PERL101 is one of our lead compounds for Niemann-Pick C. We discovered it in a whole animal screen for small molecules that suppress the delayed onset of adulthood in NPC mutant nematode larvae. The path from primary screening hit to mouse validation studies usually requires chemical optimization to improve compound stability and oral availability or to remove metabolic and pharmacokinetic/PK liabilities (“soft spots”) from a hit structure.
This path can be a fraught journey. One of the more dastardly aspects of making a small molecule drug is when efficacy, potency and metabolic stability are initially out of sync in a chemical series, and then wander off in different directions during optimization. One easy preemptive measure is to purchase libraries of lead-like (Rule of 3 compliant) compounds that lack reactive or unstable substituents. The hard part is implementing therapeutically relevant disease models in statistically robust assays that also anticipate preclinical development hurdles.
An advantage of cell-based screens over cell-free screens is that efficacy, and hopefully single-digit micromolar potency, are selected in tandem with other pharmaceutical properties a compound requires to successfully traverse the preclinical gauntlet, starting with cell permeability. But there’s still a gulf between cell culture and a rodent, let alone a person.
The ability of a compound to diffuse into, and not be pumped out of, cells is a wee component of the march from in vitro to in vivo. In an ideal world, a drug discovery platform outputs chemotypes that possess efficacy and potency across multiple disease models, as well as PK properties that demand little to no med chem in advance of mouse studies.
You may wonder like I do what is the probability of a cell-based phenotypic screens yielding hits with intrinsically good PK, metabolic stability and 100% oral bioavailability? Is the probability higher for organism-based phenotypic screens? We still don’t know the answers to those questions. But we do know that PERL101 has unusually good pharmaceutical properties for an unoptimized primary screening hit.
Let’s start with the first in vitro metabolic stability studies we performed at the end of the summer. We measured the stability of PERL101 and seven closely related analogs in mouse liver microsomes. This standard preclinical assay is used to determine if a compound can be oxidized by cytochrome P450 enzymes, which are responsible for drug metabolism in animals — nematodes and flies included. The eight compounds form three groups, each with related substitution patterns. PERL101 is in the high-stability group, though it’s not the most stable:
An analog that slightly outperforms PERL101 has one extra methyl group appended to it. We tested both compounds in in vivo PK experiments assessing two routes of administration — IV at 1mg/kg and PO at 20 mg/kg. PERL101 levels were nearly double the methyl analog levels after IV injection, demonstrating that PERL101 is more stable than the mouse liver microsomes results suggested:
We then subjected PERL101 to its biggest early challenge: oral bioavailability. While a drug can be injected it’s much easier to swallow a pill, ideally once a day. Usually it takes optimization of a screening hit to get reasonable oral bioavailability. Remarkably, PERL101 is 100% absorbed by the mouse gut, has a half-life over 3 hours and appears to have a large volume of distribution:
PERL101 reaches peak plasma concentrations just above at its EC50 concentration in NPC patient cells. A single-dose PK experiment in mice is incredibly informative, but not always predictive of how a drug will behave in humans. The first step in that direction is to measure compound stability on human hepatocytes, which express the cytochrome P450 enzymes that drugs will encounter in the field. PERL101 shows excellent stability:
84% of compound remained after three hours, indicating a calculated half-life of 13 hours, i.e,. once-a-day dosing. Let’s now remember that PERL101 was discovered in a nematode screen. What is it specifically about nematode physiology and drug exposure in the assay conditions that selected for a compound possessing oral bioavailability in a rodent?
Turns out nematodes have a primordial stomach and liver all rolled into one intestine, with pulses of acidity and cytochrome P450’s to boot. Thanks to a cuticle that serves as Fort Knox, compounds can only get into nematodes through the front entrance, i.e., their mouths. In the primary screen, nematode larvae swim in liquid media and imbibe compound over days. In other words, we shouldn’t be surprised that oral bioavailability in an invertebrate can be predictive of oral bioavailability in a mammal.
The time it takes to go from primary screen to mouse validation is on average 3-5 years, and cost as many millions. Orphan disease patients can’t afford to wait for the incumbent decade-long, billion-dollar, failure-prone R&D process. Accelerating new orphan drug approvals starts with rapid, lean drug discovery resulting in higher quality lead compounds out of the gate.
