Statistically, the prime cause of endometrial cancer (EC) is obesity. Its incidence is three to 10 times greater in obese woman. A lesser but telling risk factor is nulliparity - never having borne a child.

"When colon cancer runs in a family, endometrial cancer may be suspect," said molecular geneticist Paul Goodfellow at Washington University in St. Louis.

"Endometrial cancer - cancer of the uterus," Goodfellow observed, "is the commonest gynecologic cancer and the fourth most common cancer in women. An estimated 39,300 women developed the disease last year," he noted, "and 6,600 died from it."

Goodfellow is director of the Siteman Cancer Center, Cancer Genetics Research Program, at the university. He is lead author of a paper in the current Proceedings of the National Academy of Sciences (PNAS), dated May 13, 2003. Its title: "Prevalence of defective DNA mismatch repair and MSH6 mutation in an unselected series of endometrial cancers."

"Our finding," Goodfellow told BioWorld Today, "is that a high-fraction frequency of women with uterine endometrium cancer have an inherited mutation in the DNA MSH6 repair gene. Heretofore," he continued, "its MSH6 gene had been seen as involved in a very small fraction of cases of familial cancer risk. Our research results suggest that just a minimum of seven in 441, or about one in 60, women developed uterine endometrial cancer. All studies of the MSH6 gene show it as part of a family of DNA repair genes. This scenario of cancer risk focused on families that had cancer concentrated in the older generation. These studies," Goodfellow went on, "suffered perhaps from the problem of looking under the lamppost where the brighter light is."

Mutated MSH6 Gene Causes DNA Repair Problems

"It had been a study of families that have features associated with mutation in other DNA repair genes - typically the colon-cancer connection. And interestingly enough, in those extraordinarily rare families with MSH6 mutations there seems to be an excess of endometrial cancer. So we turned it around and said, Well, what about endometrial cancers in general? What fraction of those are attributable to inherited mutations in the MSH6 DNA repair gene?' To that end we did some molecular tricks on a very large patient cohort, looking at their tumors. We didn't select for the patients' genetic history, but just as they came to the clinic. By studying molecular features of their tumors, we developed an index of suspicion that this group of women had a DNA repair problem. Therefore, MSH6 may account for that problem we observed in the tumor. And when we did molecular stratification of cases based on tumor alone - seven of 30 of the high-risk patients actually had a mutation.

"Then we started to look at features associated with genetic disease," Goodfellow recounted. "Our cohort of 441 women was unselected for age of onset or family history. The seven who had mutations had a mean age of onset almost a decade younger than the rest of the population. Two of the seven women," Goodfellow recalled, "had a second malignancy that's associated with DNA repair defects. One of the women incurred a colon cancer six or seven years after her endometrial malignancy. Another one had a synchronous ovarian carcinoma. Those additional data suggest that these molecular features of a tumor can point to the group that is genetically at risk.

"Although not included in the paper," he went on, "we've continued our studies by going to the at-risk family members, and have demonstrated that, of course, the relatives are at significant risk for cancer. But it's not the kind of cancers and family histories that have been under the lamppost. They're not the colon cancer-prone families that people have focused on in the past for DNA repair gene mutations.

"You can't take a clinical genetic history in your routine gynecologic service," Goodfellow pointed out. "Not every woman that has breast cancer has a genetic history taken. And it's not been done for patients with endometrial cancer. Until recently, the heritability of EC was not appreciated. We identified mutations in naturally existing populations - women who came to medical attention because they had an initial uterine cancer. They are all inactivating mutations - an admixture of insertion, deletion or point mutations.

"This MSH6 gene was known and linked to cancer risk in studies of familial disease, so-called colon cancer-prone families. It was really a candidate for investigation in those families that didn't have mutations in the likely suspects. And the mutation rate was actually shockingly low in that group of patients - referred to as hereditary non-polyposis colorectal cancer. MSH6 encodes a protein that participates in the recognition of mismatches in newly synthesized DNA. So we have a human genome of 3 billion base pairs. And every time a cell divides it's important that all 3 billion base pairs be copied with high fidelity. Our DNA synthesis machinery, like all biologic systems, is error prone. MHS6 and its molecular partners scan the newly synthesized DNAs and look for differences between the starting material and the new DNA sequence. If they recognize a difference, they can recruit additional enzymes to take out the bad stuff and replace it with a good copy, exactly near the parent DNA."

Only Lately Endometrial Cancer Seen As Heritable

"The colon-endometrial association may have to do with the type of mutations that are naturally occurring," Goodfellow suggested. "Of course, like most cancers, the tumors evolve through mutations taking place in cells that divide to repopulate. The endometrium is a tissue that gets rebuilt every month during a woman's reproductive menstrual life. With every cell division there's a finite probability of getting a bad mutation. We replace the lining of our colon very frequently. All that cell division sets you up for the accumulation of mutations and ultimately the development of disease - namely carcinoma.

"Because it's only recently recognized that endometrial cancer can be inherited, there's been little follow-up for the relatives who might be at risk for endometrial and other malignancies.

"I'm sure the companies that are doing DNA repair diagnosis," Goodfellow observed, "will begin to include our molecular tricks to help tip the odds of getting to the gene and mutations in the most economical way - saving money for everybody. This is a clear tumor pattern associated with the genetic makeup of the patient. I think we always benefit for the better proof of principle," he concluded, "as we move forward with molecular pathology."