(4) REEVE AE, HOUSIAUX PJ, GARDNER RJM, ET
AL: LOSS of a Harvey ras allele in sporadic Wilms' tumour. Nature 309:174-176, 1984 (5) FEARON ER, VOGELSTEIN B, FEINBERG AP:
Somatic deletion and duplication of genes on chromosome 11 in Wilms' tumours. Nature 309:176-178,1984 (6) NAYLOR SL, JOHNSON BE, MINNA JD, ET AL:
Loss of heterozygosity of chromosome 3p markers in small-cell lung cancer. Nature 329:451^154,1987 (7) BRAUCH H, JOHNSON B, HOVIS J, ET AL:
Molecular analysis of the short arm of chromosome 3 in small-cell and non-small-cell carcinoma of the lung. N Engl J Med 317:1109-1113,1987 (8) KOK K, OSINGA J, CARITT B, ET AL: Deletion
of a DNA sequence at the chromosomal region 3p21 in all major types of lung cancer. Nature 330:578-581,1987 (9) YOKOTA J, WADA M, SHIMOSATO Y, ET AL:
(10) JOHNSON BE, SAKAGUCHI AY, GAZDAR AF, ET
AL: Restriction fragment length polymorphism studies show consistent loss of chromosome 3p alleles in small-cell lung cancer patients' tumors. J Clin Invest 82:502-507, 1988 (//)
ALI IU, LIDEREAU R, THEILLET C, ET AL:
Reduction to homozygosity of genes on chromosome 11 in human breast neoplasia. Science 238:185-188, 1987 (12) LUNDBERG C, SKOOG L, CAVENEE WK, ET AL:
Allele losses on chromosome 17 in human epithelial ovarian carcinoma. Oncogene 5:1599-1601,1990 (24) LEE W-H, BOOKSTEIN R, LEE EY-H: In Tumor
Suppressor Genes (Klein G, ed). New York: Decker, 1990, pp 169-200 (25) HARBOUR JW, LAI S-L, WHANG-PENG J, ET AL:
Abnormalities in structure and expression of the human retinoblastoma gene in SCLC. Science 241:353-357,1988 (26) HENSEL CH, HSIEH C-L, GAZDAR AF, ET AL:
Altered structure and expression of the human retinoblastoma susceptibility gene in small-cell lung cancer. Cancer Res 50:3067-3072, 1990 (27) LEE EY-H, TO H, SHEW J-Y, ET AL: Inactiva-
tion of the retinoblastoma susceptibility gene in human breast cancers. Science 241:218221,1988 (28) T'ANG A, VARLEY JM, CHAKRABORTY S, ET
AL: Structural rearrangement of the retinoblastoma gene in human breast carcinoma. Science 242:263-266,1988 (29) MANIATIS
T,
FRITSCH
EF,
SANSBROOK
J:
Molecular cloning. In A Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 1982 (50) SCHWINN DA, LOMASNEY JW, LORENZ W, ET
AL: Molecular cloning and expression of the cDNA for a novel al-adrenergic receptor subtype. J Biol Chem 265:8183-8189, 1990 (31) KIDD KK, BOWCOCK AM, SCHMIDTKE J, ET AL:
Report of the DNA committee and catalogs of cloned and mapped genes and DNA polymorphisms. Cytogenet Cell Genet 51:622-947,1989
(32) LEE W-H, BOOKSTEIN R, HONG F, ET AL: Loss of heterozygosity in human ductal breast Human retinoblastoma susceptibility gene: tumors indicates a recessive mutation on Cloning, identification, and sequence. Science chromosome 13. Proc Natl Acad Sci USA 235:1394-1399,1987 84:2372-2376,1987 (33) WEBER JL, MAY PE: Abundant class of human (13) VOGELSTEIN B, FEARON ER, HAMILTON SR, ET DNA polymorphisms which can be typed AL: Genetic alterations during colorectal-tumor using the polymerase chain reaction. Am J development. N Engl J Med 319:525-532, Hum Genet 44:388-396, 1989 1988 ET AL:
(34) BUDOWLE B, CHAKRABORTY R, GIUSTI AM, ET
Chromosome 5 allele loss in human colorectal carcinomas. Nature 328:616-619, 1987
AL: Analysis of the VNTR locus Dl S80 by the PCR followed by high-resolution PAGE. Am J Hum Genet 48:137-144, 1991
(14) SOLOMON
E,
VOSS
R,
HALL
V,
(15) OKAMOTO M, SASAKI M, SUGIO K, ET AL: LOSS
of constitutional heterozygosity in colon carcinoma with familial polyposis coli. Nature 331:273-277,1988 (16) LAW DJ, OLSCHWANG S, MONPEZAT J-P, ET AL:
Concerted nonsyntenic allelic loss in human colorectal carcinoma. Science 241:961-965, 1988 (17) KNUDSON AG JR: Mutation and cancer: Statistical study of retinoblastoma. Proc Natl Acad Sci USA 68:820, 1971 (18) YOUNG JL, PERCY CL, ASIRE AJ: Surveillance
Epidemiology, and End Results: Incidence and mortality data, 1973-1977. DHHS Publ No. (NIH)81-2330. Bethesda, Md: National Cancer Institute, 1987
(35) BOOKSTEIN R, RIO P, MADREPERLA SA, ET AL:
Promoter deletion and loss of retinoblastoma gene expression in human prostate carcinoma. Proc Natl Acad Sci USA 87:7762-7766, 1990 (36) Wu S, SEINO S, BELL GI: Human collagen,
type II, alpha 1 (COL2A1) gene: VNTR polymorphism detected by gene amplification. Nucleic Acids Res 18:3102,1990 (37) SASANO H, COMERFORD MS, SILVERBERG SG,
ET AL: An analysis of abnormalities of the retinoblastoma gene in human ovarian and endometrial carcinoma. Cancer 66:2150-2154, 1990
Gastric cancer can be divided into two histologic types: intestinal and diffuse. To determine whether Helicobacter pylori, a bacterium linked with gastritis, was associated with either cancer type, we reviewed histologic sections from stomachs of patients who had undergone gastrectomy for gastric cancer. Of 37 of the sections with evidence of intestinal-type cancer, 33 (89.2%) contained H pylori in noncancerous tissue compared with 7 (31.8%) of 22 of the sections with evidence of diffusetype cancer (odds ratio = 17.7; /^.OOl). This association remained strong when controlled for age, sex, site, and number of sections reviewed. The prevalence of H pylori in intestinal-type gastric cancer far exceeded the prevalence of H pylori in diffuse disease and that described in the normal US population. This finding suggests that H pylori may be a cofactor in development of intestinal-type gastric cancer. [J Natl Cancer Inst 83:640-643,1991]
Gastric cancer is a leading cause of cancer death worldwide. Although its incidence has steadily decreased over the last five decades in the United States and Western Europe, in countries in Latin America and Asia, gastric cancer remains among the most common malignancies in both men and women (1,2). "Epidemic" foci where gastric cancer rates are extraordinarily high exist around the world. There are many ways to classify gastric cancer histologically. From an epidemiologic standpoint, one helpful classification is the Lauren classification (5). This divides gastric cancer into two histologic types: intestinal and diffuse. The intestinal type of gastric cancer occurs more commonly in males than in females and
(19) LEE JH, KAVANAGH JJ, WHARTON JT, ET AL:
Allele loss at the c-Ha-rasl locus in human ovarian cancer. Cancer Res 49:1220-1222, 1989 (20) LEE JH, KAVANAGH JJ, WILDRICK DM, ET AL:
Frequent loss of heterozygosity on chromosomes 6q, 11, and 17 in human ovarian carcinomas. Cancer Res 50:2724-2728, 1990 (2/) EHLEN T, DUBEAU L: LOSS of heterozygosity
on chromosomal segments 3p, 6q, and 1 lp in human ovarian carcinomas. Oncogene 5:219223,1990 (22) RUSSELL SE, HICKEY GI, LOWRY WS, ET AL:
Allele loss from chromosome 17 in ovarian cancer. Oncogene 5:1581-1583, 1990
640
Helicobacterpylori Infection in Intestinal- and Diffuse-Type Gastric Adenocarcinomas Julie Parsonnet,* Dan Vandersteen, Jeffrey Goates, Richard K. Sibley, Jonathan Pritikin, Yuan Chang
Received November 29. 1990; revised February 28, 1991; accepted March 1, 1991. J. Parsonnet, J. Pritikin, D. Vandersteen (Department of Medicine), J. Goates, R. K. Sibley, Y. Chang (Department of Pathology), Stanford University School of Medicine, Stanford, Calif. We thank Drs Alice Whittemore, Richard Olshen, and Patrick Moore for reviewing this manuscript. *Correspondence to: Julie Parsonnet, MD. Stanford University School of Medicine, Health Research and Policy Bldg, Rm 109A, Stanford, CA 94305-5425.
Journal of the National Cancer Institute
Downloaded from http://jnci.oxfordjournals.org/ at University of Western Ontario on October 30, 2014
Loss of heterozygosity on chromosomes 3, 13, and 17 in small-cell carcinoma and on chromosome 3 in adenocarcinoma of the lung. Proc Natl Acad Sci USA 84:9252-9256, 1987
(23) ECCLES DM, CRANSTON G, STEEL CM, ET AL:
Patients and Methods Subjects included all patients who had undergone gastrectomy for gastric cancer between 1985 and 1989 at Stanford University Hospital, Stanford, and Santa Clara Valley Hospital, San Jose, Calif. The age and sex of patients treated with Vol. 83, No. 9, May 1, 1991
gastrectomy were recorded, and available slides and tissue blocks from gastrectomy specimens were obtained. Hematoxylin and eosin (H&E)-stained sections from identified cases were separated into two groups: sections of malignant gastric tissue and sections of nonmalignant gastric tissue. To avoid bias, these groups were labeled so that pathologists reviewing the material could not relate malignant tissue sections with nonmalignant tissue sections from the same patient. Malignant tissue was evaluated only for tumor type. It was not inspected for H pylori infection. Histologic classification of gastric carcinoma into diffuse and intestinal types conformed to criteria set forth by Lauren (3) and was based on architectural structure, cytologic features, connective-tissue stromal response, and associated inflammatory infiltrates. H&E-stained nonmalignant sections from each subject were evaluated for H pylori infection. If H pylori was not evident in tissue sections from an individual patient, the original blocks were recut, stained with Giemsa stain, and re-examined. Any nonmalignant tissue section that contained curved, slender bacteria between the gastric mucus layer and the gastric epithelium by H&E or Giemsa stain constituted an H pylori gastrectomy tissue section. Three pathologists examined the sections in a blinded procedure. They reviewed the discrepant findings until a consensus was achieved. Data were then entered in the PC SAS statistical program and analyzed using Wilcoxon rank sum test, Fisher's exact probability test, chisquare test, and stratified analysis with logit-derived common odds ratios (ORs) and confidence intervals (CIs).
Results Sixty patients who were treated with gastrectomy for gastric cancer were identified. The mean and median ages of the patients were 65 and 68 years, respectively (range, 32-84 years). A median of 3 nonmalignant tissue sections (range, 1-4) was available from cases for review. Thirty-seven cases of gastric cancer were classified as intestinal type (Fig 1) and 23 as diffuse type (Fig 2). Age was similar in the two groups, but intestinal-
Fig 1. Intestinal-type carcinomas are characterized by prominent, well-formed glands lined by neoplastic epithelial cells with some degree of polarity. Cytologically, cells are large and display cytoplasmic as well as nuclear pleomorphism. Mitoses are frequently seen, as are polymorphous inflammatory infiltrates (H&E, original magnification x200).
Fig 2. Diffuse carcinomas are characterized by singly infiltrating malignant cells with poor cohesion. Although poorly formed glandlike structures rarely may be discerned, they are comparatively smaller and malignant cells lining these "glands" lack polarity. Infiltration of malignant cells with indistinct cytoplasm and pyknotic nuclei often produces a connective tissue proliferation and a chronic, lymphocytic type of inflammation (H&E, original magnification x 200).
type cancers were more likely to be from males than were diffuse-type cancers (Table 1). The number of nonmalignant sections available for review were similar for both groups. H&E-stained sections REPORTS 641
Downloaded from http://jnci.oxfordjournals.org/ at University of Western Ontario on October 30, 2014
normally arises in the gastric antrum. The diffuse type of cancer is only slightly more common in males than in females and may on occasion develop in relatively young persons; it generally arises in the fundus and is associated with blood group A. Regional and ethnic group variations in gastric cancer incidence appear to be related to the prevalent histologic types of gastric cancer seen in these groups (4,5). Intestinal-type cancer predominates in "epidemic" zones (6). When gastric cancer incidence declines, as it has in Europe and the United States, it has been predominantly the intestinal type that has diminished in frequency. The incidence of diffuse type has remained unchanged or has decreased to a much lesser degree (4,7-9). Because the incidence of intestinal-type cancer can change dramatically over a relatively short period, it has been hypothesized that its pathogenesis may be linked to environmental and perhaps socioeconomic factors. Additional support for the environmental hypothesis comes from studies of migrants, which demonstrate retention of parent-country risk in firstgeneration migrants but not in later generations (10). In contrast, diffuse-type disease has been assumed to be largely genetically linked. In the evolution of intestinal-type gastric cancer, the mucosa progresses through the steps of superficial gastritis, chronic atrophic gastritis, and intestinal metaplasia before malignancy appears (11,12). Because both superficial gastritis and chronic atrophic gastritis are strongly linked with Helicobacter pylori infection (13,14), we hypothesized that H pylori is an environmental cofactor involved in the transformation of gastric mucosa from normal epithelium to malignancy. We explored this hypothesis by retrospectively comparing the association of gastric H pylori infection among intestinal carcinomas with that among diffuse carcinomas.
Table 1. Characteristics of patients with diffuseand intestinal-type carcinomas and association of the tumor types with H pylori infection
Characteristic
Intestinal cancer*
Median age, y 70 % male 73.0 Median No. of 3 sections of nonmalignant tissue % with H pylori in 89.2 nonmalignant tissue sections
Diffuse cancerf
P
65 34.8 3
.29 .004 .45
31.8
AL: LOSS of a Harvey ras allele in sporadic Wilms' tumour. Nature 309:174-176, 1984 (5) FEARON ER, VOGELSTEIN B, FEINBERG AP:
Somatic deletion and duplication of genes on chromosome 11 in Wilms' tumours. Nature 309:176-178,1984 (6) NAYLOR SL, JOHNSON BE, MINNA JD, ET AL:
Loss of heterozygosity of chromosome 3p markers in small-cell lung cancer. Nature 329:451^154,1987 (7) BRAUCH H, JOHNSON B, HOVIS J, ET AL:
Molecular analysis of the short arm of chromosome 3 in small-cell and non-small-cell carcinoma of the lung. N Engl J Med 317:1109-1113,1987 (8) KOK K, OSINGA J, CARITT B, ET AL: Deletion
of a DNA sequence at the chromosomal region 3p21 in all major types of lung cancer. Nature 330:578-581,1987 (9) YOKOTA J, WADA M, SHIMOSATO Y, ET AL:
(10) JOHNSON BE, SAKAGUCHI AY, GAZDAR AF, ET
AL: Restriction fragment length polymorphism studies show consistent loss of chromosome 3p alleles in small-cell lung cancer patients' tumors. J Clin Invest 82:502-507, 1988 (//)
ALI IU, LIDEREAU R, THEILLET C, ET AL:
Reduction to homozygosity of genes on chromosome 11 in human breast neoplasia. Science 238:185-188, 1987 (12) LUNDBERG C, SKOOG L, CAVENEE WK, ET AL:
Allele losses on chromosome 17 in human epithelial ovarian carcinoma. Oncogene 5:1599-1601,1990 (24) LEE W-H, BOOKSTEIN R, LEE EY-H: In Tumor
Suppressor Genes (Klein G, ed). New York: Decker, 1990, pp 169-200 (25) HARBOUR JW, LAI S-L, WHANG-PENG J, ET AL:
Abnormalities in structure and expression of the human retinoblastoma gene in SCLC. Science 241:353-357,1988 (26) HENSEL CH, HSIEH C-L, GAZDAR AF, ET AL:
Altered structure and expression of the human retinoblastoma susceptibility gene in small-cell lung cancer. Cancer Res 50:3067-3072, 1990 (27) LEE EY-H, TO H, SHEW J-Y, ET AL: Inactiva-
tion of the retinoblastoma susceptibility gene in human breast cancers. Science 241:218221,1988 (28) T'ANG A, VARLEY JM, CHAKRABORTY S, ET
AL: Structural rearrangement of the retinoblastoma gene in human breast carcinoma. Science 242:263-266,1988 (29) MANIATIS
T,
FRITSCH
EF,
SANSBROOK
J:
Molecular cloning. In A Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, 1982 (50) SCHWINN DA, LOMASNEY JW, LORENZ W, ET
AL: Molecular cloning and expression of the cDNA for a novel al-adrenergic receptor subtype. J Biol Chem 265:8183-8189, 1990 (31) KIDD KK, BOWCOCK AM, SCHMIDTKE J, ET AL:
Report of the DNA committee and catalogs of cloned and mapped genes and DNA polymorphisms. Cytogenet Cell Genet 51:622-947,1989
(32) LEE W-H, BOOKSTEIN R, HONG F, ET AL: Loss of heterozygosity in human ductal breast Human retinoblastoma susceptibility gene: tumors indicates a recessive mutation on Cloning, identification, and sequence. Science chromosome 13. Proc Natl Acad Sci USA 235:1394-1399,1987 84:2372-2376,1987 (33) WEBER JL, MAY PE: Abundant class of human (13) VOGELSTEIN B, FEARON ER, HAMILTON SR, ET DNA polymorphisms which can be typed AL: Genetic alterations during colorectal-tumor using the polymerase chain reaction. Am J development. N Engl J Med 319:525-532, Hum Genet 44:388-396, 1989 1988 ET AL:
(34) BUDOWLE B, CHAKRABORTY R, GIUSTI AM, ET
Chromosome 5 allele loss in human colorectal carcinomas. Nature 328:616-619, 1987
AL: Analysis of the VNTR locus Dl S80 by the PCR followed by high-resolution PAGE. Am J Hum Genet 48:137-144, 1991
(14) SOLOMON
E,
VOSS
R,
HALL
V,
(15) OKAMOTO M, SASAKI M, SUGIO K, ET AL: LOSS
of constitutional heterozygosity in colon carcinoma with familial polyposis coli. Nature 331:273-277,1988 (16) LAW DJ, OLSCHWANG S, MONPEZAT J-P, ET AL:
Concerted nonsyntenic allelic loss in human colorectal carcinoma. Science 241:961-965, 1988 (17) KNUDSON AG JR: Mutation and cancer: Statistical study of retinoblastoma. Proc Natl Acad Sci USA 68:820, 1971 (18) YOUNG JL, PERCY CL, ASIRE AJ: Surveillance
Epidemiology, and End Results: Incidence and mortality data, 1973-1977. DHHS Publ No. (NIH)81-2330. Bethesda, Md: National Cancer Institute, 1987
(35) BOOKSTEIN R, RIO P, MADREPERLA SA, ET AL:
Promoter deletion and loss of retinoblastoma gene expression in human prostate carcinoma. Proc Natl Acad Sci USA 87:7762-7766, 1990 (36) Wu S, SEINO S, BELL GI: Human collagen,
type II, alpha 1 (COL2A1) gene: VNTR polymorphism detected by gene amplification. Nucleic Acids Res 18:3102,1990 (37) SASANO H, COMERFORD MS, SILVERBERG SG,
ET AL: An analysis of abnormalities of the retinoblastoma gene in human ovarian and endometrial carcinoma. Cancer 66:2150-2154, 1990
Gastric cancer can be divided into two histologic types: intestinal and diffuse. To determine whether Helicobacter pylori, a bacterium linked with gastritis, was associated with either cancer type, we reviewed histologic sections from stomachs of patients who had undergone gastrectomy for gastric cancer. Of 37 of the sections with evidence of intestinal-type cancer, 33 (89.2%) contained H pylori in noncancerous tissue compared with 7 (31.8%) of 22 of the sections with evidence of diffusetype cancer (odds ratio = 17.7; /^.OOl). This association remained strong when controlled for age, sex, site, and number of sections reviewed. The prevalence of H pylori in intestinal-type gastric cancer far exceeded the prevalence of H pylori in diffuse disease and that described in the normal US population. This finding suggests that H pylori may be a cofactor in development of intestinal-type gastric cancer. [J Natl Cancer Inst 83:640-643,1991]
Gastric cancer is a leading cause of cancer death worldwide. Although its incidence has steadily decreased over the last five decades in the United States and Western Europe, in countries in Latin America and Asia, gastric cancer remains among the most common malignancies in both men and women (1,2). "Epidemic" foci where gastric cancer rates are extraordinarily high exist around the world. There are many ways to classify gastric cancer histologically. From an epidemiologic standpoint, one helpful classification is the Lauren classification (5). This divides gastric cancer into two histologic types: intestinal and diffuse. The intestinal type of gastric cancer occurs more commonly in males than in females and
(19) LEE JH, KAVANAGH JJ, WHARTON JT, ET AL:
Allele loss at the c-Ha-rasl locus in human ovarian cancer. Cancer Res 49:1220-1222, 1989 (20) LEE JH, KAVANAGH JJ, WILDRICK DM, ET AL:
Frequent loss of heterozygosity on chromosomes 6q, 11, and 17 in human ovarian carcinomas. Cancer Res 50:2724-2728, 1990 (2/) EHLEN T, DUBEAU L: LOSS of heterozygosity
on chromosomal segments 3p, 6q, and 1 lp in human ovarian carcinomas. Oncogene 5:219223,1990 (22) RUSSELL SE, HICKEY GI, LOWRY WS, ET AL:
Allele loss from chromosome 17 in ovarian cancer. Oncogene 5:1581-1583, 1990
640
Helicobacterpylori Infection in Intestinal- and Diffuse-Type Gastric Adenocarcinomas Julie Parsonnet,* Dan Vandersteen, Jeffrey Goates, Richard K. Sibley, Jonathan Pritikin, Yuan Chang
Received November 29. 1990; revised February 28, 1991; accepted March 1, 1991. J. Parsonnet, J. Pritikin, D. Vandersteen (Department of Medicine), J. Goates, R. K. Sibley, Y. Chang (Department of Pathology), Stanford University School of Medicine, Stanford, Calif. We thank Drs Alice Whittemore, Richard Olshen, and Patrick Moore for reviewing this manuscript. *Correspondence to: Julie Parsonnet, MD. Stanford University School of Medicine, Health Research and Policy Bldg, Rm 109A, Stanford, CA 94305-5425.
Journal of the National Cancer Institute
Downloaded from http://jnci.oxfordjournals.org/ at University of Western Ontario on October 30, 2014
Loss of heterozygosity on chromosomes 3, 13, and 17 in small-cell carcinoma and on chromosome 3 in adenocarcinoma of the lung. Proc Natl Acad Sci USA 84:9252-9256, 1987
(23) ECCLES DM, CRANSTON G, STEEL CM, ET AL:
Patients and Methods Subjects included all patients who had undergone gastrectomy for gastric cancer between 1985 and 1989 at Stanford University Hospital, Stanford, and Santa Clara Valley Hospital, San Jose, Calif. The age and sex of patients treated with Vol. 83, No. 9, May 1, 1991
gastrectomy were recorded, and available slides and tissue blocks from gastrectomy specimens were obtained. Hematoxylin and eosin (H&E)-stained sections from identified cases were separated into two groups: sections of malignant gastric tissue and sections of nonmalignant gastric tissue. To avoid bias, these groups were labeled so that pathologists reviewing the material could not relate malignant tissue sections with nonmalignant tissue sections from the same patient. Malignant tissue was evaluated only for tumor type. It was not inspected for H pylori infection. Histologic classification of gastric carcinoma into diffuse and intestinal types conformed to criteria set forth by Lauren (3) and was based on architectural structure, cytologic features, connective-tissue stromal response, and associated inflammatory infiltrates. H&E-stained nonmalignant sections from each subject were evaluated for H pylori infection. If H pylori was not evident in tissue sections from an individual patient, the original blocks were recut, stained with Giemsa stain, and re-examined. Any nonmalignant tissue section that contained curved, slender bacteria between the gastric mucus layer and the gastric epithelium by H&E or Giemsa stain constituted an H pylori gastrectomy tissue section. Three pathologists examined the sections in a blinded procedure. They reviewed the discrepant findings until a consensus was achieved. Data were then entered in the PC SAS statistical program and analyzed using Wilcoxon rank sum test, Fisher's exact probability test, chisquare test, and stratified analysis with logit-derived common odds ratios (ORs) and confidence intervals (CIs).
Results Sixty patients who were treated with gastrectomy for gastric cancer were identified. The mean and median ages of the patients were 65 and 68 years, respectively (range, 32-84 years). A median of 3 nonmalignant tissue sections (range, 1-4) was available from cases for review. Thirty-seven cases of gastric cancer were classified as intestinal type (Fig 1) and 23 as diffuse type (Fig 2). Age was similar in the two groups, but intestinal-
Fig 1. Intestinal-type carcinomas are characterized by prominent, well-formed glands lined by neoplastic epithelial cells with some degree of polarity. Cytologically, cells are large and display cytoplasmic as well as nuclear pleomorphism. Mitoses are frequently seen, as are polymorphous inflammatory infiltrates (H&E, original magnification x200).
Fig 2. Diffuse carcinomas are characterized by singly infiltrating malignant cells with poor cohesion. Although poorly formed glandlike structures rarely may be discerned, they are comparatively smaller and malignant cells lining these "glands" lack polarity. Infiltration of malignant cells with indistinct cytoplasm and pyknotic nuclei often produces a connective tissue proliferation and a chronic, lymphocytic type of inflammation (H&E, original magnification x 200).
type cancers were more likely to be from males than were diffuse-type cancers (Table 1). The number of nonmalignant sections available for review were similar for both groups. H&E-stained sections REPORTS 641
Downloaded from http://jnci.oxfordjournals.org/ at University of Western Ontario on October 30, 2014
normally arises in the gastric antrum. The diffuse type of cancer is only slightly more common in males than in females and may on occasion develop in relatively young persons; it generally arises in the fundus and is associated with blood group A. Regional and ethnic group variations in gastric cancer incidence appear to be related to the prevalent histologic types of gastric cancer seen in these groups (4,5). Intestinal-type cancer predominates in "epidemic" zones (6). When gastric cancer incidence declines, as it has in Europe and the United States, it has been predominantly the intestinal type that has diminished in frequency. The incidence of diffuse type has remained unchanged or has decreased to a much lesser degree (4,7-9). Because the incidence of intestinal-type cancer can change dramatically over a relatively short period, it has been hypothesized that its pathogenesis may be linked to environmental and perhaps socioeconomic factors. Additional support for the environmental hypothesis comes from studies of migrants, which demonstrate retention of parent-country risk in firstgeneration migrants but not in later generations (10). In contrast, diffuse-type disease has been assumed to be largely genetically linked. In the evolution of intestinal-type gastric cancer, the mucosa progresses through the steps of superficial gastritis, chronic atrophic gastritis, and intestinal metaplasia before malignancy appears (11,12). Because both superficial gastritis and chronic atrophic gastritis are strongly linked with Helicobacter pylori infection (13,14), we hypothesized that H pylori is an environmental cofactor involved in the transformation of gastric mucosa from normal epithelium to malignancy. We explored this hypothesis by retrospectively comparing the association of gastric H pylori infection among intestinal carcinomas with that among diffuse carcinomas.
Table 1. Characteristics of patients with diffuseand intestinal-type carcinomas and association of the tumor types with H pylori infection
Characteristic
Intestinal cancer*
Median age, y 70 % male 73.0 Median No. of 3 sections of nonmalignant tissue % with H pylori in 89.2 nonmalignant tissue sections
Diffuse cancerf
P
65 34.8 3
.29 .004 .45
31.8
