ATYPICAL RESPONDERS LANDSCAPE REVIEW ∙ OCTOBER, 2017 19 Genetic polymorphisms such as SNPs can also predict the development of side effects to therapy. Two SNPs have recently been shown to predict taxane-induced peripheral neuropathy. Patients of African descent have a higher risk for this neuropathy [33], and a second SNP is associated with Avastin® (bevacizumab)-induced hypertension [34]. More studies are needed to precisely predict a response to therapy by leveraging pharmacogenomics. Tissue analysis of genetic explanations for an atypical response should involve investigation of anomalies in both tumor tissue and non-tumor samples. The Role of the Host Environment Response to therapy is impacted not only by the biology of the tumor, but also by the biology of the environment in which the tumor is located (“microenvironment”). Tumor cells interact with surroundings that include vasculature, immune cells, extracellular matrix, stroma, hormones, various secreted growth factors, cytokines, and chemokines [35- 37]. These factors are dynamic and are likely to contribute to tumor behavior and response or resistance to therapy [37, 38]. Indeed, therapies such as Nexavar® (sorafenib), Sutent® (sunitinib), Gleevec® (imatinib), and Avastin® (bevacizumab) are aimed in part at modulating these tumor microenvironment factors [39]. The details of how the tumor microenvironment may specifically mediate an atypical response are not entirely clear, although efforts are currently underway to better understand how tumor surroundings affect progression of different cancer subtypes. In 2015, a research team found that the progression of different types of breast cancer in mice is influenced by the tissue—the tumor microenvironment—in which the tumor is embedded. In this study, matrix metalloproteinase 9 (MMP9) was deleted in two mouse models [38]. The absence of the MMP9 protein delayed tumor onset only in one mouse model (triple-negative/basal-like C3(1)–Simian virus 40 large T antigen), whereas it had no effect in a luminal mouse mammary tumor virus–Neu model. The reason for this inconsistency appeared to be mediated by insulin-like growth factor binding protein (IGFBP)-1. If IGFBP-1 was absent, MMP9 had no effect, but if IGFBP-1 was present, then MMP9 became active. This suggests that IGFBP-1 interacts with MMP9 to promote tumor formation. Thus, trials of MMP inhibitors may need to be focused on patients whose tumor microenvironment contains IGFBPs [38] (read more). In addition, a recent review described Response to therapy is impacted not only by the biology of the tumor, but also by the biology of the environment in which the tumor is located (“microenvironment”). Co-morbidities and the drugs that patients take for them may impact atypical responses and survival in cancer patients. Additionally, some co-morbidities preclude MBC patients from taking the very drugs that may be of most benefit to them.