Sunday, November 24, 2019

Antibiotics - Innovation or Clinical Utility?

Lately I have become intrigued with the word, innovation. Innovation is used a criterion for funding grants, for status in reviews of our antibiotic pipeline, for intelligence, imagination and many other great things. I want to explore the meaning and utility of innovation in the particular context of antibiotic discovery and development and of our antibiotic pipeline.

A story I tell frequently comes from my days as a practicing infectious diseases physician at the Cleveland VA Medical Center. In the early 1980s our surgical ICU suffered an outbreak of bacteremia and pneumonia caused by Serratia marcescens resistant to all the antibiotics we tested at the time (colistin was not on that list). But imipenem was undergoing its phase 3 testing and I was able to obtain some under a compassionate use protol that Merck had established. I am sure that we saved lives. But imipenem is really just a B-lactam – albeit a very special one.  Is it innovative? I think so.  But more importantly, its clinical utility was infinite for the care team and our patients. 

To me, innovation stands for imagination, daring, intelligence and the ability to see things beyond what we know and think we understand. To innovate is to navigate to where we have not gone before, and, hopefully, to see a way to get there. So, yes, innovation can be a good thing. In the context of antibacterial discovery, Theuretzbacher et. al. note that for many, innovation connotes a novel target or novel chemistry or novel mode of action. In the end, Theuretzbacher et. al. end up defining innovation as simply meaning that the therapy shares no cross resistance with existing therapies. 

I would like to introduce a more useful concept – clinical utility. Clinical utility clearly would include a lack of cross-resistance, but also may include other advantages.  A new therapy might avoid the need for monitoring drug levels, it might be more safe than existing treatments, it might reduce the numbers of doses required, it might be orally bioavailable, or it might avoid the need for new, experimental diagnostic tests. All of these speak to the utility of any new therapy to the physicians who prescribe it and the patients they treat. 

In my world, innovation also means risk. And risk in the pursuit of new and important therapies is fine as long as everyone understands that this is the case. Therapies directed at novel targets, those that use novel chemistry and those that exploit new modes of action all are subject to increased risk.  The risks include the risk of scientific failure early in the discovery process, the risk of failure from non-clinical safety studies and the risk of clinical failure either because of safety or efficacy issues. 

On the other hand, the use of known targets, known chemistries and known modes of action reduces risk. Sometimes, this lack of “innovation” might also lead to a lack of clinical utility.  But, historically, while this does occur, there have been many very useful but not so innovative therapies to come forward over the last several decades. The B-lactamase inhibitors recently introduced to market (avibactam, vaborbactam) have a much broader spectrum of inhibition than their predecessors. They are innovative in that they utilize new chemistries to achieve their improved spectrum. Of greater importance to me, they have increased clinical utility based on this broad spectrum of activity and still avoid most cross-resistance. A major addition to our clinical armamentarium will be aztreonam-avibactam that will have activity against Class B B-lactamases, a group of enzymes that have so far eluded the BLI-BLA strategy. At this point, one could argue that this combination is not so novel or innovative and that would be true.  But look at how clinically useful it might be. Another pipeline combination that achieves this goal is VNRX5133-cefepime from VenatoRx. In this case, the inhibitor is still based on boron chemistry like vaborbactam, but is able to assume different binding modes to inhibit class A and B B-lactamases – an innovative mode of action.

When I look at drug discovery and development plans and proposals, although I consider “innovation,” what I truly evaluate and value is potential clinical utility. These two characteristics do not always go together. I suggest that we all prioritize providing better (but not necessarily innovative) therapies to our patients with unmet medical needs as our ultimate goal.


  1. David, Spot on ! Thanks. Clinical utility is our goal, we do not innovate to innovate, we want to develop drugs and technologies that a actually solve a medical problem. It appears this is a real issue currently in the AMR space in particular but also beyond this space.

  2. Hi David, this is useful input. We will reflect on this with our expert group for the next iteration of the clinical pipeline in 2020. Peter Beyer

  3. You continue to provide thoughtful input. Thank you for it. About innovation, in cancer chemotherapy and AIDS, hypertension, and other disease areas, combination therapies have been pursued to advantage. Indeed, BLI/BLA are combinations that have demonstrated great clinical utility. Trimethoprim/ sulfamethoxazole is another winner. But isn't there more room in antibacterial therapy than these two classes?

  4. Hi David, great post. This is a critical point that is unfortunately lost on many folks in academia, industry, and government. The assumption is often made that new equals better. In particular, it is frequently assumed that inhibiting new biological targets will produce better outcomes. Why is this heuristic so widespread? Perhaps it is partly human nature. The grass is always greener on the other side of the fence, and we easily forget that change can it is often for the worse. However, the new equals better heuristic is actually quite useful for many important thereaputic areas. In the fields of oncology, autoimmunity, cardiovascular disease, and neurodegeneration, the need for new targets is self-evident because highly effective inhibitors of existing targets are not curative and often fail to even durably mitigate disease progression (this is by no means meant to imply that these drugs are not highly valuable to the patients who need them). However, this logic fails in the field of antibiotic discovery where the standard of care in most cases is a rapid and safe CURE. In the field of antibiotics, the data tells us that the current targets are great and may in fact to be the best, as folks like Lynn Silver have long argued. Patients would be better served by focusing on clinical utility as you outlined above, but changing hearts and minds will be challenging so long as "new antibiotic target" is the golden ticket to publishing in high impact journals, securing R01s, getting coverage in the lay press, and getting buy-in from industry R&D governance committees

  5. I agree completely. "Innovation" has been a bete noir for me. For one thing, it has been used to describe drug products - or candidates (either as non-cross resistant/new target/NCE - or Clinical Utility) - but I also think innovative merits being used to describe the process of drug discovery, as you have - involving the imagination, creativity, risk of going outside the box. It may not be immediately successful, but it is still innovative, to my mind, and it may ultimately prevail. More relevant to your essay - the narrow definition of non-cross resistant is not as meaningful as clinical utility. Among all the beta-lactam/beta-lactamase inhibitor combinations, there are differences that will have strong effects on clinical use. They will probably remain the mainstay of therapy - keeping up (incrementally) with resistance. As the previous commenter noted, new targets may be unlikely to be panaceas. New chemical scaffolds are more relevant. Thanks for bringing this up, Dave.

  6. Great post. Innovative and novel seem to be the buzzwords in antibiotic development. I have heard people say that novel antibiotics won't have issues with resistance. I have also heard that we can charge premium pricing and have high uptake making a viable business case. Both of these are erroneous statements. In the end we should develop to clinical utility.