Cri de Coeur

The BBC has a health news item on gaps in research on breast cancer. Research into "genetic factors" and "life-style changes" are always safe because they place the onus for disease on the victim. What is conspicuously missing is any search for a pathogen. ("We've already ruled that out, haven't we?")

This hits me at a particularly bad time because even as I write a good friend is in surgery for a problem caused by a recurrence of breast cancer which appeared when she thought she was in the clear. It seems to have metastasized to bone, which is never a good sign.

The friend who told me about this got more than he bargained for in response. I told him that the cause of 95% of mammary tumors in mice was first discovered in 1936, under the name "the Bittner factor". This was promoted to a real virus some years later, and eventually named mouse mammary tumor virus (MMTV). It took a little longer to admit this was really an active retrovirus. Part of the problem was that MMTV is both endogenous and exogenous, and most virions are defective, in the absence of helper viruses.

Aside: in evolutionary terms there is good reason to believe beta retroviruses have been active in humans more recently than gamma retroviruses, as recent research on HERV-K111 has shown. It doesn't rule out gamma retroviruses, it just means that we are more likely to find beta retroviruses active in humans after the separation of humans from other apes, particularly with reference to modern humans -- in the last 100,000 years. Cutting things finer than this in evolutionary terms is tricky. For reasons why this is relevant to human breast cancer I would suggest reading "Sexual Selection and the Descent of Man". (A search of Google images for the subject of "boobs" may also be informative on the subject of sexual selection and human mammary hyperplasia. Has anyone investigated chimpanzees to see if they are as sensitive to this aspect of appearance? I think not.)

Fast forward into a long-running controversy about human breast cancer. A possible parallel to the "Bittner factor" in humans was long considered possible, at least by some experts. Since about 1987 we have been finding sequences resembling MMTV in human breast cancer patients. Most of these sequences are incomplete and defective. Virologists have done a capital job of undermining any treatments based on a possible retroviral cause. (It was obviously of great political importance to prevent panic which would result from the discovery that another retrovirus was widespread and active in humans after the debacle with HIV.)

There are serious gaps in a virological explanation of how such an infection could operate in humans. Then again, there have been serious gaps in our explanation of how MMTV infection works in mice. There was strong evidence the infection was passed from mother to offspring by milk, but the causative agent could not be isolated. Second there was no good explanation of how and where the infection managed to hide until those offspring reached sexual maturity, when they could pass it on. These gaps are being filled in -- for mice. Talk about vertical transmission in humans is apparently verboten, but genetic factors are fair game, the patient isn't infected, she just chose the wrong parents.

It seems, at least to me, fairly obvious that it is possible to transfer infected maternal cells in milk. If the milk also contains antibodies which would kill actively infected cells and destroy free virions, that would help to produce a latent infection in offspring which would not become apparent until sexual maturity. This is exactly what the virus needs. Beyond that, and beyond cell-to-cell transmission within the body via viral synapses, we have the whole subject of transmission via nucleic acids in exosomes, which would escape most current attempts at detection. Nothing in nature suggests that diseases must cooperate with current human detection technology. Assumptions which work when transmission and replication is rapid and reliable are unlikely to be true in chronic infections spanning decades. It is particularly hard to decide what latent infections might be passed to immunologically-naive neonates, if these are not immediately apparent.

New drugs for breast cancer targeting specific receptors have made it through phase II trials, and are now recommended for fast-track approval within a year (assuming the U.S. government doesn't fold up). This doesn't help my friend at all, late-stage breast cancer has many problems beyond those which started the process.

What kinds of receptors are important? Some you might expect are receptors for estrogen or progesterone, others involve steroids. Do we know of viruses with receptor elements also sensitive to these hormones? Others respond to chemokines or cytokines like epidermal growth factor (EGF) or insulin-like growth factor (IGF). Some, like CCR5, were specifically targeted in therapy for HIV infection. Finally, there are those potential helper viruses mentioned above. Do herpes group viruses and papilloma viruses play any role in other forms of cancer? Are they transmitted by breast feeding?

There are a variety of reasons to try antiviral and antiretroviral therapies against breast cancers. One reason is the lower rate of breast cancer (and prostate cancer) among HIV+ patients on ARV therapy, when almost every other cancer rate is higher. There is abundant evidence that current therapies are inadequate. It doesn't matter if the pathology is neatly packaged in a conspicuous virion, or spread out over several viruses and modes of transmission; the important thing is to interrupt the pathology at an early stage. This point of view has been neglected in the last 25 years.


Added: when I wrote the above, I used the figure 100,000 years for the time in which modern humans have been separate from Neanderthals and Denisovans. In fact, Neanderthal and Denisovan genes do appear in some populations; for example there are Denisovan genes in modern Papua New Guinea populations. Both subspecies disappeared about 30,000 years ago, so none of the samples sequenced for genome analysis can be younger than that. I suspected it would be hard to extract nuclear DNA from samples much older. It turns out that the Neanderthal genome came from individuals who lived about 45,000 years ago, while the Denisovan genome came from an individual who lived about 50,000 years ago. This places sharp limits on the time in which HERV-K111 has been most active. There is one copy in chimpanzee genomes, several in Neanderthal and Denisovan genomes, and more than 100 full-length copies in modern humans. It appears active beta retroviral infection took place between 50,000 years ago, and the point in prehistory when modern humans spread to most continents. This is cutting things even finer than I said above. We are no longer talking about millions of years, or even 100,000. Whatever disadvantages this virus would cause via breast cancer would be of little consequence in the brutal logic of evolution if they came after reproductive age. For most of prehistory, and even much of history, mean life expectancy was around 35 years.

The idea that breast cancer is caused by a fairly well-adapted retroviral infection is not nearly as great a stretch as often stated.
It is even odder to consider that most strains of papilloma virus do not cause cancer. It was really hard to track down those responsible. Exactly how these strains originate is a good question. I suspect a co-factor.

Blog entry information

Last update

More entries in User Blogs

More entries from anciendaze