Regional, racial, gender, and tumor biology disparities in breast cancer survival rates in Africa: A systematic review and meta-analysis

Abstract

Background: The survival rates from breast cancer in Africa are poor and yet the incidence rates are on the rise. In this study, we hypothesized that, in Africa, a continent with great disparities in socio-economic status, race, tumor biology, and cultural characteristics, the survival rates from breast cancer vary greatly based on region, tumor biology (hormone receptor), gender, and race. We aimed to conduct the first comprehensive systematic review and meta-analysis on region, gender, tumor-biology and race-specific 5-year breast cancer survival rates in Africa and compared them to 20-year survival trends in the United States.

Methods: We searched MEDLINE, EMBASE, and Cochrane Library to identify studies on breast cancer survival in African published before October 17, 2018. Pooled 5-year survival rates of breast cancer were estimated by random-effects models. We explored sources of heterogeneity through subgroup meta-analyses and meta-regression. Results were reported as absolute difference (AD) in percentages. We compared the survival rates of breast cancer in Africa and the United States.

Findings: There were 54 studies included, consisting of 18,970 breast cancer cases. There was substantial heterogeneity in the survival rates (mean 52.9%, range 7-91%, I2 = 99.1%; p for heterogeneity <0.0001). Meta-regression analyses suggested that age and gender-adjusted 5-year survival rates were lower in sub-Saharan Africa compared to north Africa (AD: -25.4%; 95% CI: -34.9 - -15.82%), and in predominantly black populations compared to predominantly non-black populations (AD: -25.9%; 95% CI: 35.40 - -16.43%). Survival rates were 10 percentage points higher in the female population compared to male, but the difference was not significant. Progesterone and estrogen receptor-positive breast cancer subtypes were positively associated with survival (r = 0.39, p = 0.08 and r = 0.24, p = 0.29 respectively), but triple-negative breast cancer was negatively associated with survival. Survival rates are increasing over time more in non-black Africans (55% in 2000 versus 65% in 2018) compared to black Africans (33% in 2000 versus 40% in 2018); but, the survival rates for Africans are still significantly lower when compared to black (76% in 2015) and white (90% in 2015) populations in the United States.

Conclusion: Regional, sub-regional, gender, and racial disparities exist, influencing the survival rates of breast cancer in Africa. Therefore, region and race-specific public health interventions coupled with prospective genetic studies are urgently needed to improve breast cancer survival in this region.

Fig 1. PRISMA flowchart of a systematic review of the breast cancer survival rate in Africa.

Fig 2. Survival rate (%) of breast cancer shaded by the country from which the papers used in the meta-analysis were published.

No data in black shade.

Fig 3. Pooled 5-year survival rate in breast cancer population in Africa.

Fig 4. Pooled 5-year survival rate in breast cancer population in Africa stratified by country from which the papers used in the meta-analysis were published.

Fig 5. Pooled 5-year survival rate in breast cancer population in Africa stratified by gender.

Fig 6. Pooled 5-year survival rate in breast cancer population in Africa stratified by region (sub-Saharan Africa vs. non-sub-Saharan Africa) from which the papers used in the meta-analysis were published.

Fig 7. Pooled 5-year survival rate in breast cancer population in Africa stratified by sub-regions (East Africa, West Africa, South Africa and North Africa) from which the papers used in the meta-analysis were published.

Fig 8. Pooled 5-year survival rate in breast cancer population in Africa stratified by black versus non-black race.

Fig 9. Correlation of survival rates in breast cancer patients and hormone receptor status.

Fig 10. Trends in 5-year survival rate in the United States compared to Africa stratified by race (white vs. black in the US, and black and non-black in Africa).

Bands are 95% confidence intervals.