“We caught it early.”
For a woman just diagnosed with breast cancer, those are words she hopes to hear.
Cancer screening is built on a simple premise: Find cancer early, treat it early. Save a life.
Sometimes that’s exactly what happens. A mammogram finds a tumor before a woman can feel a lump, before the cancer spreads. Finding it may save her life.
However, that’s not always the case. Some tumors grow so slowly that they never would have caused symptoms during a woman’s lifetime. She would have lived with the cancer, unknowingly, and died of something else. Instead, she has surgery, perhaps radiation, and all the risks and side effects that go along with them. The cancer might have left her alone. The treatments may not.
The cancer is there, and the treatments might be appropriate. The problem is that doctors can’t reliably tell which cancers will become dangerous. On a screening program’s scorecard, both count as cancers caught early.
Treating them all may be a trade-off worth making. If screening a woman saves her life, maybe some unnecessary treatment for others is a price worth paying. However, it’s hard to weigh that benefit against the risk of treating many women who may never have fallen ill.
Two new studies in the Journal of Medical Screening examine those questions from different directions. One asks how often mammograms find cancers that would never have made a woman sick. The other asks whether finding cancer earlier really means that screening is saving lives.
The questions extend beyond mammograms. The Food and Drug Administration is weighing a new blood test designed to detect early signs of more than 50 kinds of cancer, including those not routinely screened for. The promise is appealing. However, as medicine develops more ways to find cancer earlier, how do we know when finding it actually helps?
The Cancers That Never Needed Finding
Researchers have a name for finding a cancer that never would have made someone sick: overdiagnosis. It doesn’t mean that the mammogram was wrong or the cancer wasn’t real. It means that, without screening, a woman would have lived and died without knowing that she had it.
However, overdiagnosis is difficult to measure. Once a cancer is found, it’s usually treated. It’s impossible to go back and find out what would have happened if it had been left alone.
Norway offered researchers an unusual opportunity to estimate how often this happens.
In 1995, Norway began sending invitations to women aged 50 to 69 to get a mammogram every two years. The program started as a pilot in four counties and then gradually spread to the remaining counties over the next decade. The staggered rollout let researchers compare breast cancer diagnoses in areas that had organized screening with those that didn’t.
Using Norway’s national cancer registry, researchers analyzed more than 63,000 breast cancer diagnoses over three decades. As screening spread, breast cancer diagnoses went up. That was expected. Mammograms were finding tumors that women didn’t yet know that they had.
Were mammograms simply finding cancers years before they would cause symptoms—or were they also finding cancers that would never have made the women sick?
Consider a woman whose mammogram found cancer when she was 60. Without screening, that cancer may not show up as a lump until she’s 72. Because it was already found by mammogram at 60, she wouldn’t receive a new diagnosis of cancer at 72.
Across thousands of women, screening therefore should produce more diagnoses at younger ages, followed by fewer diagnoses later in life. For every extra cancer found during screening, there should be roughly one fewer diagnosis when screening stopped.
However, that’s not what happened. Diagnoses did fall after women stopped receiving screening invitations at age 70, but not enough. Even after the researchers followed them through age 84, the decline hadn’t made up for the extra cancers found during screening.
The researchers concluded that some cancers weren’t simply being found sooner. Without mammograms, they likely never would have caused symptoms during the women’s lifetimes.
The researchers estimated that 18 of every 100 breast cancers found through screening were overdiagnosed. The estimates included ductal carcinoma in situ (DCIS), often referred to as precancer or stage zero cancer, in which abnormal cells stay inside the breast’s milk ducts but can sometimes develop into an invasive cancer.
Combining their findings with earlier estimates of how many breast cancer deaths screening prevents, the researchers calculated that approximately three women were overdiagnosed for every breast cancer death prevented.
Harald Weedon-Fekjaer, a biostatistician at Oslo University Hospital, senior author of the study, and member of BreastScreen Norway’s advisory board, still supports screening because it saves lives. However, he recognizes that women may weigh the trade-off differently.
“People have different values,” he told The Epoch Times in an email. For some women, a needless cancer diagnosis would upend their lives. Others would endure hard treatment, even if possibly unnecessary, for the chance to watch their grandchildren grow up.
When Early Detection Isn’t Enough
Finding cancer early sounds like a good thing, but does it mean that fewer people die?
In a separate study in the Journal of Medical Screening, researchers sought to answer that question. The researchers examined 58 randomized trials covering 10 types of cancer. The trials compared people offered screening with those receiving usual care. The researchers wanted to know whether trials that found more early-stage cancers also showed greater reductions in cancer deaths.
If finding more cancers was a reliable measure of lives saved, the trials that found more early-stage cancers should also show fewer deaths.
However, that’s not what researchers found. Across the trials, finding more early-stage cancers wasn’t reliably associated with fewer cancer deaths.
The authors point to several possible explanations. Some cancers found early would never have caused harm, as the Norwegian study showed. Others may be so aggressive that finding them sooner doesn’t change the outcome. And some remain difficult to treat even when caught at an early stage.
As the researchers explained in their paper, detecting cancer early “could decrease mortality, make no difference to outcomes, or cause harm.”
That doesn’t mean that screening doesn’t save lives. It does for certain cancers. However, finding cancer early is only helpful if it changes what would have happened to the person otherwise.
“Early detection is essential for cancer screening to work to reduce cancer deaths,” Adam Brentnall, a statistician at Queen Mary University of London who coauthored the study, told The Epoch Times in an email.
However, screening tests are designed to detect cancer early.
“So that alone does not tell us if it works,” he said.
Brentnall said a better sign of screening’s effectiveness is whether fewer people are diagnosed with advanced cancers—those that have grown or spread and are harder to treat.
Earlier research by Brentnall and his team found that trials showing fewer advanced cancers also tended to show fewer cancer deaths. A drop in advanced cancers is a more promising measure of screening’s benefit than simply counting how many cancers are found early.
A new study of more than 105,000 women published on Nov. 6 in JAMA Network Open raises similar questions about breast MRIs, which are sometimes used alongside mammograms for women with dense breasts or those at higher risk of breast cancer.
The researchers found that MRI finds four to six times as many early-stage cancers as mammography. Still, researchers didn’t see fewer women developing more serious, advanced cancers. And the MRIs also led to more false alarms. Lower-risk women who had an MRI needed a biopsy 93 times for every 1,000 screenings, even though no cancer was found. This only happened 15 times with mammograms.
What researchers really want to know is whether fewer women die as a result of screening. However, answering that question takes many trials and years of follow-up.
How Much Screening Is Enough?
For an American woman wondering when to schedule her mammogram, it may depend on which doctor you ask.
The U.S. Preventive Services Task Force recommends a mammogram every two years from age 40 to 74. The American College of Radiology recommends annual screening starting at 40 and continuing as long as a woman is healthy.
Both recommend starting at 40, although screening younger women brings its own challenges. Women younger than 50 tend to have denser breasts, making mammograms harder to read and false alarms more common, Weedon-Fekjaer said.
The groups differ on how much additional screening is worth doing. More mammograms offer more chances to find dangerous cancers early, but also more chances to find cancers that would never have caused harm. Weedon-Fekjaer said overdiagnosis generally goes up with more frequent checks, particularly as women get older.
Dr. Karla Kerlikowske, a breast cancer screening researcher at the University of California–San Francisco, pointed to U.S. computer models that put the trade-off into numbers.
Consider 1,000 women who get 3D mammograms from age 50 to 74. For women screened every two years, the models estimate that about seven of them will be spared a breast cancer death compared with no screening. If screened every year, about nine deaths would be prevented.
That’s approximately two additional lives saved for screening every year versus every other year.
However, those two extra lives come at a price. The models show that the yearly schedule would mean about 400 more callbacks after mammograms that turned out not to show cancer. Fifty additional women would undergo biopsies only to find no cancer. And six more women would be diagnosed with cancers that never would have made them sick, rising from 12 to 18 women.
Kerlikowske cautioned that these numbers are projections. No randomized trial has ever shown that yearly mammograms prevent more deaths than screening every two years.
For a woman overdiagnosed, though, the consequences can be painfully real. It could mean having part or all of her breast removed, radiation appointments, financial stress from missing work, and the list goes on.
“If there is added benefit from annual screening, it is small,” Kerlikowske told The Epoch Times in an email, especially when weighed against the extra tests, false alarms, biopsies, and overdiagnosis.
Dr. Stamatia Destounis, chair of the American College of Radiology’s Commission on Breast Imaging, sees the balance differently.
She favors annual screening because it offers more chances to find dangerous cancers before they spread. Detecting cancer earlier may allow women to undergo less intensive treatment. She cited U.S. modeling studies that project fewer breast cancer deaths when women get a mammogram once per year.
“I would want the patient to understand both the benefits of yearly screening and the risks of biennial screening,” she told The Epoch Times in an email.
The U.S. Preventive Services Task Force says there isn’t enough evidence to say whether screening after 74 does more harm than good. The American College of Radiology recommends continuing screening as long as a woman is healthy.
As women age, though, the chance of finding a cancer that never would have caused trouble in their lifetime increases.
Still, “for a healthy woman, age alone is not a reason to stop screening,” Destounis said.
Weedon-Fekjaer put it bluntly. An 80-year-old with advancing Alzheimer’s disease should never be screened, he said; treatment would likely take her best remaining years without adding to them.
This leaves women with another decision beyond whether screening works. It’s about how much additional screening they are willing to take on for the possibility of greater benefit.
Weedon-Fekjaer said when to stop depends on “both personal preferences and individual health.”
Can Some Cancers Be Left Alone?
How often to look for cancers is one decision. Deciding what to do when cancer is found is entirely different.
No one—pathologists, radiologists, or any other doctor—can reliably predict which cancers will become dangerous. However, researchers have begun investigating whether some low-risk conditions can be monitored rather than treated immediately.
One example is DCIS, often called stage zero breast cancer. Because doctors can’t reliably tell which cases will progress, women diagnosed with DCIS generally undergo surgery, sometimes followed by radiation or medication.
What if some could safely avoid surgery?
A clinical trial called COMET tested that possibility. Researchers assigned carefully selected women with low-risk DCIS to either standard treatment or active monitoring. Rather than getting immediate surgery, women in the monitoring group got regular breast exams and mammograms with further testing if anything changed.
After two years, 4.2 percent of the women assigned to monitoring were diagnosed with invasive cancer in the affected breast, compared with 5.9 percent of those who received standard treatment. By the trial’s own measure, monitoring was no worse than usual care over that period.
Destounis called the results encouraging but cautioned against applying them too broadly.
“It does not establish that every patient with DCIS can safely forgo treatment indefinitely,” she said.
However, for women whose diagnoses meet the trial’s criteria, she said, active monitoring is an option worth discussing with their medical team.
And even when monitoring is medically reasonable, choosing it may not be easy.
“It would probably be very hard to leave a cancer untreated if you know about it,” Weedon-Fekjaer said.
For now, doctors still can’t confidently identify every cancer that can be left alone. However, for some women with low-risk DCIS, the COMET findings could mean leaving the doctor’s office without scheduling surgery. Instead, they could be watched closely, potentially avoiding treatment for something that wouldn’t have harmed them.
The Next Test
The questions raised by mammograms are becoming more urgent as researchers develop new tests that can find cancers doctors don’t routinely screen for.
One such test is Galleri, made by Grail, which looks for more than 50 types of cancer in a single blood draw. Instead of screening one organ or system at a time, it can look for signs of cancer throughout the body.
The promise is appealing. However, a large UK trial shows how difficult it can be to turn more detection into better outcomes.
More than 142,000 people participated in the trial, which compared usual cancer screening with usual screening plus Galleri. The blood test found additional cancers, including some at early stages.
However, it missed its main goal. Among 12 types of cancer the researchers focused on, Galleri did not significantly reduce the combined number diagnosed at stages three and four, when cancers have grown larger, deeper into the tissues, or spread throughout the body.
The test did find fewer people with stage four cancers on its own. Researchers are continuing to follow participants to learn whether that difference continues and, ultimately, whether the test prevents cancer deaths.
Brentnall, whose research unit helped conduct the trial, said the results show why detecting cancer is only part of evaluating a screening test.
Brentnall said detecting cancer isn’t enough to determine whether a screening test works, but that “just focusing on cancer mortality” isn’t enough, either. What happens after a positive result matters, too.
In September, an FDA advisory committee voted 7–2, with one abstention, that Galleri’s benefits outweighed its risks. However, members were more divided on whether the evidence established its effectiveness, voting 6–4 in favor. The FDA has not yet made a final approval decision.
For women deciding whether to undergo screening, Brentnall suggested that they ask their doctors, “For someone like me, in concrete terms, what are the potential benefits and harms of having this screening test, and how do you know?”
Many women reading this have already been treated for a cancer a mammogram found. Asked what they should take from the Norwegian findings, Weedon-Fekjaer didn’t pretend to have an answer for every woman.
“Mammography screening might have saved your life, but we can never know,” he said.
