The International Genomics Consortium’s first large-scale gene expression project is picking up steam. Last week, the IGC, a non-profit organization based in Phoenix, Ariz., said that genetic security firm First Genetic Trust of Chicago will build and handle a banking system for its planned Expression Project in Oncology, expO.
ExpO’s initial goal is to collect 10,000 tumor tissue samples spanning a large number of cancer types, as well as 1,000 normal tissues, from various academic institutions over the next three years, then perform gene expression analyses on the samples, and to make the data freely available to the public. So far, 18 academic cancer centers – including Johns Hopkins and the MD Anderson Cancer Center – have signed on, and tissue collection is slated to begin early next year, according to Mike Bittner, a cancer researcher at NHGRI and an advisor to the IGC.
One of First Genetic’s roles will be to provide different degrees of data access to the parties involved, he said. The expression data would be stored in a format compliant with MIAME, and made public together with anonymous medical records and outcome data, according to Bittner. “It is becoming quite clear that the ability to get well-characterized samples with good medical information associated with them . . . is a terrible bottleneck for research, in any ’omic’ format,” he said.
However, several questions need to be solved before expO can really take off. For a start, IGC’s current commitments total less than half of the $35 million it aims to raise by the end of this year. According to Bittner, the consortium is discussing sponsorship opportunities with pharmaceutical companies, research foundations, as well as the National Cancer Institute.
Also, the microarray platform or platforms to be used for the gene expression analyses have yet to be chosen. Affymetrix, Agilent, and Amersham each have a chip in the fire, after their technologies were used in a three-month pilot study the IGC conducted last winter on a series of brain tumors. However, “it was clearly written into all agreements that this would not in any way guarantee selection of their platform,” said Bittner.
The data of the pilot study, which will not be published, is available to prospective funding partners. But the chip companies have also been courting the IGC in other ways: Affymetrix last year sponsored one of its employees, Grace Colón, to work as an interim COO for IGC for a few months; Agilent has sponsored legal help; and Motorola Life Sciences – which recently sold its CodeLink Array business to Amersham – had promised its help in laboratory management in the future.
In the end, more than one platform might be the lucky winner. “It’s still an open question as to what the strength[s] and weaknesses of all the different approaches are,” said Bittner. “But obviously, there are cost and preference issues.” The tissue providers, the funding partners, and the IGC all have an interest in choosing the technology used, he said, and the final decision would probably be made by an operations board. “This is all conceived along the same general lines as the SNP Consortium,” Bittner said.
The Issue of Consent
Another issue will be to get consent from all the institutional review boards of the participating institutions, which have to approve the use of medical samples for research purposes.
“[This] has been fairly problematic for anyone who wants to do a broad sampling because frequently you have to deal with two or even three IRBs within a single institution to get approval, and they all have different requirements,” said Bittner.
To change that, the IGC is currently working on uniform standards that might be adopted by all the IRBs.
Technical procedures for processing the tissue samples already exist, Bittner said. The plan is to have one IGC employee at each participating center, who would collect the tissues, obtain the consents and relevant medical records, and send the samples to IGC in Phoenix.
At a central laboratory, the samples would then be processed and their expression profiles measured. “If it’s done institution by institution, everyone does things a little bit differently, and that’s still a barrier in expression profiling,” said Bittner. A standard method for isolating the RNA has already been devised and was tested during the pilot study, he said.
Though gene expression analysis will be expO’s first priority, other studies might follow. Samples of the isolated RNA will be kept for future improvements in expression profiling, and “we will retain part of the [tumor] sample purely for the totally unknown,” Bittner said.
Long-term, the IGC has even more ambitious aims: “It’s also interested in doing this kind of study across other types of disease,” Bittner said.
“ExpO is a module, and we would like to create other modules.”