The evolution of parasites is not limited to their virulence, or even to their potential to be transmitted to new hosts, etc. An additional consideration is how parasites are affected by their adaptation to the host environment plus how the host environment affects their evolution non-adaptively. That is, how parasites change, genetically and physiologically, in response to a combination of natural selection and genetic drift. Note that the ideas that I will now consider are not limited to parasite evolution, but instead can be applicable to any organism that finds itself adapting to a new, especially resource-rich environment and/or environments in which an organism's gene pool becomes highly constrained.
The impact of natural selection is simpler to consider. Here the expectation is that parasites will adapt to their host environment, becoming both more streamlined and, if possible, more coevolved with their hosts. While these steps can involve evolutionary innovations, including development or acquisition of mechanisms for avoiding the host immune response, what also can occur is a loss of metabolic breadth. That is, the host environment can be quite resource rich and organisms may become better competitors, within that environment, by reducing their metabolic capacity (i.e., loss of genetic information due to functional redundancies, here redundancies with functions associated with the host organisms). At an extreme, some of the smallest known cellular genomes are those of parasitic bacteria that produce few organic factors but instead display a diversity of cell membrane receptors for importing those factors from host tissues. Indeed, it is almost cliché to expect parasites to have a reduced functional ability to obtain resources but which at the same time are highly proficient at converting those resources into progeny, plus quite good at avoiding being eliminated from what otherwise can be an impressively "cushy" environment (e.g., tape worms).
It is important to note that natural selection is not the only mechanism that can lead to such decay of a parasite's genotype. That is, drift can eliminate even genes that are beneficial but not essential. Drift, furthermore, can operate not just as the familiar genetic bottlenecking and founder effects, but also as a consequence of hitchhiking and periodic selection. These various mechanisms can be exacerbated if genetic exchange is minimized, and particularly while population sizes are reduced (i.e., which otherwise favor drift over selection). The result is an accumulation of deleterious alleles, a phenomenon that can be viewed as a gradual deterioration of an organism's metabolic or indeed adaptive versatility. In other words, parasites can experience both drift-mediated genomic decay and selection-mediated decay.
Ironically, it can be acquisition of additional DNA, in the guise of pathogenicity islands, that can convert a non-pathogen to a pathogen. Thus, we can speculate (Lawrence, 2005) that there exists a typical pathway in pathogen development which can involve first acquisition of additional DNA and then, subsequently, loss of either non-essential or less-essential genes driven by a combination of streamlining (natural selection) and Muller's ratchet (drift in the absence of genetic exchange). If sexual exchange is ongoing, however, then we instead can have the more familiar equilibrium between gene acquisition and gene loss, plus a relative absence of at least Muller's Ratchet (drift though can still operate given bottlenecking or founder effects, and even periodic selection if selection occurs rapidly in comparison with gene exchange).
With gene loss, reduced potential to receive DNA from other organisms (such as due to adhering to an obligately parasitic lifestyle), and increased specialization, Lawrence (2005) explicitly postulates a four step process, one leading from free-living organisms (he provides E. coli as an example) to opportunistic pathogens with relatively broad host ranges (e.g., Salmonella typhimurium) to more-virulent pathogens with narrower host ranges (e.g., Salmonella typhi) to virulent pathogens with markedly greater dependence on their hosts for ongoing replication (e.g., intracellular pathogens such as Mycobacterium tuberculosis) to obligately host-dependent pathogens (e.g., Rickettsia spp. or Mycoplasma spp.). It is worth noting that M. tuberculosis, for all its infamy as a pathogen, is nonetheless a "basket case" in terms of its ongoing genetic integrity.
The various lessons gleaned from consideration of parasite evolution are highly applicable to considering the population biology of cancer. What is cancer? From the perspective of medicine it is many diseases, all involving a combination of uncontrolled cell division and an inappropriate invasiveness of those cells into other tissues. From the perspective of population biology, however, cancer cells represent a peculiar form of virulent defection within an otherwise clonal population of cooperating entities. That is, these are cells that grow faster, take more than their share of resources, contribute less to the inclusive fitness of the body's cells, and go places where they otherwise do not belong. In fact, just as the acquisition of parasites is an inevitable consequence of evolutionary success, so too is cancer-like defection an inevitable consequence of multicellularity. Multicellular organisms, as a consequence, not only possess, but in fact must possess numerous adaptations that have the effect of minimizing cancer occurrence or consequences. Generally these mechanisms are ones which have the effect of assuring the clonality of body cells, as I consider in the concluding chapter of this text, on the evolution of "Greater Size and Complexity".