Francis A. Asamoah , MD * , † , Joel Yarney , MD * , Shivanshu Awasthi , PharmD, MPH ‡ , Verna Vanderpuye , MD * , Mary A. Dadzie , MD * , Angelina Fink , MPH ‡ , Arash O. Naghavi , MD, MS † , Afua Abrahams , MD § , James E. Mensah , MD ∥ , Evans Sasu , MPhil * , Samuel N. Tagoe , PhD * , Jasreman Dhillon , MD ¶ , Peter A.S. Johnstone , MD † , Kosj Yamoah , MD, PhD , † , ‡
03 October 2019
To evaluate early-stage prostate cancer (PCa) radiotherapy treatment patterns and outcomes among Ghanaian men (GM) compared with US men (USM).
This retrospective study consists of 987 National Comprehensive Cancer Network low risk, favorable intermediate risk, and unfavorable intermediate risk PCa patient subgroups; GM (173) and USM (814). Differences in baseline covariates and clinical characteristics between GM and USM were analyzed using χ 2 and Mann-Whitney test while Cox Proportional Hazards model was used to assess freedom from biochemical failure differences between the study groups.
Median follow-up for this study was 40 months. GM were diagnosed at a younger median age (64 vs. 68 y, P<0.001) with heavier unfavorable intermediate risk disease burden (32.4% vs. 19.2%) compared with USM. Significant differences were identified in median external beam radiotherapy dose (72.4 vs. 78 Gy, P<0.001); brachytherapy utilization (49.7% vs. 80.6%, P<0.001) and androgen deprivation therapy for intermediate risk disease (48.4% vs. 21.0%, P<0.001) between GM and USM, respectively. GM with low risk and favorable intermediate risk PCa were at increased risk of biochemical recurrence compared with USM with adjusted hazard ratio: 5.15 (1.27 to 20.7), P=0.02 and 4.64 (1.20 to 17.92), P=0.02, respectively.
Compared with USM, GM with low and favorable intermediate risk PCa may experience less durable disease control following standard treatment recommendations. Results suggest differences in radiation treatment and possible inherent differences between the 2 populations. This data will aid in developing research strategies to improve treatment outcomes in GM.