Editor

For years, the drug industry has known its cost of developing a new product – estimated at $1.3 billion – is not sustainable. For years, companion diagnostics have been considered as a potential solution to that problem. And for years, the result has been a lot of talk, but very little action.

Until now. Last month's FDA approval of melanoma drug Zelboraf (vemurafenib, Daiichi Sankyo Co. Ltd. and Roche AG) and lung cancer drug Xalkori (crizotinib, Pfizer Inc.) proved the point: both drugs are targeted, both were approved alongside companion diagnostics, and both advanced from Phase I trials to market in about five years.

While most drugs languish in clinical trials for upwards of a decade, Zelboraf and Xalkori were able to move quickly, largely due to the fact that both were paired with companion diagnostics very early in their clinical development, allowing them to find the right patient population, establish efficacy and present a convincing case to regulators.

"We're really at the tipping point of pharmaceutical companies seeking out diagnostics companies earlier in development," said Kathryn Becker, director of Abbott Molecular Oncology.

It's been a long time coming.

"If you look back almost a decade ago, to the sequencing of the human genome, people thought things would happen almost immediately," said Abbott spokesman Don Braakman. "It's been a longer, more costly process than anticipated."

Everyone points to Roche's breast cancer drug Herceptin (trastuzumab) and its accompanying HER2 tests as a shining example of success in personalized medicine, but few have followed in its footsteps.

In some cases, the specific mechanism of a drug is not fully understood early enough in the development process to improve clinical trial stratification. In other cases, the mechanism is known, but drug makers are reluctant to limit their potential market to patients with a specific genetic mutation. In still other cases, drug companies use lab-developed tests to find the right patients in the clinic, but when it comes to translating those tests into an FDA-approved diagnostic – well, that's someone else's problem.

Even Herceptin was not developed in tandem with a companion diagnostic, Abbott's Becker noted. But what companies seem to be realizing is that – even beyond the fact that companion diagnostics result in better treatment for patients and more efficiency for the healthcare system – they can also save money in R&D, especially if you incorporate them from the beginning.

That's what Zelboraf and Xalkori illustrate.

Case Studies in Efficiency

Zelboraf, originally developed by Plexxikon Inc., targets tumors that carry the BRAF V600E mutation. BRAF mutations are thought to occur in about half of melanoma patients and in 8 percent of solid tumors. Plexxikon understood the biology of its drug and approached Roche about developing a companion diagnostic in 2005, even before Zelboraf had entered the clinic.

Roche liked what it saw. In 2006, the big pharma signed on as Plexxikon's partner in a $706 million deal and geared up to start human trials. Phase I data showed a stunning 81 percent response rate, prompting the initiation of simultaneous Phase II and Phase III trials in 2009. (See BioWorld Today, Oct. 5, 2006, and Sept. 30, 2009.)

By late 2010, Roche had good Phase II data, and in early 2011, the Phase III study was halted early due to strong efficacy. Daiichi Sankyo bought Plexxikon a few months later, and approval applications were filed. The FDA, seemingly determined not to be the wrench in this well-oiled machine, delivered its approval two months earlier than anticipated. (See BioWorld Today, Nov. 8, 2010, May 12, 2011, and Aug. 18, 2011.)

The story of Xalkori's development is similar. Pfizer scientists discovered the molecule in 2005 and advanced it into the clinic in 2006. Although they knew Xalkori targeted cMET and ALK, it wasn't until 2007 that researchers identified the EML4-ALK mutation as a cause of non-small cell lung cancer. That finding "allowed us to link the biology to what we were seeing in the clinic," said Mace Rothenberg, Pfizer's senior vice president of clinical development and medical affairs in the Oncology Business Unit.

Rothenberg explained that the company had already seen three patients respond well in its Phase I trial. When subsequently tested, all three had the EML4-ALK mutation. Pfizer shifted its focus on Xalkori's development from cMET to ALK, but since the EML4-ALK mutation only occurs in about 3 percent to 5 percent of non-small cell lung cancer patients, the pharma knew it would need a companion diagnostic.

"Focusing your development on a genetically characterized subset of patients in Phase I is very unusual," Rothenberg said. But that was Pfizer's plan. The pharma started evaluating potential diagnostic partners, signing a deal with Abbott in 2009. By 2010, the companies had Phase I/II data showing a nearly 60 percent complete or partial response rate in patients with the EML4-ALK mutation. The FDA approved the drug – again ahead of schedule – based on data from two single-arm Phase II trials. Two confirmatory Phase III trials are underway.

Becker noted that when Pfizer and Abbott first teamed up in 2009, their goal was to see Xalkori and its companion diagnostic approved in 2012. Rothenberg added that the fact that the approval came a year early "shows what can be accomplished when government, academia and industry work together" and "sets the bar" in terms of efficient development of targeted agents.

Beating Standards

That Zelboraf and Xalkori advanced from Phase I in 2006 to FDA approval in 2011 is nothing short of astounding compared to standard drug development timelines.

Vertex Pharmaceutical Inc.'s Incivek (telaprevir), approved earlier this year for hepatitis C, took 20 years and $4 billion of investment to get to market. Human Genome Sciences Inc.'s lupus drug Benlysta (belimumab), also approved earlier this year, and Dendreon Corp.'s prostate cancer vaccine Provenge (sipuleucel-T), approved last year, each spent more than a decade in the clinic. (See BioWorld Today, April 30, 2010, March 11, 2011, and May 24, 2011.)

Incivek, Benlysta and Provenge are not examples where a companion diagnostic would have been relevant, but they illustrate the significant time and expense required to get new drugs to market.

When it comes to targeted drugs and companion diagnostics, codevelopment of the two might not just be faster and cheaper – it might be required.

Last year, an FDA panel recommended the agency postpone its decision regarding approval of ChemGenex Pharmaceuticals Ltd.'s Omapro (omacetaxine) for chronic myeloid leukemia patients with the T315I genetic mutation until a diagnostic test for that mutation was also available. The biotech had used a variety of assays to detect the mutation during clinical trials, but such assays can't be used as diagnostics. (See BioWorld Today, March 23, 2010.)

ChemGenex later decided to drop its focus on the T315I mutation and pursue approval in the broader population of CML patients who have failed prior treatments. (See BioWorld Today, July 15, 2010.)

Earlier this year, the FDA released a draft guidance laying out a tentative regulatory pathway for companion diagnostics. Although separate divisions will review companion diagnostics and drugs, the agency plans to conduct the reviews simultaneously, as it did for both Zelboraf and Xalkori. The guidance also noted that, in some cases, the drug may be approved ahead of the diagnostic, but the agency's preference appears to be for co-development.

Becker noted that having the FDA's support and guidance was "a big piece" of what allowed the review of Pfizer's Xalkori and Abbott's companion diagnostic ALK test to progress so quickly. The FDA worked closely with both Abbott and Pfizer to "help us navigate this new path," she said.

In the past, diagnostics companies and drug companies developed their products individually because the "process wasn't in place" to bring them together, Becker said. But with the FDA guidance and some examples that prove the system can work, "I think we'll see more and more partnerships [between drug developers and device makers] occurring earlier and earlier," she predicted.