Improved risk scoring systems for colorectal cancer screening in Shanghai, China

A novel colorectal cancer screening framework with feature interpretability to identify high-risk populations for colonoscopy

BACKGROUND AND AIM

Risk assessment is of paramount importance for the detection and treatment of colorectal cancer. We developed and validated a feature interpretability screening framework to identify high-risk populations and recommend colonoscopy for them.

METHODS

We utilized a training cohort consisting of 1 252 605 participants who underwent colonoscopies in Shanghai from 2013 to 2015 to develop the screening framework. We incorporated Shapley additive explanation values into feature selection to provide interpretability for the framework. Two sampling methods were separately employed to mitigate potential model bias caused by class imbalance. Furthermore, we employed various machine learning algorithms to construct risk assessment models and compared their performance. We tested the screening models on an external validation cohort of 359 462 samples and conducted comprehensive evaluation and statistical analysis of the validation results.

RESULTS

The external validation results demonstrated that the models in the proposed framework achieved sensitivity over 0.734, specificity over 0.790, and area under the receiver operating characteristic curve ranging from 0.808 to 0.859. In the predictions of the best-performing model, the prevalence rates of colorectal cancer were 0.059% and 1.056% in the low- and high-risk groups, respectively. If colonoscopies were performed only on the high-risk group predicted by the model, only 14.36% of total colonoscopies would be needed to detect 74.86% of colorectal cancer cases.

CONCLUSIONS

We developed and validated a novel framework to identify populations at high risk for colorectal cancer. Those classified as high risk should undergo colonoscopy for further diagnosis.

Age- and sex-stratified detection rates and associated factors of colorectal neoplasia in the Tianjin colorectal cancer screening program from 2012 to 2020

PURPOSE

Colorectal cancer (CRC) screening has been implemented in Tianjin, China since 2012. The objective was to estimate the neoplasia detection rate in a high-risk population by age and sex and to investigate the potential factors associated with colorectal neoplasia.

PATIENTS AND METHODS

This study is based on data of the Tianjin CRC screening program from 2012 to 2020. Residents with a positive high-risk factors questionnaire (HRFQ) or a positive faecal immunochemical test (FIT) were identified as high-risk participants and were subsequently recommended for a free colonoscopy.

RESULTS

A total of 4,117,897 eligible participants aged 40-74 years completed both a HRFQ and FIT, and 217,164 (5.3%) of them were identified as high-risk participants. Positive rates of preliminary screening increased with age and were higher in females than in males. For 57,971 participants undertaking colonoscopy, the detection rates of nonadvanced adenoma, advanced adenoma and CRC were 37.8%, 5.7% and 1.6%, respectively. Detection rates of advanced neoplasia increased from the age of 50 and were higher in males. For nonadvanced neoplasia, a strong increase was observed in males from the age of 40 and in females from the age of 50. Male sex had a greater impact on individuals aged 40-49 than on older individuals. Several factors including current smoking, drinking, and higher body mass index (BMI) were significantly associated with the presence of neoplasia, whereas, these associations were mainly restricted to individuals aged above 50 but not those aged 40-49 years.

CONCLUSIONS

These findings support that age-specific risk stratification and sex-specific initiating ages for CRC screening should be recommended to improve the accuracy and effectiveness of current screening strategy.

Diagnostic accuracy of risk assessment and fecal immunochemical test in colorectal cancer screening: Results from a population-based program and meta-analysis

BACKGROUND

Fecal immunochemical test (FIT) is a commonly used initial test for colorectal cancer (CRC) screening. Parallel use of FIT with risk assessment (RA) could improve the detection of non-bleeding lesions, but at the expense of compromising sensitivity. In this study, we evaluated the accuracy of FIT and/or RA in the Shanghai CRC screening program, and systematically reviewed the relevant evaluations worldwide.

METHODS

RA and 2-specimen FIT were used in parallel in the Shanghai screening program, followed by a colonoscopy among those with positive results. Sensitivity, specificity, detection rate of CRC, positive predictive value (PPV), and other measures with their 95% confident intervals were calculated for each type of tests and several assumed combined tests. We further searched PubMed, Embase, Web of Science, and Cochrane Library for relevant studies published in English up to January 5, 2022.

RESULTS

By the end of 2019, a total of 1,901,360 participants of the screening program completed 3,045,108 tests, with 1,901,360 first-time tests and 1,143,748 subsequent tests. Parallel use of RA and 2-specimen FIT achieved a sensitivity of 0.78 (0.77-0.80), a specificity of 0.78 (0.78-0.78), PPV of 0.89% (0.86-0.92), and a detection rate of 1.99 (1.93-2.05) for CRC per 1000 among participants enrolled in the first screening round, and performed similarly among those who participated for several times. A meta-analysis of 103 published observational studies demonstrated a higher sensitivity [0.76 (0.36, 0.94)] but a much lower specificity [0.59 (0.28, 0.85)] of parallel use of RA and FIT for detecting CRC in average-risk populations than in our subjects. One-specimen FIT, the most commonly used initial test, had a pooled specificity comparable to the Shanghai screening program (0.92 vs. 0.91), but a much higher pooled sensitivity (0.76 vs. 0.57).

CONCLUSION

Our results indicate the limitation of FIT only as an initial screening test for CRC in Chinese populations, and highlight the higher sensitivity of parallel use of RA and FIT. Attempts should be made to optimize RA to improve effectiveness of screening in the populations.

A risk scoring system for advanced colorectal neoplasia in high-risk participants to improve current colorectal cancer screening in Tianjin, China

Background

Given the limited effectiveness of the current Chinese colorectal cancer (CRC) screening procedure, adherence to colonoscopy remains low. We aim to develop and validate a scoring system based on individuals who were identified as having a high risk in initial CRC screening to achieve more efficient risk stratification and improve adherence to colonoscopy.

Methods

A total of 29,504 screening participants with positive High-Risk Factor Questionnaire (HRFQ) or faecal immunochemical test (FIT) who underwent colonoscopy in Tianjin from 2012–2020 were enrolled in this study. Binary regression analysis was used to evaluate the association between risk factors and advanced colorectal neoplasia. Internal validation was also used to assess the performance of the scoring system.

Results

Male sex, older age (age ≥ 50 years), high body mass index (BMI ≥ 28 kg/m2), current or past smoking and weekly alcohol intake were identified as risk factors for advanced colorectal neoplasm. The odds ratios (ORs) for significant variables were applied to construct the risk score ranging from 0–11: LR, low risk (score 0–3); MR, moderate risk (score 4–6); and HR, high risk (score 7–11). Compared with subjects with LR, those with MR and HR had ORs of 2.47 (95% confidence interval, 2.09–2.93) and 4.59 (95% confidence interval, 3.86–5.44), respectively. The scoring model showed an outstanding discriminatory capacity with a c-statistic of 0.64 (95% confidence interval, 0.63–0.65).

Conclusions

Our results showed that the established scoring system could identify very high-risk populations with colorectal neoplasia. Combining this risk score with current Chinese screening methods may improve the effectiveness of CRC screening and adherence to colonoscopy.

Modifiable lifestyle factors have a larger contribution to colorectal neoplasms than family history

BACKGROUND

 Screening recommendations for colorectal cancer (CRC) are mainly based on family history rather than lifestyle risk factors. We aimed to assess and compare risk factors for colorectal neoplasm (CRN) and evaluate trends in neoplasm detection rates during the three rounds of screening from 2012 to 2020 in Tianjin, China.

METHODS

This study was based on 89,535 first-recorded colonoscopies in Tianjin CRC screening program, 2012-2020. Of these, 45,380 individuals with complete family history and lifestyle factors were included for population attributable fraction (PAF) estimation.

RESULTS

 The overall detection rate of nonadvanced adenomas, advanced adenomas and CRC was 39.3%, 5.9% and 1.5%, respectively. The PAFs of current smoking, alcohol consumption, physical activity, higher BMI and family history of CRC, respectively, were 8.9%, 2.6%, 1.9%, 5.8%, and 1.1% for males with nonadvanced CRN; 12.3%, 7.3%, 4.9%, 7.2%, and 0.8% for males with advanced CRN; 3.4%, 0.4%, 2.1%, 7.8%, and 0.7% for females with nonadvanced CRN; and 4.3%, 0.2%, 8.2%, 8.5%, and -0.6% for females with advanced CRN. The PAFs of selected lifestyle factors were 19.9% for males with nonadvanced CRN, 29.0% for males with advanced CRN, 9.7% for females with nonadvanced CRN and 13.8% for females with advanced CRN.

CONCLUSIONS

 Modifiable lifestyle factors, including smoking, alcohol consumption, physical activity and BMI, have a larger contribution to CRN than family history of CRC. Our findings will provide references for developing guidelines of CRC prevention and control in China.

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.