Transcript:Keith Flaherty, MD: I’ve been fascinated to watch, over the past five years, with the introduction of new effective systemic therapies, the way that radiation therapy and surgery are being reevaluated in terms of their role in patients with metastatic disease. There’s actually some new data to consider, even in surgical approaches for patients with earlier stage disease.

I wanted to focus a little bit on radiation therapy, pivoting a little bit from this injectable-lesion concept to radiation as an adjunct to systemic therapy. This has been particularly popular in terms of immune therapy. Jason, what’s your current take on emerging data and current clinical practice considerations in terms of weaving radiation therapy into a patient’s management?

Jason Luke, MD, FACP: I think it’s a two-fold, two-pronged approach. One is we have highly effective therapies now, but directed radiation to a lesion that’s symptomatic is also highly effective. So the question becomes: is that palliative approach useful and is there more beyond that? Because, as you’re alluding to, there’s pretty good preclinical data suggesting that you can actually get some distant effects with radiation with upregulation of other immune molecules like PD-L1.

Perhaps that has the potential to augment the activity of other immunotherapy. In my practice, I don’t necessarily go around looking for lesions that I want to radiate. But if it comes up that a patient has a symptomatic lesion, I certainly don’t hesitate to say I’m going to add radiation to that lesion, both palliatively, as well as hopefully, to sort of amplify an immune response that’s already ongoing.

I would say also in the context of targeted therapies, we’ve learned, as well, that patients can have disease control at sites and then lose that in only certain parts of the body. And certainly that’s where a palliative approach to get that lesion is also highly effective—and then you still retain the benefit of the targeted therapy.

Keith Flaherty, MD: Rene, there are trials ongoing investigating immune checkpoint antibody therapy with radiation to try to understand how reproducible an effect there might be in terms of systemic impact. Absent that data, because we don’t really have rigorous data, do you feel comfortable giving radiation therapy during the course of the checkpoint antibody treatment, be it CTLA4, PD-1 monotherapy, or even combination treatment?

Rene Gonzalez, MD: Yeah, I’ve done that before, and yes, I agree with everything Jason said. I think another thing that’s possibly contributing to drive this is just the greater availability of better technology for radiation, such as the stereotactic radiation that’s started in the brain. So, in the brain, I think it’s certainly a critical thing.

But now, that’s more widely applicable in the body. I think that kind of radiation might be beneficial. I was a little bit surprised today to see a poster—I think it was from Australia—at the SMR meeting. It was a retrospective look at the abscopal effect. My sense of it was that it’s not very common. I don’t remember the details of the poster, but they were suggesting as high as 50%. It seemed high.

Keith Flaherty, MD: Jeff, maybe again for this audience, it might be useful to kind of visit this issue of how is it or why is it that people would think that radiation and the type of effect it has on tumor cells could be potentiating to the immune system in a systemic way.

Jeffrey Weber, MD, PhD: There’s a long history to the so-called abscopal effect. I guess you could define it as a scenario where you deliver a local therapy that causes local destruction of tumor, and at some point, subsequently, there’s a systemic effect where you have regression of distant disease.

There was a 2012 article in the New England Journal of Medicine, where Michael Postow and Jedd Wolchok’s group published a very interesting case of a person who had obvious progression in multiple lesions on ipilimumab, in a clinical trial—I forget the dose, I think it was probably around 1 or 3 mgs/kg—who then had a growing painful pleural-based lesion that was posterior, so it was invading into the ribs.

After local radiation, it appeared in the timeline that the tumors, distantly in other locations, began to shrink in the absence of further ipilimumab. Although, if you look carefully at the timeline, ipilimumab was given very close to the time of the radiation subsequently. So, could this have been a late ipilimumab response? Probably not. I think this was a true case of an abscopal effect.

That is, in the absence of further systemic treatment, the ipilimumab, the active radiation therapy to one isolated lesion, caused an immune response, which induced regression distantly, which was then amplified because the investigators realized what was happening and said, “Oh, let’s continue the ipilimumab.” And I think we agree, looking at the scans and all the data from that article in the New England Journal of Medicine, there was progressive disease beforehand that occurred six months to a year after starting therapy.

I was once at a meeting of melanoma doctors, there must have been 100 doctors in the room, and I said, “Could people raise their hands if they’ve ever seen an example of this abscopal effect, where radiation in the face of no other systemic therapy induced a systemic response in someone who had had ipilimumab?” and I think I had one hand out of about 80 physicians describing one patient. I personally would be hard-pressed to describe, in any patient I’ve treated with ipilimumab, or nivolumab, or pembrolizumab, a true abscopal effect.

I think Rene is right. It’s a real, but relatively rare, phenomenon. I have seen bona fide data in breast cancer where ipilimumab has no real activity. I’ve been shown this by colleagues at NYU, where there was a PET scan indicating clear-cut progression of breast disease, there was radiation of a chest wall lesion, and regression of liver disease subsequently. It doesn’t happen with ipilimumab. So, I really buy into that. How often it occurs is unclear to me.

At one point, after the Postow article was published, everybody was going nuts deliberately, stereotactically treating every brain metastases in sight, then giving them ipilimumab.

Jason Luke, MD, FACP: And I guess I would second that to say that this is not a clinical like algorithm. If it comes up to radiate the patient, that’s palliative, and if you get such an effect, that’s great. But the idea that we should be doing this prospectively, I think that’s not how we should be thinking.

Jeffrey Weber, MD, PhD: And it has been looked at. I’m not sure if it was the best conceived study. There was a prostate study with ipilimumab and radiation, and it wasn’t obvious that there was any abscopal effect at all because the ipilimumab didn’t really add to the effect of the radiation. You figure if ipilimumab with radiation together were included, you would get some abscopal effect, which promote the activity of ipilimumab in prostate cancer, which you really didn’t see in that randomized study.

So, that was kind of a minus for the abscopal effect. And the other point worth making is that with ipilimumab in patients who have had stereotactic radiosurgery, it is my feeling, based on my discussions with my neurosurgical colleagues at my institution and elsewhere, that there is a late inflammatory effect in the brain causing radiation necrosis way beyond what you would expect with radiation necrosis, which typically is six to 12 months after stereotactic.

We’ve seen two to three years later in patients who had abnormal uptake on an MRI in the area of a previously radiated tumor. And they would develop serious symptoms with cerebral edema when they came to surgery because the surgeons assumed they had progressive disease which required debulking. No tumor. Radiation necrosis. Radiation necrosis three years later? I could see it happening once or twice, but we probably have assembled half a dozen cases at my institution.

And what they have in common is they’ve all had ipilimumab or ipilimumab/nivolumab that was preceded by stereotactic radiosurgery (SRS) or, in some cases, in the midst of treatments, they got one lesion SRS’d. So, I think it’s something that I have encouraged my neurosurgical colleagues to write up. Perhaps you all in your institutions have seen this, too, but it should generate a little bit of caution in interpreting an MRI scan in a patient who’s had stereotactic radiosurgery who’s also had ipilimumab or ipilimumab with nivolumab.

Transcript Edited for Clarity