Introduction

Since currently utilized screening tools including fecal immunochemical test and serum tumor markers cannot reveal sufficient diagnostic powers for colorectal cancer (CRC) due to their low sensitivities. It is thus hopeful to discover non-invasive diagnostic biomarkers for CRC. Urine is one of the most attractive samples for mass-screening because of the non-invasiveness and easy handling. We have previously reported many types of urinary biomarkers for gastrointestinal malignancies.

Aims & Methods

We conducted the current study to discover and establish urinary protein biomarkers that enables early detection of CRC. Among 475 patients including 299 healthy controls (HCs) and 175 stage 0-III CRC patients, age- and sex-matched 373 patients were enrolled in this case-control study. A whole cohort was randomly divided into the discovery cohort (n=32), training cohort (n=220) and validation cohort (n=111). In the discovery cohort, we have performed the isobaric tags for relative and absolute quantitation (iTRAQ) -coupled liquid chromatography-mass spectrometry (LC-MS) to identify urinary protein biomarker candidates. After that, the urinary protein biomarkers were established with ELISA in the training cohort. Finally, the established biomarkers were validated with ELISA in the independent validation cohort.

Results

In the discovery cohort, we identified 78 urinary proteins with aberrant expression between HCs and CRC patients. Among them, urinary levels of 11 proteins were significantly higher in the CRC patients than in the HCs, and 4 urinary proteins were independent biomarkers on the multivariate analysis in the training cohort. In the validation cohort, these urinary proteins also showed significantly higher levels in the CRC patients than in the HCs. Of these urinary biomarkers, DPEP1 and TFF1 were top two urinary protein biomarkers for the diagnosis of CRC, and the combination biomarker panel using these two proteins showed an excellent AUC of 0.870 with 94.3% sensitivity and 63.4% specificity for the diagnosis of stage 0-III CRC. Notably, this combination biomarker panel could distinguish stage 0-I CRC that are curable with endoscopic resection, with an AUC of 0.852, 93.8% sensitivity and 61.6% specificity.

Conclusion

The established novel urinary biomarker panel provides a promising method for early detection of CRC with completely non-invasive manner.

Introduction

Liver steatosis represents the hepatic manifestation of increased adiposopathy, whose pathogenetic features have been proposed as tumorigenic triggers, specifically in colorectal cancer (CRC) [1]. Fatty liver diagnosis challenges clinicians since it is clinically silent and may be complicated with liver fibrosis. Since liver biopsy is not feasible in all patients, non-invasive and point-of-care indexes have been proposed to detect liver steatosis and fibrosis [2].

Aims & Methods

We performed a retrospective analysis among 1145 metabolic subjects to reveal any condition that may predict and concur to CRC development during a 8-years period. Anthropometric and bio humoral parameters were recorded at time zero. Non-invasive liver fibrosis scores were calculated as follows: AAR=AST/ALT; FIB-4=(age x AST)/(platelet count x √ALT); mFIB-4=(10 x age x AST)/(platelet count x ALT); Forns=7.811 - 3.131 x  ln(platelet count)  + 0.781 x ln(GGT)  + 3.467 x ln(age) - 0.014  x (total cholesterol); APRI= AST/Upper limit of normal values x 100/platelet count; AARPRI=AST/ALT/platelet count x 150. Comparisons were carried out by Student-T test (p-value <0.05 was considered significant) and ROC analysis was performed to assess cut-off values. Chi-square test was used to calculate the associated Odds Ratio (OR) with 95% confidence interval (CI).

Results

During a 8 years period, 28 patients (2.4%) developed CRC. No association between CRC development and visceral and general obesity was detected, while fasting plasma glucose was significantly higher and Vitamin D was significantly lower in patients who later were diagnosed with CRC, compared to those who did not develop cancer. We found that all non-invasive liver fibrosis scores significantly identified CRC-developer at time 0 (see Table). In ROC analysis, these scores also showed good sensitivity and specificity in predicting colon cancer (AAR AUC= 0.59, sensitivity 44% and specificity 78%; FIB-4 AUC= 0.74, sensitivity 84% and specificity 57%; mFIB-4 AUC=0.69, sensitivity 80% and specificity 53%; Forns AUC= 0.67, sensitivity 73% and specificity 54%; APRI AUC= 0.63, sensitivity 57% and specificity 75%; AARPRI AUC =0.64, sensitivity 44% and specificity 83%). We then calculated ORs for values above the ROC cut-offs finding that higher FIB-4 was associated with a 6-fold increased risk of developing CRC (95% CI=2.2-16.5), while mFIB-4 showed a OR=3.6 (95% CI=1.4-8.7), Forns OR was 3.1 (95% CI=1.3-7.9), APRI OR was 3.2 (95% CI=1.4-6.9) and AARPRI OR was 3 (95% CI=1.3-6.4).

	No Cancer (n=1117) mean ± SD	Colon cancer (n=28) mean ± SD	p-value
Fasting glucose (mg/dL)	100.8±29.1	119.5±49.8	0.0011
25-OH Vitamin D (pg/mL)	22.8±11.2	16.3±7.8	0.0099
AAR	0.8±0.3	1±0.5	0.0064
FIB-4	1.1±0.7	1.8±1	<0.0001
mFIB-4	2.2±1.5	3.3±2.3	0.0004
Forns	5.3±1.8	6.3±1.3	0.0039
APRI	0.3±0.2	0.5±0.4	0.0003
AARPRI	0.6±0.3	0.8±0.5	0.0011

Conclusion

These findings support the hypothesis that liver steatosis and fibrosis may play a role in the clinical background of CRC and bring to light the fascinating possibility of a reversed gut-liver axis communication in the pathogenesis of CRC. Thus, the use of non-invasive scores of liver fibrosis may be helpful to predict the risk of CRC and serve as novel prognostic factors for prevention and therapeutic strategies.

References

[1] Calle EE, Rodriguez C, Walker-Thurmond K, Thun MJ. Overweight, obesity, and mortality from cancer in a prospectively studied cohort of U.S. adults. N Engl J Med 2003;348:1625–38. https://doi.org/10.1056/NEJMoa021423.
[2] Hernandez Roman J, Siddiqui MS. The role of noninvasive biomarkers in diagnosis and risk stratification in nonalcoholic fatty liver disease. Endocrinol Diabetes Metab 2020;3:e00127. https://doi.org/10.1002/edm2.127.

Introduction

Over the world, fecal occult blood testing (FOBT) is used as a colorectal cancer (CRC) screening method for middle age and older age, but there are problems such as sensitivity/specificity issues, low participation rates in primary and secondary screenings, and that the incidence/mortality rates of CRC are still high. Fusobacterium nucleatum (Fn) has been reported to be frequently detected in the feces of CRC patients in recent years, and it has been reported that screening sensitivity can be increased by combining it with Fn PCR testing. However, due to the cost and technical complexity of PCR, it has not yet been widely used in clinical practice. In this study, we developed a simple method using ELISA to measure antibodies against FadA expressed by Fn and evaluated its usefulness.

Aims & Methods

We evaluated FOBT, Fn PCR quantification, and Fn (ELISA) in 41 individuals diagnosed with CRC from April 2020 to December 2022 and 41 healthy volunteers without CRC. Subgroup analysis was also performed in CRC patients at each stage.

Results

The correlation coefficient between Fn PCR quantification and Fn (ELISA) was good at r=0.722, p<0.001. FOBT alone had a sensitivity of 61% and a specificity of 100%. In Fn ELISA, when cut off#1 was set to maximize the AUC in ROC, the sensitivity and specificity were 76% and 78%, respectively, and when combined with FOBT, the sensitivity was 90% and the specificity was 78%. In screening, high specificity is required, so we set cut off#2 Fn ELISA to achieve a specificity of 93% and performed analysis. When combined with FOBT, the sensitivity was 73% and the specificity was 93%. When targeting advanced cancer of stage II or higher, the sensitivity by combining FOBT and Fn ELISA was 100% at cut off#1 and 79% at cut off#2, while for early cancer, the sensitivity was 76% at cut off#1 and 59% at cut off#2, both of which were better than the sensitivity of FOBT alone.

Conclusion

By combining Fn measurement using ELISA with FOBT, it was possible to increase the diagnostic sensitivity and reduce false negatives compared to FOBT alone. There are challenges such as setting appropriate cut-off values for use in screening, but measuring Fn in addition to FOBT may improve screening accuracy and increase the participation rate in secondary screening, and in the future, we would like to evaluate its effectiveness even in populations with low prior probability of screening participation.

Introduction

Cold snare polypectomy (CSP) for polyps 10-14 mm is unsatisfactory, with higher rates of unclear/positive histologic margins compared to small polyps,¹ which is similar for CSP with submucosal injection.² Therefore, hot snare polypectomy (HSP) or endoscopic mucosal resection (EMR) has been widely performed for polyps measuring ≥10 mm.³ However, these procedures often cause adverse events such as delayed bleeding and perforation due to deep thermal injury related to the use of electrocautery.⁴ A recent large-scale study reported a low incidence (0.9%) of invasive cancer in 10-19 mm colorectal polyps.⁵ Therefore, the development of less invasive therapeutic options for this size category is demanding.
Thus, we have developed HSP with low power pure-cut current (LPPC-HSP), a polypectomy technique with low power, low voltage, and no coagulation current (UMIN Clinical Trials Registry; UMIN000037678). This method uses pure-cut current and reduces the power of electrocautery, expecting less deep thermal damage compared with conventional HSP with blend or coagulation current, resulting in a lower risk of perforation and bleeding. This study aimed to evaluate the efficacy and safety of LPPC-HSP for colorectal polyps 10 to 14 mm, compared with conventional EMR.

Aims & Methods

In this multicenter, retrospective, observational study, records of patients who underwent EMR or LPPC-HSP for nonpedunculated colorectal polyps 10 to 14 mm at two Japanese institutions between January 2021 and March 2022 were evaluated consecutively. EMR was performed with blend cut current (Endocut Q; effect 3, time interval 2, time duration 2), while LPPC-HSP with a low power pure-cut current (Autocut; effect 1, 10 W in VIO300D or Autocut; effect 0.4 [Pmax 13 W] in VIO3). We analyzed the treatment outcomes of the two techniques using propensity score matching to reduce selection bias. The following variables expected to affect treatment outcomes were selected: age, sex, polyp size, morphology, lesion location, endoscopists (experienced endoscopists or gastrointestinal fellows), and antithrombotic agents. R0 resection was defined as en bloc resection with histologically tumor-free margins.

Results

We retrospectively enrolled 203 EMR (168 patients, 203 lesions) and 208 LPPC-HSP (169 patients, 208 lesions) cases. After propensity score matching, the baseline characteristics between the groups were comparable, with 131 pairs. The en bloc and R0 resection rates were not significantly different between EMR and LPPC-HSP groups (93.9% vs. 96.9%, P=0.38; 90.8% vs. 91.6%, P=1.00). The vertical margin was negative in all lesions except one in the EMR group. Immediate bleeding requiring hemostasis was observed in 10.7% of cases in the EMR group versus 2.3% in the LPPC-HSP group (P=0.010). The rates of delayed bleeding and perforation didn’t differ between the groups (0% vs. 0.8%, P=1.00; 0% vs. 0%).

Table 1. Baseline characteristics and treatment outcomes after propensity score matching
Baseline characteristics	EMR n=131	LPPC-HSP n=131	P value	standardized difference		Treatment outcomes	EMR n=131	LPPC-HSP n=131	P value
Age, years (interquartile range)	72 (67-76)	72 (68-78)	0.29	0.105		En bloc resection (%)	123 (93.9)	127 (96.9)	0.38
Male sex (%)	91 (70)	96 (73)	0.59	0.085		R0 resection (%)	119 (90.8)	120 (91.6)	1.00
Lesion size, mm (interquartile range)	10 (10-12)	10 (10-12)	0.93	0.017		Immediate bleeding (%)	14 (10.7)	3 (2.3)	0.01
Location right colon/left colon/rectum (%)	61 (47) / 53 (41) / 17 (13)	63 (48) / 53 (41) / 15 (12)	0.95	0.049		Delayed bleeding (%)	0 (0)	1 (0.8)	1.00
Morphology 0-Is/0-IIa (%)	89 (68) / 42 (32)	87 (66) / 44 (34)	0.90	0.033		Perforation (%)	0 (0)	0 (0)	N/A
Antithrombotic agents (%)	24 (18)	35 (27)	0.14	0.202
Gastrointestinal fellows performed (%)	42 (32)	44 (34)	0.90	0.033

Conclusion

LPPC-HSP showed treatment outcomes comparable to conventional EMR, with fewer adverse events. These results suggest that this technique is a safe and effective treatment for nonpedunculated colorectal polyps 10 to 14 mm.

References

1. Hirose R, Yoshida N, Murakami T, et al. Histopathological analysis of cold snare polypectomy and its indication for colorectal polyps 10-14 mm in diameter. Dig Endosc 2017;29:594-601.
2. Yabuuchi Y, Imai K, Hotta K, et al. Efficacy and safety of cold-snare endoscopic mucosal resection for colorectal adenomas 10 to 14 mm in size: a prospective observational study. Gastrointest Endosc 2020;92:1239-1246.
3. Ferlitsch M, Moss A, Hassan C, et al. Colorectal polypectomy and endoscopic mucosal resection (EMR): European Society of Gastrointestinal Endoscopy (ESGE) Clinical Guideline. Endoscopy 2017;49:270-297.
4. Oka S, Tanaka S, Kanao H, et al. Current status in the occurrence of postoperative bleeding, perforation and residual/local recurrence during colonoscopic treatment in Japan. Dig Endosc 2010;22:376-380.
5. Parsa N, Ponugoti P, Broadley H, et al. Risk of cancer in 10 - 19 mm endoscopically detected colorectal lesions. Endoscopy 2019;51:452-457.

Disclosure

Authors declare no Conflict of Interests for this study.

Introduction

Sessile Serrated Lesions (SSLs) show an “ultra flat” morphology, a mucus cap and are predominantly located proximally to the splenic flexure (PSF) (1). Hypothetically, previous knowledge in detection of flat or non-protruding conventional adenoma could be a predictor factor for endoscopic detection of SSLs.

Aims & Methods

Aim: To investigate in a South American Private Hospital, the relationship between Flat Adenomas Detection Rate (FADR) and Sessile Serrate Lesion Detection Rate (SDR)
Methods: This study is a retrospective, single-center analysis carried out at Centro Medico Docente La Trinidad Caracas-Venezuela from total selected colonoscopies performed by the gastroenterologists between 2008 and 2018 in patients 51-80 years old (YO). For this study, Adenoma Detection Rate (ADR) includes the whole morphologic spectrum of colon & rectum traditional adenomas from Paris Classification of Colorectal Lesions (2). We will consider the “General ADR” as “GADR”, including Non-protruding or flat ADR: “FADR” plus Protruding ADR: “PADR”, as well as the last two separately. By using logistic regression generalized estimating equation models, the location in colon & rectum and morphology of all types of adenomas (ADs) were assessed. The relationship between the global number of colonoscopies (GNC), GADR, FADR, PADR and the SDR was investigated calculating a correlation coefficient (CC) weighted by the number of total colonoscopies. In addition, 3 linear regression equation models to estimate the relationship between the 3 different types of ADR and SDR and the probability of occurrence of SDR based on the proportion of colonoscopies with SDR presenting every type of ADR weighted by the number of colonoscopies with SDR were calculated.

Results

Results of the linear regression models used to estimate the relationship between the different types of ADR and SDR and the probability of occurrence based on the number of Centro Medico Docente La Trinidad colonoscopies
Category of ADR	R2	β	P	SDR Correlation coefficient	Probability of occurrence (%)	Colonoscopies in each category of ADR validated for SDR calculation
General ADR	0.9687	0.1632	<0.001	0.9842	94.26	148
Flat ADR	0.759	1.888	0.002	0.8712	92.35	145
Protruding ADR	0.3186	0.1664	0.113	0.5644	37.58	59
R2 : Determination coefficient. This evaluates the strength of the linear relationship between the two variables.
β: Partial regression coefficient. It indicates the average effect that the increase in one unit of the predictor variable has on the dependent variable.
ADR: Adenoma Detection Rate. SDR: Serrated Detection Rate. 157 colonoscopies were valid for the calculation of the SDR

For this study, 8432 colonoscopies were validated for 9 gastroenterologists (G1-G9). For patients between 51 and 80 YO, GADR for all 9 Gastroenterologists was 19.76%. GADR-SDR individually for 9 gastroenterologists (G1-G9) varies from as high as G9 GADR-SDR: 33.21%-4.19% to as low as G1: 12.67%-0.50%. The most frequent location for ADs was PSF: OR 1.95, IC 95% (1.30-2.84), P<0.001. SSLs were mostly located PSF OR 2.18, IC 95% (1.44, 2.92) P<0.001. The morphology of ADs was mainly pedicle OR 1.00 IC 95% P<0.001 and for SSL it was flat by far: OR 6.84 IC 95% (2.08-9.02) P<0.001. The CC demonstrated a good association between GADR and SDR: 0.98 P<0.05; and for FADR: 0.87, P<0.05 but not for PADR: 0.56 P=0.113. The GNC was not associated with the SDR CC: -0.197, P=0.6117. The GADR and FADR show robust strength of the linear relationship with SDR and probability of occurrence, but not for PADR (table)

Conclusion

A sine qua non factor for a good SDR for the gastroenterologist is previous knowledge about “FADR”. To the best of our knowledge, this is the first time that ADR has been divided into “General, Flat and Protruded” in order to investigate the impact of the endoscopic skills on flat colorectal lesions detection for a good SDR.

References

1. Rex DK, Ahnen DJ, Baron JA et al. Serrated lesions of the colorectum: review and recommendations from an expert panel. Am J Gastroenterol 2012; 107: 1315-1329; quiz 1314, 1330 [PMID: 22710576 DOI: 10.1038/ ajg.2012.161]
2. Lambert R et al. Superficial Neoplastic Lesions in the Digestive Tract. Endoscopy 2005; 37: 570±578