The contribution of cell phenotype to the behavior of gastric cancer.

BACKGROUND: Several histochemical studies suggest a role of tumor cell phenotype and related differentiation markers in the prognostic assessment of gastric cancer. Unfortunately, most studies have dealt with single or a few markers and have paid limited attention to their interplay with tumor histological types, which are potentially informative of prognosis. METHODS: In this study, 292 invasive (T1b to T4) gastric cancers with prolonged follow-up and carefully analyzed histotype, inclusive of histotype-based grade, were investigated histochemically with a panel of 14 phenotypic markers known to be expressed in normal gut tissues and gastric cancer. RESULTS: Three of seven intestinal type markers investigated showed a trend for improved prognosis, one of which, CDX2, was stage independent. Three among gastric and pancreatobiliary duct markers (MUC1, MUC6, and pepsinogen II), predicted more severe prognosis stage independently, as did a combination of eight potentially informative (p < 0.1 at univariable Cox analysis) markers. Cancers with predominantly intestinal phenotype had significantly better prognosis than those with predominantly gastric, mixed, or poorly defined phenotypes; among the latter, those with high lymphocyte response, with favorable outcome, were separated from anaplastic cancers, with ominous prognosis. At multivariable analysis, CDX2 and the eight marker combination proved to be stage- and grade-independent predictors. CONCLUSIONS: When individually considered, and with the exception of CDX2, the biomarkers investigated gave an appreciable, although moderate, contribution to the prognostic evaluation of gastric cancer. Combined analysis of all potentially informative markers gave more important information, highly additive to both stage and histotype-based grade.

Improving the AJCC/TNM classification for use in early gastric cancer.

PURPOSE: The current TNM classification is still unsatisfactory for collecting all the prognostic information from the clinical presentation of early gastric cancer: "T" is limited to two levels, the classes of "N" are still wide and "M" is generally absent. PATIENTS AND METHODS: This study involved 99 patients who underwent radical gastric resection for early gastric cancer. Clinical and histological parameters were prognostically analyzed for both observed and relative survival. Univariate and multivariate analyses were applied to the proportional hazards model. RESULTS: Number of metastatic lymph nodes and measure of the largest diameter of the tumor were the only independent prognosticators of observed and relative survival. Their similar relative hazards allowed an additive use of them in the N class. Two cut-off values of this composite clinical parameter are proposed for a good discrimination of the relative survival. DISCUSSION: The number of metastatic lymph nodes is the cornerstone of the current TNM system and was confirmed as adequate. The possibility of adding tumor size to the number of the involved lymph nodes improves and amplifies the prognostic ability, which is presently limited by the rarity of lymph node involvement and the small number of the lymph nodes usually involved.

A combined histologic and molecular approach identifies three groups of gastric cancer with different prognosis.

The limited prognostic value of currently used histologic classifications of gastric cancer and their failure to account for the complexity of the disease as revealed by more recent investigations prompted a combined reinvestigation of histologic, molecular, and clinicopathologic patterns in 294 extensively sampled, invasive gastric cancers representing all main histotypes and stages of the disease and followed for a median of 150 months. Among histologic parameters tested, only cellular atypia, angio-lympho- or neuroinvasion, Ki67 proliferation index, expansile/infiltrative type growth, and T8 cell-rich high lymphoid intra-/peritumor response (HLR) proved to be stage-independent predictors of patient survival. Among molecular tests, p53 gene exon 7 (loop 3) and 8 (loop-sheet-helix motif and S-10 band), but not p53 protein overexpression, TP53 LOH or 18qLOH, were found to worsen prognosis. Microsatellite DNA instability was a favorable prognostic factor when coupled with HLR. Patient survival analysis of the main histotypes and their subtypes confirmed the favorable prognosis of HLR, well-differentiated tubular, muconodular, and low grade diffuse desmoplastic cancers, and highlighted the worse prognosis of anaplastic and infiltrative-lymphoinvasive mucinous cancers compared to ordinary cohesive and diffuse cancers. Distinct roles of individual morphologic and molecular factors in tumor progression of the different histotypes have been recognized. The combination of survival-predictive histotypes and individual histologic or molecular parameters allowed us to develop a classification of all gastric cancers into three grades of increasing malignancy which proved to be of high prognostic value.

Lower- and higher-grade subtypes of diffuse gastric cancer.

Diffuse gastric cancer, characterized by poorly cohesive, diffusely infiltrating cells with no or little gland formation, is known to show several morphologic variants, but their prognostic value, if any, is poorly documented. In this article, 119 cases of invasive (T1b to T4) diffuse gastric cancer, which had undergone potentially curative surgery and were followed postoperatively for a median time of more than 10 years, were investigated for histologic or histochemical patterns possibly predictive of survival. Among 5 histologic groups identified, a low-grade subtype (17 cases) with prominent desmoplasia closely surrounding individual tumor cells (tumor embedding desmoplasia) and no or scarce angio-lympho-neuroinvasion showed stage-independent improved survival compared with 36 non-low-grade desmoplastic, 24 signet ring, and 28 diffuse cancers not otherwise specified. Fourteen cases with anaplastic cells showed clinicopathologic patterns and outcome of highly malignant neoplasms. None of the tumor cell differentiation markers (including 6 mucins and 3 proteases) nor proliferative index or p53 protein expression had independent predictive power, although MUC1 was significantly less expressed in low-grade desmoplastic cases. Cox survival analysis showed the significantly better prognosis of 17 low-grade desmoplastic and worse prognosis of 14 anaplastic cancers compared with the remaining 88 cases. In conclusion, a low-grade desmoplastic and a high-grade anaplastic subtype should be separated histologically from the bulk of diffuse gastric cancers owing to their distinctive histologic, clinicopathologic, and prognostic aspects.

Intracellular, intercellular, and stromal invasion of gastric mucosa, preneoplastic lesions, and cancer by Helicobacter pylori.

BACKGROUND AND AIMS: It is not clear how Helicobacter pylori, an apparently extracellular pathogen colonizing the luminal side of the gastric epithelium, invariably causes an immune-inflammatory response on the stromal side of the mucosa. Penetration of H pylori into epithelial cell lines and its interaction with immune-inflammatory cells have been documented in vitro. Several investigations also showed in vivo bacterial penetration into the epithelium up to the lamina propria; however, the identification as H pylori of the bacteria-like bodies observed in unchanged, metaplastic, or neoplastic mucosa remained sometimes questionable. METHODS: To search for bacteria-like organisms, we used transmission electron microscopy on endoscopic biopsy specimens from 20 dyspeptic subjects and surgical specimens of neoplastic and nonneoplastic mucosa from 20 cancerous stomachs. To ascertain the H pylori nature of the organisms found, we used 6 different antibodies directed against bacterial lysates, purified vacuolating cytotoxin A, or purified cytotoxin-associated antigen A in immunogold tests. The results were compared with those of H pylori strains cultivated in vitro. RESULTS: In nonmetaplastic gastric epithelium, cytochemically proven H pylori were detected, in the majority of cases, inside cytoplasm of epithelial cells, in intraepithelial intercellular spaces, and in underlying lamina propria, often in direct contact with immune-inflammatory cells and sometimes inside small blood vessels. Cytochemically proven H pylori were also observed inside 6 of 8 intestinal metaplasias and 9 of 20 cancers. CONCLUSIONS: H pylori penetrates normal, metaplastic, and neoplastic gastric epithelium in vivo, intracellularly, or interstitially to cause a strong immune-inflammatory response and promote gastric carcinogenesis.

Type or extension of intestinal metaplasia and immature/atypical "indefinite-for-dysplasia" lesions as predictors of gastric neoplasia.

At present, no information exists on the neoplastic potential of the immature hyperproliferative and atypical lesions of the gastric mucosa, which have been recently labeled "indefinite for dysplasia." In addition, uncertainties still exist concerning the risk contribution of intestinal metaplasia (IM) type and extension, as well as Helicobacter pylori infection. In this study, 471 dyspeptic patients showing IM 10% or higher (median, 40; 25th-75th percentile, 20-60) in antral, angulus, or corpus endoscopic biopsies were submitted to repeated examinations (median, 3; 2-5) over 52 (26-85) months of follow-up, during which 44 neoplastic cases were recorded. IM extension, incomplete, sulfomucin-positive, or CAR5 antigen-positive IM; H pylori infection; and indefinite-for-dysplasia lesions (IDLs), as found at first examination, all showed significant neoplastic potential. However, only IDL, ongoing H pylori infection, and patient's age retained independent predictive power in a multivariate model. On the other hand, IM extension 20% or higher proved to be more sensitive as first screening parameter for identification of subjects with increased neoplastic risk. We suggest that patients with IM, when infected, should undergo H pylori eradication to reduce their cancer risk; only those bearing IDL or very extensive IM (which strongly correlates with IDL) should be followed up with endoscopies and biopsies.

Identification of a lower grade muconodular subtype of gastric mucinous cancer.

Until now, survival analysis of gastric mucinous cancers showed either no difference or an even worse prognosis than stage-adjusted non-mucinous tumours. In the pancreas and breast, mucinous cancers showing well-demarcated mucin deposits (muconodular pattern), expansile growth and predominance of MUC2 mucin are known to have a more favourable prognosis. In this study, an attempt was made to separate, among 41 gastric mucinous cancers, a subgroup of tumours with muconodular expansile pattern, possibly predictive of a more favourable outcome. A group of 15 tumours was identified, which were characterised by overwhelming (80-100%) mucinous component, predominance of mucus over tumour cells inside muconodules, moderately aggressive growth of their epithelial component (reduced proliferative rate, moderate anaplasia, lack of angioinvasion and limited lymphoinvasion) and dominant expression of intestinal goblet cell markers, with special reference to MUC2 mucin. Univariate and multivariate survival analysis showed a significantly improved outcome of these lower grade muconodular tumours compared with the remaining mucinous cancers as well as with non-mucinous cancers of cohesive, diffuse (signet-ring cell included) and undifferentiated high-grade types.