The Role of Kaiso as a predictive breast cancer biomarker in Africa and across the African Diaspora

Around the world, women of African ancestry and throughout the African Diaspora, continue to perish from breast
cancer at significantly higher rates than any other racial/ethnic group. Although much of these disparate
outcomes have been attributed to inequities in access to healthcare, socioeconomic disadvantage, and other
crippling exposures linked to institutional racism, most studies also indicate a prominent role in tumor biology.
Causal factors include i) higher frequencies of aggressive breast cancer subtypes associated with genetic
ancestry; ii) more aggressive biological behavior of low-risk breast cancers based-on race/ethnicity and ancestry,
and iii) decreased predictive value of prognostic gene signatures when applied to women of African heritage.
Recently we have shown the subcellular localization of the breast cancer biomarker Kaiso (ZBTB33) in both the
nucleus and cytoplasm of breast cancer cells were each more associated with biologically aggressive cancers
and independent predictors of poor breast cancer survival and response to neoadjuvant therapy, revealing that
cytoplasmic Kaiso was highly correlated with an “immune-evasive” or “immune-suppressed” tumor
microenvironment. Mechanistically cytoplasmic Kaiso was shown to be very highly correlated with the machinery
involved in extracellular vesicle (EV) loading and secretion, a process known to have a dramatic influence on
the micro-environment of the tumor bed and metastatic niche. In this study, we will validate and extend these
observations of the subcellular localization of Kaiso as an independent predictive biomarker of breast cancer
survival and the tumor micro-environment using two diverse breast cancer cohorts from Africa and across the
African diaspora. One from the Columbia University Herbert Irving Comprehensive Cancer Center (N=487) and
the other from the Aga Khan University Hospital, Nairobi, Kenya (N=237) (Specific Aim 1). Using a multi-omic
approach that will integrate the analysis of tissue microarrays (TMAs), quantitative multiplex
immunofluorescence (qMIF), gene expression (RNA-seq), tumor mutational signatures and frequency (whole-
exome sequencing), deep annotation of patient clinicopathological features, environmental exposures, and
genetic ancestry, we will define prognostic models that will demonstrate the role of Kaiso in predicting breast
cancer survival, response to therapy and the role of genetic ancestry in breast cancer evolution and outcome
(Specific Aim 1). We will also define how Kaiso modulates the influence of EV secretion on breast cancer growth,
invasion, metastasis, and immune evasion in murine xenograft models using cells engineered for depleted or
enhanced Kaiso expression (Specific Aim 2). These goals will be accomplished through a transatlantic
collaboration of a highly-specialized multi-disciplinary team of breast cancer pathologists, tumor biologists,
molecular biologists, biostatisticians, bioinformaticians, epidemiologists, genome researchers, and cancer health
disparities investigators.