Background: Studies of clinical relevance of human papillomavirus (HPV) DNA load are often limited by one time measurement of prevalent infection.
 
Methods: The nested case-control study was conducted among women enrolled in the ALTS-LSIL Triage Study who were followed semiannually over 2 years for HPV and cervical lesion detection. Cases were women who had intraepithelial neoplasia 2 or 3 (CIN2-3) initially diagnosed during follow-up and HPV16 DNA detected at one or more follow-up visits prior to or at the time of CIN2-3 diagnosis. Controls were selected from those who had HPV16 DNA detected during follow-up but without a diagnosis of CIN2-3 in the 2-year study period. HPV16 DNA was measured by real-time PCR on one follow-up sample per woman and enrollment samples of those who were positive for HPV16 at study entry.

Results: The mean value (SD) of log10 HPV16 E7 copy number per nanogram of cellular DNA at the follow-up visit was 2.81 (±1.72) for 92 cases and 1.60 (±1.63) for 152 controls. The odds ratio associating risk of CIN2-3 with per log₁₀-unit increase in follow-up HPV16 DNA load was 1.52 (95%CI, 1.29-1.78). Of 159 women with HPV16 infection initially detected during follow-up, 149 (48 cases and 101 controls) were tested for HPV16 DNA load on the first positive sample. The mean value of HPV16 DNA load of the "incident" infection was 1.52 for controls and 3.59, 3.02, and 2.11 for cases who had CIN2-3 diagnosed at the same time of or 6 and 12-18 months, respectively, after the initial HPV16-positive detection (P<0.001). Among those with HPV16 infection at study entry, the mean value of viral load between enrollment and follow-visit was comparable among 41 cases (3.27 versus 3.08, P=0.40), but dropped substantially among 44 controls (2.76 versus 1.78, P=0.004).

Conclusion: Risk of CIN2-3 increases as increasing HPV16 DNA load. The time to lesion development correlates with viral load of the initial infection. The patterns of HPV16 DNA load over time differ substantially between women with and without subsequent CIN2-3.