Improving pre-clinical experimental models to test anti cancer drugs

Share :
Published: 21 Jul 2011
Views: 6506
Dr Robert Kerbel - Sunnybrook Research Institute, University of Toronto, Canada

Dr Robert Kerbel talks to ecancertv at the Worldwide Innovative Networking in personalized cancer medicine (WIN) meeting, initiated by the Institut Gustave Roussy (France) and The University of Texas MD Anderson Cancer Center (USA) Paris, 6-8 July, 2011.

Dr Kerbel discusses the frustration of pre-clinical models not delivering later on down the line, when promising molecules fail in the clinic.

He has approached this problem by developing models for treating tumours that are metastatic, rather than primary tumours. These include advanced breast cancer, malignant melanoma, liver cancer, and ovarian cancer. These models have been used to test anti-angiogenic drugs and metronomic chemotherapy.

WIN 2011, 6-8 July, 2011, Paris

Improving pre-clinical experimental models to test anti-cancer drugs

Dr Robert Kerbel – Sunnybrook Research Institute, University of Toronto, Canada

The topic that I addressed involves, I would say, one of the most frustrating and enduring problems in cancer research. What I’m referring to is the use of pre-clinical experimental models for testing various anti-cancer drugs; these models generally involve mice, they have for over fifty years and what frequently happens is that investigators evaluating a new drug or a new therapy or a combination of drugs, not infrequently report obtaining extremely exciting results in mice. When obtaining these results, you can appreciate that that could be a major factor in whether to move forward into the much more expensive and complicated process of clinical trial testing of that drug or that therapy. Clinical trials are very expensive, especially when you get to the more advanced phase III clinical trial. In oncology there is a tremendously high attrition rate, things that often work in these pre-clinical models fail completely when they are evaluated in the clinic and this is one of the major reasons why anti-cancer drugs are so expensive, so that companies want to recoup the losses that they incurred as a result of all the previous failures. It also means that large numbers of patients are being exposed to therapies and drugs that are really of no use, were thought to be maybe useful based on the pre-clinical work that had been done.

As a result of this there has been a great deal of debate and discussion about how to improve pre-clinical models so that they become better at predicting what might happen in the clinic; at least that if you get a positive result pre-clinically you’ll get some kind of positive result, it may not be as good but some kind of positive result rather than a complete failure when it’s tested in the clinic. The approach that we’ve been taking, which I would say is almost unique in the world, surprisingly so I would say, is that almost all of the previous pre-clinical work done over the last fifty years has involved growing and treating tumours in mice which are not metastatic or which are metastatic but the time that the therapy was started did not involve treatment of advanced metastatic disease. That’s the type of patient that is typically enrolled in almost all phase I clinical trials, phase II clinical trials and most phase III clinical trials – patients who have liver metastases, lung metastases, brain, bone, whatever. And metastatic disease is much harder to treat than localised primary tumours which are often removed anyway by surgery.

So we began an approach about six or seven years ago where we have begun to develop models of advanced metastatic breast cancer, malignant melanoma, renal cell cancer, ovarian cancer, among others and then we’re using these models to evaluate the impact of various types of drugs that I’m interested in – antiangiogenic drugs that target blood vessel development in tumours or chemotherapy, given either in a standard way or at very low doses every day, something called metronomic chemotherapy. We have uncovered quite a few observations, which we have reported mostly, that seemed to indicate that these models of advanced metastatic disease more closely recapitulate what you see in the clinic when we do retrospective analyses. So if we take drugs, for example, that on their own, some of these new so-called targeted therapies, have very little activity in cancer patients with metastatic disease, that’s what we’re finding in these new models that we’re developing as opposed to the more traditional model that has been used for the last fifty years. So that type of observation, amongst others, leads us to believe that we may be on the right track, that although these models are more expensive to develop and to use, at the end of the day they may be far more informative with respect to providing information that will tell, inform oncologists and drug companies whether to move forward or not with a particular drug or a particular combination of drugs into a clinical trial.

We haven’t been developing new drugs ourselves, we take new drugs that have been developed recently, for example some of the antiangiogenic drugs like Avastin or other drugs that have been recently approved by the FDA and other regulatory agencies around the world for treating cancer by inhibiting blood vessel formation, and we’ve been evaluating again these drugs in a different way from what other investigators have historically done. We’re evaluating their impact in a setting of more advanced stage metastatic disease. And, with some exceptions, especially the drug known as Avastin or Bevacizumab, it is essentially not used up front at the beginning of therapy on its own, it is used in conjunction with chemotherapy. Because it was found, in early clinical trials, that when it’s used on its own for treating patients with advanced metastatic disease it is largely ineffective, it has hardly any discernible meaningful activity. It doesn’t look like a good drug but when it is combined with certain other drugs, particularly chemotherapy, then it adds to the effect that chemotherapy normally brings about so that a combination is better. We’re seeing the same sort of thing now in these models of advanced metastatic disease, in other words, when we take these antiangiogenic drugs and we evaluate them on their own they seem to be largely ineffective when treating advanced metastatic disease. When we start to combine them with certain chemotherapy drugs then we begin to see efficacy that’s improved, a chemotherapy efficacy that’s improved by the addition of the antiangiogenic drug.

So we’re now also investigating different ways of giving chemotherapy, not necessarily in the traditional way of giving the highest dose you can, what is called the maximum tolerated dose, separated by a long break period between each successive course of therapy; instead we’ve been working for over a decade now on giving chemotherapy drugs at much lower doses in a much more frequent schedule, maybe every day by tablet with oral drugs, something called metronomic chemotherapy. It’s still very investigational, so it’s taking old drugs like, for example, cyclophosphamide that has been around for decades and decades, and evaluating that drug where it’s given in a non-traditional way, less toxic but maybe just as efficacious or maybe more so. But again we’re evaluating that possibility using these models of advanced metastatic disease.