Why Do Physicians Overtreat Elderly Breast Cancer Patients?

Kathy D. Miller, MD; Reshma Jagsi, MD, DPhil


February 01, 2018

Kathy D. Miller, MD: I'm Kathy Miller, professor of medicine at the Indiana University School of Medicine in Indianapolis. Welcome to Medscape Oncology Insights, coming to you from the 2017 annual meeting of the American Society of Clinical Oncology (ASCO).

As oncologists, we all want to give our patients the best treatment available. However, with some patients, especially our older patients, we often give them more treatment than they really need.

Here to talk about the issue of overtreating elderly breast cancer patients is Dr Reshma Jagsi, a professor and deputy chair of radiation oncology at the University of Michigan in Ann Arbor. She's also contributed to the National Comprehensive Cancer Network's guidelines on older adult oncology.[1] Welcome, Reshma.

Reshma Jagsi, MD, DPhil: Thanks for having me, Kathy.

Relearning CALGB 9343 and PRIME II Lessons

Dr Miller: We use radiation in several different ways in breast cancer—certainly after breast-conserving surgery, more commonly after mastectomy, and in the palliative setting of metastatic disease. Where do you see overtreatment having the biggest issue?

Dr Jagsi: In terms of overtreatment, the biggest issue really lies in the adjuvant setting after breast-conserving surgery. We now have high-quality evidence from not only CALGB 9343[2,3] but also early results from the PRIME II randomized controlled trial,[4] both of which have demonstrated that in select women who are older and have biologically favorable disease, we may be able to get away with less treatment if this is something that patients would prefer.

Dr Miller: I have to ask you about the term "select patients." How do we select them?

Dr Jagsi: The women in CALGB 9343 were 70 years of age or older, and in PRIME II they were 65 or older. These results shouldn't be generalized to patients much younger than that, but they're patients with early-stage disease, stage I disease, tumors that are ≤ 2 cm that haven't spread to the lymph nodes, at least based on clinical examination.

An important point that often gets missed is that people don't realize that more than 60% of the patients on the CALGB trial actually had only clinical stage in their axilla. We assume that patients must have pathologic staging if they are to be eligible to omit radiation therapy, and that's not the case. Those trials included patients who had estrogen receptor-sensitive disease and who intended to and were willing to receive endocrine therapy.

Dr Miller: That's still a really large group of women, perhaps one of the most common types of new breast cancer patients that we see.

It's been over a decade since we've had the CALGB data. What impact has that study had on practice?

Dr Jagsi: It's been a decade since we've had the early 5-year data[2] from the CALGB trial. These showed that the women who had radiation therapy after breast-conserving surgery and went on to receive tamoxifen had a risk of only 1% for local, regional breast tumor recurrence at 5 years. The risk was 4% in the women who didn't have radiation therapy. It was a difference, but a small one, and the 4% was considered by many to be acceptable [at that time].

We didn't have the 10-year CALGB data[3] until somewhat more recently. But you're right, it has been years now since we've had the 10-year data. Unfortunately, practice pattern studies have not shown much of a decrease in the utilization of radiation therapy in those patients.

Dr Miller: Just to be clear for our listeners, what did the CALGB trial's 10-year data look like, because that is an issue for local recurrence, right?

Dr Jagsi: Absolutely. There was a significant difference between receiving radiation therapy or not, but the absolute risks were low enough that this presents a reasonable option for women to consider whether or not to have radiation treatment.

The risk was 10% without radiation therapy, which was reduced to 2% with radiation therapy. So it certainly remains quite reasonable and important to offer these patients radiation therapy as an option.

Why Overtreatment Continues Among Providers

Dr Jagsi: What's wonderful about these data is that we can now allow women to make individualized preference- and value-concordant decisions; the decision that's right for them, which is what we all strive to do for our patients.

Dr Miller: How do we know what's right for them? More importantly, how do we know what the discussions and the decisions are? We can get data about who does or does not get radiation. We can see if the proportion of women in those groups getting radiation changes. How do we know when we've gotten it right, that we're at the right proportion?

Dr Jagsi: It's a real challenge. We absolutely can't just look at the rate of utilization alone. The right rate of utilization shouldn't be 100%, and it shouldn't be 0% but somewhere in between. In order to understand if we've gotten it right, we really do need to know what's going on in the exam room.

I had the great privilege of working with the Cancer Surveillance and Outcomes Research Team (CanSORT), which takes population-based samples of patients diagnosed with early-stage breast cancer and asks them what they were told, how they decided. They also survey the physicians who've been caring for these patients and ask them their impression of the data, what they offer the patients.

The insights that have emerged from the type of research we've done in older adult patients in these settings have been that many women are not offered the option of omitting radiation therapy. Many surgeons and radiation oncologists tell us that they believe it's unreasonable to offer [this].

We've done a very good job as a radiation oncology community in educating everyone about the importance of radiation therapy as an adjuvant to breast-conserving surgery in general. But with these newer data, we need to pull that back just a little bit, while not robbing the many women who definitely benefit from radiation therapy [of it]. We need to make sure that those who can reasonably consider omission are actually given that option.

Dr Miller: What do we know about the surgeons and the radiation oncologists who thought it would be wrong to not offer radiation? Are they aware of the data? Is this just an issue of their believing a 10% [risk without radiation therapy] is too high, or is this just an issue of not being aware of the study data?

Dr Jagsi: It's a little bit of both. There is a wonderful presentation at this meeting by a faculty colleague of mine, Dean Shumway, which reports the results of a very large national survey that investigated exactly that question.[5] It demonstrates that about 40% of the surgeons and 20% of the radiation oncologists think that it's unreasonable. In fact, approximately 30% of the surgeons think that radiation therapy actually affords a survival benefit, which was not demonstrated in the CALGB study, and 10% of the radiation oncologists think that there is a survival benefit. Many of these individuals do overestimate the benefit of radiation therapy in terms of reducing local regional recurrence risks; and, in addition, many of these providers were unable to accurately estimate life expectancy. We found that those who were less able to estimate life expectancy and less aware of the results of the randomized trials were less likely to view omission of radiation as a reasonable option. It's a little bit of both things.

Challenging Patient Interactions

Dr Miller: What did you hear from the patients? I assume that many of those women had radiation. You're now asking them about those decisions, and perhaps many are now just hearing for the first time as part of that study that they perhaps didn't need it, or at least maybe it should have been a longer discussion.

Dr Jagsi: Because our sample includes not just older patients, but younger patients, we only asked them: If they didn't receive radiation, why not? We asked them what they were offered. Short courses of radiation? Omission of radiation? We hopefully didn't put them in the position of second-guessing their decision making.

But I think that it is a very challenging discussion. It is easy to walk into a room and say to a patient, "I have something that is going to help you. It's going to be minimal in terms of morbidity. We can get it done relatively efficiently; and, gosh, let's get going."

It is much more challenging to do the complex risk communication that is involved in sharing what a 10% versus a 2% local recurrence risk really means, right? What does local recurrence mean? Should we frame it as 10% and 2%, or should we frame it as 98% and 90%?

There's some wonderful work going on at my institution where decision scientists are looking at the best ways to frame this information in terms of pictographs where you see shaded numbers of bodies that actually give you a sense of the difference between treatment A and B.

Framing the Discussion

Dr Miller: Even thinking about how you frame it, though, becomes a bit of a minefield. You may get quite different proportions of patients making different choices based on how you frame it and how you present exactly the same numbers. How are you going to know which one is better?

Dr Jagsi: We need to frame things multiple ways because we know that framing matters. The best we can do to try to fully inform our patients is to give them both valences and to try to fight against this instinct that both doctors and patients have towards overtreatment. We have anticipatory regret. We worry about what we will later be sorry we didn't do. We are risk-averse by nature.

We need to recognize those issues. We need to fully inform our patients, give them time to deliberate, and remove that sense of pressure. We also need to really contribute to the quality of the decision making by providing tools to patients that remove some of that heterogeneity in what they hear by using standardized language and information presentation. Again, this should be framed one way and another, repeated so that they can refer back to and reflect upon it, and then really deliberate before coming to the right decision [for them].

Dr Miller: I'm thinking about a busy radiation oncologist in the community. You may have just described a 2-hour initial consultation and at least another hour for a second visit to go through all of this information in detail with multiple ways and to get a sense of this patient's values and where she might fall on that spectrum. That just doesn't seem possible.

Dr Jagsi: I agree with you. That doesn't seem realistic. That's where the tremendous power of decision tools [comes into play].

A colleague of mine, Sarah Hawley, presented at this meeting results from a randomized trial of an interactive decision tool.[6] The study randomized patients with breast cancer to receive either static information from high-quality sources like ASCO or an interactive tool that actually engaged them to share what their preferences were and then fed back tailored content. [The results] are quite exciting for showing what we can do with these types of decision tools.

I think that developing content specific to decisions like this will be the important next step in that research.

Dr Miller: You've been looking at overtreatment in this one situation, but there are a lot of treatment decisions that we make where the differences between very different therapies are quite small, yet differences in toxicity and cost may be quite substantial.

It seems like this sort of research could apply to a lot of the more complex decisions that we make in other settings that we commonly deal with, probably with some of the same problems.

Dr Jagsi: Absolutely. I completely agree.

Dr Miller: I think we'll be hearing a lot more about decision science and how we can use the technology to help us with some of these discussions. Thank you, Reshma.

To you, our listeners, thank you for joining us today. This is Dr Kathy Miller reporting from ASCO 2017.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.