CK8/18 and Claudin Profile of Female Breast Cancers from Botswana.

BACKGROUND: Globally, breast cancer mortality is rising particularly in transitioning countries where it is exacerbated by late diagnosis and poor diagnostic infrastructure. The dearth of data on the biology of breast tumours from under-studied populations compromise management of breast cancer patients resulting in poor patient outcomes. We conducted a study to determine the prevalence and impact of Claudin-low and CK8/18 in the Botswana female breast cancer cohort. DESIGN: This laboratory experimentally retrospective cross-sectional study was carried out on archived formalin-fixed paraffin embedded (FFPE) tissue obtained from mastectomy specimens. OBJECTIVES: To determine the protein expression of CK8/18 and Claudin-3, Claudin-4 and Claudin-7 on breast mastectomies from Botswana female breast cancer cohort. METHODS: CK8/18, Claudin 3, Claudin 4 and Claudin 7 were Immunohistochemical determined on 125 previously molecular classified Botswana female mastectomies. Statistical software STATA and SPSS were used to determine the relationship between CK8/18 & Claudin expression, clinicopathological variables, breast cancer molecular subtypes, CK5/6 and EGFR. RESULTS: The prevalence of Claudin-low tumours in our cohort was 22% (27/125) had a significant relationship with histological subtype, p=0.002. Claudin 3 and Claudin 4 were highly expressed in 54% (67/125) and 53% (66/125) of breast tumours respectively. Claudin 3 and claudin 4 significantly correlated with molecular breast cancer subtypes, p= 0.001. Claudin 7 was expressed in 47% (59/125) of all breast tumours and had a significant relationship with CK5/6, p=0.029. CK8/18 was highly expressed by 64% (80/125) of mastectomies and correlates with; Tumour grade (p= 0.005) and breast cancer molecular subtypes (p=0.002). CONCLUSION: Diagnostic stratification of breast tumours should include Claudin-low tumours as they respond differently to therapy when compared to the intrinsic breast cancer molecular subtypes. There is also a significant CK8/18 signaling that warrants deployment of anti-CK8/18 antagonists where ER is poorly expressed.