Thanks, Elizabeth . I would really like to thank the organizers. Like Elizabeth I am thrilled to see epidemiology take a lead position in the symposium. This morning I would like to talk to you about the causes of testis cancer and no, this is not my high school graduation photo for those of you who asked, but thank you for asking. What this is many of you may recognize is this is a very famous painting of the Myth of Oedipus, and you remember that the sphinx was terrorizing the people of Thebes . The sphinx was a monster with the body of a lion and the head of a woman and she had a riddle, and the riddle was what goes on four legs in the morning, two legs at noon and three at night, and for everyone who couldn’t answer the riddle the sphinx devoured them. Well, Oedipus could answer the riddle, and the answer was man because in the morning he crawls as an infant. At midday, youth, he walks on two legs, and when he is older he walks with the aid of a cane, and upon hearing this the sphinx furiously killed herself and threw herself off a cliff ending the reign of terror in Thebes . Why am I telling you this? I am telling you this because the clues from the descriptive epidemiology of testis cancer are very much like the sphinx’s riddle and I absolutely believe that if we could frame the answer to that riddle we could eliminate scores of testis cancer and what I would like to do this morning is simply to remind you in the first half of the talk what that riddle is and then in the remaining 7 minutes or so give you a hypothesis about what I think could answer those questions.
Now, you all know that testis cancer is the commonest cancer among young men, but the epidemiology of testis cancer shows some very unusual features, and the first unusual feature that oncologists and urologists all know is that testis cancer peaks at the age of around 25 to 35. So, most cancers go up logarithmic with age. Testis cancer does not. It is a disease of young men. Also, rates vary 10-fold even within Europe , even with countries that are closely related to one another geographically and culturally. Now, since the 1940s the incidence rate for testis cancer has gone up 2 percent each year. The result of this is that incidence rates for testis cancer double every 20 years.
These are some data from my colleague and friend Heinrich Moller now in London , and they show two points. One is there is this dramatic curve where the peak occurs in young adulthood and the second is that for countries that are fairly closely related to one another there is enormous variation in the shape of this age incidence curve with the highest rates in the world found in Denmark and they are five times higher than they are in Finland.
What do we know for sure about testis cancer? Ask any urologist, and he will say, “Cryptorchidism is a risk factor for testis cancer.” Oh, sure, it is but 95 percent of cases of testis cancer do not occur among boys with testicular maldescent. So, cryptorchidism occurs and is a risk factor with testis cancer but the vast majority of testis cancer is not caused or associated with cryptorchidism.
There has got to be something as Hamlet would say, rotten in Denmark because Denmark has the highest incidence of testicular cancer in the world. This rate as you just saw is five times higher than in nearby Finland .
Affluence. Testis cancer is a disease of the affluent. Also, men with reduced fertility, men seen at infertility clinics have a two-to-threefold increased risk of testicular cancer later in life and it is not that the testis cancer caused their infertility because they clinically had no testis cancer at the time. So, some factor causes both infertility and testis cancer.
The last clue, and it is very tantalizing is this. The Danish cancer registry is among the best in the world and examination of those data by Heinrich Moller have shown that the risk of testis cancer has gone up as I mentioned 2 percent each year since the 1940s except during World War II. Boys born during the years 1940 to 1945, 30 years later had a significantly reduced risk of testis cancer. This clearly suggests that whatever the cause of testis cancer is it may be associated with goods and provisions that were restricted during the German occupation of Denmark and Norway .
What have people thought causes testis cancer? There is a large range of hypotheses and they are mostly that. They range from in utero exposure to endogenous estrogens, environmental estrogens, to sedentary life style and tight underwear. I don’t think I am going out too far on a limb to say that none of these hypothesis have produced convincing data, certainly none have produced data that would allow us in any way to reduce the incidence of testicular cancer which I say is still increasing every year.
So, the heart of the question is this, and I am dead serious about the idea of the sphinx’ riddle. Here is modern day Oedipus and here is the sphinx, except her riddle now is not what goes on four legs, two legs and three legs but what is one, a carcinogen, two, high in Northern Europe especially in Denmark; what is increasing over time with the exception of World War II in Scandinavia? This factor is associated with affluence and it causes infertility.
Now, whatever you may think is the answer whether you agree or disagree with what I am about to suggest to you the answer to this riddle is the heart of the question. So, if you can come up with something that satisfies these five criteria like Ross Perot, I am all ears.
What I want to suggest to you now and I offer this as a hypothesis in the hope of stimulating some discussion today is that there is something that will satisfy these criteria.
A couple of years ago we wrote a paper where we suggested that ochratoxin A, a mycotoxin that is a widespread contaminant of food would satisfy these criteria.
How is that going to work? Remember there are five things,but before we get to them what in the world is ochratoxin A anyway? Ochratoxin A is a mycotoxin. It is a toxin produced by the molds Aspergillus and Penicillium. It is found in moldy grains and in animals that consume moldy grains, mostly pigs.
The main products contaminated with ochratoxin A are cereals, pigs, pork and coffee. In Canada 36 percent of the slaughterhouse pigs have significant ochratoxin contamination of their blood and in fact human exposure to ochratoxin is also widespread.
In fact, in Denmark examination of Danish blood supplies shows that 50 percent of human blood samples in Denmark are contaminated with ochratoxin.
Is ochratoxin going to meet those five tests? What is the first test? It must be a carcinogen. Ochratoxin is absolutely a genotoxic carcinogen. If you feed mice and rats ochratoxin you will produce kidney and liver tumors.
Interestingly many urologists will recognize this substance because ochratoxin is the same substance that produces an unusual nephropathy in pigs known incidentally as Danish nephropathy because it was identified in Denmark as a major cause of agricultural losses in the pork industry.
Danish nephropathy is probably the veterinary equivalent of the human disease Balkan endemic nephropathy, also, known as Balkan uropathy. This is a tubular interstitial disease of the kidney, a fatal disease and one-third of patients dying with Balkan uropathy have tumors of the urinary tract. So, ochratoxin is for sure a carcinogen and it is a urinary tract carcinogen.
The second test, distribution high in Denmark , ochratoxin contamination of foods is especially high in Northern Europe . This is because the conditions that foster the growth of mold in crops are what produces ochratoxin. Countries like Denmark , Sweden and the British Isles and Germany are the prime places for ochratoxin to grow.
As I mentioned ochratoxin is a contaminant of rye and pork. Denmark is the largest per capita consumer of these products in the world, also, the highest mean value for ochratoxin in blood actually has been reported in Denmark where of 144 samples 50 percent of them were positive for ochratoxin and they are positive for ochratoxin at a value within the same range of the porcine data in pigs in Canada. So, human exposure to ochratoxin is widespread and significant.
A third factor, which is increasing prevalence over time with the exception of the war years, well, this trend is actually quite consistent with trends in the consumption of pork. During 1960 to 1994, those 30 years meat consumption increased at a rate of 2 percent a year. This was actually largely pork consumption, the other white meat.
I would say that these data are also consistent with the idea that something, whatever it is should be reduced during the war years because during the occupations of Norway and Denmark meat, meat and coffee but meat especially was in very scarce supply.
A fourth clue was high socioeconomic status. Now, this is likely, I would think, to reflect a higher meat consumption of more affluent women and the possibility that these women breast feed much more frequently.
Breast milk is highly associated with higher socioeconomic status for the obvious reason that women who work have a difficulty breast feeding.
Breast milk is heavily contaminated with ochratoxin so that in Norway one-third of all the breast milk samples are contaminated with ochratoxin and the converse, there was no ochratoxin in infant formula.
The fifth which is actually quite interesting, is the idea of reduced fertility. If you inject ochratoxin into rats you will cause cytolysis of the seminiferous epithelium.
If you feed ochratoxin at clinically relevant doses to mice you cut their sperm count in half and if you treat these animals early in life, weanling rats fed diets that are naturally contaminated with ochratoxin they develop testicular hyperplasia and produce no mature sperm and this is interesting to clinicians I would hope because it is commonly observed that men with testicular cancer have slightly hypoplastic testes even among the non-affected gonad.
Well, I hope you find that interesting. It was interesting to us because my feeling was how could you do it in testis; it is very hard to take any further test in this just yet because ochratoxin has been measured by very, very different techniques in different countries and they can’t be compared.
However, what you could do is you could compare the risk of testis cancer in these countries relevant to foods that are highly contaminated with ochratoxin and there at least measures have been made in equivalent fashion.
So, we used data on the per capita consumption of foods commonly contaminated with ochratoxin, coffee and pork.
More persuasive I would think are the data actually on pork consumption which the older data calls pig meat, probably because it was translated from the German. It is actually quite interesting here that countries that historically avoid pork products actually have quite low testicular cancer.
Now, these are ecologic data and are subject to a large number of problems. I simply want to say that the data such as they are are consistent with the idea that foods contaminated with ochratoxin are associated with risk of testicular cancer.
Is this idea plausible? I think so, and I think so for the following reasons: These are all statements that are known about ochratoxin. Animals exposed to ochratoxin will show ochratoxin in testis. Ochratoxin is transferred to the fetus transplacentally. Ochratoxin is also transmitted to newborns lactationally.
Ochratoxin causes adducts in testicular DNA. This is known. It was observed as an also ran in studies looking at adducts in kidney, but this is actually quite fascinating because I believe that if you find something that will cause adducts in testicular DNA it is not a large step from adducts in DNA to testis cancer and that is because adducts in other tissues principally the kidney predict later tumor development in these organs and finally it has been a long time since we all had embryology but I would remind you that during embryogenesis the fetal testis was the kidney. So agents that are known to cause kidney cancer and to cause DNA adducts in the fetal testis are a reasonable supposition for cause of testicular cancer.
How do I envision this to work? Here is what would happen. We know that consumption of ochratoxin during pregnancy or childhood will cause ochratoxin to be taken up by the testis. This has been observed repeatedly in rats, mice and rabbits. Ochratoxin that is taken up by the testis will cause DNA adducts in testis. This has been amply confirmed.
Here is the only hypothesis part. The hypothesis is that DNA adducts in testis will be promoted to testicular cancer given the anabolic effects of androgens at puberty and in fact what this would produce is the unusual age-specific incidence curves we discussed earlier because if you have no adducts in your testis there will be no testis cancer.
If you have adducts in your testis when those things are promoted they will be promoted at or after puberty and then testis cancer will be rarely seen again.
So, the hypothesis could account for the unusual age distribution of testis cancer.
What is the take-home message of this? Well, I would argue and again I am arguing for the point of discussion that ochratoxin provides a plausible explanation, a coherent explanation for much of the descriptive epidemiology of testis cancer.
We know that ochratoxin induces adducts in DNA in the testis and this produces a plausible and testable model for testicular cancer and we are in fact currently testing this hypothesis by treating pregnant mice with ochratoxin, surviving their male offspring and examining their testes and I would just like to remind you that this is a modern interpretation of the Myth of Oedipus and I am very grateful to the Lance Armstrong Foundation for supporting our experimental work on testis cancer and ochratoxin, and I thank you for the chance to talk about this.