For the first time, researchers have found genetic material from uterine or ovarian cancers in Pap smears, meaning that it may become possible to detect three diseases with just one routine test.
But the research is early, years away from being used in medical practice, and there are caveats. The women studied were already known to have cancer, and while the Pap test found 100 percent of the uterine cancers, it detected only 41 percent of the ovarian cancers. And the approach has not yet been tried in women who appear healthy, to determine whether it can find early signs of uterine or ovarian cancer.
On the other hand, even a 41 percent detection rate would be better than the status quo in ovarian cancer, particularly if the detection extends to early stages. The disease is usually advanced by the time it is found, and survival rates are poor. About 22,280 new cases were expected in the United States in 2012, and 15,500 deaths. Improved tests are urgently needed.
Uterine cancer has a better prognosis, but still kills around 8,000 women a year in the United States.
These innovative applications of the Pap test are part of a new era in which advances in genetics are being applied to the detection of a wide variety of cancers or precancerous conditions. Scientists are learning to find minute bits of mutant DNA in tissue samples or bodily fluids that may signal the presence of hidden or incipient cancers.
Ideally, the new techniques would find the abnormalities early enough to cure the disease or even prevent it entirely. But it is too soon to tell.
“Is this the harbinger of things to come? I would answer yes,” said Dr. Bert Vogelstein, director of the Ludwig Center for Cancer Genetics and Therapeutics at Johns Hopkins University, and a senior author of a report on the Pap test study published on Wednesday in the journal Science Translational Medicine. He said the genomes of more than 50 types of tumors had been sequenced, and researchers were trying to take advantage of the information.
Similar studies are under way or are being considered to look for mutant DNA in blood, stool, urine and sputum, both to detect cancer and also to monitor the response to treatment in people known to have the disease.
But researchers warn that such tests, used for screening, can be a double-edged sword if they give false positive results that send patients down a rabbit hole of invasive tests and needless treatments. Even a test that finds only real cancers may be unable to tell aggressive, dangerous ones apart from indolent ones that might never do any harm, leaving patients to decide whether to watch and wait or to go through surgery, chemotherapy and radiation with all the associated risks and side effects.
“Will they start recovering mutations that are not cancer-related?” asked Dr. Christopher P. Crum, a professor at Harvard Medical School who was not involved in the research.
But he also called the study a “great proof of principle,” and said, “Any whisper of hope to women who suffer from endometrial or ovarian cancer would be most welcome.”
DNA testing is already performed on samples from Pap tests, to look for the human papillomavirus, or HPV, which causes cervical cancer. Dr. Vogelstein and his team decided to try DNA testing for cancer. They theorized that cells or DNA shed from cancers of the ovaries and the uterine lining, or endometrium, might reach the cervix and turn up in Pap smears.
The team picked common mutations found in these cancers, and looked for them in tumor samples from 24 women with endometrial cancer and 22 with ovarian cancer. All the cancers had one or more of the common mutations.
Then, the researchers performed Pap tests on the same women, and looked for the same DNA mutations in the Pap specimens. They found the mutations in 100 percent of the women with endometrial cancer, but in only 9 of the 22 with ovarian cancer. The test identified two of the four ovarian cancers that had been diagnosed at an early stage.
Finally, the team developed a test that would look simultaneously for cancer-associated mutations in 12 different genes in Pap samples. Used in a control sample of 14 healthy women, the test found no mutations — meaning no false-positive results.
Dr. Luis A. Diaz, the other senior author of the report and an associate professor of oncology at Johns Hopkins, called the research a step toward a screening test that at first blush appears very effective at detecting endometrial cancer, though obviously less so at finding ovarian cancer.
“Probably one of the most exciting features of this approach,” Dr. Diaz said, “is that we wanted a test that would seamlessly integrate with routine medical practice that could be utilized with the same test that women get every day all over the world, the Pap smear.”
But, he added: “We can’t say it’s ready for prime time. Like all good science, it needs to be validated.”
He and other members of the team said it might be possible to improve the detection rate for ovarian cancer by looking for more mutations and by changing the technique of performing Pap tests to increase the likelihood of capturing cells from the ovary. The change might involve timing the test to a certain point in a woman’s monthly cycle, using a longer brush to collect cells from deeper within the cervix or prescribing a drug that would raise the odds of cells being shed from the ovary.
The technique also needs to be tested in much larger groups of women, including healthy ones, to find out whether it works, particularly at finding cancers early enough to improve survival. And studies must also find out whether it generates false positive results, or identifies cancers that might not actually need to be treated.
Michael H. Melner, a program director in molecular genetics and biochemistry for the American Cancer Society, called the research “very promising,” in part because it is based on finding mutations.
“It tells you not just that cancer is there, but which mutation is there,” Dr. Melner said. “As we learn more and more about which mutations are associated with more or less severe forms of cancer, it’s more information, and possibly more diagnostic.”
