Bruce Chabner, M.D., Former Director, DCTD
Mary Wolpert, Ph.D., Mary Wolpert, Ph.D.
Joe Tomaszewski, Ph.D., Deputy Director, DCTD
Jerry Collins, Ph.D., Associate Director, DTP

Bruce Chabner, M.D.
Former Director, DCTD

The question is… we are faced with an ethical dilemma. And that is that we are treating a lot of patients to help a few. And the question further is, "Are we making progress toward changing that situation so that we're actually designing therapies specific for patients?" Absolutely. I think it's happened with hormonal therapy. It's happened with receptor-based therapy, with the HER2 ligand receptor, for example, with rituximab.

It's happening even with small molecules, with the EGFR story, EGFR success of Tarceva and Iressa. And while its success is limited, I think it demonstrates the principle well. Now, we haven't gotten ... we haven't made a lot of progress with standard, cytotoxic drugs.

We really don't understand the determinants of success and failure in the common solid tumors. But we're getting there. We have the tools. We have the approaches, I think, that can teach us a lot, particularly the influences of pharmacogenetics. That is the host genetic disposition, how he handles the drug or she handles the drug, and how the drug damage is repaired. It maybe an inherited issue. And I think pharmacogenetics has a lot to teach us about that.

Secondly, studies of tumor selective therapies are now expedited through microarrays and through immunohistochemistry. And hopefully, we'll find patterns, if not specific molecular markers, for drug sensitivity and resistance for the cytotoxic compounds in the same way.

I think that for the time being we're destined to use both cytotoxics and targeted drugs together. Neither one is with some rare exceptions going to be adequate for treating the major solid tumors. So we really have a big challenge ahead of us in terms of designing appropriate selective therapy, particularly for the cytotoxic compounds. We're really not there yet. But as I said, I think we have the tools. And I think we will get there.

Mary Wolpert, Ph.D.
Chief, GCOB, DTP

I'm most excited about the future of chemotherapy, because we're going into a new era where we're going to be able to use biomarkers and more information that will be generated from new technology breakthroughs like all of the genomics, proteomics and other "omic" sciences.

We're going to be heading for the era of personalized medicine where we're going to be able to treat the disease and treat the patient. Already, we're beginning to see early stages of that.

It's been very gratifying to me, when I started in the field of drug metabolism, to know that early experiments in the laboratory showed that drugs were all metabolized by different enzymes. And now, we're beginning to learn that all people are not the same and I was really excited to learn about new results with the camptothecin series of drugs to learn that they now can tell that there are fast metabolizers and slow metabolizers.

And certain people who are slow metabolizers probably shouldn't receive camptothecin because their body just cannot handle the drug and they'll get toxic symptoms. More and more, we're going to find this to be true with many other drugs so that the very large randomized clinical trials that we used to do in the past are going to be changed to more rational subset analysis and also more rational selection of patients.

So I think the future is very bright and it will result from the pioneering spirit that our forebears have shown as well as a good marriage of the drug discovery teams as well as the technology that is just now beginning to appear.

Joe Tomaszewski, Ph.D.
Deputy Director, DCTD

1. If you're looking at traditional drug development when I first came into the program, in those situations, you didn't know what your target was. You may not have known, if you didn't do any PK work, you didn't know what concentrations were necessary. Because something is active in a mouse and you can attain concentrations that may be effective in a mouse, doesn't necessarily mean that in a rat, a dog, non-human primate, or the human primate, that you can achieve those same concentrations.

So if you never knew anything about kinetics in the first place and you simply went in blindly, you would be escalating up to what's called the maximum tolerated dose. So when you talk to a variety of cancer patients who have been treated with a number of the cytotoxics, they get very sick, there's a lot of nausea and vomiting, loss of hair, a lot of side effects for those drugs.

Once you start applying kinetics to the situation, now you can define what levels are necessary to have the impact on tumor that you're looking for and you can take that methodology and then apply it across the animal models that you're going to be looking at and eventually into man.

2. The whole issue in relation to PK/PD that we're trying to emphasize at this point in time is that you want a complete work up in your animal models, whether it's from an efficacy point of view, from a pharmacology point of view or from a toxicology point of view, you want to know what levels are necessary, whether you need what's called a peak level, which would be a high level of the drug for a short period of time or whether or not you need a longer exposure at a lower concentration. You want to be able to define that. So you do your kinetics in your animal models in order to define that. And you do it in your efficacy model first because you need a point of reference. You need to know what levels had impact on the tumor.

You then take that information and put it in additional animal models that you're going to look at for toxicity and see whether or not there's any toxicity associated with that. If not, fine, you can continue to move forward. Pharmacodynamics is a similar situation but now you're not simply looking at drug in plasma.

You're looking at some indicator of activity, in primarily tumor, in a particular pathway. So again, if it were in a PI3 kinase inhibitor, you'd be looking for phospho AKT, so you would have to take a tumor sample from the patient, which means that prior to drug administration, you need to take a biopsy, and after drug administration, you need to take another biopsy.

So what we're going to try to do in our PK/PD paradigm is to do real-time analyses, not take samples and put them in the freezer and wait until the 30 or so patients are finished. But we're going to try to limit the patient population to maybe six or 12.

3. You also find out whether or not there's any impact on the marker and that will then allow you to make a dose escalation adjustment for your next patient. So you're using real time analysis of data to adjust your dosing paradigm, so for what's called a cytostatic agent, which simply stops tumors from growing, that is involved with a particular pathway, in most cases, as I indicated earlier, you're going to have to take that for your lifetime.

But you don't want to have to do the lifetime clinical studies in order to find out whether you're having that impact on the tumor. So you will define in your animal models, in your efficacy models in particular, what duration of exposure is necessary to have an impact on that marker that's associated with either tumor stasis or tumor regression and that's what you're going to use as your target for what's called a Phase 0 clinical trial where you're going to do your PK/PD determinations.

So, theoretically, what you're going to gain from the whole experience is that what you so sorely want is human data. So you're going to, by going through this paradigm, you're going to be able to move into man sooner with less data, you're not going to endanger anyone because you're going to define where toxicity is in relation to the impact on your marker or the kinetics.

4. So what are the advantages of the PK/PD paradigm that we're going to shift to? The real advantage is that you're going to have both human pharmacokinetics, human pharmacodynamic results within a very short period of time that will then help you, allow you to define what your additional clinical development paradigm should be, whether or not you should quit at that point and kill the drug or whether or not you have a good chance of success.

Jerry Collins, Ph.D.
Associate Director, DTP

1. Phase 0 clinical study is the interface between all the studies that went on to drug development prior to entry in human beings in the beginning or the first in-human experience with a new molecule. Well, the idea generally is that the investigators and the drug sponsor have very specific questions that they need answered about the particular candidate molecule that they're looking at now.

So the clinical trial is designed not to do a full-scale clinical evaluation of efficacy and toxicity, but to answer very specific questions that will impact upon whether this molecule is likely to succeed or fail in later trials. To make decisions about whether to continue with this molecule or to go back to the lab and refine it before trying a successor molecule.

2. In drug development, we think of pharmacokinetics, PK, and pharmacodynamics, PD, as the underlying basis for how a drug works. During the initial human experience, we want to confirm that the pharmacokinetics and pharmacodynamics that we observed in pre-clinical non-human systems is fairly similar in humans.

So the goal of Phase 0 is to look at the PK to see its similarities and/or differences to other systems and to look at the beginnings of drug response, pharmacodynamics, and most importantly to see if the link between PK and PD is within the range that we expected based on everything that we did before entering humans.

3. Phase 0 clinical evaluations are at the beginning of a fairly long process of clinical evaluation. But the most important part are the questions that investigators and sponsors want to answer. The information obtained from Phase 0 studies must impact on decision-making. What is the most important thing that we need to know about a molecule as it enters humans? It might be a very simple question, such as whether the molecules in a pill actually get absorbed and circulate in the body. Might be a more complicated question like whether that drug substance changes the way a receptor works in a tumor.

So we have to be able to identify, articulate what is the most important question and we have to design a fairly simple clinical study that will answer that question and provide go, no go information in terms of whether we should proceed with this molecule or whether we should abandon it and replace it with another candidate.