Introduction

Quantitative fluorescence molecular endoscopy (qFME) enables real-time in vivo visualization and quantification of fluorescence and can thereby highlight tumor tissue. qFME has proven its potential in detection of lesions and residual disease but is limited to the mucosal side only^1,2. Ultrasound guided needle biopsy (USNB/SFF) single fiber fluorescence spectroscopy may enable assessment of lymph nodes and all tissue layers of the esophageal wall. This pilot study evaluated the clinical impact of USNB/SFF spectroscopy using the fluorescent tracer bevacizumab-800CW for the identification of residual tumor in patients with esophageal cancer.

Aims & Methods

Patients diagnosed with locally advanced esophageal adenocarcinoma scheduled for neoadjuvant chemoradiotherapy followed by surgery were included. Patients were intravenously injected with either 4.5, 10 or 25 mg bevacizumab-800CW 2-3 days prior to the qFME to establish the optimal dose. FME collected real-time white light and fluorescence images of the primary tumor site and normal esophageal tissue. Quantitative fluorescent measurements were acquired superficially using multi-diameter single fiber reflectance / single fiber fluorescence (MDSFR/SFF) and inside the esophageal wall and within lymph nodes using USNB-SFF spectroscopy. Additionally, the reduced scattering coefficient and absorption coefficient were determined from the MDSFR/SFF data.

Results

Fifteen patients were included in the dose-escalation, where 25 mg showed the best differentiation between primary tumor site and normal tissue and was extended with 3 additional patients (TBR: 2.63±0.60, p=0.0002). VEGF immunohistochemistry confirmed the higher uptake of VEGF at the primary tumor site. USNB was evaluated in 6/8 patients from the 25mg cohort. The primary tumor site (p=0.071) and lymph nodes (p=0.090) showed higher fluorescence intensity compared to normal tissue. Fluorescence measurements of normal tissue were comparable between the luminal MDSFR/SFF and USNB/SFF measurements. In 33% of all patients residual tumor was found only below a tumor-free mucosa signifying the need to analyze the deeper tissue layers. Last, the optical properties, the reduced scattering coefficient and absorption coefficient. The combination of the intrinsic fluorescence, absorption and scattering coefficient was converted into the new variable omega and improved the differentiation between tumor and normal tissue even further.

Conclusion

We showed the potential of bevacizumab-800CW combined with qFME and USNB/SFF for response monitoring in esophageal cancer. This combination allows for both in vivo superficial and deeper tissue measurements to ensure that partial responders with a tumor-free mucosa are also adequately selected. A phase II study should establish its value for stratification of patients for active surveillance or surgery.

References

1. Tjalma JJJ, Koller M, Linssen MD, Hartmans E, de Jongh S, Jorritsma-Smit A, et al. Quantitative fluorescence endoscopy: an innovative endoscopy approach to evaluate neoadjuvant treatment response in locally advanced rectal cancer. Gut. 2020;69(3):406-10.
2. Nagengast WB, Hartmans E, Garcia-Allende PB, Peters FTM, Linssen MD, Koch M, et al. Near-infrared fluorescence molecular endoscopy detects dysplastic oesophageal lesions using topical and systemic tracer of vascular endothelial growth factor A. Gut. 2019;68(1):7-10.

Disclosure

Nothing to disclose

Introduction

Clostridioides difficile infection (CDI) is the most common cause of healthcare associated infectious diarrhea in hospitalized patients, affecting primarily elderly patients with comorbidities after antibiotic exposure. Despite introduction of novel antibiotic options, recurrent CDI (rCDI) remain a major clinical challenge. Bezlotoxumab, an anti-Toxin B monoclonal antibody used in conjunction with standard of care antibiotic therapy, has been shown to reduce the frequency of rCDI in registration trials. To date, there is limited post-marketing data on this compound concerning its clinical efficacy and safety outside trials. The aim of the BODEGA study was to assess the clinical efficacy, safety and logistics of administration of bezlotoxumab in a real-life setting.

Aims & Methods

In this observational, prospective and retrospective multicenter study, we collected data from 100 patients receiving bezlotoxumab in combination with standard of care for the prevention of rCDI. Follow-up with respect to response and safety was performed at days 10, 45 and 90. For each patient, variables with respect to demography, fitness, previous CDI treatments and CDI associated hospitalizations, diagnostic and outcome were documented. Documentation was performed by use of the online platform ClinicalSurveys.Net. For statistical analysis, SPSS version 29.0.0.0 (241) was used.

Results

At the time of this analysis, data from 88/100 unique patients with documented CDI was evaluable, as 12 had not yet completed follow-up. The evaluable study population consists of 62,5% (n=55) females with an average age of 59 years (range: 20-91). Overall, 94,3% (n=83) of patients presented with additional comorbidities. Gastrointestinal (n=33, 37,5,4%) and cardiovascular diseases (n=38, 43,2%) were most frequently documented. Patients experienced a median of 3 episodes (range=0-10) of CDI prior to the episode in which bezlotoxumab was administered. The most recent CDI episode was non-complicated in 90,9% (n=80) and community acquired in 52,9% (n=46). Patients received fidaxomicin standard (n=36, 40,9%), fidaxomicin extended (n=16, 18,2%), vancomycin standard (n=13, 14,8%), vancomycin tapered (n=13, 14,8%) and others (n=4, 4,5%) as antibiotic treatment for the episode during which bezlotoxumab was administered. In their CDI patient history, most patients received the following treatments sequentially: vancomycin and fidaxomicin (n=34, 38,6%) fidaxomicin only (n=25, 28,4%, multiple administrations possible) and vancomycin only (n=16, 18,2%, multiple administrations possible). In 76/88 patients (86,4%) and 67/83 (76,1%) there was no further recurrence on day 45 and 90, respectively.

Conclusion

Preliminary data from the BODEGA study demonstrates comparatively low recurrence rates at days 45 and 90 in a real-life setting. Our study population was characterized by cases of multiply recurrent CDI with an uncomplicated course of CDI, that was often acquired in an ambulatory care setting. In these aspects, the study population differs significantly from patients included in registration trials. Although follow-up is not yet complete for all study patients, we conclude that bezlotoxumab is a promising option for treating patients with multiply recurrent CDI.

References

N/A

Disclosure

Funded with an unrestricted grant from Merck/MSD

Introduction

Chemoradiotherapy (CRT) is a standard non-surgical treatment for esophageal squamous cell carcinoma (ESCC). One of the major disadvantages of CRT is poor local control, and salvage esophagectomy is generally indicated for local failure; however, it is invasive with a high rate of postoperative complications and mortality so that patients often refused or are intolerant to salvage esophagectomy. Alternatively, photodynamic therapy (PDT) with talaporfin sodium (TS) is less invasive salvage treatment for local failure after CRT/ radiotherapy (RT) and was reported to have relatively high local complete response (L-CR) rate and few adverse events (AEs) in the short term. However, long-term outcomes of PDT with TS are unknown.

Aims & Methods

This retrospective study aimed to clarify the long-term outcomes of PDT with TS for local failure after CRT/RT for ESCC. Patients who underwent PDT with TS for histologically proven local failure of ESCC between December 2012 and December 2021 at the National Cancer Center Hospital East were enrolled. Patients with other active cancers or a prior history of PDT were excluded. Basic indication criteria of PDT were as follows: invasion depth limited to the shallow muscle layer, <1/2 of the esophageal circumference, absence of metastasis, and patient’s refusal or unsuitable physical condition for salvage surgery. We assessed L-CR rate, incidence of AEs, clinical courses after PDT, 3-year overall survival (OS), progression free survival (PFS) and major PFS (mPFS) in which major disease progression after PDT which defined as disease progression requiring more invasive treatment than repeated endoscopic treatment and death from all causes were considered as events. We also assessed pre-PDT factors associated with L-CR. L-CR was defined as no apparent residual tumor, disappearance of post-PDT ulcer and confirmation of the scar formation, and absence of cancer cells in biopsy.

Results

108 patients (92 men; median age 74 years [range, 52-88 years]) were enrolled. Patients with pre-CRT/RT cT1/ T2/ T3/ T4/ unknown were 50/ 22/ 24/ 9/ 3. Those with residual lesion/ local recurrence/ unknown were 63/ 39/ 6. Patients with pre-PDT ycT1a/ T1b/ T2 were 6/ 85/ 17. Pre-PDT median lesion size was 20 mm (range, 4-50 mm). 60 patients achieved L-CR after PDT, 41 had L-nonCR, and 7 were not evaluable. L-CR rate of PDT was 58.8% (60/101). During median follow-up of 23 months (range, 1-90 months), 39 patients (36.1%) finally had no recurrence after PDT. 18 patients, including 8 with L-nonCR after PDT and 10 with local recurrence after achieving L-CR, underwent re-endoscopic treatments alone and remained disease-free afterward. The other 44 patients had major disease progression, and 15 received salvage surgery among them. As for the lesion characteristics associated with L-CR, the lesions with <1/2 circumference and with <20 mm in size were more likely to achieve L-CR than those with ≥1/2 circumference (64.4% vs. 16.7%, P=0.003) and with ≥20 mm (80.0% vs. 45.2%, P=0.001), respectively. The 3-year OS/ PFS/ mPFS of all patients were 78.1%/ 29.8%/ 48.1%. The 3-year OS of L-CR cases was better than that of L-nonCR cases (89.3% vs. 67.7%, P=0.003). There was no grade 3 < AEs. Non-cardiac chest pain (60.2%), skin phototoxicity (4.6%), and esophageal stricture requiring balloon dilatation (4.7%) occurred.

Conclusion

PDT with TS for local failure after CRT/RT for ESCC showed the preferable long-term outcomes. Moreover, achieving local control with PDT could improve prognosis of patients even with local failure after CRT/RT.

Introduction

Colorectal cancer is one of the most common types of cancer worldwide. The majority of patients are operable at the time of diagnosis. One of the most severe surgical complications is anastomotic leakage (AL). The incidence of AL depends on where the anastomosis is located (1-2% for ileocolic anastomoses, increasing to 10-15% of anastomoses to the rectum). AL leads to prolonged hospital stay, decreased quality of life, increased risk of cancer recurrence, and increased mortality.
Experimental studies have shown that certain bacteria can contribute to the development of AL.
In Denmark, people aged 50-74 years are offered free of charge screening for colorectal cancer. They collect a stool sample at home to be analyzed for the presence for blood (a fecal immunochemistry sample). It is possible to extract bacterial DNA from the fecal sample, to characterize the gut microbiome at the time of sampling.

Aims & Methods

The objective of the study was to use third-generation sequencing methods to analyze screening fecal samples in a cohort of colorectal cancer patients, and to characterize and compare the gut microbiome between patients with or without postoperative AL.
Fecal samples sent to the Colorectal Cancer Screening Program that tested positive for blood, from 2016-2018, were place in a biobank and stored at -80 ⁰C. By combining the biobank with data from the Danish Colorectal Cancer Group, we identified 18 patients who had surgery for colonic or rectal cancer in the period and suffered a postoperative AL (AL group). We matched these patients in a 1:2-ratio, with 36 patients without AL (Control group). We analyzed the 16S ribosomal RNA genes in the screening fecal samples with third-generation sequencing (Oxford Nanopore Technologies).

Results

The data are preliminary, as the data analysis has not yet been completed.
The AL-group and Control group were similar on basic demographic parameters (age, gender, type of resection, smoker status and ASA classification). The gut microbiome composition was significant different in the two groups, based on the two main components of a principal coordinate analysis (PCoA) chart (PERMANOVA, p-value = 0.01).
The Bacteroidota phylum was more abundant in the Control group (p < 0.05). On the order level, Campylobacterales and Bacteroidales were more abundant in the Control group (p < 0.001). On family level, Campylobacteraceae were more abundant in the Control group (p < 0.001), whereas Burkholderiaceae were abundant in the AL-group (p < 0.001). The AL-group also had more Ralstonia and Burkholderia at genus level (p < 0.001), whereas the Control group had more Campylobacter and Oridobacter genera (p < 0.001).

Conclusion

The preoperative gut microbiome composition in colorectal patients with subsequent anastomotic leakage were significantly different from the control group without anastomotic leakage. Campylobacter and Oridobacter seem protective against AL, whereas Ralstonia and Burkholderia seem harmful. We are awaiting further analysis for the species level.
Further studies should focus on validating our results, including changes in the gut microbiome prior to surgery - for instance the impact of bowel preparation – and on estimating the predictive value of our findings.

Clinical Case Summary

A 67 years old male patient had orotracheal intubation due to COVID-19 infection and respiratory insufficiency. This led to the onset of a tracheoesophageal fistula (caused by the continuous overpressure of the cuff). At admission to our Hospital, the patient was in poor clinical condition and was sent to the intensive care unit. At endoscopy, a 30 mm fistula was found.
After a first multidisciplinary evaluation, surgical treatment was attempted, but this was unsuccessful in the closure of the fistula.
In second place endoscopic suturing with Apollo Overstitch was done (double-channel gastroscope), but also this was unsuccessful in completely closing the fistula and was repeated after 5 days. On the control endoscopy after 3 days, the fistula was still there
A multidisciplinary team composed of endoscopists, anesthesiologists, bronchoscopists, and cardiologists discussed the case again. The decision was to place an Amplatzer atrial septal occluder to close the fistula (Abbott, Illinois, USA). The procedure was technically and clinically successful, and the fistula was finally closed. The patient was weaned off from intubation and then O2 therapy and was discharged after a long rehabilitation.
The stent remained in place. The patient could eat and breathe without difficulty. At 12 months follow-up the patient is completely autonomous, eats everything and is in excellent clinical condition.

Introduction

Barrett’s mucosal neoplasia can be endoscopically cured with minimal risk of lymph node metastasis while submucosally invasive neoplasia poses higher risk and usually requires surgery. Endoscopic optical diagnosis is often challenging to differentiate between the two. We aim to develop an artificial intelligence (AI) system using a deep neural network to stage Barrett’s neoplasia into mucosal (LGD, HGD, pT1a) and submucosally invasive (pT1b) lesions.

Aims & Methods

A VGG16-based deep neural network was used for the staging AI system. The model was trained with 69 images which were carefully selected from 79,159 frame images with histological ground truth. As a pre-trained model for training, the model from initial Barrett’s detection and delineation system was used, which was trained for 1,090,171 images from 161 Barrett's patients¹. We tested staging AI on pre-recorded images and its performance was compared to two experts who regularly perform Barrett's endoscopic resections.

Results

68 neoplastic images (white light, enhanced imaging, near focused/magnification) were used. 43 were mucosal (2 LGD, 31 HGD, 10 pT1a) and 25 were submucosal (6 pT1bSM1 and 19 pT1bSM2/3). The sensitivity, specificity and accuracy of AI were 80%, 84% and 82% while that of experts were 74%, 78% and 76% with no statistical difference (p=0.507).

Conclusion

This suggests that the staging AI system is good at differentiating submucosal neoplasia and it can perform as well as experts. It has the potential to help guide expert clinicians to decide if endoscopic resection of Barrett's neoplasia can be curative while helping non-expert to achieve more yield in targeted biopsy. However, more work is still needed to further refine and validate this staging AI model.

References

1. Mohamed Abdelrahim, Masahiro Saiko, Naoto Maeda, Ejaz Hossain, Asma Alkandari, Sharmila Subramaniam, Adolfo Parra-Blanco, Andres Sanchez-Yague, Emmanuel Coron, Alessandro Repici, Pradeep Bhandari, Development and Validation of Artificial Neural Networks Model for Detection of Barrett’s Neoplasia, a Multicenter Pragmatic Non-Randomized Trial, Gastrointestinal Endoscopy (2022)

Introduction

Gastric emptying (GE) of liquids and solids is differentially regulated. GE of liquids follows an exponential kinetic and mainly depends on fundic tone, GE of solids requires trituration by antral contractions for passage through the adequately opened pylorus into the duodenum. Adler et al. showed that liquid GE was more often delayed in systemic sclerosis patients compared with GE of solids (Rheumatology 2022), but knowledge on the association between GE of solids and liquids in dyspeptic patients, in general, is limited.

Aims & Methods

To investigate this association and the correlation of GE times with symptoms, we compared results of breath tests using 13C-acetate (13C-ABT, GE-liquids) and 13C-octanoic acid (13C-OBT, GE-solids) in consecutive dyspeptic patients undergoing both tests (Keller et al. UEG J 2021). Normal ranges at our institution for 13C-ABT are 20-35 min (T lag-liquids), for 13C-OBT 50-150 min (T ½-solids). All patients additionally answered the gastroparesis cardinal symptom index questionnaire (GCSI, Revicki APT 2003).

Results

302 patients were included (230 women, age 45.9 (31.7-59.5) yrs, median(IQR)). Median T lag-liquids was 25 (20-35) min, median T ½-solids 78 (60-104) min. T lag-liquids and T ½-solids were weakly associated (R=0.165, p=0.002). Both were indirectly correlated with age (R=-0.280, p<0.0001 and R=-0.179, p=0.0009, resp), and similar in men and women. Concordant results for both tests were achieved in 172 subjects (57.6%): Both were normal in 161 (53.3%), both showed accelerated GE in 10 (3.3%) and both delayed GE in 3 (1%) subjects. Discrepant results were observed in 128 subjects (42.4%), in the vast majority of these, one of the tests was normal, but in 7 subjects (2.3%), GE was accelerated in one test, delayed in the other. 67 patients (22.2%) had delayed GE in at least one of the tests. Only in this subgroup, GE time correlated significantly with GCSI (R=0.242, p=0.024), in particular with the nausea/vomiting subscale (p=0.009).

Conclusion

In a large group of dyspeptic patients, more than 45% had disturbances of liquid and/or solid GE. In less than 5% of the total group both tests were found to be either accelerated or delayed in the same patient. Instead, measurements of liquid and solid GE often differred. In patients with delayed GE in at least one of the tests, severity of gastroparesis symptoms was associated with GE time. Thus, measurements of liquid and solid GE may be complementary and identify patients with different pathomechanism leading to GE disturbances and symptoms.