What If What You ’Survived’ Wasn’t Cancer?

Bloomberg View , August 17, 2013

You’re feeling fine when you go for your annual physical. But your mammogram looks a little funny, or your PSA test is a little high, or you get a CT lung scan and a nodule shows up. You get a biopsy, and the doctor delivers the bad news: You have cancer. Because you don’t want to die, you agree to be sliced up and irradiated. Then, fortunately, you’re pronounced a “cancer survivor.” You’re glad they caught it early.

But maybe you went through all that pain for nothing.

For decades, the reigning theory has been that the earlier a cancer is spotted and treated, the less likely it is to be lethal, because it won’t have time to grow and spread. Yet this theory infers causality from correlation. It implicitly assumes that cancer is cancer is cancer, even though we now know that even in the same part of the body, cancer is many different diseases -- some aggressive, some not. Perhaps people survive early-stage cancers not because they’re treated in time, but because their disease never would have become life-threatening at all.

This isn’t just logical nit-picking. Thanks to widespread screening, the number of early-stage cancers identified has skyrocketed. In many instances -- including types of breast, prostate, thyroid and lung cancers -- more early diagnoses haven’t led to proportionate decreases in mortality. (New drugs, not early detection, account for at least two-thirds of the reduction in breast-cancer mortality.) The cancers the tests pick up aren’t necessarily life-threatening. They’re just really common. So more sensitive tests and more frequent screening mean more cancer, more cancer treatment and more cancer survivors.

“We’ll all be cancer survivors if we keep going at the rate that we’re going,” says Peter Carroll, the chairman of the department of urology at the University of California at San Francisco and a specialist in prostate cancer.

In a well-intended effort to save lives, the emphasis on early detection is essentially looking under the lamp post: Putting many patients who don’t have life-threatening diseases through traumatic treatments while distracting doctors from the bigger challenge of developing ways to identify and treat the really dangerous fast-growing cancers.

“Physicians, patients, and the general public must recognize that overdiagnosis is common and occurs more frequently with cancer screening,” argues a recent JAMA article by the oncologists Laura J. Esserman (a surgeon and breast-cancer specialist), Ian M. Thompson Jr. (a urologist) and Brian Reid (a specialist in esophageal cancer). They argue for limiting the term “cancer” to conditions likely to be life-threatening if left untreated.

That’s going to be a tough change for a lot of people to swallow. For patients and the rest of the public, getting tested offers a sense of control, encouraging an almost superstitious belief that frequent screening will ward off death. (A few years ago, when the actress Christina Applegate was making the talk-show rounds urging young women to get breast MRIs, my own oncologist told me he was getting calls from women who thought the tests would not merely detect but prevent breast cancer.)

Early detection of non-life-threatening cancers also produces a steady supply of “cancer survivors,” who work to support cancer charities and make their efforts look successful. There’s an entire industry devoted to celebrating “breast cancer survivors” in particular, and many women are heavily invested in that identity. It offers a heroic honorific as a reward for enduring horrible treatments. A term originally coined to remind cancer patients that their disease need not be fatal has become a badge of personal achievement.

Physicians, meanwhile, fear making a mistake. It seems safer to treat someone who doesn’t really need it than to miss something potentially fatal. But, warns Esserman, director of the Carol Franc Buck Breast Care Center at UCSF, “the cancers that grow and spread very quickly are not the ones that you can catch in time with screening.” If anything, emphasizing early detection misdirects research and funding. “We have to come up with better treatments, we have to figure out who’s really at risk for those and figure out how to prevent them,” she says. “We’re not going to fix it with screening.”

There are plenty of scientific unknowns. Take the commonly diagnosed breast cancer called ductal carcinoma in situ, which accounts for about a third of new U.S. diagnoses, 60,000 a year. In these cases, the cells lining the walls of milk ducts look like cancer, but they haven’t invaded the surrounding breast tissue. DCIS was a rare diagnosis before the introduction of mammograms, which are highly sensitive to milk-duct calcifications, and the JAMA article labels it a “premalignant condition” that shouldn’t even be called cancer. Arguably, a lot of women who think of themselves as “breast cancer survivors” have survived treatment, not cancer.

Yet oncologists who identify DCIS have been surgically removing it (and in many cases the entire surrounding breast) for 40 years, so it’s hard to know how dangerous it actually is. “Since we really don’t know the true natural history of DCIS we do not know if DCIS always progresses to invasive cancer or not,” says Colin Wells, a radiologist at the University of California at Los Angeles specializing in breast imaging. “There are some reasons to think not, but this needs to be worked out” with further research. If DCIS does spread to invade breast tissue, the question remains whether that cancer threatens to go beyond the breast, becoming lethal if untreated.

By contrast, we do know that a lot of prostate cancer isn’t dangerous. Autopsy studies show it’s quite common in older men who die from unrelated causes. “Out there in the street, if you remove the prostates in men over the age of 50, 30 to 40 percent would have some kind of cancer,” Carroll says, “most likely, low grade and low volume.”

Thanks to more sensitive tests, he notes, the prostate “cancers we’re detecting today are totally different than the cancers we saw two decades ago. And our ability to distinguish these tumors is much better. We have the wherewithal now to be able to tell a patient that your cancer is highly likely confined to your prostate, of small volume, slow growing, and something that may not need immediate treatment at all.”

Carroll has more than 1,000 patients under “active surveillance,” getting regular PSA tests, imaging and biopsies. Only about one in three turns out to need treatment within five to 10 years. (An additional 10 percent opt for surgery simply because they get tired of all the tests or can’t take the anxiety.) The program is also working, Carroll says, to “decrease the burden of testing,” ideally by eliminating the need for repeated biopsies.

Prostate cancer illustrates the cultural barriers to abandoning what Esserman calls today’s “scorched earth policy.” Despite the widespread awareness that many prostate cancers aren’t life-threatening, many physicians are determined to find and treat it any time a PSA score comes in a little high. “I saw a gentleman this week who had had 12 biopsies, no cancer, and they said there must be cancer in there and they did 24,” says Ian Thompson of the University of Texas Health Science Center at San Antonio, who is one of the JAMA authors.

A prostate-cancer diagnosis is still terrifying to patients and their families. Thompson describes many of his conversations with patients -- and especially with their wives -- as “talking them off the ledge.” When he tells patients they’re likely to be fine without immediate treatment, they often worry how they’ll explain the good news to their children or neighbors. People expect a cancer diagnosis to entail trauma.

Although Carroll thinks calling slow-growing prostate tumors “cancer” is important to encourage vigilance, Thompson wants to change the nomenclature, using the term IDLE (indolent lesions of epithelial origin) to describe low-risk cases where waiting isn’t likely to make a difference. Just using the word “cancer,” he argues, creates unnecessary suffering.

“The number of people that will die from those slow-growing prostate cancers is really low,” he says, but the unacknowledged costs of giving them a cancer diagnosis are huge: “the person who can’t sleep for two weeks before his next test results, and all the follow-up biopsies and all the lost wages, and the people who can’t get life insurance because they now have a new cancer diagnosis, the person whose firm says, ‘Well, we’re concerned you have cancer and therefore you can’t be promoted to this job.’”

It’s a compelling case, but changing the vocabulary finesses the fundamental cultural issue: the widespread and incorrect belief that “cancer” is a single condition, defined only by site in the body, rather than a broad category like “infectious disease.” Someone doesn’t develop “cancer” but, rather, “a cancer.” How frightening that diagnosis should be depends on which one.