Why Do Cancer Omics Attract Clinicians So Much?

Introduction

Since its first steps, Clinical Oncology has looked to basic research in the hope of fully understanding carcinogenesis and finding a better way to cure patients. Today, to the many who devote their lives to taking care of patients with cancer, Omics represent the biggest hope and challenge for tumor prevention, diagnosis, and treatment. The great body of information that has been produced over the last few years has given us the opportunity to study cancer from different viewpoints, thus opening up a wide range of perspectives for the future. Nevertheless, the consequences in clinical practice are still modest, reminding us that the Omics revolution has not yet achieved all its goals.

The main hopes of Genomic Medicine were to define new subtypes of disease identified by different biomarkers, different outcomes and different response to therapy. These goals were expected to be reached by comparing patterns of DNA alterations and of RNA or protein expression in different classes of tumors. At present, although Genomic Medicine has greatly increased both the quantity of data and the rate at which we can generate data, we still need the skills to bring together all these data in order to understand both physiological and pathological processes (www.nature.com/humangenome).

Thus, the possibility of quickly going from the bench to the bedside with improved tools to cure our cancer patients will depend on the ability of the researchers to overcome the “cons” and take advantages of the “pros.”

Cons

Studies of gene expression profiling have shown that cancers vary greatly depending on the specific combination of mutations, polymorphs, and deletions present in that patient, thereby amplifying the need for the development of tailored treatment. Tumor cell gene expression is also modulated by epigenetic mechanisms such as DNA methylation, histone acetylation, and/or methylation (Cheng and Blumenthal, 2010), and by small endogenous RNA molecules, such as miRNA and ncRNA (Taft et al., 2010), thus making it more difficult to modify these complex pathways. Genome-wide association studies (GWAS) have suggested that there may be associations between specific polymorphisms and certain types of cancer, but none of these studies is presently being used to improve cancer prevention. Individual genetic and epigenetic characteristics are also influenced by the environment we live, by our diet, and by the prenatal milieu (Fontham et al., 2009), leading to cancer prevention programs that are limited to small geographical areas and/or ethnic groups.

Over the last few years, the search for reliable biomarkers of patient outcome that might discriminate between aggressive and indolent cancers and that could predict the response to specific therapeutic regimens has been intense, but the complexity of the pathways that were revealed by proteomic and metabolomic studies have reduced the expectations of reaching this goal within a short time.

Thus, the complexity of the mechanisms involved in the regulation of tumor cell proliferation, differentiation and death have made targeted drugs hard to find.

Pros

Despite these considerations, however, the gap between basic research and clinical application could be filled in a short time by translational research projects linking bench to bedside. A great contribution will be given by the dramatic drop in the cost of sequencing and by the implementation of sequencing technologies (www.nature.com/humangenome). Further success will derive from the launch of the cancer genome project (http://www.sanger.ac.uk/genetics/CGP) under the umbrella of the international cancer genome consortium (http://www.icgc.org). This project will make sequences of 25,000 cancer cells freely available to researchers, thus allowing for the identification of a genetic profile for each specific cancer phenotype. The use of molecular signatures will allow us to define homogeneous subgroups of cancer with similarly altered pathways. Cancer genome-wide analysis (CGWA) by using high-density oligonucleotide arrays has already allowed the identification of copy number variation (CNV) in several types of cancers. As a consequence, specific CNV profiles have been associated with different cancer phenotypes. The genome-wide study has led to the discovery of several nonrandom chromosome numerical variations as well as structural abnormalities. These nonrandom abnormalities contribute to defining the so-called “genome signature of cancer” and have helped to more precisely assess the risk of individual cancer patients and to find targeted drugs. Broad application of these approaches to large numbers of patients affected by different types of cancer will certainly allow us to find either powerful, targeted drugs, as has occurred for chronic myeloid leukemia patients (imatinib) or targeted antibodies, as is the case for breast cancer (trastuzumab and cetuximab) or myeloma (rituximab) patients.

GWAS have also improved the search for new prognostic markers, and different types of CNVs have been found to be associated with diverse clinical behavior and with different disease outcome. Recently, evaluation of tumor CNV at diagnosis in patients with low or intermediate risk tumors has been added to a new protocol for neuroblastoma (https://www.siopen-r-net.org/), a pediatric solid tumor with an incidence of 10.2 cases per million children (Maris, 2010). Patients whose tumors have at least one structural abnormality will undergo more aggressive treatment than those lacking chromosomal aberrations.

As well as the genome, the transcriptome has also been deeply investigated in order to find gene signatures associated with poor outcome. With regards to neuroblastoma, meta-analysis has been used to match several public gene expression profiling databases with experimental microarray data from various European laboratories (Vermeulen et al., 2009). A specific 59-gene signature that includes several important genes for neuroblastoma, such as MYCN and TRK1, is able to predict patient outcome. It is interesting to note that the search for molecular predictive factors in neuroblastoma dates back several decades. In the early 1990s, MYCN oncogene amplification (Schweigerer, 1990) was included as a negative prognostic marker in localized neuroblastoma and, nowadays, CNV signature has become the most important marker associated with tumor progression. Thus, at least with regard to neuroblastoma, Omics allowed the move from “one gene” to a “genome” prognostic marker, and some of the newly discovered genome signatures have been introduced to stratify patients into new therapeutic protocols. The success of new therapeutic approaches that use genome signatures greatly depends on the optimal interaction among several groups of people such as oncologists, nurses, pathologists, biologists, and experts in bioinformatics. As shown in Figure 1, the coordinated activity of various teams can bring genome signature evaluation back to the bedside in less than 10 days: a number of days that does not impair the starting of tumor treatment. The oncologist admits the patient, the patient undergoes surgery, the tumor sample is evaluated by the pathologist who sends a sample to the biologist to extract DNA, the genome is studied by various methodologies, then reports are uploaded into a database and a final report is sent to the oncologist who chooses the best treatment for the patient.

Conclusions

Translational research represents an indispensable link between basic and clinical research by fueling innovations that will certainly have great impact on the prevention, diagnosis, and treatment of cancer over the next decades, thus helping clinicians in all aspects of cancer patient management. With this in mind, many outstanding experts in all the areas related to cancer Omics met at the International School of Erice to discuss their work, which is now presented in this special issue of OMICS.