Unorthodox Parenteral β-Lactam and β-Lactamase Inhibitor Combinations: Flouting Antimicrobial Stewardship and Compromising Patient Care

In India and China, indigenous drug manufacturers market arbitrarily combined parenteral β-lactam and β-lactamase inhibitors (BL-BLIs). In these fixed-dose combinations, sulbactam or tazobactam is indiscriminately combined with parenteral cephalosporins, with BLI doses kept in ratios similar to those for the approved BL-BLIs. Such combinations have been introduced into clinical practice without mandatory drug development studies involving pharmacokinetic/pharmacodynamic, safety, and efficacy assessments being undertaken.

ABSTRACT

In India and China, indigenous drug manufacturers market arbitrarily combined parenteral β-lactam and β-lactamase inhibitors (BL-BLIs). In these fixed-dose combinations, sulbactam or tazobactam is indiscriminately combined with parenteral cephalosporins, with BLI doses kept in ratios similar to those for the approved BL-BLIs. Such combinations have been introduced into clinical practice without mandatory drug development studies involving pharmacokinetic/pharmacodynamic, safety, and efficacy assessments being undertaken. Such unorthodox combinations compromise clinical outcomes and also potentially contribute to resistance development.

TEXT

Antibiotic resistance in Gram-negative pathogens has reached catastrophic proportions, leading to severe erosion in the effectiveness of mainstay antibiotics (1). In countries such as India and China, the situation is even more challenging, as the multidrug resistance (MDR) rates are much higher than in the United States and Europe. For instance, rates of resistance to 1st- and 2nd-line antibiotics used to treat Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacterales in the United States are 45% to 50%, 15% to 20%, and 10% to 20%, respectively, while in India and China, they are 60% to 80%, 30% to 35%, and 30% to 90%, respectively (2, 3). The higher resistance rates observed in India and China may be ascribed to nonjudicious use and overuse of antibiotics, especially the 2nd- and 3rd-line therapies in empirical settings (4–7). Such practices highlight an inadequate implementation of antimicrobial stewardship involving the selection of an appropriate drug and the optimal dose and therapy duration to ensure the best clinical outcome along with fewer side effects and a low risk for resistance development. Another widespread practice that flouts antimicrobial stewardship in these countries is the use of irrational parenteral β-lactam and β-lactamase inhibitor (BL-BLI) combinations (8–11) whose dose ratios have been selected without scientific basis. The clinical use of such combinations seems to have contributed to a dramatic rise in extended-spectrum β-lactamase (ESBL)-mediated resistance after the introduction of such irrational combinations in 2003 (Central Drugs Standard Control Organisation, Ministry of Health & Family Welfare, Government of India [https://cdscoonline.gov.in/CDSCO/Drugs]). For instance, resistance to cefotaxime increased from 54.8% to 83% in Escherichia coli strains and from 43.7% to 76% in Klebsiella spp. from 2000 to 2018 (see Table S1 in the supplemental material) (12–15). In contrast, the rates of cefotaxime resistance were lower in the United States, where such combinations are not approved (5% to 20% resistance among E. coli strains and 20% to 30% resistance among Klebsiella spp.) (16).

In well-regulated markets, such as the United States and Europe, combinations of clavulanic acid, tazobactam, and sulbactam with, respectively, amoxicillin-ticarcillin, piperacillin, and ampicillin are available. These BL-BLI combinations were originally developed according to a rational approach in the selection of dose ratios. Similarly, the novel BL-BLI combinations ceftolozane-tazobactam, ceftazidime-avibactam, meropenem-vaborbactam, and imipenem-cilastatin-relebactam have also gained U.S. Food and Drug Administration (FDA) and/or European Medicine Agency (EMA) approval; their dose ratios were justified through rigorous pharmacokinetics/pharmacodynamics (PK/PD) studies (17, 18). Inadequate access to such novel MDR and active therapies in countries such as India and China, coupled with a significant compromise in the effectiveness of older BL-BLIs, has resulted in a crippling therapeutic void in the treatment of contemporary Gram-negative infections (9, 19–22). Meanwhile, indigenous pharmaceutical firms in these countries resorted to marketing several unorthodox BL-BLI combinations, namely, ceftriaxone-tazobactam, ceftazidime-tazobactam, cefoperazone-tazobactam, cefepime-tazobactam, cefotaxime-sulbactam, ceftazidime-sulbactam, ceftriaxone-sulbactam, cefepime-sulbactam, and meropenem-sulbactam (Table 1) (9, 23–27). The doses of the BLIs in all these combinations were retained at the same ratios used in the older scientifically developed combinations, such as piperacillin-tazobactam and ampicillin-sulbactam. For instance, the tazobactam-based combinations are formulated in a BL/BLI dose ratio of 8:1, a ratio similar to the ratio for piperacillin-tazobactam (4 g piperacillin plus 0.5 g tazobactam). Likewise, sulbactam-based combinations were formulated in a 2:1 dose ratio, similar to the ratio for ampicillin-sulbactam (2 g ampicillin plus 1 g sulbactam).

TABLE 1.

Clinical doses of irrational β-lactam–β-lactamase inhibitor combinations available in India and China^a

BL-BLI combination	Clinical dose (mg) of:
	BL	BLI
Ceftriaxone-tazobactamb	1,000	125
	500	62.5
	250	31.25
Ceftazidime-tazobactamb	1,000	125
	500	62.5
	250	31.25
Cefoperazone-tazobactamb	1,000	125
	500	62.5
Cefepime-tazobactam	1,000	125
	500	62.5
Cefotaxime-sulbactamb	1,000	500
	500	250
	250	125
Ceftriaxone-sulbactamb	1,000	500
	500	250
	250	125
Ceftriaxone-sulbactam + EDTA at 37 mg (CSE 1034)	1,000	500
Cefepime-sulbactam	1,000	500
Meropenem-sulbactam	1,000	500

^aBL, β-lactam; BLI, β-lactamase inhibitor; EDTA, ethylenediaminetetraacetic acid. All tazobactam-based combinations are in an 8:1 ratio, like that of piperacillin-tazobactam. All sulbactam-based combinations are in a 2:1 ratio, like that of ampicillin-sulbactam.

^bCombination available in China.

In conceptualizing a new BL-BLI combination effective against highly resistant contemporary ESBL-expressing pathogens, identifying a suitable BLI to be combined with a β-lactam partner depends on several factors. These include the degree of antibacterial potency of the partner β-lactam, its relative enzymatic stability toward contemporary β-lactamases, the inhibitory potency of the BLI against these enzymes, and the PK/PD adequacy of unbound plasma concentrations of a β-lactam and BLI at the proposed clinical dose. Eventually, the BL-BLI dose and the ratio should be optimal for providing clinical coverage of the envisaged target pathogens. For instance, the significance of BLI plasma concentrations in selecting the BLI dose is evident when comparing the dose ratios of three tazobactam-based combinations, namely, piperacillin-tazobactam (8:1), ceftolozane-tazobactam (2:1), and cefepime-tazobactam (1:1). Depending on the PK interaction between the respective β-lactam and tazobactam (there is a positive PK interaction between piperacillin and tazobactam but no such interaction between ceftolozane and tazobactam and between cefepime and tazobactam) and the target pathogens’ coverage profile, the dose ratio of tazobactam varies with each of these partners (28–30). For instance, in cefepime-tazobactam, a high 1:1 dose ratio (2 plus 2 g, every 8 h [q8h]) that delivers 6 g tazobactam/day helped provide comprehensive coverage against ESBL/class C-producing pathogens because of the ability of high and sustained concentrations of tazobactam to spare cefepime from the hydrolytic attack of multiple ESBLs and class C and KPC (Klebsiella pneumoniae carbapenemase) β-lactamases (31–33). Such a coverage profile is not possible with ceftolozane-tazobactam (2:1), which delivers a lower dose (3 g tazobactam/day) in combination with a highly labile β-lactamase drug. Therefore, for a given BL-BLI combination, the appropriate BLI dose ultimately depends upon complementary features of the partner BL and the envisaged pathogen coverage profile. However, for arbitrarily combined BL-BLI combinations, the above-mentioned aspects were completely ignored.

For justifying a selected clinical dose ratio for a BL-BLI combination to regulatory agencies, such as the U.S. FDA and EMA, several nonclinical and clinical studies are required to be undertaken (34) (illustrated in Fig. S1). These studies include in vitro/in vivo PK/PD studies to identify pharmacodynamic targets (e.g., the percentage of time that the free drug is above the MIC [% fT>MIC] for a β-lactam and the percentage of time that the free drug is above the threshold concentration [C_T] for a BLI) and a phase I trial to assess safety and pharmacokinetic interactions and also to develop a population pharmacokinetic model to undertake Monte-Carlo simulation (MCS) for supporting clinical doses. Finally, a phase III study is necessary to establish substantial evidence of safety and efficacy in a patient population. Such studies, unfortunately, have not been undertaken for unorthodox combinations.

Since the dose ratios of unorthodox BL-BLI combinations are merely copied from scientifically developed BL-BLIs, the clinical attainment of β-lactam pharmacodynamic targets (based on a requirement of a % fT>BL-BLI MIC of 50% to 70%) for contemporary pathogens is not known. For these unorthodox combinations, our assessment revealed that tazobactam or sulbactam is not comprehensively effective in restoring the activity of a respective partner β-lactam against contemporary clinical isolates expressing multiple class A ESBLs and class C enzymes. Table 2 shows the MICs of these unorthodox BL-BLI combinations against contemporary Gram-negative Indian isolates. Though there was a drop in the MIC₅₀ of a BL-BLI combination compared to that of a β-lactam alone, MIC₉₀s remained significantly higher (≥16 mg/liter). The concentration-time profile of these β-lactams at the selected dose indicate that the stipulated 50% to 70% fT>MIC₉₀-based pharmacodynamic target is not attainable (Fig. S2 and S3). This PK/PD deficit is associated with all the irrational combinations listed in Table 1, except meropenem-sulbactam (1 g plus 0.5 g, q8h) versus ESBL/class C-expressing Enterobacterales. However, for ESBL/class C-expressing Enterobacterales, the addition of sulbactam to meropenem is not justifiable, as standalone meropenem is highly stable against ESBLs and class C β-lactamases (35). Furthermore, in the cases of carbapenem-hydrolyzing, OXA-expressing Acinetobacter baumannii and class B (NDM, VIM) carbapenemase-expressing Enterobacterales, meropenem has limited activity, and addition of sulbactam cannot be justified, as it does not effectively inhibit these carbapenemases (36, 37). One may argue that sulbactam has clinical utility against A. baumannii due to its intrinsic antibacterial activity against this pathogen. However, for the coverage of A. baumannii, the sulbactam dose needs to be higher to meet the 50% fT>MIC target, since sulbactam is modestly active against OXA carbapenemase-expressing A. baumannii isolates (MIC, 16 to 32 mg/liter), as shown in Table 3. In light of the paucity of epidemiological data regarding the prevalence of these enzymes in various clinical isolates, Shafiq et al. previously raised this concern of the potential misuse of this combination (9).

TABLE 2.

MICs of irrational BL-BLI combinations and ceftazidime-avibactam against contemporary Indian ESBL- and/or class C β-lactamase-producing Enterobacterales isolates^a

β-Lactam or BL-BLI (dose or ratio)	MIC50 (mg/liter)	MIC90 (mg/liter)
Ceftriaxone	>128	>128
Ceftriaxone + tazobactam (4 mg/liter)	2	128
Ceftriaxone + sulbactam (2:1)b	32	64
Ceftazidime	128	>128
Ceftazidime + tazobactam (4 mg/liter)	4	128
Cefepime	32	>128
Cefepime + tazobactam (4 mg/liter)	0.5	8
Cefepime + sulbactam (4 mg/liter)	4	32
Cefoperazone	>128	>128
Cefoperazone + tazobactam (4 mg/liter)	8	128
Cefotaxime	>128	>128
Cefotaxime + sulbactam (4 mg/liter)	32	>128
Meropenem	0.06	0.25
Meropenem + sulbactam (4 mg/liter)	0.06	0.12
Ceftazidime + avibactam (4 mg/liter)c	0.5	2

^aTotal isolates (n = 103), E. coli (n = 50), Klebsiella spp. (n = 36), Enterobacter spp. (n = 10), Citrobacter spp. (n = 5), Serratia spp. (n = 2). These isolates were collected from various Indian tertiary care hospitals from 2015 to 2018. All the MICs were determined by the agar dilution method as per CLSI guideline M07-A12 (2018) (44). MICs of BLs in the presence of a fixed concentration (8 mg/liter) of a BLI showed similarly high MIC values, except the MIC₅₀ and MIC₉₀ of cefepime-tazobactam, which were 0.5 and 8 mg/liter, respectively.

^bThe MIC₅₀ and MIC₉₀ of ceftriaxone-sulbactam combinations in the presence of EDTA (25 mg/liter) were 16 and 32 mg/liter, respectively. The EDTA concentration of 25 mg/liter represents the C_max attained after administration of CSE 1034 (ceftriaxone at 2 g plus sulbactam at 1 g plus EDTA at 74 mg) (43).

^cOne hundred percent of isolates were susceptible to ceftazidime-avibactam based on 2019 CLSI breakpoints.

TABLE 3.

MICs of meropenem in combination with sulbactam against contemporary Indian A. baumannii isolates^a

β-Lactam or BL-BLI (dose)	MIC50 (mg/liter)	MIC90 (mg/liter)
Meropenem	32	>128
Sulbactam	32	64
Meropenem + sulbactam (4 mg/liter)b	16	>128

^aThe total number of isolates tested was 34. MICs were determined by the agar dilution method as per CLSI guideline M07-A12 (2018) (44). These isolates were collected from various Indian tertiary care hospitals from 2015 to 2018.

^bSimilarly high meropenem MICs were observed in the presence of sulbactam at 8 mg/liter and at the ratio of 2:1.

Another case in point pertains to a triple-combination product consisting of ceftriaxone (1 g), sulbactam (0.5 g), and disodium EDTA (37 mg), code-named CSE 1034. This product is marketed with the objective of tackling infections caused by metallo-β-lactamase (MBL)-expressing Gram-negative isolates highly prevalent in India (38). The premise behind formulating an EDTA-based combination emanates from the common knowledge that EDTA chelates the metal ion, such as Zn²⁺, which is required for the catalytic activity of the MBL. This very principle has been used for years to detect the MBL-expressing pathogens in clinical laboratories, which, however, involve the use of 150 to 1,400 mg/liter of EDTA (39–41). Perhaps the idea of human administration of EDTA in CSE 1034 is adapted from its use as an antidote for lead/arsenic poisoning (42). The dose of EDTA in the combination is 37.5 mg, which generates a maximum concentration (C_max) of 11.8 mg/liter, with a half-life of 1 h (43). As shown in Table 4, such a low EDTA concentration was inadequate to neutralize the MBL and restore the activity of ceftriaxone. Furthermore, the sulbactam dose of 0.5 g is unlikely to protect ceftriaxone from coexpressed class A ESBLs and class C β-lactamases. This is apparent from Tables 2 and 4, which show that ceftriaxone MICs in the presence of sulbactam remained high for ESBL/class C- as well as MBL-expressing Enterobacterales. A phase 3 study (38) evaluated CSE 1034 for complicated urinary tract infections (cUTI) involving 74 and 69 microbiologic, modified intent-to-treat patients in CSE 1034 and meropenem arms, respectively. The study showed the noninferiority of CSE 1034; however, it did not assess its efficacy against carbapenem-resistant MBL-expressing pathogens. Thus, both the nonclinical and clinical evidence of the efficacy of CSE 1034 is inadequate to support its clinical use for infections caused by MBL-expressing pathogens. In a previous publication, Shafiq et al. raised a few other concerns, such as the inadequate description of demographic and baseline characteristics of subjects enrolled in a previously undertaken phase 2 study for this combination (9).

TABLE 4.

MICs of CSE 1034 against contemporary MBL- and ESBL-coproducing Enterobacterales isolates^a

β-Lactam or BL-BLI (ratio or dose)	MIC50 (mg/liter)	MIC90 (mg/liter)
Meropenem	64	128
Ceftriaxone	>128	>128
Ceftriaxone + sulbactam (2:1)	>128	>128
Ceftriaxone + sulbactam (2:1) + EDTA (25 mg/liter)b	64	>128

^aTotal number of isolates (n = 66), E. coli (n = 23), Klebsiella spp. (n = 37), Enterobacter spp. (n = 4), Citrobacter spp. (n = 2), Serratia spp. (n = 2). These isolates were collected from various Indian tertiary care hospitals from 2015 to 2018 (44). All the MICs were determined by the agar dilution method as per CLSI guideline M07-A12 (2018) (44).

^bThe EDTA concentration of 25 mg/liter represents the C_max attained after administration CSE 1034, i.e., ceftriaxone at 2 g plus sulbactam at 1 g plus EDTA at 74 mg (43).

Apart from the lack of clinical dose justification studies for these irrational BL-BLI combinations, no dose recommendations are available for pediatric populations and for patients with renal dysfunction. This may result in underdosing or excessive dosing, leading to clinical failure, resistance development, and drug toxicity. Moreover, the clinical use of these irrational combinations in the absence of scientifically developed product-specific antimicrobial susceptibility test methods may lead to uncertain clinical outcomes. Thus, from the perspective of antibiotic stewardship, these combinations have a significant potential to amplify the irrational use of antimicrobials, leading to resistance selection, and thus compromise patient care.

In summary, this commentary highlights the scientific deficiencies in these irrational BL-BLI combinations. Indiscriminate and rampant use of such combinations may significantly compromise patient care and trigger large-scale resistance development, and therefore, regulators/infectious disease experts must assess the risks versus benefits associated with these BL-BLI combinations.

The views expressed in this article do not necessarily reflect the views of the journal or of ASM.