Reply to A. Indini et al.

Indini et al state that immune-related adverse events (irAEs) are associated with improved response to immunotherapy. Although some small early trials with ipilimumab suggested that patients with irAEs were more likely to respond, this observation has not been confirmed in later, larger series with ipilimumab or nivolumab. Therefore, the impact of this association is not clear. Regarding this, Indini et al ask whether patients who experience severe irAEs had a different prognosis. As we discuss in our article, the Common Terminology Criteria for Adverse Events grading system does not adequately identify clinically significant ipilimumab immune-related toxicities. As such, we defined severe irAEs on the basis of whether the treating physician believed the patient required systemic corticosteroids. In this case, our Figures 3B and 3D address their question and show that patients experiencing irAEs severe enough to require systemic corticosteroids had similar time-to-treatment failure and overall survival compared with patients who did not require corticosteroids. They ask about the dose and timing of the systemic corticosteroids our patients received to treat irAEs. In most cases, patients received 0.5 to 1 mg/kg/d of intravenous methylprednisolone or the equivalent dose of oral prednisone. The proportion of patients who received systemic corticosteroids after each dose of ipilimumab is shown in Figure 1. Finally, Indini et al inquire if patients achieved a response before starting corticosteroids. As we explain in our article, we did not assess patients for tumor response. Even so, the majority of irAEs occurred before week 12, when a formal response would have been assessed. We decided not to assess tumor responses on the basis of the low response rate to ipilimumab and the general understanding that clinical benefit to ipilimumab does not necessarily require a response as measured by RECIST. It is worth noting that neither theWHO criteria nor RECIST were meant to play a role in making decisions about individual patient care and, in fact, have not generally correlated well with overall survival. In the penultimate sentence of their letter, Indini et al suggest that immunotherapy “should be continued even if patients experience irAEs.” Although in selected cases, immunotherapy might be continued after resolution of low-grade or asymptomatic irAEs, in most cases, we would not give additional doses of ipilimumab to patients who had required systemic immunosuppression for irAEs.