TY - JOUR
AU - Kuwano, H
AB - Abstract Background Scirrhous gastric carcinoma is characterized by excessive deposition of collagen in the stroma. However, the clinical significance of this fibrosis of the stomach has not been clarified. The aim of this study was to examine the fibrotic mechanism in several histological types of gastric carcinoma, and the combination of MUC1 and collagen type IV as a possible predictor of patient survival. Methods One hundred and two paraffin-embedded specimens of gastric carcinoma were examined by immunohistochemical staining using monoclonal antibodies against collagen type IV and MUC1. Results Collagen type IV-positive expression was significantly associated with depth of wall penetration (P = 0·025) and stage (P = 0·023). There was a significant relationship between MUC1-positive expression and interstitial collagen type IV-positive expression (P = 0·035). Survival was shorter for patients with the combination of MUC1-positive expression and interstitial collagen type IV-negative expression than for those with other expression patterns. Conclusion In patients with differentiated-type advanced gastric carcinoma, the combination of MUC1-positive and interstitial collagen type IV-negative expression may be a marker of unfavourable prognosis. Introduction Patients with scirrhous gastric carcinoma have a particularly poor prognosis1. This is a diffusely infiltrative tumour, characterized by abundant collagen deposition in the tumour stroma2,3. Mechanisms proposed for the formation of this desmoplastic stroma include the apparent increase caused by the collapse of the pre-existing matrix4, synthesis by neoplastic cells themselves5–8, and imbalance between collagen synthesis and degradation2. Many gastric carcinomas invading the submucosal layer invoke a desmoplastic response9. Barsky and Gopalakrishna10 investigated this fibrous response in a highly metastasizing mouse model, and found that the myofibroblast and collagenous composition were virtually identical to the composition of the desmoplastic response exhibited by human carcinomas. To elucidate the fibrotic mechanism in gastric carcinoma tissue, collagen type IV in both scirrhous carcinoma and other types of gastric cancer was evaluated by immunohistochemical staining using monoclonal antibodies. Collagen type IV is a major component of basement membranes11, has several binding sites with other matrix components as well as cellular receptors, and thus contributes not only to basement membrane structure but also to cell function12. Collagen type IV is involved in adhesion, migration and basement membrane invasion13. Mucins are high molecular weight glycoproteins that contain oligosaccharides14, and are major components of the mucus that protects the gastric epithelium from chemical and mechanical damage15. MUC1 is expressed widely in mucous cells of the antral epithelium, and focally in the oxyntic glands of the body and the pyloric glands of the antrum16–18. MUC1 is frequently overexpressed in carcinomas that show invasive growth, resulting in a poor prognosis for patients with gastric19 and colorectal20 cancer. Both collagen type IV and MUC1 are related to invasive growth in carcinoma13,21,22, but there are no reports specifically concerning the combined expression of collagen type IV and a mucin in advanced gastric cancer. In the present study, the expression of collagen type IV and MUC1 in advanced gastric cancer was examined by immunohistochemical staining. The objective of the study was to investigate the correlation between collagen type IV and MUC1, and to evaluate their combination in advanced gastric cancers as a possible predictor of patient survival. Methods Consecutive patients with advanced gastric carcinoma (37 differentiated and 65 undifferentiated cancers) who had undergone gastrectomy without any neoadjuvant treatment at the Department of General Surgical Science, Gunma University Graduate School of Medicine, between May 1996 and December 2000 were included. Patients who had palliative or non-curative operations were excluded. A curative surgical resection (subtotal or total gastrectomy) with D2 or more extensive lymph node dissection (curability class A or B, according to the Japanese Classification of Gastric Carcinoma23) had been performed in all patients. Characteristics of the patients are summarized in Table 1. Pathological diagnoses and classifications, such as depth of invasion, lymphatic and blood vessel penetration, and lymph node metastasis, were made according to the rules of the Japanese Research Society for Gastric Cancer23. Table 1 Characteristics of the 102 patients Mean(s.d.) (range) age (years) 63(11) (36–88) Sex ratio (M : F) 66 : 36 Tumour stage  IB 26  II 24  III 33  IV 19 Histological type  Papillary 1  Tubular adenocarcinoma   Well differentiated (tub1) 5   Moderately differentiated (tub2) 31  Poorly differentiated   Solid (por1) 30   Non-solid (por2) 23  Signet-ring cell carcinoma 12 Depth of tumour invasion  Muscularis propria 20  Subserosa 39  Serosa 37  Adjacent structures 6 Mean(s.d.) (range) age (years) 63(11) (36–88) Sex ratio (M : F) 66 : 36 Tumour stage  IB 26  II 24  III 33  IV 19 Histological type  Papillary 1  Tubular adenocarcinoma   Well differentiated (tub1) 5   Moderately differentiated (tub2) 31  Poorly differentiated   Solid (por1) 30   Non-solid (por2) 23  Signet-ring cell carcinoma 12 Depth of tumour invasion  Muscularis propria 20  Subserosa 39  Serosa 37  Adjacent structures 6 Open in new tab Table 1 Characteristics of the 102 patients Mean(s.d.) (range) age (years) 63(11) (36–88) Sex ratio (M : F) 66 : 36 Tumour stage  IB 26  II 24  III 33  IV 19 Histological type  Papillary 1  Tubular adenocarcinoma   Well differentiated (tub1) 5   Moderately differentiated (tub2) 31  Poorly differentiated   Solid (por1) 30   Non-solid (por2) 23  Signet-ring cell carcinoma 12 Depth of tumour invasion  Muscularis propria 20  Subserosa 39  Serosa 37  Adjacent structures 6 Mean(s.d.) (range) age (years) 63(11) (36–88) Sex ratio (M : F) 66 : 36 Tumour stage  IB 26  II 24  III 33  IV 19 Histological type  Papillary 1  Tubular adenocarcinoma   Well differentiated (tub1) 5   Moderately differentiated (tub2) 31  Poorly differentiated   Solid (por1) 30   Non-solid (por2) 23  Signet-ring cell carcinoma 12 Depth of tumour invasion  Muscularis propria 20  Subserosa 39  Serosa 37  Adjacent structures 6 Open in new tab Histology and immunohistochemistry The resected specimens were fixed with 10 per cent formaldehyde, and embedded in paraffin blocks. The site of deepest tumour invasion was selected for microscopic examination. Each section was subdivided into serial sections 3 µm thick and used for immunohistochemistry. The presence of a desmoplastic response was defined as a condition characterized by the formation of excessive connective tissue stroma surrounding gastric carcinomas invading the submucosal layer. Sections from one or two representative paraffin blocks at the invasive front of 102 gastric carcinomas were stained for collagen type IV (MAB1910, mouse monoclonal, 1 : 500; Chemicon International, Billerica, Massachusetts, USA) and MUC1 (Ma552, mouse monoclonal, 1 : 200; Novocastra Laboratories, Newcastle upon Tyne, UK). Sections were deparaffinized in xylene and hydrated through a graded ethanol series. Antigens were retrieved by microwaving sections on slides in a 0·01-mol/l citrate buffer, pH 6·0, for 20 min. After the sections had been rinsed in phosphate-buffered saline (PBS), endogenous peroxidase was blocked with 3 per cent hydrogen peroxide in absolute 100 per cent methanol for 30 min. After further rinsing in PBS, the sections were blocked with normal rabbit serum for 30 min and then incubated overnight at 4 °C in humid chambers with one of the primary monoclonal antibodies directed against collagen type IV or MUC1. The sections were rinsed in PBS and then subjected to the streptavidin–biotin method with Histofine® SAB-PO (mouse) kits (Nichirei, Tokyo, Japan). The sections were incubated for 20 min at room temperature with biotinylated rabbit antimouse IgG, A and M (Nichirei) for 20 min, followed by three washes in PBS, and then incubated with peroxidase-conjugated streptavidin for 10 min at room temperature, followed by a further three PBS washes. The bound antibody complex was visualized by a reaction in 0·01 per cent hydrogen peroxide and 0·05 per cent 3, 3′-diaminobenzidine for 3 min. A light counterstaining with Mayer's haematoxylin was performed. Immunostaining was scored by two observers with no previous knowledge of the clinicopathological details. Staining was evaluated in all layers of the carcinomas. Expression of collagen type IV was classified as interstitial or cytoplasmic, and considered positive if at least 10 per cent of the interstitium and the cytoplasm of the neoplastic cells was stained. Expression of MUC1 was also considered positive if at least 10 per cent of the neoplastic cells was stained16. Statistical analysis Analyses were conducted using the JMP version 5 for Windows® software package (SAS Institute, Cary, North Carolina, USA). Distributions of parameters of antigen expression were compared using χ2 and Mann–Whitney U tests; when there were fewer than five parameters, Fisher's exact probability test was used. Student's t test was used to determine group differences. Survival analyses were performed using the Kaplan–Meier method, and differences between survival curves were tested with the log rank test. Multivariable survival analysis was performed using the Cox proportional hazards model. P < 0·050 was considered statistically significant. Results There were 102 patients with advanced gastric carcinoma (37 differentiated and 65 undifferentiated cancers) who had undergone gastrectomy. The group comprised 66 men and 36 women with a mean(s.d.) age of 63(11) years. Distal subtotal gastrectomy was performed in 59 patients and total gastrectomy in 43. Patients with stage II, III or IV disease in the Japanese staging system23 received adjuvant chemotherapy. Follow-up ranged from 0·1 to 5·0 (median 3·5) years. Expression of collagen type IV and MUC1 in carcinoma tissue Positive expression of collagen type IV and MUC1 was observed in 68 (66·7 per cent) and 44 (43·1 per cent) of the 102 samples respectively. Positive expression of collagen type IV was observed in the cytoplasm and interstitium of gastric cancer cells (Fig. 1a,b). Collagen type IV was present irregularly along the thick bundles of collagen fibres in the interstitial stroma (Fig. 1b). MUC1 was expressed in the cytoplasm and apical membrane of carcinoma cells (Fig. 1c). Fig. 1 Open in new tabDownload slide Immunohistochemical staining for collagen type IV and MUC1. a Collagen type IV expressed in cytoplasm of differentiated-type gastric carcinoma cells. b Collagen type IV expressed in interstitial stroma of undifferentiated-type gastric carcinoma cells. c MUC1 expressed in cytoplasm and apical membrane of differentiated-type gastric carcinoma. (All original magnification ×200, counterstained with Mayer's haematoxylin) Relationship between interstitial collagen type IV expression and desmoplastic response A desmoplastic response was observed in 15 (41 per cent) of 37 samples of differentiated adenocarcinoma and in 47 (73 per cent) of 65 samples of undifferentiated adenocarcinoma. There was positive expression of collagen type IV in five of 15 samples of differentiated and in 19 of 47 samples of undifferentiated adenocarcinoma with a desmoplastic response. Interstitial collagen type IV-positive expression in differentiated and undifferentiated adenocarcinoma showed a significant correlation with desmoplastic response (P = 0·007 and P < 0·001 respectively) (Table 2). Table 2 Relationship between interstitial collagen type IV expression and desmoplastic response in patients with differentiated and undifferentiated adenocarcinoma . Desmoplastic response (differentiated carcinoma) . Desmoplastic response (undifferentiated carcinoma) . . Present (n = 15) . Absent (n = 22) . P* . Present (n = 47) . Absent (n = 18) . P* . Interstitial   collagen   type IV  Positive 5 0 0·007 19 0 < 0·001  Negative 10 22 28 18 . Desmoplastic response (differentiated carcinoma) . Desmoplastic response (undifferentiated carcinoma) . . Present (n = 15) . Absent (n = 22) . P* . Present (n = 47) . Absent (n = 18) . P* . Interstitial   collagen   type IV  Positive 5 0 0·007 19 0 < 0·001  Negative 10 22 28 18 * Fisher's exact test. Open in new tab Table 2 Relationship between interstitial collagen type IV expression and desmoplastic response in patients with differentiated and undifferentiated adenocarcinoma . Desmoplastic response (differentiated carcinoma) . Desmoplastic response (undifferentiated carcinoma) . . Present (n = 15) . Absent (n = 22) . P* . Present (n = 47) . Absent (n = 18) . P* . Interstitial   collagen   type IV  Positive 5 0 0·007 19 0 < 0·001  Negative 10 22 28 18 . Desmoplastic response (differentiated carcinoma) . Desmoplastic response (undifferentiated carcinoma) . . Present (n = 15) . Absent (n = 22) . P* . Present (n = 47) . Absent (n = 18) . P* . Interstitial   collagen   type IV  Positive 5 0 0·007 19 0 < 0·001  Negative 10 22 28 18 * Fisher's exact test. Open in new tab Collagen type IV and MUC1 expression, and clinicopathological features Tables 3 and 4 show the relationship between clinicopathological features and collagen type IV and MUC1 antigen expression. Sex, histological type, histological pattern, lymph node metastasis, lymphatic invasion, venous invasion and macroscopic type showed no correlation with expression of these antigens. Depth of wall penetration and stage were correlated with collagen type IV antigen expression, and age showed a positive correlation with MUC1 expression. Table 3 Collagen type IV expression and clinicopathological features in 102 advanced gastric cancers . Collagen type IV positive (n = 68) . Collagen type IV negative (n = 34) . P . Mean(s.d.) age (years) 64·2(9·7) 61·6(12·4) 0·280* Sex 1·000†  M 44 (67) 22 (33)  F 24 (67) 12 (33) Histological type 0·884†  Differentiated 25 (68) 12 (32)  Undifferentiated 43 (66) 22 (34) Histological pattern 0·826‡  Papillary 1 (100) 0 (0)  Well differentiated (tub1) 5 (100) 0 (0)  Moderately differentiated (tub2) 19 (61) 12 (39)  Solid (por1) 21 (70) 9 (30)  Non-solid (por2) 11 (48) 12 (52)  Signet-ring cell carcinoma 11 (92) 1 (8) Japanese stage 0·023‡  IB 19 (73) 7 (27)  II 17 (71) 7 (29)  III 25 (76) 8 (24)  IV 7 (37) 12 (63) Lymph node metastasis 0·653†  No 23 (70) 10 (30)  Yes 45 (65) 24 (35) Lymphatic invasion 0·372†  No 3 (50) 3 (50)  Yes 65 (68) 31 (32) Venous invasion 0·571†  No 30 (70) 13 (30)  Yes 38 (64) 21 (36) Depth of wall penetration 0·025‡  Muscularis propria 10 (50) 10 (50)  Subserosa 32 (82) 7 (18)  Serosa/adjacent structures 26 (60) 17 (40) Macroscopic type 0·771‡  1 4 (80) 1 (20)  2 27 (68) 13 (32)  3 23 (61) 15 (39)  4 13 (72) 5 (28)  5 1 (100) 0 (0) . Collagen type IV positive (n = 68) . Collagen type IV negative (n = 34) . P . Mean(s.d.) age (years) 64·2(9·7) 61·6(12·4) 0·280* Sex 1·000†  M 44 (67) 22 (33)  F 24 (67) 12 (33) Histological type 0·884†  Differentiated 25 (68) 12 (32)  Undifferentiated 43 (66) 22 (34) Histological pattern 0·826‡  Papillary 1 (100) 0 (0)  Well differentiated (tub1) 5 (100) 0 (0)  Moderately differentiated (tub2) 19 (61) 12 (39)  Solid (por1) 21 (70) 9 (30)  Non-solid (por2) 11 (48) 12 (52)  Signet-ring cell carcinoma 11 (92) 1 (8) Japanese stage 0·023‡  IB 19 (73) 7 (27)  II 17 (71) 7 (29)  III 25 (76) 8 (24)  IV 7 (37) 12 (63) Lymph node metastasis 0·653†  No 23 (70) 10 (30)  Yes 45 (65) 24 (35) Lymphatic invasion 0·372†  No 3 (50) 3 (50)  Yes 65 (68) 31 (32) Venous invasion 0·571†  No 30 (70) 13 (30)  Yes 38 (64) 21 (36) Depth of wall penetration 0·025‡  Muscularis propria 10 (50) 10 (50)  Subserosa 32 (82) 7 (18)  Serosa/adjacent structures 26 (60) 17 (40) Macroscopic type 0·771‡  1 4 (80) 1 (20)  2 27 (68) 13 (32)  3 23 (61) 15 (39)  4 13 (72) 5 (28)  5 1 (100) 0 (0) Values in parentheses are percentages unless indicated otherwise. * Student's t test; † χ2 test; ‡ Mann–Whitney U test. Open in new tab Table 3 Collagen type IV expression and clinicopathological features in 102 advanced gastric cancers . Collagen type IV positive (n = 68) . Collagen type IV negative (n = 34) . P . Mean(s.d.) age (years) 64·2(9·7) 61·6(12·4) 0·280* Sex 1·000†  M 44 (67) 22 (33)  F 24 (67) 12 (33) Histological type 0·884†  Differentiated 25 (68) 12 (32)  Undifferentiated 43 (66) 22 (34) Histological pattern 0·826‡  Papillary 1 (100) 0 (0)  Well differentiated (tub1) 5 (100) 0 (0)  Moderately differentiated (tub2) 19 (61) 12 (39)  Solid (por1) 21 (70) 9 (30)  Non-solid (por2) 11 (48) 12 (52)  Signet-ring cell carcinoma 11 (92) 1 (8) Japanese stage 0·023‡  IB 19 (73) 7 (27)  II 17 (71) 7 (29)  III 25 (76) 8 (24)  IV 7 (37) 12 (63) Lymph node metastasis 0·653†  No 23 (70) 10 (30)  Yes 45 (65) 24 (35) Lymphatic invasion 0·372†  No 3 (50) 3 (50)  Yes 65 (68) 31 (32) Venous invasion 0·571†  No 30 (70) 13 (30)  Yes 38 (64) 21 (36) Depth of wall penetration 0·025‡  Muscularis propria 10 (50) 10 (50)  Subserosa 32 (82) 7 (18)  Serosa/adjacent structures 26 (60) 17 (40) Macroscopic type 0·771‡  1 4 (80) 1 (20)  2 27 (68) 13 (32)  3 23 (61) 15 (39)  4 13 (72) 5 (28)  5 1 (100) 0 (0) . Collagen type IV positive (n = 68) . Collagen type IV negative (n = 34) . P . Mean(s.d.) age (years) 64·2(9·7) 61·6(12·4) 0·280* Sex 1·000†  M 44 (67) 22 (33)  F 24 (67) 12 (33) Histological type 0·884†  Differentiated 25 (68) 12 (32)  Undifferentiated 43 (66) 22 (34) Histological pattern 0·826‡  Papillary 1 (100) 0 (0)  Well differentiated (tub1) 5 (100) 0 (0)  Moderately differentiated (tub2) 19 (61) 12 (39)  Solid (por1) 21 (70) 9 (30)  Non-solid (por2) 11 (48) 12 (52)  Signet-ring cell carcinoma 11 (92) 1 (8) Japanese stage 0·023‡  IB 19 (73) 7 (27)  II 17 (71) 7 (29)  III 25 (76) 8 (24)  IV 7 (37) 12 (63) Lymph node metastasis 0·653†  No 23 (70) 10 (30)  Yes 45 (65) 24 (35) Lymphatic invasion 0·372†  No 3 (50) 3 (50)  Yes 65 (68) 31 (32) Venous invasion 0·571†  No 30 (70) 13 (30)  Yes 38 (64) 21 (36) Depth of wall penetration 0·025‡  Muscularis propria 10 (50) 10 (50)  Subserosa 32 (82) 7 (18)  Serosa/adjacent structures 26 (60) 17 (40) Macroscopic type 0·771‡  1 4 (80) 1 (20)  2 27 (68) 13 (32)  3 23 (61) 15 (39)  4 13 (72) 5 (28)  5 1 (100) 0 (0) Values in parentheses are percentages unless indicated otherwise. * Student's t test; † χ2 test; ‡ Mann–Whitney U test. Open in new tab Table 4 MUC1 expression and clinicopathological features in 102 advanced gastric cancers . MUC1 positive (n = 44) . MUC1 negative (n = 58) . P . Mean(s.d.) age (years) 65·8(10·5) 61·5(10·5) 0·021* Sex 0·290†  M 31 (47) 35 (53)  F 13 (36) 23 (64) Histological type 0·397†  Differentiated 18 (49) 19 (51)  Undifferentiated 26 (40) 39 (60) Histological pattern 0·087‡  Papillary 1 (100) 0 (0)  Well differentiated (tub1) 4 (80) 1 (20)  Moderately differentiated (tub2) 13 (42) 18 (58)  Solid (por1) 14 (47) 16 (53)  Non-solid (por2) 10 (43) 13 (57)  Signet-ring cell carcinoma 2 (17) 10 (83) Japanese stage 0·273‡  IB 9 (35) 17 (65)  II 7 (29) 17 (71)  III 18 (55) 15 (45)  IV 10 (53) 9 (47) Lymph node metastasis 0·070†  No 10 (30) 23 (70)  Yes 34 (49) 35 (51) Lymphatic invasion 0·177†  No 1 (17) 5 (83)  Yes 43 (45) 53 (55) Venous invasion 0·824†  No 18 (42) 25 (58)  Yes 26 (44) 33 (56) Depth of wall penetration 0·149‡  Muscularis propria 5 (25) 15 (75)  Subserosa 17 (44) 22 (56)  Serosa/adjacent structures 22 (51) 21 (49) Macroscopic type 0·069‡  1 3 (60) 2 (40)  2 11 (27) 29 (73)  3 22 (58) 16 (42)  4 8 (44) 10 (56)  5 0 (0) 1 (100) . MUC1 positive (n = 44) . MUC1 negative (n = 58) . P . Mean(s.d.) age (years) 65·8(10·5) 61·5(10·5) 0·021* Sex 0·290†  M 31 (47) 35 (53)  F 13 (36) 23 (64) Histological type 0·397†  Differentiated 18 (49) 19 (51)  Undifferentiated 26 (40) 39 (60) Histological pattern 0·087‡  Papillary 1 (100) 0 (0)  Well differentiated (tub1) 4 (80) 1 (20)  Moderately differentiated (tub2) 13 (42) 18 (58)  Solid (por1) 14 (47) 16 (53)  Non-solid (por2) 10 (43) 13 (57)  Signet-ring cell carcinoma 2 (17) 10 (83) Japanese stage 0·273‡  IB 9 (35) 17 (65)  II 7 (29) 17 (71)  III 18 (55) 15 (45)  IV 10 (53) 9 (47) Lymph node metastasis 0·070†  No 10 (30) 23 (70)  Yes 34 (49) 35 (51) Lymphatic invasion 0·177†  No 1 (17) 5 (83)  Yes 43 (45) 53 (55) Venous invasion 0·824†  No 18 (42) 25 (58)  Yes 26 (44) 33 (56) Depth of wall penetration 0·149‡  Muscularis propria 5 (25) 15 (75)  Subserosa 17 (44) 22 (56)  Serosa/adjacent structures 22 (51) 21 (49) Macroscopic type 0·069‡  1 3 (60) 2 (40)  2 11 (27) 29 (73)  3 22 (58) 16 (42)  4 8 (44) 10 (56)  5 0 (0) 1 (100) Values in parentheses are percentages unless indicated otherwise. * Student's t test; † χ2 test; ‡ Mann–Whitney U test. Open in new tab Table 4 MUC1 expression and clinicopathological features in 102 advanced gastric cancers . MUC1 positive (n = 44) . MUC1 negative (n = 58) . P . Mean(s.d.) age (years) 65·8(10·5) 61·5(10·5) 0·021* Sex 0·290†  M 31 (47) 35 (53)  F 13 (36) 23 (64) Histological type 0·397†  Differentiated 18 (49) 19 (51)  Undifferentiated 26 (40) 39 (60) Histological pattern 0·087‡  Papillary 1 (100) 0 (0)  Well differentiated (tub1) 4 (80) 1 (20)  Moderately differentiated (tub2) 13 (42) 18 (58)  Solid (por1) 14 (47) 16 (53)  Non-solid (por2) 10 (43) 13 (57)  Signet-ring cell carcinoma 2 (17) 10 (83) Japanese stage 0·273‡  IB 9 (35) 17 (65)  II 7 (29) 17 (71)  III 18 (55) 15 (45)  IV 10 (53) 9 (47) Lymph node metastasis 0·070†  No 10 (30) 23 (70)  Yes 34 (49) 35 (51) Lymphatic invasion 0·177†  No 1 (17) 5 (83)  Yes 43 (45) 53 (55) Venous invasion 0·824†  No 18 (42) 25 (58)  Yes 26 (44) 33 (56) Depth of wall penetration 0·149‡  Muscularis propria 5 (25) 15 (75)  Subserosa 17 (44) 22 (56)  Serosa/adjacent structures 22 (51) 21 (49) Macroscopic type 0·069‡  1 3 (60) 2 (40)  2 11 (27) 29 (73)  3 22 (58) 16 (42)  4 8 (44) 10 (56)  5 0 (0) 1 (100) . MUC1 positive (n = 44) . MUC1 negative (n = 58) . P . Mean(s.d.) age (years) 65·8(10·5) 61·5(10·5) 0·021* Sex 0·290†  M 31 (47) 35 (53)  F 13 (36) 23 (64) Histological type 0·397†  Differentiated 18 (49) 19 (51)  Undifferentiated 26 (40) 39 (60) Histological pattern 0·087‡  Papillary 1 (100) 0 (0)  Well differentiated (tub1) 4 (80) 1 (20)  Moderately differentiated (tub2) 13 (42) 18 (58)  Solid (por1) 14 (47) 16 (53)  Non-solid (por2) 10 (43) 13 (57)  Signet-ring cell carcinoma 2 (17) 10 (83) Japanese stage 0·273‡  IB 9 (35) 17 (65)  II 7 (29) 17 (71)  III 18 (55) 15 (45)  IV 10 (53) 9 (47) Lymph node metastasis 0·070†  No 10 (30) 23 (70)  Yes 34 (49) 35 (51) Lymphatic invasion 0·177†  No 1 (17) 5 (83)  Yes 43 (45) 53 (55) Venous invasion 0·824†  No 18 (42) 25 (58)  Yes 26 (44) 33 (56) Depth of wall penetration 0·149‡  Muscularis propria 5 (25) 15 (75)  Subserosa 17 (44) 22 (56)  Serosa/adjacent structures 22 (51) 21 (49) Macroscopic type 0·069‡  1 3 (60) 2 (40)  2 11 (27) 29 (73)  3 22 (58) 16 (42)  4 8 (44) 10 (56)  5 0 (0) 1 (100) Values in parentheses are percentages unless indicated otherwise. * Student's t test; † χ2 test; ‡ Mann–Whitney U test. Open in new tab Relationship between collagen type IV and MUC1 expression Interstitial collagen type IV-positive expression showed a significant correlation with MUC1 expression (P = 0·035), although cytoplasmic collagen type IV-positive expression did not (P = 0·691) (Table 5). Table 5 Relationship between interstitial and cytoplasmic collagen type IV and MUC1 expression . Interstitial collagen type IV . Cytoplasmic collagen type IV . . Positive (n = 24) . Negative (n = 78) . P* . Positive (n = 56) . Negative (n = 46) . P* . MUC1  Absent 9 49 0·035 33 25 0·691  Present 15 29 23 21 . Interstitial collagen type IV . Cytoplasmic collagen type IV . . Positive (n = 24) . Negative (n = 78) . P* . Positive (n = 56) . Negative (n = 46) . P* . MUC1  Absent 9 49 0·035 33 25 0·691  Present 15 29 23 21 * Fisher's exact test. Open in new tab Table 5 Relationship between interstitial and cytoplasmic collagen type IV and MUC1 expression . Interstitial collagen type IV . Cytoplasmic collagen type IV . . Positive (n = 24) . Negative (n = 78) . P* . Positive (n = 56) . Negative (n = 46) . P* . MUC1  Absent 9 49 0·035 33 25 0·691  Present 15 29 23 21 . Interstitial collagen type IV . Cytoplasmic collagen type IV . . Positive (n = 24) . Negative (n = 78) . P* . Positive (n = 56) . Negative (n = 46) . P* . MUC1  Absent 9 49 0·035 33 25 0·691  Present 15 29 23 21 * Fisher's exact test. Open in new tab Positive expression of interstitial collagen type IV was significantly correlated with MUC1 expression in undifferentiated adenocarcinoma (P = 0·025), but not in differentiated adenocarcinoma (P = 0·660) (Table 6). Table 6 Relationship between the expression of interstitial collagen type IV and MUC1 in differentiated and undifferentiated adenocarcinoma . Interstitial collagen type IV (differentiated carcinoma) . Interstitial collagen type IV (undifferentiated carcinoma) . . Positive (n = 5) . Negative (n = 32) . P* . Positive (n = 19) . Negative (n = 46) . P* . MUC1  Absent 2 17 0·660 7 32 0·025  Present 3 15 12 14 . Interstitial collagen type IV (differentiated carcinoma) . Interstitial collagen type IV (undifferentiated carcinoma) . . Positive (n = 5) . Negative (n = 32) . P* . Positive (n = 19) . Negative (n = 46) . P* . MUC1  Absent 2 17 0·660 7 32 0·025  Present 3 15 12 14 * Fisher's exact test. Open in new tab Table 6 Relationship between the expression of interstitial collagen type IV and MUC1 in differentiated and undifferentiated adenocarcinoma . Interstitial collagen type IV (differentiated carcinoma) . Interstitial collagen type IV (undifferentiated carcinoma) . . Positive (n = 5) . Negative (n = 32) . P* . Positive (n = 19) . Negative (n = 46) . P* . MUC1  Absent 2 17 0·660 7 32 0·025  Present 3 15 12 14 . Interstitial collagen type IV (differentiated carcinoma) . Interstitial collagen type IV (undifferentiated carcinoma) . . Positive (n = 5) . Negative (n = 32) . P* . Positive (n = 19) . Negative (n = 46) . P* . MUC1  Absent 2 17 0·660 7 32 0·025  Present 3 15 12 14 * Fisher's exact test. Open in new tab Relationship of survival to MUC1 and interstitial collagen type IV status in differentiated and undifferentiated adenocarcinoma with a desmoplastic response Patients with MUC1-positive antigen staining of differentiated-type gastric carcinoma had a significantly worse survival than those with MUC1-negative staining (P < 0·001) (Fig. 2a). Although MUC1-positive undifferentiated-type gastric carcinomas were associated with a poorer survival than MUC1-negative undifferentiated carcinomas, it was apparent only for years 1–3 (Fig. 2b). Fig. 2 Open in new tabDownload slide Kaplan–Meier survival curves of patients with MUC1-positive and MUC1-negative tumours in a differentiated-type and b undifferentiated-type gastric carcinomas. aP < 0·001, bP = 0·391 (log rank test) Patients with collagen type IV-negative carcinomas had a poorer survival than those with collagen type IV-positive carcinomas in both differentiated and undifferentiated adenocarcinomas with a desmoplastic response, although the difference was not statistically significant (Fig. 3a,b). Fig. 3 Open in new tabDownload slide Kaplan–Meier survival curves of patients with interstitial collagen type IV-positive and -negative expression in a differentiated-type and b undifferentiated-type gastric carcinomas with a desmoplastic response. aP = 0·135, bP = 0·411 (log rank test) Relationship of survival to combined MUC1 and interstitial collagen type IV status As interstitial collagen type IV expression correlated with the expression of MUC1, the relationship between the combined MUC1/interstitial collagen type IV status and postoperative survival was assessed in the 102 patients. Patients in each of the differentiated and undifferentiated adenocarcinoma groups were further divided into four subgroups, as shown in Fig. 4. In the differentiated group, the 5-year survival rate for MUC1 positivity/collagen type IV positivity, MUC1 negativity/collagen type IV positivity, MUC1 positivity/collagen type IV negativity and MUC1 negativity/collagen type IV negativity was 67, 100, 35 and 94 per cent respectively. Thus, patients with MUC1-positive/interstitial collagen type IV-negative expression in differentiated adenocarcinomas had a significantly lower survival after surgery (P < 0·001) (Fig. 4a). In the undifferentiated group, the 5-year survival rate in the respective subgroups was 49, 71, 43 and 52 per cent. Although MUC1-positive/interstitial collagen type IV-negative expression status was again associated with the lowest survival rate, this was not significantly different (P = 0·442) (Fig. 4b). Fig. 4 Open in new tabDownload slide Kaplan–Meier survival curves of patients in the four subgroups of MUC1 and interstitial collagen type IV co-expression in a differentiated-type and b undifferentiated-type gastric carcinomas. A, MUC1-positive/collagen type IV-positive; B, MUC1-negative/collagen type IV-positive; C, MUC1-positive/collagen type IV-negative; D, MUC1-negative/collagen type IV-negative expression. aP < 0·001, bP = 0·442 (MUC1-positive/collagen type IV-negative versus MUC1-negative/collagen type IV-negative expression pattern, log rank test) Analysis of factors related to survival In patients with differentiated adenocarcinoma, univariable analysis showed age, lymph node metastasis, Japanese stage grouping, MUC1 and MUC1-positive/interstitial collagen type IV-negative expression to be significantly related to survival (Table 7). In the multivariable analysis of these significant factors, age (P < 0·001), stage grouping (P = 0·005) and the combination of MUC1-positive/interstitial collagen type IV-negative expression (P = 0·003) were independently and significantly associated with a poor prognosis; the combined antigen status was associated with a relative risk of death 5·38 times greater than that for other patients (Table 7). Table 7 Univariable and multivariable analysis of factors related to survival in differential adenocarcinoma . Univariable P* . Multivariable† . . Hazard ratio . P . Age (years) < 0·001 1·16 (1·06, 1·30) < 0·001 Tumour diameter (mm) 0·580 Depth of wall penetration (muscularis propria versus others) 0·118 Lymph node metastasis (present versus absent) 0·022 0·97 (0·19, 3·82) 0·968 Japanese stage grouping (I/II versus II/IV) 0·022 4·89 (1·54, 26·78) 0·005 Lymphatic invasion (present versus absent) 0·944 Venous invasion (present versus absent) 0·073 MUC1 (positive versus negative) < 0·001 0·92 (0·18, 4·70) 0·906 MUC1 positive (collagen type IV negative versus others) < 0·001 5·38 (1·66, 29·34) 0·003 . Univariable P* . Multivariable† . . Hazard ratio . P . Age (years) < 0·001 1·16 (1·06, 1·30) < 0·001 Tumour diameter (mm) 0·580 Depth of wall penetration (muscularis propria versus others) 0·118 Lymph node metastasis (present versus absent) 0·022 0·97 (0·19, 3·82) 0·968 Japanese stage grouping (I/II versus II/IV) 0·022 4·89 (1·54, 26·78) 0·005 Lymphatic invasion (present versus absent) 0·944 Venous invasion (present versus absent) 0·073 MUC1 (positive versus negative) < 0·001 0·92 (0·18, 4·70) 0·906 MUC1 positive (collagen type IV negative versus others) < 0·001 5·38 (1·66, 29·34) 0·003 Values in parentheses are 95 per cent confidence intervals. * Log rank test; † Cox proportional hazards model. Open in new tab Table 7 Univariable and multivariable analysis of factors related to survival in differential adenocarcinoma . Univariable P* . Multivariable† . . Hazard ratio . P . Age (years) < 0·001 1·16 (1·06, 1·30) < 0·001 Tumour diameter (mm) 0·580 Depth of wall penetration (muscularis propria versus others) 0·118 Lymph node metastasis (present versus absent) 0·022 0·97 (0·19, 3·82) 0·968 Japanese stage grouping (I/II versus II/IV) 0·022 4·89 (1·54, 26·78) 0·005 Lymphatic invasion (present versus absent) 0·944 Venous invasion (present versus absent) 0·073 MUC1 (positive versus negative) < 0·001 0·92 (0·18, 4·70) 0·906 MUC1 positive (collagen type IV negative versus others) < 0·001 5·38 (1·66, 29·34) 0·003 . Univariable P* . Multivariable† . . Hazard ratio . P . Age (years) < 0·001 1·16 (1·06, 1·30) < 0·001 Tumour diameter (mm) 0·580 Depth of wall penetration (muscularis propria versus others) 0·118 Lymph node metastasis (present versus absent) 0·022 0·97 (0·19, 3·82) 0·968 Japanese stage grouping (I/II versus II/IV) 0·022 4·89 (1·54, 26·78) 0·005 Lymphatic invasion (present versus absent) 0·944 Venous invasion (present versus absent) 0·073 MUC1 (positive versus negative) < 0·001 0·92 (0·18, 4·70) 0·906 MUC1 positive (collagen type IV negative versus others) < 0·001 5·38 (1·66, 29·34) 0·003 Values in parentheses are 95 per cent confidence intervals. * Log rank test; † Cox proportional hazards model. Open in new tab In patients with undifferentiated adenocarcinoma, univariable analysis showed that MUC1 and MUC1-positive/interstitial collagen type IV-negative expression were not related to survival (data not shown). Discussion The relationship between expression of collagen type IV and the clinicopathological features of gastric carcinomas has not been studied previously, although others have reported that the staining patterns of collagen type IV in the basement membrane are largely related to histological type in gastric carcinomas24. In the present study, expression of collagen type IV was related to clinical characteristics such as depth of wall penetration and pathological stage, indicating that collagen type IV-positive expression is a marker of advanced disease. Positive expression of collagen type IV was observed in two-thirds of resected advanced gastric cancers, in both the cytoplasm (54·9 per cent) and the interstitium (23·5 per cent). On histological examination, interstitial collagen type IV-positive expression tended to be observed in undifferentiated adenocarcinoma. Interstitial collagen type IV positivity showed a significant correlation with desmoplastic response, although it was not observed in all differentiated and undifferentiated adenocarcinomas with a desmoplastic response. MUC1 expression is associated with invasive or metastatic properties of carcinoma cells, resulting in a poor prognosis for patients with gastric cancer19,25,26, and is correlated with histological type25. Although Akyürek and colleagues25 reported that MUC1 positivity was associated with pathological stage and worse survival, in the present study MUC1 expression was related only to patient age. This difference may be due to the different antibodies used in the two studies and definitions of positivity. Positive interstitial collagen type IV expression in all 102 carcinomas was significantly correlated with MUC1 expression. Further analysis indicated that positive interstitial collagen type IV expression in undifferentiated adenocarcinoma was significantly correlated with MUC1 expression. With regard to the correlation between MUC1 and collagen type IV, Satoh et al.27 reported that MUC1 expression in tumour cells resulted in increased in vitro invasiveness and motility, and reduced adhesion to extracellular matrix components including collagen types I and IV, and laminin. Decreased adhesion to the extracellular matrix via MUC1 expression may decrease integrin affinity to some ligands and thereby enhance motility. These changes in MUC1 and collagen type IV could facilitate tumour invasion and metastasis, and affect prognosis. The prognostic value of MUC1 status combined with collagen type IV expression relating to survival in the present study revealed that survival of those with interstitial collagen type IV-positive status was more favourable than that of patients with other expression patterns. Patients with MUC1-negative/interstitial collagen type IV-positive expression, particularly those with differentiated adenocarcinoma, had a higher survival rate than patients with other expression patterns. Conversely, the survival of patients with MUC1-positive/interstitial collagen type IV-negative differentiated adenocarcinomas was significantly shorter than that for other patients. In the multivariable analysis, this resulted in a relative risk of death 5·38 times greater than that of other patients. These immunohistochemical findings merit further evaluation to see whether they might be clinically useful when considering adjuvant chemotherapies after surgical resection. Acknowledgements The authors declare no conflict of interest. References 1 Ming S-C . Tumors of the stomach. In Atlas of Tumor Pathology , series 2, Fasc 7. Tumors of the Esophagus and Stomach, Firminger HI (ed.). Armed Forces Institute of Pathology : Washington, DC , 1973 ; 81 – 255 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 2 Matsui H , Kubochi K, Shimada A, Hasumi A. Desmoplastic response in scirrhous gastric carcinoma: imbalance between collagen synthesis and degradation . Anticancer Res 2000 ; 20 : 4733 – 4738 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 3 Minamoto T , Ooi A, Okada Y, Mai M, Nagai Y, Nakanishi I. Desmoplastic reaction of gastric carcinoma: a light- and electron-microscopic immunohistochemical analysis using collagen type-specific antibodies . Hum Pathol 1988 ; 19 : 815 – 821 . Google Scholar Crossref Search ADS PubMed WorldCat 4 Jackson JG , Orr JW. The ducts of carcinomatous breasts, with particular reference to connective-tissue changes . J Pathol Bacteriol 1957 ; 74 : 265 – 273 . Google Scholar Crossref Search ADS WorldCat 5 Takeuchi T . Collagen producing carcinoma: special reference to fibrogenesis of the stomach. In Collagen Degradation and Mammalian Collagenase , Tsuchiya M, Tamayo RP, Okazaki I, Maruyama K (eds). Excerpta Medica : Amsterdam , 1982 ; 63 – 71 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 6 Sakakibara K . An active role of neoplastic cells in desmoplasia of cancerous tissues. In Collagen Degradation and Mammalian Collagenase , Tsuchiya M, Tamayo RP, Okazaki I, Maruyama K (eds). Excerpta Medica : Amsterdam , 1982 ; 75 – 82 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 7 Sakakibara K , Suzuki T, Motoyama T, Watanabe H, Nagai Y. Biosynthesis of an interstitial type of collagen by cloned human gastric carcinoma cells . Cancer Res 1982 ; 42 : 2019 – 2027 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 8 Al-Adnani MS , Kirrane JA, McGee JO'D. Inappropriate production of collagen and prolyl hydroxylase by human breast cancer cells in vivo . Br J Cancer 1975 ; 31 : 653 – 660 . Google Scholar Crossref Search ADS PubMed WorldCat 9 Nakamura K . Histogenesis of gastric carcinoma and its clinicopathological significance. In Gastric Cancer. Springer : Tokyo , 1993 ; 112 – 132 . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC 10 Barsky SH , Gopalakrishna R. An experimental model for studying the desmoplastic response to tumor invasion . Cancer Lett 1987 ; 35 : 271 – 279 . Google Scholar Crossref Search ADS PubMed WorldCat 11 Oka Y , Naito I, Manabe K, Sado Y, Matsushima H, Ninomiya Y et al. Distribution of collagen type IV alpha1–6 chains in human normal colorectum and colorectal cancer demonstrated by immunofluorescence staining using chain-specific epitope-defined monoclonal antibodies . J Gastroenterol Hepatol 2002 ; 17 : 980 – 986 . Google Scholar Crossref Search ADS PubMed WorldCat 12 Timpl R . Structure and biological activity of basement membrane proteins . Eur J Biochem 1989 ; 180 : 487 – 502 . Google Scholar Crossref Search ADS PubMed WorldCat 13 Chelberg MK , Tsilibary EC, Hauser AR, McCarthy JB. Type IV collagen-mediated melanoma cell adhesion and migration: involvement of multiple, distinct domains of the collagen molecule . Cancer Res 1989 ; 49 : 4796 – 4802 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 14 Kim YS , Gum J, Brockhausen I. Mucin glycoproteins in neoplasia . Glycoconj J 1996 ; 13 : 693 – 707 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Dahiya R , Kwak KS, Byrd JC, Ho S, Yoon WH, Kim YS. Mucin synthesis and secretion in various human epithelial cancer cell lines that express the MUC-1 mucin gene . Cancer Res 1993 ; 53 : 1437 – 1443 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 16 Ho SB , Niehans GA, Lyftogt C, Yan PS, Cherwitz DL, Gum ET et al. Heterogeneity of mucin gene expression in normal neoplastic tissues . Cancer Res 1993 ; 53 : 641 – 651 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 17 Ho SB , Shekels LL, Toribara NW, Kim YS, Lyftogt C, Cherwitz DL et al. Mucin gene expression in normal, preneoplastic and human gastric epithelium . Cancer Res 1995 ; 55 : 2681 – 2690 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 18 Reis CA , David L, Seixas M, Burcell J, Sobrinho-Simoes M. Expression of fully and under-glycosylated forms of MUC1 mucin in gastric carcinoma . Int J Cancer 1998 ; 79 : 402 – 410 . Google Scholar Crossref Search ADS PubMed WorldCat 19 Utsunomiya T , Yonezawa S, Sakamoto H, Kitamura H, Hokita S, Akio T et al. Expression of MUC1 and MUC2 mucins in gastric carcinomas: its relationship with prognosis of the patients . Clin Cancer Res 1998 ; 4 : 2605 – 2614 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 20 Nakamori S , Ota DM, Cleary KR, Shirotani K. MUC1 mucin expression as a marker of progression and metastasis of human colorectal carcinoma . Gastroenterology 1994 ; 106 : 353 – 361 . Google Scholar Crossref Search ADS PubMed WorldCat 21 Liotta LA . Tumor invasion and metastases: role of the basement membrane. Warner–Lambert Parke–Davis Award lecture . Am J Pathol 1984 ; 117 : 339 – 348 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 22 Ohno T , Aihara R, Kamiyama Y, Mochiki E, Asao T, Kuwano H. Prognostic significance of combined expression of MUC1 and adhesion molecules in advanced gastric cancer . Eur J Cancer 2006 ; 42 : 256 – 263 . Google Scholar Crossref Search ADS PubMed WorldCat 23 Japanese Research Society for Gastric Cancer . Japanese Classification of Gastric Carcinoma (2nd English edn). Kanehara : Tokyo , 1998 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 24 Nakamura K , Mori M, Enjoji M. Distribution of basement membrane antigens in clinical gastric adenocarcinomas: an immunohistochemical study . J Clin Pathol 1987 ; 40 : 1418 – 1423 . Google Scholar Crossref Search ADS PubMed WorldCat 25 Akyürek N , Akyol G, Dursun A, Yamaç D, Günel N. Expression of MUC1 and MUC2 mucins in gastric carcinomas: their relationship with clinicopathologic parameters and prognosis . Pathol Res Pract 2002 ; 198 : 665 – 674 . Google Scholar Crossref Search ADS PubMed WorldCat 26 Kocer B , Soran A, Kiyak G. Prognostic significance of mucin expression in gastric carcinoma . Dig Dis Sci 2004 ; 49 : 321 – 332 . Google Scholar Crossref Search ADS WorldCat 27 Satoh S , Hinoda Y, Hayashi T, Burdick MD, Imai K, Hollingsworth MA. Enhancement of metastatic properties of pancreatic cancer cells by MUC1 gene encoding an anti-adhesion molecule . Int J Cancer 2000 ; 88 : 507 – 518 . Google Scholar Crossref Search ADS PubMed WorldCat Copyright © 2009 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) Copyright © 2009 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd.
TI - Prognostic significance of the expression of MUC1 and collagen type IV in advanced gastric carcinoma
JO - British Journal of Surgery
DO - 10.1002/bjs.6635
DA - 2009-07-09
UR - https://www.deepdyve.com/lp/oxford-university-press/prognostic-significance-of-the-expression-of-muc1-and-collagen-type-iv-0oE02FiPqk
SP - 901
EP - 909
VL - 96
IS - 8
DP - DeepDyve
ER -