- Email: [email protected]
Screening for Colorectal Cancer by Immunological Fecal Occult Blood Test: Number Needed to Screen and to Scope to Find One Advanced Neoplasm
result in a mean gain of 35,000 life years and 26,000 QALYs in the Dutch population per year, which translates to an amount of 1,3-5,9 billion dollar/yr for the society in the Netherlands.
AGA Abstracts
Su1103 Shared Decision-Making (SDM) for Colorectal Cancer (CRC)Screening Increases Adherence Paul C. Schroy, Karen Emmons, Ellen Peters, Michael Pignone, Julie Glick, Patricia Robinson, Maria Lydotes, Shamini Mylvaganam, Alison Coe, Stephen Evans, Marianne N. Prout, Peter Davidson, Timothy Heeren
ND: Not detected
Eliciting patient preferences within the context of SDM is a potentially effective yet unproven strategy for increasing patient adherence to CRC screening recommendations. We have previously shown that the use of a novel, interactive, computer-based decision aid facilitates effective SDM by enabling patients to identify a value concordant screening preference, increasing satisfaction with the decision-making process (SDMP) and increasing screening intentions (Med Decis Making, 2010). The goal of this study was to assess whether effective SDM increases adherence to CRC screening. Methods: Asymptomatic, average-risk patients (aged, 50-75y) were randomized to one of two intervention arms (decision aid plus personalized risk assessment [DA+PRA]) or decision aid alone [DA]) or a control arm. The interventions took place just prior to a routine office visit with their primary care providers. Test completion rates (adherence) were compared across the 3 groups using chi-square analysis. Logistic regression was used to identify predictors of adherence. Results: A total of 825 patients (84% <65y; 59% female; 62% Black, 5% Hispanics) were randomized. Patients in the DA arm were significantly more likely than controls to complete a screening test within 12 months of follow-up (43% vs. 34%; P=0.046); in contrast, test completion rates were similar for the DA+PRA group and controls (37% vs. 34%; P=0.50). Although patients who chose colonoscopy were more likely to have their preferred test ordered (77%) than those who chose tests other than colonoscopy (fecal occult blood testing [FOBT], 34%; flexible sigmoidoscopy [FS], 15%; FOBT+FS, 6%; barium enema, 20%), univariate logistic regression found no association between adherence and either test preference for those who expressed a preference (P=0.16) or concordance (i.e., agreement between patient preference and test ordered; P=0.82). Multiple logistic regression identified assignment to the DA alone intervention arm (adjusted odds ratio [aOR], 1.5; 95% confidence interval [CI], 1.1-2.1), black race (aOR, 1.6; 95%CI, 1.2-2.1) and a preference for patient-dominant decision-making approach (aOR, 1.6; 95% CI, 1.0-2.4) as independent determinants of adherence. Controlling for screening intentions in the multiple logistic regression diminished the intervention effect; conversely, controlling for SDMP had no effect. Conclusions: The use of a decision aid to facilitate SDM increased adherence to CRC screening, regardless of whether providers complied with patient preferences. Effectiveness was mediated in part by a positive impact on screening intentions. Addition of personalized risk feedback did not increase adherence over controls. An overall adherence rate of only 43% in the DA alone group, however, highlights the need for additional strategies to optimize uptake of CRC screening.
Su1101 MicroRNA Expression Alterations in Duodenal Mucosa of Patients With IPMNs: Potential Diagnostic/Prognostic Biomarker Dhananjay Kunte, Shailesh Bajaj, Ramesh K. Wali, Mark Talamonti, Michael J. Goldberg, Kevin K. Roggin, Andrew Radosevich, Vadim Backman, Hemant K. Roy Background: Diagnostic and prognostication of IPMNs represents a vexing and increasingly common clinical problem. Current diagnostic and prognostic approaches remain suboptimal. Our group has focused on field carcinogenesis detection for risk stratification (reviewed in Gastro 2011). We have noted that in the peri-ampullary duodenal biopsies, there are microarchitectural alterations as detectable by our novel optical technologies, low coherence enhanced backscattering spectroscopy and partial wave spectroscopy (Clin Cancer res 2007, Dis Biomarkers 2008, Cancer Res 2009). However, the biological basis of duodenal alterations has been LEBS (micro-architectural) alterations have been unexplored. MicroRNAs have been implicated in pancreatic carcinogenesis at both the initiation and progression phase. We wanted to compare the miRNA expression in duodenal mucosa in IPMN patients and control patients and correlate miRNA profile with the disease. Methods:We obtained endoscopic biopsies from patients with IPMNs undergoing endoscopic ultrasound for evaluation. Controls were age and gender matched. The RNA was recovered from formalin fixed paraffin embedded sections) using Ribopure kit (Ambion). The total RNA was then reverse transcribed using Taqman miRNA kit using Megaplex RT Primer set A (Applied Biosystems). The cDNA was then diluted and loaded into Taqman Low Density Array (TLDA) for miRNA (ABI). Real time RT-PCR was performed on ABI Prism 7900HT system. The data analysis was done using the SDS RQ manager and Data Assist software. Results: The comprehensive miRNA profiling of duodenal mucosal RNA from IPMN and control patients showed downregulation of 15 miRNAs and upregulation of miR-212, miR-130b and miR-100. Importantly, the miRNA reported to be upregulated in pancreatic cancer (miR-212 and MiR-100) were also seen to be upregulated in duodenal mucosa. Similarly, miR-24, which is the key regulator of p16, was also significantly modulated in the duodenal mucosa, thus underscoring the field effect for miRNA expression in pancreatic cancer. Conclusions: We report, herein, for the first time, that microRNA in the duodenal mucosa are dysregulated in patients harboring IPMNs when compared to matched controls. Importantly, these demonstrate important molecular alterations in extended field carcinogenesis of pancreas cancer. Future studies will be to develop prediction rules of duodenal microRNA with or without LEBS (microarchitectural biomarkers). The future goals are both for diagnostics and more significantly from a clinical perspective for prognosis/natural history.
Su1104 Screening for Colorectal Cancer by Immunological Fecal Occult Blood Test: Number Needed to Screen and to Scope to Find One Advanced Neoplasm Sergio Crotta, Carlo Senore, Nereo Segnan, Simona Paganin, Bruna Dagnes Background: In a screening program for colorectal cancer one method of expressing the costs and benefits is the number of people who need to be screened (NNS) and the number who need to be scoped (NNSC) to detect an advanced neoplasm (AN-advanced adenoma or cancer). We calculated the NNS and NNSC over 4 biennial rounds of immunological fecal occult blood test (FIT) screening in the same cohort of people. Population and Methods: Starting from October 2001, the 2959 subjects aged 50 to 74, resident in two towns of Aosta Valley (Italy), were invited every 2 years by letter to perform a 1-day latex FIT (OC Sensor), with a 100 ng/ml cut-off. Patients with positive test were referred for colonoscopy. We calculated NNS and NNCS for each round and over all four biennial rounds. Results: Of the 2959 subjects in the initial cohort, 2161 (73.3%) performed at least 1 test, 1559 (52.7%) 2 consecutive tests, 971 (32.8%) 3 consecutive tests and 713 (24 %) 4 tests. The yields of screening are reported in table 1. The NNS necessary to detect an AN in the first round was 61.7, higher for women than for men (89.5 vs 45.4); it was lower than when using guaiac screening1 (NNS:107) and comparable to the results of a recent study2 (NNS: 41) with FIT at the same cut-off level as our. The NNS to diagnose an AN increased in the subsequent rounds, particularly for women, but it did not change much in relation to the number of negative previous examinations. The number of cases to be examined with colonoscopy to diagnose AN was higher for women than men (3.8 vs 2.3) and remained essentially unchanged even after several negative tests, reflecting the stability of the positive predictive value. Considering the total number of exams in people having multiple consecutive tests, the NNS necessary to find an AN increased from 61.7 to 173.2 in the second round. For people who were also negative in the second test the NNS was stable between 260 and 290 in the two subsequent rounds, always with sharp difference between men and women. Conclusions: The number needed to screen to detect an advanced neoplasm was nearly three times higher in the second (173.2) and over four times in the third (291.3) as compared to the first round (61.7), but the trend was stable thereafter. The number needed to scope to find an advanced neoplasia was 2.5 initially and around 3 in the other 3 rounds, reflecting the stability of the positive predictive value. The NNT and the NNSC were always higher in women. 1.Denis B et al. Gut 2007;56:1579-1584. 2.Hol l et al. Br J Cancer 2009; 100(7): 1103 - 1110 Table 1: NNS and NNSC to find one advanced neoplasm
Su1102 The Valuation of the Increase in Quality of Life and Health-Adjusted Life Expectancy as a Result of Colorectal Cancer Screening in Future Decades; A Population-Based Study Caroline M. den Hoed, Kees Isendoorn, Wouter Klinkhamer, Anshu M. Gupta, Ernst J. Kuipers Background Colorectal cancer (CRC) affects a considerable proportion of the population above 50 years. Population screening has a major impact on morbidity and mortality due to CRC. The cost-efficacy of screening has been widely studied, but these studies mostly did not take savings of reduced CRC treatment into account. Most importantly, the societal gains of screening have not been calculated. Such knowledge is relevant to understand the benefits of screening and to decide on implementation of screening programs. Aim The aim of this study therefore was to calculate the societal gains of implementation of nationwide colon cancer screening in terms of QALY's (quality of life adjusted life years)and HALE (health-adjusted life expectancy) as well as to provide a Dollar valuation of these societal gains. Methods Literature studies were performed for data on colon cancer epidemiology, and screening outcomes over time. National / governmental databases in the Netherlands were searched to obtain the input for calculations of QALYs, HALE, and the corresponding financial gain. Results The mortality due to colon cancer is high, 26/1000 persons of 50 year and older in 2008, with a mean loss of 12.0 life years per patient diagnosed corresponding to 52,000 life years in the total Dutch population per year. Quality of life (QoL) of patients that survive colon cancer was less affected, 1 year after diagnosis the mean QoL of patients was 63/100 compared to 64/100 in the general population. Implementing FIT-screening (Fecal immunochemical test) will lead to diagnosis at an earlier stage, with 60% of cancers diagnosed in stage 1, leading to a mortality reduction of at least 50%-80%, to a mortality of 5-13/1000 patients 50 year and older. This will lead to a rise in the mean HALE with 0.025 year per person in the population. On a population of 16.8 million people this translates to a mean reduction of otherwise lost quality-adjusted life years per year of 5,000 (16.8 million*0.025/average life expectancy at birth). Until 2040 the Dutch population will lose a mean of 70,000 life years by colon cancer per year because of high incidence due to an ageing population. Implementing FIT-screening will gain 35,000 life years every year in this period, which corresponds to 26,000 QALYs per year. The value of a life year varies from $76,000 to $230,000 per person/yr and the true financial gain for society can be calculated in several ways. The implementation of FIT-screening will thus result in a gain for the Dutch society of 1,3-5,9 billion dollar/yr depending on the method used. Conclusion This study demonstrated the implementation of FIT-screening in the coming 3 decades will
AGA Abstracts
S-404
AGA Abstracts * Advanced Neoplasm Su1105 Attendance and Diagnostic Yield of Repeated Fecal Immunochemical Test Screening With Intervals of 1, 2, or 3 Years: A Comparative Population-Based Colorectal Cancer Screening Trial Aafke H. van Roon, Simon L. Goede, Marjolein van Ballegooijen, Jacqueline C. Reijerink, Hans 't Mannetje, Anneke van Vuuren, Alexandra C. van der Togt-van Leeuwen, J. Habbema, Ernst J. Kuipers, Monique van Leerdam Introduction: Fecal immunochemical tests (FIT) have suboptimal sensitivity for advanced neoplasia and require successive screening rounds for an optimal preventive effect. However, data about the influence of screening interval length on attendance and diagnostic yield are lacking, nor is it known whether the increased sensitivity of FIT compared to the conventional guaiac-based FOBT allows for longer screening intervals. We therefore performed repeated FIT screening in a population-based trial comparing 1, 2, and 3 year intervals. Methods: Three representative samples of the Dutch population (total=7,500) aged 50-75 years were randomly selected prior to invitation. Individuals were invited for two 1-sample FIT screening rounds (cut-off ≥ 50 ng Hb/mL) with intervals of respectively 1 (group A), 2 (B), or 3 years (C). Subjects with a positive FIT in the 1st screening round, and those who had moved away/died were not invited for the 2nd round. Results: In group A, attendance was 64.6% (1,544/2,391) in the 1st and 62.7% (n=1,302/2,077) in the 2nd screening round. Of the 1st round participants, 89.8% also attended the 2nd screening round. Of the non-participants in the 1st screening round, 16.3% did participate in the 2nd round. We refer to Table 1 for group B and C. Corrected for 1st round attendance, a longer screening interval improved participation in the 2nd FIT-based screening round (group A (reference) 89.8% vs. group B 90.9%, and vs. group C 91.4%; p<0.05). Attendance of 1st round non-participants was also higher in group B and C compared to the reference group A (respectively 19.3%, and 18.9%, vs 16.3%; p<0.05). The proportion of participants attending at least one screening round did not significantly differ between the screening intervals (group A: 67.4%, group B: 66.1%, group C 67.0%; p=0.60). The overall positivity rate (PR) in the 2nd screening round was significantly lower compared to the 1st (6.0% vs. 8.4%, OR 0.69; CI 0.58-0.82). However, a longer interval was not associated with higher PR at repeated screening (p= 0.36). Similarly, the overall detection rate (DR) of advanced neoplasia was significantly lower in the 2nd round (1.8%, OR 0.55; CI 0.41-0.74) compared to the 1st screening round (3.3%). Colorectal cancers were found in 0.44% of 1st round participants vs. 0.17% in 2nd round participants (OR 0.39; CI 0.16-0.98). The length of screening interval did not affect the DR of advanced neoplasia in the 2nd round (p=0.33). Conclusion: Attendance of repeated FIT screening is moderately high with uptake percentages approximating 70%. A longer screening interval of 2 or 3 years results in significantly higher participation compared to annual screening. In the 2nd screening round, positivity and detection rates were significantly lower than in the 1st screening round. However, both were not affected by the length of the screening interval. Table 1
# Data of 792 invitees are not yet available; * P-value < 0.05; Advanced neoplasia = colorectal cancer and an adenoma ≥ 10 mm, or an adenoma with ≥ 25% villous component and/or high-grade dysplasia Su1106 Gender Disparities in Performance of a Fecal Immunochemical Test for Detection of Advanced Neoplasia Sietze T. Van Turenhout, Frank A. Oort, Jochim S. Terhaar sive Droste, Arjan P. Visscher, Veerle M. Coupe, René W. van der Hulst, Pieter Stokkers, Anneke A. Bouman, Gerrit A. Meijer, Leo G. van Rossum, Chris J. Mulder Introduction Fecal immunochemical tests (FITs) are commonly used in screening for colorectal cancer (CRC) and advanced adenomas. It has been suggested that FIT performance may be different in males and females, what could lead to over- or underestimation of overall test accuracy. However, these disparities might also be caused by gender related differences in age of onset and advanced neoplasia distribution. Thus far, little data exist on potential dissimilarities in test characteristics between the sexes. Aims & Methods Aim of the present study is to compare the confounding influence of gender on the sensitivity and specificity of a frequently used FIT, for the detection of advanced neoplasia. In this prospective cohort study, all individuals referred for elective colonoscopy were invited to perform a semiquantitative FIT (OC sensor®) prior to the start of bowel preparation for colonoscopy. Subjects in which colonoscopies were performed for evaluation of rectal bleeding or anemia were excluded to correct for potential work-up bias. In addition, participants with incomplete colonoscopy were excluded. Advanced neoplasia was defined as CRC and/or an advanced adenoma. Sensitivity and specificity for advanced neoplasia were compared at different cutoff values (50-200ng/ml), and multivariate logistic regression was used to study potential confounding on sensitivity and specificity of gender, age, number of advanced adenomas, and number of adenomas ≥1cm. Results In 1682 subjects (46% males), 213 participants were found to have at least one advanced neoplasm (54% present in males), i.e. CRC in 40 and advanced adenomas in 173 subjects. At 50ng/ml, sensitivity for advanced neoplasia was 47% (55% for men, 38% for women), and specificity was 92,0% (91,4% for men and 92,4% for women). At each cut-off level, females were found to have a significantly lower sensitivity (range of difference: 17,0-20,5%), but specificity was similar (difference -1,0 - 0,0%; see table 1). Multivariate logistic regression analysis showed that gender significantly influences the sensitivity for advanced neoplasia with or without correction for age, number of advanced adenomas, and number of adenomas ≥1cm. Conclusions Gender could significantly affect sensitivity of a fecal immunochemical test for the detection of advanced neoplasia. Adjusting for gender disparity when reporting test performance should be considered in designing a screening program. Table 1. Sensitivity and specificity of a fecal immunochemical test according to gender for the detection of advanced neoplasia.
S-405
AGA Abstracts
AGA Abstracts
Su1103 Shared Decision-Making (SDM) for Colorectal Cancer (CRC)Screening Increases Adherence Paul C. Schroy, Karen Emmons, Ellen Peters, Michael Pignone, Julie Glick, Patricia Robinson, Maria Lydotes, Shamini Mylvaganam, Alison Coe, Stephen Evans, Marianne N. Prout, Peter Davidson, Timothy Heeren
ND: Not detected
Eliciting patient preferences within the context of SDM is a potentially effective yet unproven strategy for increasing patient adherence to CRC screening recommendations. We have previously shown that the use of a novel, interactive, computer-based decision aid facilitates effective SDM by enabling patients to identify a value concordant screening preference, increasing satisfaction with the decision-making process (SDMP) and increasing screening intentions (Med Decis Making, 2010). The goal of this study was to assess whether effective SDM increases adherence to CRC screening. Methods: Asymptomatic, average-risk patients (aged, 50-75y) were randomized to one of two intervention arms (decision aid plus personalized risk assessment [DA+PRA]) or decision aid alone [DA]) or a control arm. The interventions took place just prior to a routine office visit with their primary care providers. Test completion rates (adherence) were compared across the 3 groups using chi-square analysis. Logistic regression was used to identify predictors of adherence. Results: A total of 825 patients (84% <65y; 59% female; 62% Black, 5% Hispanics) were randomized. Patients in the DA arm were significantly more likely than controls to complete a screening test within 12 months of follow-up (43% vs. 34%; P=0.046); in contrast, test completion rates were similar for the DA+PRA group and controls (37% vs. 34%; P=0.50). Although patients who chose colonoscopy were more likely to have their preferred test ordered (77%) than those who chose tests other than colonoscopy (fecal occult blood testing [FOBT], 34%; flexible sigmoidoscopy [FS], 15%; FOBT+FS, 6%; barium enema, 20%), univariate logistic regression found no association between adherence and either test preference for those who expressed a preference (P=0.16) or concordance (i.e., agreement between patient preference and test ordered; P=0.82). Multiple logistic regression identified assignment to the DA alone intervention arm (adjusted odds ratio [aOR], 1.5; 95% confidence interval [CI], 1.1-2.1), black race (aOR, 1.6; 95%CI, 1.2-2.1) and a preference for patient-dominant decision-making approach (aOR, 1.6; 95% CI, 1.0-2.4) as independent determinants of adherence. Controlling for screening intentions in the multiple logistic regression diminished the intervention effect; conversely, controlling for SDMP had no effect. Conclusions: The use of a decision aid to facilitate SDM increased adherence to CRC screening, regardless of whether providers complied with patient preferences. Effectiveness was mediated in part by a positive impact on screening intentions. Addition of personalized risk feedback did not increase adherence over controls. An overall adherence rate of only 43% in the DA alone group, however, highlights the need for additional strategies to optimize uptake of CRC screening.
Su1101 MicroRNA Expression Alterations in Duodenal Mucosa of Patients With IPMNs: Potential Diagnostic/Prognostic Biomarker Dhananjay Kunte, Shailesh Bajaj, Ramesh K. Wali, Mark Talamonti, Michael J. Goldberg, Kevin K. Roggin, Andrew Radosevich, Vadim Backman, Hemant K. Roy Background: Diagnostic and prognostication of IPMNs represents a vexing and increasingly common clinical problem. Current diagnostic and prognostic approaches remain suboptimal. Our group has focused on field carcinogenesis detection for risk stratification (reviewed in Gastro 2011). We have noted that in the peri-ampullary duodenal biopsies, there are microarchitectural alterations as detectable by our novel optical technologies, low coherence enhanced backscattering spectroscopy and partial wave spectroscopy (Clin Cancer res 2007, Dis Biomarkers 2008, Cancer Res 2009). However, the biological basis of duodenal alterations has been LEBS (micro-architectural) alterations have been unexplored. MicroRNAs have been implicated in pancreatic carcinogenesis at both the initiation and progression phase. We wanted to compare the miRNA expression in duodenal mucosa in IPMN patients and control patients and correlate miRNA profile with the disease. Methods:We obtained endoscopic biopsies from patients with IPMNs undergoing endoscopic ultrasound for evaluation. Controls were age and gender matched. The RNA was recovered from formalin fixed paraffin embedded sections) using Ribopure kit (Ambion). The total RNA was then reverse transcribed using Taqman miRNA kit using Megaplex RT Primer set A (Applied Biosystems). The cDNA was then diluted and loaded into Taqman Low Density Array (TLDA) for miRNA (ABI). Real time RT-PCR was performed on ABI Prism 7900HT system. The data analysis was done using the SDS RQ manager and Data Assist software. Results: The comprehensive miRNA profiling of duodenal mucosal RNA from IPMN and control patients showed downregulation of 15 miRNAs and upregulation of miR-212, miR-130b and miR-100. Importantly, the miRNA reported to be upregulated in pancreatic cancer (miR-212 and MiR-100) were also seen to be upregulated in duodenal mucosa. Similarly, miR-24, which is the key regulator of p16, was also significantly modulated in the duodenal mucosa, thus underscoring the field effect for miRNA expression in pancreatic cancer. Conclusions: We report, herein, for the first time, that microRNA in the duodenal mucosa are dysregulated in patients harboring IPMNs when compared to matched controls. Importantly, these demonstrate important molecular alterations in extended field carcinogenesis of pancreas cancer. Future studies will be to develop prediction rules of duodenal microRNA with or without LEBS (microarchitectural biomarkers). The future goals are both for diagnostics and more significantly from a clinical perspective for prognosis/natural history.
Su1104 Screening for Colorectal Cancer by Immunological Fecal Occult Blood Test: Number Needed to Screen and to Scope to Find One Advanced Neoplasm Sergio Crotta, Carlo Senore, Nereo Segnan, Simona Paganin, Bruna Dagnes Background: In a screening program for colorectal cancer one method of expressing the costs and benefits is the number of people who need to be screened (NNS) and the number who need to be scoped (NNSC) to detect an advanced neoplasm (AN-advanced adenoma or cancer). We calculated the NNS and NNSC over 4 biennial rounds of immunological fecal occult blood test (FIT) screening in the same cohort of people. Population and Methods: Starting from October 2001, the 2959 subjects aged 50 to 74, resident in two towns of Aosta Valley (Italy), were invited every 2 years by letter to perform a 1-day latex FIT (OC Sensor), with a 100 ng/ml cut-off. Patients with positive test were referred for colonoscopy. We calculated NNS and NNCS for each round and over all four biennial rounds. Results: Of the 2959 subjects in the initial cohort, 2161 (73.3%) performed at least 1 test, 1559 (52.7%) 2 consecutive tests, 971 (32.8%) 3 consecutive tests and 713 (24 %) 4 tests. The yields of screening are reported in table 1. The NNS necessary to detect an AN in the first round was 61.7, higher for women than for men (89.5 vs 45.4); it was lower than when using guaiac screening1 (NNS:107) and comparable to the results of a recent study2 (NNS: 41) with FIT at the same cut-off level as our. The NNS to diagnose an AN increased in the subsequent rounds, particularly for women, but it did not change much in relation to the number of negative previous examinations. The number of cases to be examined with colonoscopy to diagnose AN was higher for women than men (3.8 vs 2.3) and remained essentially unchanged even after several negative tests, reflecting the stability of the positive predictive value. Considering the total number of exams in people having multiple consecutive tests, the NNS necessary to find an AN increased from 61.7 to 173.2 in the second round. For people who were also negative in the second test the NNS was stable between 260 and 290 in the two subsequent rounds, always with sharp difference between men and women. Conclusions: The number needed to screen to detect an advanced neoplasm was nearly three times higher in the second (173.2) and over four times in the third (291.3) as compared to the first round (61.7), but the trend was stable thereafter. The number needed to scope to find an advanced neoplasia was 2.5 initially and around 3 in the other 3 rounds, reflecting the stability of the positive predictive value. The NNT and the NNSC were always higher in women. 1.Denis B et al. Gut 2007;56:1579-1584. 2.Hol l et al. Br J Cancer 2009; 100(7): 1103 - 1110 Table 1: NNS and NNSC to find one advanced neoplasm
Su1102 The Valuation of the Increase in Quality of Life and Health-Adjusted Life Expectancy as a Result of Colorectal Cancer Screening in Future Decades; A Population-Based Study Caroline M. den Hoed, Kees Isendoorn, Wouter Klinkhamer, Anshu M. Gupta, Ernst J. Kuipers Background Colorectal cancer (CRC) affects a considerable proportion of the population above 50 years. Population screening has a major impact on morbidity and mortality due to CRC. The cost-efficacy of screening has been widely studied, but these studies mostly did not take savings of reduced CRC treatment into account. Most importantly, the societal gains of screening have not been calculated. Such knowledge is relevant to understand the benefits of screening and to decide on implementation of screening programs. Aim The aim of this study therefore was to calculate the societal gains of implementation of nationwide colon cancer screening in terms of QALY's (quality of life adjusted life years)and HALE (health-adjusted life expectancy) as well as to provide a Dollar valuation of these societal gains. Methods Literature studies were performed for data on colon cancer epidemiology, and screening outcomes over time. National / governmental databases in the Netherlands were searched to obtain the input for calculations of QALYs, HALE, and the corresponding financial gain. Results The mortality due to colon cancer is high, 26/1000 persons of 50 year and older in 2008, with a mean loss of 12.0 life years per patient diagnosed corresponding to 52,000 life years in the total Dutch population per year. Quality of life (QoL) of patients that survive colon cancer was less affected, 1 year after diagnosis the mean QoL of patients was 63/100 compared to 64/100 in the general population. Implementing FIT-screening (Fecal immunochemical test) will lead to diagnosis at an earlier stage, with 60% of cancers diagnosed in stage 1, leading to a mortality reduction of at least 50%-80%, to a mortality of 5-13/1000 patients 50 year and older. This will lead to a rise in the mean HALE with 0.025 year per person in the population. On a population of 16.8 million people this translates to a mean reduction of otherwise lost quality-adjusted life years per year of 5,000 (16.8 million*0.025/average life expectancy at birth). Until 2040 the Dutch population will lose a mean of 70,000 life years by colon cancer per year because of high incidence due to an ageing population. Implementing FIT-screening will gain 35,000 life years every year in this period, which corresponds to 26,000 QALYs per year. The value of a life year varies from $76,000 to $230,000 per person/yr and the true financial gain for society can be calculated in several ways. The implementation of FIT-screening will thus result in a gain for the Dutch society of 1,3-5,9 billion dollar/yr depending on the method used. Conclusion This study demonstrated the implementation of FIT-screening in the coming 3 decades will
AGA Abstracts
S-404
AGA Abstracts * Advanced Neoplasm Su1105 Attendance and Diagnostic Yield of Repeated Fecal Immunochemical Test Screening With Intervals of 1, 2, or 3 Years: A Comparative Population-Based Colorectal Cancer Screening Trial Aafke H. van Roon, Simon L. Goede, Marjolein van Ballegooijen, Jacqueline C. Reijerink, Hans 't Mannetje, Anneke van Vuuren, Alexandra C. van der Togt-van Leeuwen, J. Habbema, Ernst J. Kuipers, Monique van Leerdam Introduction: Fecal immunochemical tests (FIT) have suboptimal sensitivity for advanced neoplasia and require successive screening rounds for an optimal preventive effect. However, data about the influence of screening interval length on attendance and diagnostic yield are lacking, nor is it known whether the increased sensitivity of FIT compared to the conventional guaiac-based FOBT allows for longer screening intervals. We therefore performed repeated FIT screening in a population-based trial comparing 1, 2, and 3 year intervals. Methods: Three representative samples of the Dutch population (total=7,500) aged 50-75 years were randomly selected prior to invitation. Individuals were invited for two 1-sample FIT screening rounds (cut-off ≥ 50 ng Hb/mL) with intervals of respectively 1 (group A), 2 (B), or 3 years (C). Subjects with a positive FIT in the 1st screening round, and those who had moved away/died were not invited for the 2nd round. Results: In group A, attendance was 64.6% (1,544/2,391) in the 1st and 62.7% (n=1,302/2,077) in the 2nd screening round. Of the 1st round participants, 89.8% also attended the 2nd screening round. Of the non-participants in the 1st screening round, 16.3% did participate in the 2nd round. We refer to Table 1 for group B and C. Corrected for 1st round attendance, a longer screening interval improved participation in the 2nd FIT-based screening round (group A (reference) 89.8% vs. group B 90.9%, and vs. group C 91.4%; p<0.05). Attendance of 1st round non-participants was also higher in group B and C compared to the reference group A (respectively 19.3%, and 18.9%, vs 16.3%; p<0.05). The proportion of participants attending at least one screening round did not significantly differ between the screening intervals (group A: 67.4%, group B: 66.1%, group C 67.0%; p=0.60). The overall positivity rate (PR) in the 2nd screening round was significantly lower compared to the 1st (6.0% vs. 8.4%, OR 0.69; CI 0.58-0.82). However, a longer interval was not associated with higher PR at repeated screening (p= 0.36). Similarly, the overall detection rate (DR) of advanced neoplasia was significantly lower in the 2nd round (1.8%, OR 0.55; CI 0.41-0.74) compared to the 1st screening round (3.3%). Colorectal cancers were found in 0.44% of 1st round participants vs. 0.17% in 2nd round participants (OR 0.39; CI 0.16-0.98). The length of screening interval did not affect the DR of advanced neoplasia in the 2nd round (p=0.33). Conclusion: Attendance of repeated FIT screening is moderately high with uptake percentages approximating 70%. A longer screening interval of 2 or 3 years results in significantly higher participation compared to annual screening. In the 2nd screening round, positivity and detection rates were significantly lower than in the 1st screening round. However, both were not affected by the length of the screening interval. Table 1
# Data of 792 invitees are not yet available; * P-value < 0.05; Advanced neoplasia = colorectal cancer and an adenoma ≥ 10 mm, or an adenoma with ≥ 25% villous component and/or high-grade dysplasia Su1106 Gender Disparities in Performance of a Fecal Immunochemical Test for Detection of Advanced Neoplasia Sietze T. Van Turenhout, Frank A. Oort, Jochim S. Terhaar sive Droste, Arjan P. Visscher, Veerle M. Coupe, René W. van der Hulst, Pieter Stokkers, Anneke A. Bouman, Gerrit A. Meijer, Leo G. van Rossum, Chris J. Mulder Introduction Fecal immunochemical tests (FITs) are commonly used in screening for colorectal cancer (CRC) and advanced adenomas. It has been suggested that FIT performance may be different in males and females, what could lead to over- or underestimation of overall test accuracy. However, these disparities might also be caused by gender related differences in age of onset and advanced neoplasia distribution. Thus far, little data exist on potential dissimilarities in test characteristics between the sexes. Aims & Methods Aim of the present study is to compare the confounding influence of gender on the sensitivity and specificity of a frequently used FIT, for the detection of advanced neoplasia. In this prospective cohort study, all individuals referred for elective colonoscopy were invited to perform a semiquantitative FIT (OC sensor®) prior to the start of bowel preparation for colonoscopy. Subjects in which colonoscopies were performed for evaluation of rectal bleeding or anemia were excluded to correct for potential work-up bias. In addition, participants with incomplete colonoscopy were excluded. Advanced neoplasia was defined as CRC and/or an advanced adenoma. Sensitivity and specificity for advanced neoplasia were compared at different cutoff values (50-200ng/ml), and multivariate logistic regression was used to study potential confounding on sensitivity and specificity of gender, age, number of advanced adenomas, and number of adenomas ≥1cm. Results In 1682 subjects (46% males), 213 participants were found to have at least one advanced neoplasm (54% present in males), i.e. CRC in 40 and advanced adenomas in 173 subjects. At 50ng/ml, sensitivity for advanced neoplasia was 47% (55% for men, 38% for women), and specificity was 92,0% (91,4% for men and 92,4% for women). At each cut-off level, females were found to have a significantly lower sensitivity (range of difference: 17,0-20,5%), but specificity was similar (difference -1,0 - 0,0%; see table 1). Multivariate logistic regression analysis showed that gender significantly influences the sensitivity for advanced neoplasia with or without correction for age, number of advanced adenomas, and number of adenomas ≥1cm. Conclusions Gender could significantly affect sensitivity of a fecal immunochemical test for the detection of advanced neoplasia. Adjusting for gender disparity when reporting test performance should be considered in designing a screening program. Table 1. Sensitivity and specificity of a fecal immunochemical test according to gender for the detection of advanced neoplasia.
S-405
AGA Abstracts