DNA Tests for HPV Improve Cervical Cancer Screening

October 4, 2007

AMSTERDAM -- Testing for the DNA of human papillomavirus -- combined with a Pap smear -- reveals the lesions that lead to cervical cancer earlier than a Pap smear alone, researchers said here.

AMSTERDAM, Oct. 4 -- Testing for the DNA of human papillomavirus -- combined with a Pap smear -- reveals the lesions that lead to cervical cancer earlier than a Pap smear alone, researchers said here.

The finding, from a randomized trial of more than 44,000 women, appears to settle the controversy over whether the extra lesions detected by DNA testing are "regressive" and will not progress to cervical cancer, according to Chris Meijer, M.D., of the VU University Medical Center, and colleagues.

Adding the DNA tests to conventional cytological screening "leads to earlier detection of clinically relevant cervical lesions," Dr. Meijer and colleagues wrote online in The Lancet.

It also might mean that the five-year recommended interval between screening tests could be extended, the researchers argued.

The data came from the Population Based Screening Study Amsterdam, which enrolled 44,938 women from January 1999 through September 2002 and randomized to get either a Pap smear alone or a Pap smear with DNA tests for the high-risk strains of papillomavirus.

Five years later, at the next recommended screening, women in the study were all given both tests. The primary endpoint was the rate of cervical intraepithelial neoplasia of grade three or higher (CIN3+).

As of February 2007, follow-up of more than 6.5 years was available for 17,155 women -- or slightly fewer than half of the study participants, the researchers said.

In the first round of tests, the rate of CIN3+ among the 8,575 women who got both tests was 70% higher than among the 8,580 who just got a Pap smear -- a difference that was statistically significant at P=0.007.

On the other hand, the rate five years later was 55% lower in the intervention group, which was significant at P=0.001, although the total number of CIN3+ lesions was not significantly different over the two screening rounds.

The researchers also calculated the number of referrals for colposcopy and the number of CIN3+ lesions per referral, as a way of estimating the efficiency of the two strategies.

In the first round, the number of referrals in the intervention group was significantly higher (P<0.0001) than in the control group, but the rate of lesions per referral was not different.

In the following round of tests, the number of referrals was lower in the intervention group (again significantly at P<0.003) than among controls and the number of lesions per referral was also significantly lower (P=0.03).

On the basis of those data, the researchers argued that the interval between screenings could be extended from five to six years if both forms of test were used.

The study fills in a missing link in the understanding of DNA testing for human papillomavirus, commented Guglielmo Ronco, M.D., Ph.D., and Nereo Segnan, M.D., both of the Unit of Cancer Epidemiology at CPO Piemonte in Turin, Italy.

Writing in an accompanying editorial, they said there has been controversy over whether the additional lesions found using DNA testing for human papillomavirus had any clinical relevance.

"The proof of early detection of a substantial proportion of persistent CIN3+ lesions is the crucial step that was missing until now," they said.

The finding could "lead to greater protection against invasive cervical cancer because more lesions could be detected before they have progressed," they added.

Dr. Ronco reports financial links with GeneProbe, which is developing a test for HPV RNA.