| Driven by the need to find the right treatment for each patient, Marc Buyse of the International Drug Development Institute assesses the challenges and rewards of using genomics in clinical cancer research
Cancer clinical research is constantly looking for better identification of patients who respond to treatment. Molecular profiling may well play an important role in this respect. Genomic signatures with prognostic and/or predictive value are identified retrospectively, using material from tumour banks, and validated prospectively in randomised clinical trials. This article outlines these developments and illustrates them in the context of adjuvant therapy for early breast cancer.
THE MOLECULAR REVOLUTION IN CANCER
In-depth knowledge of molecular mechanisms underlying the development and the growth of tumours has revolutionised the development of anti-cancer agents. In the coming years, clinical research will have to adapt its methods to test so-called ‘targeted’ therapies that can shut down well-defined biological pathways for tumour growth, rather than simply kill cancer (and other) cells.
Some of these targeted agents have already shown remarkable efficacy (for instance gleevec for the treatment of gastro-intestinal stromal tumours, or herceptin for the treatment of breast cancer). However, others have yet to prove their clinical efficacy despite promising preclinical results. For instance, divergent efficacy results in lung cancer for two molecules active against the same target – erlotinib and gefitinib, two small molecules inhibiting the epidermic growth factor (EGFR) underscore the need to rethink clinical development strategies.
Erlotinib significantly increased survival compared to placebo (death rate reduced by 30 per cent, p<0.001) in a clinical study with 731 patients after two or three lines of chemotherapy for non-small cell lung cancer (1). A similar clinical trial with 1,692 patients did not identify a significant survival advantage of gefitinib compared to placebo (death rate reduced by 11 per cent, p = 0.087) (2).
Mutations in the receptors of the epidermic growth factor, which have been confirmed since, confirm the clinical observation, showing that the efficacy of gefitinib is far more important in a sub-group of patients with a specific genetic profile (2,3). This example confirms that the early and reliable identification of ‘responders’ to anti-cancer agents will be a major challenge in the near future. |
