An 18-year study from the National Cancer Institute has found that widespread screening for ovarian cancer doesn't save lives but does set up many women for needless surgery and avoidable complications.
The results, published Wednesday in the Journal of the American Medical Assn., were not a complete surprise, said study co-author Dr. Christine Berg of the National Cancer Institute. Still, experts are disappointed that yet another attempt to catch cancer early has failed to help patients beat the disease.
"I'm the chief of early detection," Berg said. "I would have loved to have found a benefit."
The study used two common screening tests for ovarian cancer: a transvaginal ultrasound scan and a blood test for CA-125, a protein released by ovarian cancer cells. Although these tests may still be helpful for women at high risk for the disease, their value for other women has long been in doubt.
"Every professional group says 'no screening for average-risk women,'" said Dr. Beth Karlan, who directs the Women's Cancer Research Institute at Cedars-Sinai's Samuel Oschin Comprehensive Cancer Institute in Los Angeles. "We need to be sober in listening to the result."
Karlan will continue to tell her patients who aren't at high risk that they shouldn't be screened. "Many women demand CA-125 testing," she said. "I have had to have that discussion with many, many patients."
During the 18-year study, researchers assigned 39,105 postmenopausal women to get annual screenings — which included both the ultrasound scan and the CA-125 blood test — and compared them with 39,111 women who received standard care at the same medical centers. Deaths from ovarian cancer numbered 118 in the screening group and 100 in the control group — statistically, there was no real difference.
However, women in the screening group faced extra risks. For instance, 1,080 women underwent surgery to remove ovarian tissue that looked suspicious but turned out to be harmless. And 163 women (15%) in that group experienced at least one serious complication, such as infections or blood clots.
Doctors certainly had cause to hope that widespread screening could save lives. Ovarian cancer is especially deadly and aggressive. It's not that common, but it still ranks as one of the top five cancers that kill women. An estimated 21,880 American women will be diagnosed with the disease this year and 13,850 will die from it, according to the American Cancer Society. Fewer than half — 46% — of women will be alive five years after diagnosis.
One reason the odds are so bad with ovarian cancer is that by the time it's diagnosed, the cancer has usually spread to other organs. Only 1 in 5 ovarian cancers is found at an early stage. When that happens, the five-year survival rate is 94%.
So, the logic goes, if the cancer could be detected earlier, it would be more treatable and more women would live longer.
But nothing about ovarian cancer is that simple. For one thing, anatomy makes early cancers hard to spot. The ovaries' deep location means tumors can't be felt, as they can in breast cancer screenings, and biopsies involve abdominal surgery, which carries its own risks. Ultrasound scans help doctors see unusual growths in the ovaries, but they can't tell by looking if it's a benign cyst or a malignant tumor, Berg said.
Growths are removed surgically, and surgery means complications. "Complications might be OK if we're successful in avoiding cancer," Berg said. But because screening doesn't seem to save lives, she believes the risks are unacceptable.
The blood test also suffers from not being specific enough, especially in the early stages of ovarian cancer. Ovary cancer cells shed the CA-125 protein into the blood. But in more than half of women with early-stage ovarian cancer, there's not enough of the marker to raise any alarm, Karlan explained. And if a blood test does detect high levels of CA-125, that doesn't necessarily mean that a woman has ovarian cancer. Endometriosis, fibroids and other cancers can sometimes cause levels to spike, she said.
When the National Cancer Institute study was designed nearly 20 years ago, doctors were aware of the tests' shortcomings. Yet many hoped that combining the two methods might work to catch cancers sooner and ultimately prevent deaths.
Ultrasound techniques have improved since then, and researchers are looking at new ways to use biomarkers. A British study in progress is monitoring CA-125, watching for rapid increases over time. Results are expected in about four years.
Karlan is conducting a study of CA-125 in combination with another marker called HE4. This combination is used to monitor recurrent ovarian cancer, she said, but her study will assess its value in screening.
Although the NCI study found no benefit for widespread screening, the tests are still valuable for women at high risk of developing ovarian cancers, Karlan said. Women should consult their doctors to figure out their risk profile; factors that go into that assessment include family history of ovarian cancer, breast cancer or colorectal cancer, obesity, never having been pregnant, certain medications, and carriers of BRCA genes.
Karlan said women should be aware of certain symptoms associated with ovarian cancer, including persistent bloating, abdominal pain, and changes in bladder or bowel function (for more information, go to http://www.cancer.org and search for ovarian cancer symptoms). Although such symptoms are not specific to cancer, they should still be checked out.
"Know your body. Know your family history," she said. "But if you're not at high risk, screening will offer more harm than good."