The websites of the National Cancer Institute and leading medical centers such as the University of Pennsylvania say cancer that starts in the fallopian tubes is ultrarare, making up less than 1 percent of all gynecological cancers.

But that assertion may be wrong. Mounting evidence points to the fallopian tubes as the origin of the fifth leading cancer killer of American women: ovarian cancer.

The implications are huge, researchers say. It could explain why ovarian cancer is so tough to detect while still confined to the ovaries, and why the disease usually has spread far beyond the ovaries when symptoms develop.

Lucy Gilbert, a gynecological oncologist at McGill University in Montreal, said it also makes sense structurally. The fallopian tubes - two thin ducts that connect the ovaries to the uterus - have feathery fringes called fimbriae "that drop cancer all over the abdominal cavity, getting it all over the place fairly early."

This new model may eventually reshape the prevention, detection, treatment, and even the name of ovarian cancer, expert say.

Already, physicians such as Robert A. Burger, a Fox Chase Cancer Center gynecological oncologist, have adjusted their prevention advice. They urge women who have finished having children and need abdominal surgery for any reason to have their fallopian tubes removed at the same time. (By using in vitro fertilization, women who are still ovulating can get pregnant even without fallopian tubes.)

"I've been talking about this for the last 15 years," said Burger. "I think the model makes perfect sense for high-grade, serous ovarian cancers. Those are the ones that tend to be the most lethal, and the most genetically unstable after diagnosis."

High-grade ovarian cancer is also the most common. Each year, it accounts for 75 percent of the 22,200 new cases - and 90 percent of the 14,000 patients who die.

When an egg is released in an ovary, the fimbriae flap to help guide the egg into the fallopian tube, where it may be fertilized by sperm, and then go on to the uterus.

The first real-life, real-time study to show that the fallopian tubes seed ovarian cancer was led by Gilbert at McGill.

The study, published last year, was designed to see whether ovarian cancer could be diagnosed while still confined to the ovaries if women promptly got gynecological tests to evaluate complaints - such as bloating, abdominal pain, and feeling full - that have been linked to the disease. Usually, women spend months being checked for digestive problems, a more common cause of such symptoms.

The 1,455 participants in the study underwent the standard diagnostic tests: a vaginal ultrasound to image the ovaries, and a blood test to measure CA125, a protein that rises with ovarian cancer as well as in benign conditions such as fibroids.

The researchers did indeed find cancer somewhat earlier - but not where they expected it to be. Surgery and tissue analysis revealed that seven of the nine women with high-grade cancers had little or no disease in their ovaries; their malignancies began in the fallopian tubes or the lining of the abdominal cavity, or both.

This explained why repeated tests found climbing CA125 levels, while their ovaries appeared normal on ultrasounds.

"The premise on which we based our early detection program was wrong," Gilbert wrote.

Other evidence implicating the fallopian tubes is more circumstantial.

For example, at Dana-Farber Cancer Institute in Boston, scientists knocked out the function of a BRCA gene in a rodent model. The animals developed high-grade fallopian tube cancer that led to ovarian cancer - mimicking what has been seen in women who carry BRCA mutations.

"In women with a BRCA mutation, we knew ovarian cancer could be a fallopian tube cancer, but we thought it was unique to this group," Gilbert said.

Better understanding ovarian cancer's origins is only the first step toward better ways to fight it, experts agree. CA125 is not recommended to screen asymptomatic women because it is unreliable, but a better marker has not been found.

One promising technology being developed at the University of British Columbia involves a minimally invasive way to check the fallopian tubes for "autofluorescence." Loss of this naturally emitted light appears to be a sign of abnormal tissue changes.

"We have a long way to go," Gilbert said, "but at least we know what we're dealing with."