Tuesday, December 21, 2004

The hits just keep on coming: All eyes on naproxen

Hot on the heels of the announcement that Celebrex can increase risk of cardiovascular adverse events discovered during a clinical trial to see if Celebrex might assist in preventing colorectal cancer, NIH has announced that naproxen sodium (trade name Aleve) has been shown in a three-year trial to increase the chance of CV events over placebo. (The latter clinical trial was the ADAPT, or Alzheimer's Disease Anti-Inflammatory Prevention Trial.)

CNN is reporting that those who took naproxen were 50 percent more likely to experience CV events than those who took a placebo, but both NIH and FDA are relatively quiet on percents thus far.

Monday, December 20, 2004

I've been waiting to see this one for a while...

I present documentation of the reason for the flu vaccine shortage and subsequent hysteria: the FDA warning letter to Chiron regarding the cGMP violations in manufacture of Fluvirin.

Pfizer to stop DTC ads for Celebrex

I find it interesting (not to mention appropriate, in this case) that Pfizer is stopping direct-to-consumer (DTC) ads for Celebrex. DTC advertising has indirectly cost the pharmaceutical industry dearly in terms of PR this year. As I mentioned in my last post, DTC ads for COX-2 inhibitors -- drugs which were intended for use in specific situations where patients showed risk for GI bleeding -- were one of the root causes of any overprescription of these drugs; and overprescription is one of the root causes for so many people feeling the side effects of the coxibs.

DTC advertising is a double-edged sword: on one hand, it obviously gets the name of the new whiz-bang drugs into people's heads (and it often hits their wallets as well, since frequently a 1-week free trial is offered) which makes great financial sense. On the other, though, it can lead people down the wrong path. Vioxx was not a blockbuster because there were that many people out there with arthritis and risk of GI bleeding; it was a blockbuster because people heard of a new and spiffy drug for arthritis, and since it's new, it must be better... and history is made.

As an interesting aside, the vast majority of FDA warning letters to pharmaceutical companies for the second half of 2004 have applied to promotional material. Not only do people have the sense that new is better, but pharma companies have been becoming bolder about making unsupported claims to boot. These are big-name companies, too, with a lot to lose; they know the gamble is worthwhile.

Obviously, eliminating DTC advertising is not the whole solution; there is still provider advertising, and there is still the human feeling (on the parts of both some doctors as well as patients!) that newer is better to contend with. The fact of the matter is that some advertising needs to exist, if not to get the word out about new therapies, then to make sure that pharma companies keep making money (for if there is no money, there is no pipeline.) It's just one piece of the puzzle.

Friday, December 17, 2004

Someone has to take the blame... don't they?

On a forum I frequent, a post was put up today regarding this study, in which a link was found to exist between use of Celebrex and major heart problems. And, of course, the blamethrowing began. Is it FDA's fault? Is it the fault of the pharmas? Is it doctors' fault for overprescribing? Whose fault is it? Everyone knows it has to come back to one entity... doesn't it?

Sorry, folks, but I don't think it does. A lot of it is left to chance. FDA approves drugs based on the data that's presented, and the amount of data presented has to fulfill certain criteria that are clearly delineated in regulations and guidance regarding those regulations. At some point, though, you need to stop evaluating and place the drug up for approval, or you're (a) never going to have a chance to help people, and, frankly, (b) never going to make money. So clearly that's one point of risk: you can never have full disclosure because you can't evaluate a drug forever before releasing it on the market.

Another point of risk is pharmaceutical direct-to-consumer advertising. Celebrex and the other coxibs didn't actually show any greater effect against arthritis pain than traditional NSAIDs such as ibuprofen and naproxen. COX-2 inhibitors were developed for a specific reason: to allow people with risk of GI bleeding to take anti-inflammatory medications for arthritis. (NSAIDs inhibit both COX-1 and COX-2, and inhibition of COX-1 is linked with greater risk of GI bleeding.) However, none of that was anywhere near the DTC advertising. All consumers saw was that there was a new, whiz-bang drug for arthritis; as we all know, new is always better, so people would trot off to their doctors asking for prescriptions. Doctors had no evidence of severe risk in prescribing the coxibs, so prescribed they were, in vast amounts. On one hand, that's helpful, as it helps us learn about the drug; on the other hand, the risks of the drug are a hard lesson for 27,000+ people to learn.

A third point of risk is insurance; a lot of people have prescription drug coverage, and that leads to a lot of people wanting prescription drugs rather than OTC drugs, which they'd have to pay for themselves. If Joe Smith has a plan that allows him to get three months of a formulary drug (or even a non-formulary drug) for one co-pay of $20 or $30, depending on how many doses he needs versus how many the doctor wrote the scrip for, Joe could be in much better financial shape with that than trying to buy three months or more of ibuprofen or naproxen, provided he goes with a brand name. On top of that, he gets the new, whiz-bang drug for less than the old, boring one; quite appealing, wouldn't you say?

Something that many people have trouble with is that there is no safe drug. Derek Lowe covered this topic quite well in one of his In The Pipeline articles, so I won't rehash the topic here, but I do suggest everyone go and read; it's worthwhile. It's a topic that causes a lot of people discomfort, and when they see problems like the coxibs are having crop up, they're desperate to blame someone because they feel the system should be foolproof. It's not. It is, I believe, a fundamentally good system, largely staffed by people who are interested in and focused on doing well and improving the human condition, but foolproof and perfect it's not.

Wednesday, December 15, 2004

"Can I ask you a question? Define validation."

...Nothing like that being the first thing out of a client's mouth when I walk into the room for what was supposed to be a meet-and-greet. What I said at the time was "making sure that a system fulfills the requirements set forth by FDA." The client, being who he was, of course responded "Do you really believe that?" to which I said that if I didn't, I was in the wrong job.

I do still believe it, but I wish I'd thought to give a more complete answer. Fundamentally, my job is to make sure that the computerized systems we use comply with FDA regulations. That having been said, in terms of computerized systems, pretty much all you have to deal with as far as FDA regulations go is 21 CFR Part 11 and predicate rules. (HIPAA needs to be considered as well, but that doesn't so much figure into our clinical trials because we don't store PHI.) Predicate rules for computerized systems can be summed up by two points:

(1) If you didn't document it, it didn't happen.
(2) Do what you say you're going to do.

That, plus the requirements set forth by 21 CFR Part 11, form the basis of what I do. It's deceptively simple; you wouldn't think it would be as difficult to document policies and procedures and a plan that complies with those policies and procedures, execute to that plan, and document the execution as it seems to be.

Tuesday, December 14, 2004

A molehill out of a mountain? 21 CFR Part 11

21 CFR Part 11 is a set of regulations put forth by FDA to restrict and guide use of computerized systems in the food, drug, and device industries. It governs electronic records, electronic signatures, access control, how to provide assurances of data accuracy, and records retention, among other things, the goal of which is to enable companies to use computerized systems for their daily work while also preserving the high quality of food, drugs, and devices developed by these companies. For someone such as myself who has always worked on newly built and/or newly implemented computerized systems, this has seemed to be somewhat of a non-event and really quite common-sensical; after all, it's not particularly intelligent to build a system that handles critical data but doesn't limit who can touch that data, for instance. On top of that, GxP, generally referred to as the predicate rules for 21 CFR Part 11, covers a great deal of ground as far as clinical, laboratory, and manufacturing controls go, so it's not as if the arena was completely uncontrolled prior to the advent of 21 CFR Part 11.

However, for companies using legacy systems or computerized manufacturing equipment, 21 CFR Part 11 was more of a big deal. There simply is no way to perform an electronic signature or have the computer automatically generate an audit trail with reliable timestamps when dealing with some manufacturing equipment; likewise with legacy computer systems that were built years before such a thing had been considered. 21 CFR Part 11 did not go into effect until 1997, well into the Information Age. A lot of companies had a lot to fix and not a lot of time to do it.

... Until FDA's guidance came out last year, at which point those companies could breathe a sigh of relief. FDA made public that they were going to exercise "enforcement discretion" with respect to legacy systems so long as they fulfilled predicate rule requirements, and, in fact, the guidance stated that it should not be read to impose any additional validation requirements beyond those needed to demonstrate compliance with predicate rules. Further, FDA stated that a risk-based approach should be taken with 21 CFR Part 11 compliance validation; meaning that the impact on patient health and product safety of a given system should be assessed, and that validation should be conducted per the results of the assessment.

This is huge news to the pharmaceutical industry, evidenced by the fact that 21 CFR Part 11 is still a key topic in journals and at conferences over a year after the final guidance narrowing its scope dramatically was issued.

The regulations found their way into the news again recently, when the scope was actually extended to include requirements set forth by the Public Health Security Act and Bioterrorism Preparedness and Response Act of 2002. I'm not sure of what the impact on the food industry has been, as I'm not part of that industry, but I think it does show that the August 2003 guidance isn't the last we've heard of 21 CFR Part 11.

Introductions are in order

My name is M (well, it begins with M,) and I manage the regulatory compliance validation department of a small contract research organization within the pharmaceutical industry. I handle mostly software validation, since electronic records and electronic submissions are the bulk of what we do, but on any given day I can be asked to advise clients, write proposals, manage my team of 3, or any other number of tasks.

Throughout my life, I've felt a connection to the healthcare and pharmaceutical industries. I had always intended to go to medical school; I spent many years in high school volunteering in the ERs of local hospitals (since no other volunteers wanted to get close that that level of grittiness and I thrived in it); and even when the path of my life took me out of the pre-medical track, I ended out working in the pharma industry in another capacity, assessing computer systems for compliance with FDA regulations.

I have every intention of going back to school eventually to change the path of my life again; but for now, I have a family to support. So I read medical and pharma blogs and dream, and it occurred to me that I may even have something to contribute. Whether or not that is ill-advised remains to be seen.