Preclinical Trials – A Nuanced Approach to Get Into the Clinic Faster

A Q&A with Dr. Mike McGarry


Oncology preclinical studies are commonly viewed as straightforward investigations run exclusively by researchers. Tasked with finding therapeutic interventions that stop cell growth, researchers often operate independently without true integration with the preclinical, regulatory and clinical teams.

In this issue of Peer Perspectives in Oncology, Dr. Mike McGarry, VP of Preclinical Studies at Medelis, describes a more nuanced preclinical process that can help a Chief Medical Officer (CMO) get into the clinic faster. This approach hinges on a strategic, well-defined rodent model, quality data, and ongoing integration between the preclinical, regulatory and clinical teams.

About Dr. Mike McGarry

Dr. Mike McGarry discusses 'Preclinical Trials: A Nuanced Approach'Dr. Mike McGarry is a preclinical research scientist with over 35 years of experience generating high quality animal-based data. For 30 years, he served as an investigator and Director of the Department of Laboratory Animal Resources at the Roswell Park Cancer Institute working primarily in experimental hematology and immunology with basic scientists and clinical colleagues. He chaired the Institute’s Animal Care and Use Committee and served on the Scientific Review Committee of the Institute’s Human Studies Review Board.

Dr. McGarry served as an IACUC member at both the Mayo Foundation and Arizona State University. While serving on the IACUC he helped numerous colleagues write and execute research protocols in a variety of species for experimental therapeutics, toxicology, pharmacokinetics, photodynamic therapy, immuno- and radiation-adjunct therapies, gene therapies, adoptive cell transfers and other leading edge technologies. He has himself worked with SCID-hu and nude immune deficient animals, many genetically altered (transgenic and knockout) and mutant mice in most common strains of mice. His experience extends to most other larger animal models routinely used in preclinical research.

Preclinical Trials- A Nuanced Approach to Get Into the Clinic Faster

Medelis: Mike, when a sponsor initiates a new preclinical study, what are the most important factors for a Chief Medical Officer to understand / evaluate during this step in the drug development process?

Dr. McGarry: The regulatory requirements, model selection and protocol are critical for transitioning into the application for clinical trials. A CMO who has a clear understanding of the nuances of the preclinical phase can extract more useful information that aids drug development and speeds overall progress to phase I.

Selecting the appropriate model for a particular protocol is perhaps the most important factor in a preclinical trial. Yet some CMOs don’t have a lot of experience with animal-based studies where even seemingly minor considerations – for example, having equal numbers of male and female mice – can have repercussions for getting through regulatory agencies.

Other pivotal issues include “Do we need a spontaneous animal model or can we use a passaged tumor?” and “When we administer the tumor, do we need to assess whether or not pre-treatment prior to inoculation with the tumor is efficacious, or do we only start interventions once the tumor has reached a certain size?”

Medelis: How rigorous is the regulatory assessment at the preclinical stage?

Dr. McGarry: Regulatory groups need to know that you’ve done your due diligence assessing the spectrum of adverse effects and efficacy.

For example, let’s say you used a particular tumor line in a genetically-engineered or mutant mouse. The regulatory agency might regard the data with some suspicion and say, “Well, did you do two other strains? Did you do it with human cells or an immune-deficient animal? What was the course of injections? IV? Infusion? Is that consistent with how this drug will be administered to the patient in trials?” You get the picture.

Medelis: How common is it for a company to hit a regulatory wall at this stage?

Dr. McGarry: It’s not as uncommon as one might think. CMOs are used to thinking of preclinical as one of those necessary steps you simply have to do. They often don’t acknowledge or recognize the complexities of this phase, and that can have far-reaching consequences.

Agencies understand and look for nuances. They might look at the data and say you have the wrong animal model or the wrong tumor line. Or they may say that the drug you’re working on is effective in a disease for which there is already a surgical procedure. So you had better be looking at the right animal model and even the correct placements of tumors. Does it make a difference if it is grown intrascapularly or on the flank? Would the results be more compelling if the tumor were grown orthotopically? Does it matter whether you have measured dimensions of tumor growth or do you want to be measuring life expectancy? Is survival the important parameter or is tumor growth?

Medelis: How can a CMO work with the preclinical team to gain a better understanding of these complexities?

Dr. McGarry: There needs to be ongoing, seamless communication and integration among the preclinical, regulatory and clinical teams. Communication affects the quality of results to be reported and thus the usefulness of the data. Each team should be actively involved in feasibility dialogue at each step in the study.

Ideally, you should also ensure that your preclinical team is knowledgeable about international standards so studies meet regulatory requirements for global registration through the International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use.

Medelis: So the sooner you bridge any gaps between preclinical and clinical, the better?

Dr. McGarry: Absolutely. I’ve spent most of my career in preclinical experimentation, and it’s an expensive funneling process. You start with thousands of compounds and eventually narrow it down to the one that warrants close scrutiny, hopefully making it into clinical study and application. As you’re narrowing the field, you also have to focus on the end goal or you can easily end up in a place that has no utility.

The best way to prevent this problem is to foster collaboration between the preclinical and clinical teams. Otherwise, what starts off being a crack can easily turn into a chasm.

Medelis: It sounds as if preclinical and clinical teams often operate independently in their own silos. True?

Dr. McGarry: Yes, that’s true. Preclinical researchers are extremely technical and scientific; they’re focused on mechanistic details of various tissue or organ side effects. On the other side, the CMO is pushing to move a drug forward while managing costs to meet the company’s targets. That forward momentum can create friction with the conservative, tentative nature of the research scientist who is watching the nuances unfold and has perhaps a very different perspective on the trial. Of course, both sides share the same objective, and close ongoing dialogue can help remove those barriers and prevent the silo effect from derailing a promising drug.

Medelis: Can you discuss the typical preclinical process and the questions a CMO should be asking during this phase?

Dr. McGarry: Let’s say you’re a CMO at a biotech that has been working in partnership with a small laboratory that has isolated a bio-reactive molecule from a mollusk. In vitro testing reveals a growth-arresting effect suggesting anti-cancer potential. You aren’t prepared to spend millions mapping out pathways; you just want to know the extent of the effect on tumor growth.

Since an in vitro assay only shows so much and you need an animal assay to assess organ toxicities, the exploratory would probably include a classically defined tumor model with testing over three weeks to determine efficacy or toxicity. The next step after exploratory and toxicity is tumor sensitivity, again in an animal study.

There are some tumors in which drugs stop working once the tumor gets beyond a certain size. So at each step, look at the data and ask the critical questions: “What can we learn from these results? Is this something that arrests or slows tumor growth? Is the tumor static? Is it tumoricidal? Is there a limit?”

If you see animals living longer, maybe you have something. If you’re using monoclonal antibodies directed against a particular surface antigen in a tumor cell, will that be sufficient to eradicate that tumor, or are other interventions required?

Pharmacokinetics is straightforward, but you also want to probe around tumor uptake because that can affect delivery. “Is there a preferential uptake in specific parts of the tumor or body?”

These lines of inquiry can open interesting channels. Photodynamic therapy is a good example. Photofrin® was found to preferentially accumulate in tumors, and laser light focused to the sensitivity of the Photofrin activated cytocidal activity in the drug. This led to an intervention in which surgery is combined with chemotherapy in patients treated with drug-activating laser light specifically aimed at where the tumor had accumulated.

Medelis: Are there any additional requirements unique to oncology preclinical trials?

Dr. McGarry: Yes — it goes back to selecting the right model for your preclinical data. In oncology, it’s much more critical because malignant disease is enormously complex. You need to make sure you have the animal model that best represents a particular malignancy and will allow you to tease apart various drug effects.

Medelis: What’s the typical time frame for a preclinical study?

Dr. McGarry: Timing mostly depends on the model and tumor type. Obviously the goal is to make the process as efficient as possible without sacrificing integrity and quality of the data.

There are some spontaneous transgenic tumor models where you may not see a tumor for 45 days. If you want to look at spontaneous tumors and whether your intervention has an effect on them, that could be a lengthy study with high costs including per diems on animal holdings.

Medelis: You have over three decades’ experience conducting animal studies. How important is depth of experience on the part of the preclinical investigator?

Dr. McGarry: Experience gets down to knowing the right animal model for the protocol. In some cases it may be necessary to craft two or three models. For instance, if you have an anti-angiogenesis agent, you need to use a solid tumor to determine whether the agent has an anti-tumor effect. Now that may seem straightforward, but it’s the kind of fundamental question that needs to be asked.

Experience also shows in the attention researchers give to the animals. For example, if an animal dies earlier than anticipated, does the researcher just report the statistic, or does s/he look for metastases or nuances that could provide valuable clues for the CMO?

Also, a preclinical investigator who is a published scientist can provide a major advantage in garnering the respect of colleagues. You might throw a leading question out there, for example, and they suddenly realize that you know something. Next thing you know, you’re engaged in deep dialogue. That level of expertise is extremely valuable.

Medelis: What would you look for in an outside preclinical team?

Dr. McGarry: Make sure the group understands your project and has the necessary facilities and expertise to meet your needs including the husbandry needs of various animal models, which can differ greatly. The group should have a deep understanding of the complexities of how strains and model characteristics affect the utility of certain animal tumor models. Special technical skills may be necessary that aren’t available at all units, such as gavage or infusion. And finally, they should demonstrate a thorough knowledge of protocol demands and the appropriate degree of compliance and documentation required to fulfill the FDA’s Good Laboratory Practices.

Medelis: Can you discuss GLPs for preclinical work – are they always required?

Dr. McGarry: GLP criteria are animal-specific directives addressing good record-keeping, husbandry, mixing of animals with other animals, and so on. During the regulatory approval process, yes, you must adhere to GLPs.

However, if you’re very early stage and are only assessing general compound parameters, you don’t need the costs associated with GLP-level studies. You obviously need good records, but you don’t need the same level of detail as a GLP study. This scenario is much less expensive and will still give a CMO or principal investigator enough information to determine whether a compound has potential.

Medelis: Many groups have pushed for the elimination of animal-based studies. Is that a likely scenario down the road?

Dr. McGarry: A variety of circles are advocating the elimination of animal-based studies, but the likelihood that permission to initiate clinical trials based on something other than animal-based data seems remote.

However, with genetic engineering, it’s possible to develop targeted knockouts and other genetic alterations that will answer almost any question. That’s a powerful scientific advance that helps us demonstrate efficacy as well as the appropriate hosts for pharmacokinetic and pharmacodynamic studies.

Medelis: Can computational toxicology diminish the reliance on animal models in the future?

Dr. McGarry: Computational toxicology is based on mathematical and computer models derived from historical information, predictive models and in vitro data like expression array analysis of cell cultures. It holds promise. However, it still hasn’t achieved a level of acceptance such that it could satisfy regulatory needs now met with animal data. Database information on families of molecular structures and their associated potential adverse reactions and efficacy still aren’t sufficient to exclude animal studies.

Medelis: How might personalized medicine affect preclinical work?

Dr. McGarry: Personalized medicine started in the early eighties with profiling a tumor for responsiveness to various chemotherapeutic agents. So what does this mean for using animal models? It remains to be seen. Gene profiling is going to make a big difference, but given the legal implications, we’re probably quite far away.

Medelis: So the animal model is here to stay?

Dr. McGarry: Since replacing the bacterium model of the 20th century, mouse models have become the gold standard for mammalian genomics and proteomics and, in turn, the development of therapeutics. It’s a very sophisticated system; given its wide utility and scientific power, I can’t see it being replaced anytime soon.

* * *

If you’d like to learn more, contact us.