Artificial Intelligence in Medicine: Ethical, Social and Legal Perspectives

Professor Joseph Sung MD, PhD Dean, Lee Kong Chian School of Medicine Nanyang Technological University Singapore

Artificial intelligence has permeated into every aspect of medicine and promise to provide accurate diagnosis, better management decision and improved outcome for patients and healthcare system. However, ethical, social and legal issues need to be resolved for successful implementation of AI tools in clinical practice. In order to gain trust and acceptance, AI algorithms should offer maximum explainability and inclusiveness. Robust evidence of benefit to patients and healthcare services has to be provided to gain justification of using these tools. Doctor-patient relationship needs to be maintained in order to gain trust and acceptance of users. Autonomy of decisions and dignity of patients need to be preserved while using machine in healthcare. Responsibility and accountability in the use of AI in medicine should be deliberated and defined in case of mishaps and damage produced. A new role of healthcare providers will emerge with the advancement of technology and changes are inevitable.

While there are numerous reports on innovations of AI in Medicine, in at least three levels, we can summarize its potential into three categories: 1. At the level individual clinician-led consultation, AI can improve the speed and accuracy of diagnosis (especially with image-interpretation assisting diagnosis), recommend and direct best choice of therapy and prognosticate outcome, 2. At the healthcare system level, AI can improve efficiency and accessibility of healthcare service, 3. At the population level, AI can help individuals in the public to modify their lifestyle and behavior, hence promoting health of the population. The power of AI in Medicine is summarized in Figure 1.

Figure 1. Potential capabilities of AI in Healthcare, from individual level, to healthcare system and population health

Successful implementation of Artificial Intelligence in gastroenterology and hepatology practice requires more than technology. There are ethical, legal and social issues that need to be settled. A group consisting of AI-developers (engineer), AI-users (gastroenterologist, hepatologist and surgeon) and AI-regulators (ethicist and administrator) formed a Working Group to draft these Positions Statements with the objective of arousing public and professional interest and dialogue, to promote ethical considerations when implementing AI technology, to suggest to policy makers and health authorities relevant factors to take into account when approving and regulating the use of AI tools, and to engage the profession in preparing for change in clinical practice. These series of Position Statements point out the salient issues to maintain the trust between care provider and care receivers, and to legitimize the use of a non-human tool in healthcare delivery. It is based on fundamental principles such as respect, autonomy, privacy, responsibility, and justice. Enforcing the use of AI without considering these factor risk damaging the doctor-patient relationship.

Investigation of non-obstructive dysphagia

Professor Prakash Gyawali Professor of Medicine Division of Gastroenterology Washington University St. Louis, Missouri USA

Dysphagia is the sensation of obstruction (food or liquid) in the chest or oesophagus. Functional dysphagia, or normal transit dysphagia, is the perception of dysphagia without mucosal, structural or motor abnormalities in the oesophagus to explain the symptoms. Normal transit dysphagia is associated with a high disease burden due to disrupted quality of life.¹

Normal transit dysphagia is uncommon as shown in a study with patients undergoing esophagogastroduodenoscopy (EGD) with dysphagia where functional dysphagia was rare and accounted for only 2.3% of all dysphagia patients.²

Patient history is critical in assessing dysphagia

There is an overlap of symptoms, which have poor specificity in differentiating organic (i.e., identifiable cause) and normal transit dysphagia.³ There are numerous differences between common clinical features among patients with organic oesophageal symptoms or functional gastrointestinal disease (FGID) (Table).³ An important distinction is that patients with organic disease have stable discrete symptoms whilst those with functional disease report multiple changeable symptoms.³

Table. Clinical features of organic versus functional gastrointestinal disease.3

Psychological factors, such as hypervigilance or anxiety, are more commonly associated with functional oesophageal disorders.³ Oesophageal Hypervigilance and Anxiety Scale (EHAS) evaluated with high-resolution manometry (HRM) was predictive of dysphagia severity, with or without motility disorders.³

Hypervigilance regarding the consumption of food resulting from previous encounters and/or anxiety can lead to food aversion and dysphagia, resulting in a decreased quality of life.⁴ Treatment can include cognitive behavioural therapy where normal eating behaviours and improved quality of life can be achieved.⁴ Patients should be evaluated thoroughly before administering treatment plans, as they may not require drug therapy.⁴

Abnormal perception of oesophageal sensations

The vagus nerve is responsible for the bidirectional perception of oesophageal sensations in the brain.⁵ Vagal afferents are most prominent in the upper gastrointestinal (GI) tract, of which there are three types: intraganglionic laminar endings, intramuscular arrays and mucosal afferents.⁵ The predominant function of vagal afferents is to relay abnormal food-related stimuli in the GI tract to the central nervous system.⁵

Both healthy individuals and those with dysphagia require multiple swallows, where mastication of food and upright postures facilitate easier clearance of bolus solids.⁶ Sensitivity for incomplete liquid and solid clearance was significantly higher in patients with dysphagia than in normal individuals, whereas the specificity was significantly higher in healthy individuals than in patients with dysphagia (Figure).⁷ In this trial, swallowed bolus stasis had a weak association with perceived dysphagia.⁷

Figure. Sensitivity (a) and specificity (b) of solid and liquid clearance in patients with dysphagia and control.7

During swallowing in healthy individuals, oesophageal distension is followed by contraction in a peristaltic manner.⁸ In patients with normal transit dysphagia, the amplitude of luminal distension is smaller, the distension waveform is more irregular than those without dysphagia, and peak distension is located earlier than contraction compared with those without dysphagia.⁸ This warrants additional study into oesophageal distension and its relation to normal transit dysphagia.⁸

Clinical diagnosis should utilize multiple tests

The pathophysiology of normal transit dysphagia focuses on peripheral triggering and perception of oesophageal symptoms and, therefore may not progress as an organic disease.⁹ An arbitrary requirement of at least 3 months of symptoms with an onset of at least 6 months before diagnosis is needed to establish chronicity.⁹ The cause of normal transit dysphagia is normally not evident in the early stages and diseases including structural or mucosal abnormalities, gastroenterology reflux disease, eosinophilic oesophagitis and motor disorders must be excluded as possible causes of symptoms.⁹

Provocative manoeuvres can identify normal transit dysphagia as demonstrated in a clinical trial using a rapid drink challenge during HRM where patients within the normal motility pattern, characterized by inhibition of pressure activity and low esophago-gastric pressure gradient, continued to report dysphagia as a symptom (n=102).¹⁰

Oesophageal motility assessed in a balloon distension study showed that patients with normal transit dysphagia had a lower median threshold to distension compared with healthy individuals.¹¹ HRM with solid swallows showed that 42% of patients with dysphagia had an abnormal sensitivity threshold and abnormal motor patterns.¹¹ In another study, 70% of the dysphagia patients experienced repeated simultaneous oesophageal contractions compared with none in normal volunteers.¹²

Functional lumen imaging probe (FLIP) is a method to assess the oesophageal motility whilst retaining similar specificity and sensitivity to supine and upright integrated relaxation pressure (as measured by HRM) in predictions of abnormal timed barium oesophagogram.¹³

FLIP can provide information on oesophageal contractility absent in manometry in patients with achalasia, as classified by high-resolution oesophageal pressure topography.¹⁴ The use of FLIP in HRM-classified normal motility patients detected 27% with abnormal EGJ distensibility, 23% with abnormal contractile response to distension, and 7% with sustained lower oesophageal sphincter contractions; this demonstrated that tests should be used together for clinical evaluation.¹⁵ Therefore, patients presenting with both normal FLIP and HRM but remaining symptomatic with dysphagia are likely to have normal transit dysphagia.

During a clinical trial, patients who had normal cardiac, HRM, and pH assessments, but still experienced chest pain symptoms, showed a significantly higher sensitivity to balloon distension compared to the control group (p=0.001) and had lower sensitivity thresholds.¹⁶ Patients with dysphagia had higher frequency and amplitude of contractions in the oesophagus despite a larger cross-sectional area during stretching, which could be attributed to a stiffer oesophagus.¹⁶ Therefore, the sensitivity due to reactivity in relation to balloon distension should be considered when evaluating patients using this method.¹⁶

Conclusion

Treatment for normal transit dysphagia should involve the assessment of the aetiology of the disease as it is often linked to a patient’s psychological state or personality traits, such as hypervigilance and/or anxiety and therefore may not require drug treatment. Diagnosis should involve multiple tests that rule out the existence of diseases other than normal transit dysphagia that may cause dysphagia.

Endoscopic treatment of obesity

Dr. Reem Sharaiha Director of Advanced & Therapeutic Endoscopy Associate Professor of Medicine, Weill Medical College Attending Physician New York-Presbyterian Hospital/Weill Cornell Medical Center USA

Treatment landscape of obesity

The Center for Disease Control estimated that 42.4% of adults in the United States have obesity, of whom 9.2% have severe obesity.¹ Obesity has been linked to a host of health complications such as type II diabetes mellitus (T2DM), cardiovascular diseases and non-alcoholic fatty liver disease (NAFLD), second only to tobacco as a leading cause of preventable deaths.¹

Treatment for obesity spans from lifestyle modification (i.e., dietary modification, exercise) to medications and surgical interventions.² Amongst the available options, bariatric surgery is the most effective treatment for obesity with long-lasting positive impact on co-morbidities such as T2DM, hypertension, cardiovascular diseases and certain cancers.^2,3 Bariatric surgery has evolved from open surgery to minimally invasive procedures using innovative laparotomy and endoscopic technology.^2,3

Endoscopic bariatric and metabolic therapies in obesity

Endoscopic bariatric and metabolic therapies (EBMTs) are gaining popularity in the treatment of obesity. They are less costly, less invasive and can be performed as an outpatient procedure; they are repeatable and reversible as well as versatile.³ EBMTs utilise an array of devices that can be applied on either the gaster or the small bowel to elicit the desired outcome of sustained weight loss and improvement of metabolic markers (Table).⁴

Intragastric balloon (IGB) functions by filling up space in the gaster, leading to reduction in food intake through the delay of gastric emptying and increasing the feeling of satiety.³ IGB can either be swallowed or placed endoscopically, filled with air or liquid, and remain fixed in place for about 4 to 12 months.³ IGB system plus lifestyle modification not just increase weight loss (15% vs 3.3% of total body weight, p<0.0001),⁵ but also improve non-alcoholic steatohepatitis (NASH) parameters significantly in comparison with lifestyle changes alone.⁶

Gastric remodelling through endoscopic sutured gastroplasty (ESG) is an incisionless, minimally invasive procedure that reduces the width and length of the stomach for weight loss.⁷ ESG has demonstrated noteworthy weight loss efficacy (20.9% total body weight at 24 months) as well as statistically significant reductions in haemoglobin A1c (HbA1c), systolic blood pressure, waist circumference, alanine aminotransferase and serum triglyceride levels.⁷ Repeating ESG is more effective than medication alone in weight recidivism with an acceptable safety profile.⁸

Table. Devices in gastric and small intestine EBMT and their effects.4

*Changes in gut hormones, gastrointestinal nutrient sensing, intestinal glucose metabolism, iron absorption and changes in microbiome. †Changes in gut hormones, gastrointestinal nutrient sensing and distal intestinal hyperstimulation. ‡Changes in bile acids, gut hormones and microbiome.

Duodenal-jejunal bypass liner is a single-use impermeable fluoropolymer sleeve implanted in the duodenum using endoscopy.⁹ This procedure mimics the traditional gastric bypass in which food travels through the sleeve, bypassing the proximal intestinal mucosa, to reduce calorie absorption.⁹ Patients who underwent this procedure had a significantly higher weight loss (24.2% vs 3.7% of total body weight) in comparison with the control group, as well as improvements in their cardiometabolic markers level.⁹

Duodenal mucosa resurfacing involves endoscopic circumferential mucosa lift and hydrothermal ablation of the mucosa to increase insulin sensitivity, potentially by increasing the unconjugated and secondary bile production.^3,10 The REVITA-1 and REVITA-2 trials showed that this technique is safe and effective in reducing HbA1c levels and improving insulin sensitivity, exerting positive effects on liver fat content as well as achieving clinically meaningful weight loss.¹⁰

NAFLD and NASH are common in patients with obesity, complications of which can lead to liver cirrhosis and fibrosis.⁴ At present, only lifestyle interventions for weight loss are recommended for NAFLD and NASH; there are no approved pharmacotherapy for both conditions.⁴ This underscores the value of EBMT intervention for sustained weight loss to mitigate hepatometabolic diseases.^4-7,9,10

Although EBMT is generally considered a costly intervention, analysis demonstrated that EBMT is cost effective for class I obesity (USD 4,105 per quality-adjusted life-year).¹¹ In addition, EBMT and bariatric surgery are found to be more cost effective for weight loss compared with medications and lifestyle interventions, underlining potential implications for improving the health systems burden due to obesity.¹¹

Innovations in EBMT have expanded into the use of robotics to automate the procedures, incisionless magnetic anastomosis systems and bariatric stents.^3,4,12 Investigations are underway to assess the effectiveness and safety of daily transient coatings of the intestine, which may be useful in patients who are ineligible for standard EBMT approaches or bariatric surgery.¹³

Conclusion

Obesity is a chronic multifactorial condition that require long-term integrated care for metabolic co-morbidities, nutrition compliance, psychological support and medication. Advances in endoscopic technology are expected to improve the safety and accessibility to effective treatments for obesity; more prospective trials are needed to assess their role in bariatric programmes and their long-term effects. Healthcare providers have a role to play in raising public awareness on obesity and maximising patient access to the most impactful interventions.

Updates on management for peritoneal surface malignancy

Dr. Simon Chu Consultant Division of Colorectal Surgery Prince of Wales Hospital Hong Kong

Peritoneal surface malignancies (PSM) consist of primary tumours including peritoneal mesothelioma or more often, peritoneal metastases of other tumours.¹

Peritoneal metastasis is conceptualised as malignant cells invading peritoneal cavities, originating from within the cavity or outside.¹ Development can occur through the spontaneous or iatrogenic detachment of primary cancer cells.¹ Subsequently, unrestricted peritoneal cancer cells adhere to the mesothelial lining or the underlying extracellular matrix and travel in the direction of the pelvis and right diaphragm.¹ During transcoelomic spread, peritoneal tumour growth often occurs at the omentum with currently unknown reasoning.¹ At the diaphragm surface, systemic metastasis arises through cancer cells in the peritoneal cavity that access the lymphatic system through specialised lymphatic stomata.¹

An analysis of data from 14 phase 3 trials in patients with metastatic colorectal cancer (CRC) showed that the overall survival was higher in patients with isolated non-peritoneal sites compared with those with peritoneal sites (adjusted hazard ratio [HR], 0.75; 95% CI, 0.63−0.91; p=0.03).² Patients with peritoneal-only metastasis experienced a significantly lower overall survival than patients with liver (p=0.012) or lung (p<0.0001) metastasis (16.3 vs 19.1 vs 24.6 months, respectively).²

Separate studies further characterise peritoneal metastatic CRC, highlighting that patients with synchronous colorectal peritoneal metastases had a lower overall survival (7−8.1 months) than those with metachronous colorectal peritoneal metastases (12−28 months).³

Cytoreductive surgery aims to reduce the disease burden

Cytoreductive surgery (CRS) utilises peritonectomy procedures that resect cancer on either visceral intra-abdominal surfaces or parietal peritoneal surfaces.⁴ PSM severity can be identified by peritoneal cavity index (PCI) defined as a quantitative combination of distributed tumours of the abdominopelvic regions with lesion size score and relate to anatomical structures in each quadrant.⁵

A retrospective review demonstrated that complete CRS and perioperative chemotherapy on patients with low-grade appendiceal mucinous neoplasms had a mean overall survival of 24.5 years.⁶ It is unknown whether these benefits can translate to colorectal or gastrointestinal peritoneal metastases so additional studies are required.

Supporting evidence and controversies in intraperitoneal chemotherapy

Systemic chemotherapy is associated with inefficiency in treating PSM due to a reduced response arising from associated impairments and potentially reduced blood flow, because of a diseased peritoneum impeding drug delivery; therefore, intraperitoneal chemotherapy (IPC) offers an additional technique.¹

Hyperthermic intraperitoneal chemotherapy (HIPEC) combines the beneficial cytotoxic effects arising from heat and the increased efficacy of certain chemotherapy molecules when heated.⁷ Intraoperative chemoperfusion can be done in an open abdomen setting or a closed setting; the latter provides reduced occupational hazard through lower risk of cytotoxic contamination from the surgical staff involved.⁷

CRS with HIPEC can have survival benefits as evidenced in a propensity score-matched cohort study that analysed data from patients with histological pseudomyxoma peritonei treated with CRS, with or without HIPEC.⁸ The weighted overall 5-year survival was 57.8% in the CRS with HIPEC group versus 46.2% in the CRS alone group (weighted HR, 0.65; 95% CI, 0.50−0.83; p<0.001).⁸

The benefits of IPCs are related to the potency of chemotherapy drugs and the degree of CRS.¹ In a single-centre retrospective study with data spanning 30 years, patients with complete CRS had significantly higher overall survival than those with incomplete resection (33.4 vs 12.7 months; p<0.0001).⁹ Within patients with complete CRS, the overall survival was negatively impacted by higher PCI, beginning from PCI score nine (HR, 1.98; 95% CI, 1.39−2.82; p=0.0001).⁹

Although randomised controlled trials have demonstrated the efficacy of CRS with IPC in comparison with standard chemotherapy (Table), the PRODIGE-7 study found that overall survival did not differ significantly between CRS plus HIPEC (41.7 months; 95% CI, 36.2−53.8) and CRS alone (41.2 months; 95% CI, 35.1−49.7) in patients with peritoneal metastatic CRC (Figure).^10-12 Also grade 3 or higher adverse events at 60 days were more common in the CRS plus HIPEC group compared with the CRS alone group (26% vs 15%, p=0.035).¹²

Table. Summary of two trials utilising cytoreductive surgery with intraperitoneal chemotherapy.10,11

CRS, cytoreductive surgery; HIPEC, hyperthermic intraperitoneal chemotherapy; HR, hazard ratio; IPC, intraperitoneal chemotherapy; SC, standard chemotherapy

Figure. PRODIGE-7 Study, Kaplan-Meier estimates of overall survival.12

In the updated American Society of Clinical Oncology guidelines, PRODIGE-7 has been an influential trial whereby CRS with oxaliplatin-based HIPEC is not recommended as a treatment for patients with colorectal peritoneal metastases.¹³

However, in the PRODIGE-7 study, CRS with HIPEC was favoured in patients with PCI scores 11 to 15 and it was inconclusive whether the use of mitomycin instead of oxaliplatin is more beneficial.¹³ Therefore, the takeaways of these studies should be that completeness of CRS is a cornerstone for successful management and patients should be selected carefully based on PCI.

Advances in management

Molecular biomarkers may help predict prognosis, in particular, mutated BRAF, as shown in a retrospective study of patients with peritoneal metastatic CRC undergoing CRS with HIPEC. Mutated BRAF was the only mutation associated with poor disease-free survival (7 months; 95% CI, 2.1−11.9) in comparison with the wildtype (16 months; 95% CI, 11.7−43.3; p=0.008).¹⁴ BRAF mutation was also found to be an independent predictor of disease recurrence (p=0.032).¹⁴ Thus patients with a BRAF mutation, even with favourable CRS, may not have a favourable treatment prognosis, so patients should be selected carefully.

Pre-operative radiological assessment is critical in management and should utilise multiple functional imaging modalities to improve the assessment of disease resectability.¹⁵ Positron emission tomography-computed tomography scans often record false-negative results due to small tumour deposits or mucinous tumours, so the combination with other techniques like magnetic resonance imaging can help increase detection sensitivity.¹⁶

Fluorescence molecular imaging offers an additional modality to distinguish between normal and malignant tissues, allowing for precise intraoperative positioning to assist in the resection of micro-cancer tissues, improving the degree of cytoreduction and overall patient outcomes.¹⁷

Incorporations of laparoscopic approaches to CRS can be beneficial in the length of hospital stay as demonstrated in an international registry-based analysis on patients who underwent laparoscopic CRS plus HIPEC; there was a 61% 5-year overall survival for colorectal originating tumours and a median length of stay of 7 days.¹⁸

Pressurised intraperitoneal aerosol chemotherapy (PIPAC), often used laparoscopically or alongside systemic chemotherapy, is a novel method of delivering chemotherapy agents and offers an alternative for patients ineligible for CRS with/without HIPEC or those with an insufficient response.¹⁹ Promising results were found in a retrospective cohort study involving patients with peritoneal metastatic CRC where three treatments of PIPAC achieved a median peritoneal regression grading score of 2.1 and a symptomatic population of 46% (vs 54% at baseline).¹⁹

Summary

The future aim of the management of peritoneal malignancy should be to create a curative treatment that can extend to colorectal and gastrointestinal peritoneal metastasis. Peritoneal malignancy treatment is dependent on multiple factors and current options include adjuvant chemotherapy, CRS, HIPEC and potentially PIPAC. Patient characteristics like PCI, BRAF mutation status and overall fitness should be considered before implementing management techniques.

Advances in the management of GERD

Dr. Marc Tin-Long Wong Associate Consultant Clinical Assistant Professor (Honorary) Department of Medicine of Therapeutics Prince of Wales Hospital Hong Kong

Gastroesophageal reflux disease (GERD) remains one of the most common diseases observed in patients by gastroenterologists, surgeons and primary healthcare physicians.^1,2 A survey in 2019 conducted in Hong Kong revealed that out of the patients who underwent endoscopic examination, a median of 46% were diagnosed with GERD.³

The diagnosis of GERD, highlighted in the Lyon Consensus, is well-validated and stringent,⁴ but in terms of treatment, not all patients respond well to pharmacological therapy or lifestyle modifications, in which case surgical interventions should be considered as an alternative. The objectives of anti-reflux surgery (ARS), which include both surgical and endoscopic approaches, in principle involves reconstruction of the oesophagogastric junction (OGJ) barrier through the strengthening of the lower oesophageal sphincter (LES) complex or realignment of the LES to the diaphragm, in order to achieve a reduction in oesophageal acid exposure.¹

Physician guidelines like the American College of Gastroenterology (ACG) recommend ARS as a long-term treatment of patients with objective GERD, with methods such as transoral incisionless fundoplication (TIF), or magnetic sphincter augmentation (MSA) as an alternative to traditional laparoscopic fundoplication.¹ The American Gastroenterological Association (AGA) recommends laparoscopic fundoplication, MSA and TIF as effective surgical and endoscopic options in proven GERD.² The Society of American Gastrointestinal and Endoscopic Surgeons (SAGES) surgical guidelines, on the other hand, suggests options including robotic laparoscopic and partial or complete fundoplication as surgical options in paediatrics and adults.⁵ Overall, there is no uniform consensus of which ARS is most preferred. Although endoscopic treatments are relatively new with a paucity of long-term data, there is intense interest as it is minimally invasive. Indeed, the decision for ARS are multidisciplinary and often require shared decision-making between surgeons and patients before ARS.^5,6

Mucosal resection is a viable endoscopic treatment for GERD

Anti-reflux mucosectomy (ARMS) is a minimally invasive endoscopic technique that relieves GERD symptoms by altering the OGJ through induction of fibrotic scar.^7,8 Efficacy in improving GERD symptoms with ARMS was shown in a clinical pilot study among 10 treatment-refractory GERD patients (Figure 1), in which the fraction of time at pH less than 4 (over 24 hours) reduced from 29.1% to 3.1% (p=0.1).⁸

Figure 1. DeMeester score before and after ARMS.8

Similar results were shown in a study retrospectively reviewing 109 patients with PPI-refractory GERD who underwent ARMS.⁹ There were significant improvements in symptom scores across 3 years of follow-up and roughly 40% to 50% were able to discontinue PPI.⁹

Anti-reflux mucosal ablation (ARMA) is essentially a scar-inducing procedure of similar concept, with a comparable.¹⁰ In 12 patients with PPI-refractory GERD, ARMA improved GERD health-related quality of life (HRQOL) scores, frequency scale for the symptoms of GERD score and DeMeester score.¹⁰

The efficacy of these endoscopic modalities was further demonstrated in a meta-analysis that highlighted a 73.8% (95% CI, 69−78%) clinical success rate overall − 68.6% in ARMS and 86.7% in ARMA.¹¹ However, it also revealed that after ARMS or ARMA, 10% of the population (with GERD) experienced dysphagia requiring endoscopic dilation.¹¹ Notably, severe hiatal hernia was a predictor of treatment failure with ARMA,¹¹ therefore dysphagia and severe hiatal hernia should be actively sought for during patient selection.

Evidence for radiofrequency ablation is inconclusive

Another endoscopic modality altering the OGJ is radiofrequency ablation used in the LES region, aiming to increase LES pressure, reduce transient LES relaxations and minimise acid exposure.¹²

Although initially recommended by SAGES (2013), a metanalysis across four trials on patients with GERD (N=163) showed that there was no difference between radiofrequency ablation (Stretta) and sham procedures or PPI therapy in GERD HRQOL, discontinuation of PPI therapy, pH less than 4 over a 24-hour period and LES pressure.¹³ Therefore, in view of varying and inconclusive data, the updated guidelines by the ACG do not recommend radiofrequency ablation as an alternative to medical treatment or ARS.¹

Transoral fundoplication provides sustained improvement

TIF is well established in the United States and allows for the reconstruction of the OGJ barrier.⁶ In a prospective, randomised study with 696 patients with troublesome regurgitation despite PPI therapy, TIF eliminated troublesome regurgitation in a larger proportion of patients (67% in intention to treat) compared with PPI therapy (45%).⁶ The mean percentage of total time that pH was less than 4 was improved after TIF (before, 9.3%; after, 6.3%) compared with sham procedures (before, 8.6%; after, 8.9%).⁶

Sustained improvements of TIF were demonstrated in a multicentre, open-label, randomised study where at the 5-year follow-up, among patients who underwent TIF, troublesome symptoms were eliminated in 86% of patients while 46% of patients discontinued PPI therapy.¹⁴ GERD HRQOL score also improved significantly (Figure 2).¹⁴

Figure 2. GERD HRQOL questionnaire at screening and 1-, 3- and 5-year follow-up assessments in patients who underwent TIF.14

Emerging endoscopic management techniques

New endoscopic devices are continuously being developed, for example, MUSE™ endoscopic stapling device demonstrated at 4 years post procedure that 69.4% of patients remained off PPI therapy and GERD HRQOL scores significantly improved (p<0.01).¹⁵

Lastly, patients having undergone peroral endoscopic myotomy (POEM), a treatment for achalasia cardia, is frequently associated with post-procedural GERD¹⁶ and poses a unique clinical challenge as they invariably have absent oesophageal peristalsis. There has been interest in related procedures such as POEM with fundoplication (POEM+F). A recent retrospective analysis of patients undergoing POEM-F achieved a 92% (23/25) technical success rate, and after 1 year, 11.1% (2/18) of the available cohort had diagnosed GERD.¹⁶

The role of phenotyping GERD

The pathophysiology of GERD is very complex and multifactorial,⁴ therefore to better triage management techniques, physicians should investigate the mechanistic phenotyping of GERD.

Determination of the oesophageal peristaltic function is of paramount importance. Ineffective oesophageal motility or absent contractility detected by high-resolution manometry (HRM) can indicate problems with oesophageal clearance and thus predict post-ARS dysphagia. In addition, provocative manoeuvres during HRM such as multiple rapid swallow (MRS) where lack of augmentation of smooth muscle contraction post multiple rapid swallows has been demonstrated to be associated with late postoperative dysphagia following ARS.¹⁷

Nevertheless, HRM was poorly associated with the gastroesophageal flap valve (GEFV) of the OGJ barrier.¹⁸ In a clinical study on 83 patients with typical GERD, GEFV grades III and IV were more common in patients with oesophagitis; however, there was no significant difference between the values of HRiM metrics and patients with different GEFV grades (p<0.05).¹⁸ Therefore, the determination of OGJ integrity may not be solely reflected by traditional HRM metrics.

Functional lumen imaging probing (FLIP), an emerging diagnostic test in the field of oesophageal motility disorders, can detect oesophageal contractility absent in manometry (as observed in achalasia patients) and phenotype oesophageal stiffness, providing information to direct treatment and OGJ reconstruction.¹⁹

In addition, FLIP can help identify the OGJ-distensibility index (OGJ-DI) which in theory has a positive correlation with the likelihood of GERD. However, in practice, there is a large overlap between normal and GERD patients, limiting clinical applicability in predictions of GERD. Nonetheless, the OGJ-DI has the potential to predict treatment outcomes, as shown in a study where patients with good treatment outcomes had higher OGJ-DI readings.^20,21

Summary

Surgery is an effective route to manage GERD through reconstruction of the OGJ barrier, while endoscopic treatment offers an attractive approach with numerous advances improving outcomes. Motility studies have value beyond the diagnosis of GERD and can help in mechanistic phenotyping to drive the selection of treatment modality.

Dynamic changes of serum cytokines and their prognostic values in patients with hepatocellular carcinoma treated by thermal ablation
(Summary of research project)

Professor Kelvin Kwok-Chai Ng Department of Surgery The Chinese University of Hong Kong Medical Centre Hong Kong

Introduction

Liver cancer is the sixth most common malignancy in the world. It is the fourth leading cause of cancer-related deaths and had a global incidence of more than 740,000 new cases in 2019.¹ Hepatocellular carcinoma (HCC) accounts for 75% – 85% of primary liver cancer. Hepatectomy, radiofrequency ablation (RFA) and orthotopic liver transplantation are the curative treatment options for early-stage HCC.² Nonetheless, hepatectomy is only indicated in patients with early-stage HCC and preserved liver function, whereas application of liver transplantation is limited globally because of the organ shortage. Thermal ablation by either RFA or microwave coagulation (MWA) is therefore the most effective treatment in patients with HCC and borderline liver function due to cirrhotic liver, especially in the Asia-Pacific region where hepatitis B infection is prevalent.

Thermal ablation kills tumors via direct thermal denaturation and coagulation of tumor proteins.³ The retention of devitalized and damaged tumor cells within the body stimulates complex and robust inflammatory responses. It has been reported that thermal ablation creates a tumor antigen source for the generation of antitumor immunity and enhances host immune responses.⁴ In the process of complete ablation by thermal energy, the exposed tumor cells in the sub-lethal ablation zone may be affected positively or negatively by the concomitant inflammatory responses including increased levels of cytokines and growth factors. Inflammation following ablation may have negative effects via the production of growth factors and cytokines by macrophages and lymphocytes, which could stimulate tumor cell growth within the sub-lethal zone.⁵ Thus, the role of systemic inflammatory reaction by the release of cytokines on the subsequent tumor recurrence after thermal ablation is yet to be defined.

Human and animal studies using immunohistochemical staining of tissues, enzyme-linked immunosorbent assay of tissue lysates, and enzyme-linked immunosorbent assay of plasma have shown a significant increase in specific inflammatory and immunomodulatory mediators after thermal ablation.^6-8 Although pre-clinical studies on inflammatory responses of thermal ablation for liver tumor are substantial, the clinical correlation of this micro-environmental reaction is limited in the literature. A study on the results of 209 patients receiving RFA as the sole treatment for HCC showed that the overall tumor recurrence rate was up to 80% with median follow-up of 26 months.⁹ Hence, the identification of high-risk patients who might develop early recurrence using surrogate biomarkers (e.g. cytokines) carries an important clinical implication on the use of adjuvant treatment (e.g. immune check-point inhibitor). It is hypothesized that thermal ablation for HCC may induce an inflammatory immune response with resulting changes in cytokines levels. This micro-environmental reaction might positively or negatively regulate the residual tumor cells, leading to subsequent tumor recurrences. The present single-centre prospective study aims to analyse the dynamic changes in serum cytokines levels in patients with HCC treated by thermal ablation. The result is used to correlate the clinical outcome of patients i.e. early tumor recurrence within 2 years. Combined with other clinicopathological parameters, a prognostic model is derived to identify high-risk patients for early tumor recurrence after thermal ablation for HCC.

Materials and methods

Study design and patient selection
This is a single-centre prospective study at a high-volume tertiary referral liver centre in Hong Kong (Department of Surgery, The Chinese University of Hong Kong). The Institutional Review Broad of The Chinese University of Hong Kong / Hospital Authority New Territories East Cluster and Hospital Authority West Kowloon Cluster reviewed and approved this study. From June 2020 to December 2021, patients with HCC receiving thermal ablation as the first treatment were recruited. The inclusion criteria were: (1) HCC of maximal diameter ≤ 5cm (2) Multiple tumor nodules ≤ 5 (3) Liver (serum total bilirubin level < 50 µmol/L and serum albumin > 30 g/L) (4) Renal function (serum creatinine level <120 µmol/L) (5) Child-Pugh class A or B10. The exclusion criteria were: (1) Evidence of extrahepatic metastasis (2) Patients receiving combined hepatectomy and other local ablative therapy (3) Patients with decompensated liver function that preclude local ablative treatment. (4) Patients with previous treatment for HCC (TACE, radiofrequency ablation, high intensity focus ultrasound or systemic chemotherapy) (5) Patients with inadequate serum sample for cytokine analysis (6) Patients who failed to sign the consent form.

Pre-treatment investigation and assessments:
The diagnosis of HCC was made in accordance with the diagnostic criteria for HCC used by the European Association for the Study of the Liver.¹¹ HCC was diagnosed when the radiologic imaging techniques (spiral contrasted CT scan or contrasted MRI) showed typical features of HCC (contrast enhancement in the arterial phase and rapid wash-out of contrast in the venous or delayed phase) and/or the serum alpha-fetoprotein (AFP) level was elevated (> 200 ηg/mL). All patients had the following pre-treatment investigations and assessments: (1) Blood tests: complete blood count, liver and renal function tests, coagulation profile, serum AFP level, hepatitis B and C serology (2) Radiological imaging study: CT scan of thorax and abdomen with contrast to exclude lung metastases or intraabdominal metastases.

Treatment Procedures:
Thermal ablation by either RFA or MWA was performed under standard protocols.^9,12 The electrode is effective in producing complete tumor necrosis with ablation of a margin of non-tumorous tissue of 1cm. The procedure will be performed percutaneously by an experienced interventional radiologist under local anaesthesia with intravenous sedation if a tumor is accessible by the percutaneous route. For tumors located at the liver dome or if there was risk of thermal injury to adjacent structures like the diaphragm and bowel, thermal ablation will be performed through a laparoscopic or open approach. Subcapsular tumors are approached by either a laparoscopic or open approach.¹³ A spiral contrasted CT scan will be performed 4 – 6 weeks after the procedure to assess the completeness of ablation. In case of incomplete ablation, repeat ablation will be attempted to achieve complete tumor ablation.

Outcome measures
The primary outcome measure is early intrahepatic tumor recurrence within two years after treatment (2-year disease-free survival). Secondary outcome measures are treatment-related complications, mortality, tumor recurrence pattern (intrahepatic recurrence and extrahepatic metastasis), and overall survival.

The following procedure-related parameters were documented. These include the type of thermal ablation (RFA or MWA), ablation time, intraoperative blood loss, blood transfusion requirement, and the treatment approach (percutaneous, laparoscopic or open). Short-term outcomes were documented. These included (1) postoperative morbidities according to the Clavien-Dindo classification¹⁴ (Documented morbidities included wound complication, pulmonary complication, hemoperitoneum, biliary complications, liver failure, and renal failure.), (2) hospital mortality (any death within the same admission for surgery), and (3) intensive care unit stay and total hospital stay.

No adjuvant treatment will be given after the procedure. All recruited patients will be followed up regularly. Liver function (complete blood picture, liver biochemistry and coagulation profile) and serum AFP level were taken every 3 months during follow-up. Spiral contrast-enhanced CT scan of thorax and abdomen will be performed every 6 months after treatment, respectively, to look for tumor recurrence. Long-term outcomes including tumor recurrence (intrahepatic and/or extrahepatic recurrence), overall and disease-free survival were documented.

Analysis of cytokines level
Serum samples were collected from each recruited patient for the analysis of cytokines levels before the procedure and 1 week after the procedure. Using a commercially available kit (MILLIPLEX MAPTM Human Cytokine/Chemokine Magnetic Bead Panel – Immunology Multiplex Assay), the levels of 39 cytokines (EGF, FGF-2, EOTAXIN, TGF-a, G-CSF, FLT-3L, GM-CSF, FRACTALKINE, IFNa2, IFNg, GRO, IFN-α2, IFN-γ, IL-1α, IL-1β, IL-1ra, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-8, IL-9, IL-10, IL-12 (p40), IL-12 (p70), IL-13, IL-15, IL-17A, IP-10, MCP-1, MCP-3, MIP-1α, MIP-1β, sCD40L, TNF-α, TNF-β, VEGF) were analyzed. This is an overnight incubation assay.

Statistical analysis
All data will be prospectively collected by a research assistant and computerized in a database. Statistical analysis will be performed by the Chi-square test or Fisher’s exact test, where appropriate, to compare discrete variables and the Mann-Whitney U test to compare continuous variables. Cumulative survival will be computed by the Kaplan-Meier method and compared by the Log-rank test. Univariate and multivariate analyses using the Cox proportional hazards regression model were performed to identify significant prognostic factors affecting the disease-free survival rate. Potential prognostic factors included patient demographics (age, sex, and comorbidities), liver function (serum bilirubin, serum albumin, and Child-Pugh grading), operative parameters (blood transfusion, and surgical approaches), pathological features (size of largest tumor, and tumor number), and the dynamic changes of selected serum cytokines. The median value was chosen to be the cut-off point of continuous variables in the Cox proportional hazards regression model.

Results

From June 2020 to December 2021, 60 patients with HCC undergone thermal ablation by either RFA or MWA were recruited. The patient demographic and tumor characteristics are described in Table 1. The median age was 69. There were more male patients than female patients (83% vs. 17%). The majority of patients were HBV carriers (70%) and had Child’s A liver function (97%). The median MELD score was low (score = 8). Tumor size ranged from 0.8cm to 5cm. The majority of patients had a single tumor (83%). Twenty-nine patients (49%) and 31 patients (51%) underwent RFA and MWA, respectively. The ablation treatment was performed through percutaneous (65%), laparoscopic (13%) and open (22%) approaches.

Table 1. Demographic characteristics

Continuous variables are expressed as median with range. Categorical variables are expressed as number of patients (percentage). MELD, model for end-stage liver disease; AFP, alpha fetoprotein; ALT, alanine aminotransferase; INR, international normalized ratio; RFA, radiofrequency ablation; MWA, microwave coagulation

The short-term outcome is described in Table 2. The procedure-related blood loss during open or laparoscopic ablation was minimal (median = 30 mL) and no patient required a blood transfusion. The postoperative complication rate was 11.7%. There was one patient who developed severe procedure-related complications. The patient had symptomatic ascites, which required image-guided abdominal drainage. There was no hospital mortality. The median hospital stay was 3 days.

Table 2. Short-term outcome measures

Continuous variables are expressed as median with range. Categorical variables are expressed as number of patients (percentage). ¶ only apply to laparoscopic and open approaches

Of 39 cytokines, 7 cytokines demonstrated significant dynamic changes in serum level before and 1 week after the procedure. They were interleukin (IL) -1, IL-6, IL-10, epidermal growth factor (EGF), eotaxin, transforming growth factor (TGF)-α (Table 3) Following the ablation process, IL-1, IL-6, IL-10 and EGF were up-regulated, whereas eotaxin and TGF-α were down-regulated.

Table 3. Dynamic changes in cytokines levels before and after the ablation procedure

All cytokines are measured in IU/ml and are expressed as median with range

With a median follow-up period of 18 months, 20 patients (33.3%) developed tumor recurrence (local recurrence n = 12, intrahepatic recurrence n = 19, and extrahepatic recurrence n = 1). The recurrence pattern is described in Table 4. Seven patients (11.7%) received hepatic resection, and 10 patients (16.7%) had transarterial chemoembolization as salvage treatment for recurrent tumors. Among 7 cytokines with significant dynamic changes following ablation, up-regulation of IL-6 and IL-10 levels were positively correlated with the overall tumor recurrence rate. (Table 5)

Table 4. Tumor recurrence pattern and treatment

HR, hepatic resection; RFA, radiofrequency ablation; TACE, transarterial chemoembolization Categorical variables are expressed as number of patients (percentage).

Table 5. Correlation between dynamic changes of cytokines and tumor recurrence after treatment

All cytokines are measured in IU/ml and are expressed as median with range.

The 1-year and 2-year overall survival rates were 96.1% and 74.8%, respectively. (Figure 1) The 1-year and 2-year disease-free survival (DFS) were 73.2% and 49.7%, respectively. (Figure 2) Table 6 describes the univariate and multivariate analyses of potential prognostic factors for DFS. The pre-treatment high serum bilirubin level (≥ 25 μmol/L) (hazard ratio [HR]: 0.965, 95% CI: 0.349 – 0.997, P = 0.032) and the upregulation of post-treatment serum IL-10 (≥ 22 IU/ml) (HR: 0.723, 95% CI: 0.383 – 0.867, P = 0.008) were the significant poor independent prognostic factors influencing DFS. (Table 6) Patients were then grouped into 3 groups (Group 1: no risk factor (n = 11); Group 2: one risk factor (n = 36); Group 3: two risk factors (n = 13)). The 1-year and 2-year DFS rates of Groups 1, 2 and 3 were 90.9% and 70.7%, 74.8% and 46.8%, and 52.7% and 39.6%, respectively. There was a statistically significant difference in DFS between Groups 1 and 3 (P = 0.003). (Figure 3)

Figure 1. Overall survival of 60 patients with thermal ablation for hepatocellular carcinoma

Figure 2. Disease-free survival of 60 patients with thermal ablation for hepatocellular carcinoma

Figure 3. Disease-free survival of Groups 1 (no risk factor), 2 (one risk factor) and 3 (two risk factors)

Table 6. Univariate and multivariate analysis of prognostic factors for disease-free survival

Conclusions

There are significant dynamic changes in serum cytokines (IL-1, IL-6, IL-10, EGF, Eotaxin and TGF-α) after thermal ablation of hepatocellular carcinoma. High pre-treatment serum bilirubin and upregulation of serum IL-10 level after treatment are independent poor prognostic factors for tumor recurrence after thermal ablation.

Funding:
This study is supported by the research grant from the Hong Kong Society of Gastroenterology.