Skip to main content

Nolan Breault

Wed, 02/23/2022 - 18:31

Hi Pierce,

Thanks for the summary of last week's MGR! This is my first time learning about non-inferiority study design, and I think the concept addresses a bunch of interesting themes related to the current state and future of medical research. These include how the funding of drug development is handled (you identify that the progress of health care is heavily dictated by how profitable a certain venture is -- should/how could this be altered to be more ethical?) and the "limits" of what medicine can achieve. I wanted to address this latter topic.

As I understand it, non-inferiority studies have gained traction in the last few decades due to incrementally smaller and smaller improvements in standards of care and outcomes -- a sort of diminishing returns scenario (while the discussion's centred around antibiotics, it would be interesting to quantify the prevalence of non-inferiority trials in other major markets of drug research, like cardiovascular and depression/anxiety medications, chemotherapeutics, etc). While I have a somewhat better-than-lay understanding of the drug development process for antibiotics, I'm not sure that I've grasped the reasons behind why our ability to develop new therapies is particularly limited. With drugs for non-infectious disease, we're typically dealing with a static adversary. That is, if we can identify what's causing a patient's ailment, the disease doesn't reflexively adapt and counteract whatever medication we prescribe (outside of some cancers). Thus, hypertension, anxiety, etc, can be managed for decades on one or two therapies. As such, the number of ways we can design an effective seems clearly finite (until precision medicine and patient-specific treatments become mainstream). We might, as you described, find ways to improve quality of life/compliance by converting an intravenous medication into an orally-administered one, but there isn't a tug-of-war with the disease.

Contrast this with a bacterial infection, where there's an entirely independent genome that experiences natural selection just like we do. As I understand it, most of our antibiotics are derived from molecules that we find in nature, as they exist as part of other organisms' immune systems. There's inherently then an arms race between the ability of bacteria to infect and propagate within a host, as well as the host's ability to stave off the infection. These forces are underlaid by proteins, which have defined structures and functions. If we throw in the capacity for favourable mutations to occur and become common in a strain of bacteria, as well as the host, could we not theoretically have many more potential molecules to draw antibiotics from? Further, many medicines are compounds that don't occur in nature, so perhaps we could use synthetics as antibiotics as well (further reading in the Nature review below).

Looking forward to everyone's thoughts on this!


Mitcheltree, M.J., Pisipati, A., Syroegin, E.A. et al. A synthetic antibiotic class overcoming bacterial multidrug resistance. Nature 599, 507–512 (2021).

Nolan Breault

Plain text

  • No HTML tags allowed.
  • Lines and paragraphs break automatically.
  • Web page addresses and email addresses turn into links automatically.