Science Editor

For the minority of lung cancer sufferers that do respond to the EGFR kinase inhibitor erlotinib (Tarceva, OSI/Genentech), the results can be dramatic - for a time.

"No one is cured," Pasi Janne told BioWorld Today. "Every patient that has a dramatic response . . . will eventually relapse."

The reason, he said, is that the development of kinase inhibitors predates a thorough understanding of how they do what they do. Erlotinib as well as gefitinib (Iressa, AstraZeneca plc) were developed in the 1990s; Janne pinpointed 2004 as the year that an understanding of kinase mutations was reached, and the major resistance mutation - T790M, which is responsible for about half of all resistance cases - was identified in 2005.

What that means is that anti-EGFR kinase drugs "were never optimized" in terms of their binding characteristics, Janne said. And as a result, compounds that are now in clinical development against resistance mutations are "quite poor" at inhibiting them.

In the joint Dec. 24/Dec. 31, 2009, issue of Nature, Janne and his colleagues reported that by using a different chemical backbone than what Iressa and Tarceva have, they were able to develop two compounds that are effective against resistant EGFR kinases, while only weakly binding wild-type EGFR - a binding profile that the researchers hope will lead to less toxicity in the clinic.

As with Gleevec (imatinib, Novartis) and the bcr-abl kinase, Iressa and Tarceva prevent binding of energy-providing ATP to the EGFR kinase. But the mutations that render the drugs ineffective are different in each case. While bcr-abl mutations by and large leave Gleevec unable to bind, EGFR kinase mutations allow EGFR kinase to bind ATP even while bound to drugs.

In their work, senior author Janne, who is assistant professor of medicine at Harvard Medical School's Dana Farber Cancer Institute, and his team focused on finding a drug that not only prevented mutated EGFR kinase from binding ATP, but that did not bind to wild-type EGFR. "We specifically focused on drugs that inhibit the resistance mutations," Janne said. The most effective compound they identified, WZ4002, bound to resistance mutations 30 to 100 times more strongly than inhibitors with a quinazoline backbone. But it also bound to wild-type EGFR 100 times more weakly than such inhibitors.

The team first synthesized a series of compounds that were made up of a kinase inhibitor core with side chains that computer modeling predicted would inhibit mutant EGFR kinase. They then screened the two compounds with the greatest selectivity for mutated EGFR kinases, WZ3146 and WZ4002, against multiple kinases to study their selectivity. WZ4002 was further tested in animals, where it inhibited the growth of xenografted lung tumors.

Janne said his hope is to get the new compounds into clinical trials "as quickly as possible." To further that goal, the compounds have been licensed to start-up Gatekeeper Pharmaceuticals. Gatekeeper, which was founded in April and expects to secure funding shortly, hopes to have the compound in the clinic within 18 months; Gatekeeper President and CEO John Chant told BioWorld Today that Gatekeeper "would like to partner with a pharmaceutical company" to achieve that goal.

Janne noted that even if the new compounds live up to their early promise as they move through development, they are still not cures. He believes that a single agent is unlikely to ever be such a cure: "Cancers are malleable," he said. "If you suppress one pathway hard enough, something else will crop up."

What might be possible, he added, is to develop a cocktail of drugs that simultaneously cuts off alternate routes of escape. Such cocktails might ultimately provide a win against cancer, instead of the prolonged stalemate that even the best kinase inhibitors currently achieve. "But you have to take it one step at a time."