A Publication of the James Buchanan Brady
Urological Institute Johns Hopkins Medical Institutions
Volume V, Winter 2000
HEREDITARY PROSTATE CANCER
"Everybody has these genes. "The crucial difference is that not everybody
has them in the same order and "some particular sequences of these genes
greatly increase the risk of developing cancer."
Imagine that your great ambition is to work a magnificent, three-dimensional jigsaw puzzle. But before you can start to make sense
of the puzzle, or even put together a little segment of it, there's something you have to do first: Turn all the pieces right-side up.
"If there are four brothers, all with prostate cancer, and they've all inherited the same region of chromosome 1
from their father who has it, too-then that is no longer what you would predict by chance"
Welcome to the world of William Isaacs, Ph.D., who has spent much of the last decade testing his self-discipline with the greatest puzzle of all,
the human genome. He is searching for the genes involved in hereditary prostate cancer (see "What is HPC?" ); his initial tests took him through the
body's 23 pairs of chromosomes at the astonishing rate of about 10 million base pairs (genetic building blocks) at a time, and gleaned a handful of
promising sites that deserved further study.
After those giant strides, however, the next part has had Isaacs creeping at what seems a snail's pace--if his previous tests were the equivalent of
flying over an area, now he's conducting a house-to-house search, examining hundreds of genes, one at a time. He is working mostly in the dark:
He doesn't know how many genes he's looking for, and -- although he and colleagues have narrowed it down considerably -- he still isn't sure exactly
where to find them. Worse, nobody has ever looked at these genes before. So before he can even determine whether or not a gene looks promising, he must
first figure out what it is. Think of an explorer sailing in uncharted waters, facing a string of islands, investigating each one and producing a
topographical map. "Each one of these genes," he explains, "could be a Ph.D. disertation in itself, just to identify and characterize the gene, get
the structure and sequence it," not to mention comparing all of this new information in the DNA of hundreds of men with HPC and their families.
And yet, slowly but surely, this hard work is paying off: So far, Isaacs and colleagues at Hopkins and the National Human Genome Research Institute
have found good evidence that at least two of these defective genes do indeed exist: One is somewhere on Chromosome 1, and the other, most recently
discovered, ties on the X chromosome -- a milestone in cancer research, this is the first time the X chromosome (which sons inherit from their mothers)
has been definitively linked to a major cancer. Isaacs would love to start figuring out the role those aberrant genes play in the cascade of events
leading to prostate cancer. Instead, he is steeling himself to keep turning over those pieces -- in this case, the pieces are thousands of undiscovered
genes--until he can see the puzzle in its entirety. Although they haven't yet pinpointed the faulty genes on either chromosome, Isaacs and colleagues
have identified certain characteristics that suggest which mutation a family may have: In families with a mutated HPC1 gene:
- At least five men in the immediate family, or multiple men in multiple generations, have prostate cancer.
- The average age of diagnosis is younger than 65.
- There is evidence of father-to-son transmission.
In families linked to a mutated HPCX gene, the defect is always passed on from the mother: In this case, a father
cannot pass the disease to his son. (X and Y are the "sex" chromosomes, and they distinct pattern of inheritance.
Fathers always pass along the Y chromosome to their sons. A man inherits the X chromosome from his mother--so
every man has one X and one Y chromosome. Women have two X chromosomes.)
"We now know of at least two different methods of transmission of prostate cancer," says Isaacs, "from father to
son, and mother to son. Say a man has prostate cancer, so does his brother, and so does his father. Then right
away we can say that we don't think this is an X-linked family, because there's no way for a father to pass
on a mutated gene on the X chromosome." However, the father can pass it on to his daughters, who then can transmit it it to their sons-the father's
grandsons. In such families, prostate cancer might seem to skip a generation. "My sons cannot inherit a mutated
X gene from me, but they could from their mother," continues Isaacs. "A classic example of an X-linked prostate
cancer would be: My mother's brothers have prostate cancer. My mother can't be affected but she passes the gene
on to me and my brothers. We can't pass it on to our sons, but my sister inherited the gene; and she passes it on to
her sons. The only offspring of my generation that could have it would be my nephews."
On the other hand, HPC1 can be transmitted from either the mother or father. "There's definitely overlap," says
Isaacs. "When prostate cancer is passed on through the mother, it's got more possibilities: It could be either HPC1
or HPCX. But if it's passed through the father, that gives us a tip-off that it's not going to be an X-linked family." The
fact that men have only one X chromosome makes looking for that gene somewhat easier. "But everybody has two
copies of chromosome 1 -- one inherited from the mother and one from the father -- so they've one good copy and
one bad copy, and sometimes it's very difficult to know which one we're looking at."
Sad legacy: This research began with one man's sad family history, and owes its success to dozens more families
ravaged by prostate cancer. In the 1980s, Urologist-in-chief Patrick C. Walsh, M.D., noticed a disturbing trend in
his patients seeking surgical treatment for prostate cancer: They seemed to be getting younger. They also tended to
have a family history of the disease. In 1986, the case of one 49-year-old man whose family had been decimated
by prostate cancer crystallized the idea in Walsh's mind that some families must carry an increased risk of
developing the disease, Every male member of this man's family -- his father, his father's three brothers, and his
grandfather -- had died of prostate cancer. So many deaths in one family had to be more than just awful coincidence,
Walsh believed. That observation led to a series of studies by Hopkins researchers that defined and characterized
HPC, demonstrating the clear link between family history and a man's risk of developing prostate cancer. Several
years ago, Walsh and behavioral scientist Sally Isaacs screened a pool of 2,500 families that met the criteria for
HPC and selected the 79 that had been hardest hit. The families filled out detailed questionnaires about their health,
occupations and family history, and sent blood samples to Hopkins.
Then, Bill Isaac and his colleagues at the National Human Genome Research Institute began scrutinizing those
blood samples, using a technique called "linkage analysis", looking for patterns of inheritance of certain portions of
chromosomes that couldn't possibly be normal. They struck paydirt first with Chromosome 1. "We can follow each copy of
Chromosome 1 as it's passed along, say from father to son, and we can see whether the distribution -- instead of
being random -- becomes skewed, so that each affected sibling gets the same copy of Chromosome 1," says Isaacs.
"So if you look in a family and there are four brothers, all with prostate cancer, and they've all inherited the same
region of chromosome 1 from their father -- who has it, too -- then that is no longer what you would predict by chance."
This year, about 180,000 cases of prostate cancer will be diagnosed in
the U.S. But about 18,000 of those are in men born with a head start
-- one or more bad genes, such as HPC1 and HPCX, that greatly increase
their odds of developing cancer, and developing it at an earlier
And if the same thing happens in many families, this pattern becomes statistically significant. The problem is that,
although finding this region marks great progress in the search for the defective genes in prostate cancer, this
region of Chromosome 1 is still huge. It's like trying to pinpoint a specific restaurant in Hong Kong, and your first
step -- identifying the continent of Asia -- although helpful, still leaves much work to be done.
"The"gene: When Isaacs and other scientists speak of finding "the gene," they don't mean some weird-looking
mutant that instantly calls attention to itself as a cause of cancer: "Everybody has these genes. For example, I have
a BRCA- 1 gene, a BRCA-2 gene (involved in breast cancer). But everybody doesn't have the same sequence of those genes.
We're looking for genes whose sequence varies in the population, and some
particular sequences of these genes greatly increase the risk of developing cancer." The sequences fluctuate, like
ingredients in a recipe -- just as bread and salt dough have the same ingredients; depending on the recipe, one is
edible, and one is not.
An advantage of Isaacs' linkage approach is that it's completely objective -- a matter of statistics. "Right now, we've put
our blinders on," he says. "We don't want to rule out a gene because we
think we know what it does; there's no question you could miss genes on that basis. We don't assume anything
about the function of that gene, or how it may work. The only thing that determines whether or not it's "the gene" is
its sequence variation. We have no idea what the function is of most of the genes we're looking at. Someday we
will, but right now we don't really care about these novel genes. If they're not the gene we're looking for, we're not
going to take the time to figure out what they do. If we can demonstrate that we've found 'the gene,' then we'll get
busy and try to figure out everything we can about it. That's the reason we started all this."
But Isaacs does allow himself to speculate about what "the genes" might do: He suspects they will turn out to be
some sort of regulatory genes, in charge of specific steps -- such as repairing damaged DNA. "In colon cancer, if
you inherit a defect in your ability to repair single-base mismatches that occur either spontaneously, or due to
carcinogens, then your DNA accumulates a lot of mutations, and that greatly accelerates the process of getting
cancer. The same thing is becoming apparent in breast cancer, with BRCA- 1 and BRCA-2." Isaacs believes that,
eventually, maybe as many as a half-dozen different genes will turn out to play a role in HPC.
Fortunately, scientists from around the world are pooling their information on prostate cancer genetics, says
Isaacs, who welcomes the help. This teamwork has resulted in an International Consortium for Prostate Cancer
Genetics; Isaacs is the acting chairman. "We have over 800 families who have been collected by this group. By
having more families, we can narrow down the region that we have to search through to find the specific genes."
Isaacs also credits, from Hopkins, Sally Isaacs, Kathy Wiley, and Patrick C. Walsh, M.D from the University of
Maryland, Jianfeng-Xu, M.D., Ph.D.; from the National Human Genome Research Institute, Jeffrey Trent, Ph.D.,
team leader; from the Mayo Clinic, Steve Thibodeau, Ph.D.; from Umea University in Sweden, Henrik Cronberg,
M.D.; from Tampere University in Finland, Olli Kalioniemi, Ph.D.
for a Prostate Cancer Susceptibility Locus on the X Chromosome."
Xu, J; Heyers, D; Freije, D; Isaacs, S; et al. Nature Genetics,
Vol. 20: 175-179, 1998.
What is HPC?
That Bill Isaacs and colleagues now have two promising leads in their search for the genes involved in hereditary
prostate cancer (HPC is remarkable -- not just because of the magnitude of their task, but because it refutes what
many scientists argued for years: That prostate cancer, so common in older men, is simply a disease that comes
with old age.
Prostate cancer is very common; this year, about 180,000 cases will be diagnosed in the U.S. But about 18,000 of
those are in men born with a head start -- one or more bad genes, such as HPC1 and HPCX, that greatly increase
their odds of developing cancer, and developing it at an earlier age. Only about 10 percent of all cases of prostate
cancer are thought to be purely hereditary But Isaacs and colleagues believe that the defective gene or
mechanisms involved in hereditary cancer are the same ones that somehow go wrong in the far more common
"Sporadic" cancer, which just develops over the course of a lifetime. In most men, scientists believe, cancer
happens because of an unfortunate chain of events -- at least one genetic aberration, plus one or more things
environmental, such as a poor diet. Say it takes three "strikes" in order for cancer to begin: Being born with a faulty
gene might be worth one or two strikes; add a lifetime of eating the wrong foods (or no eating the right ones), and
An estimated 250,000 American men may carry one of these defective genes. HPC can be inherited from either
parent: Briefly, if your father or brother has prostate cancer, your risk is two times greater than the average
American man's , which is about 13 percent. It goes up from there: Depending on the number of affected relatives
you have and the age at which they develop the disease, your risk could be as high as 50 percent if you are in a
family that meets the criteria for HPC -- if you have at least three close relatives, such as a father and two brothers
affected, or two relatives if both were younger than 55 years old when the disease was diagnosed; or if your family
has disease in three generations -- a grandfather, father or brother. In HPC families, men should have a digital rectal
examination and PSA test every year, beginning at age 40.
If three or more members of your family have bad prostate cancer you may want to register your name with the
Johns Hopkins Hereditary Prostate Cancer Study. We will provide you with regular updates about our research
progress. To receive this free information, or to find out more about this work, write to the Johns Hopkins Hereditary
Prostate Cancer Study at The Brady Urological Institute, the Johns Hopkins Hospital Baltimore, Md., 21287-2101.