Dr. Melissa L. Johnson, director of lung cancer research at Sarah Cannon Research Institute, took the podium in an education session on bispecific T-cell engagers at the 2025 American Society of Clinical Oncology (ASCO) Annual Meeting this past June. Dr. Johnson discussed some of the obstacles that community practices face in implementing these novel agents, especially regarding the potentially serious associated side effects of cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS).
ILCN: What are the specific factors that clinicians should be aware of regarding the development of CRS and ICANS for patients who have received a bispecific T-cell engager?
Dr. Johnson: I think that the most important thing to recognize about all of the bispecific agents in development is that they will all cause CRS in some patients, but the CRS happens at different times. Its severity is compound dependent and, to an extent, the management will differ based on what medicines we need to use to tamp down the immune system.
I think it’s important for those caring for patients on bispecifics to understand that every agent in this class is going to be a little different, so our ability to decide whether each is able to be given as an outpatient therapy or given as an inpatient bridge first comes with some experience using the drug.
For tarlatamab in particular, the CRS tends to display as fever, as well as hypotension and hypoxemia. In more severe cases, ICANS tends to display as a little bit of confusion in a variety of ways. For the most part, the CRS and the ICANS can be managed with steroids, and, in some cases, tocilizumab will be necessary. The other thing that is important, irrespective of the bispecific that is given, is for there to be a plan B in place.
If the patient is going home or to the hotel across the street, or if they’re going to the hospital for a couple of days and then being discharged, they should know where and what should happen if they get home and have new or worsening symptoms. Once plan B is in place, then I think we all feel a little bit more confident to let patients go back and forth from the clinic to home rather than requiring that patients be observed as inpatients.
In Part F of the phase 1 DeLLphi-300 trial with tarlatamab, we found that the CRS occurs between 8 and 12 hours after dosing. If a patient could get to day 2, and certainly day 3, they were kind of out of the woods for it. It wasn’t something that was going to pop up that weekend, for example. Likewise, we learned that if you didn’t have CRS within the first or second administration of the drug in the first or second week, you were unlikely to have it.
This made it a lot easier to manage the outpatient cohort because the patients and caregivers were like, “Okay, I get this now. This is fine. I’m doing fine.” So it is just those first few doses where patients and caregivers should be mindful, and within 30 minutes of a hospital that has the capability to manage CRS as an inpatient.
ILCN: What are your recommendations for monitoring grades 1 and 2 CRS?
Dr. Johnson: A really important point is that grade 1 CRS is just fever, and if patients don’t have fever, it’s unlikely to be CRS. As you build other symptoms on top of the fever, like fever and hypotension, for which we would give IV fluids, or fever and hypoxia, for which we might give some supplemental oxygen, then the CRS becomes grade 2.
If you think about it, we manage fevers, low blood pressure, and sometimes low oxygen saturation in the clinic all the time. This leads to a higher level of comfort with giving tarlatamab in the outpatient clinic and then watching patients, with the understanding that they would come back to the clinic frequently for follow-up, because so much of the management can be there in the outpatient setting.
ILCN: What caregiver resources or information are essential for the early diagnosis and optimal management of these events?
Dr. Johnson: Well, I guess the most important thing to realize is that we’re using vital signs to provide early indications of whether a patient is developing CRS: temperature, blood pressure, pulse, and oxygen saturation. One of the practices in the Sarah Cannon network provides a bag that has a blood pressure cuff, a pulse oximeter, and a temperature probe so that everyone has those things at home to provide data. Temperature is easy; everyone knows how to take a temperature.
Fever would be the first sign that there is something wrong. Teaching caregivers about low blood pressure and low oxygen and how these might disrupt a patient’s ability to think, ability to reason, and whether they seem like themselves or not, is important.
In my ASCO talk, I mentioned remote patient monitoring, which is being used at our center and others across the Sarah Cannon network for patients after CAR T-cell therapy. These patients may experience CRS for a long duration, and it can develop a couple of weeks after treatment, making it hard to admit the patient to the hospital for a long enough time.
These patients will wear the blood pressure cuff and be monitored remotely by nursing staff. Increasingly, these programs are using artificial intelligence to note changes in trends like temperature and blood pressure to establish an understanding of what might be more routine and what could be an issue. This alerts the nurse that there’s something that’s different, and then the nurse might just call to check in. We learned in our study that the temperature was the thing that varied the most.
Variations were not necessarily because a patient was developing a fever, but because they had a sweatshirt on, or they were under a blanket or something that changed the temperature. So it was important to not just go by objective changes but to also have a conversation, “Hey, how are you feeling? What’s going on with you today?” Caregivers were educated that, when in doubt, call as a way to learn what to look out for.
Remote monitoring is typically done by third-party vendors. It’s going to look different depending on the clinic’s resources and the agreements within a particular hospital system, but it might be possible to train a nurse to perform this task. But I wonder if patients would feel as well cared for if this were an additional responsibility to seeing patients. I don’t know if that would be as effective.
ILCN: How can Part F of the DeLLphi-300 trial serve as a model for the use of tarlatamab in the outpatient setting?
Dr. Johnson: Every trial has a protocol, and we build those protocols into order sets in our electronic medical record, which then slides very nicely into using them now that it’s an approved drug. But patients did come back to the clinic more frequently. I think the one patient who was admitted to the hospital had a 48-hour hospitalization after the first three or four doses and only came into the clinic when it was time to dose again. Patients who were randomly assigned to outpatient monitoring went home on day 1 and came back on days 2, 3, 5, and 8, and then it was time to dose again.
A lot of the tests done during those visits would be the ones that you would think of as common sense—vital signs, laboratory values, checking for liver and kidney dysfunction. CRS can be kind of a vague clinical syndrome, so some people have headaches and others have arthralgias, which makes the vital signs important.
Blood draws are also important because these will show if there’s organ dysfunction, as well as show any inflammatory markers that we would test for in the hospital or in the clinic, such as C-reactive protein (CRP), IL-6, and ferritin. Although you can’t make decisions in real time based on laboratory values, you can trend them over time to see whether the patient’s recovery is going in the right direction.
Part of the order sets were to determine whether the patient needed fluids, steroids, or tocilizumab, as determined by their vital signs. Usually, patients in the outpatient setting have smoldering CRS—they’ve had a fever at home. That was often the first thing that the caregiver said the next day when they came into the clinic, “He had a fever, but I gave him Tylenol at 7 pm and it went away.”
In those cases, I would administer steroids because I was concerned. Typically, a one-time dose of dexamethasone was enough to clamp down the fever.
ILCN: What are some best practices for that intentional collaboration regarding the administration of bispecific agents for patients with lung cancer, either on an inpatient or outpatient basis?
Dr. Johnson: It does require communication. The clinic or the practice may not always be in lockstep with the healthcare system or the hospital. We’ve had to grow that communication over time. The details around the administration of tocilizumab are an example of what comes up with every clinical trial that we’ve ever done, regardless of whether for CAR T-cell or a bispecific therapy. Specifically, who pays for tocilizumab, because the clinical trial company usually provides the reimbursement for this.
We have had to build that into budgets because, at least for patients with solid tumors, it’s not a standard drug. Increasingly now, however, it is becoming standard in hematologic malignancies, so I think it could be billed to insurance now or soon.
Another thing to consider is whether you have tocilizumab in the emergency department (ED), the inpatient unit, and the outpatient unit. You have to have access in all three of those places if you’re going to give this with tarlatamab. I don’t think lack of access to tocilizumab is the barrier that it used to be, but having it guaranteed in the hospital probably will still be a barrier for some practices.
I think education is critical because CRS can look a lot like sepsis before you know it. If an ED staff isn’t familiar with CAR T-cell therapy or doesn’t know that patients will be coming in with hypotension and hypoxia, they will assume those patients need antibiotics, IV fluids, and vasopressors, and they might not give steroids. Because of this, we have provided education to our ED, nursing team, and the hospitalists at Sarah Cannon.
Again, this is something that will vary by practice, but at the very least, open communication should be encouraged so that the patient defaults to you and is guaranteed to at least talk to a provider. Another strategy is to provide patients with a card that outlines their therapy, potential issues, management strategies, and clinic information.
I have noticed that my talk at ASCO sparked a lot of opinions, one way or the other, depending on the current standards of care within a practice. What fits one may not fit another, and the possibility of CRS and ICANS is quite intimidating to some.
Tarlatamab is well tolerated, but there will be other bispecific agents for which the associated CRS is more clinically significant, so we may decide as a practice that it can’t be given as an outpatient therapy.
I don’t mean to say that one is better than the other. It’s just that we are conscious in the community of trying to get these very new and exciting therapies into the hands of the doctors who take care of the majority of lung cancer patients in the United States, so how we can do that in this day and age has sparked a lot of innovation.
