Non-Clinical Development r Discovery - Identification of disease targets and potential therapeutic compounds:
c The most variable and least successful aspect of drug development r Non-Clinical Development - Translates discovery science into therapeutic candidates:
o Involves modifications, scale-up, purification, test methods and production
Focusing more closely on discovery, we see that there are essentially five main steps; target identification and validation, and lead identification, optimization, and validation. Figure 4.3 provides a description of these five steps. To illustrate these steps, we'll consider two different examples, a small molecule to treat AIDS and an antibody to treat psoriasis. The disease AIDS (acquired immunodeficiency syndrome) is caused by infection with HIV (the human immunodeficiency virus).
Once HIV infects cells, it produces several enzymes that are required for the replication and propagation of the virus, one of which is reverse transcriptase. This enzyme uses the viral RNA as a template and makes DNA copies of the viral genome, which then enter the nucleus where host cell enzymes are used to many more copies. Thus, because reverse tran-scriptase is a viral-specific enzyme, inhibiting the activity of this enzyme could reduce the spread of AIDS. This makes reverse transcriptase a potential "target" for therapeutic intervention .
In the early 1980s, a number of nucleotide analogues were being studied as potential anticancer therapeutics and, because these structures mimic the building blocks of DNA and RNA, many were subsequently screened for their ability to inhibit reverse transcriptase ("hits" and "leads"). One of these analogues was AZT (azidothymidine), which was found to be an effective inhibitor of reverse transcriptase (a "lead") and, when tested in patients, inhibited replication of HIV. AZT was therefore a "candidate" that became a "product."
Psoriasis is an autoimmune disease characterized by activated immune cells. Normally, the immune system acts as an internal security system, protecting the body from infection and injury. With psoriasis, however, T cells become overactive. This activity sets off a series of events that eventually make skin cells multiply so fast, they begin to pile up on the surface of the skin, forming characteristic plaques (red, scaly patches on the surface of the skin). Thus, agents that interfere with the function of T cells could reduce the signs and symptoms of psoriasis (indeed, topical steroids are used extensively), and as such are hits, leads, and candidates, depending on their stage of evaluation.
Clearly, many systemic immunosuppressive agents have been identified (cyclosporin A, methotrexate, etc.), and most provide benefit to patients with psoriasis. However, many of these agents are also quite toxic, making prolonged use difficult. As a consequence, alternative ways to interfere with the activation of T cells have been explored, and several T-cell surface structures were believed to play critical roles in this activation process. Among these structures, one (LFA-1, or lymphocyte function-associated antigen 1) appeared to be involved in T-cell activation, function, and trafficking to sites of inflammation, a new "target." Antibodies were therefore raised against human CD-11a (a subunit unique to LFA-1) and tested in vitro and in animals. These antibodies were hits. Of the antibodies that were generated, several effectively inhibited a number of T cell-mediated functions in vitro and also showed efficacy in animal models of autoimmune disease, hence "leads" . Based on these data, one antibody (MHM24) was optimized and became a candidate for human use by "humanizing" it , a process that strives to reduce the chances of generating an immune response by converting a mouse antibody sequence into a sequence commonly found in humans. The resulting antibody, termed Raptiva®, has been shown to be safe and effective in treating patients with moderate to severe psoriasis , resulting in its approval as a product.
Note that, although it is desirable to have all these elements completed prior to filing an IND, they all may not be required to do so. Some of the factors that influence how much information is needed to file an IND include (i) the clinical indication, (ii) the nature of the compound (small molecules vs. biologics), (iii) the specificity of the compound, (iv) the availability of appropriate animal models, and (v) the seriousness of the disease. Thus, small molecules and biologics for use in cancer (or other life-threatening diseases) may require less nonclinical information to file an IND than therapeutics designed for chronic or more benign diseases. Similarly, small molecules often require a more detailed safety package than biologics, in part because the later agents are often human proteins that have fairly predictable actions and degradation and clearance properties. The requirements for filing an IND are also influenced by whether the agent only interacts with a human target (and thus animal studies may be less predictive) and whether suitable animals exist for appropriately testing the new therapeutic.
The term validation has shown up several times now and is worth additional discussion as it is frequently misunderstood (Fig. 4.4). Most commonly, the term validation is used to demonstrate that a particular assay or process is well controlled and reproducible. Thus, for a company manufacturing a recombinant therapeutic protein, they must demonstrate that the fermentation process, purification process, and assays used to test the activity of the product produce similar results each time they are performed (i.e., that they are reliable). The steps involved are therefore called "process validation" and "assay validation." Such validation typically involves the preparation of standard operating procedures (SOPs) that describe in detail precisely how the process or assay is to be conducted, as well as having one person repeat the assay several times and then several people repeating the assay. Only when the results of all these assays are reproducible will that assay be considered "validated."
In contrast, validation is also now being used to support the potential validity of new targets or products. For example, an investigator might say they have identified 100 "validated" targets, by which they mean to imply that a clear linkage has been demonstrated between the presence or absence of this target and the disease in question. Whereas there can be value in these data, there is as yet no clear definition of what "validated" means when applied to new targets and potential products— some have used the term to indicate that a particular target is always absent on normal tissues but is always present in every r "Validation" has been most commonly used in biotechnology and pharmaceutical industries to reflect level of control and reproducibility for an assay or process:
o There are FDA guidelines on process and assay validation r More recently, it has been used to "suggest" that certain therapeutic targets or products are more likely to be successful than others:
o In reality, some "validated" targets or products may be weakly supported by limited in vitro data, or they may be strongly supported by knockout and disease models r Only "validated" targets are those for which clinically successful therapeutic products have been generated r Only "validated" products are those with several hundred million dollars in sales
Fig. 4.4. Validation - Frequently Misunderstood diseased tissue (a good idea), but others have used the term to indicate that certain targets are simply upregulated in a few diseased tissues (not so good). As a consequence, many people feel that the only true "validated" targets are those for which clinically successful therapeutic products have been generated (3-hydroxy-3-methylglutaryl-coenzyme A reductase, COX-2, erythropoietin receptor, CD-20, etc.), and that the only true "validated" products are those with several hundred million dollars in sales (Lipitor®, Epogen®, Rituxan®, etc.). Though validation is an important component of the product development process, it is critical to keep these distinctions in mind when listening to claims for new targets!
These key questions for discovery help guide early choices during the development process for targets (five questions) and for products at the lead stage (six questions). The target questions focus on relationships of the target with the disease and how changes in the target impact the disease (Fig. 4.5).
r Target validation:
o What does target do?
o What role does target play in the disease?
o How specific is target for the disease?
c If I inhibit target, is there an impact on the disease?
c If I inhibit target, what other effects are there (toxicity, etc.)?
r Product validation:
o How well does product work in vitro and in vivo?
o How selective is product for the target?
c How stable is product (does it break down)?
c How long is product available in vivo?
c Where does product go after administration?
o How toxic is product?
Fig. 4.5. Key "Validation" Questions
The lead questions relate to an early profile of the potential product prior to human use and hopefully suggestive of human activities for the lead. Product characteristics include activity, stability, distribution, persistence in vivo pharmacokinetics, and toxicity.
Traditionally, drug development has been viewed of as a linear, stepwise process involving a series of sequential decisions that are based on the disease, the target and the desired product properties, such as the five steps noted in the Figure 4.6 . However, as is evident from our earlier examples, this can be a long (6-12 years) and expensive process (millions of dollars per lead) that does not follow a sequential path and yields many more failures than successes.
As an illustration, let's consider the case of Lipitor, a cholesterol-lowering product that had $10.3 billion in worldwide sales for 2003 [6-9]. In the early 1980s, clinical data were accumulating that suggested a linkage between high serum levels of cholesterol and increased risk of heart attacks and stroke. Beginning in 1982, scientists at Parke-Davis (now part of Pfizer) began looking at a class of compounds called statins, which are fungal products that block cholesterol synthesis at a key step (3-hydroxy-3-methylglutaryl-coenzyme A reductase, or HMG-CoA reductase). At the time, it was unknown whether lowering plasma cholesterol levels would be beneficial and, if so, whether it could it be done safely. Thus, a clinical need appeared to exist for therapeutics that could lower serum cholesterol levels, the biosynthetic enzyme HMG-CaA reductase was a reasonable target, and statins represented an initial class of lead compounds. The particular challenges here, however, were to develop a compound that had statin activity, was safe, had potential patient benefits, and could be easily manufactured.
In 1985, a compound was developed (CI-981) that appeared to meet most of the requirements. It still had limitations, however
Transition to development
Was this article helpful?
Do You Suffer From the Itching and Scaling of Psoriasis? Or the Chronic Agony of Psoriatic Arthritis? If so you are not ALONE! A whopping three percent of the world’s populations suffer from either condition! An incredible 56 million working hours are lost every year by psoriasis sufferers according to the National Psoriasis Foundation.