Decrease in liver cancer incidence rates in Bamako, Mali over 28 years of population-based cancer registration (1987-2015).

BACKGROUND: Primary liver cancer is common in West Africa due to endemic risk factors. However, epidemiological studies of the global burden and trends of liver cancer are limited. We report changes in trends of the incidence of liver cancer over a period of 28 years using the population-based cancer registry of Bamako, Mali. AIM: To assess the trends and patterns of liver cancer by gender and age groups by analyzing the cancer registration data accumulated over 28 years (1987-2015) of activity of the population-based registry of the Bamako district. METHODS: Data obtained since the inception of the registry in 1987 through 2015 were stratified into three periods (1987-1996, 1997-2006, and 2007-2015). Age-standardized rates were estimated by direct standardization using the world population. Incidence rate ratios and the corresponding 95% confidence intervals (CI) were estimated using the early period as the reference (1987-1996). Joinpoint regression models were used to assess the annual percentage change and highlight trends over the entire period (from 1987 to 2015). RESULTS: Among males, the age-standardized incidence rates significantly decreased from 19.41 (1987-1996) to 13.12 (1997-2006) to 8.15 (2007-2015) per 105 person-years. The incidence rate ratio over 28 years was 0.42 (95%CI: 0.34-0.50), and the annual percentage change was -4.59 [95%CI: (-6.4)-(-2.7)]. Among females, rates dropped continuously from 7.02 (1987-1996) to 2.57 (2007-2015) per 105 person-years, with an incidence rate ratio of 0.37 (95%CI: 0.28-0.45) and an annual percentage change of -5.63 [95%CI: (-8.9)-(-2.3)]. CONCLUSION: The population-based registration showed that the incidence of primary liver cancer has steadily decreased in the Bamako district over 28 years. This trend does not appear to result from biases or changes in registration practices. This is the first report of such a decrease in an area of high incidence of liver cancer in Africa. This decrease may be explained by the changes and diversity of diet that could reduce exposure to aflatoxins through dietary contamination in this population.

Socio-demographic and reproductive determinants of cervical neoplasia in seven sub-Sahara African countries.

BACKGROUND: Since most human papilloma virus (HPV) infections regress without any intervention, HPV is a necessary but may not be a solely sufficient cause of cervical intraepithelial neoplasia (CIN) and cervical cancer. Hence, the influence of cofactors on progression from cervical HPV infection to high-grade CIN and invasive cervical cancer has been a subject of intensive research. OBJECTIVE: We assessed the effect of socio-demographic and sexual reproductive factors on the prevalence of invasive cervical cancer and CIN diagnosed in cross-sectional cervical cancer screening projects carried out in seven sites of different sub-Saharan countries. METHODS: Between January 2000 and August 2007, healthy women aged 25-59 who participated in the screening projects were interviewed for socio-demographic, reproductive, and behavioral characteristics, investigated for disease confirmation with colposcopy, and had biopsies directed from colposcopically abnormal areas by trained local physicians. Odds ratios (ORs) and their 95% confidence intervals (CIs) from logistic regression analyses were used to assess the effect of women characteristics on CIN 1, CIN 2-3, CIN 3, and invasive cancer outcome measures. RESULTS: Among 47,361 women screened and investigated for disease confirmation, CIN 1 was diagnosed in 1,069 (2.3%), CIN 2 in 517 (1.1%), CIN 3 in 175 (0.5%), and invasive cancer in 485 (1.0%). The site-specific prevalence of CIN 2-3 lesions ranged from 0.3 to 5.1% and from 0.2 to 1.9% for invasive cancers. Risk factors for CIN 2-3 were being widowed or separated versus currently married (OR 1.3, 95% CI 1.0-1.7 a); and having had at least four pregnancies versus zero or one pregnancy (OR at least 1.4-fold, 95% CI 1.1-1.8). Risk factors for invasive cancer were being widowed or separated versus currently married (OR 2.0, 95% CI 1.3-3.1); and having had at least three pregnancies versus zero or one pregnancy (OR at least 3.0-fold, 95% CI 2.1-4.2). Additionally, cervical cancer risk increased with increasing age, age at menarche, and age at marriage, while the risk decreased with increasing level of education and in those with some form of employment compared to housewives. CONCLUSION: The exposure of the exocervix and/or the increased levels of estrogen and progesterone for more prolonged periods during pregnancy in multiparous women and the vulnerability of widowed/separated women in society might result in increased risk of cervical neoplasia more so among women exposed to HPV infection. High parity probably explains the persistently high rates of cervical cancer in sub-Saharan Africa.

Breast cancer in pre-menopausal women in West Africa: analysis of temporal trends and evaluation of risk factors associated with reproductive life.

BACKGROUND: In West Africa, trends and risk factors for breast cancer (BC) have been rarely studied. METHODS: Here we have analyzed trends of BC over two periods in two population-based cancer registries, in Mali-Bamako (1987-1997; 1998-2009) and in The Gambia (1988-1997; 1998-2006). We have conducted a case-control study (n = 253 cases, 249 controls) on risk factors associated with reproductive life stratified by menopausal status in Bamako. RESULTS: Between the two periods, BC incidence rates increased by 20% (incidence rate ratio (IRR) 1.20 (95% CI [1.07-1.35])) in Bamako, with an annual percentage change of 2% (95% CI [0.4-3.6]). The increase was of 30% in women under 55 years (IRR 1.30 (95% CI [1.14-1.60])). A similar pattern was observed in The Gambia for women under 50 years (IRR 1.47 (95% CI [1.07-2.01])). Overall, pre-menopausal breast cancer was predominant in both countries. In contrary to what is well established, case-control study showed that late age at menarche (>14 years) increased the risk of BC among pre-menopausal women (OR: 2.02 (95% CI [1.08-3.78])) while it tended to be protective in post-menopausal women (OR: 0.61 (95% CI [0.29-1.29])). Later age at a first pregnancy (>20 years) was associated with a reduction of risk in pre-menopausal women (OR: 0.41 (95% CI [0.18-0.89])). CONCLUSION: These results indicate that the burden of pre-menopausal BC is increasing in West African countries. These cancers appear to be associated with distinct reproductive risk factors, highlighting the need for better understanding the biological bases of early BC in African populations.

Type-specific seroprevalence of herpes simplex virus type 2 and associated risk factors in middle-aged women from 6 countries: the IARC multicentric study.

GOAL: To determine type-specific seroprevalence of herpes simplex viruses (HSV-1 and HSV-2) and HSV-2 risk factors. STUDY DESIGN: Six-hundred fifty eight middle-aged control women (hospital-based in 4 of 6 countries) from a multicenter cervical cancer case-control study participated from 1985 to 1997. Type-specific serum IgG antibodies against HSV-1 and HSV-2 were detected with Western Blot. RESULTS: HSV-1 seroprevalence was 89% to 100% everywhere except Thailand (51%). HSV-2 seroprevalence ranged from 9% (Spain) to 57% (Colombia), and was independently associated with having >or=2 lifetime sexual partners overall [Odds ratio (OR), 2.1; 95% confidence interval (CI) 2.5-3.1], and in Morocco (OR, 2.7; 95% CI, 1.2-6.1) and Thailand (OR, 4.4; 95% CI, 1.3-15.4), and with being unmarried in Colombia, Peru, Spain, but not significantly in Mali. Women whose male partner's sexual debut was

Risk factors of invasive cervical cancer in Mali.

BACKGROUND: Cervical cancer is the most common cancer in women in Mali and the second commonest cause of cancer mortality. METHODS: As part of an international effort to evaluate the role of human papillomavirus (HPV) in the aetiology of cervical cancer, we conducted a hospital-based case-control study in three medical centres in Bamako during 1994-1995. A total of 82 cases (invasive cervical cancer patients) and 97 controls matched to the cases for age were included. Information on risk factors was collected through personal interview. Serum antibodies to HPV 16, 18 and 31 virus like particles (VLP) were detected using ELISA assays. Polymerase chain reaction was used to detect HPV DNA in frozen biopsies of cases. RESULTS: Human papillomavirus 6, 18, 31 VLP were detected in 60.4% of cases and 45.4% of controls (P = 0.03). Overall, HPV DNA was identified in 96.9% of the cervical cancer cases. Risk factors for cervical cancer were parity >10 versus <5 children ([odds ratio] OR = 4.8, 95% CI : 1.5-14.7), never having practised vaginal douching (OR = 17.6, 95% CI : 4.2-74.7), re-using home-made feminine napkins (OR = 45.9, 95% CI : 8.8-238.7) and having a husband with more than two wives (OR = 5.3, 95% CI : 1.3-21.3). CONCLUSIONS: These data provide further evidence on the role of HPV in cervical cancer and show that high parity and poor genital hygiene conditions were the main co-factors for cervical cancer in this population with prevalent HPV infection.