Comprehensive discussion on the presence and management of interstitial lung disease in patients being treated for HER2+ metastatic breast cancer.
Sara A. Hurvitz, MD: The downside to this drug [trastuzumab deruxtecan] is that it is quite potent, however it does also have chemotherapy-like adverse effects. Patients do have more neutropenia, there is more in the way of GI [gastrointestinal] toxicity, and visible hair loss, although I have not seen complete hair loss. In the phase 2 single-arm trial, they had around 2.7% of patients have grade 5 ILD [interstitial lung disease], so deaths due to pneumonitis or interstitial lung disease, which was really concerning. We are all eagerly awaiting the larger phase 3 data to see how the safety of this molecule would end up looking in a large comparative trial like this. I think we’re all relieved to see that in this study [DESTINY-Breast03], there were no grade 5 events, or even grade 4 events, and only 10.5% of patients had any level of ILD.
The monitoring on this study and the management of ILD was stringent. Patients with even asymptomatic ground glass opacities on a scan had to hold T-DXd [trastuzumab deruxtecan], and grade 2 or greater ILD mandated discontinuation of T-DXd. This brings me to ask you each how you’re monitoring for ILD in your patients where you’re using T-DXd. How are you managing this? Bill, can you jump in here and let me know how you’re monitoring in the metastatic setting for patients on this drug?
William J. Gradishar, MD: At first blush when we first heard these data, the exceptional activity was somewhat attenuated when we saw the toxicity, we were worried about it. As more people got experience with this drug, the general impression is that the frequency of these high-grade toxicities is much less than the initial concerns were, as you already pointed out. The most important issue is to be aware of it; you must be sensitive to any respiratory symptoms a patient has. You must probe them a bit. As you pointed out when you were describing this, a lot of patients got radiation therapy at some point in the past that affects their lung. They may have ground glass appearance of one side from a prior radiation port perhaps. Not everything that shows up in the lung automatically means it’s ILD, but I do think you have to have some suspicion for that. The greater the respiratory symptoms or changes that the patient has, particularly if they’re not unilateral, they’re more diffuse, would raise my concern about ILD being present.
The other thing to keep in mind so we don’t get all frantic about this is, if you do a survey of a lot of the drugs we use, whether it’s everolimus or CD4/6 inhibitors, there is a small fraction of patients who experience ILD. This is somewhat greater, but it’s not out of bounds from other things if we were as vigilant looking for it. There have been efforts to try to develop different ways of evaluating this or highlighting certain characteristics of a patient who might be more prone to get ILD from T-DXd. But at the present we don’t have enough information to say with certainty who it is that’s destined to develop this. We must rely on clinical suspicion and follow-up, and in severe cases of course hold the medication.
Sara A. Hurvitz, MD: Thank you, that was a very good review. Sara, are there patients in whom you won’t use T-DXd based on concern? One of the things I’ve heard clinicians ask about is, is a heavy disease burden in the lung with lung metastases a contraindication to the use of this agent? Can you talk about risk factors and who you’ll avoid this drug in, kind of where Bill left off?
Sara M. Tolaney, MD, MPH: At this point, the only thing we know of it that is contraindicated is having pneumonitis. That is a contraindication, and technically it was also excluded if you have a history of noninfectious pneumonitis. In truth, we don’t give that many drugs prior to second-line T-DXd that would have induced pneumonitis, and certainly as Bill alluded to, there are lots of drugs that can cause it. We see it in the hormone receptor-positive setting with CDK4/6, everolimus, and in triple-negative disease with immunotherapy. Sometimes with taxane therapy in the first line you can get drug-induced pneumonitis, but it’s not so common, so most of my patients are eligible. To your point of what are other contraindications, if someone comes in who has asthma, has COPD [chronic obstructive pulmonary disease], has other lung issues, a ton of lung metastases, and maybe they’re starting to get a bit short of breath from them; is that a contraindication? Technically no, that wasn’t contraindicated in the trial.
This is where we must learn more though because I do think we need to understand better predictors of developing ILD that isn’t so clear. They have done some work where they grouped a bunch of the phase 1 data and tried to look at predictors. What we saw was breast cancer seemed to be associated with the higher rates relative to other diseases, which is interesting. Dose is clearly associated with higher rates of ILD. It did seem that people of Japanese ancestry tend to have higher rates, and being more pretreated is associated with higher rates. Maybe that’s also in part why DB01 [DESTINY-Breast01 trial] had a higher rate of ILD relative to DB03 [DESTINY-Breast03], and as Bill pointed out, we’ve learned a bit more; we’ve gotten used to the drug. We can see the rates of ILD are coming down as we also get a little smarter about managing things. Again, we have a lot to learn. I’d be curious to see if we can better understand PFTs [pulmonary function tests] and if there are predictors there. The truth is, we just don’t know, and so for now I’d say the only thing that’s truly contraindicated is current pneumonitis or a history of drug-induced pneumonitis.
William J. Gradishar, MD: I wanted to make one quick comment to pick up on something that Sara [Tolaney] mentioned. These patients who have bulky disease, although we may have some trepidation and want to exercise caution, these are the exact patients who need a very active drug. You’re between a rock and a hard place, and I think in those patients, if they don’t get a response the outlook is not good for them, period.
Sara A. Hurvitz, MD: Absolutely. Both of you made the point that the aggressive monitoring and management over time as we’ve gotten better at detecting this and managing it, has led to lower rates and lower rates of death with this drug. In fact, there was an AACR [American Association for Cancer Research] poster last year that showed that after implementation of monitoring and toxicity management in 2019, across trials you saw a drop in the incidence rate and severity of ILD. To your point about patients who are Asian or Japanese having higher rates, we did look at that in the DESTINY-Breast03 trial as a subgroup. In that large phase 3 study, the rates of grade 3 ILD and the overall rates of ILD were identical in patients from the Asian subgroup. I think we’re going to continue to learn.
One last question on the ILD topic before we move on to other therapies beyond the second-line setting; how frequently are you scanning your patients? In the DESTINY trials, metastatic trials, they are following patients every 6 weeks. Are you doing this in your practice to keep an eye on ILD because it can occur later, even a year into therapy, or are you stretching it out a bit? Bill, what do you say?
William J. Gradishar, MD: Yes, we stretch it out. We’re not getting scans any more frequently than every 3 cycles, every 3 to 4 months.
Sara A. Hurvitz, MD: Sara?
Sara M. Tolaney, MD, MPH: Maybe we’re going to sound neurotic, but we have been following the DESTINY guidelines, and we have been scanning people a lot. We have been scanning them every 6 to 9 weeks. The truth is we don’t know if we should be doing this. The reason it makes me feel better to do it is that maybe I’ll catch that grade 1 ILD before it becomes a grade 3 or 4 ILD. Maybe it allows us to get it early, but we don’t know that that’s true, picking up these asymptomatic radiographic ILDs. Maybe that’s not so hot because you’re also having to pull drugs in those situations too and waiting for resolution. The truth is we don’t know, but I’ll be honest, our group has been scanning every 6 to 9 weeks. If people have been on it for a while, we do start to taper that a bit, but to your point, it is interesting to see the curve, that the ILD did happen up to 12 months, and then you see it plateau beyond that.
Sara A. Hurvitz, MD: A last question for you, Sara, because I think your group is looking at doing DLCOs [diffusing capacity for carbon monoxide tests], or doing PFTs in the beginning, or following those. Are there any updates on that, or is that something you’re doing in your practice?
Sara M. Tolaney, MD, MPH: We have not been. I know the Memorial [Sloan Kettering Cancer Center] group has been instituting doing routine PFTs, and there’s a trial that has been done at Duke [University] that’s been getting routine PFTs. It’s interesting, but our group hasn’t done that yet, and we’re all waiting to see if doing that will be helpful to predict who’s going to develop ILD. One thing I want to know is, who can be rechallenged? It’s hard to stop this drug in someone with grade 2 ILD. Everybody has such a tremendous response, and to tell someone to discontinue is just heartbreaking, so it would be nice to understand who could be rechallenged safely.
Sara A. Hurvitz, MD: I think that, and understanding better the pathophysiology, working with our pulmonology colleagues, and getting tissue; a lot of questions yet unanswered.
Transcript edited for clarity.