Industrialising microarrays

Steve Lombardi, senior vice-president of commercial operations at Affymetrix, reviews developments in microarray analysis and its potential in drug development.

With the completion of the Human Genome Project four years ago came the hope and promise that the world's most ambitious sequencing effort would revolutionise pharmaceutical research and, ultimately, give us better therapies and improved patient care. However, during the decade-long project, scientists learned that the genome is far more complex than previously thought.The first estimate of 30,000 genes has given way to estimates of hundreds of thousands of splice variants, millions of newly discovered transcripts, and tens of millions of genetic polymorphisms. But the tools needed to understand this level of complexity simply did not exist.

The microarray, invented in 1989 by Stephen P. A. Fodor and colleagues,^1-3 has emerged as a central technology that is helping to unravel much of the genome's complexity. Over the past 15 years, microarray information capacity has consistently increased, providing for a tool that allows meaningful whole-genome analysis, currently able to measure expression for nearly 50,000 transcripts, or genotype more than 500,000 polymorphisms in a single experiment. This broad-scale genetic analysis has not only helped to discover the underlying genetics for countless diseases, but has fundamentally improved drug discovery and development research.

Before whole-genome microarray analysis, many drug development assays were typically limited to answering one very focused question, often generating a single data point. To perform comprehensive drug discovery, researchers must answer hundreds or even thousands of different questions, making the process slow, expensive, and prone to variability.

Microarrays have offered a significant improvement by measuring thousands of data points in a single assay, with the ability to analyse changes in gene expression and DNA sequence variation across the whole genome. However, microarray throughput and cost-efficiency have limited its application in pharmaceutical research, which requires far more sample analysis than does biomedical research.

To enable industrialised microarray research, Affymetrix has recently developed an automated 96-array high throughput (HT) format. The system automates the most labour-intensive steps in microarray processing - sample preparation, hybridisation, washing, and scanning - dramatically reducing the cost per assay. This decrease in cost, increase in throughput, and added reliability make the HT system ideally suited for drug discovery and development applications, including target identification and validation, compound profiling, and improved clinical trial outcome.

complex challenge

The complexity of microarrays - which require multiple enzymatic reactions, specific hybridisation conditions, stringency washes, fluorescence scanning, and data analysis - presented a challenge when engineering the HT system. The Affymetrix HT array adapts the same GeneChip technology and content to a standard 96-well plate footprint. Advances in feature size reduction have allowed significantly more content to be placed on smaller-sized arrays. And, by leveraging advanced automation methods, the HT system provides the consistency required to analyse hundreds of high-content arrays simultaneously.

The current HT microarray prototype contains 96 individual arrays mounted on a single plate, with each array containing the same genomic information as the company's Human Genome U133A array, but in approximately one-fifth the surface area. For each array of the 96-array plate, more than 500,000 probes are used to measure the expression of 18,400 human transcripts, meaning that each HT plate generates more than 48 million data points. By comparison, conventional HT screening may generate only a single data point per well - a total of 96 data points per plate.

Each 96-array plate is processed and analysed on a robotic Array Station that automates the microarray processing workflow. This allows a high level of multiplexing in a single experiment and results in a significant decrease in sample-to-sample variation.

The small well size means that less sample can be used, as little as 100 µL per hybridisation, compared with the 150 µL required for corresponding cartridge-based experiments employing the same array size. With further optimisation of array-packaging design, the reaction volume can be reduced to 30 µL or less, which will reduce cost accordingly. To process an equivalent number of samples on GeneChip cartridges, a lab would not only have to dedicate extraordinary labour resources, but would require additional fluidics stations and multiple scanners as well - technologies that have been incorporated into a single HT microarray system. Modelling studies have determined that the automated system could process 20 plates or more a week, an equivalent of 1920 samples - previously an unimaginable thought for micro-array analysis.

Microarray technology has already revolutionised significant parts of the drug discovery process, but with the development of HT arrays, pharmaceutical companies can now more fully implement and apply the technology. For example, at the beginning of the process, HT technology can play a role in disease pathway identification and validation, and later on, once a target has been identified, in compound screening and lead optimisation. Researchers can then use the HT microarray system to manage clinical trials, potentially expediting the delivery of new drugs to market.

HT array analysis provides researchers with a cost-efficient way to use genome-wide expression profiling to identify drug targets and pathways for complex disease mechanisms. Additionally, GeneChip DNA analysis arrays have been used to discover the genetic basis of disease by mapping disease genes with whole-genome single nucleotide polymorphism (SNP) assays. The two platforms complement each other: gene expression arrays identify differentially regulated genes from related individuals, and DNA analysis arrays can validate those differences in fine mapping experiments.

target evaluation

Once a disease pathway is identified researchers need to validate it, and verify that disrupting the pathway will actually affect the disease. Using whole-genome expression profiling, scientists can understand a wide range of effects - desirable and undesirable - that result from disrupting a pathway, and are then able to better evaluate a potential target for drug design. As researchers manipulate a large number of genes in the validation process, they can use HT arrays to analyse simultaneously the effects of each manipulation on global gene expression. Furthermore, the system can be used for microarray-based resequencing efforts to economically pinpoint disease-causing mutations and genetic variations in large clinical populations.

After identifying and validating a target, drug researchers can use HT arrays to screen libraries of compounds to identify those that disrupt expression of intended disease genes. Whole-genome expression analysis also identifies other changes in gene expression, such as 'off-target' effects, some of which may suggest the compound produces far too many side effects to be effective. For instance, if the changes in gene expression match those of a known toxin, the compound could be eliminated from the screening process early in development, saving both time and money. On the other hand, recording off-target changes in expression may help identify treatments for other diseases, operating through a different mechanism. Despite their development to treat hypertension and depression, the respective successes of Viagra for erectile dysfunction and Wellbutrin for smoking cessation are prime examples of exploiting off-target drug action to serve other therapeutic markets.

more information

By providing more complete genetic and genomic information, microarrays are helping researchers classify disease markers, predict drug efficacy, and more successfully manage clinical trials. While the throughput and cost-efficiency of the HT system are key to industrialising microarray technology, there are already more than 40 examples of microarrays being used in large-scale trials.

A recent Phase III clinical trial by Novartis Pharmaceuticals used expression profiles to predict the success or failure of Glivec/Gleevec treatment on chronic myelogenous leukemia.⁴ Researchers analysed gene expression patterns from patients prior to treatment and found a 31-gene 'no response' signature, which predicts a 200-fold higher probability of failed therapy. Similarly, in a Phase II clinical trial conducted at the Dana Farber Cancer Research Institute for Millennium Pharmaceuticals' drug Velcade, researchers used GeneChip arrays to collect pharmacogenomic data from myeloma patients treated with the drug.⁵

They discovered a pattern consisting of 30 genes that correlate with response or lack of response to therapy. Clinical utility of biomarkers will be further assessed in a Phase III trial.

While much progress has already been made using gene-expression analysis, studies to identify genes associated with drug response, efficacy, and toxicity may become one of the most promising applications for whole-genome DNA analysis. Tools like the GeneChip Mapping 100K or 500K Array Set (which can genotype more than 100,000 or 500,000 SNPs distributed across the genome) now allow researchers readily to genotype large populations of responders and non-responders to a given drug for phenotypes including efficacy and toxicity.

With these kinds of genetic studies, scientists hope to elucidate the genes contributing to variable drug response. In key Phase III trials, microarray genotype analysis could be used to stratify patient populations to eliminate poor or toxic responders. Such stratification would help ensure maximum effectiveness through clearer statistical differentiation between drug and placebo, while also reducing trial size and costs, and improving the odds of drug approval. Once a drug is on the market, patient stratification could also be used to accelerate drug expansion into new indications through faster, smaller, and more definitive Phase IV trials or to establish medical superiority of a late-to-market drug relative to entrenched competitors in an important class of patients.

future research

Genome-wide genotype information will also fuel future research. By better understanding genetic mechanisms of drug response in patients, researchers will have made significant progress on finding next-generation drugs.

As microarray technology advances and more content can be placed on smaller-sized arrays, the application of HT microarray systems to pharmaceutical development will become even more significant and will extend beyond the traditional genetic and genomic experiments.

The ability to use microarrays representing the complete coding content of the human genome - more than 47,000 transcripts - will help accelerate discovery. Human transcriptome analysis (i.e. the complete collection of transcribed elements of the genome) is also made possible by the HT system, where an experiment can now be constructed to analyse an entire genome for structure-function relationships on a single plate. Similarly, advances in genotype analysis will be accelerated by micro-arrays that can analyse more SNPs and can sequence larger parts of the genome. Running 10 plates a week, the throughput afforded by HT analysis allows for previously inconceivable experimental scale (see table): 960 whole-genome expression profile scans, 2.5 complete human transcriptome scans (with probes positioned every 5 base pairs), or analysis of nearly 35 million SNPs.

Efforts such as these are helping researchers use the genome sequence to improve pharmaceutical research and develop new therapies for improved disease management. While the benefits of HT array analysis are only beginning to be realised, with the care taken to fit this technology into existing infrastructures, it offers the prospect of more efficient, cost-effective and personalised approaches to patient care.