Scholarlink

Objective <p>Colonoscopy is the gold standard for colorectal cancer (CRC) screening; however, its invasiveness, cost, and associated risks limit its use in population-wide programs. Therefore, effective noninvasive tools for identifying individuals at high risk for colorectal adenomas—the precursors to CRC—are needed. 2-deoxy-2-[¹⁸F] fluoro-D-glucose positron emission tomography/computed tomography (FDG PET/CT) captures systemic metabolic and inflammatory activity and may offer imaging biomarkers for adenoma risk stratification.</p> Methods <p>We retrospectively analyzed 754 asymptomatic individuals who underwent both colonoscopy and FDG PET/CT within 30 days as part of health screening. PET/CT-derived variables included standardized uptake values (SUVs) from visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), skeletal muscle, liver, spleen, bone marrow, and colorectal wall. Clinical data included age, sex, and body mass index (BMI). A least absolute shrinkage and selection operator (LASSO) logistic regression model was trained on 452 individuals and tested in a separate validation cohort of 302.</p> Results <p>The final LASSO model selected eight variables, including VAT area (positive association) and multiple tissue-specific SUV features (negative associations). In the test set, the model achieved an area under the curve (AUC) of 0.693 (95% confidence interval: 0.631–0.754), significantly outperforming individual predictors such as VAT area (AUC = 0.630, <i>P</i> = 0.011), VAT HU (AUC = 0.585, <i>P</i> = 0.001), and SAT SUVmax (AUC = 0.616, <i>P</i> = 0.046). Decision curve analysis demonstrated superior net clinical benefit compared to univariable models.</p> Conclusion <p>A multivariable model integrating FDG PET/CT-derived metabolic features with clinical parameters enables noninvasive prediction of colorectal adenomas. This imaging-based approach may help identify individuals most likely to benefit from colonoscopy, potentially improving the efficiency of CRC screening strategies in opportunistic or high-risk settings.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Development and validation of a LASSO-Based FDG PET/CT model for predicting colorectal adenoma in asymptomatic individuals undergoing colonoscopy

  • 期刊论文

  • 发表日期：

    2025年11月10日

查看全文
  • Jeonghyun Kang,
  • Youngmin Kim,
  • Yeongbeom Jeong,
  • Hye Sun Lee,
  • Young Hoon Ryu,
  • Tae Joo Jeon,
  • Jae-Hoon Lee

摘要

Objective

Colonoscopy is the gold standard for colorectal cancer (CRC) screening; however, its invasiveness, cost, and associated risks limit its use in population-wide programs. Therefore, effective noninvasive tools for identifying individuals at high risk for colorectal adenomas—the precursors to CRC—are needed. 2-deoxy-2-[¹⁸F] fluoro-D-glucose positron emission tomography/computed tomography (FDG PET/CT) captures systemic metabolic and inflammatory activity and may offer imaging biomarkers for adenoma risk stratification.

Methods

We retrospectively analyzed 754 asymptomatic individuals who underwent both colonoscopy and FDG PET/CT within 30 days as part of health screening. PET/CT-derived variables included standardized uptake values (SUVs) from visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), skeletal muscle, liver, spleen, bone marrow, and colorectal wall. Clinical data included age, sex, and body mass index (BMI). A least absolute shrinkage and selection operator (LASSO) logistic regression model was trained on 452 individuals and tested in a separate validation cohort of 302.

Results

The final LASSO model selected eight variables, including VAT area (positive association) and multiple tissue-specific SUV features (negative associations). In the test set, the model achieved an area under the curve (AUC) of 0.693 (95% confidence interval: 0.631–0.754), significantly outperforming individual predictors such as VAT area (AUC = 0.630, P = 0.011), VAT HU (AUC = 0.585, P = 0.001), and SAT SUVmax (AUC = 0.616, P = 0.046). Decision curve analysis demonstrated superior net clinical benefit compared to univariable models.

Conclusion

A multivariable model integrating FDG PET/CT-derived metabolic features with clinical parameters enables noninvasive prediction of colorectal adenomas. This imaging-based approach may help identify individuals most likely to benefit from colonoscopy, potentially improving the efficiency of CRC screening strategies in opportunistic or high-risk settings.