Cotesting With Pap, HPV Tests More Likely to Detect Cervical Cancer, Precancer

By Marilynn Larkin

July 09, 2020

NEW YORK (Reuters Health) - Human papilloma virus (HPV) or Papanicolaou test by liquid-based cytology (LBC) alone are less likely to detect cervical cancer or precancer (neoplasia or adencocarcinoma in situ) than combining both tests for the same specimen.

A 2015 Quest study (, as well as studies based on data from Magee Women's Hospital (MGW) ( and from Kaiser Permanente Northern California (KNPC) ( all found that co-testing is the most sensitive cervical cancer screening method, Dr. Harvey Kaufman of Quest Diagnostics told Reuters Health by email.

"However," he said, "while the Quest and MGW data found that Pap testing played a significant role in detecting cancer, the researchers analyzing the KPNC data concluded, 'the contribution of cytology (Pap) to screening is shown to be very small.'"

It is important to reconcile the findings, he said. The Quest Diagnostics dataset is more diverse and larger than other data sets, he noted, "with twice as many confirmed cancer cases" as KPNC, which is second largest. "Analysis of cervical screening methods must employ a nationally representative, heterogenous population," he added.

Dr. Kaufman and colleagues assessed the results of co-testing done between 2010 and 2018 by Quest Diagnostics in more than 13 million women age 30 and older. Co-test results preceding a cervical cancer (CxCa) or precancer diagnoses were analyzed and stratified by histopathology.

As reported in the American Journal of Clinical Pathology, 1,615 co-tests preceded 1,259 CxCa diagnoses, and 11,164 co-tests preceded 8,048 cervical precancer diagnoses.

More women who were subsequently diagnosed with CxCa within a year of testing were identified by the LBC than by the HPV result (85.1% vs 77.5%).

Among all women with CxCa, the overall rate of nondetection was 13.1% for cotesting (LBC negative/HPV negative); however, this rate increased substantially when testing exceeded 12 months compared to within one year prediagnosis of either CxCa or precancer.

Commenting on the increased false-negative rate when testing was done more than a year prior to diagnosis, Dr. Damian Alagia, also of Quest, told Reuters Health in an email, "Most cervical cancers develop in the absence of regular screening. Regrettably, many women are not up to date with their screening ( Ensuring women are screened with the most reliable method when they do seek screening may help identify more instances of potential or existing disease. The most reliable method is co-testing with HPV and Pap together."

Dr. Daphne Stewart, a medical oncologist at City of Hope in Duarte, California, told Reuters Health by email, "I think this result is clinically significant...and I recommend co-testing for women over 30."

As for timing of screening, she added, "currently, guidelines recommend co-testing every five years or high-risk HPV or cytology every three years."

"LBC is more likely to be positive when screening is performed within 12 months of diagnosis of cervical cancer," she noted. "This is likely due to larger lesions present closer to diagnosis, and thus more effective sampling of larger lesions. When screening is performed more remotely from the cervical cancer diagnosis, high-risk HPV is more likely to lead to a cervical cancer diagnosis."

"High-risk HPV screening is more effective in detecting CIN3+," she added. "However, 70% of these lesions will not progress to invasive cervical cancer, thus leading to 'overdiagnosis.'"

Quest did not fund the study but Drs. Kaufman and Alagia, and one other coauthor, are employees and own stock in the company.

Kaiser Permanente Northern California declined to comment.

SOURCE: American Journal of Clinical Pathology, online July 8, 2020.