What Is patientINFORM?

Women diagnosed with ovarian cancer may start having symptoms of the disease months before the cancer is found. But many of these women aren't getting the types of medical tests that could help doctors make the right diagnosis, researchers from the University of California at Davis School of Medicine report in the journal Cancer.

The finding highlights one of the most challenging aspects of ovarian cancer detection, experts say. Symptoms of the disease -- which can include abdominal swelling and digestive problems -- are very vague and easily mistaken for symptoms of other, more common conditions.

"The problem is the symptoms are not pointing to the ovaries, they're abdominal," said lead study author Lloyd H. Smith, MD, PhD, a gynecological oncologist at UC Davis. "These are very common symptoms, so the majority of people who have these symptoms are not going to have ovarian cancer, they're going to have some other problem. This is a very difficult diagnosis to make, even if the symptoms are there."

Ovarian cancer strikes more than 22,000 women in the US each year. When found at an early stage, this cancer is treatable; more than 90% of women with cancer that has not spread beyond the ovary live at least 5 years after diagnosis. Unfortunately, fewer than 20% of ovarian cancers are caught at this early stage.

Smith and his colleagues wanted to know whether it might be possible to find the disease earlier in more women. They studied the Medicare records of 1,985 women age 68 or older who had been diagnosed with ovarian cancer that had already begun to spread into the abdomen or other organs (stage IC or higher).

They looked at how many of these women had seen their doctor because of certain symptoms: abdominal pain or swelling, gastrointestinal symptoms (gas, constipation, diarrhea, and others), or pelvic pain. They also looked at which diagnostic tests the women were given to find out what might be causing the problem. Then they compared these results with the same information from women with breast cancer and women who did not have cancer.

Numerous women in all three groups had these symptoms. But in women diagnosed with ovarian cancer, these symptoms were much more common in the 3 months or so leading up to diagnosis.

Ovarian cancer patients were more likely than the other women to complain of gastrointestinal symptoms 1-3 months before their cancer was found. More than 8% of women with ovarian cancer reported these symptoms in that time period, compared to just 2% of women in the other groups. Likewise, 30% of women with ovarian cancer reported abdominal pain in the 3 months before diagnosis, while just 4%-5% of women in the control groups had this symptom. Other symptoms were also more common immediately before diagnosis: more than 5% of women with ovarian cancer had pelvic pain, and more than 16% had abdominal swelling, compared to fewer than 1% of women reporting these symptoms in the other groups. Some of these symptoms were more common in the ovarian cancer patients even as far back as 10-12 months before diagnosis.

When the researchers looked at the diagnostic tests the women had gotten, they found that ovarian cancer patients were more likely to get certain tests than women in the other two groups. However, some of the tests they got -- abdominal imaging and gastrointestinal procedures -- may miss ovarian cancer more often than other tests, though they are more likely to catch other, more common conditions that could explain these symptoms. Only about a quarter of the women with ovarian cancer had pelvic imaging (MRI, for instance, or CAT scans) and CA125 blood tests, which are more likely to find this disease.

Smith's study was not designed to look at outcomes, so it doesn't show how these women fared after their cancer diagnosis. There's no way to know if the women who did get pelvic imaging and CA125 tests had less advanced cancer or survived longer than the women who didn't. And because the study only looked at women with advanced ovarian cancer, it doesn't provide information about symptoms in women with early-stage disease.

Nevertheless, "it looks like some women could have had a potentially earlier diagnosis if they'd had the right tests," Smith said.

But that doesn't mean that all women who report these symptoms should automatically be tested for ovarian cancer, he emphasized.

"Women who get [ovarian cancer] are a very small part of the population," he explained. "But if a medical workup doesn't yield a diagnosis [of another disease] and the symptoms are still there, the workup shouldn't stop without ruling out ovarian cancer."

That's in line with what the American Cancer Society recommends, said Debbie Saslow, PhD, director of breast and gynecologic cancer at ACS. Women who have a combination of symptoms, who have persistent or worsening symptoms, or who are at high risk of developing ovarian cancer may be candidates for a pelvic ultrasound, manual pelvic exam, and the CA125 blood test, she said.

ACS does not recommend these tests for screening for ovarian cancer in women who have no symptoms of the disease and are not at high risk of getting it. That's because the tests are not accurate enough to be used in people without symptoms and because the cancer is uncommon. Smith also said that the pelvic imaging/CA125 tests are not appropriate for screening.

Another recent study, published in the Journal of Clinical Oncology, shows just why experts are reluctant to recommend ovarian cancer screening. Researchers from the United Kingdom found that ultrasound and the CA125 test were not effective at finding ovarian cancer at an early stage -- even in women already at high risk of developing the disease.

They studied more than 1,000 women at high risk either because of a family history of ovarian cancer or because they were known to have certain genetic mutations that make ovarian cancer more likely. The women were screened every year with either transvaginal ultrasound, or the CA125 blood test, or both.

Thirteen of the women developed ovarian cancer. The screening tests found 10 of these cancers, but only 3 of them were at stage I (the earliest stage) when found. Two of the cancers were at stage II, 4 of the cancers were already at stage III, and 1 was at stage IV (the most advanced stage). Most of the cancers were found by the blood test.

Of the 3 cancers that were not found by the screening tests, 2 were found because the women developed symptoms of ovarian cancer 4 and 12 months after having normal results on their screening tests. One of these women had stage IV cancer, and the other stage III. The last cancer occurred in a woman who had no symptoms and also had normal screening tests, but decided to get her ovaries removed. The cancer was found during that surgery; it was at stage III.

Twenty-nine women had false-positive results: Their screening tests indicated cancer, but when surgery was performed, no cancer was found. More than half of these women had not yet reached menopause.

The findings suggest that current screening tests are simply not effective in detecting ovarian cancer early, the researchers say. But a related editorial notes that more studies are needed to determine whether certain screening methods -- or schedules -- are better than others and more likely to find ovarian cancer early.

The number of cancers found in the new study was very small, the editorialist notes. And because the screening method wasn't always consistent -- some women only had ultrasound, others only had the blood test, and some women had both tests -- it isn't possible to say with certainty that screening doesn't work. Perhaps women must have both tests and have them more frequently in order to find ovarian cancer earlier.

One study that could help answer that question is already being done in the US. More than 2,200 women at high risk for ovarian cancer are being screened with the CA125 blood test every 3 months as part of the The Risk of Ovarian Cancer Algorithm (ROCA) Study. Another UK study is also under way, testing ultrasound and CA125 every year, with other blood tests in between.

Citations: "Ovarian Cancer: Can We Make the Clinical Diagnosis Earlier?" Published online August 22, 2005 in Cancer (print Vol. 104, No. 7). First author: Lloyd H. Smith, MD, PhD, of the University of California at Davis School of Medicine.

"Screening for Familial Ovarian Cancer: Failure of Current Protocols to Detect Ovarian Cancer at an Early Stage According to the International Federation of Gynecology and Obstetrics System." Published in the Journal of Clinical Oncology (Vol. 23, No. 24: 5588-5596). First author: Diane Stirling, Western General Hospital, Edinburgh, Scotland.

"Screening for Familial Ovarian Cancer: The Need for Well-Designed Prospective Studies." Published in the Journal of Clinical Oncology (Vol. 23, No. 24: 5443-5445). First author: Ian Jacobs, University College, London.