ATYPICAL RESPONDERS LANDSCAPE REVIEW ∙ OCTOBER, 2017 15 to novel combination therapies [13-15]. Personalized therapies are especially important in groups of patients, such as those with MBC, who have a short life expectancy (2-3 years) and a relatively inferior QOL. Most studies of atypical responses (Table 2 and Table 3) employ various types of genetic analyses, and the resulting molecular testing is expected to lead to identification of predictive biomarkers and more targeted therapies [8]. In many cases, molecular aberrations (mutations, insertions, deletions, copy number variations, rearrangements, multi-gene fusions, translocation, truncations, etc.) can predict a change in a patient such as an alteration in a molecular pathway or an increase in cellular receptors, and hence a potential therapeutic outcome. It is encouraging that rapid genetic analyses are becoming increasingly available and less costly. The study of atypical responses suggests that routine molecular testing can lead to the selection of patients for multiple types of trials. Examples include “basket trials” whereby researchers test the effect of a single drug on a specific mutation in a variety of cancer types, “umbrella trials,” which are designed to test the impact of different drugs on various mutations in a single type of cancer (one example is Lung-MAP, which is enrolling patients with advanced lung cancer; NCT02154490), and biomarker- driven trials [9]. Differently designed clinical trials may be required to optimize precision medicine [19]. One example is the I-SPY 2 trial in which breast cancer patients with Stage II, III, or “regional” stage IV, in which supraclavicular lymph nodes are the only sites of metastasis, are randomized to treatment based on response to therapy of patients with similar biomarker profiles who were previously enrolled in the same trial (NCT01042379). In addition, the investigation of atypical responses is anticipated to lead to improved selection of therapies for individual patients and avoidance of therapies that are unlikely to work, resulting in improved patient outcomes and QOL (Figure 1, p. 36). The initial successes in implementing such a model are illustrated by the published studies of atypical responses outlined in Table 3, p. 34. Published and anticipated results from these studies have identified mechanisms of drug resistance, the molecular pathways involved, and multifaceted approaches. Notably, these published studies mainly involved molecular testing of the tumor tissue without addressing the roles that normal (non- tumor) tissue, lifestyle, co-morbidities, and other factors play in an atypical response. Furthermore, the results of genetic testing may vary significantly depending on the company The investigation of atypical responses is anticipated to lead to improved selection of therapies for individual patients and avoidance of therapies that are unlikely to work. In patients who underwent genetic testing by both FoundationOne and Guardant 360 labs, only 25% agreement was found regarding their recommended drugs.