Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Helicobacter pylori and cancer among adults in Uganda

Helicobacter pylori and cancer among adults in Uganda Data from Africa on infection with Helicobacter pylori (H. pylori) are sparse. Therefore, as part of an epidemiological study of cancer in Uganda, we investigated the prevalence and determinants of antibodies against H. pylori among 854 people with different cancer types and benign tumours. Patients were recruited from hospitals in Kampala, Uganda, interviewed about various demographic and lifestyle factors and tested for antibodies against H. pylori. In all patients combined, excluding those with stomach cancer (which has been associated with H. pylori infection), the prevalence of antibodies was 87% (723/833) overall, but declined with increasing age (p = 0.02) and was lower among people who were HIV seropositive compared to seronegative (p < 0.001). Otherwise, there were few consistent epidemiological associations. Among those with stomach cancer, 18/21 (86%) had anti-H. pylori antibodies (odds ratio 0.8, 95% confidence intervals 0.2–2.9, p = 0.7; estimated using all other patients as controls, with adjustment for age, sex and HIV serostatus). No other cancer site or type was significantly associated with anti-H. pylori antibodies. The prevalence of H. pylori reported here is broadly in accord with results from other developing countries, although the determinants of infection and its' role in the aetiology of gastric cancer in Uganda remain unclear. ological study of cancer in Kampala, Uganda [1-7]. Data Background The work described in this report was part of an epidemi- from Africa on infection with Helicobacter pylori (H. pylori) Page 1 of 5 (page number not for citation purposes) Infectious Agents and Cancer 2006, 1:5 http://www.infectagentscancer.com/content/1/1/5 are sparse. Therefore, we examine here the role of antibod- the out-patient department of Mulago hospital and for ies against H. pylori in relation to the risk of cancer and 804 patients with cancer or benign tumours, for whom a investigate the prevalence and determinants of infection stored blood sample was available for testing. The latter among 854 people with different cancer types and benign group comprised people with cancers of the oral cavity tumours. (26), oesophagus (38), stomach (21), liver (52), skin (22), breast (69), cervix (190), ovary (22), prostate (10), Materials and methods penis (14), eye (63), and non-Hodgkin's lymphoma (46), Full details of the methods are provided elsewhere [1,2]. Hodgkin's disease (24), Kaposi's sarcoma (46), other can- Briefly between 1994 and 1998, we recruited adults 15 cer sites or types (126) and benign tumours (35). years or older with a new diagnosis of cancer from the wards and out-patient clinics of the main hospitals in Data were computerised by trained clerks using EPI-INFO Kampala, Uganda. After informed consent and counsel- software (CDC, Atlanta) and statistical analyses were con- ling, patients were interviewed and tested for infection ducted using STATA (STATA Corporation, Texas). Only a with HIV-1 using the Cambridge Bioscience Recombigen small proportion of those tested were seronegative for ELISA (Cambridge, MA) on sera or the GACELISA method antibodies against H. pylori (optical density <0.9) or had (Murex, Dartford, UK) on saliva. Cancer diagnoses were an indeterminate result (optical density 0.9–1.3). In all established by histology or other laboratory investigation, analyses, those with indeterminate results were consid- where possible. Diagnoses made on clinical grounds ered to be seronegative. In order to examine potential con- alone were reviewed by the investigators. The study was founding factors, the risk of being seropositive for approved by the Committee on Human Research (VA antibodies against H. pylori was examined in relation to Medical Centre and University of California San Fran- various social and demographic factors among all patients cisco) and by the Uganda National Council for Science combined (but excluding stomach cancer, which has been and Technology. associated with H. pylori infection). Odds ratios (OR) were estimated using unconditional logistic regression model- Following HIV testing, remaining sera were stored at ling with adjustment for sex, age group (<30, 30–45, 46+) minus 80 Celsius and were later shipped on dry ice to the and HIV serostatus. When calculating odds ratios in rela- Centres for Disease Control and Prevention, Atlanta, USA, tion to anti-H. pylori antibodies, for each cancer site or for H. pylori testing. Assays were performed by a single type, the comparison group included all other patients investigator who was blind to the diagnosis of the patient with the exception of stomach cancer. Tests for association from whom the blood was obtained. Briefly, H. pylori used the χ test for linear trend on one degree of freedom organisms were grown overnight in brucella broth and all p values are 2-sided. Risk factors for high titres of (GIBCO Laboratories, Madison, WS) with 10% fetal antibodies against H. pylori were examined amongst all bovine serum (Sigma, St. Louis, MO), 5 μg/ml trimetho- patients combined (excluding stomach cancer), but no prim and 10 μg/ml vancomycin (Sigma). H. pylori antigen clear associations were identified and the data are not extraction and protein isolation were done by gentle shown. freeze-thaw sonication (Heat System, Farmingdale, NY) [8,9]. A standard protein assay (Pierce, Rockford, IL) was Results used to determine the accurate and reproducible quantity H. pylori antibody status was available for 854 people; 87% (741) were seropositive, 4% (38) were seronegative of solid-phase antigen for our microtitre research ELISA [10]. Cross-reactivity and specificity of H. pylori whole-cell and 9% (75) had an indeterminate result. Table 1 shows antigens has been described previously [9,10]. Optical the association between H. pylori serostatus and selected density (OD) values at a wavelength of 492 nm were social and demographic factors among all patients exclud- determined in triplicate for each biopsy-confirmed con- ing those with stomach cancer. The prevalence of antibod- trol patient sera, using a standard 96-well microtiter plate ies did not vary with sex, but declined with increasing age ELISA spectrophotometer (Fisher Scientific, Pittsburgh, (χ = 5.1, p = 0.02) and was lower among people who PA). The mean OD values were then calculated. The ELISA were HIV seropositive compared to seronegative (χ = cut-off values were derived using known H. pylori-positive 16.2, p < 0.001). No other factor examined was associated and negative control sera as previously described [9,11]. with antibodies against H. pylori with the exception of reli- In previous validation studies the assay has demonstrated gion: the prevalence of antibodies was higher among Mus- a high and reproducible sensitivity and specificity in Afri- lims than among Christians (χ = 5.6, p = 0.02). can patients as compared to upper endoscopy and biopsy; sensitivity >88%, specificity >90% [9,11]. Table 2 shows the association between anti-H. pylori anti- bodies and specific cancer sites or types, together with the Serological results were available for 50 people with non- proportion of cancers with a laboratory verification of malignant manifestations of HIV disease, recruited from diagnosis. Overall, 62% of cancers were diagnosed on the Page 2 of 5 (page number not for citation purposes) Infectious Agents and Cancer 2006, 1:5 http://www.infectagentscancer.com/content/1/1/5 Table 1: Odds ratios (OR) for H. pylori seropositivity according to various social and demographic factors among all patients (excluding those with stomach cancer) 1 1 Variable Number positive/ OR (95% CI) Variable Number positive/ OR (95% CI) total total Sex Time to market Male 247/284 1.0 <30 minutes 323/375 1.0 Female 476/549 0.9 (0.6–1.3) 30+ minutes 278/320 1.0 (0.7–1.6) 2 2 χ = 0.5, p = 0.5 χ = 0.0, p = 0.9 1 1 Age group Size of community <36 years 269/309 1.0 >100 houses 220/262 1.0 36–50 years 250/282 1.0 (0.6–1.6) 10–99 houses 421/482 1.3 (0.9–2.1) 51+ years 204/242 0.5 (0.3–0.9) <10 houses 51/57 1.6 (0.6–4.1) 2 2 χ = 5.1, p = 0.02 χ = 1.0, p = 0.3 1 1 HIV serostatus Ever travel from home Negative 514/574 1.0 Positive 201/250 0.4 (0.3–0.6) Yes 91/109 1.0 χ = 16.2, p<0.001 No 599/688 1.4 (0.7–2.5) = 1.0, p = 0.3 Region of residence Household size Kampala 161/188 1.0 <6 people 357/410 1.0 Rest of Uganda 560/643 1.0 (0.6–1.6) 6+ people 354/411 0.8 (0.5–1.2) 2 2 χ = 0.0, p = 0.9 χ = 1.6, p = 0.2 1 1 Tribe Number of siblings Baganda 349/410 1.0 <6 siblings 379/438 1.0 Other 374/423 1.2 (0.8–1.8) 6+ siblings 333/384 0.9 (0.6–1.4) 2 2 χ = 0.8, p = 0.4 χ = 0.1, p = 0.8 1 1 Religion Number of children Muslim 95/100 1.0 <7 children 341/397 1.0 Christian 619/723 0.3 (0.1–0.8) 7+ children 290/329 1.1 (0.7–1.8) 2 2 χ = 5.6, p = 0.02 χ = 0.1, p = 0.8 1 1 Occupation Tobacco consumption Cultivator 334/387 1.0 Other 382/439 1.1 (0.7–1.9) Never smoker 562/647 1.0 χ = 0.1, p = 0.8 Past smoker 81/92 1.2 (0.6–2.5) Current smoker 68/82 0.7 (0.4–1.4) χ = 0.5, p = 0.5 Education level Alcohol consumption No school 153/175 1.0 Primary 361/415 0.9 (0.5–1.6) Never 391/443 1.0 Secondary/tertiary 191/225 0.7 (0.4–1.4) About once/week 134/157 0.8 (0.5–1.4) χ = 1.0, p = 0.3 2–4 days/week 91/106 0.9 (0.5–1.6) Most days 93/113 0.7 (0.4–1.2) χ = 1.6, p = 0.2 Household Income Lifetime number of (Ug. Sh.) sexual partners 15,000+ 464/533 1.0 1–2 partners 182/204 1.0 <15,000 196/230 0.9 (0.6–1.5) 3–9 partners 305/352 0.9 (0.5–1.5) χ = 0.1, p = 0.8 10+ partners 196/232 0.7 (0.4–1.3) χ = 1.2, p = 0.3 1. Odds ratios adjusted for age group (<30, 30–45, 46+), sex and HIV serostatus basis of histology or other laboratory investigation, but 21) had the diagnosis confirmed by a laboratory investi- the figure varied by cancer site or type, being lowest for gation. prostate cancer (44%) and highest for Kaposi's sarcoma (91%). Of those people with stomach cancer, 90% (19/ Among 21 cases with stomach cancer, one was HIV serop- ositive, 13 were women, one was aged <30 years, three Page 3 of 5 (page number not for citation purposes) Infectious Agents and Cancer 2006, 1:5 http://www.infectagentscancer.com/content/1/1/5 Table 2: H. pylori serostatus for different cancer sites or types and non-malignant conditions, together with the percentage of each cancer with laboratory verification of diagnosis 1 2 Cancer site or type Percentage with laboratory Number anti-H. pylori anti- Odds ratio (95% CI) χ and p value verification of diagnosis body positive/total Stomach 90% 18/21 0.8 (0.2–2.9) χ = 0.1, p = 0.7 All controls - 723/833 1.0 - Oral 56% 21/26 0.6 (0.2–1.7) χ = 0.9, p = 0.4 Oesophagus 45% 37/38 5.1 (0.7–38) χ = 2.5, p = 0.1 Liver 56% 44/52 0.7 (0.3–1.6) χ = 0.8, p = 0.4 Skin 77% 18/22 0.6 (0.2–1.9) χ = 0.8, p = 0.4 = 0.6, p = 0.4 Breast 62% 62/69 1.4 (0.6–3.5) χ Cervix 50% 173/190 1.6 (0.9–2.8) χ = 2.1, p = 0.2 Ovary 73% 20/22 1.1 (0.3–5.0) χ = 0.0, p = 0.9 Prostate 44% 7/10 0.3 (0.1–1.4) χ = 2.4, p = 0.1 Penis 57% 12/14 0.9 (0.2–4.3) χ = 0.0, p = 0.9 Conjunctiva 66% 33/38 1.3 (0.5–3.5) χ = 0.3, p = 0.6 Other eye 52% 22/25 1.1 (0.3–3.9) χ = 0.0, p = 0.9 Non-Hodgkin's 76% 39/46 0.8 (0.3–1.8) χ = 0.4, p = 0.5 lymphoma Hodgkin's lymphoma 83% 20/24 0.6 (0.2–1.9) χ = 0.7, p = 0.4 Kaposi's sarcoma 91% 36/46 0.7 (0.3–1.6) χ = 0.9, p = 0.4 1. Odds ratios adjusted for age group (<30, 30–45, 46+), sex and HIV serostatus 2. The comparison group for the calculation of odds ratios includes all other cancers and non-malignant conditions, excluding stomach cancer Note: H. Pylori serostatus: negative – optical density (O.D.) 0.0–0.8; indeterminate – O.D. 0.9–1.3; positive – O.D. 1.4–4.0. For the purposes of this analysis, those with an indeterminate result were considered to be seronegative. were aged between 30–45 years and 17 were aged 46+ examined were identified. The reasons for the lower prev- years. Among those with stomach cancer, 86% (18/21) alence identified in this study among people aged over 50 were seropositive for antibodies against H. pylori antigens, years and among HIV infected people are unclear. Simi- compared to 87% (761/871) of the comparison group larly, since no consistent differences between religious (odds ratio = 0.8, 95% confidence intervals 0.2–2.9; χ = groups have been identified to date in this study [1-7], it 0.1, p = 0.7). Nor was there a statistically significant asso- is likely that the differences in the prevalence of anti-H. ciation between anti-H pylori antibodies and any other pylori antibodies observed here, between Christians and cancer site or type examined. Muslims, arose by chance. Discussion It is recognised that serum antibodies against H. pylori Here we report the first data from Uganda on the seroprev- may decline because of the development of gastric alence of antibodies against H. pylori. The prevalence of changes, such as malignancy, that can suppress or kill the 87% was broadly comparable to that reported from other infection [20]. Case-control studies have therefore shown hospital series elsewhere on the African continent: 79% in inconsistent associations between antibodies against H. Algeria [12], 71% in Côte d'Ivoire [12], 79% in the Dem- pylori and gastric cancer (reviewed in reference [21]). The ocratic Republic of Congo [13], 85% in Nigeria [14] and lack of an association in this study and in the only other 86–93% in South African blacks [15,16]. In the only other from Africa in which serum antibodies were measured study of H. pylori from Uganda, Wabinga [17] identified a [22] is, therefore, unsurprising. The lack of statistical high frequency of colonisation in gastric endoscopic biop- power in this study (based on only 21 cases of stomach sies from people with upper gastrointestinal symptoms. cancer) and the incomplete diagnostic verification may have further reduced the ability to detect an association. Data on the determinants of infection with H. pylori in Indeed, the only study from Africa to find an association Africa are scant. The prevalence of infection has been between gastric cancer and H. pylori involved assessment found to rise through childhood, reaching over 70% in of infection status microscopically in tissue taken from early adulthood [15] and has been associated with mark- areas adjacent to disease and included only six people ers of Hepatitis A infection [15], premastication of infant's with the tumour [23]. food [18] and low social class in some studies, but not others (reviewed in reference [19]). In this study, few con- The apparent increase in the incidence of gastric cancer sistent associations between either the prevalence or titre seen in Uganda since the 1960s [24], though not statisti- of anti-H. pylori antibodies and any of the risk factors cally significant, may have been influenced by improve- Page 4 of 5 (page number not for citation purposes) Infectious Agents and Cancer 2006, 1:5 http://www.infectagentscancer.com/content/1/1/5 12. Megraud F, Brassens-Rabbe MP, Denis F, Belbouri A, Hoa DQ: ments in diagnosis and is at odds with the decline seen Seroepidemiology of Campylobacter pylori infection in vari- throughout much of the rest of the world. The role of H. ous populations. J Clin Microbiol 1989, 27:1870-73. pylori and other factors in the aetiology of gastric cancer in 13. Glupczynski Y, Bourdeaux L, Verhas M, DePrez C, DeVos D, Devreker T: Use of a urea breath test versus invasive methods Uganda and elsewhere in Africa remains unclear. to determine the prevalence of Helicobacter pylori in Zaire. Eur J Clin Microbiol Infect Dis 1992, 11:322-27. 14. Holcombe C, Omotara BA, Eldridge J, Jones DM: H. pylori, the Acknowledgements most common bacterial infection in Africa: a random sero- Support for this work was provided by Mulago Hospital and Makerere Med- logical study. Am J Gastroenterol 1992, 87:28-30. ical School, Kampala, Uganda, the Imperial Cancer Research Fund (now 15. Sathar MA, Simjee AE, Wittenberg DF, Mayat AM: Seroprevalence Cancer Research UK), the United States Centers for Disease Control and of Helicobacter pylori infection in Natal/KwaZulu, South Prevention (interagency agreement with the Department of Veterans Africa. Eur J Gastroenterol Hepatol 1994, 6:37-41. 16. Sitas F, Sathar MA, Simjee AE, Lombard CJ, Steyn K, Badenhorst CJ, Affairs) and the International Agency for Research on Cancer, World Jooste PL, Bourne L: Helicobacter pylori seroprevalence in the Health Organisation, Lyon. African adult population of the Cape Peninsula. S Afr J Epide- miol Infect 1997, 12:111-114. The Uganda Kaposi's Sarcoma Study Group includes named authors and V 17. Wabinga HR: Frequency of Helicobacter pylori in gastric biopsy of Ugandan African. E Afr Med 1996, 71:691-3. Sembajwe, M Kalinaki, R Byansi, C Rwatooro, S Nambooze, B Tushemeir- 18. Albengue M, Tall F, Dabis F: Epidemiological study of Helico- erwe, N Byabazaire (deceased), E Bitamazire, E Katabira, J Mugerwa bacter pylori transmission from mother to child in Africa. (deceased), D Tindyebwa, C Ateenyi-Agaba, L Marum, J Whitworth, B Rich- Abstract. Rev Esp Enferm Apar Dig 1990, 78:48. ardson, G Reeves, R Weiss and K de Cock. 19. Parkin DM, Ferlay J, Hamdi-Chérif , Sitas F, Thomas JO, Wabinga H, Whelan SL: Cancer in Africa: epidemiology and prevention, IARC Scientific Publications No. 153 IARC Press, Lyon; 2003:371-6. References 20. Forman D: The prevalence of Helicobacter pylori infection in 1. Newton R, Ziegler J, Beral V, Mbidde E, Carpenter L, Wabinga H, gastric cancer. Alimentary Pharmacology and Therapeutics 1995, Mbulataiye S, Appleby P, Reeves G, Jaffe H, the Uganda Kaposi's Sar- 9(2):71-6. coma Study Group: A case-control study of Human Immuno- 21. Danesh J: Is Helicobacter pylori infection a cause of gastric neo- deficiency Virus infection and cancer in adults and children plasia. In Cancer Surveys, Infections and Human cancer Volume 33. residing in Kampala, Uganda. Int J Cancer 2001, 92:622-627. Edited by: Newton R, Beral V, Weiss R. Cold Spring Harbor Labora- 2. Newton R, Ziegler J, Ateenyi-Agaba C, Bousarghin L, Casabonne D, tory Press; 1999. Beral V, Mbidde E, Carpenter L, Reeves G, Parkin DM, Wabinga H, 22. Louw JA, Kidd MSG, Kummer AF, Taylor K, Kotze U, Hanslo D: The Mbulaiteye S, Jaffe H, Bourboulia D, Boshoff C, Coursaget P, the relationship between Helicobacter pylori infection, the viru- Uganda Kaposi's Sarcoma Study Group: The epidemiology of con- lence genotypes of the infecting strain and gastric cancer in junctival squamous cell carcinoma in Uganda. Br J Cancer 2002, the African setting. Helicobacter 2001, 6:268-273. 87:301-308. 23. Jaskiewicz K, Lowrens HD, Woodroof CW, van Wyk MJ, Price SK: 3. Newton R, Ziegler J, Bourboulia D, Casabonne D, Beral V, Mbidde E, The association of Campylobacter pylori with mucosal path- Carpenter L, Reeves G, Parkin DM, Wabinga H, Mbulaiteye S, Jaffe H, ological changes in a population at risk of gastric cancer. S the Uganda Kaposi's Sarcoma Study Group, Weiss R, Boshoff C: The Afr Med J 1989, 75:417-419. sero-epidemiology of Kaposi's sarcoma-associated herpesvi- 24. Wabinga HR, Parkin DM, Wabwire-Mangen F, Nambooze S: Trends rus in adults with cancer, in Uganda. Int J Cancer 2003, in cancer incidence in Kyadondo County, Uganda, 1960– 103:226-232. 1997. Br J Cancer 2000, 82:172. 4. Newton R, Ziegler J, Bourboulia D, Casabonne D, Beral V, Mbidde E, Carpenter L, Parkin DM, Wabinga H, Mbulaiteye S, Jaffe H, the Uganda Kaposi's Sarcoma Study Group, Weiss R, Boshoff C: Infec- tion with Kaposi's sarcoma-associated herpesvirus (KSHV) and human immunodeficiency virus (HIV) in relation to the risk and clinical presentation of Kaposi's sarcoma in Uganda. Br J Cancer 2003, 89(3):502-504. 5. Newton R, Bousarghin L, Ziegler J, Casabonne D, Beral V, Mbidde E, Carpenter L, Parkin DM, Wabinga H, Mbulaiteye S, Jaffe H, Touzé A, Coursaget P, the Uganda Kaposi's Sarcoma Study Group: Human papillomaviruses and cancer in Uganda. Eur J Cancer Prev 2004, 13(2):113-118. 6. Ziegler JL, Newton R, Katongole-Mbidde E, Mbulataiye S, DeCock K, Wabinga H, Mugerwa J, Katabira E, Jaffe H, Parkin DM, Reeves G, Beral V, for the Uganda Kaposi's sarcoma study group: Risk factors for HIV-associated Kaposi's sarcoma in Uganda: a case-con- trol study of 1026 Adults. AIDS 1997, 11:1619-1626. 7. Ziegler J, Newton R, Bourboulia D, Casabonne D, Beral V, Mbidde E, Publish with Bio Med Central and every Carpenter L, Reeves G, Wabinga H, Mbulaiteye S, Jaffe H, the Uganda scientist can read your work free of charge Kaposi's Sarcoma Study Group, Weiss R, Boshoff C: Risk factors for Kaposi's sarcoma: a case-control study of HIV seronegative "BioMed Central will be the most significant development for people in Uganda. Int J Cancer 2003, 103:233-240. disseminating the results of biomedical researc h in our lifetime." 8. Pei Z, Ellison RT, Blaser MJ: Identification, purification and char- acterization of major antigenic proteins of Campylobacter Sir Paul Nurse, Cancer Research UK jejuni. J Biol Chem 1991, 266:1663-1669. Your research papers will be: 9. Khanna B, Cutler A, Israel NR, Perry M, Lastovica A, Fields PI, Gold BD: Use caution with serologic testing for Helicobacter pylori available free of charge to the entire biomedical community infection in children. J Infect Dis 1988, 178:460-465. peer reviewed and published immediately upon acceptance 10. Perez-Perez GI, Dwoskin BM, Chadon JE, Blaser MJ: Campylobacter pylori antibodies in humans. Ann Intern Med 1998, 109:11-17. cited in PubMed and archived on PubMed Central 11. Gold BD, Khanna B, Huang LM, Lee C-Y, Banatvala N: Helicobacter yours — you keep the copyright pylori acquisition in infancy after decline of maternal passive immunity. Pediatric Research 1997, 41:641-646. BioMedcentral Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp Page 5 of 5 (page number not for citation purposes) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Infectious Agents and Cancer Springer Journals

Helicobacter pylori and cancer among adults in Uganda

Download PDF
5 pages

Loading next page...
 
/lp/springer-journals/helicobacter-pylori-and-cancer-among-adults-in-uganda-exMcjjkVFt

References (28)

Publisher
Springer Journals
Copyright
Copyright © 2006 by Newton et al; licensee BioMed Central Ltd.
Subject
Biomedicine; Cancer Research; Infectious Diseases; Oncology
eISSN
1750-9378
DOI
10.1186/1750-9378-1-5
pmid
17150134
Publisher site
See Article on Publisher Site

Abstract

Data from Africa on infection with Helicobacter pylori (H. pylori) are sparse. Therefore, as part of an epidemiological study of cancer in Uganda, we investigated the prevalence and determinants of antibodies against H. pylori among 854 people with different cancer types and benign tumours. Patients were recruited from hospitals in Kampala, Uganda, interviewed about various demographic and lifestyle factors and tested for antibodies against H. pylori. In all patients combined, excluding those with stomach cancer (which has been associated with H. pylori infection), the prevalence of antibodies was 87% (723/833) overall, but declined with increasing age (p = 0.02) and was lower among people who were HIV seropositive compared to seronegative (p < 0.001). Otherwise, there were few consistent epidemiological associations. Among those with stomach cancer, 18/21 (86%) had anti-H. pylori antibodies (odds ratio 0.8, 95% confidence intervals 0.2–2.9, p = 0.7; estimated using all other patients as controls, with adjustment for age, sex and HIV serostatus). No other cancer site or type was significantly associated with anti-H. pylori antibodies. The prevalence of H. pylori reported here is broadly in accord with results from other developing countries, although the determinants of infection and its' role in the aetiology of gastric cancer in Uganda remain unclear. ological study of cancer in Kampala, Uganda [1-7]. Data Background The work described in this report was part of an epidemi- from Africa on infection with Helicobacter pylori (H. pylori) Page 1 of 5 (page number not for citation purposes) Infectious Agents and Cancer 2006, 1:5 http://www.infectagentscancer.com/content/1/1/5 are sparse. Therefore, we examine here the role of antibod- the out-patient department of Mulago hospital and for ies against H. pylori in relation to the risk of cancer and 804 patients with cancer or benign tumours, for whom a investigate the prevalence and determinants of infection stored blood sample was available for testing. The latter among 854 people with different cancer types and benign group comprised people with cancers of the oral cavity tumours. (26), oesophagus (38), stomach (21), liver (52), skin (22), breast (69), cervix (190), ovary (22), prostate (10), Materials and methods penis (14), eye (63), and non-Hodgkin's lymphoma (46), Full details of the methods are provided elsewhere [1,2]. Hodgkin's disease (24), Kaposi's sarcoma (46), other can- Briefly between 1994 and 1998, we recruited adults 15 cer sites or types (126) and benign tumours (35). years or older with a new diagnosis of cancer from the wards and out-patient clinics of the main hospitals in Data were computerised by trained clerks using EPI-INFO Kampala, Uganda. After informed consent and counsel- software (CDC, Atlanta) and statistical analyses were con- ling, patients were interviewed and tested for infection ducted using STATA (STATA Corporation, Texas). Only a with HIV-1 using the Cambridge Bioscience Recombigen small proportion of those tested were seronegative for ELISA (Cambridge, MA) on sera or the GACELISA method antibodies against H. pylori (optical density <0.9) or had (Murex, Dartford, UK) on saliva. Cancer diagnoses were an indeterminate result (optical density 0.9–1.3). In all established by histology or other laboratory investigation, analyses, those with indeterminate results were consid- where possible. Diagnoses made on clinical grounds ered to be seronegative. In order to examine potential con- alone were reviewed by the investigators. The study was founding factors, the risk of being seropositive for approved by the Committee on Human Research (VA antibodies against H. pylori was examined in relation to Medical Centre and University of California San Fran- various social and demographic factors among all patients cisco) and by the Uganda National Council for Science combined (but excluding stomach cancer, which has been and Technology. associated with H. pylori infection). Odds ratios (OR) were estimated using unconditional logistic regression model- Following HIV testing, remaining sera were stored at ling with adjustment for sex, age group (<30, 30–45, 46+) minus 80 Celsius and were later shipped on dry ice to the and HIV serostatus. When calculating odds ratios in rela- Centres for Disease Control and Prevention, Atlanta, USA, tion to anti-H. pylori antibodies, for each cancer site or for H. pylori testing. Assays were performed by a single type, the comparison group included all other patients investigator who was blind to the diagnosis of the patient with the exception of stomach cancer. Tests for association from whom the blood was obtained. Briefly, H. pylori used the χ test for linear trend on one degree of freedom organisms were grown overnight in brucella broth and all p values are 2-sided. Risk factors for high titres of (GIBCO Laboratories, Madison, WS) with 10% fetal antibodies against H. pylori were examined amongst all bovine serum (Sigma, St. Louis, MO), 5 μg/ml trimetho- patients combined (excluding stomach cancer), but no prim and 10 μg/ml vancomycin (Sigma). H. pylori antigen clear associations were identified and the data are not extraction and protein isolation were done by gentle shown. freeze-thaw sonication (Heat System, Farmingdale, NY) [8,9]. A standard protein assay (Pierce, Rockford, IL) was Results used to determine the accurate and reproducible quantity H. pylori antibody status was available for 854 people; 87% (741) were seropositive, 4% (38) were seronegative of solid-phase antigen for our microtitre research ELISA [10]. Cross-reactivity and specificity of H. pylori whole-cell and 9% (75) had an indeterminate result. Table 1 shows antigens has been described previously [9,10]. Optical the association between H. pylori serostatus and selected density (OD) values at a wavelength of 492 nm were social and demographic factors among all patients exclud- determined in triplicate for each biopsy-confirmed con- ing those with stomach cancer. The prevalence of antibod- trol patient sera, using a standard 96-well microtiter plate ies did not vary with sex, but declined with increasing age ELISA spectrophotometer (Fisher Scientific, Pittsburgh, (χ = 5.1, p = 0.02) and was lower among people who PA). The mean OD values were then calculated. The ELISA were HIV seropositive compared to seronegative (χ = cut-off values were derived using known H. pylori-positive 16.2, p < 0.001). No other factor examined was associated and negative control sera as previously described [9,11]. with antibodies against H. pylori with the exception of reli- In previous validation studies the assay has demonstrated gion: the prevalence of antibodies was higher among Mus- a high and reproducible sensitivity and specificity in Afri- lims than among Christians (χ = 5.6, p = 0.02). can patients as compared to upper endoscopy and biopsy; sensitivity >88%, specificity >90% [9,11]. Table 2 shows the association between anti-H. pylori anti- bodies and specific cancer sites or types, together with the Serological results were available for 50 people with non- proportion of cancers with a laboratory verification of malignant manifestations of HIV disease, recruited from diagnosis. Overall, 62% of cancers were diagnosed on the Page 2 of 5 (page number not for citation purposes) Infectious Agents and Cancer 2006, 1:5 http://www.infectagentscancer.com/content/1/1/5 Table 1: Odds ratios (OR) for H. pylori seropositivity according to various social and demographic factors among all patients (excluding those with stomach cancer) 1 1 Variable Number positive/ OR (95% CI) Variable Number positive/ OR (95% CI) total total Sex Time to market Male 247/284 1.0 <30 minutes 323/375 1.0 Female 476/549 0.9 (0.6–1.3) 30+ minutes 278/320 1.0 (0.7–1.6) 2 2 χ = 0.5, p = 0.5 χ = 0.0, p = 0.9 1 1 Age group Size of community <36 years 269/309 1.0 >100 houses 220/262 1.0 36–50 years 250/282 1.0 (0.6–1.6) 10–99 houses 421/482 1.3 (0.9–2.1) 51+ years 204/242 0.5 (0.3–0.9) <10 houses 51/57 1.6 (0.6–4.1) 2 2 χ = 5.1, p = 0.02 χ = 1.0, p = 0.3 1 1 HIV serostatus Ever travel from home Negative 514/574 1.0 Positive 201/250 0.4 (0.3–0.6) Yes 91/109 1.0 χ = 16.2, p<0.001 No 599/688 1.4 (0.7–2.5) = 1.0, p = 0.3 Region of residence Household size Kampala 161/188 1.0 <6 people 357/410 1.0 Rest of Uganda 560/643 1.0 (0.6–1.6) 6+ people 354/411 0.8 (0.5–1.2) 2 2 χ = 0.0, p = 0.9 χ = 1.6, p = 0.2 1 1 Tribe Number of siblings Baganda 349/410 1.0 <6 siblings 379/438 1.0 Other 374/423 1.2 (0.8–1.8) 6+ siblings 333/384 0.9 (0.6–1.4) 2 2 χ = 0.8, p = 0.4 χ = 0.1, p = 0.8 1 1 Religion Number of children Muslim 95/100 1.0 <7 children 341/397 1.0 Christian 619/723 0.3 (0.1–0.8) 7+ children 290/329 1.1 (0.7–1.8) 2 2 χ = 5.6, p = 0.02 χ = 0.1, p = 0.8 1 1 Occupation Tobacco consumption Cultivator 334/387 1.0 Other 382/439 1.1 (0.7–1.9) Never smoker 562/647 1.0 χ = 0.1, p = 0.8 Past smoker 81/92 1.2 (0.6–2.5) Current smoker 68/82 0.7 (0.4–1.4) χ = 0.5, p = 0.5 Education level Alcohol consumption No school 153/175 1.0 Primary 361/415 0.9 (0.5–1.6) Never 391/443 1.0 Secondary/tertiary 191/225 0.7 (0.4–1.4) About once/week 134/157 0.8 (0.5–1.4) χ = 1.0, p = 0.3 2–4 days/week 91/106 0.9 (0.5–1.6) Most days 93/113 0.7 (0.4–1.2) χ = 1.6, p = 0.2 Household Income Lifetime number of (Ug. Sh.) sexual partners 15,000+ 464/533 1.0 1–2 partners 182/204 1.0 <15,000 196/230 0.9 (0.6–1.5) 3–9 partners 305/352 0.9 (0.5–1.5) χ = 0.1, p = 0.8 10+ partners 196/232 0.7 (0.4–1.3) χ = 1.2, p = 0.3 1. Odds ratios adjusted for age group (<30, 30–45, 46+), sex and HIV serostatus basis of histology or other laboratory investigation, but 21) had the diagnosis confirmed by a laboratory investi- the figure varied by cancer site or type, being lowest for gation. prostate cancer (44%) and highest for Kaposi's sarcoma (91%). Of those people with stomach cancer, 90% (19/ Among 21 cases with stomach cancer, one was HIV serop- ositive, 13 were women, one was aged <30 years, three Page 3 of 5 (page number not for citation purposes) Infectious Agents and Cancer 2006, 1:5 http://www.infectagentscancer.com/content/1/1/5 Table 2: H. pylori serostatus for different cancer sites or types and non-malignant conditions, together with the percentage of each cancer with laboratory verification of diagnosis 1 2 Cancer site or type Percentage with laboratory Number anti-H. pylori anti- Odds ratio (95% CI) χ and p value verification of diagnosis body positive/total Stomach 90% 18/21 0.8 (0.2–2.9) χ = 0.1, p = 0.7 All controls - 723/833 1.0 - Oral 56% 21/26 0.6 (0.2–1.7) χ = 0.9, p = 0.4 Oesophagus 45% 37/38 5.1 (0.7–38) χ = 2.5, p = 0.1 Liver 56% 44/52 0.7 (0.3–1.6) χ = 0.8, p = 0.4 Skin 77% 18/22 0.6 (0.2–1.9) χ = 0.8, p = 0.4 = 0.6, p = 0.4 Breast 62% 62/69 1.4 (0.6–3.5) χ Cervix 50% 173/190 1.6 (0.9–2.8) χ = 2.1, p = 0.2 Ovary 73% 20/22 1.1 (0.3–5.0) χ = 0.0, p = 0.9 Prostate 44% 7/10 0.3 (0.1–1.4) χ = 2.4, p = 0.1 Penis 57% 12/14 0.9 (0.2–4.3) χ = 0.0, p = 0.9 Conjunctiva 66% 33/38 1.3 (0.5–3.5) χ = 0.3, p = 0.6 Other eye 52% 22/25 1.1 (0.3–3.9) χ = 0.0, p = 0.9 Non-Hodgkin's 76% 39/46 0.8 (0.3–1.8) χ = 0.4, p = 0.5 lymphoma Hodgkin's lymphoma 83% 20/24 0.6 (0.2–1.9) χ = 0.7, p = 0.4 Kaposi's sarcoma 91% 36/46 0.7 (0.3–1.6) χ = 0.9, p = 0.4 1. Odds ratios adjusted for age group (<30, 30–45, 46+), sex and HIV serostatus 2. The comparison group for the calculation of odds ratios includes all other cancers and non-malignant conditions, excluding stomach cancer Note: H. Pylori serostatus: negative – optical density (O.D.) 0.0–0.8; indeterminate – O.D. 0.9–1.3; positive – O.D. 1.4–4.0. For the purposes of this analysis, those with an indeterminate result were considered to be seronegative. were aged between 30–45 years and 17 were aged 46+ examined were identified. The reasons for the lower prev- years. Among those with stomach cancer, 86% (18/21) alence identified in this study among people aged over 50 were seropositive for antibodies against H. pylori antigens, years and among HIV infected people are unclear. Simi- compared to 87% (761/871) of the comparison group larly, since no consistent differences between religious (odds ratio = 0.8, 95% confidence intervals 0.2–2.9; χ = groups have been identified to date in this study [1-7], it 0.1, p = 0.7). Nor was there a statistically significant asso- is likely that the differences in the prevalence of anti-H. ciation between anti-H pylori antibodies and any other pylori antibodies observed here, between Christians and cancer site or type examined. Muslims, arose by chance. Discussion It is recognised that serum antibodies against H. pylori Here we report the first data from Uganda on the seroprev- may decline because of the development of gastric alence of antibodies against H. pylori. The prevalence of changes, such as malignancy, that can suppress or kill the 87% was broadly comparable to that reported from other infection [20]. Case-control studies have therefore shown hospital series elsewhere on the African continent: 79% in inconsistent associations between antibodies against H. Algeria [12], 71% in Côte d'Ivoire [12], 79% in the Dem- pylori and gastric cancer (reviewed in reference [21]). The ocratic Republic of Congo [13], 85% in Nigeria [14] and lack of an association in this study and in the only other 86–93% in South African blacks [15,16]. In the only other from Africa in which serum antibodies were measured study of H. pylori from Uganda, Wabinga [17] identified a [22] is, therefore, unsurprising. The lack of statistical high frequency of colonisation in gastric endoscopic biop- power in this study (based on only 21 cases of stomach sies from people with upper gastrointestinal symptoms. cancer) and the incomplete diagnostic verification may have further reduced the ability to detect an association. Data on the determinants of infection with H. pylori in Indeed, the only study from Africa to find an association Africa are scant. The prevalence of infection has been between gastric cancer and H. pylori involved assessment found to rise through childhood, reaching over 70% in of infection status microscopically in tissue taken from early adulthood [15] and has been associated with mark- areas adjacent to disease and included only six people ers of Hepatitis A infection [15], premastication of infant's with the tumour [23]. food [18] and low social class in some studies, but not others (reviewed in reference [19]). In this study, few con- The apparent increase in the incidence of gastric cancer sistent associations between either the prevalence or titre seen in Uganda since the 1960s [24], though not statisti- of anti-H. pylori antibodies and any of the risk factors cally significant, may have been influenced by improve- Page 4 of 5 (page number not for citation purposes) Infectious Agents and Cancer 2006, 1:5 http://www.infectagentscancer.com/content/1/1/5 12. Megraud F, Brassens-Rabbe MP, Denis F, Belbouri A, Hoa DQ: ments in diagnosis and is at odds with the decline seen Seroepidemiology of Campylobacter pylori infection in vari- throughout much of the rest of the world. The role of H. ous populations. J Clin Microbiol 1989, 27:1870-73. pylori and other factors in the aetiology of gastric cancer in 13. Glupczynski Y, Bourdeaux L, Verhas M, DePrez C, DeVos D, Devreker T: Use of a urea breath test versus invasive methods Uganda and elsewhere in Africa remains unclear. to determine the prevalence of Helicobacter pylori in Zaire. Eur J Clin Microbiol Infect Dis 1992, 11:322-27. 14. Holcombe C, Omotara BA, Eldridge J, Jones DM: H. pylori, the Acknowledgements most common bacterial infection in Africa: a random sero- Support for this work was provided by Mulago Hospital and Makerere Med- logical study. Am J Gastroenterol 1992, 87:28-30. ical School, Kampala, Uganda, the Imperial Cancer Research Fund (now 15. Sathar MA, Simjee AE, Wittenberg DF, Mayat AM: Seroprevalence Cancer Research UK), the United States Centers for Disease Control and of Helicobacter pylori infection in Natal/KwaZulu, South Prevention (interagency agreement with the Department of Veterans Africa. Eur J Gastroenterol Hepatol 1994, 6:37-41. 16. Sitas F, Sathar MA, Simjee AE, Lombard CJ, Steyn K, Badenhorst CJ, Affairs) and the International Agency for Research on Cancer, World Jooste PL, Bourne L: Helicobacter pylori seroprevalence in the Health Organisation, Lyon. African adult population of the Cape Peninsula. S Afr J Epide- miol Infect 1997, 12:111-114. The Uganda Kaposi's Sarcoma Study Group includes named authors and V 17. Wabinga HR: Frequency of Helicobacter pylori in gastric biopsy of Ugandan African. E Afr Med 1996, 71:691-3. Sembajwe, M Kalinaki, R Byansi, C Rwatooro, S Nambooze, B Tushemeir- 18. Albengue M, Tall F, Dabis F: Epidemiological study of Helico- erwe, N Byabazaire (deceased), E Bitamazire, E Katabira, J Mugerwa bacter pylori transmission from mother to child in Africa. (deceased), D Tindyebwa, C Ateenyi-Agaba, L Marum, J Whitworth, B Rich- Abstract. Rev Esp Enferm Apar Dig 1990, 78:48. ardson, G Reeves, R Weiss and K de Cock. 19. Parkin DM, Ferlay J, Hamdi-Chérif , Sitas F, Thomas JO, Wabinga H, Whelan SL: Cancer in Africa: epidemiology and prevention, IARC Scientific Publications No. 153 IARC Press, Lyon; 2003:371-6. References 20. Forman D: The prevalence of Helicobacter pylori infection in 1. Newton R, Ziegler J, Beral V, Mbidde E, Carpenter L, Wabinga H, gastric cancer. Alimentary Pharmacology and Therapeutics 1995, Mbulataiye S, Appleby P, Reeves G, Jaffe H, the Uganda Kaposi's Sar- 9(2):71-6. coma Study Group: A case-control study of Human Immuno- 21. Danesh J: Is Helicobacter pylori infection a cause of gastric neo- deficiency Virus infection and cancer in adults and children plasia. In Cancer Surveys, Infections and Human cancer Volume 33. residing in Kampala, Uganda. Int J Cancer 2001, 92:622-627. Edited by: Newton R, Beral V, Weiss R. Cold Spring Harbor Labora- 2. Newton R, Ziegler J, Ateenyi-Agaba C, Bousarghin L, Casabonne D, tory Press; 1999. Beral V, Mbidde E, Carpenter L, Reeves G, Parkin DM, Wabinga H, 22. Louw JA, Kidd MSG, Kummer AF, Taylor K, Kotze U, Hanslo D: The Mbulaiteye S, Jaffe H, Bourboulia D, Boshoff C, Coursaget P, the relationship between Helicobacter pylori infection, the viru- Uganda Kaposi's Sarcoma Study Group: The epidemiology of con- lence genotypes of the infecting strain and gastric cancer in junctival squamous cell carcinoma in Uganda. Br J Cancer 2002, the African setting. Helicobacter 2001, 6:268-273. 87:301-308. 23. Jaskiewicz K, Lowrens HD, Woodroof CW, van Wyk MJ, Price SK: 3. Newton R, Ziegler J, Bourboulia D, Casabonne D, Beral V, Mbidde E, The association of Campylobacter pylori with mucosal path- Carpenter L, Reeves G, Parkin DM, Wabinga H, Mbulaiteye S, Jaffe H, ological changes in a population at risk of gastric cancer. S the Uganda Kaposi's Sarcoma Study Group, Weiss R, Boshoff C: The Afr Med J 1989, 75:417-419. sero-epidemiology of Kaposi's sarcoma-associated herpesvi- 24. Wabinga HR, Parkin DM, Wabwire-Mangen F, Nambooze S: Trends rus in adults with cancer, in Uganda. Int J Cancer 2003, in cancer incidence in Kyadondo County, Uganda, 1960– 103:226-232. 1997. Br J Cancer 2000, 82:172. 4. Newton R, Ziegler J, Bourboulia D, Casabonne D, Beral V, Mbidde E, Carpenter L, Parkin DM, Wabinga H, Mbulaiteye S, Jaffe H, the Uganda Kaposi's Sarcoma Study Group, Weiss R, Boshoff C: Infec- tion with Kaposi's sarcoma-associated herpesvirus (KSHV) and human immunodeficiency virus (HIV) in relation to the risk and clinical presentation of Kaposi's sarcoma in Uganda. Br J Cancer 2003, 89(3):502-504. 5. Newton R, Bousarghin L, Ziegler J, Casabonne D, Beral V, Mbidde E, Carpenter L, Parkin DM, Wabinga H, Mbulaiteye S, Jaffe H, Touzé A, Coursaget P, the Uganda Kaposi's Sarcoma Study Group: Human papillomaviruses and cancer in Uganda. Eur J Cancer Prev 2004, 13(2):113-118. 6. Ziegler JL, Newton R, Katongole-Mbidde E, Mbulataiye S, DeCock K, Wabinga H, Mugerwa J, Katabira E, Jaffe H, Parkin DM, Reeves G, Beral V, for the Uganda Kaposi's sarcoma study group: Risk factors for HIV-associated Kaposi's sarcoma in Uganda: a case-con- trol study of 1026 Adults. AIDS 1997, 11:1619-1626. 7. Ziegler J, Newton R, Bourboulia D, Casabonne D, Beral V, Mbidde E, Publish with Bio Med Central and every Carpenter L, Reeves G, Wabinga H, Mbulaiteye S, Jaffe H, the Uganda scientist can read your work free of charge Kaposi's Sarcoma Study Group, Weiss R, Boshoff C: Risk factors for Kaposi's sarcoma: a case-control study of HIV seronegative "BioMed Central will be the most significant development for people in Uganda. Int J Cancer 2003, 103:233-240. disseminating the results of biomedical researc h in our lifetime." 8. Pei Z, Ellison RT, Blaser MJ: Identification, purification and char- acterization of major antigenic proteins of Campylobacter Sir Paul Nurse, Cancer Research UK jejuni. J Biol Chem 1991, 266:1663-1669. Your research papers will be: 9. Khanna B, Cutler A, Israel NR, Perry M, Lastovica A, Fields PI, Gold BD: Use caution with serologic testing for Helicobacter pylori available free of charge to the entire biomedical community infection in children. J Infect Dis 1988, 178:460-465. peer reviewed and published immediately upon acceptance 10. Perez-Perez GI, Dwoskin BM, Chadon JE, Blaser MJ: Campylobacter pylori antibodies in humans. Ann Intern Med 1998, 109:11-17. cited in PubMed and archived on PubMed Central 11. Gold BD, Khanna B, Huang LM, Lee C-Y, Banatvala N: Helicobacter yours — you keep the copyright pylori acquisition in infancy after decline of maternal passive immunity. Pediatric Research 1997, 41:641-646. BioMedcentral Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp Page 5 of 5 (page number not for citation purposes)

Journal

Infectious Agents and Cancer – Springer Journals

Published: Nov 7, 2006

There are no references for this article.