Transition from normal duodenal to adenoma in FAP
In the adenoma-normal comparison, 19 protein-coding DEGs were identified. Neoplastic processes involving 8 representative DEGs are shown in Table 4.
Enterocyte dedifferentiation can be determined by examining expression along the crypt-villus axis and in the Caco-2 cell line, which spontaneously differentiates into mature small intestine (17). In adenoma-normal, we found downregulation of DEGs involved in brush-border metabolism. Among these, expression of LCT (18) and TMPRSS15 (19) increases during the crypt-villus axis, while expression of LCT (20) increases with Caco-2 differentiation. We observed downregulation of APOA4 and APOB, which encode apolipoproteins whose expression increases during Caco-2 cell differentiation (21). The downregulation of these brush border and lipid metabolism DEGs indicates enterocyte dedifferentiation. In adenoma-normal, we found upregulation of SLC12A2, which encodes the basolateral ion transporter NKCC1. NKCC1 expression decreases in the crypt-villus axis (22), suggesting that its upregulation further implicates enterocyte dedifferentiation.
During the transition from normal to adenoma, certain DEGs implicate the Warburg effect, in which proliferating tumor cells prefer glycolysis over gluconeogenesis and aerobic respiration. We found downregulation of ALDOB, which encodes a gluconeogenesis enzyme. Of note, Aldob is downregulated in adenomas of APCMin/+ mice (14).
Decreased production of all-trans retinoic acid
In adenoma-normal, RBP2 was downregulated. In small intestine, the retinol binding protein 2 (RBP2) mediates Vitamin A (retinol) uptake. Retinol is oxidized to all-trans-retinaldehyde by alcohol dehydrogenase and then to all-trans retinoic acid (ATRA) by aldehyde dehydrogenase (23). ATRA suppresses tumorigenesis in part by blocking induction of COX-2 (24). Therefore, downregulation of RBP2 indicates that decreased ATRA production may play a role in the transition of normal duodenum to adenoma. Of note, decreased ATRA production is implicated in APCMin/+ mice adenomas, which show downregulation of Adh1 (14).
Impaired reactive oxygen species/carcinogen defense
In adenoma-normal, we found downregulation of GSTA1 and GSTA2, which encode members of the α class of gluathione-S-transferase enzymes. These enzymes have glutathione peroxidase activity, which protects cells from reactive oxygen species (ROS). GSTA1 downregulation is seen in normal duodenum from patients with FAP compared to non-FAP controls (25). Furthermore, downregulation of Gsta4 is seen in intestinal adenomas of APCMin/+ mice (14). This suggests that diminished antioxidant defense plays a role in duodenal adenoma development in FAP.
Transition from duodenal adenoma to cancer in FAP
In the cancer-adenoma comparison, there were 26 protein-coding DEGs. Neoplastic processes involving 8 representative DEGs are shown in Table 5.
Goblet and Paneth cell dedifferentiation
In cancer-adenoma, we found downregulation of CLCA1, which encodes a chloride channel expressed in intestinal goblet cells. CLCA1 expression increases with Caco-2 differentiation (26), suggesting that its downregulation may indicate goblet cell dedifferentiation.
In cancer-adenoma, we also found downregulation of DEFA5 and DEFA6, which encode α-defensins and are primarily expressed in Paneth cells of the small intestine (27). Both are upregulated in colon adenomas and cancer compared to normal tissue (28), indicating abnormal Paneth cell differentiation in colon tumors (29). Given these findings, DEFA5/6 downregulation may indicate Paneth dedifferentiation during the evolution of duodenal adenoma to cancer in FAP.
Decreased production of ATRA
In cancer-adenoma, we found downregulation of ADH1C, which again implicates decreased ATRA production in the progression of duodenal neoplasia in FAP.
Increased tumor invasiveness
In cancer-adenoma, upregulation of several DEGs involved in cell adhesion and extracellular matrix interactions was observed, including upregulation of COL12A1, which encodes for type XII collagen and is involved in the desmoplastic reaction between cancer cells and associated fibroblasts, which drives colon cancer metastases (30). Cancer tissue also exhibited upregulation of FN1 and SPP1. FN1 encodes fibronectin 1, which promotes cell proliferation and invasion by interacting with α5β1 integrin (31). SPP1 encodes osteopontin (OPN), which mediates cell migration partially through interactions with αvβ3 integrin (32). In cancer-adenoma, POSTN and IL8, which encode the pro-angiogenesis factors periostin (33) and interleukin-8 (34), respectively, were also upregulated.
DEGs with predictive potential in FAP
Among our representative DEGs, several have potential as tissue or serum biomarkers for progression of duodenal neoplasia.
Potential tissue biomarkers for duodenal cancer in FAP
We identified 13 protein-coding DEGs that distinguished FAP case and FAP control adenomas, all of which were downregulated in FAP cases (Supplemental Table 1, see Supplementary Digital Content 1, http://links.lww.com/CTG/A51). Of these DEGs, CLCA1, ADH1C, and ANXA10 have particular significance as potential tissue biomarkers.
CLCA1 encodes a chloride channel expressed in goblet cells, whereas ADH1C encodes an alcohol dehydrogenase enzyme involved in retinol oxidation. Studies have shown CLCA1 downregulation in CRC (26) and ADH1C downregulation in gastric cancer (23). In this study, CLCA1 and ADH1C are downregulated in cancer compared to adenoma and in adenoma from FAP cases compared to FAP controls, indicating that downregulation of these DEGs within adenomas may indicate increased likelihood of neoplastic progression.
ANXA10 encodes annexin 10, a calcium- and phospholipid-binding protein normally expressed in gastric mucosa that inhibits tumorigenesis by causing growth suppression and stimulation of apoptosis (35). Decreased ANXA10 expression is seen in gastric cancer (35). In this study, ANXA10 expression followed a unique pattern. Within FAP cases, ANXA10 does not differ in cancer-normal but is upregulated in adenoma-normal (FC 2.3, FDR 0.30) and downregulated in cancer-adenoma (FC −1.5, FDR < 0.10) comparisons. Furthermore, ANXA10 is significantly downregulated in adenoma from FAP cases compared to FAP controls (FC −2.1, FDR < 0.10). Considering the aforementioned roles of ANXA10, it is possible that the upregulation of ANXA10 in duodenal adenomas indicates a protective “gastric programming.” Downregulation during the transition from FAP control to FAP case adenoma and from FAP case adenoma to cancer may reflect a loss in the tumor suppressive function of ANXA10. Given these findings, determining tissue expression of ANXA10 may predict the likelihood that a duodenal adenoma progresses to cancer in FAP.
Potential serum biomarkers for duodenal cancer in FAP
Among the DEGs identified, SPP1 and CEACAM5 have potential as serum biomarkers for duodenal cancer in FAP.
SPP1 encodes OPN. SPP1 expression is 27-fold higher in sporadic ampullary cancer compared to normal duodenum and serum OPN progressively increases from healthy controls to patients with ampullary adenoma to patients with sporadic ampullary cancer (36). CEACAM5 encodes membrane-bound and secreted carcinoembryonic antigen (CEA). For CRC, serum CEA is an independent prognostic factor for recurrence and survival after curative resection (37). In this study, CEACAM5 is the only DEG upregulated in the adenoma-normal and cancer-adenoma comparisons (Table 3). Together, these findings suggest that serum OPN and CEA may help determine development of duodenal polyposis and progression to duodenal cancer in FAP.
DEGs with therapeutic potential in FAP
Certain DEGs may have significance in existing and novel chemopreventive therapies for duodenal polyposis.
Both celecoxib (5) and the sulindac/erlotinib combination (38) decrease duodenal polyp burden in FAP. We found upregulation of SPP1 in cancer-normal and cancer-adenoma comparisons. SPP1 is a Wnt/β-catenin target gene (39) and administration of the COX-2 inhibitor parecoxib to APC[INCREMENT]14/+ mice, which display a FAP phenotype, downregulates SPP1 by inhibition of Wnt/β-catenin while decreasing intestinal tumor load and mice morality (40). OPN is an upstream activator of the EGFR pathway (41). In non-small-cell lung cancer cell lines, the radiosensitizing effect of erlotinib is abolished after OPN depletion (42). Given its role as a target of PGE2 signaling and an activator of EGFR signaling, tissue levels of OPN may be of particular significance in predicting response to the sulindac/erlotinib combination regimen.
CEACAM6 upregulation in the adenoma-normal and cancer-normal comparisons was also noted. CEACAM6 encodes a membrane-bound cell adhesion molecule, which confers resistance to anoikis, the apoptosis induced by lack of correct cell/extracellular matrix attachment (43). This allows for cancer cell survival and invasiveness. Accordingly, CEACAM6 overexpression is seen in CRC (44) and pancreatic cancer (45). In a murine model of pancreatic cancer, administration of a CEACAM6-specific monoclonal antibody conjugated with immunotoxin increases tumor apoptosis and decreases tumor growth (46). In nonhuman primates, this antibody-drug conjugate has minimal toxicity, with a dose-dependent, reversible neutropenia (47). These findings implicate CEACAM6 as a potential novel therapeutic target in the treatment of duodenal polyposis in FAP.
We determined expression levels of candidate genes SPP1, CEACAM5, SI, APOA4, and ANXA10 by PCR. In certain samples from certain patients, PCR could not be successfully performed and expression levels were undefined (Supplemental Table 3, see Supplementary Digital Content 1, http://links.lww.com/CTG/A51).
For cancer-normal and cancer-adenoma comparisons, the sample size for comparison of SI expression was very low (n = 5). Therefore, we decided to exclude PCR results from SI expression. For the remaining genes, FAP cases 10 and 11 consistently did not yield results on PCR. Both FAP cases 10 and 11 had samples preserved with Hollande's fixative (Supplemental Table 4, see Supplementary Digital Content 1, http://links.lww.com/CTG/A51), which can affect RNA yield and quality (48) and therefore may explain the failure of PCR expression analysis in these samples.
Table 6 shows HTA and PCR results. For every comparison, direction of FC mirrored HTA findings. Specific magnitude of FC and statistical significance is detailed below.
- SPP1: Gene expression differences in SPP1 was fully verified by PCR.
- CEACAM5: PCR analyses verified no difference in adenoma-adenoma comparison. As in HTA analysis, PCR analysis showed upregulation in cancer-normal and cancer-adenoma, but each comparison had a trend toward significance.
- APOA4: PCR analysis verified downregulation in all comparisons; of note, for adenoma-normal, PCR analysis showed a trend toward downregulation.
- ANXA10: PCR analysis verified ANXA10 upregulation in adenoma-normal, downregulation in cancer-adenoma, and the lack of significant difference in cancer-normal. In adenoma-adenoma, PCR analysis showed a trend toward downregulation, which mirrored significant HTA results.
Duodenal cancer is a leading cause of death in FAP after colectomy. SS IV duodenal polyposis is a risk factor for duodenal cancer, yet many FAP patients have no history of SS IV polyposis (4,9,10), indicating a need for additional predictors of cancer risk. In APCMin/+ mice, gene expression changes accompany the evolution of small intestinal neoplasia (13,14). To date, no such genome-wide investigation has been performed in patients with FAP. In this study, we described the duodenal adenoma-carcinoma sequence in FAP by comparing normal, adenoma, and cancer tissue of 12 duodenal cancer cases. In the transition from normal duodenum to adenoma, we found potential roles for enterocyte dedifferentiation, the Warburg effect, decreased ATRA synthesis, and impaired ROS/carcinogen defense. In the transition from adenoma to cancer, Paneth/goblet cell dedifferentiation, decreased ATRA synthesis, and increased tumor invasiveness were implicated.
Several DEGs distinguished FAP case from FAP control adenomas. ANXA10 is unique in that it is upregulated from normal to adenoma in FAP cases but downregulated from FAP case adenoma to cancer and from FAP control adenoma to FAP case adenoma. Given its function, ANXA10 expression in adenomas may indicate a protective “gastric programming” that suppresses neoplastic evolution. We also identified DEGs upregulated in cancer compared to adenoma that may have utility as biomarkers for neoplastic progression, including SPP1 and CEACAM5 (36,49,50).
Delker et al. (15) performed gene expression analysis on normal duodenum and adenoma in patients with FAP who were either treated with sulindac/erlotinib or with placebo. In the placebo group, they performed an adenoma-normal comparison similar to the one performed in this study. Genes involved in Wnt, PGE2, and EGFR signaling were differentially expressed in the placebo group but not in the sulindac/erlotinib group, indicating a beneficial inhibition of these pathways (15). Duodenal polyps in this study also exhibited upregulation of CD44, a cancer stem cell marker associated with PGE2 signaling (51), and MMP7, which encodes a matrix metalloproteinase and is a Wnt/Beta-catenin signaling target (52). In our study, CD44 and MMP7 were both upregulated in our cancer-normal comparisons (Table 3). Furthermore, MMP1, which is also a WNT/Beta-catenin target (53), was upregulated in our adenoma-normal and cancer-normal comparisons (Table 3).
We also identified DEGs with therapeutic potential in FAP. We found upregulation of SPP1, which plays a role in both the tumorigenic effect of PGE2 (40) and in activation of EGFR signaling (41). Given its relation to both pathways, determining SPP1 expression may help predict response to sulindac/erlotinib therapy. We also identified CEACAM6 as a potential novel therapeutic target for duodenal polyposis control in FAP. CEACAM6 has been successfully targeted in animal models of pancreatic cancer (45,47).
Several limitations merit further discussion. Our RNA extraction and gene expression profiling procedures were specific for FFPE and Hollande's fixatives and all RNA samples met QC checkpoints for HTA profiling. However, during PCR verification, several samples, particularly Hollande's fixed samples, yielded undefined results. As a result, PCR comparisons involved lower sample sizes and, while FCs matched our HTA results for 4 candidate genes, P values in some comparisons did not reach statistical significance. This indicates the importance of future validation studies with independent cohorts. Another limitation is the potential for false positives. To address this, we applied a FDR < 0.10 cutoff for our DEGs. Although there is still potential for false positives despite this cutoff, it should be noted that of 52 DEGs that differed in 2+ comparisons, all 52 differed in the same direction (upregulation/downregulation) in each comparison. Similarly, of 8 DEGs that differed in 3+ comparisons, all differed in the same direction in each comparison.
In summary, we have conducted the first ever genome-wide expression analysis of duodenal neoplasia in FAP. Future validation studies with immunohistochemical staining or Western Blot analysis are needed to verify protein expression of candidate genes. Furthermore, for genes whose expression may predict response to celecoxib or sulindac/erlotinib therapy, gene knock-in or knock-out in APCMin/+ mice can be performed to determine effect on therapeutic response. Effect of the CEACAM6 antibody-drug conjugate on APCMin/+ mice can also be investigated, and if this shows therapeutic benefit and low toxicity, targeting CEACAM6 may emerge as a viable option for duodenal polyposis control in FAP.
CONFLICTS OF INTEREST
Guarantor of the article: Sushrut S. Thiruvengadam, MD.
Specific author contributions: Study concept and design: S.S.T. and C.A.B. Acquisition of data (retrospective identification and selection of patients): S.S.T., R.L., M.O. and L.L. Acquisition of data (identification and preparation of archived tissue samples for transcriptional profiling): S.S.T. and R.K.P. Acquisition of data (transcriptional profiling): M.L.V. and C.L. Acquisition of data (PCR verification): S.S.T., Z.W., and B.S. Analysis and interpretation of the data: S.S.T., R.L., Y.C., J.S.B., and C.A.B. Drafting of the manuscript: S.S.T. Critical revision of the manuscript for important intellectual content: C.A.B., J.B.S., M.L.V., B.S., and Z.W. All authors have reviewed the final submitted draft.
Financial support: Grant support provided by Cleveland Clinic Research Program Committees Award (RPC 2014–1047). Technical support for this work was provided by the Gene Expression and Genotyping Facility, a component of the Integrated Genomic Shared Resource sponsored by the Case Comprehensive Cancer Center (P30 CA43703). This work was independent of this funding.
Potential competing interests: C.A.B. reports the following relevant financial disclosures: grants from Cancer Prevention Pharmaceuticals, Ferring Pharmaceuticals, consultant royalties and personal fees from Sucampo, Aries, and Salix Pharmaceuticals. M.K. reports the following relevant financial disclosures: consulting honorarium from Helomics and TransEnterix. The other authors affirm that they have no relevant financial or personal conflicts to disclose.
WHAT IS KNOWN
- ✓ Murine models of FAP have identified DEGs in the duodenal adenoma-carcinoma sequence.
- ✓ This has not been studied in patients with FAP.
WHAT IS NEW HERE
- ✓ Transition from normal duodenum to adenoma is characterized by abnormal metabolism of brush border proteins, lipids, ROS, and retinol and transition from adenoma to cancer was characterized by upregulation of DEGs involved in cell invasion and migration.
- ✓ Certain DEGs differed between adenomas from cancer patients and controls.
- ✓ Several DEGs have potential therapeutic significance in existing chemopreventive regimens, including the sulindac/erlotinib combination for duodenal polyposis in FAP.
- ✓ In the future, physicians may be able to use differential expression of certain genes in order to determine progression of duodenal adenoma to cancer in FAP.
- ✓ In the future, physicians may be able to target novel and existing chemopreventive pathways to prevent progression of duodenal polyposis and development of cancer in FAP.
1. Klaus A, Birchmeier W. Wnt signalling and its impact on development and cancer. Nat Rev Cancer 2008;8:387–98.
2. Vasen HFA, Möslein G, Alonso A, et al. Guidelines for the clinical management of familial adenomatous polyposis (FAP). Gut 2008;57:704–13.
3. Galle TS, Juel K, Bülow S. Causes of death in familial adenomatous polyposis. Scand J Gastroenterol 1999;34:808–12.
4. Bulow S, Bjork J, Christensen IJ, et al. Duodenal adenomatosis in familial adenomatous polyposis. Gut 2004;53:381–6.
5. Phillips RKS, Wallace MH, Lynch PM, et al. A randomised, double blind, placebo controlled study of celecoxib, a selective cyclooxygenase 2 inhibitor, on duodenal polyposis in familial adenomatous polyposis. Gut 2002;50:857–60.
6. Samadder NJ, Neklason DW, Boucher KM, et al. Effect of sulindac and erlotinib vs placebo on duodenal neoplasia in familial adenomatous polyposis. JAMA 2016;315:1266.
7. Brosens LAA, Keller JJ, Offerhaus GJA, et al. Prevention and management of duodenal polyps in familial adenomatous polyposis. Gut 2005;54:1034–43.
8. Johnson MD, Mackey R, Brown N, et al. Outcome based on management for duodenal adenomas: Sporadic versus familial disease. J Gastrointest Surg 2010;14:229–35.
9. Groves CJ, Saunders BP, Spigelman AD, et al. Duodenal cancer in patients with familial adenomatous polyposis (FAP): Results of a 10 year prospective study. Gut 2002;50:636–41.
10. Björk J, Akerbrant H, Iselius L, et al. Periampullary adenomas and adenocarcinomas in familial adenomatous polyposis: Cumulative risks and APC gene mutations. Gastroenterology 2001;121:1127–35.
11. Bianchi LK, Burke CA, Bennett AE, et al. Fundic gland polyp dysplasia is common in familial adenomatous polyposis. Clin Gastroenterol Hepatol 2008;6:180–5.
12. Moser AR, Pitot HC, Dove WF. A dominant mutation that predisposes to multiple intestinal neoplasia in the mouse. Science 1990;247:322–4.
13. Paoni NF, Feldman MW, Gutierrez LS, et al. Transcriptional profiling of the transition from normal intestinal epithelia to adenomas and carcinomas in the APCMin/+ mouse. Physiol Genomics 2003;15:228–35.
14. Leclerc D, Deng L, Trasler J, et al. Apc Min/+
mouse model of colon cancer: Gene expression profiling in tumors. J Cel Biochem 2004;93:1242–54.
15. Delker DA, Wood AC, Snow AK, et al. Chemoprevention with cyclooxygenase and epidermal growth factor receptor inhibitors in familial adenomatous polyposis patients: mRNA signatures of duodenal neoplasia. Cancer Prev Res 2018;11:4–15.
16. Burke CA, Beck GJ, Church JM, et al. The natural history of untreated duodenal and ampullary adenomas in patients with familial adenomatous polyposis followed in an endoscopic surveillance program. Gastrointest Endosc 1999;49:358–64.
17. Sambuy Y, Angelis IDe, Ranaldi G, et al. The caco-2 cell line as a model of the intestinal barrier: Influence of cell and culture-related factors on caco-2 cell functional characteristics. Cell Biol Toxicol 2005;21:1–26.
18. Goda T, Yasutake H, Tanaka T, et al. Lactase-phlorizin hydrolase and sucrase-isomaltase genes are expressed differently along the villus-crypt axis of rat jejunum. J Nutr 1999;129:1107–13.
19. Imamura T, Kitamoto Y. Expression of enteropeptidase in differentiated enterocytes, goblet cells, and the tumor cells in human duodenum. Am J Physiol Gastrointest Liver Physiol 2003;285:G1235–41.
20. Van Beers EH, Al RH, Rings EH, et al. Lactase and sucrase-isomaltase gene expression during Caco-2 cell differentiation. Biochem J 1995;308(Pt 3):769–75.
21. Reisher SR, Hughes TE, Ordovas JM, et al. Increased expression of apolipoprotein genes accompanies differentiation in the intestinal cell line Caco-2. Proc Natl Acad Sci U S A 1993;90:5757–61.
22. Jakab RL, Collaco AM, Ameen NA. Physiological relevance of cell-specific distribution patterns of CFTR, NKCC1, NBCe1, and NHE3 along the crypt-villus axis in the intestine. Am J Physiol Liver Physiol 2011;300:G82–98.
23. Kropotova ES, Zinov’eva OL, Zyrianova AF, et al. Expression of genes involved in retinoic acid biosynthesis in human gastric cancer [in Russian]. Mol Biol (Mosk) 2013;47:317–30.
24. Subbaramaiah K, Cole PA, Dannenberg AJ. Retinoids and carnosol suppress cyclooxygenase-2 transcription by CREB-binding protein/p300-dependent and -independent mechanisms. Cancer Res 2002;62:2522–30.
25. van Heumen BW, Roelofs HM, te Morsche RH, et al. Duodenal mucosal risk markers in patients with familial adenomatous polyposis: Effects of celecoxib/ursodeoxycholic acid co-treatment and comparison with patient controls. Orphanet J Rare Dis 2013;8:181.
26. Yang B, Cao L, Liu B, et al. The transition from proliferation to differentiation in colorectal cancer is regulated by the calcium activated chloride channel A1. PLoS One 2013;8:e60861.
27. Mastroianni JR, Costales JK, Zaksheske J, et al. Alternative luminal activation mechanisms for paneth cell α-defensins. J Biol Chem 2012;287:11205–12.
28. Radeva MY, Jahns F, Wilhelm A, et al. Defensin alpha 6 (DEFA 6) overexpression threshold of over 60 fold can distinguish between adenoma and fully blown colon carcinoma in individual patients. BMC Cancer 2010;10:588.
29. Joo M, Shahsafaei A, Odze RD. Paneth cell differentiation in colonic epithelial neoplasms: Evidence for the role of the apc/β-catenin/Tcf pathway. Hum Pathol 2009;40:872–80.
30. Karagiannis GS, Petraki C, Prassas I, et al. Proteomic signatures of the desmoplastic invasion front reveal collagen type XII as a marker of myofibroblastic differentiation during colorectal cancer metastasis. Oncotarget 2012;3:267–85.
31. Mitra AK, Sawada K, Tiwari P, et al. Ligand-independent activation of c-Met by fibronectin and α5β1-integrin regulates ovarian cancer invasion and metastasis. Oncogene 2011;30:1566–76.
32. Eltanani M, Campbell F, Kurisetty V, et al. The regulation and role of osteopontin in malignant transformation and cancer. Cytokine Growth Factor Rev 2006;17:463–74.
33. Ouyang G, Liu M, Ruan K, et al. Upregulated expression of periostin by hypoxia in non-small-cell lung cancer cells promotes cell survival via the Akt/PKB pathway. Cancer Lett 2009;281:213–9.
34. Yoshida S, Ono M, Shono T, et al. Involvement of interleukin-8, vascular endothelial growth factor, and basic fibroblast growth factor in tumor necrosis factor alpha-dependent angiogenesis. Mol Cell Biol 1997;17:4015–23.
35. Kim JK, Kim PJ, Jung KH, et al. Decreased expression of annexin A10 in gastric cancer and its overexpression in tumor cell growth suppression. Oncol Rep 2010;24:607–12.
36. Van Heek NT, Maitra A, Koopmann J, et al. Gene expression profiling identifies markers of ampullary adenocarcinoma. Cancer Biol Ther 2004;3:651–6.
37. Kim CG, Ahn JB, Jung M, et al. Preoperative serum carcinoembryonic antigen level as a prognostic factor for recurrence and survival after curative resection followed by adjuvant chemotherapy in stage III colon cancer. Ann Surg Oncol 2017;24:227–35.
38. Samadder NJ, Neklason DW, Boucher KM, et al. Effect of sulindac and erlotinib vs placebo on duodenal neoplasia in familial adenomatous polyposis. JAMA 2016;315:1266.
39. Rohde F, Rimkus C, Friederichs J, et al. Expression of osteopontin, a target gene of de-regulated Wnt signaling, predicts survival in colon cancer. Int J Cancer 2007;121:1717–23.
40. Zagani R, Hamzaoui N, Cacheux W, et al. Cyclooxygenase-2 inhibitors down-regulate osteopontin and Nr4a2—New therapeutic targets for colorectal cancers. Gastroenterology 2009;137:1358–66.e3.
41. Lamour V, Henry A, Kroonen J, et al. Targeting osteopontin suppresses glioblastoma stem-like cell character and tumorigenicity in vivo
. Int J Cancer 2015;137:1047–57.
42. Wang M, Han J, Marcar L, et al. Radiation resistance in KRAS-mutated lung cancer is enabled by stem-like properties mediated by an osteopontin-EGFR pathway. Cancer Res 2017;77:2018–28.
43. Beauchemin N, Arabzadeh A. Carcinoembryonic antigen-related cell adhesion molecules (CEACAMs) in cancer progression and metastasis. Cancer Metastasis Rev 2013;32:643–71.
44. Jantscheff P, Terracciano L, Lowy A, et al. Expression of CEACAM6 in resectable colorectal cancer: A factor of independent prognostic significance. J Clin Oncol 2003;21:3638–46.
45. Duxbury MS, Ito H, Ashley SW, et al. CEACAM6 as a novel target for indirect type 1 immunotoxin-based therapy in pancreatic adenocarcinoma. Biochem Biophys Res Commun 2004;317:837–43.
46. Duxbury MS, Matros E, Clancy T, et al. CEACAM6 is a novel biomarker in pancreatic adenocarcinoma and PanIN lesions. Ann Surg 2005;241:491–6.
47. Strickland LA, Ross J, Williams S, et al. Preclinical evaluation of carcinoembryonic cell adhesion molecule (CEACAM) 6 as potential therapy target for pancreatic adenocarcinoma. J Pathol 2009;218:380–90.
48. Zhou M, Bronner M, Magi-Galluzz C, et al. Optimized RNA extraction and RT-PCR assays provide successful molecular analysis on a wide variety of archival fixed tissues. Cancer Res 2007;67:4423.
49. Slentz K, Senagore A, Hibbert J, et al. Can preoperative and postoperative CEA predict survival after colon cancer resection? Am Surg 1994;60:528–31.
50. Ni XG, Bai XF, Mao YL, et al. The clinical value of serum CEA, CA19-9, and CA242 in the diagnosis and prognosis of pancreatic cancer. Eur J Surg Oncol 2005;31:164–9.
51. Wang D, Fu L, Sun H, et al. Prostaglandin E2 promotes colorectal cancer stem cell expansion and metastasis in mice. Gastroenterology 2015;149:1884–95.e4.
52. Dey N, Young B, Abramovitz M, et al. Differential activation of Wnt-β-catenin pathway in triple negative breast cancer increases MMP7 in a PTEN dependent manner. PLoS One 2013;8:e77425.
53. Jean C, Blanc A, Prade-Houdellier N, et al. Epidermal growth factor receptor/-catenin/T-cell factor 4/matrix metalloproteinase 1: A new pathway for regulating keratinocyte invasiveness after UVA irradiation. Cancer Res 2009;69:3291–9.
Supplemental Digital Content
© 2019 by Lippincott Williams & Wilkins, Inc.