Re: misconceptions about FDA requirements for drug approvals

[Guest guest]

Hi , the requirement would be the same for personalized medicine - that

there is clear evidence of meaningful clinical benefit. The difference would be

for the indication - which would be narrower - such for CLL with xyz biomarker.

The studies would select for these patients in clinical trials. If successfully

the marketing Label (assuming it proves effective) would be for that subset of

CLL. Which means less profitability, but most likely a easier path to approval -

assuming the rate of benefit will be higher for such therapies.

Actually this has already happenwd with the approval of revlimid for a type of

MDS with a specific gene deletion - based on phase II data.

Best,

Karl

>

> I don't know if anyone has read the guidance documents from FDA's website on

> " Clinical Trial Endpoints for Approval of Cancer Drugs " or " Providing Clinical

> Evidence of Effectiveness for Human Drugs and Biological Products " .  There

are

> others as well, but these will give you the perspective of what drug companies

> must do to get approval.  It is also what the advisory panel members look at

to

> weigh in if a drug's benefit is greater than its risk.  I do not think there

> are very many if any guidance documents from FDA on personalized medicine

that

> drug companies can follow in order to get approval at this point.  I have

been

> in FDA meetings with reviewers at FDA and they clearly do not want to set a

> precidence of letting a drug get approvel with little efficacy and risks.   

> Just read the latest on the two obesity drugs that the advisory panel shot

> down.  There is no such thing as a risk free drug.

> Until the FDA puts out more guidance documents on approval of personalized

> medicine, I don't think you will see a lot of drugs approved in this

> area.  Provenge is the only one that I am aware of.  I would assume FDA is

> working on it with other guidance documents as well.  Karl, you may have more

> insite than I on that.

>

>

>

>

>

> ________________________________

> From: " karls@lymphomation " <KarlS@...>

> - CNS-PAL <cns-pal >; - PAL-parent-support

> <PAL-parent-support >; ;

> mantlecell ; nhl-dlc ;

nhl-follic ;

> nhl-info ; nhl-malt ; pal-bod ;

> pal-datafork

> Sent: Wed, September 22, 2010 10:13:17 AM

> Subject: RE: misconceptions about FDA requirements for drug

> approvals

>

>  

> re: " if the novel treatment has shown hints of effectiveness and absolute

> lack of any side effects, approval requirements should be lowered. All the

> more reason when dealing with truly personalized medicine. "

>

> Thanks for commenting. Interesting perspective for sure. If you mean lower

> but still providing clear evidence that it improves outcomes in some? - I

> think it would depend if that benefit could not be explained by chance or

> biased study methods.

>

> Approving non-toxic drugs (the exceptions) that don't actually work would

> then lead to routine delay of effective therapy. Timing of therapy

> sometimes counts - sometimes it's critical. For indolent lymphomas we

> could argue that the harm is less - but only if the unproven therapy is

> given when treatment is not needed -- (a good time, btw, to consider

> participation in studies of low toxic investigational agents.)

>

> As you know, for cancers most active agents will also have toxicities. So,

> more than for other indications we need biomarkers to predict individual

> response to therapy - such as unproductive toxicity.

>

> Standards for efficacy provides also a foundation for secondary studies. So

> which of the many hundreds of studies *suggesting* benefit do you follow up

> on? Lowering standards would have many side effects, not touched on by the

> Wall Street Journal.

>

> Meanwhile, many corporations remain focused on one-size-fits-all testing ...

> because that is how they want to also market it. They are businesses, after

> all.

>

> That might have been okay when nothing was effective, but the shift has to

> be towards personalized medicine (the right drug for the right patient)

> particularly in a landscape like lymphoma/CLL where we have dozens of active

> drugs. Do we find out about response to Rituxan by giving it? The sponsor

> doesn't have a problem with that model, but patients should. And when you

> select patients for studies based on the right biomarker, the studies can be

> smaller because the rate of benefit will rise - and such studies may not

> even require randomized design.

>

> So where is the comparative testing? It seems we get a lot of strategic

> trials aimed at expanding the marketing label, sometimes by avoiding

> comparisons with the strongest protocols. And here we have to be

> particularly alert to unethical design. The test will provide reliable

> answers, but is the comparator sub-standard - less likely than something

> else to meet the clinical needs of the participants?

>

> It's serious stuff, and we don't need untrained journalists misrepresenting

> the problem - spreading simplistic, toxic ideas about FDA motives, which has

> no other intended purpose but to get the decision right.

>

> For example, when you have a 30% response rate for a drug that causes side

> effects, 70% get ONLY the side effects, which, importantly, can limit the

> range of next choices. But the sponsor and sometimes the media will hum on

> about how patients are being shortchanged by FDA standards. Well those 70%

> of non-responders getting only the toxicity, they are patients too.

>

> The commercial model for development of drugs is important and necessary,

> but it also has significant flaws, not the least of which is a biased intent

> to find the fastest and surest way to market. I guess you take the good

> with the bad, especially when the choice would be no drug development. So

> I'm not a communist, folks. I get it that the industry is the ONLY entity

> capable of doing these big and expensive projects, which many times fail.

> But I look to the NCI to fill in these gaps and the FDA and patients to

> raise expectations.

>

> Note: Please know that there are many exceptional and caring persons who

> work in the drug industry, who do great science for all the right reasons.

> Truly, and I've met a good many. But I don't agree with the Supreme Court -

> that a company is like an individual. Its mission is one-dimensional: to

> maximize shareholder profits. Great people within it? Yes. But should we

> trust (without oversight and regulation) these instruments of commercial

> activity? No. And should we trust the perspective of journalists at Wall

> Street Journal in these matters? I think not.

>

> Excuse my ranting on about this. I'm done, I promise.

>

> Karl

>

> ==

> Re: reactions to " New Drugs Stir Debate on Basic Rules of Clinical Trials "

> http://nyti.ms/amxLkS

>

> Greetings,

>

> I've observed that the editors of the Wall Street Journal are prone to

> demonizing the FDA; and advocating for reducing standards for evaluations of

> new drugs for marketing approval. But I think this is a biased industry

> perspective . profit-driven, and not in the best interest of patients.

>

> Responsibility for ethical study design is shared with the investigators

> (disease experts) and drug sponsors who design the study and submit it to

> FDA for approval. If the study described in the NY Times was unethical,

> there are plenty to blame for it.

>

> Regrettably, sometimes the patient perspective is missed and unethical

> studies are opened for enrollment. These are sad events, which are more

> likely to occur in my view when informed patient representatives are not

> included in the concept phase of study design.

>

> What often drives it is the sponsor's marketing objectives and what they

> perceive is needed to achieve that goal, aided and abetted by disagreements

> about what is ethical design and perhaps judgments about the potential risks

> and benefits of the compared protocols.

>

> So to err is human. There is sometimes a tension between doing good science

> and acting in the best interest of the participants. But the patient

> interest must come first. Participation in a study must have the potential

> to benefit us - equivalent or better than any standard option. And there

> must be GENUINE uncertainty regarding which arm of a controlled study is

> superior.

>

> However, if your methods for testing a new drug are not reliable, the

> patients who later use these drugs can suffer or even die. Further without

> sufficient testing we could not know which treatment really works, is

> better, or which is even safe.

>

> So the study method must also be able to demonstrate that a protocol

> provides clinical benefit - else it is also unethical as you've subjected

> participants to an insufficently tested drug without getting answers to

> benefit future patients.

>

> So a study must be good science AND good medicine.

>

> Throughout history tragedies have occurred when drugs have been

> insufficiently tested. Thalidomide being but one example, which led to

> Congressional legislation. So the alternative to evidence-based standards

> is medicine by opinion, theory and sales pitch. Chaos. I nice business

> model? Perhaps, but not good for patients.

>

> See End Points and United States Food and Drug Administration Approval of

> Oncology Drugs http://www.lymphomation.org/endpoints.pdf

>

> It provides an inventory of the numerous agents approved by FDA based on

> non-survival endpoints (surrogate endpoints, such as response rate or PFS,

> which reasonably predict survival benefit). This document shows that the

> sponsor and experts it consulted had other choices for study design.

> However, this does not absolve the FDA from its responsibility to identify

> and reject unethical study design.

>

>

>

>

>

>

>

>