Date of Award


Document Type


Degree Name

Master of Medicine (MMed)


Pathology (East Africa)


Introduction: Breast cancer is a heterogeneous group of diseases classified into various subtypes according to clinical, histological and molecular features as well as by the expression of prognostic and predictive biomarkers. New biomarkers are being studied which may be therapeutic targets in various breast cancer subtypes particularly triple-negative breast cancer. Recent studies show that androgen receptor (AR) expression can be used as a prognostic marker and there is evidence that anti-androgen therapy may be beneficial in oestrogen receptor negative breast cancer. Retrospective studies show that patients with breast cancers co-expressing androgen receptor and oestrogen receptor have a favourable prognosis and improved survival. On the other hand, oestrogen receptor-negative breast cancers which express the androgen receptor in the presence of the human epidermal growth factor-2 (HER2) over-expression are associated with an unfavourable prognosis. In addition there may be regional or even racial differences in the expression of the AR in breast cancer. The objective of this study was, therefore, to determine AR expression in breast cancers from Kenya and secondarily, to determine the association of AR expression with prognostic and predictive clinico-pathologic factors.

Materials and Methods: The study was nested within a larger study based at AKUHN which is exploring the prevalence of triple-negative breast cancer in Kenya. Tissue blocks from 222 breast cancer specimens collected prospectively from September 2012 to March 2014 within the larger study were analysed for the presence of the androgen receptor. Tumour blocks were stained for AR on automated IHC platform. Nuclear staining of >10% was used as the cut-off for a positive AR stain. Clinical pathological features, including the ER/PR/HER2 status, were captured from data recorded in the larger study.

Results: Of the total 222 cases of breast cancer analyzed for the androgen receptor, 78.8% were found to be positive for AR expression (confidence interval 72.86 – 84.01). A third of the triple-negative breast cancer cases (33%) and 94% of HER2 subtype breast cancer cases were positive for androgen receptor expression. Androgen receptor expression was significantly associated with estrogen receptor expression (p = 0.000001) and Her2 subtype of breast cancer (p = 0.04). Androgen receptor expression followed the age distribution of the sample population and tended to be inversely related to tumour grade. However these findings were not statistically significant. No association was found between androgen receptor expression, tumour stage, lymph node status or specimen type.

Conclusions: The androgen receptor is expressed in 79% of breast cancers in this study. This is similar to the levels of AR expression in published literature. The association between AR expression and ER expression is expected and has frequently been reported in literature. The strong association between AR expression and the HER2 subtypes of breast cancer is stronger than that reported in literature and should be explored. Further studies are needed to investigate patients with ER negative, AR positive breast cancers in view of the poorer prognosis in a subset of these patients and their potential to benefit from anti-androgen treatment.

Included in

Pathology Commons