If treatment prevents invasive breast cancer, why have the rates of invasive cancer gone up in the mammography era?
—Shelley Hwang, MD
There are a few things about ductal carcinoma in situ (DCIS) on which everyone agrees: Incidence increased dramatically with the advent of mammography screening, not all cases detected will go on to cause symptoms in the patient’s lifetime, and there’s no proven way to tell which cases will progress.
But agreement often ends when it comes to clinical management. Should all cases be treated with surgery and radiation in the absence of proven markers to show which cases will progress? Or is watch-and-wait an option? Can existing tools be used to guide the decision?
Experts put their views on the line in a debate at the 66th Annual Cancer Symposium of the Society of Surgical Oncology, held March 6–9 in Washington, DC.
Battle Lines Drawn
“All DCIS patients benefit from standard therapy,” declared Thomas B. Julian, MD, a Professor of Surgery at Drexel and Temple Universities in Philadelphia, who took the pro-treatment position in the debate-format session.
DCIS is a heterogeneous disease that is overdiagnosed and overtreated, argued Shelley Hwang, MD, Professor of Surgery and Chief of Breast Surgery at Duke University. “Not all DCIS patients benefit from standard therapy.”
The increase in DCIS in the United States has been dramatic, rising from 1% in the 1970s to 25% in 2009. It now accounts for 30% to 40% of lesions found by mammography. And MRI is likely to find even more.
“As our technology moves, this will be a continuing problem…. We will have to move on to a better way to interpret the taxonomy of this disease,” Dr. Julian said. The current standard of care, according to National Comprehensive Cancer Network (NCCN) guidelines, is either lumpectomy followed by radiation, or mastectomy, or lumpectomy alone. However the level of evidence supporting the last option is considered “low-level.”
It’s Cancer—Treat It as Such
Dr. Julian began with the basics: DCIS shares many characteristics and biomarkers with invasive cancer. And in the many major, randomized trials of DCIS, adjuvant treatment has reduced the recurrence rate. In trials by the National Surgical Adjuvant Breast and Bowel Project, for instance, radiation therapy after lumpectomy reduced recurrence by 50% (though it had no effect on overall survival). In other trials, treatment with tamoxifen reduced recurrence rates when compared with a placebo, and radiation was superior to observation. Even participants deemed “good risk” benefited to a degree in these trials.
In fact, although some indicators of risk—margin status, presence of comedo necrosis, estrogen receptor status, Ki67 status, and others—are under study, none has so far been able to predict treatment benefit. In analyses of the large randomized trials, “there was no subset that didn’t seem to benefit from use of radiation therapy, even those with free margins and no comedo necrosis,” Dr. Julian said. Biomarker trials will need to be completed, including long-term follow-up and validation, “before we start moving away completely from use of radiation therapy in a subset of patients with DCIS,” he insisted.
“Certainly molecular profiling is very important and will probably help us identify where we might not have to have a draconian full-court press for this disease. But this is still a work in progress,” he said.
One Size Does Not Fit All
This approach ignores some important facts, argued Dr. Hwang. First, it is uncertain whether treating DCIS prevents the incidence of invasive breast cancer overall. “If treatment prevents invasive breast cancer, why have the rates of invasive cancer gone up in the mammography era? This is in direct contrast to colon cancer surveillance, where colonoscopy and removal of polyps has reduced the incidence of colon cancer.”
Second, treatment doesn’t come without costs—financial costs to the medical system and personal costs to the patient. Surgery and radiation leave their marks and may have complications. (On her obligatory financial disclosure slide, Dr. Hwang included what she said was the most important disclosure: “I hate overtreatment.”)
Third, there are markers, even if not perfect, that can help in gauging a patient’s risk of DCIS recurring or progressing. Dr. Hwang highlighted clinical markers such as grade, tumor size, estrogen receptor status, age, comorbidities, and so forth; the Van Nuys Prospective Score, a conglomerate of clinical data and other markers such as margin status; and the newer “nomogram” from Memorial Sloan-Kettering Cancer Center that uses 10 variables to predict the likelihood of recurrence in different subsets of patients.
She also pointed to emerging molecular tools, such as the “12-gene score,” and a “DCIS score,” which identifies low-, intermediate-, and high-risk patients based on the presence of cancer-related genes.
Patient’s Role in Decision-making
More validation of markers is needed, Dr. Hwang said, agreeing with Dr. Julian on this point. But, she argued, what we know so far can still serve as a starting point for discussion with patients.
And finally, what about the patient? “All the preceding discussion has almost assumed that we’re in the driver’s seat and patients are going to do whatever we say. But this is no longer the reality and I think it’s essential that we consider the patient’s very important central role.”
It is important that women be given the chance, and the necessary information, to choose alternatives to standard therapy, she concluded, “and it really is our responsibility to start offering them options and inform them of the possible outcomes for each option. Every patient is unique, and the days of one treatment pathway for every patient are long gone. These conversations will lead the way toward real shared decision-making, and will likely be relevant for many of the disease we screen for today.” ■
Disclosure: Dr. Julian reported no potential conflicts of interest. Dr. Hwang has served as a consultant or advisor for Genomic Health and has received research funding from Merck.