A Target for Iressa

The fall and rise (and fall) of a pharmacogenetics poster child

David Secko
Apr 1, 2006

It emerged a seductive idea: Bridging the gap between genetics and pharmacology could allow clinicians to hit cancer cells where it hurts them most. In May 2004, Iressa, known generically as gefitinib, rose from the ashes of poor clinical trials as an example of how this might work. But this early poster child for pharmacogenetics has yet to live up to its hype.

A tyrosine kinase inhibitor, Iressa blocks the epidermal growth factor receptor (EGFR), a protein overexpressed in many tumors. Results of clinical trials were disappointing in 2003, as most patients with non-small-cell lung cancer did not benefit from the drug. Roughly 10% of patients, many of them Japanese, considerably improved, however, as their tumors seemed to melt away. "The results were puzzling," says Matthew Meyerson of the Dana-Farber Cancer Institute in Boston. Everyone wanted to know: What was different about this 10%?

In the three Hot Papers featured...

Together the results positioned Iressa as a drug that works better in certain genetic contexts. People had thought about this before, but "nobody had a signpost for how to develop these drugs," says Haber. "These papers pointed to a predictable way to think about it." Still, disappointment returned for AstraZeneca, the drug's maker; last year, the Food and Drug Administration restricted Iressa's use. Its effectiveness questioned, Iressa nevertheless spurred a trend towards developing drugs based on preexisting mutations in the population.

Sensitivity Versus Resistance

Patrick Zarrinkar, from Ambit Biosciences in San Diego (a company that uses protein kinase screens to develop drugs), was immediately excited by the findings that various EGFR mutations, some of which are deletions and substitutions that cluster around the protein's ATP-binding pocket, correlated with Iressa sensitivity. "The first question that came to my mind is, how do these mutations work?" says Zarrinkar. "Two different hypotheses seemed possible," he adds. "Either the mutations affect target binding ... or they affect the signaling of the receptor."

Data derived from the Science Watch/Hot Papers database and the Web of Science (Thomson Scientific, Philadelphia) show that Hot Papers are cited 50 to 100 times more often than the average paper of the same type and age.

T.J. Lynch et al., "Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib," N Engl J Med, 350:2129-39, 2004. (Cited in 709 papers).

R. Sordella et al., "Gefitinib-sensitizing EGFR mutations in lung cancer activate anti-apoptotic pathways," Science, 305:1163-7, 2004. (Cited in 151 papers).

J. G. Paez et al., "EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy," Science, 304:1497-500, 2004. (Cited in 632 papers).

Work by Haber's group had already revealed that Iressa-sensitive EGFR mutations selectively activate anti-apoptopic pathways,2 suggesting these receptors promote survival signals. Zarrinkar and his colleagues then found that a set of the EGFR mutations does not affect Iressa binding to the receptor in vitro.4 "This suggests the effect of these mutations is through biology, not biochemistry," says Zarrinkar.

The realization that sensitivity to Iressa appears to be linked to anti-apoptopic signaling has put stock in a model called "oncogene addiction" for how these drugs work,5 says Haber, the idea being that certain mutations are more important to maintaining a cancer cell phenotype than others. "The argument is that, if Bcr-Abl mutations in leukemia, which Gleevec targets ... or the EGFR mutations in lungs cancer cells are taken away," says Haber, "these cells die."

Regardless, cancer cells are masters of evasion, and it is now apparent that mutations can work both ways. For example, in 2005, Bal#193;zs Halmos' lab at Harvard Medical School found that a secondary mutation in EGFR could cause Iressa-sensitive tumors to become resistant.6 "This mutation is interestingly at the same residue as the common resistance mutation for the drug Gleevec," says Zarrinkar. How these mutations exactly affect the structure/function of EGFR is a big question in need of an answer, he says.

Next Generation Inhibitors

<figcaption>Chest X-rays of a 56-year-old female who had never smoked revealed a mass in the right lower lung (A). Although marked decrease in tumor size occurred two months after gefitinib was initiatiated (B), the tumor progressed nine months later. (From C.H. Gow et al., PLoS Medicine, 2: e269, 2005.) Credit: © 2005 C.H. GOW</figcaption>
Chest X-rays of a 56-year-old female who had never smoked revealed a mass in the right lower lung (A). Although marked decrease in tumor size occurred two months after gefitinib was initiatiated (B), the tumor progressed nine months later. (From C.H. Gow et al., PLoS Medicine, 2: e269, 2005.) Credit: © 2005 C.H. GOW

Despite the molecular insights, the idea of Iressa as a pharmacogenetic drug is growing murky, as clinical data begin to reveal no survival differences in patients taking it. Although these results apply to all patients, and not just the 10% with EGFR mutations, they do suggest that "it's an open question whether gefitinib should be used in only people with particular mutations," says Meyerson, "since we don't really know what clinically determines who should benefit." Indeed, it's not clear what an EGFR-dependent cancer is exactly, in terms of secondary resistance mutations and other factors, says Meyerson, so how do we know who should get Iressa?

Nevertheless, Iressa-sensitive EGFR mutations brought to the fore what genetics can do for drug development. "In developing the next Iressa, the goal is to try and select ahead of time who a [drug] is likely to work on," says Bruce Johnson, from the Dana-Farber Cancer Institute.

Still, some important pharmacogenetics questions remain unanswered, says Haber: How well do genetic mutations actually predict drug responses; whether drugs need to be linked to genetic tests; and whether pharmaceutical companies will lose interest if their drug treats only a subset of cancers. Still, the slightest hints of a pharmacogenetic drug discovery continue to excite. Pfizer's Sutent, a tyrosine kinase inhibitor, works in 30% of kidney cancers and was recently FDA approved. Haber muses: "Why this 30% and not the other 70%?"

dsecko@the-scientist.com

References

1. T.J. Lynch et al., "Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib," N Engl J Med, 350:2129-39, 2004. (Cited in 709 papers) 2. R. Sordella et al., "Gefitinib-sensitizing EGFR mutations in lung cancer activate anti-apoptotic pathways," Science, 305:1163-7, 2004. (Cited in 151 papers) 3. J.G. Paez et al., "EGFR mutations in lung cancer: Correlation with clinical response to gefitinib therapy," Science, 304:1497-500, 2004. (Cited in 632 papers) 4. M.A. Fabian et al., "A small molecule-kinase interaction map for clinical kinase inhibitors," Nat Biotechnol, 23:329-36, 2005. 5. I.B. Weinstein, "Cancer. Addiction to oncogenes - the Achilles heal of cancer," Science, 297:63-4, 2002. 6. S. Kobayashi et al., "EGFR mutation and resistance of non-small-cell lung cancer to gefitinib," N Engl J Med, 352:786-92, 2005.