Goin’ up the liver

Let’s talk about the liver, with the understanding that medicinal chemists and drug discovery folks in general are always ready to talk about it. We have what must be described as a wary relationship with the organ, because (thanks to the work of the hepatic portal vein) everything that makes it into the bloodstream after being ingested goes straight to the liver before it goes anywhere else. This includes our orally-dosed drug candidates, and the problem is that the liver is absolutely stocked with enzymes to tear such interloping compounds to bits. That’s ‘first pass metabolism’ in the lingo, and it can be a brutal process.

Unfortunately, the organ can be overmatched at times, and at all points during drug development you have to be watchful for signs of liver toxicity and damage. Paracelsus was right – the dose makes the poison – which means paying particular attention at the higher doses used for detecting signs of toxicity in general. ‘Liver tox’ is one of the most common red flags, and depending on the dose and the severity, it can truly kill your project right where it stands. Very rarely is there any coming back from a bad liver tox readout.

How do you know? The warning signs are elevated levels of several key enzymes such as alanine transaminase, aspartate transaminase, gamma-glutamyl transpeptidase and alkaline phosphatase. All three of these are found in other tissues, but their sudden appearance at higher-than-normal levels in the blood after dosing a new drug substance generally means just what you think it means. In addition, there are often changes in bilirubin and serum albumin, among other signs.

There’s no substitute for dosing human patients. And if you aren’t holding your breath when that happens, then you must not have been doing this work for very long

The other key indicator is whether these go back to normal after the dose of compound has cleared. The liver, in its taking-on-all-challengers function, is overall a rather resilient organ, and liver tissue can regenerate to a large extent after injury. The usual drug-induced hepatotoxicity signs go right back down, but if these enzymes and proteins stay high, you might be looking at a case of liver disease instead, or (still worse) your drug might have done even more serious damage than you realised.

That can happen, unfortunately. The classic example is the combination of alcohol and paracetamol, which is a terrible mistake. Alcohol itself is a well-known source of acute liver damage, an attack that uses up stocks of detoxifying glutathione. But the glutathione pathway is how paracetamol is usually cleared. If there’s not enough of it around because of alcohol consumption, a very reactive and toxic metabolite gets formed instead, and it causes immediate and severe damage. High doses of this combination can leave a person with no other option for survival but a liver transplant after just a few hours.

Even under more normal conditions, one obviously does not want to keep challenging the liver’s powers of recovery. That is what happens during alcoholism, and it leads eventually to the fibrotic scarring and loss of function that we call cirrhosis. So real liver toxicity, as mentioned, is generally a showstopper. Why, then, don’t we avoid it more often in drug development? The problem is that the liver’s cellular architecture is complex and (so far) impossible to completely reproduce in tissue culture. Keeping actual liver samples alive and functional has not been feasible, either, which is a general problem with so-called ‘primary tissue’. You can grow colonies of hepatocyte cells themselves, but they quickly lose much of their resemblance to a real liver (another general problem with primary cells kept under culture conditions). There’s a great deal of work going into making robust ‘organoids’ that can better mimic real organs, but that has some distance to go.

In the end, there is no substitute for dosing actual laboratory animals. And if things go well there, there is then no substitute for dosing human patients. As I like to say, if you aren’t holding your breath when that last part happens, then you must not have been doing this work for very long. There are constant surprises, as witness Eli Lilly’s recent success with a small-molecule GLP-1 agonist compound (orforglipron) while Pfizer had two of its own candidates crash out due to liver problems. You cannot let your guard down even after a drug is approved and makes it to market, either, since no clinical trial can cover the range of the patients out there in the real world. A wide variety of livers await you and your drug!