By Julia Karow
This article has been updated with additional information from Nicholas Schork, a RainDance collaborator.
As the first example of a targeted assay for its sequencing platform that may be used clinically in the future, Roche's 454 Life Sciences last week launched kits for genotyping HLA genes at high and medium resolution.
Independently, RainDance Technologies this week launched a kit to amplify a large panel of HLA genes using its microdroplet PCR platform and to analyze them by next-gen sequencing. It developed the assay in collaboration with Expression Analysis.
The human leukocyte antigen, or HLA, locus encodes more than 200 genes with key roles in the immune system. It is highly polymorphic, with approximately 6,000 known alleles. HLA genotypes are associated with autoimmune and infectious diseases, as well as some cancers, and HLA typing is important in transplant medicine for matching donors and recipients.
The two 454 kits — called GS GType HLA MR and HR — are available immediately for research use only. They are the first in a series of assays 454 plans to launch in other disease areas including immunogenetics, infectious disease, and cancer.
According to the company, the assays offer several advantages over HLA typing by Sanger sequencing including increased throughput — depending on the 454 platform and resolution, between 5 and 80 samples can be analyzed per run — and savings in time and labor because polymorphisms can be linked, or phased, unambiguously within exons.
According to 454 CEO Chris McLeod, there is a "huge use" of HLA typing just in disease research, and there is potential for its use in other areas, such as forensics or population genetics.
The current level of resolution is "great for a lot of research applications," he said. The plan is "to build on this in the future to potentially launch a registered product" for clinical use, he added.
The company is not yet announcing when it plans to file an application with the US Food and Drug Administration, but McLeod said it aims to file for a sequencing platform, as well as an assay, both based on the current research versions. For many clinical applications of HLA typing, he said, the resolution of the assay needs to be higher than it is currently. The clinical sequencing platform is "not going to be totally new," but similar to the existing one.
According to McLeod, clinical HLA typing is "a material market" for 454 that today is primarily served by Sanger sequencing, although no registered diagnostic product exists.
Last month, Life Technologies said it has started clinical trials for HLA typing on its 3500 Dx Genetic Analyzer — a Sanger capillary sequencer — and plans to apply for 510(k) clearance of the platform and assay kits from the FDA (CSN 3/22/2011).
McLeod said 454 sees clinical HLA testing as a market "where we would be able to have significant advantages over Sanger sequencing." Compared to other next-gen sequencing platforms, 454's relatively long reads help to determine the phase of alleles unambiguously. "A lot of the other [platforms] have to do that virtually because their read lengths are just not long enough," he said.
454 and Roche have tested their HLA assay in collaboration with researchers at Stanford University, Technical University Dresden in Germany, Children's Hospital of Philadelphia, the Red Cross Transfusion Service of Upper Austria, Children's Hospital and Research Centers in Oakland, and the Institute of Immunology and Genetics in Kaiserslautern in Germany, and published the results in Tissue Antigens last month.
According to Elizabeth Trachtenberg, director of the HLA & Immunogenetics Laboratory at the Children's Hospital and Research Center Oakland, an author on the paper, one of the main advantages of the 454 platform is that it reduces the amount of ambiguity from heterozygous combinations of alleles. With Sanger sequencing, "we have to do a lot of secondary testing to resolve these ambiguous heterozygous combinations, and this new clonal sequencing drastically reduces that," she told Clinical Sequencing News, "which is really incredibly helpful for an HLA clinical laboratory. We spend at least a quarter of our time resolving ambiguities."
The cost per sample is also "significantly less than with Sanger sequencing," she said, because many samples can be processed in parallel — her lab currently sequences 11 loci for 40 samples per run — and because ambiguity is reduced, though she has not done an exact cost calculation yet.
To be sure, though Trachtenberg's 454 GS FLX runs in a CLIA laboratory, it is not used for clinical work at the moment but for disease research studies. "But as time goes on, we will use it for clinical trials and for patient samples," she said, probably even before the FDA approves a 454 instrument. While she would not be able to use Roche's reagent kits for clinical work, she said she could develop her own reagents.
[ pagebreak ]
Two years ago, several HLA testing labs were still skeptical about using the 454 platform for clinical HLA typing, citing the complex and labor-intensive sample prep, read lengths that were not long enough to phase polymorphisms across exons, and the need to pool many samples in a run to make the assay cost effective (IS 3/17/2009).
McLeod said the company has worked with users to automate the sample prep with liquid handling systems. Also, read lengths have improved since two years ago with the introduction of the Titanium chemistry and are now around 400 bases. Finally, with the introduction of the GS Junior, smaller numbers of samples can be processed in one run.
According to Trachtenberg, the sample prep is indeed "complex and involved," but with automation — her lab runs liquid handling robots from Tecan and Beckman Coulter — "it's definitely do-able."
Even longer reads would help reduce ambiguities further, she said, but "it's not critical to have those longer reads now."
Trachtenberg said that researchers are trying to develop HLA sequencing on the Illumina platform, which has shorter reads than the 454, but she has not yet seen any data showing that it is possible. "Illumina, of course, is much cheaper than 454, but they have not been able to show they can do these big gene complexes," she said, noting that whole-genome datasets generated on the Illumina platform omit the HLA locus.
Each of 454's research HLA assays comprises a primer set to PCR-amplify between one and three exons of up to 10 HLA genes, as well as software to correct errors and to optimize the 454 sequence data for use with third-party interpretation software. The kits come with four PCR microtiter plates with dried-down primers, each for 10 samples. While the medium-resolution assay requires one plate for 10 samples, the high-resolution assay requires two plates.
454 declined to disclose pricing for the HLA kits, but said it is "competitive with capillary sequence-based assays."
The PCR amplicons are designed to feed directly into the 454 sequencing sample prep workflow, either for the GS FLX or the JS Junior. At medium resolution, 80 samples can be analyzed per GS FLX run and 10 per GS Junior run. At high resolution, the FLX can handle 40 samples per run, and the Junior 5 per run.
Besides the HLA assay, 454 has several other targeted sequencing assays in development that "will have great utility in human healthcare," McLeod said, many for specific diseases. Each will consist of an optimized primer set and software. The plan is to launch these assays on a regular basis, approximately one every three to six months. Many of these assays will have the potential to be used in a clinical environment in the future, he said.
RainDance Technologies launched its own HLA gene amplification kit this week, called HLASeq Research Screening Panel, for use with its microdroplet PCR platform and subsequent analysis by next-generation sequencing. According to a company spokesperson, the panel was designed to work "equally well" with any of the current next-generation sequencing platforms.
Like 454, RainDance is marketing its HLASeq kit for research use only, and the company did not respond to a request about a potential future clinical use.
RainDance's kit covers more genes than either of 454's assays. HLASeq tiles amplicons across the entire 3.8-megabase HLA superlocus, covering all Class I and Class II genes as well as several minor HLA antigen genes. According to the company, it provides "the most extensive coverage of genes in the MHC region."
RainDance developed and validated the panel in collaboration with genomics service provider Expression Analysis, which offers services on RainDance's RDT 1000 platform and plans to include the HLASeq panel in its offering starting in May.
Expression Analysis CEO Steve McPhail told CSN via e-mail that the company sequences the HLASeq panel on the Illumina platform, which will also be used in the service.
Nicholas Schork, a professor of psychiatry at the University of California, San Diego, who has had early access to Illumina sequencing data for the HLASeq panel, said in a statement issued by RainDance that he has been "very impressed with the initial results ... particularly its coverage of the entire HLA super locus and the completeness of the data from all of the genes that encompass this complex region."
Schork told CSN by e-mail that his team assembled the reads and called variants and is still analzying the data. "Longer reads (like 454) do have advantages, especially for phasing the HLA region ... but there may be ways of overcoming problems with short reads by using more sophisticated bioinformatics tools."
McLeod said that RainDance's panel would compete with its own HLA assays, noting that the companies did not collaborate on HLASeq, as they are doing for another RainDance product, an ADME gene screening research panel (IS 5/8/2011).
The RainDance spokesperson said that its panel "is more complementary than competitive" to 454's assays because it was developed for researchers "looking more broadly at the [HLA] region," whereas the 454 assays "will be useful to researchers interested in only those particular areas" they cover.
Have topics you'd like to see covered in In Sequence? E-mail the editor at jkarow [at] genomeweb [.] com.