center,Surgery,Chicago, IL, USA 2University Of Illinois At Chicago,Pediatric Surgery,Chicago, IL, USA 3University Of Illinois At Chicago,Pharmacology,Chicago, IL, USA

Introduction: Neuroblastoma is the most common solid extracranial tumor in childhood and accounts for 15% of all pediatric cancer deaths. Intersectin 1 (ITSN1) protein is involved in phosphoinositide 3-kinase (PI3K) signaling, which has been shown to stabilize the MYCN oncogenic transcription factor. Silencing ITSN1 in vitro significantly reduced the anchorage-independent growth of neuroblastoma cells. We hypothesize that silencing ITSN1 with shRNA in neuroblastoma cells leads to decreased tumor growth in an orthotopic mouse model.

Methods: We have previously created stable SK-N-AS neuroblastoma cell lines with empty vector (pSR) and vector that contains scrambled shRNA (pSCR), that served as control lines, or vector that contains shRNAs to ITSN1 (Sh#1 and Sh#2). Orthotopic neuroblastoma tumors of each cell line were established in three immunocompromised mice by injecting 1×106 cells directly into the adrenal gland. Tumor volume was monitored weekly with ultrasound, using tumor volume >1000mm3 as the end point. Harvested tumors were analyzed with anti-ITSN1 antibody on Western blot. Tumor growth was analyzed using Kaplan Meier curves and student’s T-test, with p<0.05 deemed statistically significant.

Results: Orthotopic tumors were successfully created in all cell lines. At day 25-post injection, tumor size for pSR was 1235.7±761.2mm3, pSCR was 1311.4±296.5mm3, Sh#1 was 574.4± 539.9mm3, Sh#2 was 439.5±95.3mm3. Sh#2 was significantly smaller than pSCR (p=0.027), and there was a trend toward significance that Sh#1 was smaller than pSCR (p=0.13). Overall survival was superior in Sh#2, 31±1.7 days, compared to pSCR, 25±0 days (p=0.025). Again, a trend towards overall survival in Sh#1, 31±7.2 days, compared to pSCR, 25±0 days (p= 0.11). There was no difference in tumor growth between pSR with the other groups, likely due to high variance within this group, overall survival of 29.67±14.1 days. Western blot analysis of the harvested tumor showed decreased ITSN1 expression in experimental groups Sh#1 and Sh#2 compared to the control groups pSR and pSCR.

Conclusion: Silencing ITSN1 expression in neuroblastoma cells leads to decreased tumor growth in vivo. We demonstrate that orthotopic animal models can serve as an authentic ‘read out’ of molecular manipulation of ITSN1 signaling in vitro. Thus, understanding the role of ITSN1 signaling pathways in tumor development may provide new insight into neuroblastoma tumorigenesis.

03.03 Sphingolipid Signaling in Enteric Glia: Implications for Intestinal Disorders of Inflammation

Posted on January 18, 2016

B. D. Bauman1, J. Meng1, S. Banerjee1, S. Roy1, W. Kennedy3, B. J. Segura1,2 2University of Minnesota,Pediatric Surgery,Minneapolis, MINNESOTA, USA 3University Of Minnesota,Neurology,Minneapolis, MINNESOTA, USA 1University Of Minnesota,General Surgery,Minneapolis, MINNESOTA, USA

Introduction:
Our studies presently underway are aimed at better understanding the underlying risks of pediatric intestinal disorders including inflammatory bowel disease and necrotizing enterocolitis (NEC) in premature infants. Our central hypothesis is that the relative prematurity of patients with NEC puts at them at greater risk of intestinal inflammation due to the underdeveloped nature of their enteric nervous system. Evidence from our lab and others suggests enteroglia may play a significant role in modulating intestinal inflammatory disorders through the maintenance of gut barrier function. Specifically, we have found that sphingosine-1-phosphate (S1P), a bioactive lipid, and FTY720, a known S1P agonist, mitigates intestinal inflmmation. To more fully understand the link between enteric glia, bioactive lipid signaling and NEC, we are now extending our studies into models of pediatric disease.

Methods:
Using in-vivo approaches involving DSS induced colitis in the mouse model, and in-vitro approaches involving the study of rat intestinal epithelial IEC-6 cells we investigated the putative signaling properties of the bioactive lipid sphingosine-1-phosphate (S1P)–known to cause calcium signaling in enteric glia, which bear its receptors based on our prior studies.

Results:

The S1P analog FTY720 (0.3 mg/kg daily by oral gavage) markedly attenuated the inflammatory response induced by 3.5% dextran sodium sulfate (DSS) by H&E analysis compared with controls in a mouse model of colitis at 3 and 5 days. Cultured rat intestinal epithelial cells (IECs) demonstrated enhanced tight junction formation (Occludin staining and transepithelial resistance) in response to enteric glial cell (EGC)-cultured media within 48 hours. By flow cytometry, Rat EGCs display traditional markers and the Toll like receptor 4 (TLR4). By QPCR, Rat EGCs show diminished expression of TLR4, TNFa, IL-1b, and iNOS response to S1P (1uM) treatment.

Conclusion:
Our data suggest that the sphingolipid analog FTY720 and S1P have effects upon inflammation in-vivo and in-vitro, respectively. This suggests that sphingolipids may modulate intestinal inflammation through a mechanism involving the enteric nervous system, potentially serving as a mode of preventive therapy in adults and children alike

03.04 The Feasibility of Spring Mediated Extra-Peritoneal Intestinal Lengthening

Posted on January 18, 2016

A. Scott1, J. D. Rouch1, N. Huynh1, E. Chiang1, S. L. Lee1, B. M. Wu2, J. C. Dunn1,2, S. Shekherdimian1 1University Of California – Los Angeles,Department Of Surgery, Division Of Pediatric Surgery,Los Angeles, CA, USA 2University Of California – Los Angeles,Department Of Bioengineering,Los Angeles, CA, USA

Introduction: Currently, models used for mechanical intestinal lengthening achieve lengthening within the intra-abdominal cavity. Here we show that spring-mediated lengthening can be achieved outside the intra-abdominal cavity through a stoma.

Methods: Using Sprague-Dawley rats, the defunctionalized limb of a Roux-en-y jejunojejunostomy was exteriorized from the peritoneal cavity through a defect created in the anterior abdominal wall (n=8). An encapsulated polycaprolactone spring was placed into the extra-abdominal segment. The stoma containing the spring was secured under the skin to protect it from desiccation and destruction. After 4 weeks, segments were retrieved for histological analyses. Animals who had blank capsules placed in the extra-peritoneal de-functionalized limb served as controls (n=3).

Results: Stomal segments were successfully lengthened from 1.0 cm to 2.5 ± 0.4 cm, whereas control segments were lengthened from 1.0 cm to 1.5 ± 0.1 cm (p<0.05). After lengthening, the exteriorized bowel remained viable and patent. Lengthened segments had increased smooth muscle thickness and crypt depth when compared to normal jejunal mucosa.

Conclusion: Spring-mediated mechanical lengthening is not limited to the abdominal cavity. Extra-peritoneal lengthening yields greater than a 2-fold increase in intestinal length. Lengthening through a stoma removes the need to enter the abdomen for subsequent re-lengthening of intestinal segments.

03.05 Escherichia coli 07:K1 CE10 Colonizes Neonatal Rats and Protects from Necrotizing Enterocolitis

Posted on January 18, 2016

J. D. Bowling1, J. Lim1, J. Golden1, A. Dossa1, B. Bell1, L. Chase1, J. Wang1, A. Grishin1, H. R. Ford1 1Children’s Hospital Los Angeles,Pediatric Surgery,Los Angeles, CA, USA

Introduction: Early post-natal microbiota is presumed to play a critical role in the pathogenesis of necrotizing enterocolitis (NEC), a severe gastrointestinal disease found often in premature infants. Whereas some colonizing bacteria may act as opportunistic pathogens, other bacteria might be innocuous and even protective. In our previous studies, the presence of E. coli CE10 in 4 day old rats negatively correlated with NEC. Here we tested the hypothesis that CE10 is capable of colonizing neonatal rats and protecting them from NEC.

Methods: E. coli CE10 was transformed with a plasmid (pJK_proK14_eGFR) conferring ampicillin resistance and expression of green fluorescent protein (GFP). Transformed bacteria were introduced to neonatal rats with the first feeding. Afterwards, the animals were subjected to 4 days of the NEC-inducing formula-feeding/hypoxia (FF/H) regimen. NEC was scored histologically. Stool and ileum samples were plated for total bacterial load and the numbers of E. coli C10. The level of apoptosis was analyzed using TUNEL staining and fluorescence microscopy.

Results: E. coli CE10 established itself as a first colonizer in animals that received it. FF/H alone caused NEC in 60% of neonatal rats. Introduction of E. coli CE10 reduced the incidence of NEC to 39% (p = 0.027). FF/H yielded a higher rate of apoptosis than E. coli CE10 via TUNEL staining. EC was scored histologically. Stool and ileum samples were plated for total bacterial load and the numbers of E. coli C10. The level of apoptosis was analyzed using TUNEL staining and fluorescence microscopy

Conclusion: E. coli 07:K1 CE10 is an efficient first colonizer in neonatal rats. CE10 significantly reduced the incidence of NEC. Bacteria similar to this strain can be used for the prophylaxis of NEC.

03.02 FX11 Inhibits Aerobic Glycolysis and Promotes Apoptosis in MYCN-amplified Neuroblastomas

Posted on January 18, 2016

E. J. Rellinger1, B. T. Craig1, J. Qiao1, K. Kim1, D. H. Chung1 1Vanderbilt University Medical Center,Pediatric Surgery,Nashville, TN, USA

Introduction: Neuroblastoma (NB) is a heterogeneous pediatric solid tumor with diverse outcomes ranging from spontaneous regression to clinical progression despite our most aggressive therapies. MYCN-amplifications are common with high-risk NB, a designation that portends <50% survival. Myc-driven tumors are ‘glucose addicts’ relying predominantly upon glycolysis for energy production even in the presence of adequate oxygen, a phenomenon known as the Warburg effect. Lactate dehydrogenase is the terminal enzyme of glycolysis. One of its isoforms, lactate dehydrogenase A (LDHA), is preferentially expressed in Myc-driven tumors and avidly converts pyruvate to lactate. FX11 [3-dihydroxy-6-methyl-7-(phenylmethyl)-4-propylnaphthalene-1-carboxylic acid] is a gossypol derivative that selectively inhibits the LDHA isoform and demonstrates preclinical efficacy in C-myc driven pancreatic cancer and lymphomas models. However, the efficacy of FX11 in N-Myc driven NBs is unknown. We hypothesized that MYCN-amplified NBs would be susceptible to LDHA inhibition using FX11.

Methods: We first evaluated the protein expression of LDHA in twelve of our NB cell lines (eight MYCN-amplified and four MYCN-single copy). Two MYCN-amplified cell lines (BE(2)-C and IMR-32) and two MYCN-single copy cell lines (SK-N-SH and SK-N-AS) were used to evaluate the effects of FX11 on NB growth. Cell viability was examined using a tetrazolium-based assay with validation completed by manual cell counts. Immunoblotting for PARP and caspase 3 cleavage products was our measure of apoptosis induction. Lactate concentration was quantified from culture supernatants using a colorimetric assay.

Results: LDHA was ubiquitously expressed in both MYCN-amplified and MYCN-single copy NB cell lines. Treatment with FX11 (10 μM) significantly decreased the levels of lactate at 6 h, validating that FX11 effectively blocks lactate production in NB cells. Both tetrazolium-based measures and manual cell counts demonstrated that FX11 decreased cell viability of NB cells after three days of treatment in our MYCN-amplified cell lines (BE(2)-C and IMR-32) and four days for MYCN-single copy cell lines (SK-N-SH and SK-N-AS). These time courses coincided with induction of PARP and Caspase 3 cleavage in MYCN-amplified cell lines. Notably, no Caspase 3 products were detectable with FX11 treatment in MYCN-single copy cell lines (Fig.).

Conclusion: FX11 has comparable in vitro efficacy in MYCN–amplified NBs to those previously reported in C-Myc driven pancreatic and lymphoma models. LDHA blockade successfully induced apoptosis in our MYCN-amplified tumors. In vivo trials are now underway to evaluate the preclinical efficacy of FX11 in MYCN-amplified NBs.

09.18 Do minority women wait longer for their definitive breast cancer surgery after initial diagnosis?

Posted on January 18, 2016

L. Steel1, Y. R. Li1, E. Carrigan1, J. Tchou1 1University Of Pennsylvania,Department Of Surgery,Philadelphia, PA, USA

Introduction: Prompt and appropriate surgical management following diagnosis significantly impacts long-term survival in breast cancer patients. However, access to care can often be delayed. Some studies have observed that there are racial discrepancies in the length of time between initial cancer diagnosis and definitive cancer surgery, and race has been repeatedly shown to be an independent predictor of survival.

Methods: To test the hypothesis that the length of time between breast cancer diagnosis and definitive surgery is associated with race, we performed a retrospective chart review of a pilot breast cancer patient cohort (n=424) who have received their surgical care at a single institution between 1995 and 2015 to establish feasibility. Women were stratified into the following categories based on self-reported race: Asian (16), Black (112), White (283), and Other (9). The interval from cancer diagnosis to definitive cancer surgery was defined as the number of days from the date of the patient’s initial positive core biopsy or surgical biopsy to the date of definitive surgical intervention (i.e., lumpectomy or mastectomy).

Results: The average diagnosis to surgery time was the longest for African Americans (65.29 days, median = 42) and significantly shorter for White women (49.76 days, median = 34) (two-sided t-test p <0.008). Asians (48.56 days, median = 35.5) and patients of other ethnicities (52.33 days, median =39.92) had comparable diagnosis to surgery times as compared to White women, though their cohort sizes were small and precluded statistical analysis.

Conclusion: Our findings are consistent with other studies showing that there is a prolonged diagnosis to surgery time for women of African American descent. It is not clear whether this is due to delays in seeking care after diagnosis or if there are delays in scheduling surgery. Socioeconomic and cultural differences as well as disparities in health literacy and education may all contribute to these differences. Further analysis is planned to expand our pilot cohort to include more than 3,000 patients consecutively treated at our institution during this study period to evaluate the impact of other covariates, which are not commonly available in national databases but are readily available from our electronic medical record, on the relationship between race, prolonged definitive surgery time interval and clinical outcomes (i.e., age, socioeconomic status, insurance, and comorbidities such as body mass index). Our goal is to address whether a delay in receiving definitive surgery following diagnosis significantly impacts overall or cancer-specific survival, as narrowing this time window may be a targetable measure for breast cancer care providers in reducing existing racial disparities in cancer care outcomes.

22.09 Toll-like Receptor 4 Regulates the Expression of Intracellular Serpins in Necrotizing Enterocolitis

Posted on January 18, 2016

D. F. Nino1, S. Sho3, C. P. Sodhi1, M. Good2, C. J. Luke2, G. A. Silverman2, D. J. Hackam1 1Johns Hopkins University School Of Medicine,Pediatric Surgery,Baltimore, MD, USA 2Children’s Hospital Of Pittsburgh Of UPMC,Newborn Medicine,Pittsburgh, PA, USA 3University Of California – Los Angeles,Surgery,Los Angeles, CA, USA

Introduction: Necrotizing enterocolitis (NEC) is the most frequent and lethal pathology that afflicts the gastrointestinal tract of premature infants. Our lab has previously demonstrated the critical role that the innate immune receptor for lipopolysaccharide (LPS), toll-like receptor 4 (TLR4), and subsequent exaggerated inflammatory signaling play in the pathogenesis of NEC and intestinal necrosis. Intracellular serpin B3 has been identified as a pro-survival factor with a novel role in necrosis inhibition. Given the characteristic necrosis observed in NEC, we hypothesized that TLR4 signaling in the gut plays a role in the regulation of intracellular serpin expression.

Methods: Under IRB approval, intestinal samples were obtained from patients diagnosed with NEC and evaluated for the expression of serpin B3 by qRT-PCR and immunohistochemistry (IHC). The role of TLR4 was determined using a model of endotoxemia (LPS 1mg/kg IP 3 h) or experimental NEC in neonatal (7 – 12 day old) C57Bl/6, TLR4 KO or mice lacking TLR4 in the intestinal epithelium (TLR4ΔIEC). NEC was induced by feeding formula supplemented with bacteria isolated from human NEC via gavage five times/day and exposure to hypoxia (5%O2, 95%N2) for 10 min in a hypoxic chamber twice daily for 4 days. TLR4 and serpin B3 deficient enterocytes (IEC-6 cells) were generated using lentivirus-TLR4-shRNA or lentivirus-serpin B3-shRNA, respectively. The role of serpin B3 in the inflammatory response of enterocytes challenged with LPS was evaluated by IL-6, iNOS and IL-1b mRNA expression determined by qRT-PCR.

Results: Human and murine NEC intestinal samples display a significantly increased expression of serpin B3 by mRNA (58 fold and 6 fold, respectively) and evidenced by IHC. These findings were determined to be TLR4-dependent, since serpin B3 expression was significantly lower in TLR4-deficient enterocytes compared to wild-type controls (0.16 vs. 8 fold relative to actin) when challenged with endotoxin. These results are further supported by the fact that endotoxemia leads to upregulation of serpin B3 in wild-type mice (2 fold) but have no effect on TLR4 deficient mice. Moreover, TLR4ΔIEC mice were protected from developing NEC and had decreased levels of serpin B3. Strikingly, when compared to wild-type IEC-6 cells exposed to LPS, serpin B3-deficient cells were found to have increased expression of the inflammatory mediators iNOS (4 fold), IL-6 (2 fold) and IL-1b (3 fold).

Conclusion: Intracellular serpin B3 levels are significantly increased in the intestine of both human and murine NEC. Normal TLR4 expression in the intestinal epithelium is required for serpin B3 expression. Our findings suggest a potential role for TLR4 regulation of serpin B3, a molecular modulator of intestinal inflammation and necrosis, hallmarks of this devastating disease.

22.10 Bladder Tumor Initiating Cells as Predictors of Outcome

Posted on January 18, 2016

K. Skowron1, J. Namm5, S. Pitroda3, M. Beckett3, N. Khodarev3, M. Posner1,4, G. Steinberg2, R. Weichselbaum3 1University Of Chicago,Department Of Surgery / Section Of General Surgery,Chicago, IL, USA 2University Of Chicago,Department Of Surgery / Section Of Urology,Chicago, IL, USA 3University Of Chicago,Department Of Radiation And Cellular Oncology,Chicago, IL, USA 5Loma Linda University Health,Department Of Surgery,Loma Linda, CA, USA 4University Of Chicago,Department Of Surgery / Section Of Surgical Oncology,Chicago, IL, USA

Introduction:
Bladder cancer is the 9th leading cause of cancer death in the U.S. Treatment consists of radical cystectomy with adjuvant chemotherapy or bladder-sparing chemo-radiotherapy in selected cases. Volkmer et al. identified bladder tumor initiating cells (BTICs) in progressive stages of differentiation (1). We established patient derived xenografts (PDX) in order to study BTIC biology in vivo. We hypothesized that patient tumors with a greater population of basal or poorly differentiated BTICs would have a poor clinical outcome compared to patients with well-differentiated tumors. Gene expression analysis was performed with the goal of identifying prognostic markers.

Methods:
Patients undergoing cystectomy are recruited under an IRB-approved protocol. Tumor samples are received from pathology; fragments are injected subcutaneously into the flank of NOD/SCID mice and monitored for engraftment. Remaining tissue is analyzed with flow cytometry to quantify the BTIC populations. Any tumors which engraft are excised, reimplanted into further generations of mice, and sorted for RNA extraction of the BTIC populations. Subsequent xenograft tumors are treated with either radiation or cisplatin, and analyzed for tumor growth and changes in BTIC populations. Clinical data are gathered using the REDCap web-based data server. RNA expression is assessed using Illumina Human HT-12 array.

Results:
Of 70 patient samples, 41 tumors have engrafted in mice. There was no correlation between engraftment and survival. However, a correlation between greater basal BTIC populations and overall survival approached significance (p=0.09). If patients did not receive systemic chemotherapy (n=28), the basal BTICs conferred a greater risk to overall survival (LR 5.5, p=0.06) than among patients (n=21) who received any systemic chemotherapy (LR 0.62, p=0.74). The basal BTICs comprise an average of 1% of the tumor cells (range 0-17%). Early results indicate a differential response to cisplatin and radiation among PDX tumors for both agents. Analysis of the RNA array identified a signature of 54 genes suppressed in basal BTICs relative to the more differentiated tumor cells. This gene signature correlates with overall survival (p=0.0047) and disease-specific survival (p=0.0085) in a large cohort of patients.

Conclusion:

Our results suggest that the presence of a small percentage of basal BTICs portends a poor clinical outcome, which may be improved by the addition of chemotherapy. RNA evaluation of these BTICs has revealed potential biological markers and targets for future therapy, as well as a gene signature which may provide a tool for improved patient selection for therapy.

(1) Volkmer JP, Sahoo D, Chin RK, et al. Three differentiation states risk-stratify bladder cancer into distinct subtypes. PNAS 2012;109:2078-2083.

22.06 Sphingosine Treatment Prevents Lung Pseudomonas aeruginosa Infection in Burn Injured Mice

Posted on January 18, 2016

T. C. Rice1, E. F. Midura1, E. Gulbins1,2, M. J. Edwards1, C. C. Caldwell1 1University Of Cincinnati,Division Of Research,Cincinnati, OH, USA 2University Hospital Essen,Department Of Molecular Biology,Essen, NORTH RHINE – WESTPHALIA, Germany

Introduction: Burn-injured patients are susceptible to severe pulmonary infection by Pseudomonas aeruginosa (PA). Here, we investigate a potential role of the ceramide/sphingosine axis in increased susceptibility. Ceramide levels are regulated by acid sphingomyelinase (Asm) activity that converts the membrane phospholipid sphingomyelin to ceramide. Subsequently, ceramide is hydrolyzed via acid ceramidase to the long chain base, sphingosine. Beyond its cellular localization, sphingosine can be observed extracellularly in microparticles (MPs). Although it is known that sphingosine treatment can be used to combat infection, the role of cellular sphingosine or sphingosine-containing MPs in the defense against pulmonary infection in burn patients has not been elucidated. Altogether, we hypothesize that 1) injury alters Asm activity and subsequently sphingosine levels such that susceptibility to infection is increased and 2) sphingosine treatment will prevent lung PA infection in burn injured mice.

Methods: Outbred mice were subjected to a dorsal 28% total body surface area, full thickness scald injury. First, we characterized sphingosine expression in the lungs (by immunohistochemistry) and in bronchoalveolar lavage (BAL) derived MPs (by Nanoparticle Tracking Analysis) on post burn day one (PBD1). Second, Asm activity and MP numbers were determined in BAL fluid in mice 24 hours following inoculation with PA. Finally, to determine the effects of sphingosine on infection, some PBD1 mice were pre-treated with sphingosine. All mice were then inoculated with PA and four hours later, colony forming units (CFUs) were enumerated.

Results: Similar to what is seen clinically, infected burn-injured mice had increased mortality as compared to infected uninjured mice (50.0% vs 18.2%, p=0.048). To determine underlying mechanisms of this mortality, we first observed on PBD1, there is a four-fold reduction in bronchial epithelial sphingosine expression in burn injured mice compared to sham. Further, these mice have a significant reduction in total MPs in BAL fluid. Secondly, there was a significant reduction in Asm activity and total MPs in BAL fluid in infected burn mice compared to infected uninjured mice. The BAL MPs from burn infected mice have significantly reduced sphingosine expression. Finally, following PA inoculation, burn mice have a three-log increase in pulmonary infection compared to sham. However, when burn mice are treated with inhaled sphingosine prior to inoculation, there is a significant reduction in pulmonary infection to a level similar to control mice.

Conclusion: The data demonstrate that 1) burn decreases bronchial sphingosine levels and MP numbers, 2) upon infection, injured mice have decreased Asm activity and MP sphingosine levels, and 3) sphingosine treatment prevents lung infection of PA in burn injured mice. These data suggest that in burn patients, treatment with sphingosine will reduce subsequent pulmonary infection.

22.07 Intestinal Alkaline Phosphatase Prevents Burn Induced Intestinal Permeability and Bacterial Translocation

Posted on January 18, 2016

F. Adiliaghdam1, M. Gharedaghi1, S. Yan1,2,3, S. R. Hamarneh1, S. A. Morrison1, L. G. Rahme1,2,3, R. A. Hodin1 1Massachusetts General Hospital,General Surgery,Boston, MA, USA 2Shriners Hospitals For Children,Boston, MA, USA 3Harvard School Of Medicine,Department Of Microbiology And Immunobiology,Boston, MA, USA

Introduction: Burn patients are at increased risk for systemic sepsis and multi-organ failures. Burn-induced gut barrier dysfunction and bacterial translocation is thought to be one of the sources of bacterial presence in the blood stream. Previously, we have shown that IAP detoxifies bacterial pro-inflammatory factors, prevents endotoxemia, enhances gut barrier functions, and preserves intestinal microbial homeostasis. We have hypothesized that supplemental intestinal alkaline phosphatase (IAP) could be used as a therapeutic modality to prevent gut-derived sepsis.

Methods: 6-week old CD-1 mice were divided into 4 groups (n=5 in each group). We inflicted a 30% total body surface area burn on the back of mice +/- intradermal Pseudomonas aeruginosa (PA) infection mid-eschar of burn site to induce gut barrier dysfunction. IAP or vehicle were administered orally by gastric gavage in a dose of 200U per gavage at 3 and 12 hours post infection. The results were compared to a non-burned control group that underwent a sham procedure. The change in intestinal permeability was assessed by measuring FITC-Dextran 20kD presence in the blood stream at 6 hours post gavage. Bacterial burden in the blood and in intestinal samples was assessed. Intestinal inflammation and cytokine levels were also measured using ELISA.

Results:TThe experimental group underwent the burn procedure and infection had significantly higher intestinal permeability to FITC-Dextran compared to burn alone or control group mice. (p = 0.01) Oral IAP supplementation significantly reduced the gut barrier dysfunction seen in the treated group. (FITC-Dextran concentration in blood: 0.255 µg/ml in infected group compared to 0.074 µg/ml in IAP treated group, p = 0.04) Consistently, infected mice showed significantly lower serum endotoxin levels in IAP-treated group (2.69 ± 0.60 EU/ml) compared to infected and no treated group (17.51 ± 6.14 EU/ml, p = 0.003). IAP treatment reduced the bacterial load in the blood of burn and infected mice (6.3 x 105 CFU/ml of blood in infected group compared to 4.2 x 103 CFU/ml in IAP treated infected mice) and in the ileal tissue (1.4 x 106 CFU/g of ileal tissue in infection group compared to vs 2.8 x 104 CFU/g in IAP group, p = 0.049). Lastly, the burn procedure combined with infection resulted in significant levels of intestinal inflammation, as evidenced by elevated levels of TNF-a in the intestinal tissue. IAP treatment reduced TNF-a level.

Conclusion:IAP is a major regulator of the gut mucosal barrier and is able to ameliorate burn-induced gut barrier dysfunction. It appears that IAP functions through preventing intestinal inflammation. Oral IAP supplementation may represent a novel approach to prevent sepsis in burn patients.

22.08 Prospective, Randomized, Controlled, Blinded Trial: ICG Angiography In Abdominal Wall Reconstruction

Posted on January 18, 2016

C. R. Huntington1, B. A. Wormer1, S. W. Ross1, P. D. Colavita1, T. Prasad1, A. E. Lincourt1, I. Belyansky2, S. B. Getz1, B. T. Heniford1, V. A. Augenstein1 1Carolinas Medical Center,Charlotte, NC, USA 2Anne Arundel Medical Center,Annapolis, MD, USA

Introduction: Indocynanine green angiography (ICG-A) has been utilized to measure tissue perfusion during surgical reconstruction procedures and intestinal anastomosis, despite a lack of high quality evidence to support its use. While unsubstantiated, ICG-A has also been proposed to reduce complications in abdominal wall reconstruction (AWR). Two tertiary referral hernia centers conducted a prospective, randomized, controlled, blinded trial to investigate the utility of ICG-A in reducing wound complications in complex AWR.

Methods: After IRB approval, all consented patients underwent ICG-A utilizing the SPY Elite deviceTM prior to flap/skin closure after AWR. In the control group, both the Hernia Surgeon and Plastic Surgeon were blinded to ICG-A images. In the experimental group, the surgery team viewed the images and modified tissue flaps if warranted. ICG-A videos were saved and reviewed by independent, blinded surgeons to ensure correct interpretation. Outcomes included medical, surgical and wound complications and reoperation over 3 months. Groups were compared with Chi square and Wilcoxon rank sum analysis.

Results: Among 95 patients, n=49 control and n=46 experimental, preoperative characteristics were similar: age (58.3vs.56.7years,p=0.4), BMI (34.9vs. 33.6kg/m2,p=0.8), tobacco use (8.2%vs. 8.7%,p=0.9), diabetes (30.6%vs.37.0%,p=0.5), and previous hernia repair (71.4%vs.60.9%,p=0.3). The mean hernia defect was 293cm2 and mesh size 1033cm2. Operative characteristics were also equivalent, including rate of panniculectomy (69.4% vs. 58.7%,p=0.3), component separation (73.5%vs. 69.6%,p=0.6), estimated blood loss (160vs.180mL,p=0.4) and mean operative time (204vs.217minutes,p=0.4). The experimental group was more likely to have skin and subcutaneous flaps modified (37%vs.4.1%,p<0.0001). However, there was no significant difference between control vs. experimental groups in wound infection (10.2%vs.21.7%,p=0.12), skin necrosis (6.1% vs. 2.2%,p=0.3), fat necrosis (10.2% vs. 13.0%,p=0.7), overall wound-related complications (32.7% vs. 37.0%,p=0.7), reoperation rates (14.3%vs. 26.1%,p=0.7), or long-term hernia recurrence (4.1% vs. 2.2%,p=0.6) with mean follow-up of 8.3 months. When limited to significantly at-risk patients (obese, diabetic, concomitant component separation, or panniculectomy), there was no significant difference in wound-related complications between groups. Patients with hypoperfusion on ICG-A (below a threshold of 10 units) had higher rates of wound infection (28%vs.9.4%,p<0.02), however flap modification after viewing images did not improve wound-related complications in these patients (15.6%vs.12.5%,p=0.99).

Conclusion: Though intra-operative ICG-A use during complex AWR may aid in identifying patients at risk of wound infection, it did not decrease wound-related complications or reoperation rates in complex abdominal wall reconstruction. The use of ICG-A in complex AWR is not warrented in ventral hernia repair.

22.02 Hospital Resource Characteristics Associated with Improved Surgical Performance

Posted on January 18, 2016

R. S. Hoehn1, D. J. Hanseman1, D. Go1, K. Wima1, A. L. Chang1, A. E. Ertel1, S. A. Shah1, D. E. Abbott1 1University Of Cincinnati,Surgery,Cincinnati, OH, USA

Introduction: We have previously shown that inferior outcomes at safety-net hospitals are largely dependent on hospital factors. We hypothesized that variation in surgical outcomes is associated with differences in hospital financial and human resource capabilities.

Methods: The University HealthSystems Consortium Clinical Database and the American Hospital Association Annual Survey were used to examine hospitals performing 8 complex surgical procedures from 2009-2013. After excluding centers performing fewer than 10 procedures during the study period, between 66 and 166 centers were included in each procedure analysis. Hospitals in the lowest quartiles of both mortality rate and cost were characterized as high-performing (n=45), and hospitals in the highest quartiles of cost and mortality as low-performing (n=45) (Figure 1). High-performing hospitals for one procedure were never low-performing for another procedure.

Results: On average, high-performing hospitals had lower proportions of Medicaid patient days (17.5% vs 29.7%; p<0.01), higher proportions of outpatient surgery (62.9% vs 53.3%), and spent more on capital expenditures per bed ($155,710 vs $62,434; p<0.05). Also, high-performing hospitals employed more hospitalists (0.08 vs 0.04 per bed), had more privileged physicians (2.03 vs 1.25 per bed), and had more full-time equivalent personnel (8.48 vs 6.79 per bed; all p<0.05). As a result, these hospitals appeared to be more efficient; high-performers had more total admissions per bed (45.8 vs 38.4), fewer days per admission (5.20 vs 5.77), and more inpatient surgeries per bed (15.7 vs 12.6; all p<0.05).

Conclusion: High-performing hospitals invest in more human resources and demonstrate increased throughput compared to low-performing hospitals. Linking hospital reimbursement to outcomes and value-based purchasing may exacerbate, rather than improve, disparities in surgical care that currently exist.

22.03 CDK4/6 and MEK Inhibition Overcome STAT3-mediated Chemoresistance in KRAS mutant Pancreatic Cancer

Posted on January 18, 2016

J. A. Castellanos1, M. Van Saun2, N. Nagathihalli2, Y. Xiong1, C. Kasmai3, N. Merchant2 1Vanderbilt University Medical Center,Surgical Sciences,Nashville, TN, USA 2University Of Miami,Surgery,Miami, FL, USA 3Meharry Medical College,Nashville, TN, USA

Introduction:

KRAS is mutated in ~95% of pancreatic ductal adenocarcinoma (PDAC) and is the initiating genetic event in the development of this lethal cancer. Despite great efforts, we have been unable to effectively target KRAS and it is felt that Ras proteins are ‘undruggable’. We now propose a novel approach to overcome this resistance by targeting MEK and CDK 4/6, two key downstream effectors of KRAS.

Methods:

Expression of Retinoblastoma (Rb) and ERK protein levels in KRAS wild-type and mutant human PDAC cell lines was determined at baseline and with MEK (MEK162) and/or CDK4 (LEE011) inhibition. The effects of combined therapy on cell-cycle progression, colony formation, and invasion were determined. Ptf1acre/+;LSL-KrasG12D/+;Tgfbr2flox/flox (PKT) mice were treated with CDK4/6 and MEK inhibition and transcriptomic profiles were obtained by performing RNAseq on PKT mouse tumors after 2 weeks of treatment with vehicle, LEE011, MEK162, or the combination in addition to wild type Ptf1acre/+;LSL-KrasWT/WT;Tgfbr2flox/flox. Differentially expressed genes (DEGs) were identified using baySeq, DESeq2, and EdgeR in a pairwise fashion, and the lists of genes were inputted into WebGestalt (http://bioinfo.vanderbilt.edu/webgestalt/) for pathway enrichment analysis using gene ontology (GO), KEGG, and Wikipathways. Gene set enrichment analysis (GSEA) was performed to predict therapeutic targets and assess therapeutic response.

Results:

The CDK4/6 inhibitor LEE011 effectively inhibits phosphorylation of Rb in PDAC cell lines. Combined inhibition of CDK4/6 and MEK decreases pRB and pMAPK expression, delays cell cycle progression, decreases colony formation, and decreases invasion in the KRAS mutant cell lines. GSEA of wild type mouse pancreata vs. control mouse tumor revealed significant upregulation in the MYC targeted pathway, a key regulator of Ras mediated therapeutic resistance. Overall survival (OS) in PKT mice treated with MEK162 or LEE011 alone was not significantly different compared to controls, but mice receiving combined CDK4/6 and MEK inhibition exhibited a 400% increase in OS (59 vs. 251.5 days). Combined CDK4/6 and MEK inhibition significantly downregulated IL-6/STAT3, Kras, EMT, and IL-2/STAT5 related pathways compared with control mice. Combination treated mice also had significant downregulation in IL-6/STAT3 signaling compared to MEK treated mice.

Conclusions:

Targeting KRAS in PDAC through downstream inhibition of MEK remains ineffective due to upregulation of STAT3 signaling. Targeting two key downstream effectors of KRAS signaling with combined inhibition of CDK4/6 and MEK overcomes STAT3 mediated chemoresistance and results in significantly enhanced therapeutic efficacy in the aggressive PKT genetic mouse model of PDAC.

22.04 Does AAS Membership Improve the Academic Productivity of its Constituent Members?

Posted on January 18, 2016

N. Valsangkar1, D. Milgrom1, P. J. Martin1, J. Parrett1, M. M. Joshi1, T. A. Zimmers1, L. G. Koniaris1 1Indiana University,Department Of Surgery,Indianapolis, IN, USA

Introduction: To evaluate the relationship between Association for Academic Surgery (AAS) membership and academic productivity for surgical faculty in the United States.

Methods: Academic metrics including numbers of publications, citations, and NIH funding history were determined for 4,015 surgical faculty at the top-55 NIH funded departments of surgery, using SCOPUS, NIH RePORT, and the Grantome online databases. AAS membership was determined from a past and present membership registry.

Results: Overall, 912 (22.7%) of all 4,015 surgical faculty examined were AAS members. Median publications (P) and citations (C) (±SD) were: 54±96 and 985±3321 for AAS members compared with 31±92 and 528±3001 for non-members (p < 0.001). The greater academic productivity of AAS members was observed across all subspecialties and was especially pronounced for assistant and associate professors [Table 1]. AAS membership was also associated with significantly increased rates of NIH funding both for training grants such as those of the K-series and R-awards. Among AAS members, 13.4% had current or former R01/P01/U01grants compared with only 9% of AAS non-members (p < 0.05). AAS members also had two times as many current or previous smaller NIH awards (13.9% vs. 6.2%, p < 0.05) Analysis of AAS membership by subspecialty revealed that AAS members were most commonly general surgery faculty (58%). AAS membership was lowest in cardiothoracic surgery, plastic surgery and among Ph.D. faculty working in departments of surgery; however, AAS members from these subspecialties had significantly greater academic output compared with AAS non-members as well.

Conclusion: AAS membership is associated with substantially better academic metrics and NIH funding rates for junior and midlevel surgical faculty across all subspecialties. Individuals and academic departments interested in improving publications and funding should encourage participation in the AAS. Causes for the low membership rate for certain subspecialties in AAS should be examined and addressed to potentially improve academic metrics within these subspecialties.

« Previous Page
Next Page »