Human papillomavirus testing as triage for atypical squamous cells of undetermined significance and low-grade squamous intraepithelial lesions: sensitivity, specificity, and cost-effectiveness.

OBJECTIVE

Our purpose was to evaluate the cost-effectiveness of the use of a Food and Drug Administration-approved human papillomavirus test in triaging patients with Papanicolaou smears showing atypical squamous cells of undetermined significance or a low-grade squamous intraepithelial lesion for colposcopy compared with an algorithm that used cytologic follow-up.

STUDY DESIGN

Four hundred sixty-two women referred to our Colposcopy Clinic with a Papanicolaou smear report of atypical squamous cells of undetermined significance or a low-grade squamous intraepithelial lesion underwent repeat Papanicolaou smear, cervical colposcopy, directed cervical biopsy, and endocervical curettage. In addition, human papillomavirus testing by the Food and Drug Administration-approved HPV Profile (Digene Diagnostics, Silver Spring, Md.) test was done. A comparison of sensitivity, specificity, and cost-effectiveness of an algorithm determining the need for colposcopy on the basis of repeat cytologic testing versus an algorithm that incorporated repeat cytologic testing and human papillomavirus screening was done. The cost-effectiveness of both of these triage algorithms was also compared.

RESULTS

As expected, high-risk human papillomavirus deoxyribonucleic acid was detected with greater frequency in relation to increasing severity of cervical intraepithelial neoplasia. In 268 women, the follow-up smear obtained in our clinic was reported as negative. High-risk human papillomavirus types were found in 23.5% of these women. In the human papillomavirus-negative women, 5.9% had grade 2 or 3 cervical intraepithelial neoplasia confirmed on cervical biopsy. In comparison, 20.6% of those with a positive result of the human papillomavirus test had grade 2 or 3 cervical intraepithelial neoplasia on biopsy (p < 0.001). Despite this difference, the sensitivity of a positive result of a high-risk human papillomavirus test in predicting the presence of grade 2 or 3 cervical intraepithelial neoplasia was only 52%. Among the women for whom a follow-up clinic Papanicolaou smear was reported as showing atypical squamous cells of undetermined significance or a low-grade squamous intraepithelial lesion, there was no difference in the frequency of biopsy-proved grade 2 or 3 cervical intraepithelial neoplasia between those women with a positive human papillomavirus test result and those with a negative test result. Colposcopy would have been recommended for 194 women because of a repeat clinic smear revealing atypical squamous cells of undetermined significance, a low-grade squamous intraepithelial lesion, or a high-grade squamous intraepithelial lesion, and in 21.6% of these grade 2 or 3 cervical intraepithelial neoplasia was shown on biopsy (sensitivity 63%, specificity 62%). Colposcopy would have been recommended for 180 women because high-risk human papillomavirus or a high-grade squamous intraepithelial lesion was detected at the clinic visit, and in 25% of this group grade 2 or 3 cervical intraepithelial neoplasia was shown on biopsy (sensitivity 67%, specificity 66%). Sensitivity and specificity were virtually identical for the two algorithms, but the cost of human papillomavirus testing was nearly double that of triage based on repeat cytologic testing alone ($692 vs $1246 per case).

CONCLUSION

The Food and Drug Administration-approved HPV Profile test is not a cost-effective triage for patients referred with Papanicolaou smears reported as showing atypical squamous cells of undetermined significance or low-grade squamous lesions.