TY - JOUR
T1 - Worldwide distribution of human papillomavirus types in cytologically normal women in the International Agency for Research on Cancer HPV prevalence surveys
T2 - A pooled analysis
AU - Clifford, G. M.
AU - Gallus, S.
AU - Herrero, R.
AU - Muñoz, N.
AU - Snijders, P. J F
AU - Vaccarella, S.
AU - Anh, P. T H
AU - Ferreccio, C.
AU - Hieu, N. T.
AU - Matos, E.
AU - Molano, M.
AU - Rajkumar, R.
AU - Ronco, G.
AU - De Sanjosé, S.
AU - Shin, H. R.
AU - Sukvirach, S.
AU - Thomas, J. O.
AU - Tunsakul, S.
AU - Meijer, C. J L M
AU - Franceschi, S.
PY - 2005/9/17
Y1 - 2005/9/17
N2 - Background: The proportion of women infected with human papillomavirus (HPV) varies greatly across populations, as might the distribution of HPV types. We aimed to compare HPV-type distribution in representative samples of women from different world regions. Methods: Women were randomly selected from the general population of 13 areas from 11 countries (Nigeria, India, Vietnam, Thailand, Korea, Colombia, Argentina, Chile, the Netherlands, Italy, and Spain). A standardised protocol was used for cervical specimen collection. All HPV testing was by GP5+/6+ PCR-based EIA. The proportion of HPV-positive women infected with different HPV types was compared by study area and between pooled regions with age-adjusted odds ratios (ORs) with corresponding 95% floating CIs. Findings: 15 613 women aged 15-74 years without cytological abnormalities were included in a pooled analysis. Age-standardised HPV prevalence varied nearly 20 times between populations, from 1.4% (95% CI 0.5-2.2) in Spain to 25.6% (22.4-28.8) in Nigeria. Although both overall HPV prevalence and HPV16 prevalence were highest in sub-Saharan Africa, HPV-positive women in Europe were significantly more likely to be infected with HPV16 than were those in sub-Saharan Africa (OR 2.64, p=0.0002), and were significantly less likely to be infected with high-risk HPV types other than HPV16 (OR 0.57, p=0.004) and/or low-risk HPV types (OR 0.44. p=0.0002). Women from South America had HPV-type distribution in between those from sub-Saharan Africa and Europe. Heterogeneity between areas of Asia was significant. Interpretation: Heterogeneity in HPV type distribution among women from different populations should be taken into account when developing screening tests for the virus and predicting the effect of vaccines on the incidence of infection.
AB - Background: The proportion of women infected with human papillomavirus (HPV) varies greatly across populations, as might the distribution of HPV types. We aimed to compare HPV-type distribution in representative samples of women from different world regions. Methods: Women were randomly selected from the general population of 13 areas from 11 countries (Nigeria, India, Vietnam, Thailand, Korea, Colombia, Argentina, Chile, the Netherlands, Italy, and Spain). A standardised protocol was used for cervical specimen collection. All HPV testing was by GP5+/6+ PCR-based EIA. The proportion of HPV-positive women infected with different HPV types was compared by study area and between pooled regions with age-adjusted odds ratios (ORs) with corresponding 95% floating CIs. Findings: 15 613 women aged 15-74 years without cytological abnormalities were included in a pooled analysis. Age-standardised HPV prevalence varied nearly 20 times between populations, from 1.4% (95% CI 0.5-2.2) in Spain to 25.6% (22.4-28.8) in Nigeria. Although both overall HPV prevalence and HPV16 prevalence were highest in sub-Saharan Africa, HPV-positive women in Europe were significantly more likely to be infected with HPV16 than were those in sub-Saharan Africa (OR 2.64, p=0.0002), and were significantly less likely to be infected with high-risk HPV types other than HPV16 (OR 0.57, p=0.004) and/or low-risk HPV types (OR 0.44. p=0.0002). Women from South America had HPV-type distribution in between those from sub-Saharan Africa and Europe. Heterogeneity between areas of Asia was significant. Interpretation: Heterogeneity in HPV type distribution among women from different populations should be taken into account when developing screening tests for the virus and predicting the effect of vaccines on the incidence of infection.
UR - http://www.scopus.com/inward/record.url?scp=24944504307&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=24944504307&partnerID=8YFLogxK
U2 - 10.1016/S0140-6736(05)67069-9
DO - 10.1016/S0140-6736(05)67069-9
M3 - Article
C2 - 16168781
AN - SCOPUS:24944504307
VL - 366
SP - 991
EP - 998
JO - The Lancet
JF - The Lancet
SN - 0140-6736
IS - 9490
ER -