RESUMEN
BACKGROUND & AIMS: Colonoscopy-based surveillance to prevent colorectal cancer (CRC) causes substantial burden for patients and health care. Stool tests may help to reduce surveillance colonoscopies by limiting colonoscopies to individuals at increased risk of advanced neoplasia. METHODS: This cross-sectional observational study included individuals aged 50-75 years with surveillance indication. Before bowel preparation, participants collected samples for a multitarget stool DNA test and 2 fecal immunochemical tests (FITs). Test accuracy was calculated for all surveillance indications. For the post-polypectomy indication only, which is the most common and is associated with a relatively low CRC risk, long-term impact of stool-based surveillance was evaluated with the Adenoma and Serrated Pathway to Colorectal Cancer model. Stool-based strategies were simulated to tune each test's positivity threshold to obtain strategies at least as effective as colonoscopy surveillance. RESULTS: There were 3453 individuals with results for all stool tests and colonoscopy; 2226 had previous polypectomy, 1003 had previous CRC, and 224 had a familial risk. Areas under the receiver operating characteristic curve for advanced neoplasia were 0.72 (95% CI, 0.69-0.75) for the multitarget stool DNA test, 0.61 (95% CI, 0.58-0.64) for the FIT OC-SENSOR (Eiken Chemical Co, Tokyo, Japan) and 0.59 (95% CI, 0.56-0.61) for the FIT FOB-Gold (Sentinel, Milan, Italy). Stool-based post-polypectomy surveillance strategies at least as effective as colonoscopy surveillance reduced the number of colonoscopies by 15%-41% and required 5.6-9.5 stool tests over a person's lifetime. Multitarget stool DNA-based surveillance was more costly than colonoscopy surveillance, whereas FIT-based surveillance saved costs. CONCLUSIONS: This study found that stool-based post-polypectomy surveillance strategies can be safe and cost-effective, with potential to reduce the number of colonoscopies by up to 41%. CLINICALTRIALS: gov, Number: NCT02715141.
RESUMEN
Over the past years, insights in the cancer neuroscience field increased rapidly, and a potential role for neurons in colorectal carcinogenesis has been recognized. However, knowledge on the neuronal distribution, subtypes, origin, and associations with clinicopathological characteristics in human studies is sparse. In this study, colorectal tumor tissues from the Netherlands Cohort Study on diet and cancer (n = 490) and an in-cohort validation population (n = 529) were immunohistochemically stained for the pan-neuronal markers neurofilament (NF) and protein gene product 9.5 (PGP9.5) to study the association between neuronal marker expression and clinicopathological characteristics. In addition, tumor and healthy colon tissues were stained for neuronal subtype markers, and their immunoreactivity in colorectal cancer (CRC) stroma was analyzed. NF-positive and PGP9.5-positive nerve fibers were found within the tumor stroma and mostly characterized by the neuronal subtype markers vasoactive intestinal peptide and neuronal nitric oxide synthase, suggesting that inhibitory neurons are the most prominent neuronal subtype in CRC. NF and PGP9.5 protein expression were not consistently associated with tumor stage, sublocation, differentiation grade, and median survival. NF immunoreactivity was associated with a worse CRC-specific survival in the study cohort (P = .025) independent of other prognostic factors (hazard ratio, 2.31; 95% CI, 1.33-4.03; P = .003), but these results were not observed in the in-cohort validation group. PGP9.5, in contrast, was associated with a worse CRC-specific survival in the in-cohort validation (P = .046) but not in the study population. This effect disappeared in multivariate analyses (hazard ratio, 0.81; 95% CI, 0.50-1.32; P = .393), indicating that this effect was dependent on other prognostic factors. This study demonstrates that the tumor stroma of CRC patients mainly harbors inhibitory neurons and that NF as a single marker is significantly associated with a poorer CRC-specific survival in the study cohort but necessitates future validation.
RESUMEN
PURPOSE: Colonoscopy and the fecal immunochemical test (FIT) are currently the most widely used screening modalities for colorectal cancer (CRC), however, both with their own limitations. Here we aim to identify and validate stool-based DNA methylation markers for the early detection of CRC and investigate the biological pathways prone to DNA methylation. METHODS: DNA methylation marker discovery was performed using The Cancer Genome Atlas (TCGA) colon adenocarcinoma data set consisting of normal and primary colon adenocarcinoma tissue. The performance of the five best candidate markers and a previously identified marker, NDRG4, was evaluated on tissues and whole stool samples of healthy subjects and CRC patients using quantitative MSP assays. The results were compared and combined with FIT data. Finally, pathway and gene ontology enrichment analyses were performed using ToppFun, GOrilla and clusterProfiler. RESULTS: GDNF, HAND2, SLC35F3, SNAP91 and SORCS1 were ranked as the best performing markers. Gene combinations of all five markers, NDRG4 and FIT were evaluated to establish the biomarker panel with the highest diagnostic potential, resulting in the identification of GDNF/SNAP91/NDRG4/FIT as the best performing marker panel. Pathway and gene ontology enrichment analyses revealed that genes associated with the nervous system were enriched in the set of best performing CRC-specific biomarkers. CONCLUSION: In silico discovery analysis using TCGA-derived data yielded a novel DNA-methylation-based assay for the early detection of CRC, potentially improving current screening modalities. Additionally, nervous system-related pathways were enriched in the identified genes, indicating an epigenetic regulation of neuronal genes in CRC.