Personalized Medicine Could Shake Up PharmaDavid Bogoslaw
In the game of pharma, companies may have to give up on the home run known as the blockbuster drug—products like Lipitor and Avastin that were marketed to large swaths of the population and generated billions in sales—and hope that well-timed singles and doubles can carry the day.
Drug manufacturers have been loath to surrender broad markets and efficient marketing campaigns for more targeted patient populations and, presumably, higher marketing costs. But the number of prospective blockbusters has dwindled—and some drug companies are slowly coming around to the potential benefits of targeting certain treatments to smaller patient populations who genetic tests show will be most receptive to them. The hope is that increased safety and quality assurance—and the possibility of moving drugs up into earlier phases of treatment—could bolster sales and help make up for narrower markets for each drug.
There's no certainty, however, as the markets for specific drugs shrink, that drug companies will be able to make up for the size of their lost blockbusters by developing enough new products to treat the same population their drugs previously treated, says Isaac Ro, a biotech analyst at Leerink Swann & Co.
At a time when the success of U.S. health-care reform legislation largely hinges on the ability to slash medical costs, the promise of saving billions of dollars by not prescribing drugs to patients who genetic testing shows won't respond to them could make a compelling case for an accelerated shift toward personalized medicine.
While the benefits of personalized medicine are already apparent to pharmacy benefit managers like Medco Health Solutions (MHS), the upside to the much larger world of drugmakers is just starting to be explored.
Greater use of genomic testing is enabling physicians to increasingly choose the best treatment for a patient with less guesswork. Drug manufacturers such as Genentech, acquired by Roche Pharmaceuticals (RO:SW) in 2009, and Amgen (AMGN) have started to develop companion diagnostic tests alongside new drugs early on in their research and development processes, whether in-house or with outside partners. Companion diagnostics can identify patients who aren't likely to respond to certain treatments based on genetic information. The U.S. Food & Drug Administration (FDA) so far has identified 32 distinct genomic biomarkers in current product labeling of assorted drugs.
Amgen partnered with Qiagen NV (QGEN), a Dutch producer of diagnostic tests, to develop a companion diagnostic for Vectibix, Amgen's colorectal cancer drug that was approved by the FDA for third-line treatment of metastatic, or late-stage, cancer. Prescribing the drug only for late-stage treatment offers a drug company a smaller market opportunity than drugs such as Avastin, which is approved for first-line treatment.
There's already evidence that using companion diagnostics to exclude patients based on genomic analysis can improve clinical trial data for a new drug and boost the odds of FDA approval for use in targeted patient populations, says Peer Schatz, chief executive of Qiagen. He believes enriching clinical data with the aid of biomarkers will become a standard process that will make it easier to market new drugs instead of limiting their market possibilities. And it may even enable certain drugs to expand their market size by seeking a higher standard of FDA approval as first-line therapeutics that would allow them to compete directly against existing first-line treatments like Avastin, he adds.
Narrowing Clinical Trials
Diagnostic tests based on biomarkers like K-RAS for certain mutations of colorectal cancer tumors will also make it possible for drug companies to salvage data from unsuccessful clinical trials and "turn past failures into future successes" by resubmitting drugs for approval for smaller subsets of patients, FDA Commissioner Dr. Margaret Hamburg said at a meeting of the Personalized Medicine Coalition in Washington on Feb. 25.
When Vectibix was first approved for the general colorectal cancer patient population in 2006, there was no selection for mutations of the K-RAS biomarker. Clinical trials were already under way for the use of Vectibix in early treatment when Amgen had hints in preclinical studies for approval of the drug for late-stage treatment that tumors with the K-RAS mutation wouldn't respond to the drug. The company altered the Phase 3 clinical trials for first- and second-line treatment approval to exclude these patients, and "those were the data that may potentially support the use of Vectibix" earlier in the treatment of a subset of colorectal cancer patients, says David Reese, executive medical director of the company's Medical Science Div. Roughly 60% of colorectal cancer patients have the "wild type," or normal form of the K-RAS gene that's been shown to respond to Vectibix.
As a third-line treatment, worldwide sales of Vectibix totaled $233 million in 2009. Amgen plans to file for approval this year with both the FDA and the EMEA, its European regulatory counterpart, for the drug to be used as a first-line treatment, says Reese. While Amgen expects expanded approval to result in a larger market for Vectibix, Reese wouldn't speculate on how much sales might improve. "A lot depends on what the FDA label ends up being," he says.
But cancer patient advocate Nancy Roach, chair and co-founder of the Colorectal Cancer Coalition, says she's not convinced that more compelling clinical trial results will make it any easier to market targeted drugs. "The data I've seen don't show Vectibix or other drugs are effective enough to compete with Avastin as a first-line treatment," she says.
Market size is a less telling metric by which to gauge the value proposition of developing drugs for subsets of patients than development costs, says Ross Muken, director of health-care services and technology research at Deutsche Bank Securities. Ten to 15 years ago, the cost to develop a new drug was $300 million to $400 million, vs. more than $1 billion today, he says. That's money down the drain if a drug fails to get regulatory approval.
Developing new drugs based on information about patients' genetic receptivity "allows that equation to normalize again and your cost of developing a new molecule will go down significantly," says Muken. "So your market size doesn't have to be as big."
Not spending money on drugs that have a poor chance of being approved increases a drug company's return on R&D spending, and would also be beneficial to HMOs and other insurers, who generally balk at higher-cost treatments, he says. According to Muken, Medco has already shown how personalized medicine can lower the overall cost of patient care, and insurers will realize the value of eliminating prescriptions to patients who have adverse reactions to ill-suited drugs.
In the case of the breast cancer drug Herceptin, Genentech knew from its market research early on that the market would be limited to women with the HER-2 mutation—20% to 25% of breast cancer patients. The drug was approved for HER-2-positive patients in the metastatic, or most advanced, stage in 1998 and approved for the adjuvant, or early diagnosis, stage in November 2006.
Genentech's U.S. sales of Herceptin jumped by more than 65% in 2006 to $1.23 billion even before early-stage treatment was approved by the FDA late that year. Use of the drug in early-stage cases rose on the strength of the clinical trial data presented at the May 2005 meeting of the American Society of Clinical Oncology, which provided the basis for subsequent FDA approval, Krysta Pellegrino, a Genentech spokeswoman, said in an e-mail message.
"Treating someone in the adjuvant setting—that's your best chance of curing somebody," Pellegrino said in an interview with Bloomberg BusinessWeek. "We're always looking at moving our drugs up the line" into earlier lines of therapy where it would be more beneficial to patients.
Some drug companies, while still shooting for blockbusters, are taking steps to ensure at least a smaller market for certain products if initial clinical trials aimed at the broader patient population end in failure. Pfizer has an ongoing large clinical trial for its new kidney cancer drug Sutent. Separately, the company has asked Genomic Health (GHDX), a developer of clinical laboratory services for cancer, to develop a prognostic test to predict the risk of recurrence in patients with renal cell carcinoma. That test could potentially identify a smaller population of patients who could benefit most from the drug, says Randy Scott, Genomic Health's executive chairman.
"If Pfizer hits a home run and Sutent works broadly in all early renal cancer patients or has dramatic effects, you'd want to give it to all patients as a precaution" after surgery to remove the original tumor, he says. But if not, Pfizer won't lose time in salvaging the drug for a more focused population, he adds. "It's a good example of the drug industry beginning to recognize that [sometimes] these drugs only work in a subset of patients and they need to do a better job with companion diagnostics."
Dr. Robert Epstein, Medco's chief medical officer, cites other kinds of drugs that have sold better when paired with a companion diagnostic test. One example is the generic drug abacavir, an oral treatment for HIV, to which 6% of patients had a hypersensitivity reaction. Once a certain gene was uncovered that predicts with more than 99% certainty which patients will have the reaction—and the findings were published in The New England Journal of Medicine in February 2008, urging doctors not to prescribe the drug to that subgroup—sales of the drug increased 28% worldwide over the next eight months, according to an article in the November 2008 issue of Personalized Medicine by Felix Frueh, who has since become head of personalized medicine research at Medco.
"You take the worry away [that the drug won't work] and the market share grows," says Medco's Epstein.
Genentech is always looking for biomarkers to help identify patients for its new drugs and builds biomarkers into all of its pipeline products, says Pellegrino. Amgen also touts its "extensive biomarker program," which it views as integral to its development process, but won't disclose how many other drugs it is developing in tandem with companion diagnostics. Bristol-Myers Squibb (BMY) says it's committed to biomarker research but wouldn't say whether that research is being built into its drug development programs.
For the FDA's part, Commissioner Hamburg said on Feb. 25 that she understands that "for [drug] developers to make a substantial investment in [more targeted therapies], they need clear guidelines setting out our expectations and standards for approval."