Why Late Systemic Therapy in Cancer Persists

Systemic therapy is often given late in cancer because of optimism, treatment momentum and the difficulty of having end-of-life conversations. Patients receiving late therapy are more likely to be hospitalized or admitted to the ICU, prompting The University of Texas MD Anderson Cancer Center in Houston, for example, to implement mortality reviews, approval checks and improve communication with patients and caregivers.

The discussion features Maureen Canavan, an epidemiologist at Yale Cancer Outcomes, Public Policy and Effectiveness Research Center, and Dr. Kerin Adelson, healthcare executive, chief quality and value officer, and professor of Breast Medical Oncology at MD Anderson Cancer Center.

CURE: What are some of the key factors that drive oncologists and patients to pursue systemic therapy so late in the disease course?

Adelson: I think the first thing to really make clear is that most oncologists don’t go into these decisions saying, “I’m going to give systemic therapy to a dying patient.” For the most part, they hope that treatment is going to work. One of the factors is that oncologists don’t always recognize that death is imminent or that treatment is unlikely to have benefit.

There are multiple reasons. Oncologists are optimistic. We call that optimism bias — the bias that we’re more likely to think treatment will have benefit and less likely to recognize when it won’t. There are also biases toward novel therapies, the idea that something new and exciting holds promise. And then there are other biases, like bandwagon bias, where if other people are doing something, we feel that we should too, and that doing more is always better than doing less.

And then I think there’s also the fact that it is much harder to say to a patient, “I don’t have anything left to offer that will help” than it is to say, “Your scan shows progression, let me give you X, Y and Z next.” The easier thing is to say, “It looks like you have progression, let’s switch your chemo to taxol.”

Canavan: I think there’s also something about the momentum effect — if you are actively treating patients, making the transition to what might be more prognostically appropriate end-of-life care is a hard decision. You stick with the momentum and stay on that path. Which brings us back to one of the big picture items we always talk about: the importance of goals-of-care conversations and open, honest and ongoing communication between the patient, the provider and the caregivers to make sure we’re aligning with what the patient wants and what’s prognostically appropriate.

The data show a strong association between end-of-life systemic therapy and higher rates of hospitalization and ICU stays. What does this tell us about how treatment decisions are being made and where communication gaps may lie?

Adelson: I think there’s an important point to make about our data. We looked at everything that happens to a patient in the last 30 days of life, and we saw that patients who stay on systemic therapy are more likely to have a medicalized death. They’re more likely to go to the ED, be admitted to the hospital, go to the ICU and even die in the hospital, and they’re less likely to use hospice services. But it does not tell us that systemic therapy definitively caused those other things to happen.

There are clues that it might. Patients who get systemic therapy usually get it in the outpatient setting, and then all of those other outcomes occur in the inpatient setting. It suggests that, in a large population, the treatment may come first while they’re outpatient, and then they end up in the hospital. But our analysis doesn’t prove that. What it shows is that patients on systemic therapy are more likely to also have these other outcomes, and because we didn’t look directly at timing, it reflects a more aggressive pattern of end-of-life care that includes systemic therapy as well as those acute inpatient events.

Canavan: Yeah, and just to piggyback on that, the big point Karen’s highlighting is that, although we didn’t demonstrate a cause-and-effect relationship, this constellation of more medical care and higher system utilization points to the idea that if you’re getting systemic therapy at the end of life, there’s often a belief that more treatment is better. You get the systemic therapy, and then if there are side effects or disease progression, you’re more likely to end up in the acute care system.

I think one communication gap is pausing to ask: Is more treatment also leading to higher rates of hospitalization and ICU care? That’s something we are hoping to tease out more. But this idea stringing back to communication. Providers should clearly convey to patients and caregivers that there may be limited benefit and also potential downsides such as toxicity or increased likelihood of emergency care.

You mentioned that MD Anderson has implemented structural changes to reduce chemo overuse. What do those interventions look like in practice, and how have they affected patient care?

Adelson: Some of this work is still underway, so I can talk about what we’ve done and what we’re doing. We took a two-pronged approach. One is cultural. There has been a long-standing bias that if patients come to MD Anderson looking for one more treatment or a trial, we are obligated to offer treatment — even when they are nearing the end of life or coming to us for the first time through the ER.

What we’ve done is start having more conversations with our oncologists and require review of all cases where patients received chemotherapy within 14 days of death. We call them mortality reviews. In those reviews, we ask: Was the treatment appropriate? Could we tell the patient was nearing the end of life? Did we have open prognostic conversations and present no further treatment as an option? Did we discuss hospice? All of those reviews are discussed at department-level meetings to raise awareness and hold each other accountable.

Then, from a structural standpoint, we are putting checks and balances in place. More than 80% of patients who get chemotherapy late in life receive it inpatient, and that’s easier to control. We now require approval prior to giving treatment in the hospital. We classify chemotherapy regimens by whether they are standard-of-care inpatient or outpatient. If a doctor wants to give an outpatient regimen inpatient, they have to justify it and get approval from a departmental decision maker.

We also developed clinical criteria that flag signs of progressive cancer or frailty, like weight loss, needing supplemental oxygen, spending most of the day in bed, or organ dysfunction. If patients meet criteria suggesting they are nearing the end of life, those cases require departmental approval before treatment can be given. It’s a way to put more oversight in place to reduce non-beneficial or potentially harmful systemic therapy.

Transcript has been edited for clarity and conciseness.

For more news on cancer updates, research and education, don’t forget to subscribe to CURE®’s newsletters here.