In order for a compound to work (i.e. have an appropriate effect on its target) it needs to reach sufficiently high concentrations and engage with the target in the body for long enough to either prevent a reaction happening or to trigger downstream events which have a biological effect.

For tablets or capsules which are taken by mouth, there are various barriers to prevent this from happening.

  • The tablet reaches the stomach where it starts to degrade and release the drug particles from the binding agents and coatings which bulk out the tablet
  • Then the stomach opens and moves the acidic stomach fluid into the intestinal tract where the drug needs to be sufficiently fat-loving to get across the mucus lining the gut and through the intestinal cells so it can enter the blood stream (Absorption).
  • The presence of food in the stomach can affect the rate that the stomach churns and mixes its contents up as part of the digestive process. Also how much food passes into the intestine and how often this happens.
  • The acidic nature of the stomach contents is neutralised by bile salts introduced from the bile duct as the food travels down the intestine this can affect how much of the drug is absorbed in different sections of the intestine.

Once the drug is in the blood stream on the other side of the intestinal cells, it is taken directly to the liver where enzymes are present in the highest levels which can breakdown (Metabolise) the drugs into products which are more easily excreted in urine or faeces. These enzymes usually either

  • introduce or unmask N(itrogen) or O(xygen) atoms which (usually) inactivate the drug and make it more water soluble so it can be excreted by the kidney

or

  • add bulky substituents (conjugates) which adds considerably to its weight, allowing it to be excreted in the faeces. It usually takes ~24 hours for stomach contents to be excreted in faeces, but this time isn’t the same for everyone and can vary quite a lot depending on age and health, as can how efficiently the kidney produces urine (Excretion).

Usually only a small proportion of the absorbed drug is metabolised every time the drug passes through the liver in the bloodstream and only then does the unmetabolised drug have the opportunity to get into organs, tissues, cells and organelles (Distribution) and reach its target and have the effect that we want. If the metabolism happens too quickly, then the chances that the drug will get to the target are lower, and the drug might not have the effect that we want to treat the person effectively.

In the Drug Discovery Unit (DDU) we can measure many of the factors which affect the absorption, distribution, metabolism and elimination (ADME) characteristics of drugs. We can also predict these factors by comparing the characteristics of our new chemicals to marketed drugs which have similar properties. ChEMBL or SwissADME  are free access databases holding this type of information. They can be used (in conjunction with our results) to build local models which predict what the likely outcome of an experiment might be, before or without needing to actually do that experiment. For example, predicting the permeability of drugs at different pHs usually encountered along the GI tract or, predicting the likely metabolism rate in non-rodent species where fewer studies are performed due to the high costs associated with higher species or human trials.

We also use software packages (such as Maestro or Stardrop) to predict some of the more important chemical descriptors for virtual compounds (which we haven’t made yet) which have the right mix of properties to get the PK profile we want, before deciding which drugs to make. If we have enough existing data from a chemically similar series, we can make reasonably good predictions of different parameters, which can be used until the results are able to be measured directly.

Ideally we are trying to find drugs which are/have:

  • Fat loving so that they can get across membranes and into tissues
  • Soluble in the acidic liquids encountered in the gut and in the water portion of blood at more neutral pH.
  • Lower binding to blood cells or plasma proteins since this will affect their distribution; if they bind too strongly to blood components they will not get into tissues/cells
  • Not be metabolised too quickly by enzymes in the liver (and other organs) so that their levels remain high enough to have an effect at the target

We use this data to help us decide which of all the compounds we could make have the highest chance of having a positive biological effect in our test system. We check and profile the ADME characteristics of those compounds which have shown the best biological activity in vitro and may perform virtual mouse or rat dose studies using software to check that the drug levels are high enough to have the desired effect (by comparing to biology results). If levels are predicted to remain high enough in blood over a long enough time we would test whether the drug works in an animal model of the disease. If it does work we can  use the results to predict whether the drug would also work in a human by allometric scaling the results from animals to man, and what dose might be needed to achieve this effect. We use computational software (Gastroplus) to combine the separate effects of ADME properties for a new drug in specific tissues, based on compartmental models of how quickly blood moves through tissues in the body, and this is helping us to predict whether the drug levels are high enough in those tissues to have the biological effect.