New method targeting hypermethylation detects ovarian cancer in the blood

Sep 16 2004

Fox Chase Cancer Center researchers have found a new way to detect ovarian cancer in the blood. Reported in the Sept. 15, 2004, issue of Cancer Research, the new method targets hypermethylation - one mechanism used by cancer cells to turn off genes that protect against tumor development.

When these tumor-suppressor genes are inactivated by hypermethylation, they cannot do their job, which then allows cancer cells to develop. This research marks the first time hypermethylation has been examined for the detection of ovarian cancer.

Fox Chase molecular biologist Paul Cairns, Ph.D., and his colleagues tested for hypermethylation of BRCA1 and RASSF1A, two genes strongly associated with ovarian cancer.

"In normal cells, BRCA1 and RASSF1A are unmethylated, meaning they are able to do their job. We found these genes to be frequently hypermethylated in the blood and peritoneal fluid from patients with ovarian cancer," explained Cairns.

Tumor samples, preoperative blood and peritoneal fluid DNA were obtained and matched from 50 patients with ovarian or primary peritoneal cancer. The blood from 20 healthy age-matched women, normal ovary tissue from 10 women, and tissue, blood and peritoneal fluid from 10 women with benign ovarian cysts were used for controls.

Thirty-four of the 50 tumors (68 percent) showed hypermethylation of one or both genes. The remaining 16 tumor samples, which did not show hypermethylation for RASSF1A or BRCA1, had hypermethylated forms of other tumor-suppressor genes: APC, p14, p16 and DAP (death-associated protein-kinases), which provided a target for screening.

An identical pattern of gene hypermethylation was found in the matched blood DNA from 41 of 50 patients (82 percent sensitivity), including 13 of 17 cases of stage I disease. No hypermethylation was observed in the non-cancerous tissue, peritoneal fluid or blood from control samples (100 percent specificity).

"We were keenly interested in the potential of hypermethylation for the detection of early ovarian tumors," said Cairns, "so we tested an additional 21 stage I tumor DNA samples." Twenty of the 21 showed hypermethylation of at least one of the six genes in the panel.

"Hypermethylation is a common event in these tumors and appears to happen relatively early because we found it in early-stage cancers," said Mitchell Edelson, M.D., a gynecological oncologist in the department of surgical oncology at Fox Chase and a co-author of the study.

"What's encouraging about this research is that there were no false-positive results. A gene negative for hypermethylation in the tumor DNA was always negative in the matched serum or peritoneal fluid DNA.

"There is no other obvious body fluid available for ovarian cancer testing except the blood," Edelson added. "In the past couple of years, we have seen detection attempts with proteomics and now with this method. It is encouraging to see so much activity regarding one of the most feared diseases for women."

The potential application of this finding could be significant because there is no reliable screening method available for ovarian cancer. Most women are diagnosed with late-stage disease after developing symptoms. The cure rate for women with advanced ovarian cancer is low, but doctors are often successful in treating early-stage disease.

"It would appear that hypermethylation is a very specific marker for this disease, and if confirmed in larger studies, methylation may be useful in ovarian cancer diagnosis," said Cairns.

In earlier research, Cairns and his colleagues have demonstrated the ability to detect prostate and kidney cancer by targeting hypermethylation.