Cervical screening adherence is relatively low, but safe. Extended intervals are very safe.
Castle PE, Kinney WK, Xue X, et al. Effect of several negative rounds of human papillomavirus and cytology co-testing on safety against cervical cancer: an observational cohort study. Ann Intern Med. 2018;168(1):20-29.
Rendle KA, Schiffman M, Cheung LC, et al. Adherence patterns to extended cervical screening intervals in women undergoing human papillomavirus (HPV) and cytology cotesting. Prev Med. 2018;109:44-50.
Patients who have been screened for cervical cancer for a long time--decades, even--have a diminishing likelihood that cancer will ever be detected. Furthermore, highest-risk patients already have been triaged into further testing or procedures, such as a loop excision electrosurgical procedure or hysterectomy. Two recent studies examined the implications of repeated negative screening and patients' acceptance of extended screening intervals.
Details of the studies
Several negative rounds of cotesting (HPV and cytology) might justify changes to the screening interval. To determine the rate of detection of CIN3, adenocarcinoma in situ, and cervical cancer (≥CIN3) in routine practice after successive negative screening at 3-year intervals, Castle and colleagues looked at records of more than 990,000 women in an integrated health care system who underwent cotesting (HPV and cytology) between 2003 and 2014. They determined that the risk of invasive cervical cancer and ≥CIN3 declined with each round of cotesting; the absolute risk fell more between first and second rounds than between second and third rounds.
At any given round of cotesting, Castle found that the ability to reassure a patient about cancer and cancer risk was similar when looking at an HPV result alone, whatever the cytology or HPV-cytology cotest result was. The investigators concluded that similar patterns of risk would have been seen had stand-alone HPV testing been used, instead of co-testing, (HPV testing alone might have missed a few cases of CIN3 and adenocarcinoma in situ leading to cancer). A single negative cotest was so effective at ruling out ≥CIN3 and cervical cancer that, after a second round of cotesting, they found that no interval cancer cases were detected among women who had a negative HPV result.
Women aged 50 years or older had a 5- to 6-fold lower risk after their third consecutive negative cotest than women aged 30 to 39 years had after their first negative cotest. These data support the ideas, Castle noted, that 1) assigning screening intervals based on both age and number of previous negative screens and 2) extending the screening interval even further than 3 years after 2--perhaps even after 1--negative cotests or HPV tests are worth entertaining. Screening women of this age becomes inefficient and cost-ineffective, even at 5-year intervals.
Is patients' adherence to an extended interval of cotesting reliable enough to change practice? Rendle and colleagues examined the records of more than 491,000 women (in the same integrated health care system that Castle studied) who had undergone routine cervical cancer screening between 2003 and 2015. Their goal was to determine how high adherence had become to the system's recommendation of an every-3-year screening interval--an interval that mirrors long-standing guidelines elsewhere.
In short, researchers observed increasing and relatively rapid clinical adoption of every-3-year cotesting for routine cervical screening over time; between 2003 and 2009, the cohort grew significantly less likely overall to come in early for screening. In this setting, adoption of an extended screeninginterval appears to run counter to earlier understanding that patients are likely to resist such extension.
Women aged 60 to 64 were most likely to screen early across 2 consecutive intervals. What Rendle termed a "modest" decrease in the percentage of late screeners (but still within a 5-year interval) was also noted during adoption of the 3-year interval.
What next?
Molecular-based testing. Research, mostly outside of the United States, is taking us in the direction of molecular-based technologies as at least a component of cervical cancer screening. Today, we rely mostly on Pap tests and colposcopy, but these are subjective screens, with a human operator. With molecular testing (mostly of components of HPV), results are objective--a "Yes" or "No" finding based on clinically validated thresholds. Methods such as genotyping, P16INK4a/Ki-67 gene product dual-stain cytology, and testing for E6 and E7 HPV mRNA transcripts are in development, and hold promise to allow us to screen safely using almost completely molecular testing, thus eliminating human error and subjectivity and enriching the population that needs further management with very sensitive and potentially specific testing.
We are being presented with the possibility that almost all aspects of screening can be done without a provider, until the patient needs treatment.
Access to screening. Research is also looking at improving access, such as self-sampling for primary screening. That includes home cervical and vaginal sampling, with specimens mailed to the laboratory, from where results and follow-up instructions as communicated to patients. The Netherlands and the United Kingdom are moving to self-sampling primary screens; the United States is not--yet. But that is the direction research is taking us.
Modified guidelines. Eyes are on the work of the USPSTF. Last year, the Task Force issued draft recommendations (https://www.uspre ventiveservicestaskforce.org/Page/Document/draft-recommendation-statement/cervical-cancer-screening2#clinical), followed by a comment period (now closed), for updating 2012 cervical cancer screening guidelines in a way that would trigger a major change in clinical practice. Those draft recommendations and public comments are under review; final recommendations are possible within this calendar year.
WHAT THIS EVIDENCE MEANS FOR PRACTICE
Continue to follow current screening guidelines; they are safe and effective for preventing cervical cancer. This assumes adherence to intervals, which is both the provider's and the patient's responsibility: First, less is more; too much screening ("I've always done it this way") can be harmful. Second, screening at intervals set by the guidelines is extremely safe, despite earlier reports or provider concerns that suggest otherwise.
Patients who have undergone several rounds of negative screening have a markedly diminished risk of cervical cancer. Serve them best by performing this underutilized gyn procedure: Sit on your hands.
Be aware that winds of change are blowing: What constitutes appropriate screening intervals is up for discussion this year, and molecular-based testing technologies that are under investigation have the potential to someday be a vast improvement over current good, but subjective, interpretations of results.
Last, promote primary prevention of cervical cancer with HPV vaccination in your practice to increase the percentage of protected patients. Doing so will contribute not only to their long-term health but also, at a societal level, to a herd immunity effect.5 Any positive HPV infection in a future of a well-vaccinated population will be significant, and HPV-targeted technologies to identify the highest risk women will be the most efficient screening.