1B
ORAL ABSTRACTS: EVALUATING SCREENING
1B
Method(s): We constructed a Markov state-transition model based on previous economic modelling carried out in the context of the UK NHS Bowel Cancer Screening Programme (NHSBCSP). We used results from a 2014-2015 pilot study of the introduction of FIT screening in two of the five screening hubs in England to update the model with test performance data for FIT vs. gFOBT at a range of cut-off values for positive FIT. Other model parameters were taken from NHSBCSP data, national cost datasets and published sources. Results were extrapolated to a lifetime time horizon and probabilistic sensitivity analyses were carried out to assess the effect of parameter uncertainty on the study conclusions.
Result(s): The results from the economic model suggest that FIT is associated with lower costs and better quality of life outcomes than gFOBT at all FIT cut-off values considered in the analysis. The findings were driven by lower rates of cancer incidence in the long term with FIT screening, as predicted by the modelled extrapolation. The model predicted an increase in the total incremental colonoscopy costs by a factor of 25 between the highest (180µg/g) and the lowest (20µg/g) FIT cut-off value in the model. However, a five-fold increase in cancer management savings over the same range resulted in greater overall cost savings at lower cut-off values. Total incremental quality-adjusted life years also increased as the FIT cut-off value decreased.
Conclusion(s): Although more favourable quality of life and cost outcomes compared to gFOBT are predicted at lower FIT cut-off values, the large number of referrals at these levels estimated by the model may present a significant challenge to colonoscopy services. Therefore, the choice of FIT cut-off value upon introduction of the test must take into account short-term capacity constraints in the healthcare system.
In Austria, current opportunistic cervical cancer screening program includes annual Pap cytology for women as of age 18 years. However, new primary screening tests with increased test-sensitivity in combination with risk-based follow-up algorithms may improve the trade-off between benefits and overtreatment. Our aim was to systematically evaluate the benefit-harm balance of different cervical cancer primary screening strategies for the Austrian context.
Method(s):
We used a validated Markov-state-transition model calibrated to the Austrian epidemiological setting and clinical context of the disease to evaluate different screening strategies that differ by primary screening test (cytology, p16/Ki-67-dual stain, and HPV-testing alone or in combinations), screening interval, age, and specific follow-up algorithms for women with positive test results. Austrian clinical and epidemiological data, as well as test accuracy data from international meta-analyses and trials were used. Predicted outcomes were reduction in cervical cancer incidence, -mortality, overtreatment (defined as conization with histological diagnosis of no lesion or a lesion grade CIN1), and the incremental harm-benefit ratios (IHBR) measured in numbers of overtreatment per additional prevented cervical cancer death. Comprehensive sensitivity analyses were performed.
Result(s):
Based on our results, HPV-based primary screening strategies are more effective compared with cytology or with p16/Ki-67-testing alone. Adopting risk-based follow-up algorithms including p16/Ki-67 triage for women with ASCUS/LSIL and colposcopy referral for women with HSIL or p16/Ki-67-positivity can reduce overtreatment. In the base-case analysis (31-43% screening adherence in women below 60 years of age), biennial HPV+cytology cotesting seemed to be the optimal screening strategy (IHBR: 45 unnecessary conizations per additional prevented cancer death). Annual screening strategies resulted in much higher IHBRs (131-355 unnecessary conizations per additional prevented cancer death). Based on the IHBRs, the screening interval may be extended to 3 years in populations with screening adherence of 40%-60% and to 5 years in populations with higher adherence rate. The age for screening initiation could be extended from 18 to 24 years without significant loss in effectiveness, but with reduced overtreatment.
Conclusion(s):
Based on our benefit-harm analysis, HPV-based screening in women of the age 30 years or older and cytology in younger women at screening intervals of at least 2 years incorporating a risk-based follow-up algorithm can be recommended for the Austrian screening setting. Screening should be initiated in women of the age 20-24 years.
Repeated participation is important in faecal immunochemical testing (FIT) screening for colorectal cancer (CRC). However, a large number of screening invitations over time may lead to screening fatigue and consequently, decreased participation rates. We evaluated the impact of screening fatigue on overall screening programme effectiveness.
Method(s):
Using the ASCCA model, we simulated the Dutch CRC screening programme consisting of biennial FIT screening in individuals aged 55-75. We analysed several screening scenarios differing in participation pattern, number of negative screens after which screening fatigue occurs and decrease in participation rate due to screening fatigue. Outcomes were reductions in CRC incidence and mortality compared to no screening.
Result(s):
Assuming 63% participation which was homogenous across the population, i.e. each round each individual was equally likely to attend screening, thirty years of screening reduced CRC incidence and mortality by 39% and 53%, respectively, compared to no screening. When assuming clustered participation, i.e. three subgroups of individuals with a high, intermediate and low participation rate, screening was slightly less effective; reductions were 33% for CRC incidence and 43% for CRC mortality. Screening fatigue considerably reduced screening effectiveness; if individuals would refrain from screening after three negative screens, model-predicted programme effectiveness was reduced by 37% and 46% under homogenous and clustered participation, respectively.
Conclusion(s):
Screening will substantially decrease CRC incidence and mortality. However, screening effectiveness can be seriously compromised if screening fatigue occurs. This warrants careful monitoring of individual screening behaviour and consideration of targeted invitation systems in individuals who have been repeatedly screened.
The Fecal Immunochemical Test (FIT) is a test commonly used to screen for Colorectal Cancer (CRC) and adenomas. Liquid based FIT’s permit user specification of the threshold (T) for test positivity with effects on the number testing positive, the likelihood of detecting a neoplasm and the costs of screening. We used the Canadian Risk Management Model CRC sub-model (CRMM-CRC) to predict the effectiveness and cost-effectiveness of CRC screening based upon different choices for T.
Method(s):
The CRMM-CRC is a micro-simulation model for Colorectal Cancer in Canada. The model includes a natural history spine where CRC’s are assumed to develop from adenomas which are initiated, grow and regress according to age and sex-specific rates. The probabilities of treatment utilization, costs of testing and disease treatment were taken from Canadian data sources. Published literature on FIT testing by threshold in asymptomatic individuals were reviewed to derive the probability of a positive test (PPT - sensitivity and specificity) for subjects by disease status: CRC, size of adenoma (0-5mm, 6-9mm, 10+mm) and no neoplasm. Models were fit to provide a range of estimates expressing uncertainty in actual PPT values for the different disease states as a function of T. Effectiveness and cost-effectiveness predictions were obtained by simulating the lifetime of cohort age 45 in 2014. Screening was assumed to occur biennially between ages 50 and 74 with evaluation of abnormal tests by colonoscopy using different threshold values. Costs and QOL adjusted life-years were both discounted at 3%.
Result(s):
Uncertainty in threshold specific PPT’s were modeled for T between 50 and 225ng/ml in two dimensions: slope – rate of change in positivity rates by T and average – positivity rates at T=100 ng/ml. There were 7 resulting parameter sets evaluated representing joint sensitivity analyses by varying slope and average. All individual scenario combinations (parameter & T) resulted in predicted reductions in CRC incidence (range=22%-53%) and mortality (35%-65%). Costs per QOL varied between CAD$-530 and $5,558 across the individual scenarios. Within each parameter set the lowest threshold evaluated had the highest effectiveness but was not necessarily the most cost-effective.
Conclusion(s):
In the Canadian context screening with FIT for CRC is predicted to be a cost-effective intervention. Low thresholds for abnormality would appear preferable where colonoscopy resources allow.
In Austria opportunistic cervical cancer screening with annual cytology starting at age 18 years is the current standard. However, new screening tests together with risk-based screening and follow-up algorithms may have the potential to improve both, the effectiveness and the efficiency of the cervical cancer screening program. Our aim was to systematically evaluate the long-term effectiveness and cost-effectiveness of different cervical cancer primary screening strategies for the Austrian health care context.
Method(s):
A Markov-state-transition model was developed for the Austrian health care context simulating the disease including HPV-infection and different pre-invasive as well as invasive cervical cancer stages. The model was applied to evaluate different screening strategies that differ by primary screening test (including cytology, p16/Ki-67-dual stain, and HPV-testing alone or in combinations), screening interval, age, and specific follow-up algorithms for positive test results. We used Austrian clinical, epidemiological and economic data, and test accuracy data from international meta-analyses and trials. All costs and effects were discounted at 5% annually. Index year was 2014. Predicted outcomes were reduction in cervical cancer cases and deaths, remaining life expectancy (in life years [LY]), total lifetime costs (in Euro), and the incremental cost-effectiveness ratios (ICER; in Euro/LY gained). Comprehensive sensitivity analyses were performed.
Result(s):
Within the same screening interval, HPV-based primary screening strategies are more effective (relative reduction cancer death: 56%-79% for 5-2 yearly screening intervals) compared with cytology (42%-69%) or with p16/Ki-67 testing alone (50%-76%). Adopting risk-based follow-up algorithms including p16/Ki-67 triage for women with ASCUS or LSIL and colposcopy referral for women with HSIL or p16/Ki-67-positivity can improve efficiency. In the base-case analysis (31-43% screening adherence in women below 60 years of age), optimal balance between benefits and costs achieved biennial HPV-testing (with cytology triage of HPV-positive women) at the age of 30 years and biennial cytology (with p16/Ki-67-triage of women with ASCUS/LSIL) at younger age with an ICER of 43,700 Euro/LY gained. In sensitivity analyses results were sensitive with regard to HPV-test cost, cytology test accuracy, screening adherence rate and annual discount rate.
Conclusion(s):
Based on our results, biennial primary HPV screening with cytology triage in women age 30 years and older and biennial cytology with p16/Ki-67-triage in younger women can be considered as a cost-effective screening option for the Austrian context.
