Latest Drugwonks' Blog
As Harry Truman opined, “I have found the best way to give advice to your children is to find out what they want and then advise them to do it.”
If only it were that easy with FDA advisory committies.
Matt Herper (Forbes) reports that an analysis of FDA advisory committee recommendations compared to actual FDA actions (from 2007 through 2010) shows that FDA followed adcomm advice 74% of the time. The study, by Concept Capital, looked at a total of 120 product-specific advisory committee votes and the ensuing FDA actions.
Interestingly, the FDA overruled “no” votes only three times: (Tarceva for maintenance therapy in lung cancer, Avastin for breast cancer, and Micardis to lower blood pressure.)
As Herper writes, “In other words, a no vote from an advisory panel is likely to stick, but a yes vote does not mean the product will be approved.”
“Advice is judged by results, not by intentions.”
-- Cicero
"Securing funding for "Regulatory Science" has been a top priority for Commissioner Margaret Hamburg, highlighted most recently during an October 6 luncheon address to the National Press Club, coinciding with a white paper outlining the agency’s vision for enhancing its science base.
The initiative may be the last hope for the agency to play a central role in encouraging drug development as a core part of its mission, after the underwhelming results from prior efforts like the Critical Path Initiative and the Reagan/Udall Foundation. For industry, "Regulatory Science" funding may be the only hope to secure a pool of resources for the agency to craft policy in areas like companion diagnostics, surrogate markers, etc. without another significant step-up in user fees."
The problem is what money that has been spent is being steered to academic institutions that are not collaborating with either other, with other industry consortium or things on the Critical Path opportunities list. Further, lip service is being paid to adoption of new tools and pathways even as the FDA puts out guidances that ignore or fail to encourage their use. The muddled 510k guidance discussion and recent FDA guidance on submission of Phase I safety data are examples of guilt by omission.
I find myself agreeing with Senator Tom Harkin who expressed concerned about the FDA's increasing set of duties and inability to establish common protocols to speed development of anti-infectives and vaccines within the bio-defense sphere:
Stressing that he was “just thinking out loud,” Harkin suggested that maybe “we need to take something out of FDA, something out of Defense, that would be put under BARDA, and let BARDA be the lead agency.” (Harkin previously noted the “hard work” he and others in Congress put into creating BARDA, and wondered how the new HHS initiative fit within that framework.)
Harkin noted his ongoing frustration with the failure to achieve licensure of cell-based flu vaccines. “FDA just—institutionally, I don’t know if they can do it,” he said. “It is just that they have so much to do and they have other responsibilities, and mostly they are focused on drugs that we take for illnesses.”
That's quite a statement. Unfortunately the organized interests in Washington are more interested in rent-seeking than getting things done.
Just thinking out loud...maybe everyone needs to redirect efforts towards the development of collaboration to produce the science and regulatory culture required to sustain advances in clinical development and the goal of increasingly effective use of products once on the market. That's what the Critical Path is and was about. Discussions about what can be done with fees collected under The Prescription Drug User Fee Act are starting.
www.fda.gov/ForIndustry/UserFees/PrescriptionDrugUserFee/default.htm
That might be a good time and place to introduce changes in priority and purpose.
www.nature.com/news/2010/101012/full/467766a.htm
Cancer-gene testing ramps up
Thousands to get personalized medicine in Britain's National Health Service.
Ewen Callaway
In an approach that many doctors and scientists hope will form the medical care of the future, Massachusetts General Hospital in Boston has for the past year and a half been offering people with cancer a novel diagnostic test. Instead of assessing tumours for a single mutation that will indicate whether a drug is likely to work or not, the hospital tests patients for some 150 mutations in more than a dozen cancer-causing genes, with the results being used to guide novel treatments, clinical trials and basic research. This form of personalized medicine tailors treatments on the basis of the molecular and genetic characteristics of a patient's cancer cells, potentially improving the treatment's outcome.
Now Britain is set to test whether an entire health-care system is ready for the approach. Plans were unveiled this week to deploy broad genetic testing for selected cancer patients in Britain's government-run health-care provider, the National Health Service (NHS). This form of 'stratified medicine' uses genetic information to group patients according to their likely response to a particular treatment.
"The United Kingdom is really the ideal place to do this," says James Peach, who heads the programme for Cancer Research UK, the charity that is leading the effort. As the NHS treats millions of people each year, unprecedented numbers of suitable patients could be enrolled in the genetic-profiling programme. "The idea is to scale this up to every patient in the NHS," says Peach. In its first phase, the programme will be rolled out to as many as 12,000 NHS cancer patients over two years, beginning in early 2011. By contrast, Massachusetts General has tested about 1,600 patients, and other hospitals' efforts each number in the hundreds.
The tests, which will look for several dozen mutations in about a dozen genes linked to cancer, will be carried out on people with lung, breast, colorectal, prostate or ovarian cancers, or metastatic melanoma, who are being treated at six NHS hospitals. Therapies that target specific tumour-causing mutations have already been approved, or are on the verge of approval, for most of these conditions, says Peach.
Testing a clinical sample for so many mutations at once is a challenge in itself. Because most existing clinical tests probe individual genes, the NHS programme is working with the Technology Strategy Board, a government agency that supports technology development, and several companies to design a customized test that detects all of these mutations in one go. The partnership, which includes the pharmaceutical multinationals Pfizer and AstraZeneca, will also design software to make the results useful to researchers and clinicians. By genotyping patients for a broad array of cancer-causing mutations, the new tests will make it easier to assign subjects to clinical trials, Peach says.
That is already happening at Massachusetts General, where the test is helping to establish clinical trials that wouldn't otherwise have happened, says Leif Ellisen, a geneticist who helps lead the hospital's cancer testing programme. For example, its broad genetic test detects a mutation in a gene called BRAF that is already known to be commonly mutated in metastatic melanoma. Finding such mutations in people with lung and colon cancer made it possible to put them in a trial of an experimental treatment targeting that gene, Ellisen explains.
Basic research should also benefit from the NHS programme, says Peach. Researchers will have access to consenting patients' genetic data as well as to medical records of the outcomes of the treatment. These data could reveal how drugs targeting one molecular pathway are affected by mutations in another gene, says Andy Futreal, a cancer geneticist at the Wellcome Trust Sanger Institute in Hinxton, UK, and an adviser to the programme.
Peach hopes that the first phase of the cancer programme will pave the way for expanding genetic testing to more patients and other conditions, such as diabetes, AIDS and even psychiatric disorders. Cancer offers a good testing ground for personalized medicine, because numerous targeted therapies already exist, but "there's no reason why this should be restricted to cancer", says Peach.
Fabrice André, who runs a similar cancer diagnostic programme that has so far been offered to about 100 patients at the Gustave Roussy Institute in Villejuif, France, says the NHS programme could point the way to implementing personalized medicine across an entire population. "It can really change the landscape of how molecular testing is being done for cancer," he says. "If they succeed, then it's going to be a major step forward."
The Office of Personnel Management (OPM) is planning a new database that will store health care claims information from three federal programs - the Federal Employees Health Benefits Program, the National Pre-Existing Condition Insurance Program and the forthcoming Multi-State Option Plan.
OPM (in a 10/5 Federal Register notice, says the database will allow OPM to "actively manage all three programs to ensure the best value for the enrollees and taxpayers." The database will be effective Nov. 15 "unless comments are received that would result in a contrary determination."
OPM? Healthcare “value?” This raises two issues: (1) expertise and (2) mission creep. Are people being hired to do this? Who are they? Who’s choosing them? Transparency is required – if not a Congressional hearing.
Information collected will include personal identifying information, address, dependent information, employment information, health care provider details including debarred provider information, health care coverage information, health care diagnosis information and provider changes and reimbursement on the aforementioned coverage, procedures and diagnosis.
OPM? Really? Sounds like a CMS program – or even AHRQ. Also, will this data be shared with other federal agencies? And if so, to what end.
Per the FR notice, "the data will be de-identified for specific analysis that provide flexible queries of the data set for general demographic queries, risk-adjusted profiles, and comparison of chronically ill patients and other useful analytics; and engage in econometric modeling of, among other things, health trends, risk adjustment methodologies, pharmacy pricing and negotiation."
Hm – “econometric modeling?” That sounds menacing.
Yes – definitely Congressional hearing material.
So if you want to claim just the opposite: that spending on health care is not related to life expectancy or want to show that the US healthcare system needs to control costs by allowing the government to "coordinate care?"
You do what Sherry Glied and Peter Muening did in their widely publicized article in Health Affairs, "What Changes In Survival Rates Tell Us About US Health Care". You come up with a 15 year survival rate which understates the effect of treatment since most studies track 5 year survival rates after diagnosis and treatment and overstate behavioral issues. Then you simply assert -- because there is more spending in the US than Europe -- that the difference is the result of our lousy, inefficient system of care.
content.healthaffairs.org/cgi/content/full/hlthaff.2010.0073v1#SEC1
How do the authors get away from making the exact obvious conclusion that the rest of the epidemiological literature has shown? One reason is that Health Affairs is carrying a torch for Obamacare and whatever peer review took place was passive or unconcsious. Another reason: such statistical skewing is easy to get away with because most in the media simply report the conclusion without looking at the methods.
The authors state: "We measured fifteen-year survival rather than life expectancy because the latter can be biased by the survival experiences of a small number of elderly people, among whom coding errors are common. Focusing on survival also allowed us to distinguish between the experiences of specific cohorts. We explored fifteen-year survival for men and women separately because risk-factor profiles differ greatly by sex and country.
By looking at overall survival after 15 years the authors can go back to a time when medical innovations essential to survival by disease were non-existent but detection in the US was more prevalent The effect of higher levels of detection -- in the absence of innovations -- are what appears to be lower rates of survival. At the same time they ignore mortality rates because the US had a faster decline in mortality from major diseases (many of which Glied and Muenning ignore) than in other countries.
This hatchet job has yet to be questioned by anyone in the media. If anyone is interested they can compare the Health Affairs j'accuse with other studies that are less biased. In particular, look at Low Life Expectancy in the United States: Is the Health Care System at
Fault? by Samuel H. Preston and Jessica Y. Hoy of UPenn.
repository.upenn.edu/cgi/viewcontent.cgi
Been warning about this from the very beginning, and now it’s starting in earnest: Cost-centric strategies leaving patient-focused medicine in the dust.
Last week the Medicare Payment Advisory Commission met to consider recommendations that empower Medicare to reinstate the option to base Part B drug reimbursement on the least costly alternative (LCA) among products.
Note please, that “least costly” in no way means “best for the patient.”
On October 7th, MedPAC heard two proposals outlined by commission staffer Nancy Ray.
The first was that Congress should give CMS authority to apply least costly alternative policies in setting payments for items and services covered under Medicare Parts A and B, and CMS should periodically assess the clinical similarity of Medicare-covered services and apply LCA policies for those services deemed clinically similar.
The second was that Congress should direct CMS to set the payment rate for a newly covered service that lacks evidence demonstrating better outcomes than existing treatment options at a level that is no higher than the LCA.
The policy could end up relying heavily on data from comparative effectiveness research conducted under the auspices of the Patient-Centered Outcomes Research Institute, which is being created under the Affordable Care Act.
A provision of the ACA states that HHS cannot deny coverage of items based solely on the results of comparative clinical effectiveness research, presenting a possible gray area should LCA determinations favor one product over another.
Ray predicted that both of the LCA recommendations would decrease spending relative to current law and also would lower beneficiary cost sharing in the short term and Medicare premiums in the long term.
While the commission did not vote on either recommendation, comments from members were generally supportive. Surprised?
If you needed another reason to understand why the upcoming elections are so crucial to 21st century patient care – you’re welcome.
You can see the first evidence of Obamacare repeal by Obamacare here:
www.hhs.gov/ociio/regulations/patient/appapps.html
Who "lost" medical technology?
At an FDA-sponsored public meeting in Irvine, California, CDRH Director Jeff Shuren shared that new regulations for approving medical devices may include a new category for medium-risk products. Products in this category would require more studies on safety to win agency clearance. According to Shuren, changes will be phased in and that changing the rules may not result in more studies being required of device makers initially. “We are still addressing that on a case-by-case basis where it’s appropriate and necessary, he said.
Is more data always better? Sometimes. But how does this weigh against the growing “device gap” between the US and the EU? It seems that what’s old is new again – and not necessarily for the better.
When the agency determines it needs more data for an approval or that a device application should fall under a new designation, consideration must also be given to what’s going on “over there.” We cannot and must not create a regulatory environment for medical technology (or, for that matter, pharmaceuticals) that exists in isolation from the rest of the world. Harmonization mustn’t mean “do it our way,’ but rather, “let’s talk.”
It even discusses the need for collaboration:
Support for mission-critical applied research, both at FDA and collaboratively
Support within the FDA is critical to expanding the field of regulatory science. An active research program, directly connected to the FDA review process, will not only bring needed advances in regulatory science straight to FDA review, product development, and evaluation but will also add value to guidance and policy development.
In addition, the discipline of regulatory science must be developed though support from both partnerships and external research and collaboration. There are substantial opportunities to enhance and expand current FDA programs and to develop new ones that support effective, more robust, external and collaborative efforts to advance regulatory science. Some projects are already under way:
• A Joint Leadership Council recently created by FDA and NIH to promote the expansion of regulatory science through enhanced scientific collaboration and jointly supported and administered extramural research grants in regulatory science.
• Creation and support of academic Centers of Excellence in Regulatory Science to carry out applied regulatory science research both independently and in collaboration with the FDA and as a locus for scientific exchange and training opportunities for both FDA and academic scientists
• Enhanced strategic collaboration and coordination with other governmental agencies to develop new programs to advance regulatory science and innovation
• Enhanced support and focus for the Critical Path Initiative to catalyze and enable partnerships and consortia that advance regulatory science and public health through innovation and modernization of the medical product development and evaluation process
• Partnership with the Reagan-Udall Foundation on projects in support of regulatory science
Good words. But two important words are missing ... "with industry."
Obviously, the one-line reference to the Reagan/Udall Foundation means "with industry" and it's possible that these words which must not be spoken also reside within "partnerships and consortia." But why the strange silence about such a valuable partner? Advancing regulatory science without regular and robust collaboration from the regulated? Not likely.
Fortunately, Peggy Hamburg often speaks about working with industry on Critical Path issues. Enquiring minds want to know if any references to "industry" were blue-penciled out prior to publication -- and if so, by whom.
The white paper intelligently comments that:
There is no single discovery — no magic bullet — to address our unique set of modern scientific regulatory challenges. But one thing is clear: if we are to solve the most pressing public health problems we face today, we need new approaches, new collaborations and new ways to take advantage
of 21st century technologies. And we need them now.
All said, this paper is a step in the right direction and people who work in (expletive deleted) should read it.
There’s no crying in healthcare reform.
In today’s New York Times, David Leonhardt (discussing the Presidential rhetoric used to assuage citizen uncertainty) writes, “Mr. Obama went so far as to suggest there would be no disruptions, saying that people could keep their current plan if they liked it. But that’s not quite right. It is not possible to change a system as huge, and as hugely flawed, as ours without some disruptions.”
Now that the legislation has passed, it seems as though the new talking point is, “It’s gonna get worse before it gets better.” And whether or not it’s going to get better is still very much theory (at best) and hoping (at worst).
Remember all that money that healthcare reform was going to save us? Well, since a swiftly deprogrammed Peter Orszag put that shibboleth to bed (again, in the pages of the New York Times), the debate seems to be about how insurance is going to be made more affordable. One way we’re supposed to achieve this, according to Leonhardt, is that “people will be required to buy insurance, to spread costs among the sick and the healthy.”
Sure – except that this mandate (if it doesn’t turn out to be unconstitutional) (1) doesn’t even kick in until 2014 (the same time the theoretical state exchanges come in to play) and (2) will likely penalize offenders less than even a low-cost health insurance premium. Oops.
Specific to the state exchanges that are the foundation of the theory, Leonhardt writes, “the new markets for health insurance, known as exchanges, won’t be up and running until 2014. This timetable has its problems, and the Obama administration will probably need to grant some more temporary exemptions.”
Call me when the revolution starts.
Not to say that it’s at all about politics! No penalties until 2014? If we’re in such dire straits – why not, say, right now. “It’s coming and it’s going to be great,” as a rallying cry is, plainly speaking, tanking since the reform plan is already leaking like a sieve. And it’s not going to cut it before the November elections. The truth is that it’s likely to get worse. Whether or not it gets “better” (whatever that means) depends largely on “fixes” that will necessarily be made in the next Congress.
Just what we need, healthcare sausage.
All this to say that we’d better all stop crying and start trying to think about what needs to be done beyond the talking points. After all, if it was easy, anyone could do it.
Center for Medicine in the Public Interest is a nonprofit, non-partisan organization promoting innovative solutions that advance medical progress, reduce health disparities, extend life and make health care more affordable, preventive and patient-centered. CMPI also provides the public, policymakers and the media a reliable source of independent scientific analysis on issues ranging from personalized medicine, food and drug safety, health care reform and comparative effectiveness.
