Possible etiologies of cancer of the cervix other than herpesvirus.

Abstract

Summary Various hypotheses concerning both infectious and noninfectious causes have been proposed for the etiology of cervical cancer. The evidence for noninfectious agents and for infectious agents other than herpesvirus is reviewed here. Although some carcinogenic potential for smegma may be inferred from experimental studies, epidemiological studies do not indicate a significant role for either smegma or circumcision in human cervical cancer. Coppelson and colleagues have developed a hypothesis to explain the importance of early coitus in cervical cancer, proposing that sperm is the nucleic acid vector and the potential mutagen for tissues undergoing active metaplasia. The most attractive aspect of this hypothesis is the suggestion that any sperm may be a potential mutagen, not because of some special intrinsic property but because of a particular susceptibility of the target tissue at certain times. Six infectious agents have an established association with cervical cancer. The association with Trichomonas , syphilis, and gonococcal infections in all probability merely reflects the fact that these infections are strongly associated with promiscuity. The correlation of mycoplasmal infection with mild cervical dysplasia, as well as with promiscuity, calls for more serious consideration of this infection, but there is no consistent association between mycoplasmal infection and cervical cancer. Chlamydial infections may be important because of their chronicity, their association with chronic inflammatory disease, and their venereal mode of transmission. Cytomegalovirus infections may be implicated upon similar grounds. This virus is known to be associated with chronic inflammatory disease, but evidence for a venereal mode of transmission and for male genital infection is less convincing. Whatever the evidence for the association of these agents with cervical cancer, none appears to be so likely a candidate for its cause as herpesvirus type 2. However, future analytic studies should include examination of male partners and the study of semen for infectivity and such studies should be designed to test multifactorial hypotheses. Too often our hypotheses are single-agent oriented; the design and analysis of studies should both be structured to study multiple possibilities. Any consideration of etiological hypotheses for cervical cancer must begin with consideration of the epidemiological characteristics of that disease. These characteristics, which have been summarized by Rotkin (44), are now universally accepted, since there is remarkable uniformity among studies of diverse populations done at different times (Table 1). Thus, at highest risk of cervical cancer are women who have had 1st coitus at an early age (43), have had multiple sex partners, are sexually promiscuous (50), and are of lower socioeconomic status (28). Any hypothesis must also account for those women at lowest risk, in particular the celibate (14), and to a lesser degree the nulliparous and Jewish women (50). There are many further refinements which might be made to a list of positive and negative risk factors, such as early marriage, early pregnancy, multiple marriages, and multiple pregnancies (55), some of which appear more important in some studies than in others, but the most important variables are generally accepted to be the 1st ones stated.