Cancer can be seen as a genetic disease resulting from mutations in a subset of genes that confer growth advantage to the cells in which they occur. The recent boom of sequenced cancer genomes illustrates the potential of next‐generation sequencing to provide unprecedented insights into the mutational processes, cellular repair pathways and gene networks associated with cancer. Studies such as the massively parallel sequencing of a small‐cell lung cancer cell line (Pleasance et al, 2010), a clear cell renal cell carcinoma (CCRCC, the most common form of adult kidney cancer) (Dalgliesh et al, 2010), several molecular subtypes of breast cancer (Stephens et al, 2009) and B‐cell chronic lymphocytic leukaemia (Campbell et al, 2008) highlight the mutation diversity occurring in different cancers (Figure 1). These studies also revealed other interesting features of the cancers studied. For example, mutations in small‐cell lung cancer may correspond to a given mark of exposure as cigarette smoking (Pleasance et al, 2010) and the sequencing of the CCRCC identified novel inactivating mutations in genes like histone H3 lysine 27 demethylase (Dalgliesh et al, 2010). The detailed analysis of chromosomal rearrangements in breast cancer has also revealed their complexity (Stephens et al, 2009) and Campbell et al provided evidence for heterogeneity and the existence of a molecular clock to track the tumour development (Campbell et al, 2008). The sequencing of the coding exons of 518 protein kinase genes (a total of 274 Mb of deoxyribonucleic acid (DNA)) revealed more than 1000 somatic mutations in 210 different human cancers (Greenman et al, 2007) and the pattern of mutations observed reflected the cancer specificity of individual cancers, likely determined by different exposures to carcinogens, DNA repair defects and cellular origins. The systematic sequencing of cancer genomes will likely reveal the enormous diversity expected to be present in cancers and may show an even broader spectrum of affected genes than we can now anticipate. Currently more than 90,000 individual mutations in 13,423 genes in almost 370,000 tumours are described in COSMIC, a database of Somatic Mutations in Cancer (Forbes et al, 2010).
The enormous clinical implication of revealing novel mutations has been thoroughly discussed by Dias‐Santagata et al in this issue; these mutations are also the tumour's ‘Achiles’ heel' as they can specifically ‘mark’ the tumour cells and direct the mechanism of their destruction (Figure 1). They are at the basis of targeted cancer therapies that will interfere with cancer proliferation in a specific manner. Many of these therapies focus on proteins that are involved in cell signalling pathways, which form a complex communication system that governs basic cellular functions and activities such as cell division, cell movement, how a cell responds to specific external stimuli and even cell death. By blocking signals that tell cancer cells to grow and divide uncontrollably, targeted therapies can prevent cancer progression and may induce controlled cancer cell death
»targeted therapies can prevent cancer progression and may induce controlled cancer cell death«
(apoptosis). Other targeted therapies can cause cancer cell death directly, by specifically inducing apoptosis, or indirectly, by stimulating the immune system to recognize and destroy cancer cells and/or by delivering toxic substances to them. In fact an increasing number of so‐called ‘smart drugs’ or targeted therapies, which may block oncogenic pathways or stimulate pathways specifically inactivated in tumour cells are either approved or in trial for clinical use. The current paradigms are EGFR mutations in adenocarcinoma of the lung that can be treated with gefitinib, KRAS mutations in colon cancer with respect to treatment with EGFR antibodies as well as others reviewed by Harris et al. Introducing systematic clinical screenings for mutations affecting these pathways is essential to identify targets for targeted therapies and the patients that will respond to each treatment. In this issue Dias‐Santagata et al present a significant step forward in this direction by describing an optimized assay for identification of the most common oncogenic mutations (including EGFR, KRAS, NRAS, BRAF and PIK3CA). The assay is based on an easy to use and widely available technology, the SNaPshot from Applied Biosystems. Primary cancers (n = 250) from 26 different human malignancies were analysed. The immediacy of the application of mutation analysis in clinical practice is appealing (2–3 weeks) and the author's discussion of their experience shows it is feasible and useful. Another essential value of the method is the possibility to retrospectively analyse huge series of samples of archive material since it is optimized for DNA from paraffin embedded tissues. Virtually, all clinical laboratories have the necessary equipment to perform this analysis and clinical researchers can interrogate their collected archival materials to test their own hypothesis thus allowing for new ideas to be generated and pursued also outside huge technological centres. Although deep sequencing is likely to become less costly and increasingly available for immediate use in the clinic, the sparseness of biological material from small biopsies will still be a limitation for its use and the system by Dias‐Santagata et al is highly efficient in this regard. Regardless of the technology used, screening for relevant mutations in clinical settings is of pivotal importance to help the oncologist to design the appropriate treatment for each patient.
»screening for relevant mutations in clinical settings is of pivotal importance to help the oncologist to design the appropriate treatment for each patient.«
This approach and others, such as OncoMap (MacConaill et al) or Oncotype DX, fill in the gap between elite science and clinical application. Dias‐Santagata et al have chosen to design assays for recurrent mutations that activate oncogenic signalling pathways targeted by either FDA (Food and Drug Administration) approved drugs or in advanced stages of clinical trials. The scientific discovery of these mutations therefore dates from around 5 years ago. The high‐throughput deep sequencing studies we highlighted above have described an unprecedented number of novel mutations within less than a year. Leary et al (2010) suggest an even more proactive use of massively parallel sequencing for personalized medicine by estimating the tumour burden from the fraction of mutant DNA present in plasma samples to monitor the effect of new drugs. While the clinical relevance of newly identified mutations needs to be further validated, the speed with which cancer genomes can now be sequenced underlines the need for continuous updating of the mutation assays used in the clinics. In that respect, the modular system presented by Dias‐Santagata et al can be upgraded in a simple manner by including additional parallel assays.
All these developments call for an accelerated and versatile approval system to allow clinical applications to follow the rapid accumulation of mutation data provided by new generation sequencing. Only then personalized medicine can indeed reach the individual person.
The authors declare that they have no conflict of interest.
- Copyright © 2010 EMBO Molecular Medicine