UCseek: ultrasensitive early detection and recurrence monitoring of urothelial carcinoma by shallow-depth genome-wide bisulfite sequencing of urinary sediment DNA

Hypermethylated TAGMe as a universal-cancer-only methylation marker and its application in diagnosis and recurrence monitoring of urothelial carcinoma

BACKGROUND

Urothelial carcinoma (UC) is the second most common urological malignancy. Despite numerous molecular markers have been evaluated during the past decades, no urothelial markers for diagnosis and recurrence monitoring have shown consistent clinical utility.

METHODS

The methylation level of tissue samples from public database and clinical collected were analyzed. Patients with UC and benign diseases of the urinary system (BUD) were enrolled to establish TAGMe (TAG of Methylation) assessment in a training cohort (n = 567) using restriction enzyme-based bisulfite-free qPCR. The performance of TAGMe assessment was further verified in the validation cohort (n = 198). Urine samples from 57 UC patients undergoing postoperative surveillance were collected monthly for six months after surgery to assess the TAGMe methylation.

RESULTS

We identified TAGMe as a potentially novel Universal-Cancer-Only Methylation (UCOM) marker was hypermethylated in multi-type cancers and investigated its application in UC. Restriction enzyme-based bisulfite-free qPCR was used for detection, and the results of which were consistent with gold standard pyrosequencing. Importantly, hypermethylated TAGMe showed excellent sensitivity of 88.9% (95% CI: 81.4-94.1%) and specificity of 90.0% (95% CI: 81.9-95.3%) in efficiently distinguishing UC from BUD patients in urine and also performed well in different clinical scenarios of UC. Moreover, the abnormality of TAGMe as an indicator of recurrence might precede clinical recurrence by three months to one year, which provided an invaluable time window for timely and effective intervention to prevent UC upstaging.

CONCLUSION

TAGMe assessment based on a novel single target in urine is effective and easy to perform in UC diagnosis and recurrence monitoring, which may reduce the burden of cystoscopy. Trial registration ChiCTR2100052507. Registered on 30 October 2021.

Development and validation of machine learning models for diagnosis and prognosis of cancer by urinary proteomics, based on the FLEMENGHO cohort

The current study aims to develop and validate machine learning (ML) models for the prediction of cancer status by the non-invasive urinary proteomic in a population-based cohort. In this retrospective study, urinary proteome profiles in 804 cases from the FLEMENGHO cohort were measured by mass spectrometry. After feature selection by LASSO on both clinical variables and urinary proteome profile, benchmark models by clinical variables were built with six different ML algorithms. Proteome-based models and combined models were built and compared with the benchmark models. The models' performance, i.e. area under the curve (AUC) was compared by Delong method. The 95% confidence interval was estimated by the bootstrapping method. The best-performing model was explained by Shapley Additive Explanations (SHAP) method. The predictive role of proteome biomarkers in longitudinal cancer diagnosis was also explored. A clinical model, based on age, blood sugar and blood lipid profile, yielded the best AUC of 0.75 (0.68-0.82), with 0.80 (0.72-0.91) for the proteome model based on 13 selected biomarkers and 0.83 (0.77-0.90) for the combined model (P=0.01 for comparison with clinical model). SHAP on the support vector machine in the combined setting showed that except for age, proteome biomarkers contribute to the final prediction of the model. After adjusting with clinical factors, three proteome biomarkers are independent risk factors for longitudinal cancer development. Urinary proteome profiling, together with fine-tuned machine learning algorithms, demonstrates the predictive potential for cancer diagnosis transparently.