In This Report Introduction Illustration Effective Treatments The New Colostomy Other Colon Woes Key Weapons: Diet and Exercise Resources

By Rick Weiss
Washington Post Staff Writer
Tuesday, February 10, 1998; Page Z06

Stephan Spitzer was shocked in 1991 when his 37-year-old brother was diagnosed with colon cancer. "It came out of the blue," said Spitzer, a 45-year-old computer programmer living in Brookville. "They found it pretty advanced. He had a year of chemotherapy, then he died."

The tragedy of his younger brother's death raised a host of difficult questions for Spitzer, who, like most people his age, had never thought much about colon cancer -- much less about its genetics.

Did his brother's death mean that Spitzer too was at increased risk from colon cancer? Was there any way for Spitzer to find out whether he carried a potentially deadly gene? Most worrisome: Were Spitzer's son and daughter, then 10 and 7, at increased risk from the disease? And if so, what could they do about it?

Five years ago, doctors could provide few answers to those questions. But since then, scientists have discovered several of the key genetic mutations that can lead to colon cancer, and a number of research institutes and companies now offer tests for those mutations. Two months ago, Spitzer made the decision to take a newly devised blood test that detects a mutation that predisposes some people to colon cancer. His story illustrates the rapidly advancing world of colon cancer genetics, where scientists are still sorting out the risks, benefits and lingering uncertainties of genetic testing.

Genetic vs. Inherited

Like all cancers, colon cancer is a genetic disease. The uncontrolled proliferation of cells that causes tumor growth always starts with a glitch in the genetic machinery of a single cell.

Sometimes that tiny glitch occurs in a gene that normally puts the brakes on cell division. When that brakeman is damaged for one reason or another, the cell starts dividing uncontrollably. The resulting mass of cells is a tumor.

In other cases, colon cancer arises because there is a breakdown in a gene that normally repairs other damaged genes. When that repair gene is damaged and can no longer do its job, molecular errors start to accumulate within that cell's genetic material. Eventually, one of those uncorrected errors may strike a gene that regulates cell division. Once again, a cancer is born.

Everyone is at some risk of having a colon gene go bad; like everything else in the body, genes break down sometimes. That's why about 5 or 6 percent of the people in this country can expect to get colon cancer at some point in their lives.

The question that concerns people like Spitzer is whether a relative's cancer began because of a new, spontaneous genetic error that happened in that person's body some time after birth, or because of a preexisting glitch that was inherited from a parent and may also have been inherited by other family members.

A family tree can provide some answers. If a bad gene were being passed from parent to child to grandchild, a clear pattern of colon cancers would soon emerge. But many people don't know much about what killed their relatives, especially going back more than a generation or so.

What's more, most genetic errors that predispose to colon cancer do not guarantee that a cancer will develop. Genetic mutations raise the odds. But whether a cancer will actually appear often depends on a range of other factors, including diet, general health, and the condition of other genes that interact with the damaged gene.

As a result, it can be difficult to tell whether a case of colon cancer is a sporadic case or is part of a family history of cases. Indeed, scientists are still not sure what percentage of all colon cancers are due to inherited factors and what percentage are due to new mutations in individuals. Some estimate that as many as 45 or 50 percent of all colon cancers have an inherited component; others say it's more like 10 percent.

Gradually, however, researchers have begun to pin down the specific inherited genetic mutations that have the greatest impact on one's risk of colon cancer.

Familial adenomatous polyposis (FAP) is an inherited condition characterized by the appearance of thousands of polyps, or finger-like projections, in the colon. These polyps are not cancerous, but they are inclined to become cancerous over time. Virtually everyone with FAP develops colon cancer at some point, sometimes as early as their teens or twenties and almost always by age 40. Moreover, FAP colon cancers tend to be rather aggressive; more than half of affected people will die from their disease.

Doctors first recognized FAP as an inherited cancer syndrome in the 1880s, but it wasn't until 1991 that scientists discovered the culprit gene. That gene, called APC, is a so-called tumor suppressor gene -- a brakeman gene. When damaged, it leads to unrestrained cell division.

So high is the risk of death from FAP that children known to have inherited the mutation must have their colon examined annually for the first signs of polyps, starting when they are still in elementary school.

"For these kids the surveillance starts around age 10 or 11, and often we do surgery by their early teens," said Kenneth Offit, chief of the clinical genetics service at Memorial Sloan-Kettering Cancer Center in New York. "Prophylactic surgery [even before cancer is found], which is controversial for other entities like breast cancer, is not only not controversial for FAP," Offit said, "but is the recommended policy."

Hereditary non-polyposis colon cancer (HNPCC), also known as Lynch syndrome, is the most common form of inherited colon cancer and accounts for at least 5 percent of all colon cancers. While it rarely strikes as early as FAP, it is characterized by a relatively early age of onset; on average, the cancer appears at age 44, as compared to 66 for sporadic colon cancers.

While polyps usually take up to 40 years to become cancerous, HNPCC polyps may do so within three to five years, said Bert Vogelstein, the Johns Hopkins geneticist who has done much of the recent groundbreaking work on colon cancer genes. The majority of cases involve the highest reaches of the colon (known as the "proximal" or "right-sided" colon), which means that only an exam of the entire large intestine with a colonoscope (as opposed to the shorter sigmoidoscope) is likely to reveal the cancer.

About half of people with HNPCC get additional colon tumors within 10 years, and all are at increased risk of several other cancers. Women with HNPCC, for example, are at substantially increased risk of endometrial and ovarian cancer. And both men and women with HNPCC must be on the lookout for stomach, small intestine, kidney and other cancers.

The 19th-century Michigan pathologist Aldred Warthin first noted the syndrome when he asked his seamstress about her depression and learned that she felt certain she would someday succumb to cancer of the uterus or bowels, since that was the fate of so many in her family. Sure enough, she died of endometrial cancer while still in her youth. In another heavily studied HNPCC family, the patriarch was an alcoholic who told doctors he didn't care to stop drinking because "everyone in the family dies of cancer."

In 1993, 98 years after Warthin's first description, U.S. scientists linked the syndrome to a mutated version of a DNA repair gene called MSH2. Since then, researchers have found three other DNA repair genes (MLH1, MSH6 and PMS2) that, when mutated, can lead to HNPCC.

Another colon cancer syndrome -- one that apparently affects only Jews of Eastern European, or Ashkenazi, descent -- was discovered by researchers at Johns Hopkins University last August. Like FAP, this variant of colon cancer is caused by a mutation in the APC gene, but it does not lead to the "carpets" of polyps typical of FAP and does not lead to cancer as surely as does the FAP mutation. Relatively little is known about this syndrome, but preliminary studies suggest that people who inherit this mutation have about twice the average odds of developing colon cancer.

To Test or Not To Test

The availability of these genetic tests has raised difficult scientific and ethical questions. How reliable are they? Who really stands to benefit from them? Are there some people to whom genetic information may be more of a psychological burden than a help, either because the information provided will be too vague to be helpful or because it tells them more than they want to know? And are there cases when the potential benefits of knowing one's genetic makeup are outweighed by the risk that insurers and employers may use that information as a discriminatory tool?

The first step in deciding whether to get a genetic test is to take a very close look at one's family tree, said Vogelstein. Imperfect a tool as it is, he said, family history is often underutilized.

"Many physicians are unaware on a practical level of how important the family history is, especially since the genetics are still so new," Vogelstein said. "It's very important to investigate who in the family, if anyone, has contracted cancer. In general, that information is as valuable as any of this fancy, sophisticated genetic information."

That is often best done with the help of a genetic counselor, a specialist trained in recognizing and interpreting familial medical patterns. "There's an art to taking a family history," said Jill Brensinger, a genetic counselor at Hopkins's Hereditary Colorectal Cancer Clinic and Registry. "When you ask the questions right, bells start to ring."

In some cases, that information can lead to an easy decision about genetic testing. In families with a clear history of FAP, for example, scientists know that each child has a 50 percent chance of having inherited the mutation. Without the gene test, all children must be considered at risk and submit to exams from an early age, enduring all the expense and discomfort of regular testing.

But if a child tests negative, "you know you're home free," said Gloria Petersen, an epidemiologist at Hopkins's School of Hygiene & Public Health and a professor of oncology at the medical center.

That doesn't mean that the test isn't misused. A study last year found a disconcertingly high rate of inappropriate testing, inadequate counseling and misinterpretation of results in a population of 177 patients who underwent APC testing.

The study, led by Johns Hopkins gastroenterologist Francis M. Giardiello, found that nearly one-fifth of those tested were not actually at risk of the disease so should not have been tested, and only one-fifth of the group received genetic counseling before the test. Most disturbing, doctors misinterpreted the results in nearly one-third of the cases.

"Of particular concern is that some patients at risk for familial adenomatous polyposis would have been given a false negative result," the researchers wrote in the March 20, 1997, issue of the New England Journal of Medicine.

Still, explained Giardiello, when done correctly, genetic testing can change lives drastically for people with a family history of FAP. "It used to be that the only way to do surveillance on children was by performing endoscopies every year. There's a cost to that, and discomfort. Now we can genetically test these children," and those who test negative can avoid the exams until they are 50, when colon cancer becomes a risk for all.

For other colon cancer syndromes, such as HNPCC, the risks and benefits of testing are more difficult to assess. For one thing, doctors disagree about how many cases of colon and other cancers relatives must have to classify as an HNPCC family. A fairly strict definition, called the Amsterdam criteria, requires a minimum of three cases of colon cancer (at least two of which must be in one's parents, children or siblings) in at least two successive generations with at least one case occurring in a person younger than 50 years old. Other criteria give more credit for the appearance of other, related cancers in the family tree, such as endometrial cancer, which is often associated with the syndrome.