Epidemiological studies have established human papillomavirus (HPV) infection as the central cause of invasive cervical cancer (ICC) and its precursor lesions (Walboomers et al, 1999). However, only a fraction of precancerous lesions progress to ICC. A strong candidate factor for differential progression is HPV type (Lorincz et al, 1992). Identifying HPV types that preferentially progress from high-grade squamous intraepithelial lesions (HSIL) to ICC has implications not only for follow-up protocols in ICC screening programmes, but also for prophylactic type-specific HPV vaccine trials. For ethical reasons, final outcome measures in such trials will be the prevention of HSIL. However, it is important to know whether the HPV type distribution in HSIL is representative of those that go on to cause cancer. Articles presenting HPV type-specific prevalence data were identified from Medline. Studies had to include at least 20 cases of squamous cell or histologically unspecified cervical cancer (Clifford et al, 2002) and/or 20 histologically verified cases of HSIL. In this study, HSIL refers both to lesions classified by the Bethesda system, that is, CIN2/3, and those classified separately as CIN2 and CIN3. Studies had to use polymerase chain reaction (PCR)-based assays to identify HPV, and to present prevalence of at least one type other than HPV6, 11, 16 or 18 (Clifford et al, 2002). This report includes 8594 squamous cell carcinoma of the cervix (SCC) cases (including 2725 of unspecified histology), as previously reported in Clifford et al (2002), and 4338 HSIL cases (1733 reported as CIN2/3, 1824 as CIN3, 729 as CIN2 and 52 as cervical carcinoma in situ)(detailed information on the HSIL studies is reported in the Appendix). Compared to SCC, cases of HSIL were more likely to be from (i) Europe and South/Central America rather than other regions, (ii) studies that detected HPV from exfoliated cells rather than biopsy specimens and (iii) studies that used ‘broad’-spectrum (MY09/11, GP5+/6+ and SPF10) rather than other PCR primers (Table 1 Table 1 Distribution of SCC and HSIL cases by region and study characteristics Lesion No. of studies No. of cases Source region (% of cases) Cervical specimen for HPV testing (% of cases) PCR primers used (% of cases) SCC 78 8594 Africa (6.9), Asia (31.7), Broad spectruma (61.9) Europe (32.0), North America/Australia Biopsies (83.4) Narrow spectrumb (15.5) (13.0), South/Central America (16.5) Exfoliated cells (16.6) Combination/other (16.3) Type-specific only (6.4) Broad spectruma (80.8) HSIL 53 4338 Africa (1.8), Asia (16.7), Europe Biopsies (34.1) Narrow spectrumb (7.9) (52.4), North America (10.3), Exfoliated cells (65.9) Combination/other (7.4) South/Central America (18.8) Type-specific only (3.9) HPV=human papillomavirus; SCC=squamous cell/unspecified carcinoma of the cervix; HSIL=high-grade squamous intraepithelial lesion; PCR=polymerase chain reaction; a ‘Broad’-spectrum PCR primers include MY09/11, GP5+/6+ and SPF10. b ‘Narrow’-spectrum PCR primers include GP5/6, L1C1/C2 and PU1M/2R. ). Type-specific prevalence is presented for the 14 most common high-risk (HR) types identified in SCC (Table 2 Table 2 Comparison of overall and type-specific HPV prevalence between SCC and HSIL cases SCC HSIL SCC : HSIL HPV type n HPV (%) n HPV (%) prevalence ratioa All 8550 87.6 4338 84.2 1.04 (1.03, 1.06) 16 8594 54.3 4338 45.0 1.21 (1.16, 1.26) 18 8502 12.6 4338 7.1 1.79 (1.56, 2.10) 33 8449 4.3 4302 7.2 0.59 (0.53, 0.68) 45 5174 4.2 2214 2.3 1.85 (1.35, 2.91) 31 7204 4.2 4036 8.8 0.48 (0.43, 0.54) 58 5646 3.0 2175 6.9 0.43 (0.37, 0.52) 52 5304 2.5 2153 5.2 0.48 (0.40, 0.60) 35 6223 1.0 2690 4.4 0.22 (0.18, 0.27) 59 4488 0.8 1636 1.5 0.55 (0.38, 0.97) 56 4493 0.7 2110 3.0 0.23 (0.18, 0.31) 51 4580 0.6 2171 2.9 0.20 (0.16, 0.27) 68 4148 0.5 1763 1.0 0.50 (0.33, 1.04) 39 3899 0.4 1841 1.1 0.35 (0.24, 0.66) 66 4799 0.2 1778 2.1 0.10 (0.08, 0.15) HPV=human papillomavirus; SCC=squamous cell/unspecified carcinoma of the cervix; HSIL=high-grade squamous intraepithelial lesion. a With 95% confidence intervals. ). As not all studies tested for all 14 types, sample size varies between type-specific analyses. Type-specific prevalence is expressed as a percentage of all cases tested for HPV, and thus represents the prevalence in either single or multiple infections. Overall, HPV prevalence was slightly higher in SCC cases (87.6%) than HSIL (84.2%) (SCC : HSIL ratio 1.04, 95% CI 1.03–1.06) (Table 2). HPV16 was the most common type in both SCC (54.3%) and HSIL (45.0%), but was more prevalent in SCC (ratio of 1.21, 95% CI 1.16–1.26). HPV18 was also more prevalent in SCC (12.6%) than in HSIL (7.0%), with a ratio of 1.79 (95% CI 1.56–2.10). HPV45 was associated with a ratio of 1.85 (95% CI 1.35–2.91), similar to that of HPV18. All other HR types included in the analysis had ratios between 0.1 and 0.6 (Table 2). The SCC : HSIL ratios for the eight most common HPV types were additionally calculated within more homogeneous study subgroups: (i) studies that did not report any multiple infections (6558 SCC, 2182 HSIL), (ii) studies testing for HPV from biopsies (7128 SCC, 1483 HSIL) and (iii) studies using ‘broad’-spectrum PCR primers (5318 SCC, 3502 HSIL). The SCC : HSIL ratios were also calculated separately for HSILs classified by the Bethesda system and for CIN3 only. Across all these subanalyses, SCC : HSIL ratios remained consistent for HPV16 (range: 1.04–1.25), HPV18 (1.46–1.93) and HPV45 (1.20–4.61). HPV31, 33, 35, 52 and 58 were consistently associated with ratios of 0.3–0.9, with the exception of HPV58 for biopsy studies (1.06, 95% CI 0.73–2.08). Where sample size permitted, subanalyses were also stratified by region. When estimated from studies within Asia, Europe and South/Central America, respectively, there was no material difference in SCC : HSIL ratios for HPV16 (1.46, 1.17, 1.40), HPV18 (1.74, 2.02, 1.46), HPV45 (4.35, 1.39, 1.20), HPV33 (0.56, 0.62, 0.76), HPV52 (0.39, 0.26, 0.64) or HPV58 (0.55, 0.24, 0.30). However, notably high ratios were observed for HPV31 in South/Central America (1.13, 95% CI 0.84–1.70) in comparison to Europe (0.41, 95% CI 0.36–0.48) and Asia (0.43, 95% CI 0.31–0.68), and for HPV58 in China (including Taiwan and Hong Kong) (1.27, 95% CI 0.85–2.51) in comparison to non-Chinese Asian countries (0.37, 95% CI 0.27–0.58), raising the possibility of localised variation in the malignant potential of particular HPV types (Chan et al, 2002). Our findings suggest that worldwide, HSIL infected with HPV16, 18 or 45 are more likely to progress to SCC than HSIL infected with other HR types. This could be interpreted in two ways: either these types have a greater potential to induce fully malignant transformation, and/or these infections somehow preferentially evade the host immune system. Compared to other HPV types, HPV18 has been associated with increased oncogenic potential in cell culture, screening failures and poorer cancer prognosis (Hildesheim et al, 1999; Schwartz et al, 2001; Woodman et al, 2003). Thus, HPV18 enrichment in SCC may reflect its greater oncogenic potential. Given its genetic similarity to HPV18, this may also be true for HPV45. Conversely, compared to other HPV types, HPV16 infections are more likely to persist and progress to HSIL (Molano et al, in press). Both persistence of infection and progression to HSIL have been shown associated with HPV16 variants (Londesborough et al, 1996). Thus, HPV16 enrichment in SCC may be related to its greater ability to escape immune surveillance compared to other types. Even in countries with established screening programmes, women still die from rapidly progressing cancers that escape periodic examination. Given that HPV16, 18 and 45 appear to have greater progressive potential, and in the event that future cervical screening programmes include HPV typing, women infected with HPV16, 18 and 45 may require closer surveillance than women infected with other HR HPV types. The demonstration that the HPV type distribution in HSIL is not entirely representative of those that go on to cause cancer also has important implications for prophylactic type-specific HPV vaccine evaluation. This is because any beneficial effect identified by randomised trials from the proportion of HSIL preventable by HPV16 or HPV16/18 vaccines may be an underestimate of the beneficial effect on the prevention of ICC.
