Risk-adjusted colorectal cancer screening using the FIT and routine screening data: development of a risk prediction model

Differences in treatment of stage I colorectal cancers: a population-based study of colorectal cancers detected within and outside of a screening program

BACKGROUND

 Screen-detected colorectal cancers (CRCs) are often treated less invasively than stage-matched non-screen-detected CRCs, but the reasons for this are not fully understood. This study evaluated the treatment of stage I CRCs detected within and outside of the screening program in the Netherlands. METHODS : Data from the Netherlands Cancer Registry for all stage I CRCs diagnosed between January 1, 2008 and December 31, 2020 were analyzed, comparing patient, tumor, and treatment characteristics of screen-detected and non-screen-detected stage I CRCs. Multivariable logistic regression was used to assess the association between treatment (local excision only vs. surgical oncologic resection) and patient and tumor characteristics, stratified for T stage and tumor location.

RESULTS

 Screen-detected stage I CRCs were relatively more often T1 than T2 compared with non-screen-detected stage I CRCs (66.9 % vs. 53.3 %; P < 0.001). When only T1 tumors were considered, both screen-detected colon and rectal cancers were more often treated with local excision only than non-screen-detected T1 cancers (odds ratio [OR] 2.19, 95 %CI 1.93-2.49; and OR 1.29, 95 %CI 1.05-1.59, respectively), adjusted for sex, tumor location, lymphovascular invasion (LVI) status, and tumor differentiation. CONCLUSIONS : Less invasive treatment of screen-detected stage I CRC is partly explained by the higher rate of T1 cancers compared with non-screen-detected stage I CRCs. T1 stage I screen-detected CRCs were also more likely to undergo less invasive treatment than non-screen-detected CRCs, adjusted for risk factors such as LVI and tumor differentiation. Future research should investigate whether the choice of local excision was related to unidentified cancer-related factors or the expertise of the endoscopists.

Improved use of faecal immunochemical tests for haemoglobin in the Scottish bowel screening programme

OBJECTIVES

This study aimed to develop a risk-scoring model in the Scottish Bowel Screening Programme incorporating faecal haemoglobin concentration with other risk factors for colorectal cancer.

METHODS

Data were collected for all individuals invited to participate in the Scottish Bowel Screening Programme between November 2017 and March 2018 including faecal haemoglobin concentration, age, sex, National Health Service Board, socioeconomic status, and screening history. Linkage with The Scottish Cancer Registry identified all screening participants diagnosed with colorectal cancer. Logistic regression was performed to identify which factors demonstrated significant association with colorectal cancer and could be used in the development of a risk-scoring model.

RESULTS

Of 232,076 screening participants, 427 had colorectal cancer: 286 diagnosed following a screening colonoscopy and 141 arising after a negative screening test result giving an interval cancer proportion of 33.0%. Only faecal haemoglobin concentration and age showed a statistically significant association with colorectal cancer. Interval cancer proportion increased with age and was higher in women (38.1%) than men (27.5%). If positivity in women were mirrored in men at each age quintile interval cancer proportion would still have remained higher in women (33.2%). Moreover, an additional 1201 colonoscopies would be required to detect 11 colorectal cancers.

CONCLUSIONS

Development of a risk scoring model using early data from the Scottish Bowel Screening Programme was not feasible due to most variables showing insignificant association with colorectal cancer. Tailoring the faecal haemoglobin concentration threshold according to age could help to diminish some of the disparity in interval cancer proportion between women and men. Strategies to achieve sex equality using faecal haemoglobin concentration thresholds depend considerably on which variable is selected for equivalency and this requires further exploration.

Diagnostic yield of a risk model versus faecal immunochemical test only: a randomised controlled trial in a colorectal cancer screening programme

BACKGROUND

Combining the faecal immunochemical test (FIT) result with risk factors for advanced neoplasia (AN) may increase the yield of colorectal cancer (CRC) screening without increasing the number of colonoscopies. We conducted a randomised controlled trial in the Dutch CRC screening programme to evaluate a previously developed risk model including FIT, age, sex, smoking status, and CRC family history.

METHODS

A total of 22,748 individuals aged 56-75 years were pre-randomised to the risk-model group or the FIT-only group. Both groups received the FIT; those allocated to the risk-model group also received a single-page questionnaire. Study participants with a positive result (FIT ≥ 15 µg Hb/g faeces and/or risk ≥0.10) were referred for colonoscopy. The primary outcome measure was the proportion of invitees in whom AN was detected.

RESULTS

In the risk-model group, 3113/11,364 invitees (27%) returned the FIT and questionnaire versus 3061/11,384 invitees (27%) in the FIT-only group (p = 0.40). The yield of AN was 3.70/1000 invitees in the risk-model group versus 3.43/1000 in the FIT-only group (absolute difference: 0.27/1000, 95%CI: -1.30 to 1.82, p = 0.82).

CONCLUSIONS

Combining FIT with risk factors for CRC did not increase the yield of AN compared to FIT-only in an existing CRC screening programme. There was no difference in participation between groups.

CLINICAL TRIAL REGISTRATION

NCT04490551 (ClinicalTrials.gov).

Cross-sectional risk models using quantitative fecal hemoglobin in colorectal cancer screening: a systematic review

INTRODUCTION

Fecal Immunochemical Testing (FIT) is a central tool in colorectal cancer (CRC) screening. To improve the selection of individuals for colonoscopy, risk models combining FIT with additional CRC risk factors have been developed. This systematic review aims to provide an overview of the current noninvasive FIT-based risk models for CRC screening to facilitate future implementation.

METHODS

We performed a systematic literature search for risk models that combined quantitative fecal hemoglobin with clinical data or noninvasive biomarkers and that were intended for CRC screening. Risk of bias was assessed using the PROBAST tool.

RESULTS

Twenty risk models reported across 29 publications were included. The overall risk of bias was high. In studies that compared risk models to FIT, 11/12 (92%) risk models had a significantly higher c-statistic than FIT only. 16/20 risk models (80%) had not been externally validated and only one model has been implemented so far.

CONCLUSION

FIT-based risk models have the potential to improve the yield of CRC screening. Unfortunately, all included publications had a high risk of bias and most risk models have not yet been externally validated. The prospect of improved CRC screening with risk models should encourage more rigorous evaluation in existing screening programs.

Impact of time between faecal immunochemical tests in colorectal cancer screening on screening results: A natural experiment

Repeated rounds of faecal immunochemical testing (FIT) for occult blood is a common method for screening for colorectal cancer (CRC). However, the time interval between FIT rounds is not thoroughly investigated. In a CRC screening trial in South-Eastern Norway, individuals were invited for biennial FIT between 2012 and 2019. The positivity threshold was >15 mcg haemoglobin/g faeces (mcg/g). Due to organizational challenges, the interval between screening rounds randomly varied between 1.5 and 3.5 years, forming a natural experiment. We investigated the detection rate of CRC and advanced neoplasia (AN: CRC or advanced adenoma) at the subsequent round (FIT₂ ), according to the faecal haemoglobin concentration (f-Hb) at the initial screening round (FIT₁ ), and time between the two screening rounds. 18 522 individuals with negative FIT₁ who attended FIT₂ were included in this study. 245 AN were detected at FIT₂ , of which 34 were CRC. The CRC detection rate at FIT₂ for participants with FIT₁  = 0 mcg/g was 0.09% while it was 0.28% for participant with 0 > FIT₁  ≤ 15 mcg/g; odds ratio (OR) 3.22, 95% CI 1.49-6.95. For each 3 months' increment between FITs, the OR for detecting CRC was 1.33 (95% CI 0.98-1.79), while the OR was 1.13 (1.02-1.26) for AN. Individuals with FIT₁ -value of 0 mcg/g, had a lower AN detection rate compared with participants with 0 > FIT₁  ≤ 15 mcg/g, irrespective of time between tests. Although CRC and AN detection rates increase with increasing time interval between FITs, individuals with undetectable f-Hb at first screen have substantially lower risk of CRC at the next screening round compared with individuals with detectable f-Hb.

Colorectal Cancer Risk Assessment and Precision Approaches to Screening: Brave New World or Worlds Apart?

Current colorectal cancer (CRC) screening recommendations take a "one-size-fits-all" approach using age as the major criterion to initiate screening. Precision screening that incorporates factors beyond age to risk stratify individuals could improve on current approaches and optimally use available resources with benefits for patients, providers, and health care systems. Prediction models could identify high-risk groups who would benefit from more intensive screening, while low-risk groups could be recommended less intensive screening incorporating noninvasive screening modalities. In addition to age, prediction models incorporate well-established risk factors such as genetics (eg, family CRC history, germline, and polygenic risk scores), lifestyle (eg, smoking, alcohol, diet, and physical inactivity), sex, and race and ethnicity among others. Although several risk prediction models have been validated, few have been systematically studied for risk-adapted population CRC screening. In order to envisage clinical implementation of precision screening in the future, it will be critical to develop reliable and accurate prediction models that apply to all individuals in a population; prospectively study risk-adapted CRC screening on the population level; garner acceptance from patients and providers; and assess feasibility, resources, cost, and cost-effectiveness of these new paradigms. This review evaluates the current state of risk prediction modeling and provides a roadmap for future implementation of precision CRC screening.

Machine Learning in Colorectal Cancer Risk Prediction from Routinely Collected Data: A Review

The inclusion of machine-learning-derived models in systematic reviews of risk prediction models for colorectal cancer is rare. Whilst such reviews have highlighted methodological issues and limited performance of the models included, it is unclear why machine-learning-derived models are absent and whether such models suffer similar methodological problems. This scoping review aims to identify machine-learning models, assess their methodology, and compare their performance with that found in previous reviews. A literature search of four databases was performed for colorectal cancer prediction and prognosis model publications that included at least one machine-learning model. A total of 14 publications were identified for inclusion in the scoping review. Data was extracted using an adapted CHARM checklist against which the models were benchmarked. The review found similar methodological problems with machine-learning models to that observed in systematic reviews for non-machine-learning models, although model performance was better. The inclusion of machine-learning models in systematic reviews is required, as they offer improved performance despite similar methodological omissions; however, to achieve this the methodological issues that affect many prediction models need to be addressed.

Risk Scoring Systems for Predicting the Presence of Colorectal Neoplasia by Fecal Immunochemical Test Results in Chinese Population

INTRODUCTION

Adherence to colonoscopy screening for colorectal cancer (CRC) is low in general populations, including those tested positive in the fecal immunochemical test (FIT). Developing tailored risk scoring systems by FIT results may allow for more accurate identification of individuals for colonoscopy.

METHODS

Among 807,109 participants who completed the primary tests in the first-round Shanghai CRC screening program, 71,023 attended recommended colonoscopy. Predictors for colorectal neoplasia were used to develop respective scoring systems for FIT-positive or FIT-negative populations using logistic regression and artificial neural network methods.

RESULTS

Age, sex, area of residence, history of mucus or bloody stool, and CRC in first-degree relatives were identified as predictors for CRC in FIT-positive subjects, while a history of chronic diarrhea and prior cancer were additionally included for FIT-negative subjects. With an area under the receiver operating characteristic curve of more than 0.800 in predicting CRC, the logistic regression-based systems outperformed the artificial neural network-based ones and had a sensitivity of 68.9%, a specificity of 82.6%, and a detection rate of 0.24% by identifying 17.6% subjects at high risk. We also reported an area under the receiver operating characteristic curve of about 0.660 for the systems predicting CRC and adenoma, with a sensitivity of 57.8%, a specificity of 64.6%, and a detection rate of 6.87% through classifying 38.1% subjects as high-risk individuals. The performance of the scoring systems for CRC was superior to the currently used method in Mainland, China, and comparable with the scoring systems incorporating the FIT results.

DISCUSSION

The tailored risk scoring systems may better identify high-risk individuals of colorectal neoplasia and facilitate colonoscopy follow-up. External validation is warranted for widespread use of the scoring systems.

Optimizing the Design of a Repeated Fecal Immunochemical Test Bowel Cancer Screening Programme With a Limited Endoscopy Capacity From a Health Economic Perspective

OBJECTIVES

In 2016, it was announced that the fecal immunochemical test (FIT) would replace the guaiac fecal occult blood test in the UK Bowel Cancer Screening Programme. England has limited endoscopy capacity. This study informed decision making by determining the most cost-effective FIT screening strategy (age range, frequency, and FIT threshold) under a constrained endoscopy capacity.

METHODS

An economic model with a colorectal cancer natural history component was used to model 60 221 screening strategies with first screening at age 50 to 60 years, screening interval of 1 to 6 years, 3+ screening episodes, and FIT integer threshold of 20 to 180 μg hemoglobin/g feces. Screening strategies requiring the same endoscopy capacity were compared to determine the characteristics of the most cost-effective strategies.

RESULTS

With 50 000 annual screening referral colonoscopies, the 20 most cost-effective strategies had a starting age of 50 to 53 years, 2-yearly screening, 7 or 8 rounds of screening, and FIT threshold of 127 to 166. Compared with a 2-yearly screening interval, screening less frequently (3-, 4-, 5-, or 6-yearly) with a more sensitive FIT was less cost-effective.

CONCLUSIONS

The UK Bowel Cancer Screening Programme should use a 2-yearly FIT screening interval. When endoscopy capacity increases, the screening starting age should be reduced first followed by reducing the FIT threshold. These findings are relevant for other colorectal cancer screening programs with constrained endoscopy capacity.

One or two faecal immunochemical tests in an organised population-based colorectal cancer screening programme in Murcia (Spain)

OBJECTIVE

Roll-out of population-based colorectal cancer (CRC) screening with faecal immunochemical test (FIT) is limited by availability of further investigations, particularly colonoscopy and examination of excised lesions. Our objective was to assess whether variation in number of faecal samples and threshold adjustment can optimise resource utilisation and CRC detection rate.

METHODS

Three different screening strategies were compared for the same FIT threshold using a quantitative FIT system: one FIT, positive when >20 µg Hb/g faeces; two FIT, positive when either was >20 µg Hb/g faeces; and two FIT, positive when the mean was >20 µg Hb/g faeces. We calculated changes in the size of population the provider could invite to screening for an equal number of screening positive results, and CRC and adenoma detected.

RESULTS

In our setting, Region of Murcia, south of Spain (not fully rolled out screening programme), changing the usual strategy of two FIT, positive when either to positive when the mean was >20 µg Hb/g faeces, would increase population invited by 37.81% with the same number of positive results (which would generate a CRC detection rate of 19.2%). In a fully rolled out programme, changing the strategy from one to two FIT (positive when the mean is >20 µg Hb/g faeces), would increase CRC detection rate by 4.64% with an increase of only 13.34% in positive FIT.

CONCLUSIONS

In a population-based CRC screening programme, smart use of number of FITs and positivity threshold can increase population invited and CRC detection without increasing the number of colonoscopies and pathological examinations needed.

Risk-stratified selection to colonoscopy in FIT colorectal cancer screening: development and temporal validation of a prediction model

BACKGROUND

Faecal immunochemical tests (FITs) yield many false positives and challenge colonoscopy capacity in colorectal cancer (CRC) screening programmes. We aimed to develop a risk-based selection of participants to undergo diagnostic colonoscopy.

METHODS

The study was observational and used registry data from the Danish CRC screening programme. We included all participants invited 2014-2016 with a positive FIT (≥ 20 μg fHb/g) who underwent colonoscopy (n = 56,459). We predicted the risk of CRC or advanced neoplasia (AN) from age, gender and FIT value using logistic regression. We evaluated calibration and discrimination and conducted temporal validation. We compared the number of CRCs and adenomas identified by risk cut-offs and by a corresponding FIT cut-off.

RESULTS

AUCs were 74.9% (95% CI: 73.6; 76.3) and 67.4% (95% CI: 66.8%; 68.0%) for the models predicting CRC and AN in the validation dataset. The cut-off of CRC risk calculated from age, gender and FIT value identified 1.03 times (95% CI: 1.02; 1.05) more CRCs and 1.01 times (95% CI: 1.01; 1.01) more medium/high-risk adenomas compared with the corresponding FIT cut-off.

CONCLUSIONS

With existing data, risk-stratified FIT screening using a risk cut-off instead of a FIT cut-off can slightly improve the selection to colonoscopy of those at highest risk of cancer and adenomas.

A scoping review of risk-stratified bowel screening: current evidence, future directions

PURPOSE

In this scoping review, we examined the international literature on risk-stratified bowel screening to develop recommendations for future research, practice and policy.

METHODS

Six electronic databases were searched from inception to 18 October 2021: Medline, Embase, PsycINFO, CINAHL, Cochrane Database of Systematic Reviews and Cochrane Central Register of Controlled Trials. Forward and backwards citation searches were also undertaken. All relevant literature were included.

RESULTS

After de-deduplication, 3,629 records remained. 3,416 were excluded at the title/abstract screening stage. A further 111 were excluded at full-text screening stage. In total, 102 unique studies were included. Results showed that risk-stratified bowel screening programmes can potentially improve diagnostic performance, but there is a lack of information on longer-term outcomes. Risk models do appear to show promise in refining existing risk stratification guidelines but most were not externally validated and less than half achieved good discriminatory power. Risk assessment tools in primary care have the potential for high levels of acceptability and uptake, and therefore, could form an important component of future risk-stratified bowel screening programmes, but sometimes the screening recommendations were not adhered to by the patient or healthcare provider. The review identified important knowledge gaps, most notably in the area of organisation of screening services due to few pilots, and what risk stratification might mean for inequalities.

CONCLUSION

We recommend that future research focuses on what organisational challenges risk-stratified bowel screening may face and a consideration of inequalities in any changes to organised bowel screening programmes.

Behavioural Challenges Associated With Risk-Adapted Cancer Screening

Cancer screening programmes have a major role in reducing cancer incidence and mortality. Traditional internationally-adopted protocols have been to invite all 'eligible individuals' for the same test at the same frequency. However, as highlighted in Cancer Research UK's 2020 strategic vision, there are opportunities to increase effectiveness and cost-effectiveness, and reduce harms of screening programmes, by making recommendations on the basis of personalised estimates of risk. In some respects, this extends current approaches of providing more intensive levels of care outside screening programmes to individuals at very high risk due to their family history or underlying conditions. However, risk-adapted colorectal cancer screening raises a wide range of questions, not only about how best to change existing programmes but also about the psychological and behavioural effects that these changes might have. Previous studies in other settings provide some important information but remain to be tested and explored further in the context of colorectal screening. Conducting behavioural science research in parallel to clinical research will ensure that risk-adapted screening is understood and accepted by the population that it aims to serve.

Predictive Modeling of Colonoscopic Findings in a Fecal Immunochemical Test-Based Colorectal Cancer Screening Program

BACKGROUND

The fecal immunochemical test (FIT) is the primary modality used by the Los Angeles County Department of Health Services (LADHS) for colorectal cancer (CRC) screening in average-risk patients. Some patients referred for FIT-positive diagnostic colonoscopy have neither adenomas nor more advanced pathology. We aimed to identify predictors of false-positive FIT (FP-FIT) results in our largely disenfranchised, low socioeconomic status population.

METHODS

We conducted a retrospective study of 596 patients who underwent diagnostic colonoscopy following a positive screening FIT. Colonoscopies showing adenomas (or more advanced pathology) were considered positive. We employed multiple logistic and linear regression as well as machine learning models (MLMs) to identify clinical predictors of FP-FIT (primary outcome) and the presence of advanced adenomas (secondary outcome).

RESULTS

Overall, 268 patients (45.0%) had a FP-FIT. Female sex and hemorrhoids (odds ratios [ORs] 1.59 and 1.89, respectively) were associated with increased odds of FP-FIT and fewer advanced adenomas (β = - 0.658 and  - 0.516, respectively). Conversely, increasing age and BMI (ORs 0.94 and 0.96, respectively) were associated with decreased odds of FP-FIT and a greater number of advanced adenomas (β = 0.073 and 0.041, respectively). MLMs predicted FP-FIT with high specificity (93.8%) and presence of advanced adenoma with high sensitivity (94.4%).

CONCLUSION

Increasing age and BMI are associated with lower odds of FP-FIT and greater number of advanced adenomas, while female sex and hemorrhoids are associated with higher odds of FP-FIT and fewer advanced adenomas. The presence of the aforementioned predictors may inform the decision to proceed with diagnostic colonoscopy in FIT-positive patients.

A user-friendly objective prediction model in predicting colorectal cancer based on 234 044 Asian adults in a prospective cohort

BACKGROUND

Prediction models of colorectal cancer (CRC) had limited application for not being user-friendly. Whether fecal immunochemical tests (FITs) can help predict CRC has been overlooked.

PATIENTS AND METHODS

With 1972 CRCs identified, 234 044 adults aged ≥40 years were successively enrolled between 1994 and 2008. Prediction models were developed by questionnaire/medical screening and quantitative FIT. NNS (number needed to scope to find one cancer) is time dependent, spanning entire study period. Significant 'risk factors' were family history, body mass index, smoking, drinking, inactivity, hypertension, diabetes, carcinoembryonic antigen, and C-reactive protein.

RESULTS

Positive FIT (≥20 μg/g hemoglobin/feces) had cancer risk 10-fold larger than negative FIT, and within each age group, another 10-fold difference. The C statistic of FIT (0.81) with age and sex alone was superior to the 'common risk-factors' model (0.73). NNS, stratified by age and by FIT values, demonstrated a scorecard of cancer risks, like 1/15 or 1/25, in 5 years. When FIT was negative, cancer risk was small (1/750-1/3000 annually). The larger the FIT, the sooner the appearance of CRC. For every 80-μg/g increase of FIT, there were 1.5-year earlier development of CRC incidence and 1-year earlier development of CRC mortality, respectively. Given the same FIT value, CRC events appeared in the proximal colon sooner than the distal colon.

CONCLUSIONS

A simple user-friendly model based on a single FIT value to predict CRC risk was developed. When positive, NNS offered a simple quantitative value, with a better precision than most risk factors, even combined. When FIT is negative, risk is very small, but requiring a repeat every other year to rule out false negative. FIT values correlated well with CRC prognosis, with worst for proximal CRC.

Faecal immunochemical testing in bowel cancer screening: Estimating outcomes for different diagnostic policies

OBJECTIVES

The National Health Service Bowel Cancer Screening Programme (NHS BCSP) in England has replaced guaiac faecal occult blood testing by faecal immunochemical testing (FIT). There is interest in fully exploiting FIT measures to improve bowel cancer (CRC) screening strategies. In this paper, we estimate the relationship of the quantitative haemoglobin concentration provided by FIT in faecal samples with underlying pathology. From this we estimate thresholds required for given levels of sensitivity to CRC and high-risk adenomas (HRA).

METHODS

Data were collected from a pilot study of FIT in England in 2014, in which 27,238 participants completed a FIT. Those with a faecal haemoglobin concentration (f-Hb) of at least 20 µg/g were referred for further investigation, usually colonoscopy. Truncated regression models were used to explore the relationship between bowel pathology and FIT results. Regression results were applied to estimate sensitivity to different abnormalities for a number of thresholds.

RESULTS

Participants with CRC and HRA had significantly higher f-Hb, and this remained unchanged after adjusting for age and sex. While a threshold of 20 μg/g was estimated to capture 82.2% of CRC and 64.0% of HRA, this would refer 7.8% of participants for colonoscopy. The current programme threshold used in England of 120 μg/g was estimated to identify 47.8% of CRC and 25.0% of HRA.

CONCLUSIONS

Under the current diagnostic policy of dichotomising FIT results, a very low threshold would be required to achieve high sensitivity to CRC and HRA, which would place further strain on colonoscopy resources. The NHS BCSP in England might benefit from a diagnostic policy that makes greater use of the quantitative nature of FIT.

Comparative yield and efficiency of strategies based on risk assessment and fecal immunochemical test in colorectal cancer screening: A cross-sectional population-based analysis

Objective

Integration of risk stratification into fecal immunochemical test (FIT) might aid in the suboptimal detection of advanced neoplasms by FIT in colorectal cancer (CRC) screening. A comparative study was conducted to evaluate the participation and diagnostic yield of the parallel combination of questionnaire-based risk assessment (QRA) and FIT, FIT-only and QRA-only strategies in a CRC screening program in China.

Methods

The study included 29,626 individuals aged 40−74 years and invited to participate in a CRC screening program in China. Participants were first invited to undertake QRA and one-time FIT (OC-sensor). Participants with positive QRA or FIT were deemed to be high-risk individuals who were recommended for subsequent colonoscopy. Participation, detection rate, and resource demand for colonoscopy were calculated and compared.

Results

Of the 29,626 invitees, 20,203 completed the parallel combination, 8,592 completed the QRA-only, and 11 completed the FIT-only strategy. For the parallel combination, FIT-only, and QRA-only strategies, the overall positivity rates were 10.2% (2,928/28,806), 5.4% (1,096/20,214), and 6.8% (1,944/28,795), respectively; the yield of advanced neoplasm per 10,000 invitees were 46.9 [95% confidence interval (95% CI): 39.8−55.4], 36.8 (95% CI: 30.5−44.4), and 12.2 (95% CI: 8.8−16.8), respectively; the positive predictive values for detecting advanced neoplasms among participants who completed colonoscopy were 4.7% (95% CI: 4.0%−5.6%), 9.9% (95% CI: 8.3%−11.9%), and 1.9% (95% CI: 1.3%−2.6%), respectively; the number of colonoscopies required to detect one advanced neoplasm was 11.4 (95% CI: 9.8−13.4), 5.7 (95% CI: 4.8−6.7), and 28.4 (95% CI: 20.7−39.2), respectively.

Conclusions

The parallel combination of QRA and FIT did not show superior efficacy for detecting advanced neoplasm compared with FIT alone in this CRC screening program.

Risk-stratified strategies in population screening for colorectal cancer

Colorectal cancer (CRC) screening has been demonstrated to reduce CRC incidence and mortality. However, besides such benefits, CRC screening is also associated with potential harmful effects. In an ideal world, screening would only be directed to the small proportion of the population that might potentially benefit. Risk‐based screening can be seen as a first step towards this ideal world, by redistributing screening resources from low‐risk to high‐risk individuals. In theory, this should result in scarce resources being used in individuals who benefit most, while intensity of screening is reduced in individuals who benefit less, hence improving the benefit‐harm ratio among all invitees. Available strategies that have been proposed for risk‐based CRC screening include using information on age, sex, prior screening history, lifestyle and/or genetic information. Implementation of risk‐based screening requires careful consideration of reliable risk prediction models, participation with screening and informed decision‐making. While it is important to recognise the limitations of current approaches, available evidence suggests that it might be feasible to start planning the introduction of tailored strategies within screening programmes. Implementing risk‐based screening based on age, sex and prior screening history alone would already represent a substantial improvement over current uniform screening approaches. We propose that it is time that screening programmes start there and continue striving towards more comprehensive approaches embedding primary prevention as an effective approach to lower risk for everyone.

Individualized faecal immunochemical test cut-off based on age and sex in colorectal cancer screening

The risk of having colorectal cancer (CRC) or its precursors vary with age and sex. Yet, most CRC screening programs using the quantitative faecal immunochemical test (FIT) use a uniform FIT cut-off. We aimed to calculate individualized FIT cut-offs based on age and sex. Data from a study of 1,112 asymptomatic average-risk screening participants undergoing colonoscopy without preselection were used to build a logistic regression model to calculate the risk of having advanced neoplasia (AN) at colonoscopy using age, sex, and FIT concentration as variables. We calculated age- and sex-adjusted FIT cut-off concentrations based on a uniform risk threshold. In a total of 101 of the 1,112 participants AN was detected at colonoscopy. We selected a risk threshold that would produce a specificity of 96.9% in the study group, matching the specificity of FIT at a cut-off of 20 µg Hb/g faeces. At this threshold, age- and sex-adjusted FIT cut-off concentrations ranged from 36.9 µg Hb/g faeces for 50-year-old women to 9.5 µg Hb/g faeces for 75-year old men. At this level of specificity, the risk-based model reached a sensitivity for AN of 28.7% (95%CI: 20.8 to 38.2) versus 27.7% (95%CI: 19.9 to 37.1) for FIT only. Using a risk threshold instead of a uniform FIT-based threshold for inviting screening participants to follow-up colonoscopy ensures that everyone has a comparable risk of AN prior to colonoscopy and may improve the detection of advanced neoplasia, although the absolute magnitude of the increase is likely to be limited.

Impact of changing from a guaiac faecal occult blood test to a faecal immunochemical test in a national screening programme: Results from a pilot study within the national bowel cancer screening programme in England

OBJECTIVE

Testing for occult blood in faeces is widely used in bowel cancer screening around the world. In many programmes, the faecal immunochemical test (FIT) is replacing the traditional guiaic faecal occult blood test (gFOBT). There have been a number of reports on the clinical impact of making this change; yet, no-one has considered the pre-analytical and analytical impact of moving from a gFOBT to a FIT bowel cancer screening programme.

METHODS

We interrogated data obtained in a FIT pilot carried out in England in 2014 to assess the timeliness of specimen collection device return time and analysis for gFOBT and FIT, the impact of time to analysis on faecal haemoglobin (f-Hb) concentration, and any differences observed between analyses carried out at two different testing laboratories.

RESULTS

FIT kits were returned on average 5.6 days sooner than gFOBT. The time to analysis for FIT leads to an overall rise in f-Hb concentration within the manufacturer's stated 14-day stability period.

CONCLUSION

Both these factors are important considerations for laboratories when considering setting up a bowel cancer screening programme, especially if transitioning from gFOBT to FIT. Our data also support previous evidence of males having a higher f-Hb than females and demonstrate that after adjusting for sex, age and screening hub, neither index of multiple deprivation nor screening episode significantly affected f-Hb.

Assessing Individual Risk for High-Risk Early Colorectal Neoplasm for Pre-Selection of Screening in Shanghai, China: A Population-Based Nested Case-Control Study

Objective

To identify people with high-risk early colorectal neoplasm is highly desirable for pre-selection in colorectal cancer (CRC) screening in low-resource countries. We aim to build and validate a risk-based model so as to improve compliance and increase the benefits of screening.

Patients and Methods

Using data from the Shanghai CRC screening cohort, we conducted a population-based nested case–control study to build a risk-based model. Cases of early colorectal neoplasm were extracted as colorectal adenomas and stage 0-I CRC. Each case was matched with five individuals without neoplasm (controls) by the screening site and year of enrollment. Cases and controls were then randomly divided into two groups, with two thirds for building the risk prediction model and the other one third for model validation. Known risk factors were included for risk prediction models using logistic regressions. The area under the receiver operating characteristic curve (AUC) and Hosmer–Lemeshow chi-square statistics were used to evaluate model discrimination and calibration. The predicted individual risk probability was calculated under the risk regression equation.

Results

The model incorporating age, sex, family history and lifestyle factors including body mass index (BMI), smoking status, alcohol, regular moderate-to-intensity physical activity showed good calibration and discrimination. When the risk cutoff threshold was defined as 17%, the sensitivity and specificity of the model were 63.99% and 53.82%, respectively. The validation data analysis also showed well discrimination.

Conclusion

A risk prediction model combining personal and lifestyle factors was developed and validated for high-risk early colorectal neoplasm among the Chinese population. This risk-based model could improve the pre-selection for screening and contribute a lot to efficient population-based screening in low-resource countries.

Microbiome Analysis of More Than 2,000 NHS Bowel Cancer Screening Programme Samples Shows the Potential to Improve Screening Accuracy

PURPOSE

There is potential for fecal microbiome profiling to improve colorectal cancer screening. This has been demonstrated by research studies, but it has not been quantified at scale using samples collected and processed routinely by a national screening program.

EXPERIMENTAL DESIGN

Between 2016 and 2019, the largest of the NHS Bowel Cancer Screening Programme hubs prospectively collected processed guaiac fecal occult blood test (gFOBT) samples with subsequent colonoscopy outcomes: blood-negative [n = 491 (22%)]; colorectal cancer [n = 430 (19%)]; adenoma [n = 665 (30%)]; colonoscopy-normal [n = 300 (13%)]; nonneoplastic [n = 366 (16%)]. Samples were transported and stored at room temperature. DNA underwent 16S rRNA gene V4 amplicon sequencing. Taxonomic profiling was performed to provide features for classification via random forests (RF).

RESULTS

Samples provided 16S amplicon-based microbial profiles, which confirmed previously described colorectal cancer-microbiome associations. Microbiome-based RF models showed potential as a first-tier screen, distinguishing colorectal cancer or neoplasm (colorectal cancer or adenoma) from blood-negative with AUC 0.86 (0.82-0.89) and AUC 0.78 (0.74-0.82), respectively. Microbiome-based models also showed potential as a second-tier screen, distinguishing from among gFOBT blood-positive samples, colorectal cancer or neoplasm from colonoscopy-normal with AUC 0.79 (0.74-0.83) and AUC 0.73 (0.68-0.77), respectively. Models remained robust when restricted to 15 taxa, and performed similarly during external validation with metagenomic datasets.

CONCLUSIONS

Microbiome features can be assessed using gFOBT samples collected and processed routinely by a national colorectal cancer screening program to improve accuracy as a first- or second-tier screen. The models required as few as 15 taxa, raising the potential of an inexpensive qPCR test. This could reduce the number of colonoscopies in countries that use fecal occult blood test screening.

Artificial Intelligence Techniques That May Be Applied to Primary Care Data to Facilitate Earlier Diagnosis of Cancer: Systematic Review

BACKGROUND

More than 17 million people worldwide, including 360,000 people in the United Kingdom, were diagnosed with cancer in 2018. Cancer prognosis and disease burden are highly dependent on the disease stage at diagnosis. Most people diagnosed with cancer first present in primary care settings, where improved assessment of the (often vague) presenting symptoms of cancer could lead to earlier detection and improved outcomes for patients. There is accumulating evidence that artificial intelligence (AI) can assist clinicians in making better clinical decisions in some areas of health care.

OBJECTIVE

This study aimed to systematically review AI techniques that may facilitate earlier diagnosis of cancer and could be applied to primary care electronic health record (EHR) data. The quality of the evidence, the phase of development the AI techniques have reached, the gaps that exist in the evidence, and the potential for use in primary care were evaluated.

METHODS

We searched MEDLINE, Embase, SCOPUS, and Web of Science databases from January 01, 2000, to June 11, 2019, and included all studies providing evidence for the accuracy or effectiveness of applying AI techniques for the early detection of cancer, which may be applicable to primary care EHRs. We included all study designs in all settings and languages. These searches were extended through a scoping review of AI-based commercial technologies. The main outcomes assessed were measures of diagnostic accuracy for cancer.

RESULTS

We identified 10,456 studies; 16 studies met the inclusion criteria, representing the data of 3,862,910 patients. A total of 13 studies described the initial development and testing of AI algorithms, and 3 studies described the validation of an AI algorithm in independent data sets. One study was based on prospectively collected data; only 3 studies were based on primary care data. We found no data on implementation barriers or cost-effectiveness. Risk of bias assessment highlighted a wide range of study quality. The additional scoping review of commercial AI technologies identified 21 technologies, only 1 meeting our inclusion criteria. Meta-analysis was not undertaken because of the heterogeneity of AI modalities, data set characteristics, and outcome measures.

CONCLUSIONS

AI techniques have been applied to EHR-type data to facilitate early diagnosis of cancer, but their use in primary care settings is still at an early stage of maturity. Further evidence is needed on their performance using primary care data, implementation barriers, and cost-effectiveness before widespread adoption into routine primary care clinical practice can be recommended.

The use of electronic healthcare records for colorectal cancer screening referral decisions and risk prediction model development

BACKGROUND

The database used for the NHS Bowel Cancer Screening Programme (BCSP) derives participant information from primary care records. Combining predictors with FOBTs has shown to improve referral decisions and accuracy. The richer data available from GP databases could be used to complement screening referral decisions by identifying those at greatest risk of colorectal cancer. We determined the availability of data for key predictors and whether this information could be used to inform more accurate screening referral decisions.

METHODS

An English BCSP cohort was derived using the electronic notifications received from the BCSP database to GP records. The cohort covered a period between 13th May 2009 to 17th January 2017. Completeness of variables and univariable associations were assessed. Risk prediction models were developed using Cox regression and multivariable fractional polynomials with backwards elimination. Optimism adjusted performance metrics were reported. The sensitivity and specificity of a combined approach using the negative FOBT model plus FOBT positive patients was determined using a probability equivalent to a 3% PPV NICE guidelines level.

RESULTS

292,059 participants aged 60-74 were derived for the BCSP screening cohort. A model including the screening test result had a C-statistic of 0.860, c-slope of 0.997, and R² of 0.597. A model developed for negative screening results only had a C-statistic of 0.597, c-slope of 0.940, and R² of 0.062. Risk predictors included in the models included; age, sex, alcohol consumption, IBS diagnosis, family history of gastrointestinal cancer, smoking status, previous negatives and whether a GP had ordered a blood test. For the combined screening approach, sensitivity increased slightly from 53.90% (FOBT only) to 58.82% but at the expense of an increased referral rate.

CONCLUSIONS

This research has identified several potential predictors for CRC in a BCSP population. A risk prediction model developed for BCSP FOBT negative patients was not clinically useful due to a low sensitivity and increased referral rate. The predictors identified in this study should be investigated in a refined algorithm combining the quantitative FIT result. Combining data from multiple sources enables fuller patient profiles using the primary care and screening database interface.

A risk-stratified approach to colorectal cancer prevention and diagnosis

Population screening and endoscopic surveillance are used widely to prevent the development of and death from colorectal cancer (CRC). However, CRC remains a major cause of cancer mortality and the increasing burden of endoscopic investigations threatens to overwhelm some health services. This Perspective describes the rationale for and approach to improved risk stratification and decision-making for CRC prevention and diagnosis. Limitations of current approaches will be discussed using the UK as an example of the challenges faced by a particular health-care system, followed by discussion of novel risk biomarker utilization. We explore how risk stratification will be advantageous to current health-care providers and users, enabling more efficient use of limited colonoscopy resources. We discuss risk stratification in the setting of population screening as well as the surveillance of high-risk groups and investigation of symptomatic patients. We also address challenges in the development and validation of risk stratification tools and identify key research priorities.

Towards risk-stratified colorectal cancer screening. Adding risk factors to the fecal immunochemical test: Evidence, evolution and expectations

With increasing incidence and mortality, colorectal cancer (CRC) is a growing health problem worldwide. An effective way to address CRC is by screening for fecal (occult) blood by the fecal immunochemical test (FIT). However, there is room for improvement since precursor lesions and CRC bleed intermittent and can therefore be missed by the FIT (false negatives) or, the detected blood did not result from precursor lesions or CRC (false positives). This review provides the latest evidence on risk prediction models using FIT combined with additional risk factors before colonoscopy, which risk factors to include and if these models will better discriminate between normal findings and CRC compared to the FIT-only. Many prediction models are known for CRC, but compared to the FIT, these are less effective in detecting CRC. The literature search resulted in 645 titles where 11 papers matched the inclusion criteria and were analyzed. Comparing the FIT-only with the risk prediction models for detecting CRC resulted in a significantly increased discrimination for the models. In addition, 2 different risk-stratification categories before colonoscopy were distinguished, namely the 1-model approach which combined risk factors with FIT results in a prediction model while the 2 step approach used risk factors apart from the FIT. Finally, combining FIT with CRC risk factors by means of a model before colonoscopy seems effective regarding discriminative power, however, more research is needed for validation combined with transparent and standardized reporting to improve quality assessment, for which suggestions are reported in this study.

Performance of two faecal immunochemical tests for the detection of advanced neoplasia at different positivity thresholds: a cross-sectional study of the Dutch national colorectal cancer screening programme

BACKGROUND

Faecal immunochemical tests (FITs) are recommended for colorectal cancer screening. Two frequently used FIT methods (FOB-Gold, Sentinel Diagnostics, Milan, Italy and OC-Sensor, Eiken Chemical, Tokyo, Japan) perform similarly in detecting advanced neoplasia (ie, colorectal cancer and advanced adenoma) at a fixed positivity cutoff for faecal haemoglobin concentration. It is unclear whether the performance of the two methods is also comparable at other thresholds. We compared the accuracy of the two assays in detecting advanced neoplasia across various thresholds.

METHODS

In a cross-sectional study in the Dutch national screening programme, individuals who were screening naive in 2016 (aged 55-75 years) living in the southwest region of the Netherlands were invited to use two different FIT assays on the same bowel movement. Eligible participants were randomly selected from municipal registers. Participants were referred for colonoscopy if either FIT assay result met the predefined positivity threshold (≥15 μg haemoglobin per g faeces). We compared the respective distributions of reported haemoglobin concentration and positivity rates with various FIT positivity thresholds. The performance of each FIT for identifying advanced neoplasia at colonoscopy in FIT-positive assays was compared with the area under the receiver operating characteristic curve.

FINDINGS

21 078 (50·0%) of 42 179 invitees completed both FIT assays. The distribution of haemoglobin concentrations differed significantly between the two FITs (p<0·0001), with higher positivity rates for OC-Sensor at FIT thresholds of 5 and 10 μg haemoglobin per g faeces, similar positivity rates at 15 and 20 μg haemoglobin per g faeces, and higher rates for FOB-Gold at FIT thresholds of 25-150 μg haemoglobin per g faeces. 2046 (9·7%) of 21 078 participants had at least one FIT assay that was positive and of these, 1724 (84·3%) attended colonoscopy. The accuracy of results in individuals undergoing colonoscopy did not significantly differ between the FITs, with an area under the receiver operating characteristic curve of 0·675 (95% CI 0·649 to 0·702) for FOB-Gold and 0·686 (0·661 to 0·712) for OC-Sensor (p=0·40). At identical positivity rates, the positive predictive value of the two FIT assays was similar (difference varying from 0·5% [95% CI -2·6 to 3·7] at a positivity rate of 3·5% to 2·4% [-2·5 to 7·3] at a positivity rate of 2·0%).

INTERPRETATION

The two widely used FITs have significantly different distributions of reported haemoglobin concentration and yield different positivity rates at equal thresholds. However, they perform similarly in detecting advanced neoplasia at a preset positivity rate. When implementing either FIT in a screening programme, the desired positivity rate that identifies participants to be referred for colonoscopy should first be set, guided by available resources and feasibility.

FUNDING

The Netherlands Organisation for Health Research and Development.

Inflammatory genes are novel prognostic biomarkers for colorectal cancer

Inflammatory genes serve a crucial role in the pathogenesis of inflammation-associated tumors. However, as recent studies have mainly focused on the effects of single inflammatory genes on colorectal cancer (CRC), but not on the global interactions between genes, the underlying mechanisms between inflammatory genes and CRC remain unclear. In the current study, two inflammation-associated networks were constructed based on inflammatory genes, differentially expressed genes (DEGs) in CRC vs. normal samples, and protein-protein interactions (PPIs). These networks included an inflammation-related neighbor network (IRNN) and an inflammation-related DEG network (IRDN). Notably, the results indicated that the inflammatory genes served as important CRC-associated genes in the IRNN. Certain inflammatory genes were more likely to be network hubs and exhibited higher betweenness centralities, indicating that these inflammatory hub genes had central roles in the communication between genes in the IRNN. By contrast, in the IRDN, functional enrichment analysis revealed that genes were enriched in numerous cancer-associated functions and pathways. Subsequently, 14 genes in a module were identified in the IRDN as the potential biomarkers associated with disease-free survival (DFS) in CRC patients in the GSE24550 dataset, the prognosis of which was further validated using three independent datasets (GSE24549, GSE34551 and GSE103479). All 14 genes (including BCAR1, CRK, FYN, GRB2, LCP2, PIK3R1, PLCG1, PTK2, PTPN11, PTPN6, SHC1, SOS1, SRC and SYK) in this module were inflammatory genes, emphasizing the critical role of inflammation in CRC. In conclusion, these findings based on integrated inflammation-associated networks provided a novel insight that may help elucidate the inflammation-mediated mechanisms involved in CRC.