What this covers

This Business Breakdowns episode has host Patrick questioning Mark Smith, who founded OpenBiome before starting Finch Therapeutics. It begins with biology: why the microbiome holds roughly a thousand times more genes than the human genome, why the small bowel stays low in bacteria while the colon hosts them, and why germ-free animals have dysfunctional immune systems. Smith then moves to C. difficile, which takes over after antibiotics clear competing bacteria and infects about half a million people a year in the US. The first randomized fecal microbiota transplantation trial was stopped early because withholding the transplant was judged unethical.

The second half is about turning that evidence into drugs. Smith contrasts the current "shotgun" method, which reseeds a gut with a donor's whole community, with more defined, strain-level products, and he explains how reverse translation uses clinical data to choose strains. He also explains why probiotics adapted to foods like milk rarely colonize. Manufacturing cost limits how many strains a product can hold. Phage therapy comes up as a precise way to subtract specific bacteria.

The more contested claims concern disease. Smith casts the microbiome as a continuous signal to the immune system, like a dead man switch, and links its erosion to allergy and autoimmunity. He argues that antibiotic use is associated with worse outcomes on checkpoint immunotherapy, that transferring a responder's microbiome can raise response rates, and that autism, with its frequent GI symptoms, is a planned indication. The episode closes on drug development economics: low approval odds, early safety failures, roughly billion-dollar costs, and the case for mining existing data, as Foundation Medicine did in oncology, before trials begin.

Sharpest takeaway

Smith argues the gut microbiome is a regulable organ system whose disruption by antibiotics and low-fiber diets drives immune-mediated disease, and that Finch can turn fecal-transplant clinical data into standardized microbial drugs through reverse translation.

  • Microbes pay for their gut niche with metabolites that calibrate immune tone, so losing them shifts the immune system toward inflammation and allergy
  • Fecal transplant data for C. diff and other conditions can be mined to select specific strains for defined drugs
  • Manufacturing cost, not biology alone, is the main constraint that limits how many strains a product can contain

The argument · threads2 threads · 62 claims
0.85

Reliably resetting the microbiome requires whole-community transplant, not food changes or supplements.

2 pointscentrality 5/5
  • Changing food or taking supplements does not reliably reset the microbiome to a desired state, which is why a whole-community transplant is needed.

    “The tools that are available today, like changing food or supplements, don't really have the desired outcomes, resetting your microbiome.”

  • The current shotgun approach to microbiome therapy is to take an intact functional community from a healthy donor, encapsulate it in a capsule that releases in the gut, and reseed the patient's gut, accepting empirical benefit before knowing which specific strains matter.

    “we'd have an intact, functional ecosystem harvested from a healthy donor that can then be encapsulated and delivered in a targeted capsule”

0.44

Microbes' deep integration in human biology suggests broadly applicable microbiome therapies.

2 pointscentrality 2/5
  • Mitochondria inside every human cell are descended from bacteria that became embedded in our cells long ago and now help extract energy from food.

    “It's actually a bacteria that's just lived with us for such a long time that got embedded into all of our cells.”

  • Because the immune system is regulated by the microbiome and touches almost every disease, microbiome-based therapies could have broad applicability across seemingly unrelated indications.

    “Your immune system touches almost every disease and that's the common thread.”