LOS ANGELES – Testing for high-risk human papillomavirus appears to be more reliable than performing liquid-based cytology for assessing a woman’s risk of developing advanced cervical neoplasia, based on the final results of the ATHENA trial.

In the prospective cohort study, both types of testing were performed in 42,209 women aged 25 years or older presenting for routine cervical cancer screening. A subset of these women had annual follow-up including repeat testing and colposcopy over 3 years.

The cumulative incidence rate of grade 3 cervical intraepithelial neoplasia (CIN3) or more advanced neoplasia was 0.33% in women having negative results for high-risk HPV, but much higher at 0.77% in women having negative liquid-based cytology results, regardless of HPV results. In addition, combining the two types of testing yielded very little gain over HPV testing alone; the rate was 0.29% in women with negative results on both tests.

"The adjusted 3-year cumulative incidence rate of CIN3 or greater in high-risk HPV-negative women is less than half that of women who have liquid based cytology that is negative," said lead author Thomas C. Wright Jr. at the annual meeting of the Society of Gynecologic Oncology. "Co-testing with both liquid-based cytology and high-risk HPV provides very little benefit over using high-risk HPV testing alone for screening."

"These results, we believe, suggest that high-risk HPV testing and genotyping is superior to liquid-based cytology for cervical cancer screening. And one of the goals of this trial is to go to the FDA and to get approval for HPV with genotyping for primary screening," added Dr. Wright, director of the division of gynecologic and obstetrical pathology at Columbia (N.Y.) University Medical Center.

"Certainly, the new American Cancer Society, ASCCP [American Society for Colposcopy and Cervical Pathology], and ASCP [American Society for Clinical Pathology] guidelines recommend co-testing, which is cytology and HPV, as a preferred screening approach for women 30 years and older," Dr. Wright noted. "I think the data that I showed today suggest that cytology adds so little that we would really do well to look at primary HPV screening. It’s going to be up to regulatory agencies though, as well as societies such as the American Cancer Society to make risk-benefit analyses. But, certainly, the data look very promising."

The 3-year cumulative incidence rate of CIN3 or worse was about 10% overall among patients having positive test results for any of the 14 high-risk types of HPV assessed. The main driver, however, was positivity for HPV 16 – women positive for this HPV type had a CIN3 or worse rate of 25%. HPV 18 was a lesser but still major contributor, associated with a rate of nearly 11%. The rate was 5% for women positive for any of the other 12 high-risk HPV types.

The data show "the disproportionate impact of HPV 16 positivity on the 3-year cumulative incidence rate among individuals who are high-risk HPV positive," Dr. Wright commented. "It is actually well over double that of women who have the 14 other types of high-risk HPV pooled, and compared to the other high-risk types, you have a fivefold increase."

The trial was the U.S. regulatory trial to obtain FDA clearance for the cobas HPV Test (manufactured by Roche Molecular Systems), which detects 14 types of HPV associated with a high risk of cervical neoplasia and cancer.

The analyses reported were based on 42,209 women aged 25 years or older on enrollment. They had a median age of 41 years, and 83% were white. At baseline, 10% had a high-risk HPV infection as assessed by the cobas test and 6% had atypical squamous cells of undetermined significance (ASCUS) or more advanced cervical neoplasia as detected by cytology.

The widely differing rates of CIN3 or worse for women who were high-risk HPV negative versus liquid-based cytology negative "are very similar to what has been seen in a number of large European studies," Dr. Wright noted.

And the temporal trends for the various high-risk HPV groups generally mirror those observed in a previous series of women seen at a Kaiser facility who had 10 years of follow-up (J. Natl. Cancer Inst. 2005;97:1072-9). "High-risk HPV-negative women remain very low [risk] in the ATHENA as well as in the Kaiser series," he pointed out.

In his post-presentation discussion, Dr. Wright was asked whether the study findings lend "a further push for school-based HPV vaccines in the United States."

"I’m not sure how my data address that," he responded. "However, I firmly believe that we should have school-based HPV vaccination in the United States. I think that most people in this room would believe that the percentage of our vaccinated cohort in the United States is much smaller than it should be; some people would use the word ‘appalling’ to refer to the rate at which we have vaccinated. And I think that societies such as the Society of Gynecologic Oncology should take the lead in trying to" improve vaccine uptake.

Dr. Wright disclosed that he is a consultant for Roche Molecular Systems, BD Diagnostics, Cepheid, and Hologic. The trial was sponsored by Roche Molecular Systems.

HPV testing best predicts cervical neoplasia risk