Analysis of family history data for evidence of non-Mendelian inheritance resulting from vertical transmission.

A number of infections of man, as well as of other animal and plant species, are heavily dependent upon 'vertical transmission'--that is direct parent-to-progeny transfer--for their maintenance in host populations. Such vertical transmission may be considered as a form of inheritance. It is usually non-Mendelian. This paper discusses the implications of such inheritance for the distribution of disease in families. A method is described for making quantitative predictions of prevalence rates of infection and of disease within different classes of relatives of either infected or uninfected probands. It is pointed out that, whereas a maternal line excess is to be expected among relatives of positive probands, the opposite should be found in families of negative probands. Expected differences between maternal and paternal line prevalence rates of observable disease decline rapidly with distance of relationship from the proband, and are greatly reduced by diagnostic insensitivity (analogous to penetrance). The implications of this analytic method for the design of family history studies are discussed. Published data on familial breast cancer are reviewed, and found to show no evidence that this condition is associated with a non-integrated vertically transmitted agent.