Redefining cancer to reduce unnecessary treatment
“We desperately need a 21st century definition of cancer versus the 19th century definition we’ve been using.”
As chief medical officer for the American Cancer Society, seeks to Dr. Otis Brawley heighten public awareness about early detection and prevention of cancer, the need for more research, and the benefits of smoking cessation and dietary improvements in reducing cancer deaths. He also works to eliminate disparities in access to cancer care. Brawley is a professor of hematology, oncology, medicine and epidemiology at Emory University and is a member of the Centers for Disease Control and Prevention advisory committee on breast cancer in young women. He previously served as assistant director of the National Cancer Institute. In his 2012 book, Brawley wrote about How We Do Harm, overtreatment and undertreatment of cancer patients, as well as financial conflicts of interest. Modern Healthcare reporter Sabriya Rice recently spoke with him about how changing the language and perceptions surrounding cancer could improve treatment. The following is an edited excerpt.
Modern Healthcare: What is cancer?
Dr. Otis Brawley: Cancer is uncontrolled cell growth. When a cell goes from one to two mitosis, and then two to four, that’s usually normal growth. If there’s not a stop signal, that’s when you have a cancer. The fact that people had these uncontrolled growths is talked about in the Bible and is even seen on the walls of caves in Egypt from 3,000 or 4,000 years ago.
MH: What do you think is problematic about this definition of cancer?
Brawley: We’ve used the biopsy a lot over the past 160 years, since the Germans gave us the microscopic definition of cancer in the 19th century. They used autopsies of people who clearly died of breast, lung, colon and prostate cancers. They were the first to do biopsies and what we now call H&E staining. They looked at biopsies under their microscopes, drew pictures of what they saw and defined what cancer looks like.
Now, 160 years later, keep in mind X-ray was invented in 1895, mammography in 1955, ultrasound in the 1960s, CT in the ’70s, MRI in the ’80s, and some of our biopsy techniques have just been patented in the past 10 to 15 years. Now we can see a five-millimeter tumor in a woman’s breast, stick a needle into the tumor, sample it and send it to a pathologist. The pathologist uses the same H&E staining techniques that were used 160 years ago and looks at it under a 2014 version of the microscope. That pathologist says this looks just like what the Germans defined as cancer 160 years ago.
The big difference is that what the Germans saw was a tumor that clearly had spread and caused harm and killed the patient. This 5 millimeter thing may very well not be genomically or genetically programmed to grow, spread and cause harm, and that’s what we call an overdiagnosed cancer. In breast cancer, with mammography screening, we believe 10% to 30% of all of our localized cancers are overdiagnosed cancers. These are women who will be treated needlessly.
In prostate cancer, some people estimate 40% to 60% of prostate cancers that we are currently
diagnosing are these over-diagnosed cancers. We desperately need a 21st-century definition of cancer versus the 19thcentury definition we’ve been using.
MH: What would that new definition look like?
Brawley: I think it’s going to be looking at specific cancers, and it’s going beyond looking at what the cancer looks like under the microscope and actually looking into the genes. We’ve got some good leads.
We have a couple of studies and a couple of tests that are available now in breast cancer. They have not been perfected yet. One looks at 21 different genes in breast cancer and asks the question, “Is this specific gene turned on or turned off? If it’s turned on, is it overexpressed or not?” Every one of these 21 specific genes is examined and the same question is asked for each gene. And they actually say that some tumors tend never to be harmful, whereas other tumors always seem to be aggressive. That’s moving us already to the point where we will say, “Ma’am, you have a cancer, but we’re going to treat it less aggressively, because our genomic studies indicate that this may not be a very aggressive tumor.”
MH: Is that similar to the case of ductal carcinoma in situ?
Brawley: DCIS is something that looks like cancer, but it hasn’t penetrated through the basement membrane. Most DCIS’ are in this overdiagnosis category. There are some DCIS’, especially ones that we call highgrade DCIS’, that are destined to become invasive cancers later on.
I was always taught when I took pathology that the term “in situ” essentially means “not.” It’s ductal carcinoma, but it has not invaded through the basement membrane. That’s what the “in situ” means. Unfortunately, many doctors, and especially many patients, when they hear ductal carcinoma in situ, they tend to get more concerned than perhaps they should, and they treat it like cancer. The amazing thing is, today we have women who are getting bilateral mastectomies because of ductal carcinoma in situ. But the same women, if diagnosed with true breast cancer, would choose to get a lumpectomy and radiation—far less treatment for the real cancer.
MH: Last July, the National Cancer Institute published a
“I believe if we can get a better definition of real cancer, versus something that looks like cancer but actually will never cause harm, we can do a much better job for the patient.”
paper proposing that the term “cancer” be reserved for lesions with the likelihood of killing the patient. Is DCIS an example where that type of definition would be applicable?
Brawley: Ductal carcinoma in situ is such a frightening word, and it causes so much emotion on the part of doctors and patients. Some experts have said maybe we should take that frightening word carcinoma out of the phrase and instead call it an indolent lesion of undetermined origin.
MH: Would this kind of clarification help in terms of quality of care and patient safety?
Brawley: I believe if we can get a better definition of real cancer, versus something that looks like cancer but actually will never cause harm, we can do a much better job for the patient. Ultimately, we’re going to be able to tailor our care to the people who need the treatments versus the people who don’t need the treatments. We’re closer to that in breast cancer than in any other disease. The second cancer where we’re close is prostate cancer.
MH: When the NCI paper was published last year, critics said this is just another way to deny care. What would you say to those people?
Brawley: It’s almost impossible to combat statements that are that ignorant, quite honestly. These are folks who honestly don’t understand medicine. I would much rather talk to the population that’s interested in this versus the population that’s trying to fan political flames.
MH: For cancer survivors, what message would this send to them if we use the terminology the authors recommended in that NCI paper?
Brawley: This is an important question. In the prostate cancer literature over the past two years, there have been at least a dozen papers about whether Gleason Score 5 and 6 prostate cancers (scores in the mid-range on a 1-to-10 scale of what experts believe are tumors that will spread) should be called cancer.
Some of the people, as they answer these questions in our medical journals, say they don’t think we should change the name, because we don’t want to tell half of the men who’ve been treated for prostate cancer over the past 50 years that a committee has now decided that they didn’t have cancer. It is very, very tricky.
In his 2012 book, How
We Do Harm, Brawley wrote about overtreatment and undertreatment of cancer patients, as well as financial conflicts of interest.