FIRST DO NO HARM
All These Cancer Screenings Probably Won’t Save Your Life
Millions are suffering through needless and destructive treatments that follow early-detection testing results—and all for cancers that likely wouldn’t kill you anyway.
Some cancers are easy to prevent. For example, sun block can prevent skin cancer; the hepatitis B vaccine can prevent liver cancer; the human papillomavirus (HPV) vaccine can prevent cervical cancer as well as head, neck, anal, and genital cancers; and cessation of cigarette smoking can prevent lung cancer.
Other cancers are harder to prevent, like those of the prostate, thyroid, and breast. The reason, as explained by Gilbert Welch, a professor of medicine at Dartmouth, is best explained using a barnyard analogy. Imagine, writes Welch, that three animals in a barn are trying to escape: a bird, a turtle, and a rabbit. When you open the door, they escape at different rates.
• The bird, which will fly out before you can close the door, is analogous to a cancer that is so aggressive that it will kill you no matter what you do. Even if detected early, it doesn’t matter—this cancer is invariably fatal.
• The turtle, which is so plodding that it will never escape, is analogous to a cancer that is so slow growing, so non-virulent that it is never fatal. People die with this cancer, not from it.
• The rabbit, which can be caught if the door is closed quickly enough, is analogous to a cancer worth detecting. If not detected early, this cancer is fatal. If it is detected early, then the screening test will save lives.
Screening tests are valuable only if they are detecting mostly rabbits, like Pap smears to detect cervical cancer or colonoscopies to detect colon cancer. For thyroid, prostate, and breast cancers, however, the value of early screening tests isn’t so clear. Atul Gawande, a surgeon at Johns Hopkins University School of Medicine and prolific author, sums it up best: “We now have a vast and costly health-care industry devoted to finding and responding to turtles.”
We’ll start with thyroid cancer.
In 1999, the South Korean government initiated a large-scale screening program consisting of ultrasonography followed by needle biopsy to detect thyroid cancer. As a result, South Korean physicians detected 40,000 new thyroid cancers, 15 times more than had been detected before the program started. Thyroid cancer became the most common cancer in South Korea.
South Korean doctors treated these newly diagnosed thyroid cancers by completely removing the thyroid—a thyroidectomy. People who undergo these surgeries require thyroid replacement hormones for the rest of their lives. And adjusting the dose can be difficult. Patients suffer from too much thyroid replacement hormone (sweating, heart palpitations, and weight loss) or too little (sleepiness, depression, constipation, and weight gain). Worse, because of nerves that travel close to the thyroid, some patients suffer vocal-cord paralysis, which affects speech.
Initially, South Korean health officials were excited that they had detected all of these new thyroid cancers. Then they looked at the mortality rates. No difference. The incidence of death from thyroid cancer was the same as before the screening program. The only tangible outcome was that now tens of thousands of people had suffered the side effects of thyroid surgery and hormone replacement therapy. As the Koreans had learned, the problem with thyroid cancer is that it’s mostly turtles. There just weren’t enough rabbits to make the screening program worthwhile.
Overdiagnosis and overtreatment of thyroid cancer hasn’t been limited to South Korea. In France, Italy, Croatia, Israel, China, Australia, Canada, and the Czech Republic, the rates of thyroid cancer have more than doubled. In the United States, they’ve tripled. In all of these countries, as had been the case in South Korea, the incidence of death from thyroid cancer has remained the same.
Early detection of prostate cancer has also been problematic.
In 1970, Richard Ablin, a professor of pathology at the University of Arizona, created the PSA blood test. PSA, which stands for Prostate-Specific Antigen, is an enzyme made by cells in the prostate gland. At first, doctors used the PSA test to determine whether prostate cancers had relapsed. Then they used it to predict whether someone had prostate cancer. If the level of PSA in the blood was high, then urologists recommended a prostate needle biopsy. If the biopsy showed prostate cancer, then men either had a total removal of the prostate or radiation therapy.
Prostate cancer is now the most commonly diagnosed non-skin cancer in the United States. So, what’s happened to the incidence of death from the disease? Nothing. The risk of dying from prostate cancer hasn’t changed in the last 10 years. Indeed, about 50 percent of men older than 60 have been found at autopsy to have prostate cancer after they had died from something else; in men more than 85 years old, that number climbs to 75 percent. In other words, as had been the case for thyroid cancer, people were more likely to die with prostate cancer than from it. Prostate cancer, like thyroid cancer, is mostly turtles.
In 2012, the United States Preventive Services Task Force recommended against PSA-based screening tests for prostate cancer. In 2014, the American College of Preventive Medicine followed suit. But not before a lot of harm had been done. Prostate surgeries and radiation therapy commonly cause incontinence and erectile dysfunction. Worse, five of every 1,000 men will die from prostate surgery. And all for potentially nothing.
Mammography screening for breast cancer is also being reevaluated. Although it is clear that mammography, which was introduced in the U.S. in the mid-1970s, saves lives; the question is how many and at what cost.
In 2012, Archie Bleyer and Gilbert Welch published a study in the New England Journal of Medicine titled, “Effect of Three Decades of Screening Mammography on Breast-Cancer Incidence.” They found that, with the advent of screening mammographies, the incidence of breast cancer in the U.S. had doubled. For every 100,000 women screened, the number of women diagnosed with breast cancer had increased from 112 to 234. At the same time, the number of women presenting with late-stage breast cancer (the kind that often results in death), decreased from 102 to 94 (per 100,000). In other words, only eight of the 122 women diagnosed with breast cancer appeared to benefit from the screening. Eight. The others had been treated with mastectomy, radiation therapy, and chemotherapy without clear benefit. The authors concluded that, although the incidence of death from breast cancer had clearly decreased during the era of screening mammographies, most of that decrease was due to better treatment, not better screening. They also estimated that during the three decades of mammography, about 1.3 million women had been diagnosed with a cancer that would never have killed them.
As a consequence of this and other studies, the recommendations for mammography have changed. Previously, doctors had recommended that all women between 40 and 74 years of age get a mammogram every two years. Now the United States Task Force for Prevention recommends screening starting at age 50, not 40. There were just too many non-cancers that were called cancers—and women were suffering needlessly, both physically and psychologically, as a result.
Until scientists are able to find genetic or biochemical markers that clearly distinguish aggressive cancers from innocuous ones, we are going to continue to suffer from the overdiagnosis and overtreatment of cancers that aren’t really cancers at all. In 2012, a group of physicians at the American Board of Internal Medicine in Philadelphia founded Choosing Wisely. This group provides educational materials to help patients choose therapies that are free from unnecessary harms and supported by the evidence. In no arena is this advice more useful than in cancer-screening programs.
Paul A. Offit is a professor of pediatrics at the Children’s Hospital of Philadelphia and the author of Pandora’s Lab: Seven Stories of Science Gone Wrong (National Geographic Press, April 2017).