PDA Letter Article

Quality Metrics to Impact Different Pharma Sectors

by Rebecca Stauffer, PDA

The recent 2017 PDA Pharmaceutical Quality Metrics and Quality Culture Conference in Bethesda, Md. offered attendees an opportunity to discuss the current state of quality metrics in light of the recently revised US FDA guidance on metrics. In the second plenary, panelists representing different segments of the industry offered their perspectives on the Agency’s quality metrics initiative. Below is a sampling of the panelists’ responses.

Large Pharma

Barbara Allen
Barbara Allen,
PhD, Senior Director, Global Quality Systems, Eli Lilly

Many companies that I’m involved with on the large pharma side have quite a diverse range of products, and one type of product that’s coming up quite a bit in conversation is how to address drug/device combination products and how to report them. In general, most are interpreting that if the product is registered with CDRH, that it would not be within the scope of reporting, and that it would be the products registered under CDER that are reported. Then for the metrics, trying to distinguish the drug element and the device element is under consideration, where I would say a lot of the emphasis is on the drug part and the device part comes into play at the very end, but the constituent parts of the device are not being considered as intermediates. I think it would be a complicating factor at this point…in addition, similarly for the API reporting, an emerging common interpretation I’m hearing is all steps of the API, from the first registered starting material to intermediate steps are in scope. This typically extends the supply chain significantly, which increases the number of sites and data points...so we need some clarity on that scope, and perhaps...it may be wise to start with the API step instead of all the individual steps. I think some clarity around that would be very helpful.


Deborah AutorDeborah Autor, Head of Strategic Global Quality and Regulatory Policy, Mylan, representing the Association for Accessible Medicines (formerly Generic Pharmaceutical Association)

I do think it’s important to make sure everybody knows that generics are actually 89% of the drugs prescribed in the United States...so, I speak from that perspective, with the understanding it’s that scale, that volume, and that low cost model...that, of course, means you have to figure out ways to operate more effectively and be sensitive to cost....

I think the No. 1 challenge for us is that timeline for implementation...I think the Agency has a lot between the first draft and the second draft of the guidance to clarify. There are still a lot of open questions as to how to implement that...I’ll say opening it up for a month, for us to submit our data, that’s a short window of time, and realigning APR submissions—whatever we do to connect with FDA’s annual reporting, which the Agency wants so it can match the datapoints, is going to be challenging. For a proposed solution, I’d say slow down that phased approach. Start with a small number of metrics, a small number of players, and have verification to go along... especially when you have senior management who say “I want to be on the Reporters List.” I think my suggestion would be to perhaps scale back the Reporters List. I really do worry about companies taking resources from the quality unit because those are the resources that are doing this, and taking them from something they were doing and changing their tasking toward metrics because that may not be the most impactful thing they could do from a quality standpoint...my point is that it has a huge impact when you look across a large complex company, and I think even the smaller companies, it’s going to have a really big impact...do I think this is the right course to pursue? I think from a benefit burden analysis, I’d say at this point, I don’t think we’re there...If I could know three things about my sites, it wouldn’t be lot acceptance rate, invalidated OOS, and product quality complaint rate...I think I’d look at it more like an MHRA approach... things that really get to qualitatively what’s happening at a site.

API Supplier

Guy Villax, CEO, Hovione

To answer the question on where is quality metrics going, I have a dream...I wrote about the Dean’s List, that the FDA should have a Dean’s List, in 2013. And the reason for this is that FDA is amazingly effective at using the stick and getting the wrong people to be out of business or to improve themselves or to stop that product from reaching the pharmacies. And that’s wonderful. They do a really good job. The other really good job they do is to push this industry, which takes great risks on new products but is very conservative in manufacturing, and it pushed us with the risk assessment and it pushed us with PAT, it pushed us with QbD. They’ve done an amazing job to revolutionize the industry. And when they speak, it’s much louder than what they think, and they have a far greater reach than what they think but they have not been able to help us in industry by telling us what’s good. So at one end of the distribution it’s very, very clear what’s bad. Crystal clear. Form 483s, Warning letters, etc. At the other end, which is [made up of] the ones that go beyond [just compliance] and do it better and want to improve, FDA has no way of telling us because they have no Form XYZ to say “well done!”...now I think [if] the quality metrics remains voluntary as opposed to compulsory, it means they’re going to ignore the majority of the companies doing okay but don’t try to do that much better. This is great, let’s just ignore them...because what we want is to push those that want to be much, much better, and to be rewarded by being singled out in the Dean’s List.

Let me add something, I felt that...the three quality metrics picked—I find them incredibly 20th Century. Incredibly 20th Century. I mean, when we have such amazing, powerful computer [programs], we have huge amounts of data [available] about our [manufacturing] sites, why do we pick three measures that you’re certain everyone’s going to game...what I would imagine is the right way is the use of Big Data, so that you can measure risk [holistically and in an unbiased manner]. You can measure the key weaknesses and strengths of the sites. And if you use Big Data, then you can’t game them..


Carol Montandon, Chief Quality Officer, Vice President, Quality and Compliance, Johnson & Johnson Consumer

There are certain consumer products that you would logically think would be well within the FDA’s target for metrics, and those are some of the ones that my company sells, such as OTC medicinal products for pain or cough and cold symptoms. But, I also have other OTC drugs within the business that I support that are nonmedicinal, such as acne wash, toothpaste, and mouthwash. Some of FDA’s stated goals around metrics are to focus on risk to the patients and to avoid drug shortages from a public health perspective. These nonmedicinal, nondose-limiting OTC products would not provide benefit from this program at all. It’s hard to imagine a world in which a shortage in toothpaste from a particular manufacturer would cause a public health issue. Same thing with a mouthwash or deodorant. And these products, just by the nature of them, there are hundreds and hundreds of them. All of us are consumers. All of us stand in front of those shelves in our pharmacy or our supermarket and see the plethora of products that are out there, and they constantly are changing—adding a new flavor or adding a new scent. The number of products we launch on a yearly basis is in the hundreds. If we were required to collect this metric data in a very specific manner for submission to the FDA, it will be quite burdensome.


Melissa SeymourMelissa Seymour, Vice President, Global Quality Control, Biogen

From a challenge perspective, and as an industry working with the Agency, there are a couple of challenges in getting people involved you’re going to have—most of the people in this room here probably have an intention of participating. There are hundreds, thousands of companies not sitting in this room. So, how do you get information from those companies? How do you get buy-in from those companies? I think the Reporter’s List could be beneficial, but it could also be detrimental. It could force companies to not want to be on the lower tier or to not understand what it means to be put out there and therefore not participate. It could have a reverse effect in that consumers may think that just because a company is on that list that the Agency endorses that organization [that it means all is well, but that may not be the case]. Certainly taking a paused approach on that list would be smart until we get a more scientific understanding of the data that we’re getting—the analytics and what it tells us we’re getting about those companies. Later on, I think that could be an incentive. I question starting out with the Reporter’s List now. The other challenge I think is on the benefits side. Historically, industry has worked with the Agency on PAT, QbD, and we’ve had these grandiose ideas about the benefits we’re going to get on the backside. And, in all honesty, I have not seen many [benefits.] I think we actually have to show some mutual benefits to the organizations that are participating. And so, maybe, in reality, a small number of companies participating in a pilot will provide the opportunity for the Agency and industry to work together to be able to show that benefit because that will pull others along. This close collaboration needs continued industry/Agency meetings, talking very closely with industry... it may cause us to get to a more innovative technology-accepting situation where companies will work in collaboration with FDA.


Harry Jeffreys, VP, Regulatory Affairs and Compliance, Catalent Pharma Solutions

In terms of the two challenges that I see, this new site establishment reporting requirement, it’s a big one for contractors because certainly we’re going to have to report to the license holder about their product and that’s pretty significant in and of itself, though we’re going to have to report to FDA about the site metrics but understand that that still has to go through the license holder—we have to sit with each license holder and talk about their products and bring it all back together again and put it into a standard format and report it. That’s going to be immensely burdensome for contractors. I also think it’s going to muddy the waters a bit; there may be some duplication, some confusing metrics that are going to be reported as a result of that. There are a couple of things to support that. A contractor only sees a subset of the data that a product license holder has...your typical test lab may have a purchase order arrangement with a client, and “we’re going to use USP test such and such or we’re going to do some tests according to a validated method”—but we really don’t get the visibility of how that’s being used—commercial, clinical development, investigational or some other aspect. That sometimes is clear but often it’s not. How should a test site take that and reflect on their quality performance?

I’d like to propose that we create different categories [for site reporting requirements, based upon role, practicalities and potential benefits, up to and including exemption for CDMOs and some other nonproduct license holders from site reporting], similar to what was done for GDUFA II. When it was structured, FDA really took a look at what everybody’s role was, and sort of reallocated things. Looking at the practicalities, the roles, and potential benefits that participants will see. And I’d say up to and including exemption for CDMOs and other nonproduct license holders from site
reporting. [We’d like to see CDMO’s exempted from site reporting and have reporting be through the license holder. Quality metrics reporting should be the responsibility of the
license holder.]