Expert Commentary

The beginning of the end of the Pap?

Author and Disclosure Information

It seems so. Investigators reviewed extended data from a large prospectively followed cohort of women within the Kaiser Permanente Northern California cohort to quantify the relative contributions of the cytology and human papillomavirus (HPV) test components of cotesting for detecting cervical precancer and cancer to help guide whether cotesting, with its costs and potential harms, should be recommended. Although bringing an end to standard cotesting presents challenges for clinicians and laboratories alike, the researchers found that cytology had limited value.

Schiffman M, Kinney WK, Cheung LC, et al. Relative performance of HPV and cytology components of cotesting in cervical screening [published online ahead of print November 14, 2017]. Natl Cancer Inst. doi: 10.1093/jnci/djx225.




Realistic prospective performance data are needed to quantify the additional benefit of the cytology component of cotesting on top of what is already known to be highly sensitive molecular HPV testing. While the addition of cytology to HPV testing can add performance, it also can add further costs and the potential for unnecessary colposcopies for what are merely cytomorphologic manifestations of an active HPV infection. Frequent invasive procedures such as colposcopy, which can be costly and lead to anxiety and distress in generally young women and the potential for overtreatment of likely regressive lesions, has been defined as a harm of screening by the US Preventive Services Task Force (USPSTF).

Details of the study

In a cohort from Kaiser Permanente Northern California, 1,208,710 women aged 30 years or older were screened with cotesting from 2003 to 2015. Those who cotested HPV negative and cytology negative were offered triennial screening. Positive cotest results were managed according to Kaiser protocol. Women with cytologic abnormalities were referred for colposcopy. Those with HPV positive/cytology negative results or HPV negative/cytology equivocal results underwent accelerated testing at 1 year. A total of 623 cervical cancers were identified and included in the analyses.

Using multiple analyses, Schiffman and colleagues demonstrated the sensitivity advantage of HPV testing. They clearly showed that the cytology component to cotesting performance over many years is very limited for detecting precancers and early curable cancers. For example, prediagnostic HPV testing (76.7%) was more likely to be positive than cytology (59.1%; P<.001 for paired comparison); 82.6% of all prediagnostic cotests were positive by HPV and/or cytology; and only 5.9% of the cotests were positive by cytology alone (HPV negative.)

Primary HPV testing is recommended as a potential screening strategy by an interim guidance group led by the Society of Gynecologic Oncology and the American Society for Colposcopy and Cervical Pathology, and it is the primary cervical cancer screening recommendation of USPSTF draft guidelines. 1 There have been reports that reliance on primary HPV testing would encourage cervical cancer mortality; Schiffman and colleagues point out, however, that according to their study data, such reports are overstated.

Despite these data, practically speaking, shifting away from standard cotesting poses numerous challenges for clinicians and laboratories alike; however, these data clearly show the limited value of cytology and, due to the overtreatment of likely regressive cervical intraepithelial neoplasia grade 2, the possible increased risk of preterm birth and its subsequent harm as well.

Study strengths and weaknesses

The authors examined the long-term relative history of HPV testing and cytology prior to cancer diagnosis in a large, prospectively followed US cohort where hundreds of women in this cohort developed cancer. There will not be a validation study of this size and scale in the near future. Further, the authors showed that the relative value of cytology to cotesting is minimal. Multiple subsequent rounds of cotesting after negative results also can be questioned.

Next Article: