SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Background  Chronic oesophageal inflammation and related oxidative stress are important in the pathogenesis of erosive oesophagitis (EO) and its malignant progression.

Aim  To study the effect of proton pump inhibitors (PPIs) on oesophageal cellular immune response and oxidative damage in EO patients.

Methods  Forty gastro-oesophageal reflux disease (GERD) patients [non-erosive reflux disease (NERD): 15, EO: 25] were included, after 7 days off antisuppressive drugs. EO patients were randomized to 20-mg rabeprazole once daily for either 4 or 8 weeks with baseline and follow-up endoscopy with distal oesophageal biopsies. T lymphocytes, macrophages and mast cells were quantified by immunohistochemistry. DNA adducts were measured by analysis of 8-oxo-deoxyguanosine levels.

Results  Erosive oesophagitis patients had more T lymphocytes and CD8+ T lymphocytes in squamous epithelium than NERD patients (= 0.001, = 0.002, respectively). Levels of DNA adducts between both groups were, however, not different (= 0.99). Four- and eight-week rabeprazole treatment in EO patients resulted in a significant decrease in number of T lymphocytes and CD8+ T lymphocytes (all P < 0.05). PPIs did not, however, affect levels of DNA adducts.

Conclusions  Short-term PPI therapy in EO patients reduces the oesophageal cellular immune response, but does not change oxidative damage. PPI therapy may therefore not be effective in reducing the risk of oesophageal cancer in GERD patients.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Gastro-oesophageal reflux disease (GERD) is characterized by symptomatic reflux of gastroduodenal contents into the oesophagus, which can result in mild-to-severe injury of the oesophageal mucosa.1 Patients with GERD can be categorized into those with non-erosive reflux disease (NERD), those with erosive oesophagitis (EO) and those with complicated GERD, such as peptic stricture and Barrett’s oesophagus (BO).2 Chronic GERD and BO are well-established risk factors for the development of oesophageal adenocarcinoma (OAC),3–5 which usually portrays a poor prognosis.6 The incidence of OAC has increased dramatically since the 1970s, at a rate faster than any other type of cancer.7 A parallel increase in incidence of detected BO in the general population predicts an even further increase in the incidence of OAC.8 As a consequence, there is growing interest in potential chemopreventive strategies that can effectively reduce the risk of developing OAC.

Profound acid suppression with proton pump inhibitors (PPIs) has been suggested to be important in the prevention of OAC. Whether PPI therapy in GERD patients can prevent the development of intestinal metaplasia of the distal oesophagus as a first step in the cascade that can lead to OAC is still unclear. Some cohort studies of patients with GERD demonstrated that long-term PPI therapy cannot prevent the development of BO;9, 10 other studies indicate that acid suppression does slow down and possibly prevents progression of this process.11–13

Insight into mechanisms leading to oesophageal injury is important to elucidate the suggested chemopreventive action of PPI therapy. Although these mechanisms are still poorly understood, an important role in the progression of GERD has been attributed to the oesophageal immunoregulatory environment and response to oxidative stress. Several studies suggest that differences exist in the cytokine profile within the oesophageal mucosa of patients with BO and those with EO,14–16 i.e. BO is characterized by Th-2 anti-inflammatory cytokines (IL-10 and IL-4), and EO by an increase in pro-inflammatory cytokines (IL-1β, IL-8 and IFN-γ).15 These observations are substantiated by others demonstrating that the inflammatory response is shifted from a cellular immune response towards a more pronounced humoural immune response when EO progresses to BO.14 This immune response is thought to facilitate the progression towards neoplasia, as a humoural environment promotes angiogenesis and may contribute to the malignant propensity of BO.17, 18

As a consequence of chronic oesophageal inflammation, reactive oxygen species (ROS) are generated, which in the presence of an imbalance between pro-oxidants and anti-oxidants promote oxidative stress.19 Indeed, increased levels of the pro-oxidant myeloperoxidase and decreased levels of glutathione, a potent anti-oxidant, have been observed in patients with EO and BO.20 Furthermore, oxidative stress may result in formation of DNA adducts, which can initiate and promote carcinogenesis. Increased levels of DNA adducts have been observed along the metaplasia–dysplasia–adenocarcinoma sequence in Barrett epithelium.20–22

So far, in spite of the existence of a strong link between chronic inflammation and oxidative stress, the inflammatory profiles and oxidative stress in GERD have not been studied simultaneously. Moreover, little is known about the effect of profound acid suppression on these processes. Therefore, in this study, we characterized the oesophageal cellular-mediated immune response and levels of DNA adducts in GERD patients in parallel, and assessed to what extent these processes are affected by PPI therapy in vivo.

Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Patients and study design

Between November 2005 and September 2007, all consecutive patients with typical GERD symptoms in the Erasmus MC, Ikazia Hospital and Sint Franciscus Gasthuis, Rotterdam, were invited to participate in this study. Typical symptoms of GERD were defined as the presence of heartburn and/or acid regurgitation. The GERD health-related quality of life (GERD-HRQL) questionnaire was used to evaluate the frequency and intensity of symptoms and their impact on patient’s quality of life. Only patients with a score ≥12 were included in this study.23, 24 Patients presenting with predominantly atypical GERD symptoms, such as chronic cough and noncardiac chest pain, were excluded. In addition, patients with a history of upper GI surgery, peptic stricture, duodenal or gastric ulcer, bleeding diathesis or coagulopathy, stroke or ischaemic attack, significant GI bleed within the past 6 months, or the presence of oesophageal varices were excluded.

All patients underwent upper GI endoscopy with biopsies and subsequent wireless 48-h pH monitoring, to characterize them as nonreflux or GERD patients with either NERD or EO. The latter two patient categories were compared with regard to the number of inflammatory cells and levels of local oxidative damage in the distal oesophagus. To assess the effect of PPI therapy on these processes, patients with EO were randomized to rabeprazole treatment (20 mg once daily) either for 4 weeks or for 8 weeks, and underwent a follow-up endoscopy with biopsy specimens at the end of treatment. All participants signed an informed consent form. The study was approved by the Institutional Review Boards of the Erasmus MC, Ikazia Hospital and Sint Franciscus Gasthuis, Rotterdam.

Endoscopy

Baseline upper GI endoscopy was performed after at least a 7-day period without antisuppressant medication. During endoscopy, the squamo-columnar junction (SCJ) was identified and the distance between the incisors and SCJ was measured. The distal part of the oesophagus was evaluated to determine the absence or presence of mucosal injury and a sliding hiatus hernia. The extent of mucosal inflammation was determined using the Los Angeles Classification System.25 Subsequently, two to four biopsy specimens were obtained from the distal oesophagus 1 cm above the SCJ. In EO patients, care was taken to take these biopsies from non-eroded sites. In addition, two biopsies from the antrum and two biopsies from the corpus of the stomach were taken for assessment of Helicobacter pylori gastritis.

In patients from the Erasmus MC, additional biopsies from the distal oesophagus 1 cm above the SCJ were obtained for determination of DNA adducts. Biopsies for histology and immunohistochemistry were fixed in 10% buffered formalin solution. Biopsies for determination of DNA adducts were immediately frozen and stored at −80 °C.

Wireless pH monitoring

After the upper GI endoscopy was completed, a BRAVO pH delivery system (Medtronic, Minneapolis, MN, USA) was introduced orally, and the pH capsule was positioned 6 cm proximal to the SCJ according to standard procedures. Proper functioning of the BRAVO probe was confirmed by a reading of pH >4 immediately after placement.

All studies were intended to be performed for 2 days (off PPI), during which time pH data were received and stored in the receiver. Patients were encouraged to engage in their usual activities and were asked to keep a diary documenting food intake, periods of sleep, and occurrence of symptoms.

Acid exposure time (AET) was defined as pathologic, if the proportion of time with a pH <4 exceeded 4.4%.26 Patients were considered to have NERD when a positive symptom association probability (SAP > 95%) was found on 48-h pH monitoring.27 Those NERD patients with an abnormal AET (% time pH <4 exceeding 4.4%) and a positive SAP were further defined as NERD pH positive, while those without abnormal AET but with a positive SAP were defined as NERD pH negative.

Histology and immunohistochemistry

Biopsy specimens obtained during the baseline and follow-up endoscopy were serially sectioned at 4 μm, mounted on adhesive slides, dried overnight at 37 °C and deparaffinized with xylene. The first of these serially sectioned slides was stained with haematoxylin and eosin, and evaluated by an experienced pathologist (HvD) for presence of microscopic oesophagitis according to established criteria.28, 29 Using these criteria, the following lesions were considered compatible with reflux disease of increasing severity: (1) basal layer hyperplasia, (2) elongation of papillae, (3) dilation of papillary vascular spaces, (4) intraepithelial inflammatory infiltration, (5) mucosal erosion and (6) granulation tissue. For practical purposes, inflammation was graded as chronic (i.e. oesophagitis grade 1, in the presence of criteria 1–3), chronic active (i.e. oesophagitis grade 2, in the presence of criterion 4 with or without criteria 1–3) or eroding ulcerating (i.e. oesophagitis grade 3, in the presence of criterion 5 or 6).30

For immunohistochemistry, antigen retrieval was performed by boiling the deparaffinized samples in either 10 mmol/L monocitric acid buffer (pH 6.0) for 10 min or in 10/1 mmol/L Tris/EDTA (pH 9.0) for 10 min. Then, samples were slowly cooled down to room temperature (RT). Prior to staining, endogenous peroxidase activity was blocked by either incubating the slides in a 0.3% solution of H2O2 in methanol or in a 40/150 mmol/L citrate–phosphate buffer for 20 min at RT. The samples were blocked with 10% rabbit non-immune serum and 10% normal human plasma in PBS for 30 min at RT. Sections were stained using antibodies specific for cells representing the cellular-mediated immune response: T lymphocytes (CD3; Dako, Glostrup, Denmark), CD8+ T cells (CD8; Dako), macrophages (CD68; Dako) and mast cells (antitryptase; Dako). For T lymphocytes and CD8+ T cells, biotin-labelled rabbit–anti-mouse antibody (Dako) was used as a second antibody, followed by the addition of a streptavidin–horseradish peroxidase complex (Dako) using 3-amino-9-ethylcarbazole as substrate. For visualization of macrophages and mast cells, biotin-labelled rabbit–anti mouse antibody was used as a second antibody, followed by the addition of a streptavidin–alkaline phosphatase complex (Dako) using new-fuchsin as substrate. Reactive lymph nodes were used as positive controls.

Following the recommendations of Wang et al.,31 the number of immune cells was quantified in the most densely populated field in the squamous epithelium of biopsy specimens, using HPF examination (magnification 400×). Counting was performed by two independent investigators (PdJ, KvZ), blinded to patient characteristics, treatment regimen and endoscopic findings. In total, six HPFs were counted and means and standard deviations were subsequently calculated.

Detection of oxidative DNA damage in oesophageal mucosa

Oxidative DNA damage was quantified by measurement of 8-hydroxy-2-deoxyguanosine (8-OHdG), a known promutagenic DNA adduct. Biopsy specimens were homogenized with a Potter homogenizer (1000 rpm) in 1% SDS/1 mmol/L EDTA/10 mmol/L TEMPO and solutions thus obtained were incubated overnight at 37 °C with 0.5-mg proteinase K. After the incubation period, DNA was isolated in the presence of 8-hydroxyquinoline to prevent artificial 8-oxo-dG, by means of repetitive extraction with phenol/chloroform/isoamyl alcohol (25:24:1) and chloroform/isoamyl alcohol (24:1). Subsequently, DNA was precipitated with two volumes of 100% cold ethanol and 1/30 volume of 3 mol/L sodium acetate (pH 5.3). Precipitated DNA was rinsed with 70% ethanol, dissolved in 2 mmol/L Tris (pH 7.4) and stored at −20 °C until analysis. High-performance liquid chromatography-electrochemical detection (HPLC-ECD) analysis of 8-oxo-dG was performed as described previously.32 Briefly, after extraction, DNA was digested into deoxyribonucleosides by treatment with nuclease P1 (0.02 U/mL) and alkaline phosphatase (0.014 U/mL). The digest was then injected into a Gynkotek 480 isocratic pump (Gynkotek, Bremen, Germany) coupled with a Midas injector (Spark Holland, Hendrik Ido Ambacht, the Netherlands) and connected to a Allsphere ODS-2 5i column (250 × 4.6 mm) (Alltech, Applied Science Group, Breda, the Netherlands) and an electrochemical detector (Antec, Leiden, the Netherlands). The mobile phase consisted of 10% aqueous methanol containing 94 mmol/L KH2PO4, 13 mmol/L K2HPO4, 26 mmol/L NaCl and 0.5 mmol/L EDTA. Elution was performed at a flow rate of 1.0 mL/min with a lower detection limit of 40 fmol absolute for 8-OHdG, or 1.5 residues/106 2′-deoxyguanosine (dG). dG was simultaneously monitored at 260 nm. The 8-oxo-dG concentration was expressed as the ratio of 8-oxo-dG/106 dG.

Statistical analysis

Continuous data were compared using the Wilcoxon rank-sum test and the Kruskal–Wallis test for independent samples, and the Wilcoxon signed-rank test for paired samples. Categorical data were analysed using the chi-squared test and Fisher exact test. Correlations between variables were evaluated by Pearson correlation analysis. Two-sided statistical significance was set at P < 0.05. Statistical analyses were conducted using spss software (10.1; SPSS Inc., Chicago, IL, USA).

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Patients

Fifty-eight of 89 eligible patients with typical GERD symptoms were included in this study (66%), the remaining were not included for reasons of unwillingness to undergo pH monitoring (= 7), unwillingness to undergo a follow-up endoscopy (= 18) or other reasons (= 6). Of those included, nine patients (16%) were not eligible for final analysis either because their histological specimens were inadequate (n = 4) such as lacking squamous epithelium in biopsies taken across the SCJ, or because the pH studies were incomplete (n = 5) because of early dislodgement of the pH capsule. Nine patients (16%) were withdrawn from the study because of non-adherence to the study protocol.

Of the remaining 40 GERD patients, 25 had EO (grade A: 13; B: 8; C: 4) confirmed by endoscopy, and 15 patients presented with NERD. Of those NERD patients, five presented AET within the normal range and were defined NERD pH-negative; 10 patients presented a pathological AET and were defined as NERD pH-positive. Baseline characteristics of the GERD patient groups are shown in Table 1. No differences were detected in age and gender. Patients with EO, however, had a significantly higher BMI than NERD pH− patients. Although GERD-HRQL scores were similar among the three groups, EO and NERD pH+ patients had significantly higher levels of intra-oesophageal acid exposure than NERD pH− patients (Table 2).

Table 1.   Demographic characteristics of patients
 NERD pH− (n = 5)NERD pH+ (n = 10)EO (n = 25) P-value*
  1. NERD, non-erosive reflux disease; EO, erosive oesophagitis; GERD-HRQL, gastro-oesophageal reflux disease health-related quality of life.

  2. P-value from Kruskal–Wallis tests/χ2-tests.

Age (mean ± s.d.)34 (13)49 (13)46 (15)0.187
Gender (% female)4 (80)7 (70)13 (52)0.384
Smoking (%)2 (40)3 (30)10 (40)0.852
Alcohol consumption (%)3 (60)6 (60)20 (80)0.390
BMI (mean ± s.d.)24 (1)25 (3)28 (5)0.042
GERD-HRQL score (mean ± s.d.)20 (6)21 (8)22 (8)0.678
Table 2.   Endoscopic and histological characteristics of patients
 NERD pH− (n = 5)NERD pH+ (n = 10)EO (n = 25) P-value*
  1. NERD, non-erosive reflux disease; EO, erosive oesophagitis.

  2. P-value from Kruskal–Wallis tests/χ2-tests.

Mean % time acid exposure (mean ± s.d.)1.7 (2)8.7 (5)9.5 (6)0.006
Endoscopy (%)
 Normal5 (100)10 (100)
 Grade A 13 (52) 
 Grade B 9 (36) 
 Grade C 3 (12) 
 Grade D 0 (0) 
Histology (%)
 Normal0 (0)0 (0)1 (4)0.947
 Grade 14 (80)8 (80)20 (80) 
 Grade 21 (20)1 (10)3 (12) 
 Grade 30 (0)1 (10)1 (4) 
H. pylori infection (%)1 (20)1 (10)4 (16)0.855

Effect of PPI therapy on the cellular immune response

At histology, 20 of 25 (80%) EO patients had oesophagitis grade 1, three (12%) oesophagitis grade 2 and one (4%) patient grade 3. Of all NERD patients, grade 1 oesophagitis was found by histology in 12 (80%), whereas grade 2 oesophagitis was present in two (13%) patients (Table 2). No statistically significant difference was found in prevalence of H. pylori gastritis between NERD and EO patients (13% vs. 16%, P = 0.819).

Results of immunohistochemically stained sections of oesophageal biopsies from NERD and EO patients are shown in Figure 1. The inflammatory infiltrate in oesophageal squamous epithelium of GERD patient groups mainly consisted of T lymphocytes. Patients with EO had significantly higher numbers of T lymphocytes in the squamous epithelium than NERD pH− patients (29 ± 2 vs. 10 ± 1, P < 0.001), and NERD pH+ patients (29 ± 2 vs. 22 ± 4, P = 0.02). The mean number of T lymphocytes was also higher in patients with NERD pH+ than in NERD pH− patients, although not statistically significant (P = 0.08). CD8+ T cells were the predominant subpopulation of T lymphocytes in all three GERD groups, with an increasing CD8+/CD3+ ratio from NERD pH− to EO (NERD pH−: 0.67 ± 0.06; NERD pH+: 0.68 ± 0.02; EO: 0.75 ± 0.02).

image

Figure 1.  Graphic representation of mean number T lymphocytes (CD3), CD8+ T lymphocytes, mast cells and macrophages (CD68) in distal oesophageal biopsy specimens from GERD patients (*P < 0.05, **P < 0.01).

Download figure to PowerPoint

The number of T lymphocytes correlated significantly with the endoscopic grade of oesophagitis (r = 0.42, P = 0.002), but did not correlate with the grade of microscopic oesophagitis (r = 0.15, P = 0.37) nor with levels of intra-oesophageal acid exposure (r = 0.18, P = 0.28). Patients with EO, NERD pH+ and pH− patients did not differ with regard to the number of mast cells, nor were any differences detected in the mean number of macrophages in the squamous epithelium.

In 10 patients with EO (grade A: 5; B: 3; C: 2), randomized to a 4-week treatment with 20 mg rabeprazole once daily, a significant reduction in the cellular-mediated immune response in the distal oesophagus was seen (Figure 2a). PPI therapy led to a significant decrease in the number of T lymphocytes and of CD8+ T lymphocytes in the squamous epithelium (28 ± 2 to 11 ± 1, P = 0.005, and 22 ± 2 to 7 ± 1, P = 0.005, respectively). In addition, the number of mast cells and macrophages in the squamous epithelium also decreased upon rabeprazole treatment (13 ± 2 vs. 9 ± 1, P = 0.074; and 5 ± 0.4 vs. 3 ± 1, P = 0.005, respectively). Mean post-treatment levels of these cell types were similar to those determined in NERD pH− patients at baseline. A similar response was seen in 10 EO patients (grade A: 6; B: 2; C: 2) randomized to an 8-week treatment regimen (Figure 2b) (all P < 0.05). Although the mean decrease in the number of immune cells was larger in the group treated for 8 weeks, the difference between the 4- and 8-week treatments was statistically not significant. In addition, the presence of H. pylori gastritis in EO patients did not influence the change in oesophageal immunological response during PPI therapy (data not shown).

image

Figure 2.  Effect of PPI therapy on the oesophageal immune response. Mean number of T lymphocytes (CD3), CD8+ T lymphocytes, mast cells and macrophages (CD68) in the distal oesophagus in (a) patients with erosive oesophagitis (EO), before and 4 weeks after treatment with rabeprazole (20 mg once daily), and in (b) patients with EO, before and 8 weeks after treatment with rabeprazole (20 mg once daily); **P < 0.01.

Download figure to PowerPoint

Effect of PPI therapy on oxidative DNA damage

In a subset of GERD patients (11 NERD, 17 EO), we determined the level of DNA adducts in the distal oesophagus. The stepwise increase in the number of immune cells within the GERD spectrum, detected at immunohistochemistry, did not, however, reflect an increased formation of DNA adducts in NERD pH+ and EO patients compared to NERD pH− patients (NERD pH−: 15 ± 4; NERD pH+: 14 ± 3; EO: 14 ± 2; P = 0.97).

In seven patients with EO, randomized to a 4-week treatment with 20-mg rabeprazole once daily, no differences in the amount of DNA adducts between pre-treatment biopsies and post-treatment biopsies were detected (13 ± 3 vs. 12 ± 2, P = 0.94) (Figure 3a). Rabeprazole treatment for a period of 8 weeks also did not result in decreased formation of DNA adducts in EO patients (n = 8) (17 ± 4 vs. 15 ± 2, P = 0.84; Figure 3b).

image

Figure 3.  Effect of PPI therapy on oesophageal oxidative DNA damage. 8-oxo-dG levels (as a measure of oxidative DNA damage) in (a) the distal oesophagus in patients with erosive oesophagitis (EO), before and 4 weeks after treatment with rabeprazole (20 mg once daily), and in (b) the distal oesophagus in patients with EO, before and 8 weeks after treatment with rabeprazole (20 mg once daily).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

To our knowledge, this is the first study in GERD patients where the oesophageal inflammatory infiltrate was characterized by and correlated to levels of local oxidative damage before and after PPI therapy. We found an increased cellular-mediated inflammatory infiltrate in the oesophageal epithelium of patients with EO and NERD pH+ compared to NERD pH− patients. This infiltrate mainly consisted of CD8+ T lymphocytes, whereas macrophages and mast cells represented only a minor proportion of the inflammatory infiltrate. Despite the stepwise increase in the number of immune cells demonstrated within the GERD spectrum, similar amounts of DNA adducts were found in EO and NERD patients. Acid suppression with PPI therapy resulted in a significant drop in the number of inflammatory cells within the distal oesophageal mucosa, especially oesophageal T lymphocytes and their CD8+ T-cell subpopulation; however, levels of DNA adducts remained unaffected.

The role of profound acid suppression with PPIs in the prevention of BO and OAC in patients with GERD is still controversial. Evidence exists that PPI therapy in BO reduces oesophageal acid exposure,33 decreases mucosal cell proliferation and increases differentiation13 and possibly reduces the length of Barrett’s segment and dysplasia incidence.11, 12, 34 This seems to be in contrast with cohort studies on long-term PPI therapy, which indicate that this therapy cannot prevent the development of BO.9, 10 Furthermore, another cohort study including 417 patients with BO did not find any effect of 4 years omeprazole treatment on the incidence of OAC.35 To elucidate the possible chemopreventive properties of PPIs, understanding of the mechanisms important in the pathophysiology of EO and its progression towards BO and OAC is necessary. The distinct inflammatory environment within the GERD spectrum has been proposed to be an important factor in the pathophysiology of oesophageal mucosal injury and may be critical in carcinogenesis.14, 15 We previously showed that when EO progresses to BO, the inflammatory response is shifted from a cellular-mediated immune response towards a more pronounced humoural immune response.14 Inhibition of the mechanisms responsible for this shift has been suggested to play an important role in the prevention of carcinogenesis in GERD.

In this study, PPI therapy resulted in a substantial decrease in the cellular-mediated immune response in the distal oesophagus, both in EO patients treated for 4 weeks and in those treated for 8 weeks (Figure 2). Oesophageal T lymphocytes and their CD8+ T-cell subpopulation were predominantly affected by acid suppression. Mean post-treatment levels of these immune cells were similar to those determined in NERD pH− patients at baseline.

The finding of increased numbers of oesophageal T lymphocytes in pre-treatment biopsies is in agreement with other studies.31, 36–40 Similar to the findings of Geboes et al.,36 a majority of oesophageal T lymphocytes possessed the CD8 suppressor T-cell phenotype, probably including a subset of cells with cytotoxic potential. The significance of this lymphocytic infiltration in GERD is yet unclear, although several hypotheses have been proposed. One hypothesis that could account for the increased number of T lymphocytes seen in oesophagitis is that gastric acid and bile acids may modulate epithelial surface antigens, which are subsequently recognized by the immune system as foreign. Alternatively, the presence of T lymphocytes in GERD may be related to the general inflammatory reaction, resulting in the release of cytokines known to induce nonspecific activation of cytotoxic T lymphocytes.41 However, others have shown that T lymphocytes follow infiltration by acute inflammatory cells, particularly at the site of metaplastic foci.42 Furthermore, T-cell infiltrates are predominantly seen in persistent areas of BO following endoscopic ablation therapy, suggesting that lymphocytes may be important in the maintenance of the metaplastic tissue.43, 44 It is thus likely that T lymphocytes play an important role in the pathogenesis of BO. Therefore, the sharp drop in particularly oesophageal T lymphocytes in EO patients observed after PPI therapy suggests that profound acid suppression in GERD may be capable of inhibiting the immunological shift along the EO–BO sequence and may be important in the prevention of carcinogenesis in GERD.

One of the consequences of chronic oesophageal inflammation is the induction of oxidative stress by production of ROS.45 Irritation of epithelial cells by gastroduodenal contents followed by excess production of ROS by inflammatory cells drawn to this area has been shown to contribute to EO and to the development of BO.19 Furthermore, excess production of ROS has been shown to result in formation of DNA adducts, which play a major role in the induction of spontaneous mutations, a prerequisite for carcinogenesis in GERD.46 As effective acid control decreased the amount of reflux, DNA damage as a result of chronic oesophageal inflammation may be reduced. In this study, however, we showed that PPI therapy did not affect the formation of DNA adducts in EO, either in those patients treated for 4 weeks or in those treated for 8 weeks (Figure 3). In addition, despite the stepwise increase in the number of immune cells from NERD to EO patients, no differences were detected in the level of DNA adducts between both patient groups at baseline. The latter finding is in agreement with a study, in which mean DNA adduct levels were similar between reflux patients with and without endoscopic oesophagitis, but significantly higher than those in healthy controls without symptoms.20 Levels of DNA adducts determined in our GERD patient groups were equivalent to those determined in another recent study in BO patients with dysplasia and patients with OAC. Moreover, these levels were again significantly higher than those determined in controls.47 This suggests that oxidative DNA damage plays an important role in the pathogenesis of GERD, in the development of BO and its progression towards OAC and is in line with the reported inverse link between intake of anti-oxidants and risk of OAC development.48 In addition, as the amount of DNA base changes related to oxidative stress depends on DNA repair mechanisms,49 it may well be that key genes involved in DNA repair are malfunctioning within the GERD spectrum.

Although we cannot rule out completely that we failed to detect any effect of PPI therapy on oesophageal mucosa DNA-damage either because the PPI dosage used was too low or the treatment follow-up period (8 weeks) was too short, this is unlikely as the cellular-mediated immune response in EO patients was already substantially decreased after 4 weeks of PPI treatment. This decrease is probably accompanied by a reduction in local ROS production, but apparently without much effect on DNA damage. Furthermore, a recent study showed that fundoplication, which can be very effective in normalizing reflux, was not capable of decreasing DNA damage in the distal oesophagus of GERD patients, after a follow-up for 6 months.50 This suggests that in GERD, a subpopulation of transformed cells with defective mechanisms against DNA-damage is present, which could be responsible for constant presence of DNA adducts despite acid suppression. This supports the finding that even long-lasting acid suppression therapy does not alter malignant transformation of BO, and may partially explain why OAC is found even after successful medical and surgical therapies for GERD.51

Some limitations of our study warrant consideration. First, the number of patients with NERD included in this study was relatively small. However, statistically significant changes between and within GERD patient groups could be found and results obtained are in agreement with others. Second, we did not include healthy controls without symptoms and without evidence of oesophageal injury. Third, we did not perform pH monitoring after 4 or 8 weeks of PPI treatment and therefore we do not exactly know the amount of oesophageal acid exposure at these time points. However, the significant decrease in the cellular-mediated immune response detected after PPI therapy indicates that acid suppression was effective in our patients, in the absence of additional medical interventions.

Strengths of our study include its prospective randomized study design and systematic data collection. In addition, GERD patients were uniformly classified with both endoscopy and wireless pH monitoring. Moreover, for the first time, the cellular-mediated immune response and amount of oxidative DNA damage were studied simultaneously and the effect of acid suppression on these processes was evaluated prospectively.

In conclusion, we have shown that PPI therapy in GERD patients reduces the oesophageal cellular-mediated immune response, thereby possibly inhibiting the progression of GERD towards BO. However, short-term PPI therapy is insufficient to reverse DNA damage caused by gastro-oesophageal reflux, suggesting that PPI therapy is perhaps not effective in the prevention of carcinogenesis because of persisting DNA damage. The relative contribution of these contrasting effects of PPI therapy to the prevention of OAC warrants further study with long-term follow-up.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We thank Corry Leunis for patient inclusion. Declaration of personal interests: None. Declaration of funding interests: This study was supported in part by an unrestricted grant form Janssen-Cilag B.V., the Netherlands.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References