Comparison of the Clinical Performance of Aptima HPV Assay and the Cobas 4800 Platform in Women with Normal Cytology and Positive High-Risk HPV.

INTRODUCTION

According to the current American Society for Colposcopy and Cervical Pathology (ASCCP) guidelines, patients with normal cytology results may be referred for colposcopy according to their high-risk human papillomavirus (hrHPV) test results. A higher positive predictive value (PPV) of hrHPV has significance for preventing unnecessary colposcopic examinations. Several studies have compared the performance of the Aptima assay and the Cobas 4800 platform among patients who had minor cytologic abnormalities. However, in our English literature search, we found no other study that had been conducted to compare these two methods in patients with normal cytology. We thus aimed to compare the PPV of the Aptima assay and the Cobas 4800 platform among women with normal cytology.

METHODS

Between September 2017 and October 2022, we retrospectively identified 2,919 patients who had normal cytology and hrHPV positivity and had been referred for a colposcopy. Among them, 882 agreed to undergo a colposcopy; on examination, 134 had target lesions revealed and underwent a colposcopic punch biopsy.

RESULTS

Among the patients who underwent a colposcopic punch biopsy, 49 (38.9%) were tested with Aptima, and 77 (61.1%) were tested with Cobas. In the Aptima group, 29 (59.2%) patients showed benign histology, 2 (4.1%) patients had low-grade squamous intraepithelial lesions (LSILs), and 18 (36.7%) patients had ≥high-grade squamous intraepithelial lesions (HSILs) biopsy results. The false-positivity rate and PPV of Aptima were 63.3% (31/49) and 36.7% (95% confidence interval [CI]: 0.232-0.502), respectively, for a histopathologic diagnosis of ≥HSIL. In the Cobas group, 48 (62.3%) biopsies were benign, 11 (14.3%) reported an LSIL, and 18 (23.4%) biopsies were ≥HSIL. The false-positivity rate and PPV of Cobas were 76.6% (59/77) and 23.4% (95% CI: 0.139-0.328), respectively, concerning a ≥HSIL tissue diagnosis. The false-positivity rate of Aptima HPV 16 positivity was 40% (4/10). The false-positivity rate of Cobas HPV 16 positivity was 61.1% (11/18). The PPVs of HPV 16 positivity for Aptima and Cobas were 60% (95% CI: 0.296-0.903) and 38.9% (95% CI: 0.163-0.614), respectively, concerning ≥HSIL tissue diagnosis.

CONCLUSION

We recommend analyzing the performances of hrHPV platforms in future, larger studies in patients with normal cytology rather than only in cases with abnormal cytology.