BOSTON — Researchers at the Center for Applied Proteomics and Molecular Medicine at George Mason University led by center co-director Lance Liotta plan to start two clinical trials to test if their reverse-phase protein array can identify patients who respond to EGFR inhibitors.
Liotta said he expects the method, which detects specific phosphorylated proteins in tumor samples, can avoid the ambiguities of fluorescent in situ hybridization and immunohistochemistry by spotlighting activated enzyme pathways. Ideally, this information can point physicians to a combination of drugs that can treat a particular tumor.
But although FISH, IHC, and mutational analysis are imperfect and indirectly indicate pathway activation, they happen to work well in certain circumstances, something clinical proteomics can't claim.
"The hope is to use [the technique] to functionally decide what is the best therapy to give a patient, and that's what we're hoping to do in clinical trials," Liotta told Pharmacogenomics Reporter.
Reverse-phase arrays contain spots of immobilized patient proteins taken from a particular tissue — in this case, tumor cells. Investigators probe the array with labeled antibodies specific to particular proteins or, as in the current example, activated states of particular proteins.
"We're having a lot of discussions with different companies about this. I'm sure versions of it will be out next year or so" as a research tool.
In comparison to antibody arrays, the technique has the advantage that its results are quantitative, it requires very little sample material, and it is very sensitive. Each spot on a reverse-phase array can correspond to tissue from a different patient or tissue, allowing a large number of samples to be interrogated at the same time.
The technique has a downside, too: The tissue being studied must be free of contaminating cells. In the case of tumor investigations, researchers commonly microdissect cancer cells from tissue samples in order to ensure quantitative results and some patient-to-patient comparability. Also, reverse-phase arrays give no information about the location of phosphorylated proteins.
The two clinical trials Liotta plans will use reverse-phase protein arrays to analyze tumor samples as a way to guide therapy with EGFR inhibitors "and other targeted inhibitors," Liotta said during a presentation at the Cambridge Healthtech Institute's Microarrays in Medicines meeting here last week. The trials will involve both primary and metastatic cancers that have already undergone therapy, he said.
"We've been using this technology in [clinical trial research] following guidelines equivalent to CAP/CLIA," which will allow CLIA labs to offer it as a service without additional hurdles if the technique works, said Liotta.
Liotta said he and his colleagues are "dedicated to get this out in use in clinical trials. … We're actually analyzing clinical specimens in the research setting today, and have everything worked out … to really do this," he said.
The technology is being developed at George Mason University, and no diagnostics companies have been involved yet, said Liotta. "We're having a lot of discussions with different companies about this. I'm sure versions of it will be out next year or so" as a research tool, he said.
In previous research, Liotta's group found EGFR to be specially phosphorylated in lung cancer samples that also carry an EGFR mutation believed to confer susceptibility to some inhibitors of this receptor. They also found downstream pathway activation, he said.
Where FISH, IHC, or gene-expression data can suggest that a particular pathway is active — by indicating genomic amplification of an oncogene or protein over-expression, for example — Liotta's proteomic method is one of a few techniques that seeks out direct evidence of pathway activation.
"If you have a gene [expression] array that shows that there's excess of some protein, that doesn't necessarily mean that that's going to be involved in a signaling cascade, because whether or not the entire cascade is connected and working is really what matters," said Liotta. "Even looking at the receptor amount is not going to tell you whether it's in use or not."
Armed with evidence indicating that specific pathways are active — such as phosphorylated proteins in the AKT pathway downstream from EGFR — clinicians would ideally be able to choose drugs to inhibit each one. "You want to see whether a drug target is working or not, and particularly before and after you give the therapy," said Liotta.
The technique may also prove useful in directing physicians to combination therapies, such as for amplifying the effects of some chemotherapies. In preliminary data from a study of breast cancer samples, Liotta and colleagues at the US National Cancer Institute found amplified Her2 — the accepted biomarker whose overexpression is required for clinicians to prescribe the drug Herceptin — to be associated with phosphorylated AKT, which suppresses apoptosis. When patients were treated with Herceptin, the level of AKT phosphorylation dropped. Those patients whose tumors grew back following chemotherapy showed ongoing AKT phosphorylation, said Liotta.
Thus, combining a drug that inhibits AKT phosphorylation with an agent promoting apoptosis might prove to be a useful strategy for enhancing drug effectiveness. The data "support the concept that one of the ways that Herceptin works is to augment apoptosis-inducing therapy," although the drug's mechanisms are not thoroughly understood, said Liotta.
Bullish on Proteomics
Clinical proteomics has made big promises in the past, and it has failed to deliver, said Jorge León, who heads the molecular diagnostics consulting firm Leomics. Liotta and George Mason colleague Emanuel Petricoin "certainly have conducted seminal research work" in proteomics, but they can be "very bullish" about the clinical applications of their research, León told Pharmacogenomics Reporter.
"Formidable" work that Petricoin and Liotta conducted on proteomic patterns related to ovarian cancer in 2001 was accompanied by an overly optimistic picture of its applications, as well as its sensitivity and specificity, said León. When the research could not be reproduced, the field of proteomics lost some credibility in the eyes of clinicians and clinical investigators, he said.
"One person's 'bullish' is another person's 'innovative,' so it depends on your point of view," said Liotta in response. "We'll just have to see what happens over time."
— Chris Womack ([email protected])