While there is a vast literature on MMTV, there are also reports of virus associated with mammary tumors in rats, dogs, cats, (probably sheep and goats) and cattle. There have even been sporadic reports of mammary tumor virus in non-human primates. The potential for transmission from domesticated animals, included pests domesticated without intent, to humans is large. Rejecting a viral etiology out of hand would be unwise. Why then does the question remain unresolved?
My view is that researchers have spent far too much time trying to force organisms into preconceived roles, and far too little paying attention to what the organisms themselves were telling them.
We now know MMTV has functioning receptor elements in a long terminal repeat (LTR) which respond to progestins, androgens and glucocorticoids. These are all hormones involved in sexual maturity, stress and lactation. This makes sense from the standpoint of viral transmission by milk. What seems less well known is that inserted provirus in non-cancerous breast cells also has a negative regulatory element. What's more, the complete viral genome requires two frame-shift events for efficient production of some proteins. These are facilitated by a pseudoknot in RNA, which frequently causes specific deletions. The provirus could be said to have several "safety" elements like the covers people put on switches that perform dangerous actions. ("Did someone just push the Destruct button for that missile? What do you mean, oops?")
The interaction between endogenous and exogenous virus is also a source of contention. It is well known in other cases that retroviruses commonly, though not always, activate sequences they insert. This says that the best way to activate a gene in an endogenous retrovirus may well be to infect with a very similar exogenous virus. An exogenous virus transmitted by sexual intercourse is the perfect trigger for replication at sexual maturity. A virus could scarcely care less about how it is classified by people.
Why this bizarre and self-contradictory set of properties? It doesn't make sense to people tightly focused on a specific model of pathogenesis developed for use with pathogens which followed a 'strategy' of rapid replication. This made detection convenient. The fact that such pathogens often killed hosts provided plenty of material for morbid pathology. This was convenient for researchers, if not for patients.
At this point I feel compelled to reiterate -- it does not benefit the pathogen to kill the host. All the examples to the contrary are exceptions caused by incomplete evolution following a jump between species, or an unusual burst of pathogen transmission involved in the final stages of disease.
Let me use an example from Ian Lipkin's work on viral hemorrhagic fever. In one outbreak three out of four infected patients died. The survivor only lived because researchers were able to identify a possible treatment in time. Without that intervention the mortality rate would have been 100%. Provided medical workers took adequate precautions that particular pathogen would have become extinct without any intervention at all. (This is not to say that intervention was worthless, just ask the survivor.)
My point is that most epidemics of pathogens following a 'strategy' of rapid replication will burn themselves out. The long-shot chance of one which will consume millions of people first is the driving force behind most epidemiology -- and one current movie.
Rapid replication may benefit a retrovirus when it jumps species, so long as it does not exhaust the pool of susceptible members of the new species before it evolves into a slower replicator depending primarily on vertical transmission and inheritance. The end result would be an endogenous retrovirus without harmful effects on the host, if evolution ever ended.
My point is that the complicated system of promoters, repressors, hormone receptor elements, endogenous and exogenous retroviruses serves a purpose, it keeps the dangerous aspects of the infection to a minimum during the reproductive period of hosts. If negative consequences fall after reproduction, they will have little or no effect on evolution. Infection may even enhance relative reproductive success of infected individuals.
One route is through modulation of immune response. In protecting itself, the virus may also allow more extensive changes associated with host reproduction. Direct immunosuppression in reproductive organs may well offer increased fertility via local immune tolerance. Elsewhere, it may allow more complete development of tissues required for lactation. While breast cancer is a dangerous form of mammary hyperplasia, there are numerous examples on the world-wide web of mammary hyperplasia which seems to contribute to promiscuous, if not reproductive, success.
We no longer have to overcome a 25% infant mortality rate to survive. We don't have to produce a dozen infants to have two survive to take care of us in old age. The effort of raising civilized children who will become more than killer apes is enormous. We could afford lower fertility in exchange for freedom from a dread cancer.
Another 50 years of doubling human populations every 25 years would almost certainly result in a catastrophic collapse. (Even if you support the projection made by the Conference of Catholic Bishops some years back, that the Earth could support 50 billion humans, this only delays disaster by 25 years. If population continues exponential growth, long term projections by planned parenthood are not all that different from opinions expressed by those bishops.)
(Perhaps this is the Gordian knot. Nobody in power dares to present such a stark choice, or even touch these explosive issues.)
Back to my primary topic today, the research muddle in relation to breast cancer. A virus very similar to current MMTV infected human ancestors resulting in a HERV. I don't think we have a good handle on how long this went on, only when it first happened. Numerous papers showing fragments of similar sequences, or activity of similar genes, in human breast cancer have been published. No one has shown a complete sequence of an exogenous virus reliably causing breast cancer.
The point I'm making is that the emphasis is wrong. The virus isn't trying to cause cancer, it is trying to replicate and persist. In the long term it is more useful to the virus to have inserted proviral sequences than active virions. Once similar endogenous sequences exist, exogenous viruses don't have to carry a complete complement of genes, some will already be present in most hosts. This strategy also pays dividends in confusing immune response. It is no longer possible to distinguish pathogens from self. Small wonder if infected hosts develop anti-nuclear antibodies.
When researchers encounter virus in tumors, I'm suggesting all such virus is likely to be defective, and deletions are one easy way to disable long sequences holding back replication. Provirus in somatic cells doesn't need to produce any virions to persist as long as the individual host. Provirus in germ-line cells doesn't need to produce any virions at all to persist and replicate for generations. There will be virions in pathological situations, but these often will not be identical to the originating sequences, which have become partially domesticated.
The search for a pure exogenous virus solely and reliably causing a cancer has resonances for me of an ancient search for "card-carrying communists". I could never figure out why someone devoted to the overthrow of systems of government would handicap themselves with an identifying card. Somehow, I suspect retroviruses are smarter.
I'll have more to say after I check the reception of this post.