Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)


Medical Sciences


William E. Rawls


A review of the available evidence relating herpes simplex type 2 (HSV-2) to cervical cancer in humans showed reasonable circumstantial evidence to suspect the participation of HSV-2 in the pathogenesis of the tumors. The best evidence was provided by seroepidemiological data, however, variations were observed among the data obtained from different populations. A mathematical model based on causal considerations and developed as an extension of a model used in chemical carcinogenesis was found to account for a large part of these variations. This model was found to reasonably describe the relation between the incidence rate of cervical cancer and the fraction of women with previous experience with the virus in different human populations. The fitting to the data suggested the existence of two kinds of etiologic factors involved in the pathogenesis of cervical cancer, one related to HSV-2 and another not related to the virus. The model described for average population values was expanded to age-specific data. This allowed the estimation of the age distribution of primary infection with HSV-2 (ϕ(v)), the distribution of the age of developing cancer unrelated to HSV-2 (Yᵤ(t)), the distribution of the age of developing cancer associated with HSV-2 (Yᵥ(t)), and the distribution of the time elapsing from the primary infection with HSV-2 up to the development of cervical cancer associated to HSV-2 (ψ(u)). The different shapes of Yᵤ(t) and Yᵥ(t) supported the existence of the two kinds of etiologic factors in the pathogenesis of the tumors. In addition, ψ(u) was found to correspond to relatively long times suggesting an important role for viral latency or viral reinfections in HSV-2 carcinogenesis. In view of the fact that some cervical cancers may not be related to HSV-2 the direct detection of HSV-2 genetic information in the cancer cells was considered fundamental for the correct assessment of the viral participation in the pathogenesis of the tumor. Large inconsistencies were found among reports of studies designed to detect HSV-2 genes in cancer cells. As a consequence a new approach was explored. The direct retrieval of viral endogenous genes was attempted from cells biochemically transformed by the virus. Contrary to reports from other systems, rescue was not detected. The frequency of rescue products was estimated as a value below 7x10^-17. These results suggested that rescue experiments which may be suitable for the detection of latent HSV, are not sensitive enough for the exposure of subgenomic viral sequence present in transformed cells. As a requirement for the rescue experiments, a new technique was developed to quantitate and clone HSV expressing the viral thymidine kinase in stocks predominantly lacking virus able to express this enzymatic activity.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."