By Turna Ray
A lively discussion ensued at a conference at Harvard Medical School last month about the timing and the confluence of events leading to the restricted marketing of a class of colorectal cancer drugs in a genetically defined subpopulation.
The dialogue, which took place at a meeting hosted by the Partners HealthCare Center for Personalized Genetic Medicine and involved pharma execs, academics, and regulators, presented a snapshot of what's really driving big pharma to get involved in personalized medicine and revealed that the industry is still unwilling to let go of the blockbuster model.
The debate, which at times was contentious, was inspired by a white paper circulated at the event, titled "Managing Drugs on the Forefront of Personalized Medicine: the Erbitux and Vectibix Story," in which Richard Hammermesh, Raju Kucherlapati, and Rachel Groden of Harvard University laid out the events leading up to European and US regulators recommending that Amgen's Vectibix and Bristol-Myers Squibb/ImClone's Erbitux should be given to colorectal cancer patients without certain KRAS mutations.
The events prior to the restricted marketing for Vectibix and Erbitux in Europe and the US, which began in 2008 and 2009, respectively, raised questions about what Amgen and Bristol-Myers Squibb/ImClone knew about the drugs and when. Conference participants were challenged to consider whether the timing and motivations behind the companies' decision to pursue a more limited indication for their drugs mattered from a business standpoint, as well as an ethical one.
The one-month time frame between when the European Medicines Agency rejected Amgen's marketing application for Vectibix, in May 2007, and when the company resubmitted its application with KRAS-related drug response data, in June of that year, was the most salient fact many conference participants upheld as a tell-tale sign that Amgen initially went for a broader indication for a drug it knew would not be efficacious in 40 percent of the general population.
"If they knew in June, then I don't see how they couldn't have known in May," said one speaker. In an effort to inspire frank discussion of the topic, speakers were not required to reveal their names or affiliations.
To that, a representative from DxS, the company that holds marketing deals with both Amgen and BMS for a companion diagnostic to their colorectal cancer drugs [see PGx Reporter 09-02-2009], suggested that while Amgen may have known KRAS' association with Vectibix response, the company didn't have the "level of evidence" on the subpopulation necessary at the time of the initial EMEA submission.
The presentation of the facts raised a number of questions during the meeting: Would Amgen have pursued a limited marketing population for Vectibix if EMEA had initially approved the drug? What were the motivations driving ImClone to pursue a similar labeling change for Erbitux? And how would this example impact other drug companies pursuing personalized medicine?
Although the answers to these questions differed based on which healthcare sector the conference participants were representing, the debate over the Erbitux and Vectibix story illustrates how the drug industry's ongoing desire to pursue the broadest indication for its drugs can sometimes result in missed business opportunities and questionable ethics.
[ pagebreak ]
Missed Markets and Blown Opportunities
Kucherlapati, author of the case study and professor of genetics at Harvard Medical School, lauded Amgen for the way it handled the EMEA's rejection of the Vectibix application and for voluntarily pursuing a labeling update with the US Food and Drug Administration, where the drug was already being marketed broadly in the colorectal cancer population for several years. Whether Amgen knew at the time of the initial EMEA filing about the impact of KRAS mutations is "irrelevant from a business standpoint," Kucherlapati asserted.
But according to some industry observers, early knowledge of which patients best respond to a drug can be a significant advantage in a crowded market, and Amgen missed this opportunity.
In early 2007, worldwide Erbitux sales had already reached blockbuster status, nearly $1.1 billion. Meanwhile, global sales for Vectibix in 2007 totaled $170 million.
"Amgen completely blew it," said a financial analyst at the Boston meeting. By waiting too long to market Vectibix in a genetically defined subpopulation, Amgen missed the chance to differentiate its product from Erbitux in the colorectal cancer market, the speaker said. When finally the issue of updating the drugs' labels came before an FDA advisory committee last year, it was too late to differentiate Vectibix from Erbitux, and the two drugs ended up with essentially the same label, the speaker said.
The FDA's class labeling update for EGFR-inhibiting monoclonal antibodies for colorectal cancer treatment came one year after the EMEA recommended Erbitux and Vectibix be marketed in non-KRAS-mutated patients. The agency's action followed an advisory committee meeting that discussed the conditions under which the FDA should accept retrospective data on genetic subpopulations from drug developers ImClone and Amgen [see PGx Reporter 07-22-2009].
The fact that sponsors marketing competing pharmaceutical products would choose to go together before an FDA advisory committee was characterized as "an unusual move" by the Harvard Business School paper that sparked the discussion at the meeting.
Alison Ayers, Pfizer's VP of Oncology Worldwide Commercial Development, pointed out that the sponsors of Erbitux and Vectibix had no choice but to re-introduce their drugs in a subpopulation of colorectal cancer patients after definitive clinical data came out at the 2008 American Society of Clinical Oncology's annual meeting.
At the ASCO annual meeting in 2008, researchers reported results from a large, multinational prospective trial showing that patients with metastatic colorectal cancer whose tumor carries the wild-type version of the KRAS gene were much more likely than patients with the mutated form of the gene to benefit from Erbitux. The results from the study led Eric Van Cutsem, lead author and professor at the University Hospital Gasthuisberg in Leuven, Belgium, to recommend that “KRAS testing be routinely conducted in all colorectal cancer patients immediately after diagnosis to ensure the best treatment strategies for the individual patient" [see PGx Reporter 06-02-2008].
The presentation of this data resulted in similar recommendations in treatment guidelines issued by ASCO and the National Comprehensive Cancer Network.
Back at the meeting in Boston, Pfizer's Ayers hinted that the Erbitux/Vectibix labeling update story displays poor pipeline planning by Amgen, BMS, and Merck. The alternative for many colorectal cancer patients with KRAS mutations is Avastin. So, without an alternative to Vectibix and Erbitux in their pipelines, these sponsors are essentially handing over business to Genentech.
Several industry representatives felt that Amgen and ImClone should be commended for voluntarily pursuing limited indications for their drugs. Others pointed out that the companies pursued labeling changes only when their backs were to the wall, suggesting that industry does not believe personalized medicine to be a significant revenue driver yet.
Ethics of Knowing
The prevailing feeling at the meeting that Amgen pursued the broadest population for Vectibix in Europe despite knowing that a large portion of patients wouldn't respond due to genetic mutations is perhaps most troubling when considered from an ethical standpoint.
The EMEA's rejection of Amgen's Vectibix application was "especially surprising" since the FDA had approved the drug in September 2006, the Harvard Business School paper noted.
[ pagebreak ]
One industry representative suggested that since KRAS mutations confer limited response in patients and not heightened risk of adverse effects, Amgen wasn't harming patients by going for all comers in the colorectal cancer population.
To that, another speaker countered that putting cancer patients on a drug that they will not respond to is harming them and therefore unethical. "That's taking away time that they can be on a drug that does work for them," the speaker said. "They can die in that time."
Amgen and BMS were both asked to shed light on the questions surrounding the timeline of their marketing applications for Vectibix and Erbitux. Amgen did not respond to questions for this article. BMS issued a statement reflecting that the way the events played out in re-labeling Erbitux contributed to the collective knowledge on the impact of KRAS mutations on drug response, and the process benefited all players involved, including the FDA, drug companies, and physicians.
"Our goal is to make sure the right patient receives the right treatment at the right time," the company told Pharmacogenomics Reporter. "As the science on KRAS progressed, [the FDA advisory committee meeting held a year ago] allowed for an important discussion that would help determine the most appropriate metastatic colorectal cancer patient population eligible for Erbitux therapy based on biomarker selection.
"It was also an opportunity for industry, physicians, patients, and the agency to collaborate and learn more about the role of the retrospective KRAS data that were currently available," the spokesperson added.
Lawrence Lesko, director of the Office of Clinical Pharmacology at FDA's Center for Drug Evaluation and Research, has acknowledged that the agency's experience reviewing retrospective data to update the labels for Erbitux and Vectibix provides a framework for other drug companies considering personalizing treatments with genomic biomarkers [see PGx Reporter 10-07-2009].
DxS CEO Stephen Little has said that the protracted regulatory debate in the US will encourage pharma companies to engage in drug/diagnostic partnerships much earlier in the drug development process [see PGx Reporter 07-22-2009].
'We All Want Blockbusters'
A few years ago, it was difficult to find a pharmaceutical company that would readily admit it was using pharmacogenomics to personalize drugs in its pipeline. Back then, a tenuous association with the very term "pharmacogenomics" for a drug company seemed to be an open acknowledgement that it had drugs in its pipelines with limited efficacy and serious adverse reactions.
However, with the dwindling number of new drug candidates in pipelines and approximately $65 billion worth of pharmaceuticals representing 25 percent of drug sales expected to go off patent between 2008 and 2013, most major pharma companies have publicly embraced the idea of using pharmacogenomics strategies to develop more efficacious and safer drugs, and to run more efficient clinical trials by reducing the number of products that fail late-stage studies [see PGx Reporter 03-28-2007; 04-04-2007].
Although pharma companies seem to be warming up to the idea of pharmacogenomically guided personalized medicine, one speaker during the meeting pointed out that the facts surrounding Vectibix's and Erbitux's re-labeling illustrate that big pharma is still in enamored with the blockbuster strategy.
A high-ranking official from Eli Lilly, which in October 2008 purchased Erbitux-maker ImClone for more than $6 billion, admitted that big pharma still dreams of finding the next Lipitor.
"We all want blockbusters," said the Lilly official. "We want them to show up in the Christmas stocking. But I'm not sure they exist anymore."
The Lilly official attempted to dispel the feeling in the audience that the sponsors of Vectibix and Erbitux unethically sat on data about the impact of KRAS mutations on drug response. The official pointed out that during early clinical trials it is necessary for drug companies to collect a diversity of data, since it is impossible to anticipate all the potential problems that will arise with a given application and what regulatory authorities will ask for. So, if they don't do it right the first time, they can go back and do it over if necessary, the official said.
Another industry representative maintained it was "unreasonable" to vilify a for-profit drug company for attempting to market a drug in the broadest possible population.
In the end, however, most discussants agreed that although Amgen and BMS/ImClone may have had to be prodded by regulatory authorities and face certain business realties before marketing their drugs in a smaller number of patients, other companies will learn from their experience and be more willing to personalize care in best responders much earlier.
Ultimately, these lessons learned bode well for the future of personalized medicine, most conference participants concluded.