Nature Reviews Drug Discovery, July 2015

Novel inhibitors of bacterial β-lactamases should restore activity to old antibiotics and hold back the tide of drug-resistant infections.

By Ken Garber

In February, US regulators approved Actavis’ Avycaz, a combination of ceftazidime, a previously approved cephalosporin, and avibactam, a novel β‑lactamase inhibitor (BLI). Like older BLIs, avibactam disables bacterial enzymes called β‑lactamases that break down multiple widely used antibiotics, thus restoring the activity of these drugs against drug-resistant bacteria. But avibactam also has greater potency against β‑lactamases than its predecessors, as well as activity against emerging classes of β‑lactamases that are driving the rising death toll from antibiotic resistance. Avycaz “can be a game-changer,” says biochemist Karen Bush at Indiana University in Indianapolis, Indiana, USA.

While avibactam is the first new BLI to be approved in 22 years, two other BLIs — relebactam and RPX7009 (table 1) — offer similar advantages over the older agents. Both are in late-stage trials, promising to provide a much-needed boost to the antibiotic pipeline. “A lot of companies are coming into this space because they see the potential to really extend the life of our current β‑lactam drugs,” says Robert Bonomo, an infectious-disease specialist at Case Western Reserve University in Cleveland, Ohio, USA.

The recent approval also signals how seriously the US Food and Drug Administration (FDA) is now taking the antibiotic resistance problem, because the agency approved the drug based on Phase II data. Although this is now common for oncology drugs, “they’ve never ever done it before for an antibiotic,” says David Shlaes, former Vice President for Infectious Diseases at Wyeth–Ayerst, who calls the approval “miraculous”. The FDA reforms that began with the 2012 Generating Antibiotics Incentives Now (GAIN) Act may be working.

Yet despite the optimism, the efficacy of these new BLIs against some of the most resistant strains of bacteria remains to be proven in the clinic. The combinations have as yet mostly been tested clinically in non‑inferiority trials in serious but not drug-resistant infections. “The big missing pieces of data are clinical data in patients with infections of organisms carrying the Klebsiella pneumoniae carbapenemases (KPC) resistance mechanism,” says Shlaes. “That’s something I think we desperately need to see.” The Medicines Company has a Phase III trial of meropenem plus RPX7009 in carbapenem-resistant Enterobacteriaceae infections, which should provide some first insight. The other companies are planning similar trials.

Battling resistance

Although the first β‑lactam antibiotic, penicillin, was isolated in 1929, this class is still among the most widely prescribed. Members of the class, which all contain a four-membered β‑lactam ring, act by binding to enzymes that are essential for bacterial cell wall synthesis, thereby causing cell lysis. However, bacterial β‑lactamase enzymes can inactivate the antibiotics by hydrolysing the β‑lactam ring, resulting in drug resistance.

Historically, drug developers have combatted this resistance in two ways. One approach has been to identify β‑lactam structures that are more resistant to inactivation; this led to the development of better cephalosporins and the carbapenems. The other approach has been to use BLIs to maintain the efficacy of β‑lactam antibiotics. The old BLIs, such as clavulanic acid, have β‑lactam rings themselves but little or no antibacterial activity, instead acting as ‘suicide inhibitors’ — they bind the pathogen’s β‑lactamase so that it cannot inactivate the companion antibiotic. Over time, however, the effectiveness of the BLIs has waned, because pathogens have evolved to express previously rare or unknown β‑lactamases that are mostly unaffected by the old BLIs, and because pathogens now express much higher levels of established β‑lactamases.

The β‑lactamases that represent the greatest threat are the extended-spectrum β‑lactamases (ESBLs), which are only unable to inactivate the last-line carbapenem antibiotics, and the carbapenemases, which inactivate even carbapenems. In 2013, the US Centers for Disease Control and Prevention estimated 26,000 cases of ESBL-producing Enterobacteriaceae caused 1,700 deaths, and 9,000 cases of carbapenem-resistant Enterobacteriaceae (CRE) caused 600 deaths. “Carbapenem resistance in Gram-negative bacteria is the single biggest threat in hospital-based infectious diseases today — period,” says Keith Kaye, an infectious disease specialist at the Detroit Medical Center, in Detroit, Michigan, USA.

Based on microbiological studies, the new antibiotic–BLI combinations should work against most such infections. Avibactam and its two nearest competitors, Merck & Co.’s relebactam and The Medicines Company’s RPX7009, all have excellent in vitro activity against the ESBLs and against the serine carbapenemases. Unlike the older BLIs, these drugs also work against common class C β‑lactamases, including the cephalosporin-hydrolyzing AmpC, which can be expressed by Pseudomonas aeruginosa (probably the most feared pathogen in hospitals and intensive care units) and many Enterobacteriaceae. With Avycaz, “we finally have something that we can use,” says Bonomo.

The success of the new BLIs was long in the making, however, and required a different chemical approach. The drugs originated in the mid‑1990s, when chemists at Roussel Uclaf proposed that diazabicyclooctanes (‘DABCOs’) — which include a five-membered ring rather than a β‑lactam ring—could be β‑lactam mimics with antibacterial activity. Although some compounds proved to be weak antibiotics, the chemists found others that could reversibly and covalently bind, acylate and inactivate β‑lactamase enzymes in a long-lived manner.

“What everybody was looking for was an irreversible inhibitor where the acylation would occur very quickly, and deacylation would occur very slowly,” says Shlaes. DABCOs fitted the bill, even if they didn’t bind irreversibly. Their kinetics make them far more potent than old sulfone BLIs like tazobactam. For example, 140 molecules of tazobactam are deacylated for every one that irreversibly inhibits the target β‑lactamase. By contrast, for avibactam deacylation is so slow that it only takes a single avibactam molecule to inactivate each enzyme molecule.

The DABCO structure provides another big advantage as well: a broader spectrum of activity against different classes of β‑lactamases. The β‑lactamase binding mode of DABCO is very different from that of the sulfones and clavulanic acid, allowing for greater interactions with the binding sites of both class A and class C enzymes.

Despite the promising chemistry, upheaval in industry delayed the development of avibactam. Hoechst acquired Roussel Uclaf in 1997 and shelved the drug. Hoechst then merged with Rhône-Poulenc in 1999 to form Aventis, which merged with Sanofi to form Sanofi–Aventis in 2004. Throughout, “the molecule just kind of sat there,” says Shlaes. Only when Novexel spun off from Sanofi–Aventis in 2004 did the programme come back to life. In 2010, AstraZeneca acquired Novexel and licensed the ceftazidime plus NXL104 combination to Forest Laboratories for North American development, and last year Actavis acquired Forest.

Meanwhile, Merck was pursuing its own β‑lactamase inhibitor. Looking to find a drug to combine with imipenem, a broad-spectrum carbapenem, the company synthesized AmpC-inhibiting leads. In the hunt for broader activity, they turned to the same bicyclic urea approach that had led Roussel Uclaf to avibactam. Merck’s DABCO relebactam adds a piperidine ring, which Merck principal scientist Katherine Young says helps prevent efflux out of the pathogen and boosts anti‑AmpC potency.

Functionally, “I don’t think there’s any important difference” between avibactam and [relebactam], says Shlaes.

The Medicines Company has taken a different chemical tack with its RPX7009 — a cyclic boronic acid that was discovered at Rempex Pharmaceuticals, which was acquired by The Medicines Company in 2013. Although boronic acids have been known since the early 1970s to inhibit β‑lactamases, “no one had ever published any report on anything in animals,” says The Medicines Company’s Chief Scientific Officer Michael Dudley. “This was very much unknown territory.” Fear of side effects may have held companies back. “Everybody thought that boron might be associated with toxicity, but that’s not proven to be true so far,” says Bonomo.

“They were able to break through on this, where nobody else was,” says Shlaes. The boron stably binds to the nucleophilic serine in the active site of the enzyme. “Molecular modelling was very successful at finding a highly potent inhibitor of the KPC enzyme with really the first target structure that was designed,” says The Medicines Company chemist Scott Hecker. RPX7009’s fast‑on, slow‑off kinetics are probably similar to that of the DABCOs, and so is the activity spectrum.

Diverging clinical strategies

The companies have so far taken one of two approaches in their development strategies. Whereas Novexel combined avibactam with a cephalosporin, Merck and The Medicines Company both combined their BLIs with carbapenems. Merck combines relebactam with imipenem, along with the peptidase inhibitor cilastatin, because a renal peptidase otherwise degrades imipenem. The Medicines Company is pairing its drug with meropenem.

This could have important commercial consequences. Carbapenems are generally reserved as a last line of defence against serious Gram-negative infections, because it’s vital to keep resistance from developing and because these drugs are expensive. Actavis’ ceftazidime plus avibactam combination spares the carbapenems, in line with these antimicrobial stewardship principles, and so is likely to be used earlier and more often than Merck’s and The Medicines Company’s combinations, says Shlaes. (The Thomson Reuters Cortellis database forecasts consensus annual sales of around $330 million in 2020 for Actavis’ combination, compared with around $270 million by 2020 for Merck’s combination.)

But that may not matter that much to Merck. “We were looking at imipenem life cycle management” by extending its spectrum of activity with a BLI, says Young. Merck took antimicrobial stewardship into account, she adds: “we wanted it to be a drug that would be reserved for serious infections”. Merck’s combination is likely to be more potent than Avycaz against the ESBLs, which are often co‑expressed by KPC‑producing pathogens.

For The Medicines Company’s part, Dudley says that Rempex was seeking the most powerful combination possible. “Combining with a carbapenem would bring together the strongest β‑lactam with the strongest β‑lactamase inhibitor and prevent resistance as well as could be done,” he says. He points out that cephalosporin combinations like Avycaz pressure the BLI to inhibit mulitple β‑lactamases to protect the antibiotic, potentially leading to mutations and the development of resistance. But resistance due to increased efflux or reduced cell membrane permeability is likely to evolve against all these BLI combinations, if it doesn’t exist already.

These first combinations may, however, be just be the first incarnations of the new BLIs. In Asia and Northern Europe, Gram-negative bacteria that express metallo-β‑lactamases (MBLs) have become a serious problem, and new BLIs — combined with different antibiotics — could provide much-needed help. Actavis and AstraZeneca are testing avibactam in combination with a monobactam antibiotic, aztreonam, which is resistant to many MBLs. “Any of the monobactam combinations with a [DABCO] will be a very powerful combination,” says Bush. Actavis and AstraZeneca are also testing avibactam with the cephalosporin ceftaroline, a combination that may be effective against methicillin-resistant Staphylococcus aureus.

Late-comers to the field have yet to announce their combination strategies, and a few companies, says Bush, are watching from the wings. “A lot will depend on how [Avycaz] does in the marketplace,” she says. “If it’s something that physicians will prescribe freely, then more companies will get back in. If it’s only used in very serious cases and infections, then companies may decide that they want to work on something else.”