Transcript:Michael B. Atkins, MD: Patients getting anti-PD-1s—to a lesser extent, anti-CTLA4—and certainly with the combination can experience pneumonitis. It happens in melanoma patients. Perhaps 1% to 2% of patients may actually have symptomatic lung toxicity, which presents with cough or shortness of breath, but more often presents with infiltrates on a chest X-ray or chest CT scans that are routinely done. There are various different patterns of abnormalities that look inflammatory. One can see, on a chest CT scan, either a honeycombed appearance or sometimes even a low bar pneumonia-type appearance. And when we see patients with that type of appearance on a scan, we’re very cautious about giving additional anti-PD-1 therapy. We ask those patients to report immediately if they’re experiencing any shortness of breath, and we tend to scan them more frequently to see if there’s progression. And we would halt therapy at the sign of any symptoms or if that inflammatory reaction is progressing.

In the rare patient who comes in with shortness of breath, particularly if they’re getting nivolumab or a combination of nivolumab plus ipilimumab, we treat that as a real emergency. We admit those patients to the hospital. We give them steroids at high doses intravenously immediately, and we’ve talked ahead of time to our pulmonary doctors about this being a very serious condition. We ask them to come by and see the patients immediately and do a bronchoscopy to evaluate whether there’s some other cause for their shortness of breath that would be treated in a different way. We find that approach has been very successful in the few patients who have had pneumonitis. We’ve been able to control it without any residual effects. But early on in the phase I studies of the anti-PD-1s, we weren’t so quick to jump on pulmonary toxicity, and there were some patients who progressed, and it actually led to the deaths of 3 patients on those phase I trials. We’re very cautious about that now, and we’ve not seen anybody who has had a long-term residual pulmonary toxicity as a consequence.

In someone who has symptomatic pneumonitis from an anti—PD-1-based therapy, we never restart the treatment. However, many of those patients don’t need additional therapy. One of the things that I think we’re finding as we follow more patients is that for many patients, even if there appears to be residual disease on their scans and even if there appears to be residual disease that lights up on the PET CT scan, there’s no actual viable tumor in those specimens. And, therefore, for many of those patients, we’re probably overtreating and exposing them to the toxicity of additional therapy, the inconvenience of coming in for regular treatment, and the healthcare system to the financial toxicity of giving treatment that isn’t needed. And so, we’re looking very hard at the question of when to stop treatment. We’re finding in many patients, particularly those who’ve experienced toxicity from the treatment, that we’ve activated their immune system as well as we can against their tumor. And if it’s going to eliminate their tumor, it’s already happened and there’s no need to resume or continue therapy in those patients.

Transcript Edited for Clarity