Interlaboratory Reliability of Hybrid Capture 2

We evaluated the interlaboratory reproducibility of the Hybrid Capture 2 (HC2; Digene, Gaithersburg, MD), a test for oncogenic human papillomavirus (HPV) DNA, using data from 4 clinical center (CC) laboratories and the quality control (QC) laboratory participating in the ASCUS (atypical squamous cells of undetermined significance) and LSIL (low-grade squamous intraepithelial lesion) Triage Study (ALTS). Residual liquid cytology specimens were tested routinely throughout the duration of ALTS at CC laboratories, and a stratified (by time in the study) random sample of specimens was retested by the HPV QC laboratory using equivalent protocols. Of the specimens selected (N = 1,175, 5.50% of all specimens obtained), 1,072 (91.23%) had sufficient specimen volume for retesting. The κ value between all CC laboratories and the HPV QC laboratory was 0.84 (95% confidence interval, 0.78–0.89), with κ values for individual CCs and the HPV QC laboratory ranging from 0.79 to 0.89. Agreement between test results was lowest among results for women with negative cytologic findings (0.73); among those with equivocal or abnormal cytologic findings, κ values were 0.80 or more. These data show that HC2 is a reliable test for detecting clinically relevant oncogenic HPV DNA.