Autoimmunity and treatment outcome in melanoma

Purpose of review Only a subset of melanoma patients with advanced disease seems to benefit from immunotherapy. Predictive markers identifying these patients are unfortunately not available. Whether immune-related side effects could serve as predictors for treatment response or just resemble unwanted side effects from immunotherapy will be outlined in this review. Recent findings Early studies suggested an association of immune-related side effects such as vitiligo and autoimmune thyroiditis with response in patients receiving IL-2 or IFNα. However, conflicting data have been reported as well, mentioning the effect of a higher rate of immune-related toxicities during prolonged administration of the drug in responders/survivors. This type of bias is also known as guarantee-time bias. Recently, a clearly significant and clinically relevant prolongation of survival was demonstrated in patients with metastatic melanoma treated with ipilimumab. Immune-related adverse events were associated with response to ipilimumab, however, at the cost of considerable toxicity. Summary Evidence for an association of immune-related toxicities and response in patients receiving IL-2 or IFNα is weak, considering guarantee-time bias. On the contrary, this association for patients receiving anti-cytotoxic T-lymphocyte antigen-4 therapy (ipilimumab) appears much stronger. Importantly, can we uncouple tumor immunity from autoimmunity in order to optimize immunotherapy in melanoma?