Biomarker Discovery Gets a Fix on Cancer

Epic Sciences’ “no cell left behind” platform can identify circulating tumor cells in an unbiased manner, without enrichment or depletion on any parameter. For example, as indicated in this image, it has been used to detect the AR-V7-positive cells among the cells collected in a simple blood draw. The AR-V7 splice variant is linked to resistance of androgen receptor-targeting drugs in metastatic castration-resistant prostate cancer patients.

Just as signposts provide information, direction, and guidance, so too cancer biomarkers can better reveal the complex, dynamic, and heterogeneous landscape of malignancies.

Such information is critical for creating better cancer diagnostics, prognostics, and therapeutics, but the journey to find just the right biomarker is often a long and winding road.

Biomarker discovery and utilization is being explored by means of various methods and technologies. These include the isolation and characterization of rare circulating cancer cells, the use of multiplexing to extract information from limited amounts of sample, and even the application of evolutionary biology to detect early cell changes.

These approaches are all being developed by companies interested in improving cancer diagnostics, prognostics, and therapeutics. For example, one of the companies cited in this article is scrutinizing circulating tumor cells (CTCs) to predict resistance to cancer drugs. (This work was presented at a scientific retreat convened by the Prostate Cancer Foundation.) Other companies represented in this article are developing novel cancer diagnostic approaches. (These were detailed at the recent Cambridge Health Diagnostics Summit.)

While a typical blood sample from a cancer patient can contain 30 million nucleated cells, it’s estimated that perhaps only 5 of those cells may be destined to form tumors. “Most present-day diagnostic technologies cannot precisely detect those rare CTCs,” observed Ryan Dittamore, vice president, translational research and clinical affairs, Epic Sciences. “Identifying robust predictive biomarkers that can be utilized in real-time with a test compatible with diagnostic workflows is one of the greatest challenges to precision medicine.

“Epic Sciences is focused on developing sensitive diagnostic tests to characterize CTCs molecularly in order to match therapies to a patient’s cancer biology.” As Dittamore indicated at the Prostate Cancer Foundation event, the company evaluated its CTC approach in a study undertaken with Howard Scher, M.D., chief of the genitourinary oncology service at Memorial Sloan Kettering.

“We focused on a form of prostate cancer called metastatic castration-resistant prostate cancer,” explained Dittamore. “In the study, we confirmed the value of the prostatic cancer biomarker AR-V7 in 193 patient samples. In men with this androgen receptor (AR) splice variant, treatment with taxanes may be more effective than treatment with AR-signaling-directed agents (enzalutamide or abiraterone).”

In Epic Sciences’ “no cell left behindTM” approach, all nucleated cells from a blood sample are placed on a slide. CTCs are stained using biomarker monoclonal antibodies and assessed via immunofluorescence and high-resolution digital pathology scanning.

“We’ve industrialized this process,” asserted Dittamore, “and can evaluate millions of patient cells in every assay. Once identified, we can then isolate individual CTCs for further single-cell evaluation, such as with next-generation sequencing.” Overall, the company’s approach can achieve the following tasks: quantifying the number of cancer cells, characterizing biomarker expression (specifying, for example, subcellular localization and genomic alterations), and assessing disease heterogeneity and clonality of cancer cell types.

Commercializating the tests will come later, indicated Dittamore. “We will continue refining and expanding tests with our numerous collaborators in academia and pharma, many of whom are using our biorepository capabilities to bank samples for later study of biomarkers for other types of cancer.”

Qiagen’s Modaplex, an automated molecular diagnostic solution, combines two established tech- nologies—PCR and capillary electrophoresis. It can be used to design complex multimodal assays that test for disparate target types such as single nucleotide poly- morphisms, expression biomarkers, microRNAs, and fusion genes. Disparate target types or assay types may be evaluated “all in one well” rather than being separated into different wells, different runs, or even different instrument types.

The ultimate goal of molecular diagnostics is to get as much information as possible, as quickly as possible, with as little sample as possible. Unfortunately, it is often necessary to make do with samples that are limited in number and extent—consider, for example, formalin-fixed, paraffin-embedded (FFPE) specimens—even though one would like to extract comprehensive information. To help solve this problem, Qiagen has developed a new platform that marries and automates two proven techniques: multiplex PCR and capillary electrophoresis.

“We created a novel and automated clinical platform called Modaplex for real-time sampling, size separation of PCR products, amplification curve building, and Ct (cycle threshold) calculations,” summarized Lilly Kong, DVM, Qiagen’s senior director, assay development. “As a result, from a very limited amount of sample, we can utilize a multimodal, multiplex, essentially ‘all in one well’ means to quickly obtain actionable results.”

Dr. Kong said that Qiagen has developed multiple oncological biomarker tests including RAS, cMET, FGFR, Cell Cycle, and DLBCL signatures. “We can multiplex up to 40 targets in a single reaction,” she asserted. “Traditional assays allow testing of only one or a few assays at a time. Further, this is an architecture that uses a modular approach.”

Dr. Kong described development of an assay evaluating cMET and epidermal growth factor receptor (EGFR) expression and copy number variation. These two players are important because the MET proto-oncogene encodes for the receptor tyrosine kinase, cMET, which is widely expressed by many cells.

Under pathological conditions, cMET confers survival and invasive properties of cancer cells while potentially blocking EGFR therapies. “This multimodal 21-plex assay,” Dr. Kong indicated, “succeeded in evaluateing nine mRNA targets and nine genomic DNA targets with appropriate standards and external process controls.”

Qiagen also examined cMET simple nucleotide polymorphisms (SNPs) in a 16-plex reaction tube. “The modular approach enables combinations of important biomarker assays that can test for disparate target types, such as SNPs, expression biomarkers, miRNAs, and fusion genes,” Dr. Kong detailed. “Further, one FFPE slide is sufficient for about 10 assays with 10–50 ng input per reaction. The automation is user-friendly while sensitivity, specificity, and precision are equal to or exceed singleplex assays.”

Although Modaplex started with oncological applications, many other panels are in development. For example, Qiagen is working on pathogen detection applications as well as in-process monitoring applications for biologicals manufacturing.

• Malignant Crosstalk

  The crosstalk between cancer cells and their local environment is a key feature of establishing and maintaining a malignant state. Said Gus Frangou, Ph.D., director of clinical diagnostics, Cellecta:“We do not exclusively focus on cancer cells, but rather endeavor to understand more holistically how the tumor microenvironment and nonmalignant cell types assist in driving tumorigenesis.”

  Cellecta has developed Driver-Map™ biomarker discovery panels as an alternative means of performing next-generation sequencing (NGS) mapping. The technology presents several advantages over whole-genome sequencing from clinical samples derived from both solid tumors and hematologic malignancies.

  “These panels can rapidly multiplex several hundred clinical samples in a single sequencing run,” Dr. Frangou noted. “The modular and customizable approach provides an overall molecular profile for characterizing the samples.

  “We take biopsy specimens and determine their RNA signature to identify key cancer mutations and clinically important gene variants present in RNA. In addition, we model tumor cell composition and the presence of infiltrating immune cells, blood vessels, and stromal cells.”

  According to Cellecta, Driver-Map is a novel multiplexed gene expression profiling service. It relies on a quantitative RNA-Seq approach that leverages the power of multiplex polymerase chain rection (PCR) technologies and next-generation sequencing. Instead of performing a routine RNA-Seq workflow, Driver-Map simultaneously measures gene expression and germline and/or somatic mutations present in transcribed RNA.”

  “We estimate there are about 600 cancer driver genes; however, only a small subset of these are currently druggable,” Dr. Frangou explained. “The current challenge is that it often takes several months for full analysis of samples for even just these cancer drivers using conventional genetic platforms. We built a digital profiling assay that expedites this tremendously and offers hundreds of times more sensitivity than other current technologies, especially for low-purity tumors.”

  The panels allow simultaneous and quantitative target gene expression analysis and mutation detection using competitive multiplex-PCR amplicons. Computational network modeling then facilitates the rapid identification of cancer driver genes, expressed pathways, and cellular subtypes present in the tumor.

  A key feature is the set of internal primer set controls. “We empirically screened thousands and thousands of primers to identify the best primer sets,” stated Dr. Frangou. “We also developed synthetic competitive template spike-in controls for all targeted genes for data normalization and cross-sequencing platform compatibility (up to 50,000 amplicons in a single reaction), and we employ bar codes to provide absolute molecule counts—analogous to digital PCR.”

• Common Workflow

  Cancers develop because of underlying alterations in the genome that may include point mutations, changes in gene copy number, gene fusions, combinations of these, and other genomic changes. “There are a number of disparate assays being used to assess these cancer drivers,” said Joe Don Heath, Ph.D., vice president, diagnostics market development, NuGEN. “But that can waste precious tissue as well as time. We’ve developed one simple, common workflow for both DNA and RNA that interrogates multiple genomic aberrations and provides flexible, custom-targeted sequencing panels with a quick turnaround time.”

  The key to the NuGEN’s Ovation^® target enrichment approach is the company’s single primer enrichment technology (SPET). “We design targeting probes that anneal to complementary sequence in the target region of interest on both strands and then extend those probes through the region of interest,” explained Dr. Heath. “The entire workflow is completed in less than 24 hours. The design allows us to tile of probes throughout large contiguous regions, as well as more closely spacing the probes, which is especially useful for samples with fragmented nucleic acid, such as FFPE tissue.”

  A key advantage of the approach is that it eliminates the difficult task of designing specific PCR primer pairs while maintaining high specificity of recovered target sequences in the final library. The overall workflow to achieve library enrichment begins by fragmenting the sample to about 500 base pairs. After an end repair step, indexed forward sequencing adaptors are ligated to the 5′ ends of the randomly fragmented DNA.

  Once individual samples are barcoded by ligating the indexed adaptors, samples are then pooled together for probe annealing and extension. During this process, primers hybridize to the probe landing zone and are subsequently extended by DNA polymerase through the target regions and the ligated forward adaptor.

  “The result is a targeted sequence that is flanked by forward and reverse adaptor sequences,” summarized Dr. Heath. “These can be amplified by PCR to complete library construction for sequencing with standard Illumina primers.”

  While the company offers catalog cancer and gene fusion panels, its focus is on custom panels. “Scientists often need only a specific set of targets,” Dr. Heath noted. “We work directly with our clients to define and optimize their targets, delivering a custom panel within three weeks of design approval. The individual design service and fast turnaround time is a winning combination for our clients.”

• Phylogenetic Analysis

  Evolutionary theory is a central tenet of biology. “Heterogeneity at all levels, including cancer development, is a product of evolution,” stated Hakima Amri, Ph.D., associate professor of biochemistry and cellular and molecular biology at Georgetown University School of Medicine and vice president of Phylomics. “During cancer development, cells transition from healthy, well-differentiated, functional cells to a much more primitive state with a closed capacity for differentiation, metabolism, and other functions.”

  Dr. Amri and colleagues have taken a novel approach to assessing cancer biomarkers by developing evolution-compatible analysis methods that examine multiple levels of heterogeneity in both healthy and diseased states. “Cancer development, progression, and maintenance are all evolutionary processes in that they all involve genetic modifications, selective pressure, and clonal production,” explained Dr. Amri. “Therefore, evolution-compatible methods of analysis can be applied not only to cancer studies but also ultimately and importantly to diagnostics.”

  Dr. Amri’s team developed a phylogenetic analytical approach for the analysis of omics data (genomics, metabolomics, proteomics). This approach came to be called Phylomics—a name that was subsequently applied to a company the team started.

  “We created a comprehensive algorithm for data mining using a systems biology approach to portray specimens on a phylogenetic tree (a cladogram),” detailed Dr. Amri. “The resulting detailed analysis presents a spectrum of similarity from the least amount of variation to the most. A tree of synapomorphies—shared, derived states—emerges, and the branching and the clades (a group of related specimens) are clarified. Importantly, this method of data modeling can show a gradient from ‘healthy’ to ‘early change’ to ‘malignant transformation.’”

  Dr. Amri said that analysis of molecular heterogeneity using phylogenetic cladograms is a paradigm shift for evaluating diseases, and such analysis could be especially useful in the early detection of cancer.

  “Because it represents evolutionary differences, the cladograms can reveal early changes that will likely lead to malignant transformation,” Dr. Amri pointed out. “These patient-stratifying dynamic trees, thus, have direction. They can provide an entirely new approach, one that can be applied to produce natural class determination, predictive biomarker recognition, and ultimately clinical diagnostics.”

  To help researchers inexperienced in understanding such cladograms, the new company offers custom services to analyze clients’ data the “phylogenetic way.” Phylomics is also seeking funding partners.

Source