Emerging technologies and cervical cancer.

Human papillomavirus (HPV) has been demonstrated to be a necessary cause of invasive cervical cancer across epidemiologic and virologic studies worldwide (1). However, the biologic mechanisms for malignant transformation by HPV were well established long before the epidemiologic studies showed any consistency among investigators or across study designs. Once sensitive biologic assays became available for the detection of HPV, correlational, case–control, and cohort studies demonstrated high relative risks, high attributable fractions, appropriate time sequence, differential risk among higher viral load and higher risk viral types, and consistency among investigators (2). Now, with the relationship of HPV to cervical cancer established, attention has turned to how to use this information to decrease the morbidity of and mortality from cervical cancer. Cervical cancer is the third most common cancer in women worldwide. Parkin et al. (3) estimated that 371 200 cases of cervical cancer were diagnosed worldwide in 1990, 80% of them in developing countries. In 1999, there were an estimated 12 800 cases of invasive cervical cancer and 4800 deaths from this cancer in the United States (4). The natural history of cervical cancer is well understood; lesions progress from dysplasia to carcinoma in situ to cancer. The cervix can be sampled cytologically by use of the Pap smear. Although never subjected to a randomized clinical trial, the Pap smear has decreased mortality in all countries in which cervical cancer screening programs have been established. Meanwhile, the natural history of HPV infections in relationship to the natural history of cervical cancer is less well understood. A study by de Villiers et al. (5) has demonstrated that sexually active women become infected and the prevalence of infection peaks between 18 and 25 years of age. Infection prevalence falls off as patients reach age 30 years, the age at which the incidence of high-grade lesions and invasive cancers begins to rise. The immunobiology of HPV is difficult to study given the epitheliotropic nature of the virus and the complicated function of the immune system. Both the antibody and cell-mediated immune system respond to HPV infection. In fact, until recently, studies on the exact relationship between viral antibody response to HPV and disease were producing very inconsistent results. The cell-mediated system also appears to be important in the control of HPV as it is in the control of the human immunodeficiency virus (HIV). Many studies [reviewed in (6)] are addressing its role in anticipation of vaccine development. The HPV prevalence data and the relationship of HPV infection to age suggest that patients who remain infected may be those most at risk for the development of high-grade lesions, but the serologic and cell-mediated immunobiologic data to support this concept are still lacking. In the developing world, the resources for detecting and treating cervical cancer are limited. The goals of cancer eradication programs include finding better screening methods, reaching unscreened populations, and developing treatment programs for cancers that are detected. In developed countries with established screening and treatment programs, the challenge of cervical cancer presents itself differently. In such countries, many of the incident cases have not been screened at regular intervals. On the other end of the spectrum, 50 million Pap smears are performed annually in the United States. Of these smears, an estimated 2.5 million reveal low-grade lesions and atypias whose natural history in more than 80% of cases includes regression. More than 6 billion dollars is spent annually in the United States on their evaluation and management (7). Thus, the challenge in developed countries is finding ways to reach the few women who are unscreened and identifying those lesions most likely to be or to become invasive cancer and using resources wisely to manage them. The present article by the ALTS Group in this issue of the Journal (8) is the first of a series of articles reporting longawaited findings from the Atypical Squamous Cells of Undetermined Significance/Low-Grade Squamous Intraepithelial Lesions (ASCUS/LSIL, or ALTS) Trial. Eligible patients included women 16 years old and older who had had a cervical Pap smear showing ASCUS or LSIL within 2 months priot to trial entry, had no history of cervical or other genital cancers, had no known HIV infection, were not pregnant, were able to give written informed consent, and were likely to commit to follow-up. All patients were to undergo a repeat Pap smear, undergo a cervigram, and have cervical specimens collected for HPV testing. The study was originally designed to recruit 3600 women with ASCUS and 3600 women with LSIL from a representative sample of U.S. women into a randomized, multicenter clinical trial with three arms: an immediate-colposcopy arm, an HPVtesting arm, and a conservative-management arm wherein patients were followed with cytology. The end point for the sample size calculation—screening efficacy at 3 years—was based on the number of high-grade squamous intraepithelial lesions (i.e., “management failures”) expected on the HPV-testing and conservative-management arms. A two-tailed significance level of .05 was assumed, and a 0.1% failure rate on the immediatecolposcopy arms and 1%, 2%, and 3% failure rates on the conservative-management or HPV-testing arms would have yielded powers of .84, .98, and .999, respectively. The trial was to include 3 years of follow-up and as many ethnic groups as possible. Many safeguards were built into the original trial design, including central review committees for Pathology Quality Control, HPV Testing Quality Control, and Colposcopy Quality Control. In addition, several committees were created to provide oversight: a Steering Committee, an Ethics and Data Monitoring Committee, and a Coordinating Unit that would oversee all aspects of the trial at the four participating clinical centers. While