XPERT(®) breast cancer STRAT4 as an alternative method of identifying breast cancer phenotype in Cape Verde (preliminary results)

INTRODUCTION: Breast cancer (BC) is a public health problem in developing countries, including Cape Verde. Immunohistochemistry (IHC) is the gold standard technique used for BC phenotypic characterisation to support efficient therapeutic decisions. However, IHC is a demanding technique that requires knowledge, trained technicians, expensive antibodies and reagents, controls, and results validation. The low number of cases in Cape Verde increases the risk of expiring the validity of the antibodies, and manual procedures often jeopardise the quality of the results. Thus, IHC is limited in Cape Verde, and an alternative technically easy solution is needed. A point-of-care messenger RNA (mRNA) STRAT4 BC assay to assess estrogen (ER), progesterone (PR), hormone growth factor 2 receptor (HER2), and Ki67, using the GeneXpert platform, has been recently validated on tissues from internationally accredited laboratories, showing excellent concordance with IHC results. To assess whether this technology can be implemented in Cape Verde to guide BC treatment we decided to study the level of agreement between the findings yielded by BC STRAT4 and the results are the same cases obtained by IHC. METHODS: Formalin-fixed and paraffin-embedded (FFPE) tissue samples from 29 Cabo Verdean BC patients diagnosed in Agostinho Neto University Hospital were analysed by applying IHC and BC STRAT4 assay. The time between sample collection and pre-analytic procedures is unknown. All the samples were pre-processed in Cabo Verde (fixed in formalin and embedded in paraffin). IHC studies were performed in referenced laboratories in Portugal. STRAT4 and IHC result concordance was assessed by calculating the percentage of results agreement and Cohen’s Kappa (K) statistics. RESULTS: STRAT4 assay failed in 2 out of the 29 analysed samples. Of the 27 successfully analysed samples, STRAT4/IHC results for ER, PR, HER2, and Ki67 were concordant in 25, 24, 25, and 18 cases, respectively. Ki67 was indeterminate in three cases, and PR was indeterminate once. The percentage of agreement between STRAT4 and IHC results for ER, PR, HER2, and Ki67 was 92.59%, 92.31%, 92.59% and 81.82%, respectively. The Cohen’s K statistic coefficients for each biomarker were 0.809, 0.845, 0.757 and 0.506, respectively. CONCLUSIONS: According to our preliminary results, a point-of-care mRNA STRAT4 BC assay may be an alternative in laboratories unable to provide quality and/or cost-efficient IHC services. However, more data and improvement on sample pre-analytic processes are required to implement this BC STRAT4 Assay in Cape Verde.