ADME/Tox: Creating a Safer Drug-Discovery Pipeline


With estimates of the cost of bringing a new drug to market topping 2 billion dollars, failure is not an option. Yet most drugs, even with proven efficacy, do fail to make it to market, and that is arguably a good thing. There is a bevy of screens and assays—collectively known as ADME/tox—designed to make sure those potential pharmaceuticals are behaving the way they should at various stages of the pipeline, before they’re allowed to progress.

Many of these, such as the Ames assay for genotoxicity, have been around for years. Some are high-throughput screening techniques to weed out “non-drug-like” molecules, whereas others amount to regulatory “box-checking.” Here we look at some trends in the field, and how they fit into the bigger picture.

What’s ADME/tox?

ADME stands for absorption, distribution, metabolism and excretion. In other words, it’s the study of what happens to a parent compound in the body. All the aspects of ADME are reflected in the pharmacokinetic (PK) profile, points out Maralee McVean, vice president of pharmacology and toxicology services at Pre-Clinical Research Services. Consider, for example, an oral medication. Its absorption through the gut is reflected in the amount of the parent compound found in the circulation. As the compound is metabolized—by the liver, for example—it starts to dissipate from the bloodstream. It will also disappear as it makes its way into various tissues or is excreted through the urine. “There are standard parameters you derive from your PK curves – Cmax, Tmax, AUC [area under the curve], elimination rates and clearance,” says McVean.

ADME—especially at the later stages of drug development, when the field has been winnowed down to a small handful of candidates, and the data is being gathered for regulatory filings—still consists of very traditional clinical and pathological assessments. “It’s a fairly well-prescribed set of endpoints that you’re expected to monitor and report on,” McVean explains.

What’s emerging is in the area further up the pipeline—in the course of drug discovery and lead optimization—during which the functions and performance of hundreds of different compounds are tested in vitro. The line between ADME and toxicology starts to grey when microtiter-based cell systems are used. “We still call it ADME, but the best term is in vitro toxicology,” says Scott Hickman, global market segment manager for BioReliance. “You can literally take a liver cell line and test for hepatotoxicity in a targeted way, whereas a few years ago we didn’t have that ability. We’re finding new cell lines and new ways to develop cell lines that are closer to the indications that researchers would see in a human population.”

Lines and fractions

The first organ an oral medication hits is the liver, and that’s usually the culprit of the metabolic fate of the compound, says McVean. First-pass in vitro ADME assays are typically performed on liver slices, hepatocytes or cell lines, or perhaps on cell-free derivatives of these (such as the cytosolic fraction, microsomes or the S9 fraction, which contains both the cytosolic and microsome fraction). Fresh and cryopreserved primary liver cells and cell lines are available from a variety of vendors, prepared from a variety of species. Each of these has its drawbacks—liver slices are not amenable to high-throughput screening; primary hepatocytes are difficult to maintain in culture; cell-free fractions lack much of the cellular machinery that can contribute to a robust assay; hepatoma-derived cell lines may display abnormal repertoire and levels of enzymes—and thus various 3D and suspension, co-culture, immortalization and other strategies have been devised and in some cases commercialized [1]. To mimic the effects of chronic exposure, longer-term cultures are necessary.

All the caveats apply, as with any cell-culture experiment, in that not all the players that would be affecting the compound in the in vivo experience are going to be involved, McVean says. Still, “you throw the compound into your cell culture … and you can get a pretty good idea what’s going to happen in vivo, and how quickly and how extensively a compound is going to get metabolized by the liver.”

This is most often done using mass spectroscopy (MS). A tandem MS platform (LC-MS-MS) with the appropriate acquisition and analysis software lets researchers kill two birds with one stone—not only quantitating the parent compound but also identifying its metabolites, says Suma Ramagiri, global technical marketing manager at SCIEX . This enables the medicinal chemist to correlate specific structures, such as the presence or placement of functional groups, with metabolism or toxicity found in the assay.

Many different cytotoxicity assays also can be run in microtiter plates, and there have been a lot of innovations in looking at multiple parameters of cell toxicity. Rather than just asking whether cells are alive or dead, researchers are looking at ATP usage, apoptosis, various DNA structural changes, mitochondrial pore formation and calcium fluxes. “You can look at a lot of different endpoints, which gives you an idea of what processes are going on before the cell is dead. That helps you figure out what the potential liabilities of your compounds are, so you kind of know what you need to fix,” McVean says.

In Vivo imaging

Beyond the standard battery of fluorescent assays, high-content imaging systems can be used to examine parameters such as texture and morphology independently of any assumptions about the etiology of observed phenotypic changes, even absent any label.

Yet, as with any in vitro toxicity assay, “the one thing you don’t often catch in cells is the metabolic aspect of the liver—liver enzymes may metabolically alter drugs to become toxic” to other organs, points out Jeffrey Peterson, director of Applied Biology at Perkin Elmer.

In vivo near infrared fluorescence imaging—primarily optical tomography—has been used to track the biodistribution of large molecules for cancer studies, for example, but has only in the past year and a half or so gotten any traction for toxicology. “Many of the same types of biology that we image for efficacy are actually involved, to some degree, in adverse effects induced by drugs,” Peterson says. “We started looking to see if any of these agents make sense for looking at early-stage screening of drug-induced liver injury,” often without needing to make prior assumptions about mode of action. For example, PerkinElmer has performed research combining Annexin-Vivo™ 750 (an indicator of cell death) with an MMP [matrix metalloproteinase]-activatable agent (that picks up inflammation) and a transferrin receptor-detecting agent for the liver (that picks up metabolic perturbations). This cocktail, in turn, can be combined with an agent used to detect vascular leaks. The company has also developed imaging agents for other indications. For example, GFR-Vivo™ 680 allows quantitation of drug-induced changes in kidney glomerular filtration rate by imaging fluorescence changes in the blood in the heart.

In addition to seeing how agents localize in a live small animal (in the case of tomography, in three dimensions), “where it used to take many, now scientists can accomplish the same results using just three to six animals with our solutions because they are doing the imaging longitudinally and can keep re-imaging the same animals,” Peterson remarks.

Get real

Some divisions of the U.S. Food and Drug Administration have started to ask that ADME/tox studies be run on a background of animals that are experiencing the disease state that the clinical human population would be experiencing, says McVean. “They’ve noticed that when they go into [a] severely diseased population, you don’t always get the same outcome.”

From gathering of early stability data to present to investors, to demonstrating the safety of a compound before filing an IND, the variety of in vitro, in vivo and ex vivo assessments that make up ADME/tox are crucial to every stage of the drug-discovery pipeline. As models become more sophisticated and tools become both more nuanced and more powerful, these assays should increasingly predict which compounds will flow though that pipeline.


[1] Ukairo, O, et al., “Long-Term Stability of Primary Rat Hepatocytes in Micropatterned Cocultures,” J Biochem Mol Toxicol, 7:204-12, 2013. [PMID: 23315828]