Metastatic Breast Cancer in Kenya: Presentation, Pathologic Characteristics, and Patterns—Findings From a Tertiary Cancer Center
Purpose: The purpose of this research was to describe the sociodemographic and clinical characteristics of Kenyan women with metastatic breast cancer diagnosed and treated at Aga Khan University Hospital in Nairobi, Kenya from 2012 to 2018.
Patients and methods: We reviewed charts of Kenyan women with metastatic breast cancer and analyzed sociodemographic data, breast cancer risk factors, and tumor characteristics associated with stage at diagnosis, receptor status (ie, estrogen receptor, progesterone receptor, and human epidermal growth factor receptor 2 [HER2]), and site of metastasis using χ2, analysis of variance, two-sample t tests, and logistic regressions.
Results: A total of 125 cases with complete medical records were included in the analysis. Forty women (32%) had metastases at diagnosis. Of the others, those diagnosed in stage III developed metastases sooner than those diagnosed in stage II (P < .001). Fifty-eight percent of patients had metastases to bone, 14% to brain, 57% to lungs, and 50% to liver. Seventy-four percent of patients presented with more than one metastatic site. Metastases to bone were associated with greater age at diagnosis (P = .02) and higher parity (P = .04), and metastases to the brain were associated with early menopause (P = .04), lower parity (P = .04), and lack of breastfeeding (P = .01). Patients whose tumors were triple negative (estrogen receptor-negative, progesterone receptor-negative, and HER2 negative) were more likely to develop brain metastases (P = .01), and those whose tumors were HER2 positive were more likely to develop liver metastases (P = .04).
Conclusion: Although our data on patterns of metastases and pathologic subtypes are similar to those in published literature, some unique findings concerning hormonal risk factors of women with metastatic breast cancer and specific metastatic sites need additional exploration in larger patient populations.