Expression of insulin-like growth factor II mRNA-binding protein 3 in human esophageal adenocarcinoma and its precursor lesions.
The oncofetal protein, insulin-like growth factor (IGF) II mRNA-binding protein 3 (IMP3), is highly expressed in fetal tissue and malignant tumors, such as pancreatic adenocarcinoma, endometrial carcinoma, endocervical adenocarcinoma, renal cell carcinoma, lung carcinoma, and melanoma. (18-30) IMP3 belongs to a family of IMPs that also includes IMP1 and IMP2 and is identical to the KH domain containing protein overexpressed in cancer protein. (31) IMP3 is shown to be expressed in various human and mouse tissues, such as developing epithelium, muscle, and placenta during early stages of embryogenesis but is not expressed or expressed in low to undetectable levels in adult tissues. (22,32) IMP3 is reported to be involved in cell growth, adhesion, invasion, and migration and is a prognostic marker associated with metastatic progression as demonstrated in several studies of renal cell carcinoma. (23-26,33-35)
The prevalence and significance of IMP3 in EAC and its precursor lesions including distinctive type BM and dysplasia are largely unknown. In this microarray study, we characterized the expression of IMP3 in EAC and its precursor lesions. In addition, correlations between IMP3 expression in EAC and various clinical parameters, such as stage, margin status, lymph node status, and overall survival, are examined.
MATERIALS AND METHODS
After receiving institutional review board approval, we retrieved esophagectomy and biopsy specimens with adequate material from the archive of the Department of Pathology at University of Rochester Medical Center. These specimens consisted of a total of 132 cases of EAC, 28 cases of ECD (16 high-grade dysplasia and 12 low-grade dysplasia cases), 28 cases of BM without dysplasia or carcinoma, and 138 cases of nonneoplastic esophageal mucosa (NNEM) without dysplasia or BM. Three cases of high-grade dysplasia were evaluated in which this diagnosis was the only finding. The other dysplasia cases were obtained from specimens that also contained invasive carcinoma. To avoid potential treatment effects, we excluded specimens exposed to preoperative treatment. All tissues had been fixed in 10% neutral buffered formalin, routinely processed, and embedded in paraffin. Hematoxylin-eosin--stained sections from each case were reviewed by 2 pathologists, and from each case we selected 2 to 3 sections with representative and adequate amount of lesional tissue to be included in the tissue microarray (TMA).
Additionally, full sections from 13 resection cases of EAC and 1 resection case of high-grade dysplasia and intestinal metaplasia were also stained with IMP3 to confirm our microarray results. All 14 full section cases contained NNEM.
Esophageal adenocarcinoma and ECD cases were graded and classified based on the World Health Organization classification. (36) Well-differentiated EACs are characterized by predominantly glandular or papillary structures, while poorly differentiated EACs are characterized by only rare glandular structures. Moderately differentiated EACs are those with intermediate glandular differentiation. In our sample set of EAC cases, if more than 1 sample from 1 EAC case was present in our microarray, the highest tumor grade was reported. These results from the microarray were compared with the original reported grading and those with discrepant grades were omitted from the outcome evaluation.
Survival and tumor characteristics data, if available, including length of follow-up, time from diagnosis to death, lymph node status, tumor site, margin status, and the American Joint Committee on Cancer stage grouping, were retrieved and deassociated with patient identifiers. The EAC cases included 57 well-differentiated, 32 moderately differentiated, and 43 poorly differentiated EACs. Of these EAC cases, 1 case was from the proximal esophagus, 15 cases were from the mid esophagus, 41 cases were from the distal esophagus, and 67 were from the gastroesophageal junction. One hundred and seven EAC cases had negative margins while 17 EAC cases had positive margins. Fifty-two EAC cases were negative for lymph node metastasis and 75 EAC cases were positive for lymph node metastasis. The EAC cases included 19 stage I (7 with submucosal invasion and 12 limited to the lamina propria), 41 stage II, 53 stage III, and 11 stage IV cases.
Construction of TMAs
The TMAs were constructed by using a manual tissue arrayer device from Beecher Instruments (Sun Prairie, Wisconsin). The areas of tissue with adequate amount of representative tissue were marked in the paraffin blocks, and two to three 1-mm diameter cores from each case were included in their respective arrays.
Immunohistochemistry was performed using mouse monoclonal anti-IMP3 antibody (1:80, Dako, Glostrup, Denmark). Briefly, 4-[micro]m thick microarray sections from routinely processed, formalin-fixed, paraffin-embedded tissues were transferred to glass slides, deparaffinized in xylene, and rehydrated in a graded series of ethanol. Heat-induced epitope retrieval was performed. The tissue was then treated with 3% [H.sub.2][O.sub.2] and then rinsed with Tween-20 buffer. A few drops of diluted normal blocking serum were placed on the tissue and incubated at room temperature. The serum was then blotted off, and the slides were incubated with primary antibody directed against IMP3 (1:80, 45-minute incubation at 4[degrees]C). The sections were then treated with a cocktail of biotinylated anti-rabbit immunoglobulin (Ig) G and antimouse IgG and IgM (Ventana, Tucson, Arizona) for 30 minutes, followed by avidin-biotin-peroxidase complex (Ventana) for 30 minutes. Sections were then rinsed, developed with diaminobenzidine and hydrogen peroxide (10 minutes), counterstained with Mayer hematoxylin, and cover slipped. Representative sections of pancreatic adenocarcinoma tissue were used as a positive control. Negative controls were performed by replacing the primary antibody with nonimmune IgG. Positive and negative controls reacted appropriately.
Scoring of Immunoreactivity of EAC, ECD, BM, and NNEM and Cellular Compartmentalization of Chromogenic Signal
Semiquantitative assessment of expression levels of IMP3 protein analytes in the cytoplasm of target cells was determined by a pathologist using bright-field microscopy. The intensity of positive staining was accessed and graded from 1+ to 3+, with 1+ being weakly positive cytoplasmic and/or membranous staining, 3+ being strongly positive cytoplasmic and/or membranous staining, and 2+ being intermediate staining between 1+ and 3+. Intensity of 0 was assigned to cases with no evidence of specific staining. The approximate percent (0%-100% in increments of 5%) of positive staining cells, defined as at least 1+ intensity of staining, within a lesion was evaluated. If more than 1 sample per case was available for evaluation, the intensity grade and percent of positive staining cells were averaged. A case was regarded as positive when there was at least 1+ staining intensity in at least 5% of the lesional cells.
Independent 2-tailed Student t test was used to determine the statistical significance of differences in IMP3 expression levels among different EAC tumor grades and between EAC and ECD. Independent 2-tailed Student t test was also used to determine the statistical significance of differences in IMP3 expression levels in EAC between TMA and full section staining. Pearson [chi square] test was used to determine statistical significance of differences in frequency of positive IMP3 expression among EAC, ECD, BM, and NNEM. Semiquantitative IMP3 staining expression level was correlated with histologic or clinical parameters, and the respective Pearson product moment correlation (r) was determined. Pearson [chi square] was used to determine the statistical significance of differences in frequency of positive IMP3 expression in subsets of EAC with varying clinical and histologic parameters. IMP3 expression level was also correlated with patient survival data. Survival curves were estimated using the Kaplan-Meier product-limit method, and the significance of differences between survival curves was determined using a log-rank test where the P value for the [chi square] statistics was determined. Multivariate analysis of IMP3 expression level, patient age, sex, lymph node status, American Joint Committee on Cancer stage, margin status, tumor site, and histologic grade on survival was performed using the Cox proportional hazards regression modeling. MedCalc (Mariakerke, Belgium) software was used for statistical analysis and P values less than .05 were considered statistically significant.
[FIGURE 1 OMITTED]
RESULTS Microarray Results
No IMP3 staining was observed in any of the NNEM and BM uninvolved by dysplasia (Figure 1, A and B). Benign gastric oxyntic glands showed consistent nonspecific cytoplasmic blush staining for IMP3. A minority (7 of 28; 25%) of dysplasia cases was positive for IMP3 (Figure 1, C and D). Three low-grade ECD cases showed positive (1+ in <15% of lesional cells) staining for IMP3, and 4 high-grade ECD cases without associated carcinoma showed more diffuse and stronger positive (1-2+ in 5%-80% of lesional cells) staining for IMP3.
Most EAC cases (93 of 132; 70%) showed positive cytoplasmic and membranous IMP3 staining (Table 1). Higher level of IMP3 expression was found in moderately differentiated EAC (mean intensity, 1.5 [+ or -] 6 0.17; mean percent of positive staining, 43 [+ or -] 5.7) and poorly differentiated EAC (mean intensity, 1.7 [+ or -] 0.17; mean percent of positive staining, 48 [+ or -] 5.5) compared with well-differentiated EAC (mean intensity, 0.86 [+ or -] 0.12; mean percent of positive staining, 22 [+ or -] 3.7). Moderately and poorly differentiated EAC cases showed statistically significant higher levels of IMP3 expression compared with well-differentiated EAC cases (P < .001). Moderately and poorly differentiated EAC cases (>80%) also more frequently showed positive IMP3 expression compared with well-differentiated EAC cases (53%). No significant difference in IMP3 expression levels was observed between moderately and poorly differentiated cases. These results are illustrated and summarized in Figure 1, E and F, and Tables 1 and 2. A comparison of clinical and pathologic parameters by IMP3 expression levels is summarized in Table 3. Although the difference in frequency of IMP3-positive EAC was not statistically significant between younger (<65 years) and older (>65 years) patients, EAC in younger patients (<65 years) showed higher IMP3 expression level compared with EAC in older (>65 years) patients (P = .03, 2-tailed Student t test). Although not statistically significant, positive IMP3 expression in EAC also appeared to be associated with higher stage and positive lymph node involvement. IMP3 expression in EAC only showed statistically significant correlation with tumor differentiation (r = 0.37, P < .001).
Multivariate analysis showed that tumor stage was the only variable predictive of overall survival (P < .001; Figure 2, A). Univariate analysis also showed that stage, regional lymph node, and margin status were predictive of overall survival (Figure 2, B and C). IMP3 expression level does not appear to be associated with overall survival by univariate analysis (Figure 2, D). However, univariate analysis showed that in the subset of stages 3 and 4 EAC cases, 1-year overall survival was 50% for IMP3-positive cases and 30% for IMP3-negative cases, and in the subset of regional lymph node-positive EAC cases, the 5-year overall survival was 20% for IMP3-positive cases and 6% for IMP3-negative cases (Figure 2, E and F).
Esophageal adenocarcinoma cases showed statistically significant higher average intensity of IMP3 expression compared with ECD (P < .001). IMP3 was more frequently expressed in EAC compared with ECD, BM, and NNEM (P < .001). IMP3 was more frequently expressed in ECD compared with BM and NNEM (P = .02 and P < .001, respectively). These results are illustrated and summarized in Figure 1 and Tables 1 and 4. Cases of BM adjacent to high-grade dysplasia and intramucosal EAC clearly demonstrated the lack of IMP3 staining within the BM and positive IMP3 staining within the dysplastic and neoplastic glands (Figure 3, A, and B).
Full Section Results
Thirteen full sections of EAC cases were also stained with IMP3 and 10 of these showed 1+ to 3+ intensity (mean, 1.27) positive staining in 10% to 100% (mean, 45%) of neoplastic cells (Table 5). The staining pattern was patchy, but no particular difference in staining was seen between neoplastic cells at different levels or depth of invasion. Two EAC cases with mucinous and goblet cell differentiation showed the most varied patchy staining pattern. One EAC case contained both well-differentiated and poorly differentiated areas. The poorly differentiated areas in this case showed positive but less intense staining for IMP3 compared with the well-differentiated area. Compared with the TMA results, the full section results showed no statistical difference in the percentage of positive cases of EAC for IMP3 (TMA, 70%; full section, 77%; Pearson [chi square] test, P = .86, degrees of freedom of 1). Independent 2-tailed Student t test showed no statistical difference in the intensity and percent of positive staining cells of EAC cases between TMA and full section staining (intensity, P = .98; percent of positive staining cells, P = .35).
A full section of high-grade ECD showed positive 2+ IMP3 staining in 80% of the lesion. Intestinal metaplasia unassociated with dysplasia did not show positive staining for IMP3 in this case. All 14 cases containing NNEM, including squamous epithelium, metaplastic columnar epithelium without goblet cells, and submucosal esophageal glands, showed no staining for IMP3. Benign gastric oxyntic glands consistently showed nonspecific cytoplasmic blush staining for IMP3. These results are summarized in Table 5.
In this study, we found that IMP3 is overexpressed in EAC and a subset of ECD but not in BM and NNEM. Low-grade ECD is noted to have only focal weak IMP3 expression in contrast to high-grade ECD, which showed more intense diffuse expression of IMP3. These results suggest that when positive, IMP3 staining is a very useful adjunct in differentiating difficult cases of neoplastic and dysplastic lesions versus BM and reactive lesions of the esophagus. Although TMA may not be completely representative, it is an efficient way to screen for tissue markers and studies have shown that TMA with 1, 2, and 3 cores will represent about 91%, 96%, and 98%, respectively, compared with whole section studies. (37-40) Additional full section staining of limited cases showed similar results compared with our microarray results. Compared with our TMA results, the full section results showed no statistical difference in the percentage of positive cases of EAC for IMP3 (TMA, 70%; full section, 77%; Pearson [chi square] test, P = .86, degrees of freedom of 1) and showed the same mean intensity of expression (mean, 1.27; independent 2-tailed Student t test, P = .98) and similar percent of lesional cells with positive staining (TMA mean, 35%; full section mean, 45%; independent 2-tailed Student t test, P = .35).
A recent study on IMP3 expression in EAC and ECD reported by Lu et al (41) also showed similar findings in IMP3 expression pattern as in the present study. However, the percent of EAC cases with positive IMP3 expression was reported to be much higher than our findings (94% versus 70% of microarray cases and 77% of full section cases). Also we noted a more significantly increased IMP3 expression in moderately and poorly differentiated EAC compared with well-differentiated EAC than was reported by Lu et al. In contrast to our study, Lu et al did report rare positive staining for IMP3 in BM (5 of 68; 7%). These differences may be due to sampling issues in a microarray study; however, the series by Lu et al also included a significant number of biopsy cases, which also would have the same issues in limited sampling of lesional tissue. These results confirm that IMP3 can be a useful supplemental immunohistochemical marker in addition to routine light microscopy evaluation of difficult esophageal surgical cases.
[FIGURE 2 OMITTED]
The present study also demonstrated a stepwise increase in IMP3 expression in NNEM and BM, ECD, and EAC, suggesting that IMP3 expression may play a role in the pathogenesis of EAC. The correlation of higher levels of IMP3 expression with higher tumor grade and the significantly better overall survival seen in IMP3-positive versus IMP3-negative high-stage EAC cases indicate that IMP3 might also have potential prognostic value in certain subsets of EAC. The reported association of increased IMP3 expression levels and worse overall survival as observed in multiple renal cell carcinoma studies (23-26) was not seen in the present study.
The role of the IGF pathway in esophageal carcinogenesis has not been well studied. IMP3 is a regulatory binding protein thought to be involved in the stabilization and intracellular trafficking of IGF-II mRNA to facilitate IGF-II production and is therefore believed to have similar roles in the modulation of other intracellular nucleotides. (42) The modulation of the IGF cell signaling pathways is thought to be associated with the acquisition of malignant features in esophageal tumors. (43,44) The signal transduction pathway leading to the ultimate expression of IGF may be modulated by multiple factors, such as epithelial growth factor and IGF binding proteins, as well as other molecular pathways such as the PI3K/AKT/mTOR and Ras/Raf/ MAPK pathways. (45) For these reasons, the isolated study of IMP3 is difficult to accurately assess for its specific role in carcinogenesis. Although our preliminary study did not show statistically significant predictive value of IMP3 for overall survival in this EAC series by univariable and multivariable analysis, the results suggest that IMP3 may have prognostic value in a high-stage subset of EAC cases.
[FIGURE 3 OMITTED]
In summary, through this TMA study, we demonstrated the expression pattern of IMP3 in esophageal tissues. Further independent studies with larger case series are needed to validate and better characterize the role of IMP3 in esophageal carcinogenesis and EAC progression.
(1.) Blot WJ, Devesa SS, Fraumeni JF Jr. Continuing climb in rates of esophageal adenocarcinoma: an update. JAMA. 1993;270(11):1320.
(2.) Devesa SS, Blot Wj, Fraumeni JF Jr. Changing patterns in the incidence of esophageal and gastric carcinoma in the United States. Cancer. 1998;83(10): 2049-2053.
(3.) Fernandes ML, Seow A, Chan YH, Ho KY. Opposing trends in incidence of esophageal squamous cell carcinoma and adenocarcinoma in a multi-ethnic Asian country. Am J Gastroenterol. 2006;101(7):1430-1436.
(4.) Lepage C, Bouvier AM, Manfredi S, Coatmeur O, Cheynel N, Faivre J. Trends in incidence and management of esophageal adenocarcinoma in a well defined population. Gastroenterol Clin Biol. 2005;29(12):1258-1263.
(5.) Vizcaino AP, Moreno V, Lambert R, Parkin DM. Time trends incidence of both major histologic types of esophageal carcinomas in selected countries, 1973-1995. Int J Cancer. 2002;99(6):860-868.
(6.) Ormsby AH, Petras RE, Henricks WH, et al. Observer variation in the diagnosis of superficial oesophageal adenocarcinoma. Gut. 2002;51(5):671-676.
(7.) Montgomery E, Bronner MP, Goldblum JR, et al. Reproducibility of the diagnosis of dysplasia in Barrett esophagus: a reaffirmation. Hum Pathol. 2001; 32(4):368-378.
(8.) Entwistle JW 3rd, Goldberg M. Multimodality therapy for resectable cancer of the thoracic esophagus. Ann Thorac Surg. 2002;73(3):1009-1015.
(9.) Malthaner RA, Wong RK, Rumble RB, Zuraw L. Neoadjuvant or adjuvant therapy for resectable esophageal cancer: a systematic review and meta-analysis. BMC Med. 2004;2:35.
(10.) Feith M, Stein HJ, Mueller J, Siewert JR. Malignant degeneration of Barrett's esophagus: the role of the Ki-67 proliferation fraction, expression of E cadherin and p53. Dis Esophagus. 2004;17(4):322-327.
(11.) Bani-Hani K, Martin IG, Hardie LJ, et al. Prospective study of cyclin D1 overexpression in Barrett's esophagus: association with increased risk of adenocarcinoma. J Natl Cancer Inst. 2000;92(16):1316-1321.
(12.) Ireland AP, Shibata DK, Chandrasoma P, Lord RV, Peters JH, DeMeester TR. Clinical significance of p53 mutations in adenocarcinoma of the esophagus and cardia. Ann Surg. 2000;231(2):179-187.
(13.) Ireland AP, Clark GW, DeMeester TR. Barrett'sesophagus: the significance of p53 in clinical practice. Ann Surg. 1997;225(1):17-30.
(14.) Cescon DW, Bradbury PA, Asomaning K, et al. p53 Arg72Pro and MDM2 T309G polymorphisms, histology, and esophageal cancer prognosis. Clin Cancer Res. 2009;15(9):3103-3109.
(15.) Liu G, Cescon DW, Zhai R, et al. p53 Arg72Pro, MDM2 T309G and CCND1 G870A polymorphisms are not associated with susceptibility to esophageal adenocarcinoma. Dis Esophagus. 2010;23(1):36-39.
(16.) Dorer R, Odze RD. AMACR immunostaining is useful in detecting dysplastic epithelium in Barrett's esophagus, ulcerative colitis, and Crohn's disease. Am J Surg Pathol. 2006;30(7):871-877.
(17.) Lisovsky M, Falkowski O, Bhuiya T. Expression of alpha-methylacylcoenzyme A racemase in dysplastic Barrett's epithelium. Hum Pathol. 2006; 37(12):1601-1606.
(18.) Pryor JG, Bourne PA, Yang Q, Spaulding BO, Scott GA, Xu H. IMP-3 is a novel progression marker in malignant melanoma. Mod Pathol. 2008;21(4):431437.
(19.) Li L, Xu H, Spaulding BO, et al. Expression of RNA-binding protein IMP3 (KOC) in benign urothelium and urothelial tumors. Hum Pathol. 2008;39(8): 1205-1211.
(20.) Li C, Zota V, Woda BA, et al. Expression of a novel oncofetal mRNA-binding protein IMP3 in endometrial carcinomas: diagnostic significance and clinicopathologic correlations. Mod Pathol. 2007;20(12):1263-1268.
(21.) Li C, Rock KL, Woda BA, Jiang Z, Fraire AE, Dresser K. IMP3 is a novel biomarker for adenocarcinoma in situ of the uterine cervix: an immunohistochemical study in comparison with p16(INK4a) expression. Mod Pathol. 2007; 20(2):242-247.
(22.) Zheng W, Yi X, Fadare O, et al. The oncofetal protein IMP3: a novel biomarker for endometrial serous carcinoma. Am J Surg Pathol. 2008;32(2):304315.
(23.) Jiang Z, Lohse CM, Chu PG, et al. Oncofetal protein IMP3: a novel molecular marker that predicts metastasis of papillary and chromophobe renal cell carcinomas. Cancer. 2008;112(12):2676-2682.
(24.) Jiang Z, Chu PG, Woda BA, et al. Analysis of RNA-binding protein IMP3 to predict metastasis and prognosis of renal-cell carcinoma: a retrospective study. Lancet Oncol. 2006;7(7):556-564.
(25.) Jiang Z, Chu PG, Woda BA, et al. Combination of quantitative IMP3 and tumor stage: a new system to predict metastasis for patients with localized renal cell carcinomas. Clin Cancer Res. 2008;14(17):5579-5584.
(26.) Hoffmann NE, Sheinin Y, Lohse CM, et al. External validation of IMP3 expression as an independent prognostic marker for metastatic progression and death for patients with clear cell renal cell carcinoma. Cancer. 2008;112(7): 1471-1479.
(27.) Yantiss RK, Cosar E, Fischer AH. Use of IMP3 in identification of carcinoma in fine needle aspiration biopsies of pancreas. Acta Cytol. 2008;52(2): 133-138.
(28.) Yantiss RK, Woda BA, Fanger GR, et al. KOC (K homology domain containing protein overexpressed in cancer): a novel molecular marker that distinguishes between benign and malignant lesions of the pancreas. Am J Surg Pathol. 2005;29(2):188-195.
(29.) Wang T, Fan L, Watanabe Y, et al. L523S, an RNA-binding protein as a potential therapeutic target for lung cancer. Br J Cancer. 2003;88(6):887-894.
(30.) Nielsen J, Christiansen J, Lykke-Andersen J, Johnsen AH, Wewer UM, Nielsen FC. A family of insulin-like growth factor II mRNA-binding proteins represses translation in late development. Mol Cell Biol. 1999;19(2):1262-1270.
(31.) Mueller-Pillasch F, Lacher U, Wallrapp C, et al. Cloning of a gene highly overexpressed in cancer coding for a novel KH-domain containing protein. Oncogene. 1997;14(22):2729-2733.
(32.) Mueller-Pillasch F, Pohl B, Wilda M, et al. Expression of the highly conserved RNA binding protein KOC in embryogenesis. Mech Dev. 1999;88(1): 95-99.
(33.) Hansen TV, Hammer NA, Nielsen J, et al. Dwarfism and impaired gut development in insulin-like growth factor II mRNA-binding protein 1-deficient mice. Mol Cell Biol. 2004;24(10):4448-4464.
(34.) Yaniv K, Fainsod A, Kalcheim C, Yisraeli JK. The RNA-binding protein Vg1 RBP is required for cell migration during early neural development. Development. 2003;130(23):5649-5661.
(35.) Vikesaa J, Hansen TV, Jonson L, et al. RNA-binding IMPs promote cell adhesion and invadopodia formation. EMBO J. 2006;25(7):1456-1468.
(36.) Hamilton SR, Aaltonen LA, eds. Pathology and Genetics of Tumours of the Digestive System. Lyon, France: IARC Press; 2000. World Health Organization Classification of Tumours.
(37.) Kononen J, Bubendorf L, Kallioniemi A, et al. Tissue microarrays for high- through put molecular profiling of tumor specimens. NatMed. 1998;4(7):844-847.
(38.) Tan D, Li Q, Deeb G, et al. Thyroid transcription factor-1 expression prevalence and its clinical implications in non-small cell lung cancer: a high- throughput tissue microarray and immunohistochemistry study. Hum Pathol. 2003;34(6):597-604.
(39.) Deeb G, Wang J, Ramnath N, et al. Altered E-cadherin and epidermal growth factor receptor expressions are associated with patient survival in lung cancer: a study utilizing high-density tissue microarray and immunohistochemistry. Mod Pathol. 2004;17(4):430-439.
(40.) Aljada IS, Ramnath N, Donohue K, et al. Upregulation of the tissue inhibitor of metalloproteinase-1 protein is associated with progression of human non-small-cell lung cancer. J Clin Oncol. 2004;22(16):3218-3229.
(41.) Lu D, Vohra P, Chu PG, Woda B, Rock KL, Jiang Z. An oncofetal protein IMP3: a new molecular marker for the detection of esophageal adenocarcinoma and high-grade dysplasia. Am J Surg Pathol. 2009;33(4):521-525.
(42.) Nielsen FC, Nielsen J, Christiansen J. A family of IGF-II mRNA binding proteins (IMP) involved in RNA trafficking. Scand J Clin Lab Invest Suppl. 2001; 234:93-99.
(43.) Adachi Y, Imsumran A, Yamamoto H, et al. IGF-I receptor is a molecular target for both esophageal squamous cell carcinoma and adenocarcinoma. AACR Meeting Abstracts. 2006;2006(1):289-a-.
(44.) Feagins LA, Susnow N, Zhang HY, et al. Gain of allelic geneexpression for IGF-II occurs frequently in Barrett's esophagus. Am J Physiol Gastrointest Liver Physiol. 2006;290(5):G871-G875.
(45.) Takaoka M, Smith CE, Mashiba MK, et al. EGF-mediated regulation of IGFBP-3 determines esophageal epithelial cellular response to IGF-I. Am J Physiol Gastrointest Liver Physiol. 2006;290(2):G404-G416.
Wei Feng, MD; Zhongren Zhou, MD, PhD; Jeffrey H. Peters, MD; Thaer Khoury, MD; Qihui Zhai, MD; Qiying Wei, MD, PhD; Camtu D. Truong, MD; Sonya Wei Song, PhD; Dongfeng Tan, MD
Accepted for publication November 22, 2010.
From the Department of Pathology and Laboratory Medicine, The University of Texas M. D. Anderson Cancer Center, Houston, Texas (Drs Feng, Truong, Wei, and Tan); the Department of Pathology and Laboratory Medicine, University of Rochester Medical Center, Rochester, New York (Drs Zhou and Peters); the Department of Pathology, University of Cincinnati, Cincinnati, Ohio (Dr Zhai); the Department of Pathology, Roswell Park Cancer Institute, Buffalo, New York (Dr Khoury); and Center of Cancer Research, the Capital Medical University, Beijing, China (Dr Song).
Drs Feng and Zhou contributed equally to this manuscript.
The authors have no relevant financial interest in the products or companies described in this article.
Reprints: Dongfeng Tan, MD, Department of Pathology and Laboratory Medicine, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Blvd, Unit 85, Houston, TX 77030 (e-mail: firstname.lastname@example.org).
Table 1. Expression of Insulin-Like Growth Factor II mRNA-Binding Protein 3 (IMP3) in Esophageal Adenocarcinoma (EAC), Esophageal Columnar Dysplasia (ECD), Barrett Mucosa (BM), and Nonneoplastic Esophageal Mucosa (NNEM) Significance of Differences in Means in EAC IMP3 Staining EAC ECD Versus ECD (a) Positive cases (b)/ total cases (%) 93/132 (70) 7/28 (25) NA Moderate to strong intensity staining (2-3+)/total cases (%) 47/132 (36) 2/28 (7) NA Diffusely positive staining cases ([greater than or equal to]50%)/ 56/132 (42) 1/28 (4) NA total cases (%) Mean intensity of staining/standard error 1.27/0.09 0.36/0.12 P < .001 Mean % of staining cells/standard error 35/2.9 7.0/3.2 P < .001 IMP3 Staining BM NNEM Positive cases (b)/ total cases (%) 0/28 (0) 0/138 (0) Moderate to strong intensity staining (2-3+)/total cases (%) 0/28 (0) 0/138 (0) Diffusely positive staining cases ([greater than or equal to]50%)/ 0/28 (0) 0/138 (0) total cases (%) Mean intensity of staining/standard error 0/NA 0/NA Mean % of staining cells/standard error 0/NA 0/NA Abbreviation: NA, not applicable. (a) Determined by 2-tailed Student t test. (b) At least 1 + staining intensity in at least 5% of lesional cells. Table 2. Expression of Insulin-Like Growth Factor II mRNA-Binding Protein 3 (IMP3) in Well-Differentiated (WD), Moderately Differentiated (MD), and Poorly Differentiated (PD) Esophageal Adenocarcinoma IMP3 Staining WD MD PD Positive cases (b)/total cases (%) 30/57 (53) 28/32 (88) 35/43 (81) Moderate to strong intensity staining (2-3+)/total cases (%) 13/53 (23) 12/32 (38) 22/43 (51) Diffusely positive staining cases (>50%)/total cases (%) 14/57 (25) 18/32 (56) 24/43 (56) Mean intensity of staining/ standard error 0.86/0.12 1.5/0.17 1.7/0.17 Mean % of staining cells/ standard error 22/3.7 43/5.7 48/5.5 Significance of Differences in Means (a) IMP3 Staining WD Versus WD Versus WD Versus MD PD MD/PD Positive cases (b)/total cases (%) NA NA NA Moderate to strong intensity staining (2-3+)/total cases (%) NA NA NA Diffusely positive staining cases (>50%)/total cases (%) NA NA NA Mean intensity of staining/ standard error P = .003 P < .001 P < .001 Mean % of staining cells/ standard error P = .001 P < .001 P < .001 Abbreviation: NA, not applicable. (a) Determined by 2-tailed Student t test. (b) At least 1+ staining intensity in at least 5% of lesional cells. Table 3. Significance of Difference in Frequency of Insulin-Like Growth Factor II mRNA-Binding Protein 3 (IMP3) Expression in Esophageal Adenocarcinoma (EAC) and Correlation With Clinical and Pathologic Parameters Pathologic IMP3+ EAC Frequency or Clinical Cases, No./ IMP3+ EAC, Parameter Total (%) P Value (df) (a) Differentiation WD 30/57 (53) <.001 (2) MD 28/32 (88) PD 35/43 (81) Age, y <65 45/59 (76) .20 (1) [greater than or equal to] 65 43/67 (64) Sex Male 74/107 (69) .67 (1) Female 19/25 (76) Tumor location Proximal 1/1 (100) .74 (3) Mid 9/15 (60) Distal 29/41 (70) GEJ 48/67 (72) Margin status Negative 77/107 (72) .75 (1) Positive 11/17 (64) Lymph node status Negative 35/52 (67) .71 (1) Positive 54/75 (72) Stage 0 1/3 (33) .68 (4) 1 13/19 (68) 2 29/41 (70) 3 39/53 (74) 4 8/11 (73) Pathologic Correlation or Clinical Coefficient r, Parameter (P Value) Differentiation WD 0.37 (< .001) MD PD Age, y <65 -0.17 (.06) [greater than or equal to] 65 Sex Male 0.09 (.32) Female Tumor location Proximal 0.12 (.19) Mid Distal GEJ Margin status Negative 0.01 (.9) Positive Lymph node status Negative 0.02 (.85) Positive Stage 0 -0.05 (.57) 1 2 3 4 Abbreviations: df, degrees of freedom;GEJ, gastroesophageal junction; MD, moderately differentiated;PD, poorly differentiated; WD, well differentiated. (a) Determined by Pearson x2 test. Table 4. Significance of Differences in Frequency of Positive Insulin-Like Growth Factor II mRNA-Binding Protein 3 Expression in Esophageal Adenocarcinoma (EAC), Esophageal Columnar Dysplasia (ECD), Barrett Mucosa (BM), and Nonneoplastic Esophageal Mucosa (NNEM) Determined by Pearson [chi square] Test [chi square] (df = 1) P Value EAC versus NNEM 145 <.001 EAC versus ECD 18.5 <.001 EAC versus BM 44.3 <.001 ECD versus NNEM 30.1 <.001 ECD versus BM 5.88 .02 Abbreviation: df, degrees of freedom. Table 5. Full Section Insulin-Like Growth Factor II mRNA-Binding Protein 3 (IMP3) Staining in Esophageal Adenocarcinoma (EAC) and Esophageal Columnar Dysplasia (ECD) Case Differentiation High-grade ECD NA EAC No. 1 Moderate EAC No. 2 Moderate EAC No. 3 Moderate EAC No. 4 Poor EAC No. 5 Poor EAC No. 6 Well EAC No. 7 Well EAC No. 8 Poor EAC No. 9 (a) Poor EAC No. 10 (a) Poor EAC No. 11 Moderate EAC No. 12 Moderate EAC No. 13 (b) Poor Positive EAC cases (c)/total EAC cases (%) Moderate to strong intensity staining EAC (2-3+)/ total EAC cases (%) Diffusely positive staining EAC cases (<50%)/ total EAC cases (%) EAC mean intensity of staining/standard error EAC mean % of staining cells/standard error Percent of IMP3 Positive IMP3 Staining Lesional Case Intensity Staining High-grade ECD 2+ 80 EAC No. 1 1+ 10 EAC No. 2 2+ 100 EAC No. 3 1+ 25 EAC No. 4 2+ 90 EAC No. 5 2+ 80 EAC No. 6 0 NA EAC No. 7 1+ 25 EAC No. 8 0 NA EAC No. 9 (a) 1+ 50 EAC No. 10 (a) 3+ 30 EAC No. 11 1 80 EAC No. 12 0 NA EAC No. 13 (b) 1-3+ 50 Positive EAC cases (c)/total EAC cases (%) 10/13 (77) Moderate to strong intensity staining EAC (2-3+)/ total EAC cases (%) 5/10 (50) Diffusely positive staining EAC cases (<50%)/ total EAC cases (%) 4/10 (40) EAC mean intensity of staining/standard error 1.27/0.28 EAC mean % of staining cells/standard error 45/10.5 Abbreviation: NA, not applicable. (a) Mucinous and patchy signet ring cell differentiation. (b) EAC case with well and poorly differentiated areas. (c) At least 1+ staining intensity in at least 5% of lesional cells.