Invasive cervical cancers in the United States, Botswana and Kenya: HPV type distribution and health policy implications

Aaron Ermel, Brahim Qadadri, Yan Tong, Omenge Orang’o, Benson Macharia, Doreen Ramogola-Masire, Nicola M. Zetola, Darron R. Brown

More deaths occur in African women from invasive cervical cancer (ICC) than from any other malignancy. ICC is caused by infection with oncogenic types of human papillomavirus (HPV). Co-infection with the human immunodeficiency virus (HIV) accelerates the natural history of ICC, and may influence the HPV type distribution. Because HPV vaccines are available, this malignancy is theoretically preventable, but the vaccines are largely type-specific in protection against infection. Data on specific HPV types causing ICC in African women is limited, and many studies utilized swab samples rather than actual cancer tissue. A previous study using archived, ICC tissue from women in Botswana identified an unusual HPV type distribution. A similar study was therefore performed in a second sub-Saharan country to provide additional information on the HPV type distribution in ICC. Archived, formalin-fixed, paraffin-embedded ICCs were acquired from women in the United States, Kenya, or Botswana. DNA was extracted and HPV genotyping performed by Roche Linear Array. HIV sequences were identified in ICCs by PCR. HPV types 16 or 18 (HPV 16/18) were identified in 93.5 % of HPV-positive ICCs from the U.S., 93.8 % from Kenya, and 61.8 % from Botswana (p < 0.0001). Non-HPV 16/18 types were detected in 10.9 % of HPV-positive cancers from the U.S., 17.2 % from Kenya, and 47.8 % from Botswana (p < 0.0001). HIV was detected in 2.2, 31.5, and 32.4 % from ICCs from the U.S., Kenya, or Botswana, respectively (p = 0.0002). The distribution of HPV types was not significantly different between HIVinfected or HIV-uninfected women. The percentages of ICCs theoretically covered by the bivalent/quadrivalent HPV vaccines were 93.5, 93.9, and 61.8 % from the U.S., Kenya and Botswana, respectively, and increased to 100, 98, and 77.8 % for the nanovalent vaccine. HPV 16/18 caused most ICCs from the U.S. and western Kenya. Fewer ICCs contained HPV 16/18 in Botswana. HIV co-infection did not influence the HPV type distribution in ICCs from African women from the two countries. Available HPV vaccines should provide protection against most ICCs in the U.S. and Kenya. The recently developed nanovalent vaccine may be more suitable for countries where non-HPV 16/18 types are frequently detected in ICC.