Deadly Perfectionism?
When a healthy woman complains to her doctor about abdominal bloating, and hears that she has a terrifying malignancy with little chance of long-term survival, she can only feel the war on cancer has passed her by. She's right--she and tens of thousands of other women stricken with ovarian cancer each year. Without a screening test to call its own, this cancer sows its malignant seeds undetected, weeping cancer-laden fluid throughout the pelvis and abdomen. What's troubling is that the tests that might spot this cancer at an early and potentially curable stage are out there but are deemed not yet good enough.
Last week offered another frustrating example. Reported in the Proceedings of the National Academy of Sciences, researchers identified a novel pattern of four proteins that signaled ovarian cancer in 95 percent of women with the disease and was normal in 95 percent of healthy women. Lance Liotta, a pioneer in the biomarker field who also developed an ovarian cancer blood test at the National Cancer Institute using new protein-analyzing technology, sees blood as a treasure-trove of protein signatures for cancer detection. Since these signatures reflect the unique genetic makeup of specific tumors, they offer hope for other cancers that are also difficult to detect until too late, such as pancreatic and lung.
The main reason promising tests for ovarian cancer are stuck in the laboratory is that they are being held to standards unseen in the other cancer screens now widely in use: a specificity of 99.6 percent. That is, near perfect with false positives of less than 1 percent. Were such perfection applied to Pap smears for cervical cancer, mammograms for breast cancer, or PSA s for prostate cancer, these screens would not exist. But they seeped into medical practice years ago. PSA, for example, was approved by the Food and Drug Administration in 1986 to monitor prostate cancer, but doctors then began using it for screening, too. Patients should know that these studies are limited, typically detecting cancer only 75 to 85 percent of the time, with false positives as high as 30 percent. But they are keen enough to tell a physician that cancer may be lurking and, when appropriate, to seek further evaluation. The real question is how these tests are used, and for the most part they have been used well. Indeed, much of the success of the war on cancer is theirs.
The reason the bar has been set so high for an ovarian cancer test is that this tumor is uncommon, about 1 in 2,500 post-menopausal women. With 5 percent false positive rates, 125 healthy women would have sonograms and perhaps minor surgery--without proof that early detection saves lives. This is deja vu all over again. Recall the battles over women having access to mammograms in their 40s? The U.S. Preventive Services Task Force and a 1997 National Institutes of Health consensus conference dismissed their use with the exact same reasoning: 2,500 younger women are screened to find about 5 cancers, but with false positive rates of 10 percent, some 250 women without disease face anxiety, more studies, and sometimes a biopsy. The cost-crunchers see the low yield as wasting money better spent elsewhere. But in 2002, after years of debate, political outrage, and research that showed women's lives were being saved, annual mammograms for 40-something women became the standard. Doctors who'd been ordering them and women who'd demanded them were vindicated. And the incidence of ovarian cancer in post-menopausal women is about as "rare" as breast cancer is in women who are in their 40s.
advertisement
