key: cord-0967983-udbvhkus authors: Tota, Joseph E.; Isidean, Sandra D.; Franco, Eduardo L. title: Defining benchmarks for tolerable risk thresholds in cancer screening: Impact of HPV vaccination on the future of cervical cancer screening date: 2020-07-16 journal: Int J Cancer DOI: 10.1002/ijc.33178 sha: 4d76c430d40c2de40cb00bb14051a82aa63e3958 doc_id: 967983 cord_uid: udbvhkus The performance of cervical cancer screening will decline as a function of lower disease prevalence—a consequence of successful human papillomavirus (HPV) vaccination. Replacement of cytology with molecular HPV testing as the primary screening test and adoption of risk‐based screening, with less intense screening of vaccinated individuals and initiated at older ages is expected to improve efficiency. However, policy officials may decide to further reduce or eliminate screening as the ratio of benefits to harms continues to decline. To evaluate the level of risk currently tolerated for different cancers in the United States (ie, for which clinical guidelines do not recommend secondary prevention though effective screening methods exist), we used US cancer registry data to compare incidence (2008‐2012) and survival (1988‐2011) associated with different cancers for which organized screening is recommended and not recommended. The most common cancer at ages 70 to 74 years (ie, age group with highest cancer incidence and reasonable life expectancy to consider screening in the US) satisfying Wilson and Jungner's classic screening criteria was vulvar cancer (incidence = 9/100 000 females). In comparison, the incidence of cervical cancer among females 65 years of age (the upper recommended age limit for screening) was 13 cases per 100 000 females (low as a reflection of effective screening), whereas 10‐year survival was 66% (similar to vulvar cancer at 67%). Our approach of defining tolerable risk in cancer screening could help guide future decisions to modify cervical screening programs. Considers psychological harms and potential morbidity and risks during the entire screening process, including diagnostic work-up and treatment. g Cancers for which organized, guideline-driven screening programs currently exist were considered highly cost-effective (ie, ++). Cancers for which screening guidelines exist for high-risk subgroups only, or in opportunistic settings were considered moderately cost-effective (ie, +). Uncertainty regarding the cost-effectiveness of screening was considered equivocally cost-effective (ie, +/−). worldwide. 1 Prior to marketing these vaccines, investigators had already begun to consider the impact that vaccination would have on cervical cancer screening. 2 Modeling studies revealed that as lesion prevalence declined, the positive predictive value of cytology (ie, probability that patients with a positive screening test truly have cervical precancer/ cancer) would become too low to maintain it as the primary screening test. 2 As the first vaccinated cohorts now become eligible for screening, there is urgency to introduce an alternative approach. Most experts agree that cytology should be replaced with more sensitive HPV DNA testing, perhaps reserving cytology (a test with excellent specificity) for triaging HPV-positive women for referral to colposcopy. [3] [4] [5] The recent arrival of Gardasil 9 (MSD), which targets five other oncogenic HPV types, responsible for an additional 20% of cervical cancer cases globally, will have an even greater impact on disease prevalence and screening performance. 1 For most other cancers that we consider, the tissue at risk is easily accessible and amenable to early detection but systematic screening is not recommended (eg, skin, thyroid, oral, vulvar, vaginal and anal cancers). We also include prostate and ovarian cancers for comparison; both of which are common but have different survival outcomes. The data that we used for this analysis are available from the SEER For cervical and colorectal cancers, detection and treatment of precancerous lesions or polyps are possible before they become invasive. The incidence of these cancers is therefore expected to be higher in the absence of screening. On the other hand, early detection is the goal of screening for breast and lung cancers and the incidence of these cancers may be lower in the absence of screening due to overdiagnosis. [16] [17] [18] In the United States, it is recommended that females aged 21-65 years regularly undergo screening for cervical cancer at least once every 3 years For colorectal cancer, screening is strongly recommended for individuals aged 50 to 75 years, 12,13 and average annual incidence at these ages ranged from 51 to 162 cases (females) and 64 to 233 cases (males) per 100 000 individuals, respectively. Due to the recent rising incidence of colorectal cancer in younger individuals in the United States, 19 the ACS issued a qualified recommendation in 2018 to initiate screening at Age 45. 13 The recommendation was issued as "qualified" rather than "strong" because evidence regarding the balance of benefits and harms remains lacking. 13 Focusing on breast cancer, guidelines are consistent in rec- We also present incidence and survival for selected cancers for which organized screening is not recommended (Figures 1 and 2 , respectively). In our assessment of these cancers ( for breast and 6 (5) for anal. The 10-year survival for these cancers was 97%, 87%, 92%, 45%, 72% and 52%, respectively. Among cancers for which organized screening is currently recommended, cervix is by far the least common ( Figure 1 ). Overall survival is also high compared to the other cancers that we examined. Among females, vulvar, breast, thyroid and melanoma cancers have better prognoses (Figure 2 ). In our initial attempt to identify cancers that could serve as benchmarks for tolerable risk, we focused on those with high incidence (≥10 cases per 100 000). According to this criterion, we identified vulvar, thyroid, melanoma, prostate and ovarian cancers. In addition to incidence, the definition of benchmarks of tolerable risk must also consider clinical outcomes, which implies focusing on cancers with poor overall prognoses (10-year survival <80%). Excluding ovarian cancer, the cancers with the highest incidence at ages 70 to 74 years for which we currently do not screen were identified to be vulvar (9 cases per 100 000 females) and oral cancers (6 and 8 cases per 100 000 individuals in females and males, respectively). While 10-year survival for cervical cancer (66%) was substantially higher compared to survival for oral cancer (52% among females and 45% among males), survival for cervical cancer was similar compared to survival for vulvar cancer (67%). In this analysis comparing cervical cancer burden with other cancers for which organized screening programs exist (breast, colorectal and lung) and do not exist (ovarian, prostate, thyroid, skin, oral, vulvar, vaginal and anal), cervical cancer ranked poorly in the former group (ie, lowest incidence and second best survival) and only moderately in the latter group (ie, among the lowest incidence and generally similar survival)-a reflection of effective screening that has prevailed in the US for several decades. Among cancers that we do not currently screen for but satisfy Wilson and Jungner's criteria (Table 1) , vulvar cancer has the highest incidence (9 cases per 100 000 at ages 70-74 years) and is arguably the best candidate for defining tolerable risk. Vulvar squamous cell carcinoma has a comparable precancerous lesion stage to cervical cancer that is amenable to detection via exfoliative cytological screening or molecular HPV testing of exfoliated cells and diagnostic biopsies can be obtained during a pelvic examination. During an exam, cytology may be performed to identify morphologic As a result, performance could be even better compared to cervical screening using cytology. 20 While a higher proportion of vulvar lesions may be attributed to HPV among younger patients, 21 In addition to providing excellent protection against cervical cancer, we also expect a reduction in the incidence of vulvar cancer due to vaccination. Approximately 25% of vulvar cancer cases globally are attributable to HPV 24 and among these cases, approximately 90% are caused by the seven oncogenic HPV types targeted by Gardasil 9. 25 In settings with successful cervical cancer screening programs, approximately 60% to 80% of cervical cancer cases may be prevented. 26, 27 If screening for vulvar cancer were to be recommended and successfully implemented with identification and treatment of vulvar intraepithelial neoplasia lesions, then we may expect a decline in the incidence of vulvar cancer that is similar. While we should acknowledge that reductions in vulvar cancer incidence may be achieved through vaccination and potentially screening, this does not impact our assessment and conclusions regarding current tolerable risk thresholds. Within a few decades from now, owing to the success of vaccination enabling protection of most birth cohorts, the annual incidence of cervical cancer will likely drop below current incidence levels for vulvar cancer, across all age groups. Arguably, screening could then be eliminated or conducted very infrequently to reduce costs and associated harms. Preterm delivery and low birth weight are important obstetric outcomes associated with excisional treatment of cervical precancerous lesions. 28, 29 Other potential harms that may result from biopsy and treatment include anxiety, pain, bleeding and discharge. 30 A recent study evaluating cervical screening benefits (incidence/mortality) and related harms suggests a substantial reduction in screeningrelated harms may be achieved in the United States by adopting a less intense screening approach that is similar to recommendations in the Netherlands (ie, cytology screening every 5 years), without any loss in benefit. 30 Investigators reported that the number of preterm deliver- 34 These results suggest that we should consider raising the age of screening initiation among vaccinated females. In 2008, the concept of developing a risk-based strategy to manage patients in cervical cancer screening was developed by Castle et al, 35 and refined by Katki et al. 36 The approach is appropriately referred to as "equal management of equal risks" and advocates applying similar management based on different combinations of test results conferring the same cancer risk, a concept that is now incorporated in new professional guidelines. 37 Although this concept of "benchmarking" risk was intended to be specific to cervical cancer, it relates to our concept of establishing tolerable risk thresholds in cancer screening. However, one of the issues in trying to identify a tolerable risk threshold is that it does not consider costs-a key criterion proposed by Wilson and Jungner. 7 In the same way that we consider an acceptable benchmark for risk of cervical cancer, we may also determine the benchmark for deciding if an intervention is cost-effective. For example, is it $50 000, $100 000 or $150 000 US dollars (USD) per qualityadjusted life-year (QALY) gained? Outside of the context of an explicit resource constraint, this is a difficult question to address; however, Neumann and colleagues suggest using either $100 000 or $150 000 USD. 38 If we are able to compare costs per QALY for different interventions, then substituting less cost-effective for more cost-effective ones would become more objective and likely more common, especially as resources become increasingly constrained. 39 In the context of cervical cancer prevention, there are two important questions: (a) "How can we achieve higher vaccination coverage?" and (b) "Once higher coverage is attained, should cervical cancer screening be discontinued?" The first question is outside the scope of this analysis, but our rationale is related to the second question. If policy decisions should eventually discontinue or reduce cervical cancer screening based on an unfavorable cost-benefit ratio in a future that includes high vaccination coverage, then it stands to reason that today's policymaking should allocate more resources towards reaching this goal. But the most likely scenario is that screening intensity will vary across settings as a reflection of different nations' budgets for health care expenditures and societal variations in risk tolerance. The process of creating screening recommendations has historically focused on evaluating evidence related to each cancer site individually. To the best of our knowledge, guideline committees do not compare risk across cancer sites. In 2015, Whitham and Kulasingam evaluated risk of cervical cancer at and after the recommended age to begin and end screening in relation to risk of breast and colorectal cancers, revealing the higher propensity to screen for cervical cancer despite lower risk. 40 Our analysis comparing incidence and survival across cancer sites for which screening is recommended and not recommended, to assess tolerable risk, is an extension of this approach. At the time this article was under consideration, the preventive measures taken in response to the COVID-19 epidemic had begun to adversely affect the entire range of activities related to cancer control, prevention and care. Justifiably, controlling the epidemic is a much greater priority, which must receive the necessary personnel and material resources to ensure a successful operation. Public health activities related to cancer screening and prevention, typically carried out in outpatient clinics, were considered of lower priority and were thus scaled back to conserve resources, as well as to prevent exposure of cancer screening participants to SARS-CoV-2. In addition, adoption of policy decisions that could improve the uptake and quality of cancer screening services may be delayed because of the epidemic. Likewise, coverage of HPV vaccination is likely to suffer in consequence of behavioral changes related to health promotion. Although a disruption in cancer control activities is likely to lead to an increase in incidence of cancers preventable via screening and vaccination it does not invalidate the arguments we presented in this article. Human papillomavirus type distribution in 30,848 invasive cervical cancers worldwide: variation by geographical region, histological type and year of publication Chapter 20: issues in planning cervical cancer screening in the era of HPV vaccination Introduction of molecular HPV testing as the primary technology in cervical cancer screening: acting on evidence to change the current paradigm Approaches for triaging women who test positive for human papillomavirus in cervical cancer screening Effect of screening with primary cervical HPV testing vs cytology testing on high-grade cervical intraepithelial Neoplasia at 48 months: the HPV FOCAL randomized clinical trial Final efficacy, immunogenicity, and safety analyses of a nine-valent human papillomavirus vaccine in women aged 16-26 years: a randomised, double-blind trial Principles and practice of mass screening for disease American Society for Colposcopy and Cervical Pathology, and American Society for Clinical Pathology screening guidelines for the prevention and early detection of cervical cancer Screening for cervical cancer: US preventive services task force recommendation statement American Cancer Society guidelines for breast screening with MRI as an adjunct to mammography Preventive Services Task Force. Screening for breast cancer: U.S. preventive services task force recommendation statement Screening for colorectal cancer: US preventive services task force recommendation statement Colorectal cancer screening for average-risk adults American Cancer Society lung cancer screening guidelines Screening for lung cancer: U.S. preventive services task force recommendation statement Overdiagnosis in cancer Effect of three decades of screening mammography on breast-cancer incidence Overdiagnosis in low-dose computed tomography screening for lung cancer Colorectal cancer incidence patterns in the United States, 1974-2013 Cytology of the vulva: feasibility and preliminary results of a new brush Prognostic significance of human papillomavirus DNA in vulvar carcinoma Clinical progression of high-grade cervical intraepithelial neoplasia: estimating the time to preclinical cervical cancer from doubly censored national registry data Natural history of cervical neoplasia and risk of invasive cancer in women with cervical intraepithelial neoplasia 3: a retrospective cohort study Global burden of cancer attributable to infections in 2018: a worldwide incidence analysis Burden of human papillomavirus (HPV)-related 58 Cervical cancer: epidemiology, prevention and the role of human papillomavirus infection Effectiveness of cervical screening with age: population based case-control study of prospectively recorded data Obstetric outcomes after conservative treatment for intraepithelial or early invasive cervical lesions: systematic review and meta-analysis Perinatal mortality and other severe adverse pregnancy outcomes associated with treatment of cervical intraepithelial neoplasia: meta-analysis Harms of cervical cancer screening in the United States and The Netherlands Optimal cervical cancer screening in women vaccinated against human papillomavirus What cervical screening is appropriate for women who have been vaccinated against high risk HPV? A simulation study Cervical cancer screening of HPV vaccinated populations: cytology, molecular testing, both or none Impact of human papillomavirus vaccination on the clinical meaning of cervical screening results Risk assessment to guide the prevention of cervical cancer Benchmarking CIN 3+ risk as the basis for incorporating HPV and pap cotesting into cervical screening and management guidelines An introduction to the 2019 ASCCP risk-based management consensus guidelines Updating cost-effectiveness-the curious resilience of the $50,000-per-QALY threshold Comparative effectiveness and health care spending-implications for reform The significantly lower risk of cervical cancer at and after the recommended age to begin and end screening compared to breast and colorectal cancer Defining benchmarks for tolerable risk thresholds in cancer screening: Impact of HPV vaccination on the future of cervical cancer screening