For personal use. Only reproduce with permission from The Lancet Publishing Group. RESEARCH LETTERS Human papillomavirus genotypes in rural Mozambique Xavier Castellsagué, Clara Menéndez, Maria-Paz Loscertales, Janet R Kornegay, Francisco dos Santos, F Xavier Gómez-Olivé, Belen Lloveras, Nayana Abarca, Neide Vaz, Avertino Barreto, F Xavier Bosch, Pedro Alonso We studied the genotype distribution of cervical human papillomavirus (HPV) infections in an age-stratified sample of 262 women in Mozambique using the PGMYO9-PGMY11 primer system in a reverse line-blot strip-based assay with high sensitivity in type-specific amplification. Despite the low precision of the estimates, we found that HPV-16 was not the dominant type. Instead, HPV 35 was the most commonly identified genotype among HPV-positive women (16/96 [17%]) and women with cervical neoplasia (7/23 [30%]). Certain genotypes might have been under-detected in previous studies, and type-specific HPV distributions might vary across populations. Therefore, the estimated proportion of cervical neoplasia that could be prevented by an HPV-16-based vaccine could be lower than expected. Lancet 2001; 358: 1429–30 Immunisation against human papillomavirus (HPV) offers the greatest hope for a long-term solution to cervical cancer in developing countries. However, little is known about the epidemiology and natural history of HPV infections in these countries. We did a prevalence study of HPV in rural southern Mozambique—a country with high incidence of, and mortality from, cervical cancer. 1 Between August and October, 1999, 262 women aged 14–61 years were selected from this population by means of a pre-established age-stratified scheme. 155 (59%) women were selected from the antenatal and family planning clinics of the Manhiça Health Center, and the rest, mostly older women, were randomly selected from the general population by use of census data. Only two women refused to participate. After signing a witnessed, informed consent form, participating women were interviewed according to a pretested standardised questionnaire for sociodemographic indicators and risk factors for HPV infection. Study participants underwent a complete gynaecological examination, and cervical samples were collected for screening of cytological abnormalities and HPV genotyping. Ethics approval was provided by the Ministerio da Saude of Mozambique. HPV DNA was detected and genotyped by the reverse line-blot strip-based detection system, 2 and further improved by use of the PGMY09-PGMY11 primer system. 3 The genotypes identified by this strip include 18 high-risk (cancer- associated) HPV genotypes (HPV 16, 18, 26, 31, 33, 35, 39, 45, 51, 52, 55, 56, 58, 59, 68 [ME180], MM4 [W13B], MM7 [P291], and MM9 [P238A]), and nine low-risk genotypes (HPV 6, 11, 40, 42, 53, 54, 57, 66, and MM8 [P155]). All samples were further tested by the hybrid capture II system (Digene, Silver Spring, MD, USA). In 14 samples for which no PCR amplification was accomplished, the results from the hybrid capture assay were used instead. Papanicolaou smears were available and interpretable for cytological reading for 245 (94%) women. Cervical neoplasia was diagnosed in 30 women (12% [95% CI 9–17]), 13 of whom had low-grade squamous intraepithelial lesions, 16 of whom had high-grade squamous intraepithelial lesions, and one of whom had carcinoma. Overall HPV DNA prevalence estimates were 36% (95% CI 30–42) by hybrid capture, 40% (34–46) by reverse line-blot, and 40% (34–46) by reverse line-blot plus hybrid capture when no PCR amplification was possible. Table 1 shows that HPV prevalence for any genotype was highest in the youngest age-group, and gradually declined with age. HPV DNA prevalence significantly increased with increasing severity of neoplastic lesions. Genotyping was possible in 96 women, for whom 165 distinct HPV isolations were obtained. Simultaneous detection of multiple (two to six) types was seen in 41% of HPV-positive women (table 2). High-risk genotypes 35, 16, 18, and 39 were the most common HPVs identified. 38 infections were detected in the 23 HPV-positive women with cervical neoplasia, of whom eight had multiple HPV types. Table 2 shows that HPV 35 was also the most common type identified in the subset of women with cervical neoplasia (30% [95% CI 15–50]), followed by type 58 (17% [7–37]), 16 (13% [5–32]), and 33 (13% [5–32]). We explored the agreement between the results of the hybrid capture and reverse line-blot assays in the 231 samples in which the HPV types considered in the two assays were similar. The total agreement was 95% and the  statistic was 0·89 (0·83–0·95). The sensitivity and specificity of the hybrid capture assay compared with the reverse line-blot assay was 91% (83–96) and 97% (92–99), respectively. In addition to the overall high prevalence of HPV infections and high-risk genotypes, the most striking features in the epidemiology of HPV infection in this population are the high frequency of multiple simultaneous infections, and the unexpected singularity of the type-specific HPV distribution. This information is essential in deciding which genotypes should be targeted when developing and testing THE LANCET • Vol 358 • October 27, 2001 1429 3 Jick H, Jick SS, Gurewich V, Myers MW, Vasilakis C. Risk of idiopathic cardiovascular death and nonfatal venous thromboembolism in women using oral contraceptive with differing progestagen components. Lancet 1995; 346:1589–93. 4 Vasilakis C, Jick SS, Jick H. The risk of venous thromboembolism in users of postcoital contraceptive pills. Contraception 1999; 59: 79–83. 5 Jick H, Kaye JA, Vasilakis-Scaramozza C, Jick SS. Risk of venous thromboembolism among users of third generation oral contraceptives compared with users of oral contraceptives with levonorgestrel before and after 1995: cohort and case-control analysis. BMJ 2000; 321: 1190–95. Boston Collaborative Drug Surveillance Program, Boston University School of Medicine, 11 Muzzey Street, Lexington, MA 02421, USA (C Vasilakis-Scaramozza MPH, H Jick MD) Correspondence to: Catherine Vasilakis-Scaramozza (e-mail: cvasil@bu.edu) HPV-positive/ High-risk Multiple tested genotypes/HPV genotypes/HPV positives positives Age (years) 14–20 28/51 (55%) 25/28 (89%) 18/28 (64%) 21–30 29/61 (48%) 28/29 (97%) 11/28 (39%) 31–40 20/53 (38%) 16/20 (80%) 4/18 (22%) 41–50 14/47 (30%) 11/14 (79%) 5/14 (36%) 51 9/41 (22%) 9/9 (100%) 1/8 (13%) Total 100/253 (40%) 89/100 (89%) 39/96 (41%) p for trend <0·0001 0·65 0·003 Cytological diagnoses Normal 63/196 (32%) 53/63 (84%) 23/60 (38%) ASCUS 9/19 (47%) 8/9 (89%) 7/9 (78%) Cervical neoplasia 24/30 (80%) 24/24 (100%) 8/23 (35%) LSIL 10/13 (77%) 10/10 (100%) 4/9 (44%) HSIL 13/16 (81%) 13/13 (100%) 4/13 (31%) Carcinoma 1/1 (100%) 1/1 (100%) 0/1 p for trend <0·0001 0·01 0·7 HSIL=high-grade squamous intraepithelial lesion. LSIL=Iow-grade squamous intraepithelial lesion. *ASCUS=atypical squamous cells of undetermined significance. Table 1: Prevalence and characteristics of HPV DNA detection by age and cytological diagnosis