Immune signatures predicting responses to immunomodulatory antibody therapy

Since the first immunomodulatory antibody was licensed by the FDA in 2011 for treating melanoma it has remained the case that only a certain proportion of cancer patients respond favourably to a particular therapy. Recent results from combining two or more different antibodies each targeting a different immune checkpoint indicate that the proportion of responding patients can be increased, but thus far there are no such therapies routinely yielding clinical benefit in 100% of patients in any cancer type. Therefore, predicting which patients will respond to a particular therapy remains of the utmost importance in order to maximise treatment efficacy and minimise side-effects and costs. Moreover, determining biomarkers predicting responses may provide insight into the mechanisms responsible for success or failure of that therapy. This article reviews seminal papers mostly from the past two years of progress in this area of intense investigation, and mostly in melanoma, the tumour type for which the largest body of data exists thus far.