May 15, 2014 (Vol. 34, No. 10)
Will Tumor Signatures, Bioinformatics Data, and Clinicians Ever Be on the Same Page?
The combination of molecular biology, engineering, and bioinformatics has revolutionized understanding of the factors precipitating human diseases, as well as potential treatment responses.
Omicss has most profoundly impacted cancer. Revelation of the connection between the molecular characteristics of the primary tumor in terms of gene expression, structural alterations of the genome, epigenetics, and mutations, as well as how these properties influence metastatic behavior and therapeutic responses will, scientists say, revolutionize cancer treatment.
In particular, advances in the field of genomics have led to remarkable paradigm shifts that will affect clinical practice, determining key characteristics of individual cancers as well as identifying novel therapeutics. These advances include, for example, the identification of sets of genes encoding proteins that act as checkpoints or on/off switches associated with cancer and novel drug development targets.
A prominent example of a checkpoint blocker drug in clinical development, Roche’s antibody MPDL3280A (RG7446) acts against an immune checkpoint blockade expressed in tumors. PDL-1, or Programmed Death-Ligand, is believed to function as a key part of the cancer-immunity cycle by acting as a stop signal that prevents the immune system from destroying cancer cells.
Cancer Diagnostics
Commercially available tests have already changed medical practice, as they help physicians identify patients with specific genetic markers and determine an appropriate course of treatment. Decipher Prostate Cancer Classifier from GenomeDx, validated specifically for men faced with treatment decisions after radical prostatectomy, classifies the patient’s tumor independently of PSA rise and other adverse features.
Decipher measures and analyzes the activity of 22 genetic markers expressed in the prostate cancer tumor to measure the tumor’s biological potential for metastasis after surgery. These 22 markers associated with aggressive disease were discovered from genome-wide search algorithms of more than a million markers and have been extensively validated in collaboration with leading academic medical institutions.
Myriad’s BRACAnalysis test confirms the presence of a BRCA1 or BRCA2 mutation. Options for patients based on the presence of these mutations include increased surveillance for ovarian and breast cancers, risk-reducing medications, or prophylactic surgery.
Myriad’s Prolaris is a risk-stratification tool for patients with prostate cancer. Designed to measure the aggressiveness of a patient’s cancers to better predict an individual’s relative risk of disease progression within ten years, the company says it can enable physicians to better define a treatment/monitoring strategy for their patients. Prolaris, the company says, is significantly more prognostic than currently used clinicopathologic variables and provides additional information that can be combined with other clinical factors to make the most accurate prediction of a patient’s cancer aggressiveness and therefore disease progression.
Foundation Medicine offers clinicians its FoundationOne genomic profile of patient solid tumors, which it says helps physicians make treatment decisions for patients with cancer by identifying the molecular growth drivers, helping oncologists match them with relevant targeted therapeutic options.
Using next-generation sequencing in routine cancer specimens, FoundationOne interrogates all genes somatically altered in human cancers that are validated targets for therapy or unambiguous drivers of oncogenesis based on current knowledge. It, the company says, reveals all classes of genomic alterations including base substitutions, insertions, deletions, copy number alterations, and select rearrangements.
One step up from test kits are interactive databases, such as the Drug-Gene Interaction database (DGIdb), which uses existing resources to generate hypotheses about how mutated genes might be targeted therapeutically or prioritized for drug development. Described by Malachi Griffith et al., in a Nature Methods paper, the database is geared toward researchers and physician-scientists who want to know whether mutations, found through genome sequencing, could be targeted with existing drug therapies. The database includes drugs approved by the FDA and other drugs in development.
Driver Genes
Adding to the complexity of data is the discovery of “driver” genes and their role in cancer biology. Elli Papaemmunuil and colleague, reporting on behalf of Chronic Myeloid Disorders Working Group of the International Cancer Genome, recently described their analysis of oncogenic mutations in large, well-characterized patient cohorts of myelodysplastic syndromes (MDS), characterized by dysplasia, ineffective hematopoiesis, and a variable risk of progression to acute myeloid leukemia.
Using previously identified mutations in genes implicated in RNA splicing, DNA modification, chromatin regulation, and cell signaling, the investigators sequenced 111 genes across 738 patients with MDS or closely related neoplasms to explore the role of acquired mutations in MDS biology and clinical phenotype.
The scientists reported that 78% of patients had one or more oncogenic mutations and that they could identify complex patterns of pairwise association between genes, indicative of epistatic interactions involving components of the spliceosome machinery and epigenetic modifiers.
This data suggests a hypothesis of genetic “predestination,” in which early driver mutations, typically affecting genes involved in RNA splicing, “dictate future trajectories of disease evolution with distinct clinical phenotypes.” Driver mutations had equivalent prognostic significance, whether clonal or subclonal and leukemia-free survival deteriorated steadily as the number of driver mutations increased. The authors concluded that analysis of oncogenic mutations in large, well-characterized cohorts of patients illustrates the interconnections between the cancer genome and disease biology, with considerable potential for clinical application.
Further illustrating the complexity confronting scientists and clinicians in translating Omics into clinically useful information is the recent development by Dutta et al., of a data-integration method to identify gene networks that drive the biology of breast cancer clinical subtypes.
The key objective of their work, the team said, was to shift the focus away from driver genes, derived from a long list of amplified genes, to identifying driver-networks. This strategy, they noted, not only includes driver genes but also reveals the associated deregulated networks/pathways. They argue that targeting individual driver genes is often difficult as not all driver genes are appropriate drug targets, and that finding a suitable drug target from members of a driver-network might be more feasible.
Their computational method simultaneously overlays gene expression and gene copy number data on protein–protein interaction, transcriptional-regulatory, and signaling networks by identifying coincident genomic and transcriptional disturbances in local network “neighborhoods.”
The scientists identified distinct driver-networks for each of the three common clinical breast cancer subtypes: estrogen receptor (ER)+, human epidermal growth factor receptor 2 (HER2)+, and triple receptor-negative breast cancers (TNBC) from patient and cell line datasets.
In one example of clinical relevance, the scientists’ TNBC analysis identified the LYN kinase as an important hub of the main driver-network in addition to EGFR in both patent tumor cells and cell-line datasets. Although EGFR was identified as the main hub gene and has been shown to inhibit the growth of TNBC cell lines, EGFR inhibitors alone or combined with carboplatin chemotherapy showed very little activity in the clinic in past studies. Their network analysis suggests potential combination therapy approaches, such as inhibiting LYN and other network EGFR partners in order to improve efficacy.
Cardiovascular Disease and Diabetes
While Omics technologies have had their most immediate impact on cancer, they have begun to influence decision-making in other medical arenas. In cardiovascular medicine, recent studies point to specific DNA variants and genes such as coxsackie virus and adenovirus receptor (CXADR), apolipoprotein E (APOE), and lipoprotein alpha (LPA) as key genetic markers for susceptibility to coronary artery disease and sudden cardiac death.
Genetic studies have helped characterize potential therapeutic responses to cardiac drugs identifying common sequence variants in at least two genes (CYP2C9 and VKORC1) that strongly affect the pharmacokinetics and pharmacodynamics of warfarin. Small clinical trials suggest that genetically based warfarin-dosing algorithms may enhance the efficiency and safety of warfarin dosing.
Likewise, the emergence of high-throughput genotyping technologies and genome-wide association studies has prompted the search for genetic markers of diabetes predisposition or susceptibility. To date, scientists say, therapeutic guidelines provide uniform management recommendations for “average” patients that rarely take into account individual variation in susceptibility to diabetes, complications, and responses to pharmacological and lifestyle interventions. A more personalized approach would combine bioinformatics with genomic, proteomic, metabolomic, and pharmacogenomic data to explore pathophysiology and to characterize more precisely an individual’s risk for disease, as well as response to interventions.
With the accumulation of massive amounts of data come problems. Analyzing, accessing, interpreting, and packaging the data into clinically useful information remains a daunting challenge. Vehicles to ensure that this valuable information is accessible and useful for physicians and patients are urgently needed.