February 24, 2005: Animal and cellular models have traditionally been used as vehicles to study the progression of a given disease. They can also be used to test the efficacy of novel compounds that may be used to treat the disease in question. By definition, however, such models only simulate a disease state and never completely replicate the exact molecular mechanisms responsible for the disease in humans.

The use of these models has had tremendous influence on the development of potent therapeutic strategies in human pathology. Unfortunately, cancer is one disease in which current drugs have had considerable trouble finding success and acceptance among its patients. One key reason may be due to the fact that these drugs target cell proliferation mechanisms relied upon by normal and cancer cells for survival. This particular feature carries with it a signature of toxicity that presents a significant physical and emotional burden for the patient to endure, and may contribute to the high failure rates of new anti-cancer drugs.

In the February issue of Nature Reviews in Drug Discovery, a critical, insightful overview on the use of animal and cell culture models in finding effective cancer therapeutics is provided by Alexander Kamb, Global Head of the Oncology Disease Area at the Novartis Institutes for BioMedical Research. The substandard efficiency exhibited by most cancer compounds currently under investigation in clinical trials, Kamb states, may be explained by a number of factors. One important factor is the unpredictable, complex physiology of the models used. Various molecular circuits that govern uncontrolled cell proliferation in these models may not be valid in the human context, thereby complicating interpretation of the results in patients.

Another distinctive element is the complexity of the cancer itself. For example, the dependence of a particular tumor on a given protein target may not necessarily be exhibited by other tumors. Moreover, the requirement of the target may ultimately depend on the genetic or epigenetic background of the specific cancer cells. Whatever the factors may be, they may all cloud the accurate assessment of a cancer drug’s performance in human patients.

On a more positive note, the imperfections of such models have revealed a straightforward yet important indicator for success: the level to which the model replicates the genetic characteristics of the cancer treated in the human patient. For example, cell lines derived from tumors containing a well-characterized genetic alteration have been used to create animal models that were subsequently responsible for successful cancer drugs in the clinic. However, again because of the models’ inherent physiological complexity, Kamb suggests that researchers must use prudence in examining and interpreting the results.

Enhancing the reliability of cancer models is difficult but not impossible, Kamb says, provided that researchers take appropriate steps to improve their molecular understanding of cancer cells, regardless of whether it is in an experimental or patient background.  Cancer models can then be chosen intelligently to more closely resemble the disease that they are intended to simulate, thereby improving the success of trial cancer drugs.