The wobbly status of ketolides: where do we stand?

Abstract

Ketolides are erythromycin A derivatives with a keto group replacing the cladinose sugar and an aryl-alkyl group attached to the lactone macrocycle. The aryl-alkyl extension broadens its antibacterial spectrum to include all pathogens responsible for community-acquired pneumonia (CAP): Streptococcus pneumoniae, Haemophilus influenzae, Moraxella catarrhalis as well as atypical pathogens (Mycoplasma pneumoniae, Chlamydia pneumoniae, Legionella pneumophila). Ketolides have extensive tissue distribution, favorable pharmacokinetics (oral, once-a-day) and useful anti-inflammatory/immunomodulatory properties. Hence, they were considered attractive additions to established oral antibacterials (quinolones, β-lactams, second-generation macrolides) for mild-to-moderate CAP. The first ketolide to be approved, Sanofi-Aventis’ telithromycin (RU 66647, HMR 3647, Ketek®), had tainted clinical development, controversial FDA approval and subsequent restrictions due to rare, irreversible hepatotoxicity that included deaths. Three additional ketolides progressed to non-inferiority clinical trials vis-à-vis clarithromycin for CAP. Abbott’s cethromycin (ABT-773), acquired by Polymedix and subsequently by Advanced Life Sciences, completed Phase III trials, but its New Drug Application was denied by the FDA in 2009. Enanta’s modithromycin (EDP-420), originally codeveloped with Shionogi (S-013420) and subsequently by Shionogi alone, is currently in Phase II in Japan. Optimer’s solithromycin (OP-1068), acquired by Cempra (CEM-101), is currently in Phase III. Until this hepatotoxicity issue is resolved, ketolides are unlikely to replace established antibacterials for CAP, or lipoglycopeptides and oxazolidinones for gram-positive infections.

Keywords::

• cethromycin
• Chlamydia pneumoniae
• Chlamydophila pneumoniae
• community-acquired pneumonia
• cost-benefit
• Haemophilus
• hepatotoxicity
• Ketek
• ketolides
• Legionella
• macrolides
• methicillin-resistant Staphylococcus aureus
• modithromycin
• Moraxella
• Mycobacterium
• solithromycin
• Streptococcus
• telithromycin

1. Introduction and overview

Bacterial resistance to antibacterial drugs has been increasing relentlessly over the past two decades driving the need for new compounds [1,2]. While nosocomial infections have received most of the publicity [3], common, community-acquired infections, pneumonia in particular, can be almost as deadly. There are currently three antibacterial classes for community-acquired pneumonia (CAP): quinolones (levofloxacin, moxifloxacin, gemifloxacin), β-lactams (amoxicillin–clavulanic acid combination) and second-generation macrolides (clarithromycin, azithromycin) [4].

Ketolides are third-generation macrolides designed specifically to act against macrolide-resistant, respiratory pathogens. They are semisynthetic derivatives of erythromycin A with the cladinose sugar at C-3 on the 14-membered lactone macrocycle removed and the resulting hydroxyl oxidized to a keto group (hence the name, ketolides). In addition, they usually have a cyclic carbamate group attached to C-11 and C-12 of the macrocycle and an aryl-alkyl chain, usually attached to the cyclic carbamate group (Figure 1) [5]. The aryl-alkyl extension broadens the antibacterial spectrum and improves activity against most macrolide-resistant respiratory pathogens [5]. In addition, ketolides have improved pharmacodynamics, pharmacokinetics (oral, once-a-day) and tissue distribution [6]. These properties established ketolides as potential contenders for the large but competitive market of mild-to-moderate CAP.

Figure 1. Structures of the 14-membered macrolide erythromycin and its 6-O-methyl derivative clarithromycin, the 15-membered azilide azithromycin and the ketolides telithromycin, solithromycin, cethromycin and modithromycin.

In the mid-90s, ketolides became the focus of intense research activity at several companies (Aventis, Johnson and Johnson, Abbott, Merck, Shionogi, Taisho, Pliva, Rib-x, Chiron, Enanta, Optimer) and academic groups [7]. Ketolides were even attached to quinolones, the latter serving as a ‘silent partner’ in the hybrid molecule, affecting pharmacokinetic properties rather than antibacterial activity [8]. Fueling the research activity was a renaissance involving the bacterial ribosome thanks to the advent of oxazolidinones (linezolid), glycylcyclines (tigecycline) and streptogramins (quinupristin-dalfopristin) [9-12].

Research activity slowed to a trickle after the tainted clinical development and controversial FDA approval of Aventis’ telithromycin (RU 66647, HMR 3647, Ketek®) [13-16]. Nevertheless, three additional ketolides progressed to non-inferiority clinical trials for mild-to-moderate CAP:

• Cethromycin (ABT-773, A-195773, Abbott/Taisho), licensed to Polymedix and subsequently to Advanced Life Sciences [17,18]. It completed Phase III non-inferiority clinical trials, and a New Drug Application was submitted to FDA (US) in 2008. It was denied approval in 2009 and has an orphan drug designation for tularemia, plague and anthrax prophylaxis. With only 2 years left in its patent life, its prospects are uncertain.

• Modithromycin (EDP-420, EP-013420, Enanta); a 6,11-bridged bicyclic ketolide, originally codeveloped with Shionogi (S-013420) and subsequently by Shionogi alone [19]. It is currently in Phase II clinical trials in Japan.

• Solithromycin (OP-1068, Optimer), licensed to Cempra (CEM-101). It is currently in Phase III clinical trials [20].

2. Mechanism of action, antibacterial spectrum, resistance

Like macrolide antibiotics, ketolides affect bacterial protein synthesis. The macrocycle of both macrolides and ketolides interacts with the peptidyltransferase center (PTC) on the 50S ribosomal subunit (proteins L4 and L22), partially blocking the polypeptide exit tunnel of the ribosome [21-24]. The sugar substituent in both subclasses interacts with A2058 of 23S rRNA, while the alkyl-aryl arm of ketolides interacts with A752 on domain II and U2609 on domain V of 23S rRNA. As would be expected from the identical orientation of the alkyl-aryl arm, telithromycin and solithromycin bind to PTC in an identical fashion, while cethromycin binds somewhat differently. The two-site binding of ketolides in the bacterial ribosome preserves binding even when A2058 is methylated by erythromycin-resistant methyltransferases (Erms). Hence ketolides maintain activity against most erythromycin- (and penicillin-) resistant isolates of Streptococcus pneumoniae.

Ketolides (telithromycin, cethromycin, modithromycin, solithromycin) have excellent activity against gram-positive aerobes (MIC₉₀s, < 0.05 mcg/ml) including macrolide-resistant (mefA and ermB) strains of S. pneumoniae [25-31]. They also have good-to-moderate activity against some gram-negative aerobes such as Haemophilus influenzae and Moraxella catarrhalis (MIC₉₀s < 8 mcg/ml), and excellent activity (MIC₉₀s, < 0.5 mcg/ml) against atypical/intracellular CAP pathogens C. pneumoniae, Legionella pneumophila, Mycoplasma pneumoniae [32,33].

While rare, ketolide (telithromycin)-resistant strains have been isolated worldwide [34,35]. Inducible resistance to both ketolides and macrolides is by energy-dependent drug efflux and through rRNA methylation [36-38]. Efflux is particularly important in gram-negative resistance [39,40]. Interestingly, the molecular details of resistance induction by ketolides differ from those by macrolides; ketolides stimulate a novel ribosomal frameshift [41].

Unlike macrolides, which are considered time-dependent killers, ketolides show concentration-dependent killing [41]. Due to their cidality, they have a long (3 – 6 h) post-antibiotic effect, both properties stemming from their mechanism of action: derailing, rather than stalling protein synthesis.

Like macrolides, ketolides also inhibit the synthesis of proinflammatory cytokines (TNF, IL-1β, IFNγ) and have anti-inflammatory/immunomodulatory properties that may be beneficial to CAP patients [42-46].

3. Pharmacokinetics, metabolism, clinical trials

Ketolides have extensive tissue distribution relative to serum and excellent pharmacokinetics allowing once daily dose administration [47-49]. Similarly to the parent compound erythromycin, they are primarily metabolized in the liver by the cytochrome P450 enzyme system, specifically the CYP 3A4 superfamily. They are eliminated by a combination of biliary, hepatic and urinary excretion. CYP3A enzymes metabolize many common drugs, such as atorvastatin (Lipitor®), and can be induced or inhibited by certain drugs, such as dexamethasone (Decadron®). Thus, ketolides have a high potential for clinically significant drug−drug interactions. In addition, they (telithromycin) decrease protein levels, and thereby activity, of CYP1A2 and CYP3A2, the result being decreased metabolic clearance of the common asthma drug theophylline [50].

Clinical trials of ketolides have focused on respiratory infections, primarily CAP. Telithromycin was found to be non-inferior to trovafloxacin and clarithromycin [51,52], cethromycin to clarithromycin [53] and solithromycin to levofloxacin [54]. There is no clinical information on modithromycin, and its development status is uncertain.

4. Adverse effects, future prospects

Like first- and second-generation macrolides, ketolides produce mild gastrointestinal effects: diarrhea, nausea, abdominal pain, vomiting. Some (telithromycin, cethromycin) also cause QT interval prolongation that could lead to irregular heart rhythm in patients with preexisting conditions. The cardiac effects, while serious, are predictable, and at-risk patients can be identified and given alternative treatments. Some (telithromycin) reportedly inhibit nicotinic acetylcholine receptors thereby blocking neuromuscular transmission and exacerbating myasthenia gravis [55]. There is also a single report of an anaphylactoid reaction in a patient, not allergic to erythromycin or azithromycin, after treatment with a single dose of telithromycin [56]. Other potential adverse effects, alluded earlier, stem from drug−drug interactions with common drugs due to ketolide metabolism by the cytochrome P450 enzyme system (CYP 3A4).

The most serious adverse effect of ketolides is rare, but irreversible hepatotoxicity. It has been extensively documented for the first ketolide to be approved, Sanofi-Aventis’ telithromycin (RU 66647, HMR 3647, Ketek) [57-59]. Telithromycin had tumultuous clinical development, controversial FDA approval in 2004 and subsequent restrictions (in 2007 and 2010) due to rare but severe, irreversible hepatotoxicity that included deaths [14-16]. This very serious adverse effect of telithromycin, discovered after FDA approval, is rare, idiosyncratic, but acute and potentially fatal. It appears to have a distinct clinical signature: rapid onset with no early signs. It thus contrasts to hepatotoxicity from other drugs, acetaminophen (Tylenol®) for example, which is predictable, dose-related and reversible, hence manageable. Hepatotoxicity has not been reported for cethromycin, solithromycin or modithromycin, but safety data are very limited.

5. Expert opinion

The issue of hepatotoxicity continues to cast a shadow over ketolide prospects. Its etiology, or whether it is a class effect, is unclear. It could be due to the imidazole in telithromycin’s structure, in which case it would be less of an issue in solithromycin, which has a triazole instead. Replacing imidazole in the antifungal ketoconazole with a triazole in itraconazole reduced – but did not eliminate – its hepatotoxicity [60]. Specific ketolide metabolites (the alkyl-aryl arm?) may also be responsible, but studies are needed to explore this possibility. Since hepatotoxicity is a rare event, genetic variability in liver sensitivity and/or ketolide metabolism may be key predisposing factors. Until the issue of hepatotoxicity is resolved, ketolides are unlikely to replace established, mostly generic, antibacterials in the treatment of CAP (Table 1). They are also unlikely to replace lipoglycopeptides and oxazolidinones for gram-positive infections.

Table 1. Key properties, issues and prospects of ketolides.

• Ketolides are oral, once-a-day macrolide derivatives with improved antibacterial activity over erythromycin against resistant respiratory pathogens and comparable efficacy to clarithromycin in non-inferiority clinical trials for community-associated pneumonia

• Rare – but potentially fatal – hepatotoxicity is a critical issue, documented extensively for telithromycin. It has not been reported for cethromycin, modithromycin or solithromycin, but safety data are limited. Hence, it is unclear whether hepatotoxicity is a ketolide class effect

• In the absence of prognostic tests identifying patients at risk for hepatotoxicity, this issue weighs heavily on the prospects of ketolides for mild-to-moderate pneumonia where safe and effective generic alternatives exist

In conclusion, the potential of ketolides, particularly solithromycin that is structurally very similar to telithromycin, hinges on the hepatotoxicity issue. To move forward, this issue needs to be investigated and resolved: risk factors identified and biomarkers for vulnerable patients developed. Once at-risk patients are identified and excluded, ketolides can be added to existing, safe and effective treatments for CAP [61]. Otherwise, in the circumspect post-telithromycin regulatory climate [62], this once promising macrolide subclass will be reserved for serious, problematic gram-positive infections for which there are few therapeutic alternatives.

Declaration of interest

NH Georgopapadakou is an independent consultant with NHG Preclinical Research Consulting, LLC and has declared that she has no conflict of interest and received no payment in preparation of the manuscript.