A model-based analysis of the health impacts of COVID-19 disruptions to primary cervical screening by time since last screen for current and future disruptions

  1. Emily A Burger
  2. Inge MCM de Kok
  3. James F O'Mahony
  4. Matejka Rebolj
  5. Erik EL Jansen
  6. Daniel D de Bondt
  7. James Killen
  8. Sharon J Hanley
  9. Alejandra Castanon
  10. Mary Caroline Regan
  11. Jane J Kim
  12. Karen Canfell
  13. Megan A Smith

  • Curated by eLife

    eLife logo

    Evaluation Summary:

    This paper describes the use of three well-established mathematical models of cervical cancer to estimate the impact of COVID-19 related-delays in screening access on cervical cancer incidence and delays in diagnosis. Consistent with previous work and the known biology of cervical cancers, the findings that short delays have relatively small effects on population-level cervical cancer risk are reassuring overall, but the impact of screening interval and screening test performance suggest that existing disparities related to screening access may be exacerbated. These results should be useful for policy makers in planning responses to future pandemics or other sources of sudden restriction of screening availability.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1 agreed to share their name with the authors.)

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

We evaluated how temporary disruptions to primary cervical cancer (CC) screening services may differentially impact women due to heterogeneity in their screening history and test modality. We used three CC models to project the short- and long-term health impacts assuming an underlying primary screening frequency (i.e., 1, 3, 5, or 10 yearly) under three alternative COVID-19-related screening disruption scenarios (i.e., 1-, 2-, or 5-year delay) versus no delay in the context of both cytology-based and human papillomavirus (HPV)-based screening. Models projected a relative increase in symptomatically detected cancer cases during a 1-year delay period that was 38% higher (Policy1-Cervix), 80% higher (Harvard), and 170% higher (MISCAN-Cervix) for underscreened women whose last cytology screen was 5 years prior to the disruption period compared with guidelines-compliant women (i.e., last screen 3 years prior to disruption). Over a woman’s lifetime, temporary COVID-19-related delays had less impact on lifetime risk of developing CC than screening frequency and test modality; however, CC risks increased disproportionately the longer time had elapsed since a woman’s last screen at the time of the disruption. Excess risks for a given delay period were generally lower for HPV-based screeners than for cytology-based screeners. Our independent models predicted that the main drivers of CC risk were screening frequency and screening modality, and the overall impact of disruptions from the pandemic on CC outcomes may be small. However, screening disruptions disproportionately affect underscreened women, underpinning the importance of reaching such women as a critical area of focus, regardless of temporary disruptions.

Article activity feed

  1. Version published to 10.7554/elife.81711 on eLife
  2. eLife

    Author Response

    Reviewer #2 (Public Review):

    Burger et al. present their compilation of 3 well established cervical cancer natural history micro-simulation models from the US (Harvard), Denmark (Miscan) and Australia (Policy 1) to evaluate what effect Covid, or any systemic problem impeding screening over a time duration, will have on cervical cancer incidence ("symptomatic cervical cancer") in the short and long term. They use the United States for the modeling example and establish that a temporary screening delay has less deleterious effect on cervical cancer incidence and morbidity than being under-screened. Screening test and previous screening frequency also impact on the outcomes.

    The authors evaluate a number of factors in their analysis:

    1. Screening test type: HPV (every five years) or cytology (every three years), as per guidelines.
    1. Screening delay such as with Covid: 1, 2, or 5 years from the participant's last screening encounter.
    1. The participant screening frequency: 1, 3, 5 or 10 yearly.
    1. Three birth cohorts: 35yo, 45yo and 55yo in 2020.

    As the Covid pandemic meant a delay of months toward a year, a key finding was the projected relative increase in symptomatic cervical cancer cases from a year delay which varied from 38% higher with Policy 1, to 170% higher with Miscan-Cervix. The comparison was for women who had not had cytology screening for 5 years before the delay versus those appropriately screened at 3 years. In the long term, over a lifetime, a one year (up to 5 year) delay, had less effect on developing cervical cancer than screening frequency or test type. This finding is reassuring for the general public. Most importantly, however, the authors showed that not being screened for a long duration (underscreened) is the most significant factor to developing cervical cancer, especially with a further systemic delay such as Covid. Being under, or never screened, is a clinically well known fact in the cervical cancer screening community. HPV screening type was also shown to be more protective against developing cervical cancer given its superior sensitivity for longer duration over cytology allowing HPV to be done every 5 years versus cytology at 3 years.

    The strength of the paper is showing the above findings through the multiple permutations of effects in detailed analytic tables for quantitative mathematical modeling experts, and summary figures simple enough for a more general reader to follow. The variation in results among the models was explained well with most of the difference due to the "dwell" time before an HPV infection develops into precancer and cancer, the Miscan model having the shortest dwell time and thus some of the higher relative rate and absolute increases in cancer. The authors emphasized that "heterogeneity" in screening history could be due to socioeconomic factors that aren't directly evaluated in the model, but women with greater socioeconomic barriers, tend to be those that are under-screened and most at risk of developing cervical cancer.

    The results are grouped into short and long term impacts. For the short term impact, the authors concentrated on showing excess cancer in women screening less frequently than guidelines, and used cytology every 3 years as a baseline. So women screening 5 or 10 years before disruption, did worse than q3 yearly guideline compliant women. There was then discussion about guideline compliant HPV screening which is done every 5 years, so only the 10 year group was non compliant. The authors discuss changing to HPV at 30 yo. Without knowing the actual guidelines for screening in the US, this section can be a bit confusing for the reader. It would be very helpful if the authors clearly state that cytology is offered every 3 years to women under 30, starting at 21 yo, and then HPV is offered q 5 yearly from 30yo. Alternatively, q3 yearly cytology can be done throughout a screening lifetime. This background information makes the short term results clearer to understand. The Figure is helpful and clear for interpretation.

    For the long term impacts, the authors are able to show that a temporary disruption in screening is less deleterious than overall poor screening history (not following guidelines). They also show that HPV testing from age 30 is better at preventing cervical cancer than 3 yearly cytology, and had less impact from a screening delay. (the notation to figure reads right panel but is likely Lower panel).

    Thank you for flagging this typo; we have changed the notation.

    Overall, the authors clearly show the effects of a temporary screening disruption in the context of a women's overall screening history, frequency and test node.

    This work is very relevant and timely in the cervical screening field and emphasizes the importance of assuring women are not under-screened, the greatest risk factor for cervical cancer. They give a comprehensive discussion of how their results are relevant for cervical cancer screening today and in the near future.

    Thank you for the nice summary and feedback.

    As alluded to earlier, clarification about the age related switch to HPV testing at 30yo would help the reader better understand the point about the two factors having to be balanced when considering HPV testing. Are the two factors the greater protective test sensitivity vs the benefit of the actual screening moment? This section was slightly confusing.

    In addition to the request for a more complete description of the US guidelines (addressed in Essential Revisions), we have clarified the description of the “2 competing factors” on page 6-7 of the manuscript.

    “Similarly, we found that the impact of disrupting an HPV-based screening program has different implications than the disruption of a cytology-based program. This can be explained by the fact that HPV screening has a higher sensitivity to detect (pre-invasive) cervical lesions than cytology; therefore, the cancer risk at time of disruption is lower (as there are fewer undetected lesions) and this may provide a greater buffer to endure temporary disruptions. On the other hand, in case of the more sensitive HPV test, disruption takes away a relatively more valuable (i.e., sensitive) screening moment. The balance between these two factors causes a greater or smaller excess risk per delay duration in case of HPV screening compared to cytology screening, which contributes to the within model differences of cytology-based versus HPV-based screening in Table 1. If in a model the first effect (HPV screening contributes to lower risk at the time of disruption) is larger than the second effect (removal of a valuable screening moment), disruption of the HPV program would have a smaller effect than that of the cytology program, which is the case for all screening frequencies in both the Harvard and Policy1-cervix models and the annual screeners in MISCAN-cervix (Table 1). The MISCAN-Cervix model predicted relatively more excess cancers for women screened with HPV 3-yearly, 5-yearly or 10-yearly due to disruptions, where delaying a more sensitive test (the second factor) seems to outweigh the first (less underlying disease at the time of the disruption). Differences in dwell time for HPV and cervical precancer among the three models predominately contributes to this balance between the two factors (Appendix), where the MISCAN-Cervix model has the shortest preclinical dwell time from HPV acquisition to cancer development (20). In addition to the shorter dwell times, the MISCAN model also assumes that some precancerous lesions are structurally missed over time by cytology-based screening because they are located deeper into the cervical canal. For women with such lesions, missing a cytology screen due to a disruption is less harmful, which increases the relative difference between primary cytology and primary HPV screening in case of a disruption, and increases the effect of women missing a very sensitive screen (second factor).”

    The authors speak to self collection as a potential solution for some underscreened women (people with cervix). It would be important to outline how self sampling is actually done. Some people believe cytology can be done on self sample. Self sampling can also include urine HPV, thus some detail about self sampling in the discussion would be helpful and give another benefit to HPV testing (DNA based for self sampling).

    Although there is research interest in urine-based HPV screening, this is not yet at the point where it has been widely implemented in screening programs; however, we agree some additional information on self-sampling would be helpful for the general reader.

    On page 7 we have added, “Importantly, vaginal HPV-based screening (unlike cytology-based screening) enables self-collection of samples at home, which may provide a tool to reduce screening barriers and facilitate outreach to under-screened people who are also most vulnerable to screening disruptions.”

  3. eLife

    Evaluation Summary:

    This paper describes the use of three well-established mathematical models of cervical cancer to estimate the impact of COVID-19 related-delays in screening access on cervical cancer incidence and delays in diagnosis. Consistent with previous work and the known biology of cervical cancers, the findings that short delays have relatively small effects on population-level cervical cancer risk are reassuring overall, but the impact of screening interval and screening test performance suggest that existing disparities related to screening access may be exacerbated. These results should be useful for policy makers in planning responses to future pandemics or other sources of sudden restriction of screening availability.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1 agreed to share their name with the authors.)

  4. eLife

    Reviewer #1 (Public Review):

    In this collaborative and comparative modeling paper, three groups of investigators with well-validated mathematical models of the natural history of cervical cancer explored the potential impact of disruptions in screening services such as those associated with COVID-19 on cancer incidence. Given known disparities in access to regular screening in the United States, the authors were particularly interested in identifying heterogeneity of effects - would externally imposed restrictions on screening have a disproportionate effect on women already at increased risk because of access issues such as prolonged intervals between screening, or reliance on less sensitive screening tests?

    Strengths:

    --The authors used three existing, well-validated cervical cancer natural history to compare results. This comparative approach, used by these authors as well as other collaborators within NCI's Cancer Intervention and Surveillance Modeling Network (CISNET), improves confidence in the overall validity and robustness of the results, given qualitatively similar findings across models that differ in terms of structure and underlying assumptions.
    --The models have previously been used in the context of US screening policy.
    --The models used birth cohorts as well as screening frequency, which accounts for age-period-cohort effects on both risk of HPV and cervical cancer as well as competing risks such as other causes of mortality and hysterectomy.
    --Cervical cancer screening both detects pre-malignant lesions and allows prevention of cervical cancer, leading to decreased incidence, and, for those lesions which have progressed to invasive cancer, detects asymptomatic lesions, leading to decreased morbidity and improved survival. The use of "symptomatically detected cancers" as the primary outcome of interest is appropriate.
    --The qualitative results are consistent with previous modeling results in the context of screening program design--the effects of a short-term delay in screening are greatest for women with a longer time since the most recent screen, or for women screened with less sensitive (cytology) compared to more sensitive (HPV) modalities. These findings were true for both short- and long-term impacts.
    --The policy recommendation to prioritize outreach and appointment availability for catch-up when restrictions are lifted to women who do not have up-to-date screening according to guidelines is supported by the findings.

    Limitations:

    --The limitations are, for the most part, those inherent in any modeling exercise and are well described and discussed by the authors.
    --As the authors note, the models do not explicitly incorporate disparate impacts by race/ethnicity or other social determinants of health, and thus cannot explicitly highlight disparities within specific groups.
    --Potential effects on cervical cancer mortality are not captured. Given the high survival of stage I cervical cancer and, in most cases, the relatively slow progression of disease, it seems plausible that even an increase in symptomatically diagnosed disease will not have a detectable effect on mortality if there is not a shift in stage distribution; however, given that treatment of invasive cervical cancer has much greater risk of short- and long-term morbidity compared to treatment of preinvasive lesions, there is likely to be an impact on quality of life if not survival.
    --Related, if the factors affecting underscreening are ALSO associated with delays in care once symptoms develop, there is a potential for disparate effects on morbidity and mortality as well.

    These results should prove useful to policy makers, clinicians, and patients, both in helping identifying women for prioritizing access to screening services when availability is constrained or restored, and for reassuring those women who do have up-to-date screening that delays are unlikely to significantly affect their risk of developing cervical cancer.

  5. eLife

    Reviewer #2 (Public Review):

    Burger et al. present their compilation of 3 well established cervical cancer natural history micro-simulation models from the US (Harvard), Denmark (Miscan) and Australia (Policy 1) to evaluate what effect Covid, or any systemic problem impeding screening over a time duration, will have on cervical cancer incidence ("symptomatic cervical cancer") in the short and long term. They use the United States for the modeling example and establish that a temporary screening delay has less deleterious effect on cervical cancer incidence and morbidity than being under-screened. Screening test and previous screening frequency also impact on the outcomes.

    The authors evaluate a number of factors in their analysis:

    1. Screening test type: HPV (every five years) or cytology (every three years), as per guidelines.
    2. Screening delay such as with Covid: 1, 2, or 5 years from the participant's last screening encounter.
    3. The participant screening frequency: 1, 3, 5 or 10 yearly.
    4. Three birth cohorts: 35yo, 45yo and 55yo in 2020.

    As the Covid pandemic meant a delay of months toward a year, a key finding was the projected relative increase in symptomatic cervical cancer cases from a year delay which varied from 38% higher with Policy 1, to 170% higher with Miscan-Cervix. The comparison was for women who had not had cytology screening for 5 years before the delay versus those appropriately screened at 3 years. In the long term, over a lifetime, a one year (up to 5 year) delay, had less effect on developing cervical cancer than screening frequency or test type. This finding is reassuring for the general public. Most importantly, however, the authors showed that not being screened for a long duration (underscreened) is the most significant factor to developing cervical cancer, especially with a further systemic delay such as Covid. Being under, or never screened, is a clinically well known fact in the cervical cancer screening community. HPV screening type was also shown to be more protective against developing cervical cancer given its superior sensitivity for longer duration over cytology allowing HPV to be done every 5 years versus cytology at 3 years.

    The strength of the paper is showing the above findings through the multiple permutations of effects in detailed analytic tables for quantitative mathematical modeling experts, and summary figures simple enough for a more general reader to follow. The variation in results among the models was explained well with most of the difference due to the "dwell" time before an HPV infection develops into precancer and cancer, the Miscan model having the shortest dwell time and thus some of the higher relative rate and absolute increases in cancer. The authors emphasized that "heterogeneity" in screening history could be due to socioeconomic factors that aren't directly evaluated in the model, but women with greater socioeconomic barriers, tend to be those that are under-screened and most at risk of developing cervical cancer.

    The results are grouped into short and long term impacts. For the short term impact, the authors concentrated on showing excess cancer in women screening less frequently than guidelines, and used cytology every 3 years as a baseline. So women screening 5 or 10 years before disruption, did worse than q3 yearly guideline compliant women. There was then discussion about guideline compliant HPV screening which is done every 5 years, so only the 10 year group was non compliant. The authors discuss changing to HPV at 30 yo. Without knowing the actual guidelines for screening in the US, this section can be a bit confusing for the reader. It would be very helpful if the authors clearly state that cytology is offered every 3 years to women under 30, starting at 21 yo, and then HPV is offered q 5 yearly from 30yo. Alternatively, q3 yearly cytology can be done throughout a screening lifetime. This background information makes the short term results clearer to understand. The Figure is helpful and clear for interpretation.

    For the long term impacts, the authors are able to show that a temporary disruption in screening is less deleterious than overall poor screening history (not following guidelines). They also show that HPV testing from age 30 is better at preventing cervical cancer than 3 yearly cytology, and had less impact from a screening delay. (the notation to figure reads right panel but is likely Lower panel).

    Overall, the authors clearly show the effects of a temporary screening disruption in the context of a women's overall screening history, frequency and test node.

    This work is very relevant and timely in the cervical screening field and emphasizes the importance of assuring women are not under-screened, the greatest risk factor for cervical cancer. They give a comprehensive discussion of how their results are relevant for cervical cancer screening today and in the near future.

    As alluded to earlier, clarification about the age related switch to HPV testing at 30yo would help the reader better understand the point about the two factors having to be balanced when considering HPV testing. Are the two factors the greater protective test sensitivity vs the benefit of the actual screening moment? This section was slightly confusing.

    The authors speak to self collection as a potential solution for some underscreened women (people with cervix). It would be important to outline how self sampling is actually done. Some people believe cytology can be done on self sample. Self sampling can also include urine HPV, thus some detail about self sampling in the discussion would be helpful and give another benefit to HPV testing (DNA based for self sampling).

  6. Version published to 10.1101/2022.07.25.22278011v1 on medRxiv