Infection with one or more high-risk HPV (hr-HPV) types is a necessary step in the development of cervical cancer, making its detection a useful tool for cervical cancer screening and prevention (Bosch et al., 2002). Hybrid Capture 2 (HC2, Qiagen, Gaithersburg, MD) was the first FDA-approved and is still the most widely used laboratory test for the presence of high-risk HPV DNA as a predictor of who is or is not at risk for cervical cancer and pre-cancer (defined here as cervical intraepithelial neoplasia grade 2 or worse). While HC2 has remarkable sensitivity for cervical intraepithelial neoplasia grade 2 or worse, the transient nature of most hr-HPV infections reduces its specificity as most infections are benign and will resolve without progressing to cervical intraepithelial neoplasia grade 2 or worse. This can be problematic for cervical cancer screening algorithms in which HPV positivity triggers more intensive follow-up; in such algorithms, good specificity is important for reducing the number of costly colposcopic follow-up visits (Cuzick et al., 2008). Although manufacturer specifications indicate a relative light unit (RLU) ratio compared with positive control of 1.0 as the threshold for determining positivity, reclassifying HC2 specimens that are equivocal and fall just above the positive cut-off is one potential way to improve the assay’s specificity (Rijkaart et al., 2010; Rebolj et al., 2011; Sasieni and Castanon, 2011). Some researchers have looked at increasing the RLU cut-off point because HC2 results at this level predict lower risk than higher values (Lorincz et al., 2002). However, equivocal results are not that common and raising the cut-point to 2.0 or 3.0 may only yield a modest improvement in specificity at some cost to sensitivity; i.e., misclassification of women with cervical intraepithelial neoplasia grade 2 or worse lesions whose RLU ratios are equivocal (Rijkaart et al., 2010). Rather than raising the cut-point, a more conservative approach is to repeat the HC2 test on specimens with equivocal results and determine the final test result based on a collective assessment of the test results (Knoepp et al., 2007), however this method requires additional laboratory time and cost to implement, weighing against any savings it yields from reducing the number of colposcopic examinations. In 2001 Kaiser Permanente Northern California added the HC2 assay to its cervical cancer screening program as a triage test for women with a cytologic diagnosis of atypical squamous cells of undetermined significance and, in 2003, as a concurrent test (“cotest”) with cytology for women 30 and older. In the cotesting scheme, women have been managed as follows: 1) women who are HPV-positive with a cytologic diagnosis of atypical squamous cells of undetermined significance, or have more severe (regardless of the HPV test results) are referred for immediate colposcopy; 2) women who test HPV-negative and have cytology of atypical squamous cells of undetermined significance, or who test HPV-positive with negative cytology are rescreened in one year; and 3) if both tests are negative, women are rescreened in three years. For cotests with borderline positive hr-HPV results, interpretation of the repeat test results is used to determine the final hr-HPV test result and hence the need for rescreening visits versus colposcopy. At the time that HC2 was introduced into the Kaiser Permanente Northern California cervical cancer screening algorithm, based on concern over the meaning of borderline-positive HC2-positive results, it was decided that any specimens with borderline or “equivocal” HC2 results, defined as a RLU ratio greater than or equal to 1.0 but less than 3.0, would be re-tested two additional times. The final hr-HPV result was determined by the majority results of the three tests i.e., only one of the two repeats needed to be positive to have the final positive hr-HPV result. To assess the utility of Kaiser Permanente Northern California’s re-testing scheme for equivocal HC2 specimens, HC2 data, cytology results, and histological outcomes (less than cervical intraepithelial neoplasia grade 2, cervical intraepithelial neoplasia grade 2 or worse) were compiled for patients undergoing routine cervical cancer screening at Kaiser Permanente Northern California.
Read full abstract