Is human papillomavirus viral load a clinically useful predictive marker? A longitudinal study.

BACKGROUND It has been suggested that in women who test positive for high-risk human papillomavirus (HPV) types, viral load can distinguish women who are at increased risk of cervical neoplasia from those who are not. METHODS Quantitative PCR (qPCR) was used to measure HPV copy number in serial samples taken from 60 and 58 young women previously found to have incident cervical HPV16 or HPV18 infections, respectively, using GP5+/GP6+ primers; women provided at least three samples for qPCR testing, at least one of which was positive. RESULTS A 10-fold increase in HPV16 or HPV18 copy number was associated with a modestly increased risk of acquiring a cytologic abnormality [HPV16: hazards ratio, 1.76 (95% confidence interval, 1.38-2.25); HPV18: hazards ratio, 1.59 (95% confidence interval, 1.25-2.03)]. However, in most women, copy number increased during follow-up, before decreasing again. In women with a HPV16 infection, the median copy number per 1,000 cells was 7.7 in their first qPCR HPV-positive sample, 1,237 in the sample yielding the maximum copy number, and 7.8 in their last qPCR HPV-positive sample; corresponding copy numbers for women with HPV18 infection were 2.3, 87, and 2.4. Maximum HPV16 and HPV18 copy number did not differ significantly between women who acquired an incident cervical cytologic abnormality and those who did not. CONCLUSION Whereas large relative increases in copy number are associated with an increased risk of abnormality, a single measurement of viral load made at an indeterminate point during the natural history of HPV infection does not reliably predict the risk of acquiring cervical neoplasia. Therefore, a single measure of HPV viral load cannot be considered a clinically useful biomarker.