Table 2.

Prevalence and attribution of HPV types in cervical cancers

YearFirst authornMultitype infections (%)Percentage of specimens testing positive for individual HPV types
Any HPV type*
SCC (31, 32, 41, 44, 45, 53, 54, 56)
Pooled prevalence1,090§
95% CI, lower3.
95% CI, upper5.70.668.614.
Multitype adjusted472§
95% CI, lower1.
95% CI, upper4.50.971.813.
Adenocarcinomas (31, 32, 40, 43, 50)
Pooled prevalence413§
95% CI, lower0.
95% CI, upper1.81.243.543.
Multitype adjusted§165§
95% CI, lower0.10.034.333.
95% CI, upper3.42.349.
All cervical cancers
Pooled prevalence1,503§
95% CI, lower2.70.057.617.
95% CI, upper3.90.261.319.
Multitype adjusted§637§
95% CI, lower1.40.058.815.
95% CI, upper**2.80.664.
  • * As determined by consensus PCR primers sensitive to positivity for at least 15 HPV types.

  • Data for additional cases and HPV typing provided by the authors. In the study, HPV testing methods changed over time. Early testing of squamous cell and adenocarcinoma specimens (n = 179) was conducted for individual HPV types 6, 11, 16, 18, and 35 (HPV testing for types 31 and 33 was combined in a single cocktail). Later testing was conducted for a much broader spectrum of individual HPV types (n = 596). In the table, prevalence data are reported across all specimens for HPV types 6, 11, 16, 18, and 35, whereas only specimens tested during the latter period are included in results for the remaining HPV types as earlier testing for these types individually was not conducted. Only specimens tested during the latter period are included in the multitype adjusted HPV data, as the earlier testing covered <8 HPV types.

  • Data for additional cases and HPV typing provided by the authors. Two small-cell (neuroendocrine) carcinomas are included among data for SCC.

  • § n refers to total number of lesions with any HPV typing data. By individual HPV type, number of observations varies depending on the number of types for which data were available within each study.

  • Lesions in these studies containing multiple HPV types were fractionally attributed to individual HPV types as described in Materials and Methods. Based on this method, figures for the attribution of individual HPV types differing from those reported in the table for prevalence were as follows: Guo et al. SCC (HPV 16, 68.6%; HPV 33, 0.0%; HPV 35, 7.2%); Schwartz et al. SCC (HPV 6, 2.6%; HPV 16, 68.0%; HPV 18, 11.1%; HPV 31, 2.8%; HPV 33, 2.8%; HPV 35, 0.0%; HPV 45, 1.4%; HPV 58, 0.0%; HPV 73, 0.0%), adenocarcinomas (HPV 6, 0.6%; HPV 16, 41.6%; HPV 18, 40.8%; HPV 33, 1.2%; HPV 35, 0.0%; HPV 39, 0.0%).

  • Prevalence across all cervical cancer cases adjusted for the relative proportion of cervical SCC versus adenocarcinomas (including adenosquamous carcinomas) observed in the U.S. population (30).

  • ** Upper confidence interval width estimated using score interval (37) where undefined from Barnett et al. (38). (for P = 0.0).