CHINA: HPV INFECTION AND CERVICAL CANCER SCREENING STUDIES.

Epidemiology of cervical cancer in China Mortality data from the 1970s to the 1990s suggest that China suffers relatively high cervical cancer mortality, particularly in the rural mid-west. Wudu in Gansu province and Yangcheng in Shanxi had age-adjusted mortality rates of > 40/100 000 in 1990-1992. Better diagnosis and treatment have reduced deaths from cervical cancer, but the improvement is not uniform and mortality rates are unchanged in some counties. Status of cervical screening There are no national screening programmes for cancer in China. In the case of cervical cancer, limitations include the nationwide shortage of cytologists. In late 2003, a national cancer prevention and control strategy was finalised and endorsed by the Ministry of Health, following consensusmeetings for early detection and treatment across nine cancer types. Two demonstration centres for cervical cancer prevention and control have been established in Shenzhen (South-East China) for high resources settings and Xiangyuan (North-West China) for low resources settings [56]. A government recommendation that all women should have at least one screen between 35 and 65 years of age is still pending. Cervical cancer research in China The first Shanxi Province Cervical Cancer Screening Study (SPOCCS I) included 1997 women who underwent a cervical evaluation using HPV self-test, optical biopsy, liquid-based cytology (ThinPrep), VIA, direct testing for HPV, or colposcopy with biopsy [57]. A second study (SPOCCS II) included > 8000 women who submitted a self-sample for HPV at their village and subsequently visited the clinic for HPV Direct Test, LBC (AutoCyte), and VIA [58]. Those with abnormalities (n = 3252) underwent colposcopy and biopsy. Women with CIN 2+ lesions were treated; those with CIN 1 lesions will be followed up in 12 months. In both studies, women with cervical cancer or CIN 2+ lesions on biopsy were likely to be infected with HPV (> 95%). Even for women with CIN 1 lesions, the attributable risk of HPV is as high as 95%. The incidence of HPV among women with normal biopsies was < 15%. Persistent HPV infection was higher among women aged 50-54 years than younger women in Shanxi, but there were marked differences in the age-related curves for women in rural areas versus cities. Risk cofactors Extensive lifestyle data were collected covering sexual history, child bearing, occupation, health, bathing practices, age at menarche and menopause, education, and income. Lifestyle had a significant effect on HPV and cervical cancer prevalence in Shanxi. Risk factors that appeared to contribute to the likelihood of cervical cancer were subject promiscuity (OR = 1.42); husband's promiscuity (1.42); bathing at a public house (1.23); postmenopause (1.22); and current smoking (1.17). Surprisingly, education appeared to slightly raise the likelihood of HPV. When self-testing was compared with direct testing for HPV, specificity for CIN 2+ was identical (85.9%), while sensitivity was higher on direct testing (97.6%) than for self-testing (83.5%). Evaluation of screening tests In both trials, each screening test was compared with pathology to determine accuracy in detecting moderate-high-grade lesions (Table 4). Conclusions With high sensitivity (> 96%) and moderate specificity (86%), the HPV direct test could be used as a primary screening test for cervical cancer risk. Self-sampling is simple and less expensive; however, improvements in instructional leaflets will be needed. Combined HPV and VIA offers potential for screening in the mid-west regions of China, but for consistent results, VIA training will be essential. [ABSTRACT FROM AUTHOR]