Time to colonoscopy, cancer probability, and precursor lesions in the Danish colorectal cancer screening program

Prescription data and demographics: An explainable machine learning exploration of colorectal cancer risk factors based on data from Danish national registries

OBJECTIVES

Despite substantial advancements in both treatment and prevention, colorectal cancer continues to be a leading cause of global morbidity and mortality. This study investigated the potential of using demographics and prescribed drug information to predict risk of colorectal cancer using a machine learning approach.

METHODS

Five different machine learning algorithms, including Logistic Regression, XGBoost, Random Forests, kNN, and Voting Classifier, were initially developed and evaluated for their predictive capabilities across various time horizons (3, 6, 12, and 36 months). To enhance transparency and interpretability, explainable techniques were employed to understand the model's predictions and identify the relative contributions of factors like age, sex, social status, and prescribed medications, promoting trust and clinical insights. While all developed models, including simpler ones such as Logistic Regression, demonstrated comparable performance, the Voting Classifier, as an ensemble model, was selected for further investigation due to its inherent diversity and generalizability. This ensemble model combines predictions from multiple base models, reducing the risk of overfitting and improving the robustness of the final prediction.

RESULTS

The model demonstrated consistent performance across these time horizons, achieving a precision consistently above 0.99, indicating high ability in identifying patients at risk. However, the recall remained relatively low (around 0.6), highlighting the model's limitations in comprehensively identifying all at risk patients, despite its high precision. This suggests additional investigations in future studies to further enhance the performance of the proposed model.

CONCLUSION

Machine learning models can identify individuals at higher risk for developing colorectal cancer, enabling earlier interventions and personalized risk management strategies. However, further studies are needed before implementation in clinical practice.

Circulating Tumor DNA in Addition to Fecal Immunochemical Test in a Dual-Test Colorectal Cancer Screening Approach

BACKGROUND

Early detection is paramount when reducing incidence and mortality of colorectal cancer (CRC). Current population-based screening programs primarily use fecal immunochemical test (FIT) to allocate individuals for colonoscopy although low specificity challenges colonoscopy capacities. We aimed to assess the potential of circulating tumor (ct)DNA markers for early CRC detection in a dual-test CRC screening approach among FIT positive individuals.

METHODS

Plasma samples from 774 FIT positive (≥100 ng Hemoglobin/mL) individuals from the Danish CRC screening program were analyzed for hypermethylated DNA in the genes Branched Chain Amino-acid Transaminase 1 (BCAT1), Ikaros-Family Zinc Finger transcription 1 (IKZF1), and Interferon Regulator Factor 4 (IRF4). Multivariate logistic regression models were generated adding the ctDNA markers and age to the FIT value. The dual-test approach was benchmarked to FIT at specific thresholds.

RESULTS

The dual-test approach improved CRC detection compared to the FIT alone (AUC of 87.2 [95% CI, 82.9-91.4] vs AUC of 72.5 [95% CI, 67.0-77.9]). This was also seen when adding advanced adenomas to the outcome resulting in AUCs of 71.8 [95% CI, 67.8-75.8] for the dual-test approach compared to 65.5 [95% CI, 61.3-69.7] for the FIT model alone. Benchmarking the dual-test approach at FIT cut-offs between 100 and 600 ng Hb/mL showed a potential for either reducing the colonoscopy requirement by up to 56% or increasing CRC detection by up to 28%.

CONCLUSIONS

As increasing FIT cutoff will decrease CRC detection rate, application of the ctDNA panel can increase the sensitivity and specificity in a dual-test approach among asymptomatic individuals.

Performance of Faecal Immunochemical Testing for Colorectal Cancer Screening at Varying Positivity Thresholds

BACKGROUND

The positivity thresholds of faecal immunochemical testing (FIT) in colorectal cancer (CRC) screening vary between countries.

AIMS

To explore the trade-off between colonoscopies performed, adverse events and lesions detected at different FIT thresholds in a Norwegian CRC screening trial.

METHODS

We included first participation in biennial FIT screening for 47,265 individuals aged 50-74 years. Individuals with FIT > 15 μg Hb/g faeces were referred for colonoscopy. We estimated the number of colonoscopies, adverse events, screen-detected CRCs, advanced adenomas and serrated lesions expected at FIT thresholds currently or recently used in other European countries ranging between 20 and 150 μg/g.

RESULTS

At the 15 μg/g threshold (Norway), 3705 participants underwent colonoscopy, of whom 203 had CRC, 1119 advanced adenomas and 256 advanced serrated lesions. Using a 47 μg/g threshold, 1826 (49.3%) individuals would have undergone colonoscopy, and 154 (75.9%) would have been diagnosed with CRC, 702 (62.7%) with advanced adenoma and 128 (50.0%) with advanced serrated lesion compared to the 15 μg/g threshold. At 150 μg/g, the corresponding figures would have been 838 (22.6%) undergoing colonoscopy, 114 (56.2%) with CRC, 345 (30.8%) advanced adenoma and 54 (21.1%) advanced serrated lesions. The detection rate of stage I CRC was 0.22% at 15 μg/g and 0.11% at 150 μg/g. Post-colonoscopy bleeding rates were 0.8% and 1.7%, respectively.

CONCLUSIONS

Increasing the FIT threshold reduces colonoscopy demand, but substantially decreases lesion detection and unfavourably changes CRC stage distribution. The risk of adverse events at colonoscopy increased with FIT threshold, requiring country-specific information on adverse events.

TRIAL REGISTRATION

Clinicaltrials.gov identifier: NCT01538550.

Association between time to colonoscopy after positive fecal testing and colorectal cancer outcomes in Alberta, Canada

OBJECTIVE

To quantify the associations between time to colonoscopy after a positive fecal immunochemical test (FIT+) and colorectal cancer (CRC)-related outcomes in the context of a provincial, population-based CRC screening program.

SETTING

Population-based, retrospective cohort study in Alberta, Canada, including Albertans aged 50-74 with at least one FIT+ in 2014-2017.

METHODS

Study outcomes were CRC diagnosis after a FIT+ and a diagnostic follow-up colonoscopy in 2014-2019 and CRC stage at diagnosis. Multivariable logistic regression models were used to evaluate the relative risk of any CRC or advanced-stage CRC. Results were presented as crude odds ratio (OR) and adjusted OR (aOR) with 95% confidence intervals (CIs).

RESULTS

Of the 787,967 participants who had a FIT, 63,232 (8%) had a FIT+ and met the study's eligibility criteria. The risk of any CRC or advanced-stage CRC stayed high and was relatively consistent for follow-up colonoscopies performed within 1-12 months of the FIT+. After 12 months, the risk of CRC was considerably higher, particularly for advanced-stage CRC. The OR and aOR for any CRC were 1.40 (95% CI: 1.13-1.73; p < 0.05) and 1.20 (95% CI: 0.96-1.49), respectively, and the OR and aOR for advanced-stage CRC were 1.42 (95% CI: 0.98-2.08) and 0.88 (95% CI: 0.59-1.32), respectively, for colonoscopy follow-up within 12-18 months versus 1-2 months.

CONCLUSIONS

For Albertans who used FIT for CRC screening, a longer time interval between a FIT+ and follow-up colonoscopy, particularly over 12 months, increases the risk of having CRC and decreases the effectiveness of CRC screening programs.

Evaluation of the "Burgenland PREvention trial of colorectal cancer Disease with ImmunologiCal Testing" (B-PREDICT)-a population-based colorectal cancer screening program

BACKGROUND

The colorectal cancer (CRC) screening program B-PREDICT is a population based invited two stage screening project using a faecal immunochemical test (FIT) for initial screening followed by a colonoscopy for those with a positive FIT. B-PREDICT was compared with the opportunistic screening colonoscopy (OPP-COL), performed in course of the nationwide screening program.

METHODS

Within B-PREDICT all residents of the Austrian federal state Burgenland, aged between 40 and 80 are annually invited to FIT testing. All individuals who underwent initial colonoscopy in Burgenland between 01/2003 and 12/2014, were included in this study. Individuals from the FIT-triggered invited screening program B-PREDICT were compared with those from the non-FIT triggered OPP-COL.

RESULTS

15 133 individuals from B-PREDICT were compared to 10 045 individuals with OPP-COL. CRC detection rates were 1.34% (CI-95%, [1.15; 1.52]) in B-PREDICT compared to 0.54% in OPP-COL (95%-CI, [0.39; 0.68] p < 0.001). The decrease in the age standardized incidence rates of CRC was more pronounced in the population screened with FIT than in the general population screened with colonoscopy. Changes in incidence rates per year were -4.4% (95%-CI, [-5.1; -3.7]) vs. -1.8% (95%-CI, [-1.9; -1.6] p < 0.001).

CONCLUSIONS

B-PREDICT shows a two-fold higher detection rate of CRC as well as HRA compared to OPP-COL.

Postponing colonoscopy for 6 months in high-risk population increases colorectal cancer detection in China

BACKGROUND AND AIMS

Colonoscopy is an important colorectal cancer (CRC) screening modality; however, not all high-risk groups identified by fecal immunochemical test (FIT) and/or high-risk factor questionnaire (HRFQ) undergo colonoscopy in time. The impact of delays in colonoscopy on CRC detection among high-risk populations remains poorly understood, warranting further clarification.

METHODS

A retrospective study was conducted among CRC high-risk population identified by Tianjin CRC screening program. According to the colonoscopy results after HRFQ and FIT, patients were classified into CRC, advanced adenoma, non-advanced adenoma, and normal groups. The time interval between CRC screening and colonoscopy was investigated and its relationship with colonoscopy results. Logistic regression was performed to explore the risk factors of CRC detection.

RESULTS

Among the high-risk population without a history of CRC or polyps, 49,810 underwent HRFQ, FIT, and colonoscopy, and a time interval of fewer than 6 months was found for 79.56% of patients (n = 39,630). People with positive FIT were more likely to undergo colonoscopy within 6 months, and detection rates of CRC and/or advanced adenoma were positively related to time intervals. Similar results were found in people with a negative FIT but positive HRFQ. A time interval longer than 6 months was a significant predictor of CRC detection in high-risk populations.

CONCLUSION

For high-risk people identified by CRC screening, especially those with a positive FIT, a time interval of 6 months was associated with an increased probability of CRC detection. Our findings emphasize that populations at high risk should undergo colonoscopy at least within 6 months.

The national FIT-based colorectal cancer screening program in the Netherlands during the COVID-19 pandemic

The COVID-19 pandemic has affected many healthcare services worldwide. Like many other nations, the Netherlands experienced large numbers of individuals affected by COVID-19 in 2020, leading to increased demands on hospitals and intensive care units. The Dutch Ministry of Health decided to suspend the Dutch biennial fecal immunochemical test (FIT) based colorectal cancer (CRC) screening program from March 16, 2020. FIT invitations were resumed on June 3. In this study, we describe the short-term effects of this suspension on a myriad of relevant screening outcomes. As a result of the suspension, a quarter of the individuals due for screening between March and November 2020 had not received their invitation for FIT screening by November 30, 2020. Furthermore, 57.8% of those who received a consecutive FIT between the restart and November 30, 2020, received it outside the upper limit of the standard screening interval (26 months). Median time between positive FIT and colonoscopy did not change as a result of the pandemic. Participation rates of FIT screening and follow-up colonoscopy in the months just before and during the suspension were significantly lower than expected, but returned to normal levels after the suspension. Based on the anticipated 2020 cohort size, we estimate that the number of individuals with advanced neoplasia currently detected up until November 2020 was 31.2% lower compared to what would have been expected without a pandemic. Future studies should monitor the impact on long-term screening outcomes as a result of the pandemic.

Association Between Time to Colonoscopy After Positive Fecal Testing and Colorectal Cancer Outcomes: A Systematic Review

BACKGROUND & AIMS

Colonoscopy is required following a positive fecal screening test for colorectal cancer (CRC). It remains unclear to what extent time to colonoscopy is associated with CRC-related outcomes. We performed a systematic review to elucidate this relationship.

METHODS

An electronic search was performed through April 2020 for studies reporting associations between time from positive fecal testing to colonoscopy and outcomes including CRC incidence (primary outcome), CRC stage at diagnosis, and/or CRC-specific mortality. Our primary objective was to quantify these relationships following positive fecal immunochemical testing (FIT). Two authors independently performed screening, abstraction, and risk of bias assessments.

RESULTS

From 1,612 initial studies, 8 were included in the systematic review, with 5 reporting outcomes for FIT. Although meta-analysis was not possible, consistent trends between longer time delays and worse outcomes were apparent in all studies. Colonoscopy performed beyond 9 months from positive FIT compared to within 1 month was significantly associated with a higher incidence of CRC, with adjusted odds ratios (AORs) of 1.75 and 1.48 in the two largest studies. These studies also reported significant associations between colonoscopy performed beyond 9 months and higher incidence of advanced stage CRC (stage III or IV) at diagnosis, with AORs of 2.79 and 1.55, respectively.

CONCLUSIONS

Colonoscopy for positive FIT should not be delayed beyond 9 months. Given the additional time required for urgent referrals and surgical planning for CRC, colonoscopy should ideally be performed well in advance of 9 months following a positive FIT.

Impact of SARS-CoV-2 Pandemic on Colorectal Cancer Screening Delay: Effect on Stage Shift and Increased Mortality

BACKGROUND & AIMS

The SARS-CoV-2 pandemic had a sudden, dramatic impact on healthcare. In Italy, since the beginning of the pandemic, colorectal cancer (CRC) screening programs have been forcefully suspended. We aimed to evaluate whether screening procedure delays can affect the outcomes of CRC screening.

METHODS

We built a procedural model considering delays in the time to colonoscopy and estimating the effect on mortality due to up-stage migration of patients. The number of expected CRC cases was computed by using the data of the Italian screened population. Estimates of the effects of delay to colonoscopy on CRC stage, and of stage on mortality were assessed by a meta-analytic approach.

RESULTS

With a delay of 0-3 months, 74% of CRC is expected to be stage I-II, while with a delay of 4-6 months there would be a 2%-increase for stage I-II and a concomitant decrease for stage III-IV (P = .068). Compared to baseline (0-3 months), moderate (7-12 months) and long (> 12 months) delays would lead to a significant increase in advanced CRC (from 26% to 29% and 33%, respectively; P = .008 and P < .001, respectively). We estimated a significant increase in the total number of deaths (+12.0%) when moving from a 0-3-months to a >12-month delay (P = .005), and a significant change in mortality distribution by stage when comparing the baseline with the >12-months (P < .001).

CONCLUSIONS

Screening delays beyond 4-6 months would significantly increase advanced CRC cases, and also mortality if lasting beyond 12 months. Our data highlight the need to reorganize efforts against high-impact diseases such as CRC, considering possible future waves of SARS-CoV-2 or other pandemics.

A delay in colonoscopy after positive fecal tests leads to higher incidence of colorectal cancer: A systematic review and meta-analysis

BACKGROUND AND AIM

A delay in performing colonoscopies after positive fecal tests in a screening program may risk neoplastic progression. Our objective is to conduct a systematic review and meta-analysis to evaluate the effects of timing of a colonoscopy after a positive fecal test on the detection of colorectal cancer.

METHODS

Cochrane guidelines and PRISMA statement were followed for this review. Digital dissertation databases were searched from inception to June 1, 2020, and all studies reporting the detection rates of colorectal cancer on the basis of different time intervals between a positive fecal test and the post-test colonoscopy were included. We compared the detection rates of colorectal cancer (overall and advanced-stage) and advanced adenoma based on different time intervals.

RESULTS

A total of 361 637 patients from six observational studies were included for the analysis. The odds of detecting any colorectal cancer (odds ratio [OR] 1.58, 95% confidence interval [CI] 1.23-2.03, P < 0.001), advanced-stage colorectal cancer (OR 2.16, 95% CI 1.47-3.16, P < 0.001), or advanced adenomas (OR 1.17, 95% CI 1.06-1.28, P = 0.001) are significantly higher if the colonoscopies are performed after 6 months from a positive fecal test, compared with within 6 months. There was no significant difference in the detection rates based on a 1-month, a 2-month, or a 3-month cut-off.

CONCLUSIONS

A delay of colonoscopies beyond 6 months after positive fecal tests is associated with a higher odds of detecting colorectal cancer. A timely follow up of patients with positive fecal tests is warranted.

Time to Colonoscopy After Abnormal Stool-Based Screening and Risk for Colorectal Cancer Incidence and Mortality

BACKGROUND AND AIMS

The optimal time interval for diagnostic colonoscopy completion after an abnormal stool-based colorectal cancer (CRC) screening test is uncertain. We examined the association between time to colonoscopy and CRC outcomes among individuals who underwent diagnostic colonoscopy after abnormal stool-based screening.

METHODS

We performed a retrospective cohort study of veterans age 50 to 75 years with an abnormal fecal occult blood test (FOBT) or fecal immunochemical test (FIT) between 1999 and 2010. We used multivariable Cox proportional hazards to generate CRC-specific incidence and mortality hazard ratios (HRs) and 95% confidence intervals (CI) for 3-month colonoscopy intervals, with 1 to 3 months as the reference group. Association of time to colonoscopy with late-stage CRC diagnosis was also examined.

RESULTS

Our cohort included 204,733 patients. Mean age was 61 years (SD 6.9). Compared with patients who received a colonoscopy at 1 to 3 months, there was an increased CRC risk for patients who received a colonoscopy at 13 to 15 months (HR 1.13; 95% CI 1.00-1.27), 16 to 18 months (HR 1.25; 95% CI 1.10-1.43), 19 to 21 months (HR 1.28; 95% CI: 1.11-1.48), and 22 to 24 months (HR 1.26; 95% CI 1.07-1.47). Compared with patients who received a colonoscopy at 1 to 3 months, mortality risk was higher in groups who received a colonoscopy at 19 to 21 months (HR 1.52; 95% CI 1.51-1.99) and 22 to 24 months (HR 1.39; 95% CI 1.03-1.88). Odds for late-stage CRC increased at 16 months.

CONCLUSIONS

Increased time to colonoscopy is associated with higher risk of CRC incidence, death, and late-stage CRC after abnormal FIT/FOBT. Interventions to improve CRC outcomes should emphasize diagnostic follow-up within 1 year of an abnormal FIT/FOBT result.

Comparative evaluation of two colorectal cancer screening campaigns using different faecal occult blood tests in a French area

OBJECTIVE

The aim of this study was to compare quality performance of the first colorectal cancer (CRC) screening campaigns (C) with the OC Sensor® Faecal Immunological Test (FIT) (C7 from 2016 to 2017) and the Hemoccult® guaiac-based test (C1 from 2004 to 2006).

METHODS

The participation rate of the eligible population, screening fecal occult blood test (FOBT) performance indices, CRC and adenoma detection rate and time interval between test positivity and colonoscopy were studied.

RESULTS

In C7, 35.9 % of the eligible population completed the screening process versus 47.6 % in C1 (p < 0.0001). The positivity rate was of 4.3 % for OC Sensor® FIT and 2.3 % for Hemoccult® test (p < 0.0001). A total of 3,252 colonoscopies were performed in C7 versus 2,005 in C1; 246 CRCs and 1,160 advanced adenomas (AA) were detected in C7 compared to 140 CRCs and 491 AA in C1 (p < 0.0001). The FOBT cancer detection rate increased significantly from 1.4 ‰ to 2.9 ‰ between the two campaigns, as did the FOBT AA detection rate, from 5.7 ‰ to 13.7 ‰. During C7, the mean time for colonoscopy after a positive FIT result was 84.3 days [95 % CI: 77.9-90.7]. There was no significant difference between the stages at diagnosis according to the time for colonoscopy within the first 6 months.

CONCLUSIONS

CRC and AA detection rates increased significantly between the two campaigns. Longer follow-up will be required to show a potential decrease in the incidence of invasive CRCs.

Preferences for different diagnostic modalities to follow up abnormal colorectal cancer screening results: a hypothetical vignette study

OBJECTIVES

In England, a significant proportion of people who take part in the national bowel cancer screening programme (BCSP) and have a positive faecal occult blood test (FOBt) result, do not attend follow-up colonoscopy (CC). The aim of this study was to investigate differences in intended participation in a follow-up investigation by diagnostic modality offered including CC, CT colonography (CTC) or capsule endoscopy (CE).

SETTING

We performed a randomised online experiment with individuals who had previously completed an FOBt as part of the English BCSP.

METHODS

Participants (n=953) were randomly allocated to receive one of three online vignettes asking participants to imagine they had received an abnormal FOBt result, and that they had been invited for a follow-up test. The follow-up test offered was either: CC (n=346), CTC (n=302) or CE (n=305). Participants were then asked how likely they were to have their allocated test or if they refused, either of the other tests. Respondents were also asked to cite possible emotional and practical barriers to follow up testing. Multivariable logistic regression models were used to investigate intentions.

RESULTS

Intention to have the test was higher in the CTC group (96.7%) compared with the CC group (91.8%; OR 2.64; 95% CI 1.22 to 5.73). CTC was considered less 'off-putting' (OR 0.66, 95% CI 0.47 to 0.94) and less uncomfortable compared with CC (OR 0.51, 95% CI 0.34 to 0.77). For those who did not intend to have the test they were offered, CE (39.7%) or no investigation (34.5%) was preferable to CC (8.6%) or CTC (17.2%).

CONCLUSIONS

Alternative tests have the potential to increase attendance at diagnostic follow-up appointments.