Patricia F. Fitzpatrick Dimond Ph.D. Technical Editor of Clinical OMICs President of BioInsight Communications
Differentiating which markers are in the driver’s seat and which are along for the ride is key.
What scientists have characterized as the deluge of information that has emerged from genome-scale studies has forced parallel development of new analytical frameworks and tools. Among these are approaches to exploring the functions of altered cancer genes in the context of a functional complex or pathway. Scientists say that the current pace of technological advances makes it increasingly clear that the ability to perform prospective and comprehensive molecular profiling of tumors will become the norm, enabling genome-informed personalized medicine.
These approaches should permit the efficient validation of genomic data to distinguish mutations responsible for disease pathogenesis from other mutations that result from genomic instability, defining those genes involved in cancer initiation, progression, or maintenance and identifying the optimal ways to exploit this information therapeutically.
Plenty of complications remain, however, in data interpretation and assigning functionality to observed mutations, as knowledge about mutations themselves and their role in cancer development piles up and gets murkier. Until recently, researchers thought that about five to seven mutations were needed to trigger the uncontrolled cellular proliferation that defines cancer. But recent estimates put this number at as high as 20 in some cancers. And according to recently published research, scientists are finding on average five to 15 mutations involved.
The majority of newly discovered mutations are rare, occurring in fewer than 5% of specific cancers. Each newly sequenced cancer genome reveals mutations that have never before been seen. Some of these mutations may be driver mutations, causally related to oncogenesis and conferring a growth advantage to cancer cells as well as positively selected in the microenvironment in which the cancer arises. Passenger mutations, on the other hand, have not been selected, have not conferred clonal growth advantage, and do not contribute to cancer development. These mutations occur within cancer genomes in which they arise, without functional consequences, during cell division.
Researching Kidney Cancer
Since many cancer genes seem to contribute to cancer development in only a small fraction of tumors, distinguishing among infrequently mutated cancer genes and those with random clusters of passenger mutations requires the analysis of large sample sets.
Investigators at the Cancer Genome Project, Wellcome Trust Sanger Institute, systematically resequenced the genomes of 210 human cancers. The investigators found more than 1,000 somatic mutations in 274 megabases (Mb) of DNA that corresponded to the coding exons of 518 protein kinase genes.
The investigators reported a substantial variation in the number and pattern of mutations in individual cancers, reflecting different exposures, DNA repair defects, and cellular origins. Most somatic mutations, they noted, may be passengers that do not contribute to oncogenesis. However, there was evidence for driver mutations contributing to the development of the cancers studied in approximately 120 genes.
The research team concluded that systematic sequencing of cancer genomes reveals the evolutionary diversity of cancers and “implicates a larger repertoire of cancer genes than previously anticipated.” Scientists had hoped, for example, that in the case of the most common form of adult kidney cancer, VHL gene alterations could serve as a prognostic and predictive marker for what seemed a fairly homogenous type of cancer. Clear cell renal cell carcinoma (ccRCC) accounts for more than 75% of this disease in adults. Germline mutations of the VHL suppressor gene cause the hereditary form of the disease, and more than 50% of sporadic ccRCCs have biallelic VHL mutations.
The protein product of VHL, pVHL, has functions including targeting hypoxia-inducible factors (HIFs) for degradation by the cell. Under the conditions of an abnormal VHL, HIF is not eliminated, becomes more active, and produces 200 or so proteins, one of them being VEGF.
Nicholas Vozelgang, M.D., a professor of medicine at the University of Nevada and a renal cancer specialist, noted that “we now know that of the clear cell tumors—70 to 80 percent of all kidney cancers are clear cell—a majority are VHL mutated. Therefore, it is probable that all these new agents—sunitinib, sorafenib, and anti-VEGF drugs—only are working on those kidney cells that produce HIF and HIF-driven proteins. The nonclear cell tumors and perhaps some of the clear cell tumors are not producing the HIF-driven proteins.”
We could group clear cell tumors into two categories: VHL mutated and VHL nonmutated, he said, but “we don’t have an easy way to quickly measure that in the blood or in the protein. The hypothesis is that the very well-differentiated renal cells will be responsive to these agents and the very poorly differentiated, aggressive tumors will not.”
However, loss of VHL alone has shown to be insufficient for tumor initiation and some of ccRCCs retain wild-type VHL alleles, indicating a requirement for additional or alternative genetic alterations for tumor development.
Studying Melanoma
At MIT’s Broad Institute, a major research effort focuses on The Cancer Genome Atlas (TCGA) project, for which the Broad is a Genome Data Analysis Center (GDAC). A long-term goal of the efforts is the comprehensive characterization of over 100,000 cancers with corresponding clinical data.
Recently scientists working at the Broad Institute, Dana Farber Cancer Institute, and other institutions reported (in a paper with 48 authors) the whole-genome sequencing of 25 metastatic melanomas and matched germline DNA. They found that their “high resolution view of the entire genetic landscape across a spectrum of metastatic melanoma tumors” revealed global genomic evidence for the role of ultraviolet mutagenesis in melanoma. Further, the research revealed several recurrent mutated and rearranged genes not previously associated with melanoma.
Their investigations showed that mutations in the PREX2 gene product, which normally interacts with PTEN, a protein that acts as a tumor suppressor, spurred tumor growth in ways that remain undefined. The PREX2 gene, the scientists said, appears to acquire oncogenic activity through mutations that perturb or inactivate one of its cellular functions. The pattern of mutations may represent, they explained, a novel category of cancer genes with “distributed mutation patterns” that may promote tumorogenicity through either dominant negative effects or more subtle dysregulation of normal protein functions.
These results were published in May in Nature. “Sequencing the whole genome certainly adds a richness of discovery that can’t be fully captured with a whole exome,” said Levi A. Garraway, a senior associate member of the Broad Institute, an associate professor at Dana-Farber Cancer Institute and Harvard Medical School, and co-senior author of the paper.
Personalizing Medicine for Cancer
But will knowing that individual patient tumors have multiple mutations help patients? It’s now clear, scientists say, that thousands, perhaps millions of genetic mutations can trigger and sustain cancer. Information about these mutations can be captured with current technology, but “we really don’t have the tools to take advantage of this information today,” notes Tyler Jacks, Ph.D., director, Koch Institute for Integrative Cancer Research at MIT, and investigator, Howard Hughes Medical Institute. “And until those tools are available no one will be able to answer the crucial question of whether a detailed map of a patient’s cancer will help that person live longer.”
Dr. Jacks’ laboratory has focused on developing new animal models of human cancer that can potentially more precisely inform the role of specific genes in cancer development and define the pathways their expression may affect. The laboratory has generated a series of novel mouse strains containing germline mutations in several genes implicated in human cancer.
These models resemble the human disease both at the genetic and phenotypic levels. And since the animals have been produced with precise mutations, the animals reflect the results of these altered genes.
Clearly, discovery of potentially clinically useful information about the multitude of aberrations that either drive or sustain human cancers will continue to require multiple approaches to mining genetic data. These endeavors will require the efforts of very large teams of scientists with bioinformatics, computational, genetic, and biochemical skills, among others, working at institutions with enormous resources to deliver the promise of personalized medicine for cancer.
Patricia F. Dimond, Ph.D. ([email protected]), is a principal at BioInsight Consulting.