Article Text

PDF

Original article
Randomised clinical trial comparing sequential and concomitant therapies for Helicobacter pylori eradication in routine clinical practice
  1. Adrian G McNicholl1,2,
  2. Alicia C Marin1,2,
  3. Javier Molina-Infante3,
  4. Manuel Castro2,4,
  5. Jesús Barrio5,
  6. Julio Ducons2,6,
  7. Xavier Calvet2,7,
  8. Cristobal de la Coba8,
  9. Miguel Montoro9,
  10. Felipe Bory10,
  11. Angeles Perez-Aisa11,
  12. Montserrat Forné2,12,
  13. Javier P Gisbert1,2,
  14. On behalf of the participant centres
  1. 1Digestive Services, Hospital Universitario de La Princesa and Instituto de Investigación Sanitaria Princesa (IP), Madrid, Spain
  2. 2Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBERehd), Instituto de Salud Carlos III, Madrid, Spain 
  3. 3Digestive Services, Hospital San Pedro de Alcántara, Cáceres, Cáceres, Spain
  4. 4Digestive Services, Hospital Ntra. Sra. Virgen de Valme, Sevilla, Spain
  5. 5Digestive Services, Hospital Universitario Río Hortega, Valladolid, Valladolid, Spain
  6. 6Digestive Services, Hospital Clínico Universitario ‘Lozano Blesa’, Zaragoza, Spain
  7. 7Digestive Services, Hospital de Sabadell, Sabadell, Spain
  8. 8Digestive Services, Hospital de Cabueñes, Gijón, Spain
  9. 9Digestive Services, Hospital de San Jorge, Huesca, Spain
  10. 10Digestive Services, Hospital del Mar, Barcelona, Spain
  11. 11Digestive Unit, Agencia Sanitaria Costa del Sol, Marbella, Málaga, Spain
  12. 12Digestive Services, Hospital Mutua de Terrassa, Terrassa, Spain
  1. Correspondence to Dr Javier P Gisbert, Servicio de Aparato Digestivo, Hospital Universitario de la Princesa, c/Diego de León 62, Madrid 28006, Spain; javier.p.gisbert{at}gmail.com

Abstract

Objectives No trial has compared non-bismuth quadruple ‘sequential’ and ‘concomitant’ regimens in settings with increasing clarithromycin rates. The study aims to compare the effectiveness and safety of these therapies for Helicobacter pylori treatment.

Design Prospective randomised clinical trial in 11 Spanish hospitals. Patients naïve to eradication therapy with non-investigated/functional dyspepsia or peptic ulcer disease were included. Randomised (1:1) to sequential (omeprazole (20 mg/12 h) and amoxicillin (1 g/12 h) for 5 days, followed by 5 days of omeprazole (20 mg/12 h), clarithromycin (500 mg/12 h) and metronidazole (500 mg/12 h)), or concomitant treatment (same drugs taken concomitantly for 10 days). Eradication was confirmed with 13C-urea breath test or histology 4 weeks after treatment. Adverse events (AEs) and compliance were evaluated with questionnaires and residual medication count.

Results 338 consecutive patients were randomised. Mean age was 47 years, 60% were women, 22% smokers and 20% had peptic ulcer. Concomitant and sequential eradication rates were, respectively, 87% vs 81% by intention-to-treat (p=0.15) and 91% vs 86% (p=0.131) per protocol. Respective compliances were 83% vs 82%. Treatment-emergent AEs were reported in 59% of patients (no differences found between treatments). AEs were mostly mild (60%), and average length was 6.1 days, causing discontinuation only in 12 patients. Multivariate analysis: ‘concomitant’ treatment showed an OR of 1.5 towards better eradication rate in a borderline significance CI (95% CI 0.9 to 2.8).

Conclusions Concomitant therapy led to a non-statistically significant advantage (5%) over sequential therapy, coming closer to 90% cure rates. Both therapies showed an acceptable safety profile.

ClincialTrials.gov: NCT01273441.

Statistics from Altmetric.com

Significance of this study

What is already known about this subject?

  • Helicobacter pylori eradication is a cost effective measure recommended in most clinical guidelines.

  • Standard triple therapies do not achieve acceptable H pylori eradication rates.

  • Non-bismuth quadruple therapies have demonstrated achieving higher eradication rates.

What are the new findings?

  • Non-bismuth quadruple therapies achieve acceptable eradication rates in regions with increasing clarithromycin resistance.

  • Both sequential and concomitant regimens offer similar and acceptable safety profiles and compliance rates.

  • Concomitant treatment achieved an over 90% per-protocol eradication rate.

How might it impact on clinical practice in the foreseeable future?

  • Results from this study may change the first-line treatment of choice in H pylori eradication in routine clinical practice.

  • Future guidelines and consensus conferences may recommend concomitant treatment of at least 10 days as first-line treatment in regions with moderate and high claritromycin resistance rates.

Introduction

Helicobacter pylori causes the most common chronic bacterial infection affecting over 50% of the world's population (more than 80% in developing countries).1 Although most H pylori-positive individuals will remain asymptomatic through life, its presence causes chronic gastritis in 100% of infected patients and is the major cause of relevant diseases, such as atrophic gastritis, peptic ulcer disease and gastric cancer.2 H pylori eradication prevents peptic ulcer recurrence and its complications.3 WHO's International Agency for Research on Cancer classified H pylori as a group 1 (definite) carcinogen.4 Eradication in patients with peptic ulcer or even functional or non-investigated dyspepsia is a cost effective approach.2

Most Consensus Conferences and Clinical Guidelines recommend the prescription of a triple therapy including a proton pump inhibitor (PPI) and clarithromycin with either amoxicillin or metronidazole, as first-line treatment.2 ,59 However, with the increasing development of H pylori resistance to antibiotics in most countries,10 the eradication rates of H pylori with triple therapies are decreasing to unacceptable levels (≤80%).11 ,12 Due to the low efficacy achieved with these treatments, they have been deemed as unethical comparators in clinical trials.13 ,14 As antimicrobial resistance becomes more prevalent worldwide, treatment failure rates are likely to continue increasing, suggesting that new regimens for H pylori eradication must be sought.

All this has led the medical and scientific community to pursue strategies that will improve treatment efficacy. One of the last therapeutic innovations in the field of H pylori eradication is sequential treatment, introduced in Italy by Zullo et al,15 which consisted in an induction phase of 5 days with amoxicillin and a PPI, immediately followed by 5 days of triple therapy (metronidazole, clarithromycin and PPI). Some authors argue that the beneficial effect of sequential treatment is only due to the use of an additional antibiotic.16 Thus, it has been suggested that all drugs could be given concomitantly, reducing the complexity of the regimen.17

However, sequential and concomitant regimens have only been compared one-to-one in three studies.1820 Two of them were performed in settings from Asia with low clarithromycin resistance, whereas the other compared treatments with different durations. No study comparing both regimens has been conducted in Europe, where clarithromycin resistance rates have increased dramatically over the years.10

Therefore, and given the differences and limitations of previous studies, the aim of the present randomised controlled trial was to compare the effectiveness, safety and compliance of sequential and concomitant therapies for H pylori treatment mimicking routine clinical practice in a large sample of patients.

Methods

Design summary and ethical issues

The study was designed as a multicentre parallel, controlled, randomised, phase IV, non-commercial, independent trial. The study was cofunded by a Spanish health ministry grant on Independent Drug Research (grant number TRA-047), and by the organising research team's own funds. Eleven Spanish hospitals from different regions participated. Patients were included from December 2010 to May 2012. The study received approval and was audited by the ethics committees of all participant hospitals, and by the Drugs National Authority. This study was conducted in accordance with the principles of the Declaration of Helsinki, ICH Guidelines for Good Clinical Practice and in full conformity with relevant regulations. Written informed consent was obtained from all patients before enrolment. All authors had access to the study data and reviewed and approved the manuscript.

Study population

Inclusion criteria: Participants with non-investigated/functional dyspepsia or gastroduodenal ulcer with indication of H pylori eradication treatment; ability and willingness to participate in the study and to sign and give informed consent; confirmed H pylori infection by at least one of the following methods: 13C-urea breath test, histology, rapid urease test or bacterial culture.

Exclusion criteria: Age below 18 years; advanced chronic disease that would not allow the patient to complete follow-up or attend to visits; allergy to any of the study drugs; previous gastric surgery; pregnancy or breastfeeding (female participants with childbearing potential were required to use medically accepted contraception for the duration of the study); alcohol or drug abuse; previous H pylori eradication treatment; and taking antibiotics or bismuth salts 4 weeks prior to inclusion.

Intervention

Sequential treatment group: Induction phase of 5 days under dual therapy (omeprazole 20 mg/12 h, and amoxicillin 1 g/12 h) immediately followed by a treatment phase of 5 days under triple therapy (omeprazole 20 mg/12 h, clarithromycin 500 mg/12 h and metronidazole 500 mg/12 h).

Concomitant treatment group: 10 days under quadruple therapy (omeprazole 20 mg/12 h, amoxicillin 1 g/12 h, clarithromycin 500 mg/12 h and metronidazole 500 mg/12 h).

All drugs used in both therapies were of generic branding.

Trial bias minimisation

Randomisation: Patients were allocated to one of the treatment arms using a 1:1 computerised random number table. Allocation was stratified by hospital and presence or not of ulcer disease, and blocked in tandems of 6.

Allocation concealment: Allocation was concealed (opaque sealed envelopes), and the coordinators and the investigators in the centres did not know the details of the allocation sequence.

Blinding: This trial was unblinded; the number of drugs to be taken and the dosing are different among arms and the principal outcome is H pylori eradication which is not influenced by the unblinded design of the protocol. Urea breath test personnel were blind to treatment given. Informed consent obtainment, patient's allocation and drug dispensation were performed by the study doctors in the outpatient clinics.

Treatment was clearly explained to all patients. Study drugs were handed to the patient in a registered box that included all the required pills, and a day-by-day intake scheme and diagram. Drugs were self-administered orally at home after meals. All non-study medication taken by the patient during the study was recorded.

Trial outcomes

Primary outcome: Confirmed H pylori eradication by intention to treat a minimum of 4 weeks after ending treatment using 13C-urea breath test or histology. Prior to testing, patients had to withdraw PPI treatment (15 days) and any antibiotic treatment (1 month).

Secondary outcomes: Compliance cut-off point was set in taking at least 90% of each study drug. Per-protocol analysis of H pylori eradication; compliance to treatment regimen (patient interrogation and residual drug count, checked by the clinical trial monitor); and treatment-emergent adverse event/adverse reaction via open answer general question, and specific questionnaire of commonly described adverse events to treatment, as well as a biochemical and haematological blood analysis. Any discomfort, even if mild, referred by the patients and evaluated by their doctors as related to treatment, was registered.

Statistical analysis

Variables and outcomes: Continuous variables are presented as arithmetic mean and SDs. Qualitative variables are presented as percentage and 95% CIs (95% CI). A level of p<0.05 was set as significance cut-off point.

Intention to treat: Includes all randomised eligible patients regardless of the correct follow-up or compliance.

Per protocol: Only patients who had done a correct follow-up and a compliance of a minimum of 90% of each study drug.

A multiple logistic regression analysis was performed. The dependent variable was H pylori eradication, and the independent variables were age, sex, reference hospital, smoking habit, diagnosis (ulcer or dyspepsia), adverse events, compliance and treatment. We used a backward modelling strategy, and the log-likelihood ratio was the statistic for model comparison.

Sample size: The hypothesis for the expected efficacy was based on previously published data on sequential and concomitant regimens. Using the formula: (Z0.95+Z0.80)2 (Ps(1-Ps)+Pn(1-Pn)}/(Ps-Pn-D)2 by Blackwelder,21 a sample size of 314 patients was calculated (level of signification α=5%, statistical power=80%, range of equivalence or hypothesised difference=10% for a conservative expected response rate of 85% in both regimens). A 10% maximum lost to follow-up has been estimated; therefore, the final sample size calculated was 345 patients.

Results

A schematic flow diagram of patient inclusion and follow-up process is shown in figure 1. All eligible patients who accepted participation were recruited. Of all included patients (343), one was not randomised and four withdrew consent and were erased from the database as required, and therefore were not included in any analysis. The intention-to-treat population was 338 patients, 170 received sequential treatment and 168 concomitant treatment. Baseline characteristics of patients are shown in table 1. There were no significant differences among baseline characteristics among treatment arms regarding any population variable (sex, age, presence of ulcer, smoking habit, nor comorbidities).

Table 1

Baseline characteristics of the studied patients

Figure 1

Trials flow-chart scheme.

As shown in table 2, the eradication efficacy of both treatments was equivalent, both in the intention-to-treat analysis: 86.9% (95% CI 82% to 92%) for concomitant versus 81.2% (95% CI 75% to 87%) for sequential, (p=0.15); and the per-protocol analysis: 91.2% (95% CI 86% to 96%) for concomitant versus 85.6% (95% CI 80% to 91%) for sequential (p=0.14). No statistically significant differences were found between treatments based on presence or not of ulcer disease, however, a trend towards higher eradication rates was found in concomitant regimen for dyspeptic patients compared with those with ulcer disease (88.1% vs 82.4%, p=0.1). Compliance for both treatments was also not significantly different: 82.7% (95% CI 77% to 88%) for concomitant versus 82.4% (95% CI=77% to 88%) for sequential.

The list and proportion of adverse reactions is shown in table 3. Sixty-five percent of the patients referred some kind of discomfort during treatment or follow-up, whereas adverse reactions appeared in 58.6% (95% CI 53% to 64%) of the patients, with an average length of 6.1 days. The most common adverse reactions were taste distortions, affecting 120 patients (35.9%, average length 7.1 days), followed by diarrhoea (including any mild increase in number or any softening of stools) in 67 patients (20.1%, average 5.3 days) and nausea in 36 patients (10.8%, average 5.3 days). Overall, 59.2% (95% CI 52% to 66%) of the adverse reactions were mild and 36.2% were moderate, whereas only 18 (5%, 95% CI 2% to 8%) of adverse reactions were of severe intensity (affecting 13 patients, 4%, 95% CI 1% to 7%).

Table 2

Efficacy and compliance

Table 3

Adverse reactions

The most common adverse reactions not predefined in the questionnaire were candidiasis (13 patients), headaches (11 patients), stomatitis (10 patients), other gastrointestinal discomforts (6 patients), skin rash (4 patients), dizziness (4 patients), nervousness-anxiety (4 patients), sleep alterations (4 patients), hepatic biochemical blood test alterations (3 patients) and flu-like symptoms (2 patients). No statistically significant differences were found between treatments in the severity or the rate of adverse reactions: 63.1% (95% CI 56% to 70%) with concomitant regimen and 54.1% (95% CI 47% to 62%) with sequential treatment (p=0.09).

Twelve patients discontinued medication due to adverse events, seven in concomitant treatment versus five in sequential treatment. Although medication was discontinued, of these patients, four of the concomitant and two of the sequential treatment still achieved H pylori eradication.

Only two serious adverse events were reported, both occurring with concomitant treatment: a vomiting cycle that required emergency room care (lasting 1 day); and a patient who required sick leave from work (10 days) 1 month after treatment, unrelated to study drugs. However, in both cases, patients completed treatment and were not withdrawn from the study.

Multivariate analysis for treatment hospital, sex, age, smoking habit, presence of ulcer, type of treatment and compliance, found association only with compliance (OR of 3.11, 95% CI 1.73 to 6.35). However, concomitant treatment showed an OR of 1.54 towards better eradication rate in a borderline significance CI (95% CI 0.85 to 2.78).

Discussion

As the efficacy of standard triple therapy does not reach the acceptable threshold of 80% eradication rate in most contexts, new strategies must be sought.11 ,12 In the search for efficacious substitutes of the commonly recommended standard triple therapy, prolonging treatment duration up to 14 days is an option. However, the increase in efficacy is small and it seems unlikely that therapies this long, which may increase side effect rates and reduced compliance, will be generally accepted.

Another alternative is bismuth containing quadruple therapy (PPI, tetracycline, metronidazole and bismuth salts). This treatment is generally recommended as second-line treatment after failure with standard triple therapy. Even though it requires taking many pills in a complex scheme, it has demonstrated that it is at least as effective, and as well tolerated, as triple therapies.22 However, this regimen is not accessible worldwide due to the lack of distribution of tetracycline and bismuth salts in several countries.

The last recommended regimens are the so-called non-bismuth quadruple regimens, which are studied in our trial. Some clinical guidelines include sequential treatment as a first-line treatment option,5 ,23 and some authors consider it as the best first-line option in countries with high resistance rates to clarithromycin.24

However, the efficacy of sequential treatment seems to be decreasing over time. A recent meta-analysis,25 promoted by the Cochrane Collaboration, pooling data from all randomised clinical trials comparing sequential versus standard triple therapy has demonstrated that, although higher efficacy is achieved with sequential treatment (with an OR of 2.11), the intention-to-treat effectiveness was not as good as expected (83.5% with sequential vs 73.8% with standard triple therapy). Moreover, the beneficial effect could not be demonstrated when comparing sequential therapy against 14-day standard triple therapy.

Regarding concomitant therapy, another meta-analysis has shown that non-bismuth concomitant regimens containing a PPI, amoxicillin, clarithromycin and metronidazole achieve almost 90% cure rate, and that its efficacy is significantly better than that of standard triple therapy (OR = 2.36) in randomised trials comparing both regimens.26 Concomitant therapy has been recommended as first-line treatment in the last Spanish Consensus Conference on H pylori infection.27

The results from our study, mimicking routine clinical practice in Spain, suggest that both sequential and concomitant regimens achieve equivalent, and higher than 80%, H pylori eradication rates in an intention-to-treat analysis. However, a borderline tendency towards better results with concomitant regimen was found (86% vs 81%, OR 1.5). Equivalent rates of compliance of approximately 82% were obtained with both regimens, which may explain why both treatments did not achieve higher intention-to-treat eradication rates. Both treatments showed similarly acceptable safety profiles: approximately 59% of patients presented that some discomfort emerged from the treatment, although intensity was mostly mild and duration was short. Only two serious adverse events were reported (a vomiting cycle and a sick leave) and were solved without leaving sequels to the patient.

Up to now, only three studies have compared both non-bismuth treatments head-to-head for H pylori eradication in a randomised clinical trial, and ours is the first one in a European context. Greenberg et al20 in a multinational study including nearly 500 patients per arm, compared a 10-day sequential versus a five-day concomitant regimen achieving 77% and 74% eradication rates, respectively. However, the low efficacy achieved with the concomitant treatment might be explained due to the short regimen prescribed (five days). The study by Huang et al18 including 85 patients per arm used more similar regimens to the ones prescribed in our study, although they selected lansoprazole instead of omeprazole. The efficacy (sequential=80% and concomitant=88%) and the rate of adverse events of approximately 63% reported by their study agree with our results. Huang's study reported a higher compliance, although probably due to the different cut-off point established (less than 70% of drug intake in their study and 90% in ours).

Finally, the Taiwanese study by Wu et al19 reported slightly higher H pylori eradication rates with both treatments (approximately 90%). However, the results from the 230 patients included in their study can hardly be compared with our results, as they used esomeprazole 40 mg twice daily which, in a recent meta-analysis, has demonstrated to improve H pylori eradication efficacy, at least in standard triple therapy.28 Moreover, the clarithromycin resistance rate was extremely low (6.6%) in comparison with the Spanish (14%) and European rates (17.5%) reported in a very recent multicentre study.10

The main limitation of our trial is not having evaluated antibiotic resistance in the studied patients; however, the recent publication of the European H pylori resistance to antibiotics study gives a good estimation of the expected resistance (in Spain, 14% against clarithromycin, and 28% against metronidazole) in the treated patients.10 Moreover, the aim of this study was to evaluate sequential and concomitant treatments mimicking routine clinical practice in Spain where routine antibiotic resistance testing is not usually performed nor recommended.29 ,27 Although acceptable compliance rates were achieved in our study, these figures may be different in clinical practice, especially in primary care. As compliance was the only factor statistically associated with efficacy, prescribing doctors should stress the importance of correct intake to patients, and/or even hand in a treatment diagram for patients to follow and check.

The H pylori eradication efficacy and safety for the sequential and concomitant treatments presented in our study also agree with those reported in the latest observational studies and meta-analyses.16 ,26 Another recent Spanish study has been published30 in which concomitant and sequential treatments were evaluated. Data from this study shows similar intention-to-treat results in the untested and antibiotic susceptible patients. For example, concomitant treatment eradication rates for clarithromycin resistant or dual clarithromycin-metronidazole resistance also achieved acceptable results (100% and 75%), although the number of subjects with resistant strains was relatively low.

In conclusion, given the unacceptable eradication rates of standard triple therapies, the results from our study suggest that sequential and concomitant regimens offer an acceptable safety profile and compliance rate, and achieve equivalent H pylori eradication rates. Like previous studies, we observed a non-statistically significant 5% increase in eradication, allowing to achieve over 90% cure rates with the use of concomitant treatment. Although further comparative studies and meta-analyses will be necessary to confirm this advantage. Current data suggest that a 10-day concomitant regimen for first-line treatment may be the most efficient strategy for the eradication of H pylori infection.

View Abstract

Footnotes

  • Collaborators Mercedes Ramas: Digestive Services, Hospital Universitario de La Princesa and Instituto de Investigación Sanitaria Princesa (IP), Madrid; Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBERehd). Raquel Millán: Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBERehd); Digestive Services, Hospital Ntra. Sra. Virgen de Valme, Sevilla. Patxi Aranguren: Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBERehd); Digestive Services, Hospital Clínico Universitario ‘Lozano Blesa’, Zaragoza. Pilar García-Iglesias: Digestive Services, Hospital de Sabadell, Sabadell. Blanca Belloc: Digestive Services, Hospital de San Jorge, Huesca. Xavier Bessa: Digestive Services, Hospital del Mar, Barcelona. Empar Sainz: Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBERehd); Digestive Services, Hospital Mutua de Terrassa, Terrassa, Spain. Jose-Luis Gisbert: Centro de Salud Hermanos Garcia-Noblejas, Madrid. Eloisa Lamas: Digestive Services, Hospital Ntra. Sra. Virgen de Valme. Sevilla. Ariadna Figuerola: Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBERehd); Digestive Services, Hospital de Sabadell, Sabadell. Cristina Álvarez: Digestive Services, Hospital del Mar, Barcelona. Eusebio S Marcos: Centro de Salud Hermanos Garcia-Noblejas, Madrid. Maria-Isabel Moreno: Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBERehd); Service of Clinical Pharmacology, Hospital Universitario de La Princesa and Instituto de Investigación Sanitaria Princesa (IP), Madrid. Francisco Abad-Santos: Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBERehd); Service of Clinical Pharmacology, Hospital Universitario de La Princesa and Instituto de Investigación Sanitaria Princesa (IP), Madrid.

  • Contributors AGM: Study concept and design; acquisition of data; analysis and interpretation of data; drafting of the manuscript; statistical analysis; obtained funding. ACM: Study design; analysis and interpretation of data; critical revision of the manuscript; statistical analysis; administrative, technical and material support; study supervision. JM-I, MC, JB, JD, XC, CdlC, MM, FB, AP-A and MF: Acquisition of data; critical revision of the manuscript. JPG: Study concept and design; acquisition of data; analysis and interpretation of data; statistical analysis; critical revision of the manuscript; obtained funding; study supervision; study group coordinator and principal investigator.

  • Funding CIBEREHD is funded by the Instituto de Salud Carlos III. This study was cofunded by a Spanish Health Ministry Grant on Independent Drug Research (Grant number TRA-047), and by the organising research team's own funds. This study was not funded by the pharmaceutical industry.

  • Competing interests None.

  • Ethics approval Comite de Investigación Biomédica del Hospital Universitario de la Princesa.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement Only the authors (intellectual contributors) of the study have access to the database and to any other unpublished data. In any case, the database is anonymous.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

References