Fax: (615) 322-7852
The combination of alisertib, an investigational Aurora kinase A inhibitor, and docetaxel promotes cell death and reduces tumor growth in preclinical cell models of upper gastrointestinal adenocarcinomas†
Article first published online: 12 SEP 2012
Copyright © 2012 American Cancer Society
Volume 119, Issue 4, pages 904–914, 15 February 2013
How to Cite
Sehdev, V., Katsha, A., Ecsedy, J., Zaika, A., Belkhiri, A. and El-Rifai, W. (2013), The combination of alisertib, an investigational Aurora kinase A inhibitor, and docetaxel promotes cell death and reduces tumor growth in preclinical cell models of upper gastrointestinal adenocarcinomas. Cancer, 119: 904–914. doi: 10.1002/cncr.27801
The contents of this work are solely the responsibility of the authors and do not necessarily represent the official views of the National Cancer Institute or Vanderbilt University.
- Issue published online: 4 FEB 2013
- Article first published online: 12 SEP 2012
- Manuscript Accepted: 1 AUG 2012
- Manuscript Revised: 2 JUL 2012
- Manuscript Received: 1 MAY 2012
- Aurora kinase A;
Upper gastrointestinal adenocarcinomas (UGCs) respond poorly to current chemotherapeutic regimes. The authors and others have previously reported frequent Aurora kinase A (AURKA) gene amplification and mRNA and protein overexpression in UGCs. The objective of the current study was to determine the therapeutic potential of alisertib (MLN8237) alone and in combination with docetaxel in UGCs.
After treatment with alisertib and/or docetaxel, clonogenic cell survival, cell cycle analyses, Western blot analyses, and tumor xenograft growth assays were carried out to measure cell survival, cell cycle progression, apoptotic protein expression, and tumor xenograft volumes, respectively.
By using the AGS, FLO-1, and OE33 UGC cell lines, which have constitutive AURKA overexpression and variable tumor protein 53 (p53) status, significantly enhanced inhibition of cancer cell survival was observed with alisertib and docetaxel treatment in combination (P < .001), compared with single-agent treatments. Cell cycle analyses, after 48 hours of treatment with alisertib, produced a significant increase in the percentage of polyploidy in UGC cells (P < .01) that was further enhanced by docetaxel (P < .001). In addition, an increase in the percentage of cells in sub-G1-phase observed with alisertib (P < .01) was significantly enhanced with the combination treatment (P < .001). Western blot analysis demonstrated higher induction of cleaved caspase 3 protein expression with the combined treatment compared with single-agent treatments. In addition, FLO-1 and OE33 cell xenograft models demonstrated enhanced antitumor activity for the alisertib and docetaxel combination compared with single-agent treatments (P < .001).
The current study demonstrated that alisertib combined with docetaxel can mediate a better therapeutic outcome in UGC cell lines. Cancer 2013. © 2012 American Cancer Society.
Upper gastrointestinal adenocarcinomas (UGCs) (ie, adenocarcinomas of the stomach and esophagus) are associated with poor patient survival rate because of inherent resistance to current therapeutic regimens.1,2 Global epidemiologic data indicate that approximately 1.4 million new UGCs are diagnosed annually, resulting in approximately 1.1 million deaths.3 Over the past several decades, the incidence rates for distal gastric cancers have declined; however, incidence rates for adenocarcinoma of the gastric cardia, gastroesophageal (GE) junction, and esophagus continue to rise.3,4
Multiple prosurvival and drug-resistant genes, such as the epidermal growth factor receptor (EGFR), human epidermal growth factor receptor 2 (HER-2), v-Ki-ras2 Kirsten rat sarcoma viral oncogene homolog (K-RAS), and Aurora kinase A (AURKA), mediate oncogenic signaling pathways in gastric and esophageal adenocarcinomas.5-9 Despite the availability of several adjuvant and neoadjuvant chemotherapeutic treatment strategies, the survival rate for patients with UGCs have only marginally improved.10 Therefore, laboratory investigations and preclinical studies specifically aimed at developing novel targeted therapies and chemotherapeutic combinations with potent antitumor activity are desperately needed to treat patients with UGCs.
We and others previously reported the amplification of the region of band 13 on the long arm of chromosome 20 (20q13) in UGC.11,12 The 20q13 chromosomal region harbors the AURKA gene, which is frequently amplified and/or overexpressed in several malignancies, including cancers of the bladder, breast, colon, liver, ovaries, pancreas, stomach, and esophagus.8,13 AURKA is a key cell cycle regulator that is critical for mitotic events.14,15 However, when overexpressed, AURKA is a bona fide oncogene and results in genetic instability, dedifferentiated morphology, and a poor prognosis in patients with UGCs.6,16 The overexpression of AURKA promotes cancer cell growth and resistance to chemotherapy by up-regulating oncogenic signaling pathways and suppressing cell-death mechanisms, respectively.13 In addition, AURKA overexpression induces growth-promoting and survival-promoting oncogenic signaling pathways, such as the phosphoinositide 3-kinase/protein kinase B (PI3K/AKT) and β-catenin pathways, in UGC cancer cells.13 The docetaxel, cisplatin, and 5-fluorouracil (DCF) chemotherapy regimen is 1 of the most efficacious chemotherapeutic regimens against advanced gastric cancer.17 It is noteworthy that AURKA overexpression reportedly mediates resistance against paclitaxel and cisplatin-induced cell death.18,19 The cell-death mechanisms, which are regulated by the tumor protein 53 (p53) family of proapoptotic proteins, are activated in cancer cells after treatment with chemotherapeutic agents. However, mutations in the p53 function are frequently observed in UGCs. In p53-mutant cancers, the related tumor protein 73 (p73) can mediate p53-like apoptotic functions and activate apoptotic pathways after treatment with chemotherapeutic agents.20 Also noteworthy is the fact that AURKA overexpression in cancer suppresses p53 and p73 protein expression and function.21,22 These findings suggest that p53-mutant UGCs with constitutively higher levels of AURKA expression can respond poorly to chemotherapy. Therefore, given the poor response of UGCs to current therapeutic regimens; novel therapeutic strategies that take into account the molecular make-up of tumors to activate cell death response are critically needed to combat UGCs.
Alisertib (MLN8237) is an investigational small molecule inhibitor developed by Millennium Pharmaceuticals, Inc. (Cambridge, Mass.) that has demonstrated the ability to selectively inhibit AURKA and thereby induce cell cycle arrest, aneuploidy, polyploidy, mitotic catastrophe, and cell death.8, 23 Currently, alisertib is being tested in various phase 1, 2, and 3 clinical trials for advanced solid tumors and hematologic malignancies (http://clinicaltrials.gov/; Accessed August 26, 2012). In addition, multiple clinical trials have indicated that docetaxel and its combination with cisplatin and 5-fluorouracil have significant antitumor activity in UGCs.17 Docetaxel is a microtubule-polymerizing chemotherapeutic agent that disrupts microtubule dynamics by binding to the β-subunit of tubulin and promoting its polymerization.24 Consequently, docetaxel causes cell cycle arrest in mitosis and subsequently induces apoptosis. In the current study, we investigated the potential therapeutic benefit of alisertib alone and in combination with docetaxel using in vitro and in vivo cell models of UGC with variable p53 status. We hypothesized that the alisertib and docetaxel combination treatment would cause enhanced cell cycle arrest, resulting in polyploidy and subsequent induction of apoptosis in UGC cancer cells, irrespective of their p53 status.
MATERIALS AND METHODS
Cell Culture and Pharmacologic Reagents
The AGS (p53 wild type) gastric adenocarcinoma cell line and the FLO-1/OE33 (p53 mutant) esophageal adenocarcinoma cell lines were maintained as a monolayer culture in Dulbecco modified Eagle medium (DMEM) (Gibco, Carlsbad, Calif) cell culture medium supplemented with 10% (volume/volume) fetal bovine serum (FBS) (Gibco).25 All cell lines were evaluated weekly to ascertain conformity to the appropriate in vitro morphologic characteristics.26 MLN8237 (alisertib) was provided by Millennium Pharmaceuticals, Inc., and alisertib stock solutions for in vitro and in vivo studies were prepared according to our previously reported methods.8 Docetaxel (Sanofi Aventis, Bridgewater, NJ) stock solution (11.6 mM) prepared in 13% ethanol (volume/volume) was provided by the TVC Outpatient Pharmacy at Vanderbilt University Medical Center (Nashville, Tenn).
Clonogenic Cell Survival Assay
AGS, FLO-1 and OE33 cells were seeded at 5000 cells per well, respectively, onto 6-well plates overnight and then were treated with various concentrations of alisertib (0.25 μM, 0.5 μM, 1.0 μM, 2.0 μM, and 5.0 μM) for 24 hours. After treatment, UGC cell survival was determined according to our previously reported protocol.8 Briefly, after treatment, the cells were incubated in drug-free cell culture medium for 10 days. Subsequently, the cells were fixed with 2% paraformaldehyde solution, stained with crystal violet dye solution, and cell survival was quantified by measuring the dye signal in each well using ImageJ analysis software (National Institutes of Health, Bethesda, Md). In addition, we selected an alisertib dose close to the 50% inhibitory concentration (0.5 μM) and treated the cells with alisertib (0.5 μM) and/or with docetaxel (0.5 nM, 1.0 nM, or 5.0 nM) for 24 hours.
Cell Cycle Analysis
AGS and FLO-1 cells were treated with alisertib (0.1 μM) and/or docetaxel (0.5 nM) in cell culture medium (2.5% FBS) for 24 hours and 48 hours, respectively. OE33 cells were treated with alisertib (0.5 μM) and/or docetaxel (1.0 nM) in cell culture medium (2.5% FBS) for 24 hours and 48 hours, respectively. After treatment, supernatant media was collected, and adherent cells were trypsinized. The supernatant and trypsinized cells were centrifuged together at ×2000g at 4°C for 10 minutes. Then, the cells were resuspended in 1 mL propidium iodide (PI) solution (PI 50 μg/mL and RNase 1 μg/mL in 1 × phosphate-buffered saline) and incubated at room temperature in the dark for 30 minutes. Subsequently, the cells were analyzed with the BD LSR III flow cytometer (BD Biosciences, San Jose, Calif), and the data were processed with BD FACS Diva software (BD Biosciences) at the VMC Flow Cytometry Shared Resource, Vanderbilt Ingram Cancer Center.
Western Blot Analysis
AGS, FLO-1, and OE33 cancer cells were plated overnight at 30% confluence in cell culture medium (10% FBS). AGS cells were treated with alisertib (0.25 μM) and/or docetaxel (1.0 nM), FLO-1 cells were treated with alisertib (0.1 μM) and/or docetaxel (0.5 nM), and OE33 cells were treated with alisertib (0.5 μM) and/or docetaxel (1.0 nM) for 48 hours in cell culture medium supplemented with 2.5% FBS. After treatment, cell lysates were prepared and evaluated for total and phosphorylated (p-) proteins; p-AURKA (threonine 288), AURKA, cleaved caspase 3, and β-actin (Cell Signaling Technology, Beverly, Mass), according to standard protocols.27
In Vivo Tumor Xenograft Inhibition
Four million FLO-1 or OE33 cells suspended in a 200-μL DMEM-Matrigel mixture (50% DMEM supplemented with 10% FBS and 50% Matrigel) were injected into the flank regions of female athymic nude-Foxn1 nu/nu mice (Harlan Laboratories Inc., Indianapolis, Ind). The tumors were allowed to grow until they measured 200 mm3 in size before the treatment was started with daily alisertib (30 mg/kg, orally) and/or once-weekly docetaxel (10 mg/kg as an intraperitoneal injection) for 3 weeks. Tumor xenografts were measured every alternate day, and tumor size was calculated according to the following formula: Tvol = L×W2 × 0.5, in which Tvol is tumor volume, L is tumor length, and W is tumor width.23
After 21 days of animal treatment, the tumors were isolated, and an immunohistochemical analysis was carried out to measure Ki-67 and cleaved caspase 3 protein expression levels, as previously reported.8 Protein expression was scored using a composite expression score (CES) that was determined by using the formula “CES = 4(intensity − 1) + frequency,” as described previously, with intensity measured on a scale from 0 to 3 and frequency measured on a scale from 0 to 4.28
Data are presented as means ± standard error of the mean (SEM). All in vitro experiments were performed in triplicate. A 1-way analysis of variance (ANOVA) with Tukey post hoc analysis was used to demonstrate statistical differences between control groups and treatment groups at the treatment endpoints. A 2way ANOVA with Bonferroni post hoc analysis was used to demonstrate statistical differences between various treatment groups and cell cycle stages. For tumor xenograft data, a 2-way ANOVA (time point-matched analysis) with Bonferroni post-test was used to compare the “mean tumor size” of a treatment group on any given treatment day with the “mean tumor size” of another other treatment groups on the corresponding treatment day. All statistical analyses described above were carried out using GraphPad Prism 5 software (GraphPad Software Inc., La Jolla, Calif). All P values ≤ .05 were considered statistically significant and are marked in the figures, in which a single asterisk indicates P < .05, and double asterisks indicate P < .01.
Alisertib Significantly Enhanced Docetaxel-Mediated Inhibition of Cell Survival
Both FLO1 and OE33 cell lines exhibited gene amplification and overexpression of AURKA at the mRNA and protein levels.8, 29 Similarly, AGS cells exhibited an increase in AURKA DNA copy number (2.26-fold) and mRNA level (4.98-fold; data not shown). Therefore, these cell models mimic the previously reported in vivo data on AURKA overexpression in primary UGCs.30 The clonogenic cell-survival assay data indicated that alisertib (0.5 μM) or docetaxel (1.0 nM) as single-agent treatments decreased the percentage of AGS cells that survived (alisertib 0.5 μM: 45.5% ± 4.6% AGS cell survival [mean ± SEM]; P < .01; docetaxel 1.0 nM, 53.6% ± 1.8% AGS cell survival; P < .01) (Fig. 1A), FLO-1 cells (alisertib 0.5 μM: 45.5% ± 4.6% FLO-1 cell survival; P < .01; docetaxel 1.0 nM: 70.1% ± 5.6% FLO-1 cell survival; P < .05) (Fig. 1B), and OE33 cells (alisertib 0.5 μM: 45.5% ± 4.6% OE33 cell survival; P < .01; docetaxel 1.0 nM: 32.4% ± 3.5% OE33 cell survival; P < .01) (Fig. 1C). Treatment with the alisertib (0.5 μM) and docetaxel (1.0 nM) combination led to a significantly enhanced inhibition of the percentage of surviving AGS cells (alisertib 0.5 μM plus docetaxel 1.0 nM: 5.5% ± 0.7% AGS cell survival; P < .01) (Fig. 1A), FLO-1 cells (alisertib 0.5 μM plus docetaxel 1.0 nM: 4.5% ± 0.5% FLO-1 cell survival; P < .01) (Fig. 1B), and OE33 cells (alisertib 0.5 μM plus docetaxel 1.0 nM: 13% ± 1.8% OE33 cell survival; P < .01) (Fig. 1C). These results suggest that the combination of alisertib with docetaxel may have a significantly greater inhibitory effect on UGC cell survival.
Alisertib-Enhanced Docetaxel Induced Polyploidy and Apoptosis in Upper Gastrointestinal Adenocarcinoma Cells
By using the AGS, FLO-1, and OE33 cell lines as in vitro models of UGC to study the effect of alisertib and docetaxel on cell cycle progression, treatment with alisertib alone or in combination with docetaxel for 24 hours significantly reduced the percentage of cells in G1-phase and S-phase and induced a significant delay in the transition from G2-phase to M-phase in AGS and OE33 cells (Figs. 2A, 3C). The 24-hour treatment with alisertib alone had a similar effect on G1-phase, S-phase, and the G2-phase to M-phase transition in FLO-1 cells; however, this effect was significantly more pronounced after 24 hours of combined treatment with combined alisertib and docetaxel (Fig. 3A). In addition, treatment with alisertib alone for 24 hours significantly increased the percentage of polyploid cells, which was enhanced further by combined alisertib and docetaxel treatment in AGS and FLO-1 cells (Figs. 2A, 3A). At the 48-hour time point, FLO-1 cells with that were treated with either alisertib or docetaxel as a single-agent demonstrated an increase in the percentage of cells in sub-G1-phase (P < .05) (Fig. 3B). This effect was significantly enhanced in FLO-1 cells that received the combined treatment (P < .01) (Fig. 3B). The cell cycle analyses indicate that FLO-1 cells were more sensitive to alisertib and/or docetaxel treatments, as evidenced by an increase in the number of cells in sub-G1-phase after 24-hour and 48-hour treatment. In addition, 48-hour treatment with alisertib and docetaxel enhanced polyploidy in AGS and OE33 UGC cells. These findings are in agreement with previously published reports, in which investigators observed that docetaxel-induced microtubule stabilization impaired mitosis, generating aneuploid and tetraploid cells that subsequently underwent apoptotic cell death.24, 31, 32 Our data suggest that the combination treatment promotes polyploidy early on and, depending on the cell line, polyploidy likely leads to cell death (sub-G1) either early on (FLO-1 cells) or at later time points (AGS and OE33 cells). These findings provide a plausible explanation and support for the clonogenic cell survival assay results that measured long-term cell viability. We also observed that the treatment with either alisertib or docetaxel alone induced the expression of cleaved caspase 3, a common apoptosis marker, a finding that was enhanced significantly after combined treatment with alisertib and docetaxel for 48 hours (Fig. 4). These results support our hypothesis that alisertib can significantly enhance docetaxel-induced apoptosis in UGCs.
Alisertib and Docetaxel Combination Treatment Exhibits Enhanced Antitumor Activity In Vivo
The in vitro results described above prompted us to determine the antitumor activity of alisertib and/or docetaxel treatments in UGC xenograft mouse models. The in vivo antitumor activity analysis demonstrated that treatment with alisertib or docetaxel alone significantly reduced the percentage tumor volume of FLO-1 cells (alisertib: 76.6% ± 6.2%; P < .01; docetaxel: 128.73% ± 15.1%; P < .01) and OE33 cells (alisertib: 101.4% ± 5.6%; P < .01; docetaxel: 44.1% ± 3.1%; P < .01). Compared with these single-agent treatments, combined treatment with alisertib and docetaxel led to an enhanced reduction in the percentage tumor volume (FLO-1 cells: 12.6% ± 1.7%; P < .01; OE33 cells: 12.9% ± 1.0%; P < .01). These results are summarized in Figure 5A,B.
In addition, the immunohistochemical analysis of FLO-1 and OE33 tumor xenografts after treatment with alisertib and/or docetaxel (day 21) demonstrated a reduction in the number of cells that were positive for Ki-67 and an increase in the number of cells that were positive for cleaved caspase 3: Data are provided in Figure 6 for FLO-1 xenografts, and similar results were obtained in OE33 xenografts. In concordance with the tumor growth results (Fig. 5), these immunostaining patterns were more significant with the combined treatment (P < .01) than with the single-agent treatments (P < .05) (Fig. 6). Therefore, the in vivo data indicate that combined treatment with alisertib and docetaxel has enhanced antitumor activity in UGC tumor xenograft models.
Despite novel therapeutic advancements, improvement in the survival rate of patients with UGC has been marginal, suggesting the presence of unique, active, intrinsic mechanisms that impart resistance to chemotherapeutic agents in UGCs.33, 34 AURKA is frequently overexpressed and/or amplified in various cancers, including UGCs.8, 13 Recent reports suggest that AURKA can induce chemotherapeutic resistance and regulate several key signaling pathways in cancer cells, suggesting its role as a central node in cancer cell signaling.13 Docetaxel has demonstrated significant in vitro and in vivo antitumor activity against a variety of UGC cell lines.35
In the current study, we used UGC cell lines to determine the therapeutic response of the recently developed AURKA selective inhibitor alisertib as a single agent and in combination with docetaxel. Although previous in vitro studies with Aurora kinase inhibitors have demonstrated their antitumor activity in combination with other chemotherapeutic agents, such as cisplatin, docetaxel, nilotinib, and vorinostat,8,36-38 alisertib, an investigational small-molecule AURKA inhibitor currently in clinical development, has not been tested in combination with docetaxel in gastrointestinal cancer models. In this regard, our current results suggest a potential therapeutic benefit of combined treatment with alisertib and docetaxel.
The p53 gene is frequently mutated in various cancers in which p53-mutant tumors exhibit inherent resistance to several chemotherapeutic drugs.39 This is of particular significance in UGC therapeutics; because, in addition to AURKA overexpression, a high frequency of defective p53 signaling caused by mutation or deletion is observed in UGCs, presenting a formidable clinical challenge.40 In this study, we used p53-mutant (FLO-1 and OE33) and wild-type (AGS) cell lines and obtained similar results. It is noteworthy that both in vitro and in vivo models suggested a promising therapeutic potential for the alisertib and docetaxel combination, as indicated by suppressed cell survival in vitro (P < .001) and significant regression of tumor growth in vivo (P < .001). On the basis of our findings, we suggest that AURKA-targeted therapy, either alone or in combination with docetaxel, is effective independent of p53 status. Our data indicate that the therapeutic effect is largely because of the induction of aberrant mitosis, leading to polyploidy and subsequent apoptosis. However, others factors, such as suppressed proliferation and nonapoptotic forms of cell death, also may be occurring. These findings are timely given the finding that alisertib is being tested actively in various phase 1, 2, and 3 clinical trials (http://clinicaltrials.gov/; [accessed August 26, 2012]).
AURKA is a serine/threonine kinase that facilitates accurate mitosis by regulating vital cell cycle events during various stages of mitosis.14, 15 It has been demonstrated that AURKA inhibition induces aneuploidy, polyploidy, and mitotic catastrophe in cells.36 After treatment with alisertib, we observed a significant increase in the percentage of cells with polyploidy. This suggests that mitotic catastrophe remains 1 of the predominant functions of alisertib, a finding that is in agreement with previously published data.8, 36 Docetaxel-based combination chemotherapeutic regimens like DCF are widely used for the treatment of advanced gastric cancers.17 Docetaxel-induced defects in spindle formation and function (tubulin polymerization) activate the spindle assembly checkpoint (SAC) and subsequently result in aneuploidy, polyploidy, and/or apoptosis.24 It is worth noting previous reports that AURKA overexpression was able to override the SAC and impart resistance to paclitaxel in HEK-293 cells, suggesting that AURKA overexpression is critical for resistance to cell cycle inhibitors.19 Treatment with docetaxel increased the percentage of cells in the sub-G1-phase, indicating late-stage cell death, which conforms to the finding that docetaxel-induced spindle defects activate SAC, which subsequently activates apoptotic pathways.24 Treatment with alisertib alone and in combination with docetaxel induced G2-M-phase arrest in vitro. It is known that AURKA regulates G2-phase to M-phase transitions, and its inhibition should result in G2/M-phase arrest.41 At low treatment concentrations, docetaxel-induced SAC is transient and weak, an effect that can be easily overridden by overexpressed AURKA, as previously reported in HeLa cells.19 However, alisertib-mediated specific inhibition of AURKA may sensitize AURKA-overexpressing cells to docetaxel, resulting in prolonged activation of SAC that is subsequently translated into accelerated mitotic slippage, polyploidy, and cell death.31 After a short, 48-hour treatment, our results demonstrated an increase in the percentage of polyploid cells in AGS and OE33 cells, whereas an increase in apoptotic cells was observed in FLO-1 cells. These data suggest that FLO-1 cells are relatively more sensitive to AURKA inhibition compared with AGS and OE33 cells when treated with alisertib alone, an effect that is further enhanced by docetaxel. Although treatment of AGS and OE33 cells with docetaxel alone for 48 hours increased the percentage of apoptotic sub-G1-phase cells, even higher percentages of polyploid cells were always observed after treatment with combined alisertib and docetaxel. It is well documented that combination chemotherapy is a common and effective therapeutic approach in the treatment of UGCs.17 In our study, combined alisertib and docetaxel treatment induced a higher percentage of polyploidy that, subsequently, resulted in increased apoptosis, as indicated by the significantly enhanced expression of cleaved caspase 3 protein levels in AGS, FLO-1, and OE33 cells. In addition, compared with single-agent treatments, the combination treatment significantly reduced Ki-67 levels and enhanced cleaved caspase 3 protein levels in tumor xenografts, providing additional evidence of the improved antitumor activity of this regimen.
In conclusion, the combination of alisertib and docetaxel results in significantly enhanced antitumor activity in cell line models, possibly mediated by apoptotic pathways induced after the activation of SAC. Thus, as clinical trials for alisertib are being actively conducted, our current findings provide a credible rationale for evaluating AURKA-targeted therapy in combination with docetaxel as a therapeutic approach for the treatment of UGCs.
This study was supported by grants from the National Institute of Health; R01CA131225 (Dr. El-Rifai), VICTR pilot project support from Vanderbilt CTSA grant UL1 RR024975; Vanderbilt Specialized Programs of Research Excellence (SPORE) in Gastrointestinal Cancer (P50 CA95103), Vanderbilt Ingram Cancer Center (P30 CA68485) and the Vanderbilt Digestive Disease Research Center (DK058404).
CONFLICT OF INTEREST DISCLOSURES
The authors made no disclosures.