Race and Mortality in Prostate Cancer: A Meta-analysis

INTRODUCTION: Prostate cancer has the highest incidence in American men and accounts for about 21% of newly diagnosed cancers (Siegel, Ma, Zou, & Jemal, 2016). African American (AA) men are more likely to be diagnosed with prostate cancer, are likely to be diagnosed at a younger age with higher prostate-specific antigen (PSA) levels and tend to have more aggressive tumors and higher recurrence rates than Caucasian men (Odedina et. al., 2009, Alanee, et.al., 2010, Kim et. al., 2010, Caso et. al., 2010). In spite of these documented disparities in prostate cancer, racial differences in mortality rates in prostate cancer gives mixed evidence. Cohen et. al. (2006) conducted a study on Medicare patients and concluded that AA men had lower disease-free survival when compared to Caucasian men. In contrast, some studies have shown that AA men have either better survival or no difference in survival when compared to Caucasian men (Shah et. al., 2011, Brawn et. al., 1993, Powell et. al., 1995). This mixed evidence regarding racial disparities and survival in this common cancer leads to difficulty in establishing treatment and follow up guidelines. The objective of this meta-analysis is to compare existing studies in literature to estimate a summary statistic (crude and adjusted HR) of mortality according to race in prostate cancer. It also aims to ascertain the consistency of the association between AA race and mortality in prostate cancer across different studies.

METHODS: An extensive literature search for articles published between 1995 and 2014 was conducted using the keywords “racial”, “disparities”, “mortality”, “death”, “prostate” and “African American” using the PUBMED search engine. Screening was done on 172 identified articles using a 2-step process, and resulted in eight articles included in the final analysis (Albain et. al., 2009, Antwi et. al., 2013, Howard et. al., 2000, Mahal et. al., 2014, Mirabeau-Beale et. al., 2013, Moses et. al., 2010, Schwartz et. al., 2009, Tewari et. al., 2009). The primary outcome was crude HR and the secondary outcome was adjusted HR. Meta-analysis was conducted on the summary statistics of these eight studies using the metagen command in R.

RESULTS: All eight articles were cross-sectional studies published between 2000 and 2014. Seven out of eight articles controlled for age, type of treatment, number of comorbidities, educational level and type of insurance, except Howard et al (2000) that only controlled for socio-economic status. All eight articles reported both crude and adjusted HR. Seven articles reported 95% confidence intervals and one reported standard error. Using meta-analysis, the overall crude HR was 1.41 (95% CI: 1.25 to 1.59) and the overall adjusted HR was 1.29 (95% CI: 1.10 to 1.50), suggesting an increased risk of mortality in African American men when compared to their Caucasian counterparts. Publication bias was assessed using the funnel plot. For crude HR, only one study showed more dispersion than the others, while for adjusted HR, three studies showed dispersion.

DISCUSSION: This study demonstrates that AA men have higher risk of mortality when compared to Caucasians. The higher hazards for African Americans are consistent for both crude and adjusted HR. Thus, in conclusion, AA men are at higher risk of mortality due to prostate cancer than Caucasian men, even after accounting for potential confounders. Further research should be conducted to stratify and summarize the hazards by stage, grade and age. This can help establish better management guidelines so that AA men receive more aggressive treatment for prostate cancer, which in turn, can combat the increased risk of mortality in these patients.

American Association for Cancer Research. vol 26, issue 2. February 2017 [Epub]

Rasmi G. Nair1, Praveen Ramakrishnan Geethakumari2, Andre A.A. Williams3 and Stephen J. Lepore1

1. Temple University, Philadelphia, Pennsylvania
2. Thomas Jefferson University Hospital, Philadelphia, Pennsylvania
3. Nemours Specialty Care, Jacksonville, Florida  

DOI: 10.1158/1538-7755.DISP16-B17 Published February 2017
email news signup