Reconciliation of Recent Helicobacter pylori Treatment Guidelines in a Time of Increasing Resistance to Antibiotics

Reconciliation of Recent Helicobacter pylori Treatment Guidelines in a Time of Increasing Resistance to AntibioticsBRIEF REVIEW
Reconciliation of Recent Helicobacter pylori Treatment Guidelines in a Time of Increasing Resistance to Antibiotics
Carlo A. Fallone,1 Steven F. Moss,2 and Peter Malfertheiner3,4
1Department of Medicine, Division of Gastroenterology, McGill University Health Center, Montreal, Québec, Canada; 2Department of Medicine, Division of Gastroenterology, Warren Alpert Medical School of Brown University, Providence, Rhode Island; 3Department of Gastroenterology, Hepatology and Infectious Diseases, Otto-von-Guericke University, Magdeburg, Germany, and 4Ludwig Maximilian University Clinic, Medical Clinic II, München, Germany
Increasing resistance to antibiotics worldwide has adverse effects on the effectiveness of standard therapies to erad- icate Helicobacter pylori infection. We reviewed guidelines developed by expert groups in Europe, Canada, and the United States for the treatment of H pylori infection. We compared the recommendations of these guidelines, reconciled them, and addressed the increasing resistance of H pylori to antibiotic therapy regimens. The guidelines recommend bismuth quadruple therapy for first-line treatment, replacing clarithromycin-based triple therapy. There is consensus for concomitant 4-drug therapy as an alternative, especially when bismuth is not available. When therapy is unsuccessful, it is likely due to resistance to clarithromycin, levofloxacin, and/or metronidazole; these drugs, if used previously, should be avoided in subsequent eradication attempts. Second-line therapies should be bismuth quadruple therapy or levofloxacin triple therapy, depending on suspected resistance, reserving rifabutin- based triple and high-dose dual amoxicillin proton pump inhibitor therapy for subsequent treatment attempts. The increasing resistance of H pylori to antibiotic therapy necessitates local availability of susceptibility tests for individuals, and establishment of regional and national monitoring programs to develop evidence-based locally relevant eradication strategies. Further studies into the development of more easily accessible methods of resistance testing, such as biomarker analysis of stool samples, are required. Options under investigation include substituting vonoprazan for proton pump inhibitors, adding probiotics, and vaccine development. Narrow-spectrum antibiotics and new therapeutic targets could be identified based on genomic, proteomic, and metabolomic analyses of H pylori.
Keywords: PPI; Gastric; Gastritis; Stomach; Bacteria; Drug.
elicobacter pylori gastritis is an infectious disease1,2
Abbreviations used in this paper: ACG, American College of Gastroen- terology; PCAB, potassium-competitive acid blocker; PPI, proton pump inhibitor.
Most current article
https://doi.org/10.1053/j.gastro.2019.04.011
H responsible for many peptic ulcers and >80% of cases of gastric cancer3; it is the third most common cause of cancer death in the world4 and is associated with several other gastric and extra-gastric diseases.5,6 Eradicating H pylori cures gastritis and peptic ulcer disease7 and has the potential of preventing gastric cancer.8
In the 1990s, 7- to 10-day triple therapies with twice daily clarithromycin, metronidazole (or amoxicillin), and proton pump inhibitors (PPIs) produced eradication rates of up to 90% in clinical trials and rapidly became the standard
treatment in most parts of the world.9,10 Since then, success with triple therapy has rapidly declined due to increasing clarithromycin resistance.11,12 The success of many alter- native regimens is also limited by increasing antibiotic resistance (eg, to levofloxacin13). As H pylori therapy has become challenging, treatment guidelines have been issued recently by 3 separate authoritative groups: the European Helicobacter and Microbiota Study Group (Maastricht V/ Florence report),6 the American College of Gastroenterology (ACG),5 and the Canadian Association of Gastroenterology/ Canadian Helicobacter Study Group (Toronto Consensus).14
Published in close succession, the 3 reports are based on thorough review of largely the same literature, but with slightly different emphases, with the target audiences and regional health care systems in mind. We review the most important recommendations of these guidelines and consider ways around the current quagmire of increasing resistance of H pylori to many currently used antibiotics.
Reconciling the Guidelines Using Antibiotic Susceptibility Data
Each guideline strongly recommends the use of anti- biotic susceptibility data whenever available or obtainable. This knowledge would limit inappropriate antibiotic use, thus preventing widespread resistance development in other organisms and reducing costs. For example, using clarithromycin triple therapy instead of concomitant quadruple therapy if the individual’s strain was known to be clarithromycin sensitive or the local resistance rate was <15% would avoid exposure to metronidazole (see section on first-line recommendations). The problem remains that these data are not available in many parts of the world, including the United States. We believe it is of the utmost importance that national and regional data should be available, as it is in several parts of Europe through the use of a registry, and that individual susceptibility testing be easily accessible. This is discussed further in this article, after the section on guideline reconciliation.
Therapy Toronto Maastricht V/Florence ACG
Bismuth quadruple therapy Recommended choice Recommended (only choice if high duala resistance)
Recommended choice
Concomitant therapy Recommended choice Recommended if high C-Res or if bismuth unavailable
Recommended choice
PPI triple therapy Restricted to areas of <15% C- Res or proven eradication success >85%
Recommended only in areas of low C-Res
Recommended if area of C-Res <15% and no previous macrolide exposure
Sequential and hybrid therapies Recommended against its use Not recommended Suggestedb
Levofloxacin regimens Not recommended No statement Suggestedb
C-Res, clarithromycin resistance. aDual refers to resistance to both clarithromycin and metronidazole. bSuggested means that the ACG finds it permissible for practice but not ideal.
July 2019 Reconciliation of H pylori Guidelines 45
RE VI EW
CT IV ES
In the meantime, the 3 consensus groups thought it was important to provide guidelines for treatment options to clinicians, even in the absence of such data. The common- alities and differences of these guidelines are presented next.
First-Line Therapy The recommended options for first-line treatment of H
pylori infection are very similar for all 3 consensus groups.5,6,14 All agree that the best approach is to succeed on the first attempt, thus avoiding retreating and retesting, and reducing cost, anxiety, and negative impacts on other gut microbiota. If the antibiotic susceptibility profile of the organism is unknown (as is usual), they all recommend choosing 1 of 3 regimens based on previous antibiotic exposure or on known local prevalence of resistant organ- isms (Table 1). Bismuth quadruple therapy is encouraged particularly in areas of high dual resistance to clari- thromycin and metronidazole, and concomitant therapy is appropriate for those patients from areas of high clari- thromycin resistance where bismuth is not available (dos- ages detailed in Table 2). The bismuth quadruple therapy is favored because it avoids the potential problem of antibiotic overuse mentioned, and has acceptable success even in strains displaying in vitro metronidazole resistance. The third alternative is clarithromycin-based triple therapy, but its use is limited to areas of known low clari- thromycin resistance (<15%), and no personal history of macrolide exposure. Because of the scarcity of known susceptibly profiles in many parts of the world, the Tor- onto Consensus accepted a recently proven high local eradication rate (>85%) with this regimen as a surrogate marker of low prevalence of clarithromycin resistance. In order to know this, it is imperative that the clinician has proof of cure or failure with all treatments offered. After a treatment attempt, testing to confirm eradication is mandatory.
The differences between the 3 consensus groups with regard to first-line therapy arise from the fact that the ACG accepts some regimens that it considers not ideal but still
permissible for clinical practice. It labels these as “suggested” regimens and thought this was necessary because of the paucity of recent US data on resistance patterns and eradi- cation rates. Among the ACG’s permissible regimens are sequential and hybrid quadruple therapies, with the caveat that their complexity could lead to lower patient compliance and lack of enthusiasm among clinicians (dosages detailed in Table 2). The Toronto Consensus and Maastricht V groups refer to evidence of superiority of concomitant over sequential therapy, and to its simpler nature, thus improving compliance without any additional unwanted effects (both regimens contain the same antibiotics). Toronto Consensus thus rec- ommends against sequential therapy, but all 3 agree that concomitant therapy is the preferred non-bismuth quadruple therapy. The ACG also lists levofloxacin therapy as a suggested first-line permissible option, but cautions that levofloxacin resistance may be quite high in North America (31% ac- cording to the only published US study of levofloxacin resis- tance, in just 110 strains15). For these reasons and because of concerns about generating increasing levofloxacin resistance in other bacterial species, Toronto Consensus recommends against levofloxacin use in first-line therapy. The European report does not consider levofloxacin as a first-line therapy, but comments that one can consider it in a patient from an area of high dual (clarithromycin and metronidazole) and low levofloxacin resistance if bismuth is not available. Overall, all guidelines agree that in general levofloxacin therapy is not a first-line option for most patients.
With regard to duration of first-line therapies, the Toronto Consensus concludes that there is sufficient ev- idence of higher success with 14 days vs shorter dura- tions. To provide simplicity for clinicians and based on the concept that the best approach is to succeed on the first attempt, they recommend 14 days for all first-line treatments. The Maastricht V/Florence Consensus is in agreement, unless 10 days has been proven locally to be as effective for quadruple therapies (Table 2). The ACG recommends 14 days for PPI triple and hybrid therapies, but lists 10–14 days as suitable for bismuth quadruple, concomitant, sequential, and levofloxacin therapies in the absence of data from the United States showing
Table 2.Regimens used for Helicobacter pylori Eradication
Regimen Dosage Duration Recommendation
Concomitant non-bismuth quadruple (PAMC)
PPIa bid Amoxicillin 1000 mg bid Metronidazoleb 500 mg bid Clarithromycin 500 mg bid
TOR: 14 d MAA: 14 d unless 10 d proven
locally ACG: 10–14 d
First line: recommended by all Rescue: MAA and ACG only
Bismuth quadruple (PBMT) PPIa bid Bismuthc qid Metronidazoleb 400 mg qid or 500 mg tid–
qid Tetracycline 500 mg qid
TOR: 14 d MAA: 14 d unless 10 d proven
locally ACG: 10–14 d
First line: recommended by all Rescue: recommended by all
PPI triple (PAC, PMC) PPIa bid Amoxicillin 1000 mg or metronidazoleb 500
mg bid Clarithromycin 500 mg bid
TOR: 14 d MAA: 14 d ACG: 14 d
First line: restricted by all Rescue: MAA only (restricted)
Levofloxacin triple (usually PAL)d
PPIa bid Amoxicillin 1000 mg bid Levofloxacin 500 mg once daily or 250 mg
bid
TOR: 14 d MAA: 14 d ACG: 10–14 d
First line: suggested by ACG only
Rescue: recommended by all
Sequential non-bismuth quadruple
PPIa bid Amoxicillin 1000 mg bid for first half only Metronidazoleb 500 mg þ clarithromycin
500 mg bid for second half only
TOR: Not recommended MAA: Not recommended ACG: 10–14 d
First line: non-ideal ACG optione
Rescue: not recommended by any
Hybrid non-bismuth Quadruple
PPIa bid Amoxicillin 1000 mg bid Metronidazoleb 500 mg þ clarithromycin
500 mg bid for second half only
TOR: Not recommended MAA: Not recommended ACG: 14 d
First line: non-ideal ACG optione
Rescue: not recommended by any
High-dose dual therapy Rabeprazole 20 mg qid Amoxicillin 750 mg qid
TOR: 14 d MAA: not stated ACG: 14 d
Rescue: considered an option by all
Rifabutin-containing therapy (PAR)
PPIa bid Amoxicillin 1000 mg bid Rifabutin 150 mg bid or 300 mg qd
TOR: 10 d MAA: not stated ACG: 10 d
Rescue: considered an option by all as third or fourth line
ACG, American College of Gastroenterology recommendation5; bid, twice a day; MAA, Maastricht V/Florence recommen- dation6; qd, once a day; qid, 4 times a day; tid, 3 times a day; TOR, Toronto Consensus recommendation.14 aThe dose depends on PPI used. Standard doses are dexilant 30 mg, esomeprazole 20 mg, lansoprazole 30 mg, omeprazole 20 mg, pantoprazole 40 mg, and rabeprazole 20 mg, although a double dose is sometimes used for dexilant, esomeprazole, omeprazole, and rabeprazole to increase eradication success. bMetronidazole may be substituted by tinidazole. cThe dose depends on formulation used. Examples include bismuth subsalicylate (262 mg), 2 tablets, colloidal bismuth subcitrate 120 mg, 1 tablet. dLevofloxacin can also be given as a non-bismuth quadruple therapy (concomitant, sequential, or hybrid) by replacing clari- thromycin with levofloxacin. eThese combinations were not thought to be ideal. Concomitant therapy would be better choice.
46 Fallone et al Gastroenterology Vol. 157, No. 1
REVIEW S AND
PERSPECTIVES
superiority of 14-day over 10-day regimens. Reconciling these guidelines, the authors believe that 14 days would be the preferred length of first-line therapy.
In patients with true penicillin allergy, the ACG and Maastricht V/Florence both recommend either clarithromycin- based triple therapy with metronidazole if there is no prior exposure to macrolides (ACG) or the patient is from an area of low clarithromycin resistance (Maastricht V) or bismuth quadruple therapy. The Toronto Consensus prefers the latter based upon superiority over the clarithromycin/metronidazole triple therapy in a prospective study.16
Treatments After One Failed Prior Attempt All 3 groups recommend avoiding the re-use of antibi-
otics that failed previously. This is particularly true for
clarithromycin and levofloxacin, where antibiotic resistance is common after exposure. For previous metronidazole exposure, the groups believe that re-use could be an option if given with bismuth because of its synergistic effect. Amoxicillin or tetracycline could be re-used because resis- tance to these agents is rare.
All 3 groups also agree that the choice of second-line therapy is between bismuth quadruple therapy and levo- floxacin triple therapy, depending on what was used pre- viously (Table 3). For example, if metronidazole was used previously, the levofloxacin regimen would be preferred for subsequent therapy, although with the rapid increase of levofloxacin resistance, susceptibility testing to guide regimen selection should be strongly considered at this juncture.
Table 3.Treatment of Patients With Prior Failed Attemptsa
Regimen Toronto14 Maastricht V/Florence6 ACG5
Bismuth quadruple therapyb
Levofloxacin therapyb Recommended choice Recommended choice Recommended choice Clarithromycin-based PPI
triple therapy Not recommended Recommended if failed bismuth quadruple
and levofloxacin and from a low Clarithromycin resistance area
Not recommended
Not enough evidence to comment
Recommended if failed bismuth quadruple and levofloxacin and from a low clarithromycin-resistance area
Recommended if previous bismuth quadruple treatment failed
High-dose dual therapy Promising but insufficient evidence to recommend
Consider if high rate of dual clarithromycin/ metronidazole resistance
Suggested
Rifabutin therapy Restricted to those who failed 3 times
Consider if failed clarithromycin based and bismuth quadruple therapies and is from high fluoroquinolone resistance area (ie, failed 2 times)
Suggested
aTreatments should be directed by susceptibility testing whenever possible. bChoice based on previous exposure (eg, if previous exposure to clarithromycin or levofloxacin, can use bismuth quadruple, and if previous exposure to clarithromycin or metronidazole, can use levofloxacin therapy).
July 2019 Reconciliation of H pylori Guidelines 47
RE VI EW
Treatments After More Than One Failed Prior Attempt
Maastricht V recommends culture with susceptibly testing after failure of second-line therapy, but this is not widely available, especially in North America. The choice of the subsequent treatment after failure of second line is where some difference exists between the consensus documents. Again, attempts with bismuth quadruple (if previous exposure to clarithromycin, levofloxacin, or even metronidazole in non-bismuth therapy) or levofloxacin-based therapy (if previous exposure to clarithromycin and metronidazole) are recommended by all 3 groups.
Clarithromycin-based triple therapy as a salvage therapy is recommended by Maastricht V only for the situation where both bismuth quadruple and levofloxacin therapies failed in an area of known low clarithromycin resistance. The ACG and Toronto Consensus groups do not recommend this treatment given the lack of evidence for its use in salvage therapy and high probability of clarithromycin resistance. The Maastricht V and ACG guidelines both accept concomitant therapy as an option for salvage, but the Tor- onto Consensus group considered the evidence insufficient to support or refute the efficacy of this regimen in this scenario.
All 3 groups were enthusiastic about the promise of high-dose dual therapy (amoxicillin and PPI) as a third- line therapy because it avoids the issue of clari- thromycin, metronidazole, and levofloxacin resistance, and has produced impressive results in some populations.17
Finally, the rifabutin regimen (dosage detailed in Table 2) as a possible salvage treatment, was positioned either as at least third (ACG and Maastricht/Florence) or fourth (Toronto Consensus) line, enthusiasm being
tempered because of potential myelotoxicity (Figure 1). Reassuringly, although myelotoxicity has been reported in about 2% of all patients undergoing rifabutin-based therapy for H pylori infection, it appears to be fully reversible and not associated with increased suscepti- bility to infections.18
With regard to duration of therapy after prior failures, Maastricht V lists 14 days for PPI triple or levofloxacin regimens, and 10 (if proven locally) to 14 days for bismuth quadruple or concomitant quadruple (Table 2). It does not recommend a specific treatment duration for high-dose dual or rifabutin regimens. The ACG recommends 14 days for salvage therapies using bismuth quadruple, levofloxacin, high-dose dual, and 10–14 days (lack of US data) with concomitant therapies. The Toronto Consensus group rec- ommends 14 days duration for all salvage treatments except rifabutin therapy, where the ACG also agrees that the rec- ommended duration is only 10 days because of possible myelotoxicity. The authors of this review would tend to agree with the Toronto Consensus recommendations in an attempt to maximize the chance of success in those who failed treatment previously.
For patients stating that they are allergic to peni- cillin, the ACG recommends referral for allergy testing after failure of first-line therapy because most patients who think they are allergic are found to not have a true penicillin allergy.19 Maastricht V states that a fluoroquinolone-containing regimen may represent a second-line rescue option in the presence of penicillin allergy (eg, with metronidazole and a PPI with or without bismuth).
Potential Weakness of the Guidelines As with most infectious diseases, the guidelines recom-
mend treatment based on the susceptibility profile of the
Figure 1. Algorithm for eradication therapies. Adopted from Fallone et al.14 PBMT, bismuth quadruple therapy; PMC, clarithromycin-based PPI triple therapy with metronidazole; PAMC, concomitant non-bismuth quadruple therapy; PAC, clarithromycin-based PPI triple therapy with amoxi- cillin; PAL, levofloxacin- based therapy; HDDT, high-dose dual therapy; PAR. rifabutin-containing therapy. *PBMT is preferred when dual resis- tance to metronidazole and clarithromycin is sus- pected and PAMC is preferred if bismuth is not available. þGiven the rapidly increasing levo- floxacin resistance in certain areas, susceptibil- ity testing if available is recommended before us- ing PAL.
48 Fallone et al Gastroenterology Vol. 157, No. 1
REVIEW S AND
PERSPECTIVES
organism, where possible. However, obtaining this infor- mation is not always practical currently, and hence the guidelines for patients whose resistance profiles are not available were developed based on the best published evi- dence, which is largely from meta-analysis of empiric ther- apy. There are limitations with these data due to heterogeneity of studies and differences of geographical resistance patterns, but these data are nonetheless very helpful in achieving an evidence-based recommendation, as confirmed by the very similar recommendations by 3 different groups of experts. The guidelines considered this heterogeneity when recommending first-line therapies (such as including the cutoff value of clarithromycin resis- tance in the population).
Helicobacter pylori Resistance to Antibiotics Has Become a Global Challenge
Antibiotic overconsumption for medical, veterinary, and livestock purposes is a worldwide problem, responsible for sharp rises in antibiotic resistance in many bacterial spe- cies.20 In addition, with the widespread use of macrolides and fluoroquinolones for all indications, there is an associ- ated increased clarithromycin and levofloxacin resistance in H pylori strains in Europe.21 Clarithromycin-resistance is now as high as 30% in Southern Europe, and 50% in China,12,22 though there are some countries (Sweden and Taiwan) with much lower rates (approximately 15%).
July 2019 Reconciliation of H pylori Guidelines 49
RE VI EW
CT IV ES
Significant differences also occur between geographical re- gions within the same country. In many countries, including the United States, there is a paucity of relevant information, for example, fewer than 500 strains from…