Home and abroad, the US Pharmacopeia is stepping up to maintain quality control. But it's not so easy. USP's Roger Williams discusses Medicare formularies, drug safety, international drug production, and the organization's changing role.
Founded 1820, the United States Pharmacopeia (USP) is the standard-setting body for many drugs throughout the world. But new pressures on cost, an increasingly globalized industry tainted with substandard drugs, and the prospects of returning to an age when doctors—or in this case, manufacturers—compound drugs depending on the patient, means USP needs new ways to ensure quality. Here, Roger Williams, USP's CEO, sits down with Pharm Exec and offers some straight talk about how the organization is changing its role from a facilitator of quality to the engine driving it. What follows is an edited transcript of that discussion.
PHARM EXEC: USP was in the spotlight because of its role in developing formulary guidelines, which outline the categories of medications covered under Medicare Part D. What was it like to be on the inside of the organization?
WILLIAMS: The first year was incredibly intense. We pulled together a committee out of the existing members of our council of experts—our standards-setting body—and I think they did a terrific job. There were many philosophical and policy and political and economic and scientific issues they had to struggle with. But at the end of the day, they delivered, on time and unanimously, the model guidelines for the first round of drug plan formularies.
It seems that USP had to make a lot of tough decisions about which drugs are truly necessary.
That gets at the core issue—access. Interestingly, a healthy tension built up in the first committee, and it continued into the second committee that just completed its deliberations [for the revised model guidelines]. One side of that tension says, "If you want to assure beneficiary access, have a very granular structure to the formulary because then you make everything available." The other says, "If you want to assure access, start watching cost." And that leads to a less granular structure with fewer categories and classes of drugs.
In the first round, the committee solved that issue by creating two approaches. One was the model guidelines, and the other was what we call the formulary key-drug types. The law said you had to have two drugs in every category class of the model guidelines, but you only have to have one in each formulary key-drug type. The difference between one and two has a lot of impact on the competitive bidding process.
Model Guidelines: Version 2.0
How do formulary key-drug types work in practice?
Formulary key-drug types create unique categories, which allow new molecules and medicines to enter into the thinking of the committee. We then say to CMS, "You should have at least that one drug in the category." But over time, as we gain experience with that molecule and start getting a sense of comparisons with other similar agents, it may move back into the model guidelines and disappear as a discrete category.
I'll give you an example—SSRIs and SNRIs. When the first SNRI was coming out, the committee would have wanted to create that category in formulary key-drug types and put that one drug in it. Pretty soon, you'd have a group of drugs there, and people would start asking, "What are the difference between the SNRIs and the SSRIs?" Over time if there is no difference, they'll be in the model guidelines just as antidepressants.
How did the drug plans react to the idea of these new types of categories?
You could almost predict how they'd respond. The plans didn't like the formulary key-drug types because they added granularity and a requirement to have one drug in each group. But the plans did like the model guidelines.
Despite that, in the first round, the committee had a lot of success. About 75 plans use the model guidelines. It'll be interesting to see how many use them the second go around, which will start this spring when CMS begins to evaluate the formularies for January '07.
What was different about developing the second model guidelines?
That tension I spoke about seemed to resolve in an interesting way. For new drugs with a defined new pharmacologic effect, the committee in this cycle wanted to create the category in the formulary key-drug types. Yet they were also willing to consolidate the model guidelines.
Another point of note is that the committees themselves decided they wanted to be more evidence-based. With the development of the first model guidelines and the formulary key-drug types, we relied on practitioner experience—doctors that say, "I really need that drug." Going forward, we would like to have it based more on evidence, and more on the literature.
I'm pleased about that, but also a little concerned. This is one of those initiatives that Congress authorized, but didn't appropriate. Every year, we have to argue with the CMS bureaucracy for support. And let's face it, they're not willing to give us much. I don't know why that is. You would think they'd give us some resources to do what could be a very powerful job on their behalf. We're a very small part of this gigantic enterprise they're engaged in, but I think not an unimportant part.
It does seem that USP flies under the radar of other stakeholders as well—most consumers have never even heard of the group.
USP is very unusual in the pharmacopeias of the world in that it's a practitioner-based organization. And yet, as part of their day-to-day lives, practitioners don't care much about the standards. But they do care about some of the healthcare information and safety programs (see "No Room for Error").
No Room for Error
Why don't they care? Do they just take high-quality medications for granted?
They do. They assume that FDA's looked at it, USP's on the beat, and that manufacturers are trying to do a good job—which for the most part is quite true. When they use their ibuprofen tablets, they don't usually question, "Do they have ibuprofen in them? Is it the right amount of ibuprofen? What about the release?" They assume that the fact that it is on the market is a manifestation of its adherence to good quality standards.
Isn't that at least kind of true?
FDA and industry and USP have done a terrific job of relieving the practitioner of that burden. But in some areas, it's not quite true. In the area of dietary supplements, FDA oversight was diminished substantially, and even now, we don't have GMPs [for manufacturers of dietary supplements]. For them, adherence to a USP monograph is optional. So all those safeguards that give practitioners assurance about the quality of a dietary supplement are diminished.
USP stepped up to the plate there, and tried to help assure the quality, given the fact FDA hadn't finalized their GMPs yet. We call that our Dietary Supplement Verification Program (see "Some Bark, Little Bite").
Some Bark, Little Bite
Where else is USP asserting its influence?
USP had its origins in compounding. And I've even heard people say that manufacturing may change so it becomes more like compounding, so that people can get the right dose at the right dose frequency. Or complex new biologics are going to be assembled at the bedside before delivery.
We want to reach back to compounding practitioners and the community pharmacists through our Pharmacists' Pharmacopeia, which was a line extension, as we call it, of USP-NF. [An official publication, issued by USP, that gives the composition, description, method of preparation, and dosage for drugs. The book contains two separate official compendia—the USP and the National Formulary.] We have to be very careful in those line extensions that we preserve our status as having two of the three official compendia of the United States: The only one we don't have is the American Homeopathic Pharmacopeia. The other two are National Formulary and United States Pharmacopeia.
FDA is obviously going through some problems. What is USP's role in resolving some of the issues that have been raised about drug safety and oversight.
I go back to that point of independent testing. Let's say Company X has a problem with their potency or their strength. Well, you don't want FDA to have to knock on their door and say, "May we use your reference standards to test the quality of your products?" I'm a strong advocate for FDA to be able to test independently.
Where do you get your reference standards?
We get what we call candidate bulk materials from all over the world—pioneers, generics, chemical supply houses. At the end of the day, however, we think of them as unknown things that we then move through a very careful process to establish a public reference standard.
That process involves us checking chemically to make sure we know exactly what we've got, sending it out for collaborative testing so that everybody can do an additional check against us, and then packaging and making it available for sale one lot at a time so that all the manufacturers throughout the world can test to a single standard.
That's the value of the "one-lot" concept. When that one lot goes down, we replace it with another lot. If you think about it, it's not the way pharmaceutical manufacturing works. There can be a multiple of lots, even from one manufacturer. And when you get to generic substitution, there could be scores of lots in the marketplace at any one time. But they all should be testing to this single standard.
It's a very nifty system if you think about it. And as much as I like PAT [process analytical technology]—I'm very supportive of the science in back of PAT—it makes me nervous when they don't talk about end-product testing. Because if you don't do end-product testing to a single reference point, pretty soon you start losing that link between all lots and all manufacturers.
But who would argue with better quality by design? In my mind, those are things people should be doing anyway without the sort of cheerleading aspect of it.
Some people would say FDA doesn't need to cheerlead, it needs to get out of the way.
I would argue that FDA needs to stand tall for pharmaceutical equivalences and bioequivalences. In other words, every manufacturer needs to maintain continuing equivalence, relative to their clinical-trial material. And if they want to do that via PAT, fine. If they want to do that by PAT and then product testing, fine. But it involves very careful change control. To me, the agency is the watchdog to say that you're pharmaceutically equivalent and bioequivalent over time, barring some intentional change.
Times change, but I still say there's a strong regulatory imperative for doing that. It also gets to this issue of generic substitution. For example, if the pioneer is wavering in terms of pharmaceutical equivalence and bioequivalence in your reference product, what does that do to your generic? And you may have five or 10 generics relative to that pioneer. What if all those pioneers start wavering? Pretty soon you don't have an interchangeable system anymore, and to me that's the bedrock of what we're doing.
I wanted to talk about some of the international initiatives USP has been doing. The published figures from emerging nations on mislabeled drugs or drugs with no active ingredient seems incredibly discouraging.
You're moving into the world of counterfeit and substandard drugs. And there have been a lot of initiatives over the last year or two that I think are going to lead to some very exciting activities.
There was a meeting in Madrid in '04 and a more recent meeting in Rome this year in March that led to the concept of a counterfeit task force with working groups. That group needs time. It needs resources. It needs energy. But I see that being marshaled now, and USP is going to try to help, if we can, either by sending knowledgeable staff or other types of resources.
What about issues of quality in the United States?
I think it's a big problem. I'm not sure we have as strong a control of the US marketplace as people might like to think.
I mean, how do you know you've got a problem? If a drug doesn't work, what do practitioners say? They say, "Gee, they were in that population group where the drug just didn't work," or, "Gee, they didn't take their drug," or, "Gee, it was the course of the disease." It goes back to that whole trust that practitioners have in the quality of their medicines.
If there were a national disaster—if 200 people died because of an excipient being incorrect, which happened in Haiti [in 1995, when glycerin contaminated with diethylene glycol was used in manufacturing acetaminophen syrup] and happened here [in 1938, when diethylene glycol was used to manufacture oral sulfa solution]— there wouldn't be enough oversight.
In addition, there are a lot of drugs in the world that I call substandard because they go back to not being pharmaceutically equivalent and bioequivalent. There are only a few countries, like the United States, that have a truly interchangeable system of pioneers, reference-listed drugs, and generics. In another country, you might have 50 copycat versions of a pioneer, none of which have ever had to document either pharmaceutical equivalence or bioequivalence. I would call those substandard, even though they're legitimately manufactured. You don't know that you're going to get the same safety/efficacy profile in the person you're about to give it to as you did in the population that was studied for the pioneer.
There are ways to solve that problem, but it's huge. Most countries don't have an interchangeable generic system. Europe is moving to it in very interesting ways. Brazil is going there. Mexico is going there. And the United States, Canada, and Northern Europe have gotten there. Japan is there. So who knows? In the next 50 years, we may see a truly interchangeable system of pioneers and generics spreading throughout the world.
At the moment, what's USP up to in this area?
First of all, USP has seen an explosion in terms of its international reference-standard sales. If it's not already, most of our reference-standard sales will come from outside this country.
Now, you might say to yourself, why is that? Well, first of all, I think both our documentary standards and our reference standards are good. Second of all, the ICH [International Conference on Harmonization] gave tremendous impetus to specifications, testing to specifications, and impurity testing. The ICH Q3C procedures focused a lot on impurities and the drug substance, then Q3A focused on the drug substance, and Q3B focused on the drug product in terms of impurities, and that leads to additional testing and additional manufacturing controls. So all of this has enhanced our products and services, not only nationally, but internationally.
Now, of course, there's another reason, which is that people want to get into the very lucrative US market. And while USP is not a gatekeeper, we're a facilitator in certain areas—namely the quality areas. It's a very strong role for us. I'd like it to expand, and certainly, we want to make sure we work appropriately with FDA as a facilitator.
What are the issues in harmonization?
I can't say that we're seeing a lot of ground swell of enthusiasm for improvement. People are happy with the way it is. We can propose improvements, but people have to be willing to accept them.
Why are countries unwilling to harmonize?
Sometimes, I'm just mystified. There's robust support for excipient harmonization, and that makes a lot of sense to me. But when you get into the nonexcipient general chapters and the drug substance, we were told quite clearly that the harmonization and the compendia connected with ICH should not focus on the drug substance. I don't know why. I would say, right now, harmonization will proceed at a very slow pace, unless something is done.
Do we have de facto harmonization with, say, Northern Europe or Canada?
Let's just say you're an excipient manufacturer. You may have to satisfy 40 compendial standards. The classic example is sodium chloride, where there are 50 tests with different procedures and different acceptance criteria. That's nightmarish. What a sodium chloride manufacturer could do is just take all the tests and create a private monograph that meets the most restrictive acceptance criteria. But that's very burdensome.
We want to work throughout the world partnering wherever we can to advance the public health mission connected with those documentary and physical standards, which, at the end of the day, relate to the quality of medicines and their safety, efficacy, and use.
We've been in a lot of places in the world through what we call our stakeholder forums. But the next step after a stakeholder forum is actually to set up a site. Our first site was in Hyderabad, India. That was very carefully considered. India is making a lot of drugs and dosage forms, and they're also coming into the United States in a big way. Hyderabad was chosen because it's a locus for a lot of bulk-drug manufacturing, but also dosage-form manufacturing. It's also in the heart of India's "Genome Valley."
So that office interacts with manufacturers and regulatory officials in India?
Right. We try to have very good relationships with the government ministries, the government regulatory agency, and the Indian pharmacopeia. I'll probably go back there over the summer to see how I can build some of those linkages. Our second site will be in Shanghai for much the same reason as India.
People are predicting that in the next five years up to 80 percent of our bulks could be coming from either India or China. Now why wouldn't we be there? I could have set some of these sites up in Howard County [in Maryland]. But we're not going to get 80 percent of our bulks from Howard County.
Can we also be a force for good in terms of what they're producing locally and for export elsewhere? So far, our receptions in both countries have been quite positive at the governmental and compendial levels. And, I like the thought of working with the Indian pharmacopeia and the Chinese pharmacopeia. They are very sophisticated. We have a lot to learn from them and hopefully they have a lot to learn from us.