Hybrid therapy as first-line regimen for Helicobacter pylori eradication in a high clarithromycin resistance area: a prospective open-label trial

Introduction

As a result of rising antibiotic resistance, the eradication rate of Helicobacter pylori (H. pylori) using standard triple therapies is unacceptably low (<80%) in most countries [1]. Therefore, bismuth-based and non-bismuth quadruple first-line options have been proposed, aiming to overcome increasing clarithromycin resistance [2,3]. A rational approach is to use a clarithromycin-free regimen (i.e., bismuth-based quadruple therapy). Nevertheless, non-bismuth quadruple therapies are the only realistic approach in countries where bismuth salts and/or tetracycline are not available [4]. An example is Greece, a country with high prevalence of resistance to clarithromycin (>27%) and metronidazole (>40%) [5-7]. Until recently, the sequential and concomitant regimens have been the competing non-bismuth quadruple treatments, with several studies assessing their efficacy, safety, and compliance rates. Finally, the 2015 Maastricht V/Florence consensus favored the use of concomitant (over sequential) therapy, by virtue of its better efficacy against antibiotic-resistant H. pylori strains supported by head-to-head trials [8,9].

Since 2011, the armamentarium of non-bismuth quadruple treatments has further expanded. This was due to the addition of an innovative 14-day, two-step regimen that functionally combines the concomitant and sequential treatments (i.e., hybrid therapy) [10]. It comprises a proton pump inhibitor (PPI) and amoxicillin for 14 days, adding clarithromycin and a nitroimidazole as a quadruple therapy for the final 7 days [2,10]. Compared to a concomitant treatment of same duration (14 days), the hybrid regimen is characterized by a lower pill burden. This could be associated with potential advantages, including a lower cost of eradication but also improved safety, convenience and better compliance. In the original pilot study, hybrid therapy produced outstanding cure rates, yielding 97% in intention to treat (ITT) and 99% in per protocol (PP) analysis [10]. However, this study was conducted in a setting (Taiwan) with low clarithromycin resistance. Thus, further validation is warranted for hybrid therapy, providing concrete evidence of its efficacy in geographical and/or national areas harboring different patterns of antimicrobial resistance [11]. To date, only a limited number of studies have evaluated hybrid therapy in a high-resistance setting, and even fewer provided pretreatment susceptibility data. Therefore, this prospective open-label trial was aimed to determine the clinical efficacy and safety of the hybrid regimen in treatment-naïve patients in Greece. The impact of H. pylori resistance to clarithromycin and metronidazole on the efficacy of hybrid therapy was also evaluated.

Patients and methods

Results

Patient data and antibiotic resistance rates

The study flow chart is shown in Fig. 1. A total of 343 patients were initially screened for eligibility and 188 were excluded. Thus, 155 patients were finally included in the study; their baseline characteristics are shown in Table 1. H. pylori culture and sensitivity test were successfully carried out in 135/155 (87.1%) patients. The rates of H. pylori strains with single resistance to clarithromycin, metronidazole and amoxicillin were 35/135 (25.9%), 42/135 (31.1%) and 0/135 (0%), respectively, whereas dual clarithromycin/metronidazole resistance was detected in 12/135 (8.9%).

Figure 1 Flowchart of the patients in the study

H. pylori, Helicobacter pylori

Table 1 Clinical and demographic characteristics of the treated population

Side effects and adherence

One hundred fifty-five patients were included in the trial, but a total of seven patients were lost to follow up (Fig. 1). Among the 148 patients who returned for follow up, a total of 58 adverse events were noted in 44 (29.7%) patients. The proportion and severity of adverse events is shown in Table 2. Overall, 2 of the 148 patients (1.3%) specified severe adverse events as the reason for treatment interruption. One patient discontinued treatment on day 11 because of abdominal discomfort; however, H. pylori was successfully eradicated, as indicated by the negative breath test on final follow up. However, the other patient, who interrupted treatment on day 3 because of an allergic skin rash, failed to eradicate the pathogen. Overall, 96.6% (143/148) of the patients took more than 90% of the total medication prescribed. Adherence to therapy, defined as the total percentage of tablets taken, was 98.3% (95%CI 98.1-98.5%).

Table 2 Adverse events resulting from eradication therapy

Eradication efficacy and associated factors

Overall, eradication rates were 85.8% (133/155; 95%CI 79.4-90.5%) and 90.2% (129/143; 95%CI 84.1-94.2%) respectively, in the ITT and PP analyses. No differences in the eradication rates were found between the clarithromycin-resistant and -sensitive strains (91.4% [32/35] vs. 94.3% [83/88], P=0.69), or between the metronidazole-resistant and -sensitive strains (90.5% [38/42] vs. 95.1% [77/81], P=0.44). However, dual clarithromycin-metronidazole resistance was associated with a significantly lower eradication rate (50% [6/12]) compared with single clarithromycin resistance (91.4% [32/35], P=0.005), single metronidazole resistance (90.5% [38/42], P=0.005) and dual susceptibility (97.8% [45/46], P=0.0001). By univariate analysis (Table 3), the following variables were not associated with H. pylori eradication: sex, age, BMI, history of smoking and alcohol consumption, indication for eradication (peptic ulcer disease vs. non-ulcer dyspepsia), study center, and single clarithromycin or metronidazole resistance. The presence of dual clarithromycin/metronidazole resistance was the only factor associated with the failure of hybrid therapy (OR 14.4, 95%CI 3.8-54.9, P=0.0003).

Table 3 Factors affecting the eradication efficacy (per protocol) of hybrid therapy as determined by univariate analysis

Discussion

Six years after it was first proposed, our understanding of hybrid therapy remains limited. Available evidence from 12 studies (2009-2015) has indeed shown a wide variation in the eradication rates, ranging from 77.6-97.4% in ITT and from 82.6-99.1% in PP analysis [12]. These heterogeneous results underscore the need for cautious interpretation of data pertaining to different regions and populations, which are likely to reflect different backgrounds of antibiotic resistance. To date, only a few studies evaluating hybrid therapy provided susceptibility data, mostly from regions with low-to-moderate clarithromycin resistance [10,13]. Moreover, the subpopulations tested were inherently small, thus raising the possibility of selection bias. This prospective trial, involving 155 patients from two hospitals located in two geographically distinct areas of Greece, indicates that hybrid therapy is a valuable (PP eradication: 90.2%) treatment option that is safe and well-tolerated by most patients. This was achieved in the face of high (>15%) clarithromycin, but acceptable (<9%) dual clarithromycin/metronidazole resistance, documented by extensive analysis of susceptibility data (n=135) [8]. To our knowledge, this is the first evaluation of hybrid therapy in Greece and represents the largest European series of antibiotic-resistant H. pylori strains treated with hybrid therapy.

We determined that the efficacy of hybrid therapy is markedly decreased by dual clarithromycin/metronidazole resistance. This premise has been also corroborated in previous studies that evaluated sequential and concomitant regimens [5,7,14,15]. Therefore, all non-bismuth quadruple therapies appear to be influenced by dual resistance, though not to the same extent. Rates of dual resistance >5%, >9% and >15% have been proposed as the thresholds associated with unacceptable performance (<90% cure rate) for sequential, hybrid and concomitant regimens, respectively [16]. Overall, our experience appears to support this notion, indicating that both the 10-day concomitant and 14-day hybrid therapies, but not the sequential therapy, are acceptable first-line regimens in Greece, a country with a >5% but <9% prevalence of dual clarithromycin/metronidazole resistance [5,7,17,18]. A head-to-head comparison between the 10-day concomitant and 14-day hybrid regimens would be useful to clarify the optimal first-line regimen in Greece. However, assuming they prove to be equally effective, the lower pill burden (80 tablets for 10-day concomitant; 84 tablets for 14-day hybrid) and less complex administration would favor the 10-day concomitant therapy. Crucially, shortening the duration of hybrid therapy (to 10 or 12 days) may be realistic in regions of moderate-to-low resistance [13]. However, this may be unfeasible in high-resistance settings, as shown by an Iranian randomized study comparing 14- and 10-day hybrid therapies (PP eradication: 92.8% and 83.8% respectively) [19]. Likewise, prolonged (14 days) exposure to amoxicillin is the key for improved eradication with hybrid therapy. Accordingly, prolonging the duration of sequential therapy (in which amoxicillin is discontinued at mid-point) does not seem to provide any substantial benefit [20]. To date, a single study from China recorded such a large dataset (n=135) of antibiotic susceptibility before hybrid therapy [21]. In sharp contrast to our data, hybrid therapy was shown to perform unsatisfactorily (ITT: 77%, PP: 86%), an unsurprising finding considering that the burden of antibiotic resistance was disproportionately higher (clarithromycin 44.9%, metronidazole 67.3%, dual 33.3%). Interestingly, the deleterious impact of dual resistance appears to be more prominent in the European as compared to the Asian population (efficacy against dual-resistant H. pylori strains: 70.2% in the Chinese study vs. 50% in present series) [21]. In keeping with this observation, only one out of three dual-resistant H. pylori strains was successfully eradicated in a Spanish/Italian study [22]. Clearly, a number of potential confounders may play a role, including racial and genetic background, differences in the metabolism of PPIs, and the nature of the underlying disease. Furthermore, variations in the type of PPI and nitroimidazole may affect the eradication outcome. Notably, tinidazole, which is known to possess a longer half-life compared to metronidazole, was used in the Chinese study. With regard to acid suppression, we used high-dose (40 mg b.i.d.) esomeprazole, a new-generation PPI that has been reported to be metabolized irrespective of the cytochrome P450 (CYP) 2C19 status [23]. In contrast, standard-dose (20 mg b.i.d.) omeprazole was used in the Chinese study, justified by the lower prevalence of rapid metabolizers among the Asian population [24]. Clearly, both studies address the same effect (i.e., detrimental impact of dual resistance on hybrid therapy). However, they addressed different settings of antibiotic resistance and distinct populations that were likely to reflect different magnitudes of this same phenomenon. Thus, rather than contrasting, the two studies should be viewed as complementary.

As sequential therapy, hybrid therapy has been devised to decrease pill burden (compared to concomitant therapy) with no efficacy cost, relying on sequential administration. It has been postulated that pretreatment with amoxicillin could markedly reduce the bacterial load and prevent the development of bacterial efflux channels [25], increasing the susceptibility of the organisms during the second phase of treatment [26]. Nevertheless, the concept of sequential administration has been seriously challenged by studies showing that “reverse hybrid” and “reverse sequential” regimens perform equally well [27,28]. Indeed, in a recent Taiwanese trial, reverse hybrid therapy was highly effective (95.7% by PP analysis) and superior to standard triple therapy [27].

In keeping with our findings, previous studies evaluating hybrid therapy reported consistently good compliance and safety [10,13,19,29-33]. A Spanish/Italian study determined that the overall rate of adverse effects of hybrid therapy was lower than that of concomitant therapy [22]. Likewise, the rate of discontinued medication was lower in a Korean study [31]. However, in both studies the differences were of borderline significance (P=0.05 and P=0.051, respectively). Overall, based on updated meta-analyses, hybrid therapy appears to display slightly better compliance and safety than concomitant therapy [12,34]. However, whether this difference may be clinically relevant, particularly regarding a routine practice setting, merits further assessment.

The present trial is not free of limitations. Firstly, by design, this was an open-label, non-controlled study, precluding conclusions on the comparative efficacy and safety of hybrid therapy. Secondly, the antimicrobial susceptibility of H. pylori was determined using the E-test, which tends to overestimate metronidazole resistance, although in a previous study this was not associated with any change in the overall metronidazole resistance pattern [7]. Thirdly, CYP2C19 polymorphisms were not evaluated. However, we expect the impact, if any, to be minimal, as the metabolism of esomeprazole has been reported to be independent of CYP2C19 status [35].

In conclusion, hybrid therapy achieved an H. pylori cure rate of 90.2% (PP) in a setting with high resistance to clarithromycin and metronidazole. Dual clarithromycin and metronidazole resistance plays a key role in the treatment failure of hybrid therapy. Further studies providing susceptibility data are warranted to better define the position of hybrid therapy in the armamentarium of current first-line regimens for H. pylori eradication.