For Melissa Resnick, Valentine’s Day 2012 holds a particular significance: It is the day she had a mammogram that led to a biopsy showing she did—yet did not—have breast cancer.
The biopsy showed that Resnick had DCIS—ductal carcinoma in situ. Breast cancer is staged according to a system that takes into account how large the tumor is and how far it has spread. DCIS is stage 0: It has not spread outside a breast duct (where virtually all breast cancer begins) into the surrounding breast tissue. For this reason, DCIS is often referred to as a precancer.
If left untreated, some DCIS lesions go on to become invasive cancers, while others remain a harmless precancer and never leave the duct. But despite decades of effort, doctors can’t yet tell a woman which type she has. As a result, breast specialists recommend that all DCIS be treated with surgery, often followed by radiation and hormone therapy. Yet many of the women treated for DCIS are likely to have not needed treatment at all.
The In Situ Impasse
The American Cancer Society estimates that this year slightly more than 60,000 women in the U.S. will be diagnosed with an in situ breast cancer. About 83 percent of these women will have DCIS; most of the others will have lobular carcinoma in situ (LCIS), also called lobular neoplasia. (See
“And Then There’s LCIS" below.)
Without mammography, DCIS would be a rare diagnosis. Unless it turns into a lump, which isn’t often the case, DCIS is detected only when a radiologist observes microcalcifications—clusters of white specks of calcium—on a mammogram. (See “But Is It Really DCIS?” below.) With mammography’s introduction as a screening tool in the late 1970s and early 1980s, DCIS diagnoses began to increase rapidly. Today, DCIS is the fourth most common cancer diagnosed in women.
Laurel Habel, an epidemiologist at Kaiser Permanente Northern California Division of Research in Oakland, has spent most of her career trying to determine the appropriate treatment for women with DCIS. “I have been doing research on DCIS since the 1990s, and it was a dilemma then and it still continues to be a dilemma today,” says Habel.
This leaves women with DCIS and their doctors in a quandary. Cancer treatments are generally given to reduce a patient’s risk of dying. Women can die when an invasive breast cancer spreads to and affects other organs in the body. But women don’t die from DCIS, because the cells can’t wreak havoc when they are inside the duct.
“We know we are overtreating DCIS,” says Eileen Rakovitch, a radiation oncologist at Sunnybrook Health Sciences Centre’s Odette Cancer Centre in Toronto. “In most women, it will not become an invasive breast cancer and it will not be life-threatening. But for some, DCIS is the first step to invasive cancer—and we don’t know who these women are.”
A nurse-midwife, Resnick, now 51, says she understood right away that having DCIS meant her disease was at a very early stage. “I was like, take a breath, it’s all right, it is not going to kill you,” she recalls. “It gave me some peace of mind in a way. I knew I was not going to die, and that is what you need to hear.”
But that didn’t make treatment decisions easy. “I remember crying and [telling the breast surgeon] just spare my breast,” Resnick says. The breast surgeon said that was possible, but that Resnick would not end up with a good cosmetic result. So, she had a bilateral mastectomy, followed by reconstruction. “I understand that is drastic,” she says. “But I didn’t do it because I was concerned I was going to die. I did it because I wanted to have a good aesthetic result.” And as someone who had often been called back for a repeat mammogram because of calcifications, and then had to have a mammogram every six months, she says, “I didn’t want to have to worry about continuing to have mammograms and having to go through this again.”
More than 7,000 of the women diagnosed with an in situ breast cancer this year will have LCIS—lobular carcinoma in situ, also called lobular neoplasia. LCIS is not typically identifiable on a mammogram, but it is usually found due to mammography, when a radiologist thinks a spot looks suspicious and it is biopsied. For that reason, incidence rates of LCIS, like those of DCIS, have risen with increased use of mammography.
LCIS is not a precancer like DCIS. The presence of LCIS in one breast means a woman is at increased risk of developing breast cancer in either breast. (The exception is a type of LCIS called pleomorphic LCIS, which does have the potential to become an invasive cancer and is treated like DCIS.) But it’s not clear precisely to what extent most LCIS increases risk.
The average woman has a 12 percent risk of developing breast cancer in her lifetime. “With LCIS, some studies say a woman’s lifetime risk could be between 15 and 20 percent and some say it’s 30 to 40 percent,” says Theresa Schwartz, a breast surgical oncologist at Saint Louis University School of Medicine. “It’s hard to say to a woman, ‘Your risk may be barely above the average risk or it could be three times the population risk.’ ’’
The ambiguity also has made it difficult to develop screening recommendations for women with LCIS. The American Cancer Society recommends that any woman with a breast cancer risk of 20 percent or higher have magnetic resonance imaging (MRI) of her breasts annually, in addition to mammography. But right now, women with LCIS are not included in that group because it’s not certain what their risk actually is or whether they would benefit from additional screening.
“We know that if you have an LCIS diagnosis, that you should have an annual mammogram and breast exam,” says Schwartz, who published a study on MRI and LCIS in the February 2015 Journal of Surgical Research. “For women [with LCIS] who have a family history of breast cancer or other risk factors, MRI is more likely to be beneficial. But for other women [with LCIS], there is no good recommendation.”
Sixty-nine percent of the women diagnosed with DCIS between 2007 and 2011 were treated with breast-conserving surgery, such as lumpectomy. Of these, 68 percent also had radiation. Seemingly, one of the easiest ways to reduce overtreatment would be to determine which of these women needed radiation to prevent a recurrence. But as a recent study suggests, the problem is not only identifying cancers that are “good risk”—those that likely don’t need radiation—but also explaining risk to patients.
The study, published in the March 1, 2015,
Journal of Clinical Oncology, randomly assigned several hundred women with what might be described as “good risk” DCIS to either radiation or observation after breast-conserving surgery. A “good risk” cancer was defined as DCIS that occurred in only one place in the breast, had a low or intermediate nuclear grade (determined by the size and shape of the cell’s nucleus), was less than 2.5 centimeters and had been removed with a clear margin around it that was at least 3 millimeters wide. After a median follow-up of seven years, two of the 287 women who received radiation had a recurrence compared with 19 of the 298 women who did not have radiation. Of those who had radiation and a recurrence, one had DCIS and one had an invasive cancer; of those who didn’t have radiation, eight had invasive cancer and 11 had a recurrence of DCIS.
On the one hand, the overall rate of recurrence was low whether the women had radiation or not—suggesting that the study’s definition of “good risk” can be used to identify many women with DCIS who probably don’t need radiation. On the other hand, the study results also suggest that radiation can significantly reduce the risk of recurrence. And different patients will interpret their own risk of recurrence in different ways.
“What I start with and then circle back to with patients is that … your decision to have radiation or not” is not going to influence survival, says Beryl McCormick, a radiation oncologist at Memorial Sloan Kettering Cancer Center in New York City, who led the study. “[Ten-year] survival rates are 99 percent. You will be fine no matter what you decide to do—and that helps women feel confident about their decision.
“I have patients who have had several benign biopsies before being diagnosed with DCIS, and they are so fed up with mammography that they want to have a mastectomy,” she continues. “There are others who have breast conservation and choose radiation because they want to minimize their chance of needing future biopsies. And there are others who say it’s not going to impact my survival and the chance of it coming back is low, and only half of those that do come back are invasive, and since I can only have radiation once, I’ll wait and do it if it comes back.”
About 1.6 million women have breast biopsies in the U.S. each year. Breast specialists recommend treatments based on pathologists’ reports of what this breast tissue looks like under a microscope. But as a new study underscores, pathologists can come to different conclusions when looking at the same cells, especially when differentiating DCIS from atypical ductal hyperplasia, an overgrowth of cells that slightly increases breast cancer risk.
In the study, published in the March 17, 2015,
Journal of the American Medical Association, 115 pathologists each reviewed 60 biopsy slides selected from a total of 240 biopsies, and their diagnoses were then compared with those of an expert panel that reviewed the same slides. The vast majority of the time, the individual pathologists correctly identified slides that the panel classified as invasive breast cancer. But on slides that the panel identified as DCIS, the pathologists diagnosed DCIS only 84 percent of the time. The pathologists either overdiagnosed the remaining cases of DCIS as invasive cancer or underdiagnosed them as atypical ductal hyperplasia.
“We need to be careful with these diagnoses because while four out of five pathologists might agree, one in five doesn’t,” says Joann Elmore, an internist and epidemiologist at the University of Washington School of Medicine in Seattle, who led the study.
The findings suggest that a woman may want to seek out a second opinion from a pathologist about her diagnosis, just as she might seek a second opinion about treatment options. “I would encourage anyone with DCIS to have a second pathology opinion,” says Beryl McCormick, a radiation oncologist at Memorial Sloan Kettering Cancer Center in New York City. “Low-grade lesions are not easy to interpret, and you need a lot of experience.”
The cost of obtaining a second pathology opinion varies, but it is likely to be $150 or more. Insurance may cover it.
The radiation study also illustrates the extent to which attempts to identify a “good risk” DCIS based on the tumor’s pathological characteristics—such as its grade and size—fall short, and why researchers are trying to find molecular markers that would provide more personalized information. Genomic Health makes OncotypeDX, a 21-gene test that provides women who have hormone-sensitive invasive breast cancer with information about their risk of recurrence. The test has allowed many women to opt for hormone therapy alone and avoid chemotherapy. That success led Genomic Health, in 2014, to begin marketing a similar test that uses a subset of OncotypeDX’s 21 genes to give women who have DCIS information about their risk of recurrence.
Rakovitch recently collaborated with Genomic Health to see how well the subset of genes predicted recurrence in a large group of women with DCIS she has followed for about a decade. At the time of their diagnoses, the women had had their DCIS removed with clear margins. None had radiation. Over the next 10 years, about 18 percent had a recurrence.
Genomic Health’s DCIS test was performed on preserved tumor tissue from these women, allowing the study team to see how well the test score—which can be a number from 0 to 100—would have predicted how these women fared. Findings that Rakovitch presented in December 2014 at the San Antonio Breast Cancer Symposium showed that the DCIS score helped differentiate those with a low risk of recurrence from those with a higher risk.
Right now, though, the test can’t distinguish between being at risk of a recurrence of DCIS or an invasive cancer. Identifying the molecular markers that could make that distinction “is research we are engaged in right now,” says Rakovitch.
Though the gene test is an independent predictor of risk, some doctors don’t think the test’s biomarkers make it easier to identify low-risk women who wouldn’t need radiation. “The number of local recurrences in the [Genomic Health test’s] lowest risk group is higher than what we are seeing with just standard pathology defining low risk,” says McCormick. “It is quite likely there will be some excellent markers, but we’re not there yet.”
All in a Name
Concerns about overtreatment of DCIS led some doctors and advocates to float the idea of branding DCIS with a new name that doesn’t label it as cancer. But that proposal never got traction. And it’s not clear how helpful it would have been if it had.
“I often will go and look at a patient’s pathology slides with the pathologist, and there’s no doubt if you are looking at cells that are really out of control growing within the duct,” says McCormick. “I know there are physicians who think patients will feel better if we call it something else, but we will still be offering the same treatment options. I think my job [explaining risk and treatments] would be more difficult if I tried to discuss radiation with a woman who had a high-grade, high-risk DCIS that I was calling a non-cancer.”
It also is not clear if changing the name would make it more or less challenging for patients to understand their risk of recurrence. “Every individual has a different tolerance for risk,” says Kaiser’s Habel. Studies indicate that most women with DCIS should be eligible for breast-conserving surgery, she notes. Yet “a younger woman may want more aggressive therapy for her DCIS because she has a longer life to get another cancer or has young children to worry about—and it’s in this group we see more bilateral mastectomies.” There also is personal choice. “Some women may prefer a mastectomy to a lumpectomy followed by six weeks of radiation, especially if they live far from a treatment center or are older and have grandchildren to take care of,” Habel adds.
And then there is what is perhaps the ultimate irony of a DCIS diagnosis. “If a lesion is truly in situ,” says Habel, “it is the one time when mastectomy is curative. It’s not curative if you have invasive disease, because once it’s invasive, there are cells that can spread to other areas, and they are the ones that will be deadly, not the ones that have been left behind in the breast.”
Viewed through that lens, some might regard Resnick’s choice to have a double mastectomy as less extreme. Ultimately, though, what matters most is how Resnick feels about her choice, not how others perceive what she did. “I understand that there are people out there who say DCIS isn’t cancer—and clearly, grammatically, because it contains the word in situ it’s not—and who think [I was overtreated]. But I feel that I made what was for me the best decision,” says Resnick. “I sleep well every night. Not on my stomach, like I used to. But I am at peace with my decision.”
Two years ago, Resnick began to facilitate a DCIS support group offered by SHARE, a New York City–based program for breast and ovarian cancer survivors. Her goal is to help others make sense of the dilemma of DCIS. “Because DCIS is somewhat of a gray area, people have to be well-educated and make the decision that is best for them,” she says. “I always felt like no woman wants to be in this sorority, and, if you are in it, it’s best to know all your options.”
June 26, 2015