To compare 3 management strategies for a diagnosis of atypical squamous cells of undetermined significance (ASCUS).
Human papillomavirus (HPV) testing is at least as sensitive as immediate colposcopy for detecting cervical intraepithelial neoplasia (CIN) grade 3—and refers about half as many women to colposcopy. Follow-up using repeat cytology is sensitive at an ASCUS referral threshold, but requires 2 follow-up visits and ultimately more colposcopic examinations than HPV triage.
A total of 3,488 women with community-acquired conventional cervical cytology interpreted as ASCUS were randomized to immediate colposcopy; triage based on enrollment HPV testing and liquid-based cytology, with referral to colposcopy for a finding of high-grade squamous intraepithelial lesion (HSIL); or repeat cytology at a referral threshold of HSIL (ie, conservative management).
After 2 years, the cumulative diagnosis of CIN grade 3 was 8% to 9% in all 3 study arms. The immediate colposcopy group had a 53.6% sensitivity for CIN grade 3, versus 72.3% for HPV triage and 54.6% for conservative management. The HPV triage strategy referred 55.6% of women for colposcopy, while the conservative strategy referred 12.3%.
Since the 1991 introduction of the Bethesda System for cervical cytology classification, there has been disagreement on the best method to further evaluate womenwith a diagnosis of ASCUS. Commonly proposed triage strategies are immediate colposcopy, repeat cytology, or HPV testing. The importance of determining the best method is clear, since 5 million women per year have an ASCUS finding on cervical cytology.
This study’s great strength is its design and low dropout rate: a randomized prospective evaluation of a large group of women, few of whom did not complete the protocol.
This trial demonstrates that each of the 3 strategies is equally effective in detecting CIN grades 2 and 3 after 2 years. The investigators estimated that, of the women in the trial who were ultimately found to have CIN 3, HPV testing would have properly triaged 92.4% and referred 53.1% for colposcopy. Repeating cytology twice would have provided the same threshold for detecting CIN 3, but would have resulted in colposcopy for 67.1% of women.
Study findings indicate that HPV testing is an effective method to triage women with ASCUS cytology. Only 1.4% of patients who were HPV-negative at enrollment were found to have CIN 3 during the 2 years of study.
Cost considerations. Unfortunately, in this report, the ALTS group did not present any data concerning cost-effectiveness. It seems reasonable, however, that because fewer colposcopies were performed in the HPVtesting arm, this strategy would be the most efficient. Since all the triage methods were equally effective at diagnosing CIN 3, the least costly should be preferred.
High- versus low-grade lesions. It is important to note that in the repeat cytology arm (which required a high-grade cytology for triage to colposcopy), one third of women with CIN 3 were diagnosed at the time of exit from the study by colposcopy or loop electrosurgical excision procedure that was performed due to persistent low-grade lesions. This finding calls into question the sensitivity of using highgrade histology to trigger colposcopy when following patients with ASCUS cytology.
For women with ASCUS cervical cytology, reflex HPV testing is as efficient as immediate colposcopy or repeat cytology and is most likely less costly. Thus, it is the preferred strategy for women with ASCUS cytology.