Long-term results from the NRG/Radiation Therapy Oncology Group (RTOG) 9804 trial continue to show that radiation significantly reduces ipsilateral breast recurrence (IBR) following lumpectomy in women with good-risk ductal carcinoma in situ (DCIS).
Beryl McCormick, MD, chief of the External Beam Radiotherapy Service of Memorial Sloan Kettering Cancer Center in New York City, and colleagues analyzed data on 636 women with good-risk DCIS, defined as DCIS that was mammogram-detected, 2.5 cm or less in size, with final margins at least 3 mm or greater, and low or intermediate nuclear grade. Patients were randomized to radiation or observation, with tamoxifen use optional.
The 15-year cumulative incidence of IBR was 7.1% with radiotherapy versus 15.1% with observation (HR 0.36, 95% CI 0.20-0.66, P=0.0007). Invasive recurrence was also significantly reduced in the radiation therapy group (5.4% vs 9.5% for observation, HR 0.44, 95% CI 0.21-0.91, P=0.27), the study found.
“In summary, breast radiation significantly reduces the incidence of IBR and specifically invasive recurrence for good-risk DCIS following lumpectomy with durable results through 15 years. Radiation is the most effective approach for reducing IBR following lumpectomy in this population,” the researchers wrote in the study online in the Journal of Clinical Oncology.
In the following interview, McCormick elaborated on some of the details and implications of the study.
Previous trials have evaluated the benefit of radiation therapy after lumpectomy for patients with DCIS. How is yours different?
McCormick: Our trial was different because we really specified and defined for the first time what we called good-risk DCIS. We actually had to go back several times to the National Cancer Institute to get funding for this, because the NCI told us they wouldn’t fund another radiation trial for DCIS because there have been so many DCIS prospective trials before this one, and they all showed a disadvantage for radiation. So they didn’t understand at first why this one was different. But it’s because we are defining good-risk DCIS as being small, no symptoms — it’s always picked up mammographically — with relatively wide margins, and low or intermediate nuclear grade.
All those prior trials by and large said as long as you had a diagnosis of DCIS you could go on the trial. So there were many patients in the previous trials, much larger than ours, that had high-grade disease, or sometimes even disease presenting as a mass, where we really wanted a woman with no symptoms and just this very tightly defined good-risk disease.
And personally I was hoping the radiation would not do as much as it did, because I was really convinced we could leave radiation out for some of those women. That was the prime driver for me — I was trying to reduce treatment in what I defined as something I thought would do very well without treatment.
What do you advise oncologists about choosing radiation therapy versus observation for this patient group based on your results?
McCormick: You know, I’ve been trying to spin these results to basically say that radiation oncologists should sit down and discuss with a patient what her idea of a risk is for a local recurrence. As you probably know, DCIS is not life-threatening, everybody does well, it doesn’t spread. It’s just really modifying the local recurrence risk, and in this trial consistently without radiation the local recurrence risk was about 1% per year.
So if I discuss that with someone who’s in her mid-60s, that sounds like something she may not have to worry about, and oftentimes some of the older patients I think still opt, once we give them the appropriate information, not to have the radiation. Other patients, knowing the radiation reduced the risk by more than half, makes them think this is something they really want to do.
So I think in contrast to how I’m really pushing for a certain treatment if I have someone with invasive breast cancer or triple-negative breast cancer, I’m really trying when I speak to these patients — and hopefully other oncologists will too — to understand what that woman’s level of comfort with risk is.
I think for some patients 1% a year, especially if they have other comorbidities and they are older, sounds fine. And that is part of the message we tried to put forth in the discussion section, that it is a statistically significant result, but because you’re only changing the local recurrence risk, and it’s still relatively low without radiation, discussing observation I think is okay too.
In your study, most patients received radiation therapy with conventional fractionation. However, in the 10% of patients who received hypofractionated radiation therapy, no IBR was observed. What do you make of this finding?
McCormick: I was very glad to see it. I used to sit on the NCCN [National Comprehensive Cancer Network] breast guidelines committee for more than 20 years, and I know we adopted a guideline for early breast cancer of hypofractionation, but it wasn’t widely taken up by many of the radiation oncologists in the country.
So I was very happy to see that these patients with DCIS did extremely well with hypofractionation, and I’m hoping this result will bring more radiation oncologists on board with using hypofractionation for these women. I think it’s great for them, they have less inconvenience, fewer trips to the radiation department, and certainly in terms of my experience the cosmetic outcome, as well as the cancer outcome, is extremely good.
Are there any outstanding questions in this area that need to be addressed with further research?
McCormick: I am certainly thinking about working with one of the other cooperative groups focusing specifically on the women who did not get radiation but who did take the tamoxifen hormone therapy for 5 years to get a better understanding of what tamoxifen or an anti-estrogen hormone treatment would do for these women if they opt for no radiation — how much is it improving their outcomes.
So I think that’s an avenue for further research. And then I think, though it’s not exactly my area of expertise, but again going back to what we were talking about, just really being able to understand how a woman looks at risk, and trying to tease out what’s important to her, and how can physicians communicate risk and that kind of information to patients, helping them guide patients to an outcome that’s good for them.
Read the study here and expert commentary about the clinical implications here.