FDA’s MANUFACTURING CHANGE REGULATIONS AND SUPPORTING GUIDANCE

...are under review in the effort to alleviate the supplement burden on already marketed products. As FDA retools its quality regulatory paradigm, industry is affirming that focusing on creating a more risk-based and quality systems-oriented regulatory approach for manufacturing changes made to existing products could help spur firms down the quality-by-design pathway. Ideas that have been put on the table in response to an FDA request for industry input include: *a simplified, two-tier system requiring only supplements for high risk changes * reevaluating the format and content of annual reports and their relationship to annual product reviews * the importance of global harmonization, and * separate consideration of the API context. [Recommendations by PhRMA at an FDA meeting soliciting input on changing the 21 CFR 314.70 regulations are included. A summary of breakout discussions on the post-approval regulatory process at a recent FDA/AAPS/ISPE conference on the quality initiative is included.]

FDA's MANUFACTURING CHANGE REGULATIONS AND SUPPORTING GUIDANCE are under review in the effort to alleviate the supplement burden on already marketed products. As FDA retools its quality regulatory paradigm, industry is affirming that focusing on creating a more risk-based and quality systems-oriented regulatory approach for manufacturing changes made to existing products could help spur firms down the quality-by-design pathway. Ideas that have been put on the table in response to an FDA request for industry input include: ? a simplified, two-tier system requiring only supplements for high risk changes ? reevaluating the format and content of annual reports and their relationship to annual product reviews ? the importance of global harmonization, and ? separate consideration of the API context. [Recommendations by PhRMA at an FDA meeting soliciting input on changing the 21 CFR 314.70 regulations are included on pp. 6-11. A summary of breakout discussions on the post-approval regulatory process at a recent FDA/AAPS/ISPE conference on the quality initiative appears on pp. 21-24].

Industry Backs 314.70 Reg/Guidance Changes To Further FDA's Quality Initiative Goals

Industry is urging FDA to focus on reducing the chemistry, manufacturing and control (CMC) supplement burden for already approved, conventional applications as a potentially important bridge in achieving the goals of its 21^st century quality initiative.

Under the initiative, a lot of the emphasis has been on developing quality-by-design-based new drug applications that will free up firms from the constraints of the current post-approval regulatory process. However, industry is stressing that retooling the 21 CFR 314.70 manufacturing change regulations and/or supporting guidance to reduce the post-approval filing burdens for existing products and to give incentives for improved risk management and quality systems would spur firms down the QbD pathway.

ä The Pharmaceutical Research and Manufacturers of America (PhRMA) was among those espousing this point of view at FDA's public meeting on "Supplements and Other Changes to an Approved Application." The meeting was held at the agency in February as part of an effort to solicit input on issues that it should consider in developing revisions to the 314.70 regulations and guidance.

In a Federal Register announcement in January, FDA explained that it would be considering the input from the meeting as well as from written comments in "evaluating how it could revise its regulations to allow for consideration of risk-based approaches based on manufacturing process understanding, including prior knowledge of similar products and overall quality systems, to provide an enhanced risk-based approach to the CMC regulatory process, which would reduce the number of supplements."

The announcement states that the current 314.70 categorizes post-approval CMC changes and their associated reporting requirements without consideration of the applicant's risk management activities or internal quality systems and practices, and "reflects a rules-based, or prescriptive, approach." As such, the current 314.70 "may create regulatory burdens and costs that discourage beneficial manufacturing changes and may not support a desirable level of innovation, modernization and flexibility for the industry as described in the cGMP initiative," FDA said.

ä The agency specified that to help reach the initiative's objectives and reduce the supplement burden, it is evaluating: ? redefining what it considers to be a major manufacturing change ? reducing the reporting burden for certain changes, and ? creating a new reporting category of manufacturing changes that do not require notification.

The January Federal Register notice called for comments to be submitted by March 7. However, FDA subsequently extended the comment period to May 18.

GlaxoSmithKline CMC Regulatory Affairs, Post-Approval, Regional Director Leo Lucisano delivered the PhRMA comments at the February public meeting (see box on pp. 6-11). The presentation included a discussion of the history and current status of FDA's post-approval change policy and recommendations for better alignment of 314.70 with the goals of the agency's quality initiative.

Lucisano expressed the general consensus that a lot of the supplemental applications being filed "involve a fairly low risk … and really don't add a lot of value and drain resources," stating that PhRMA supports a revision of 314.70 that would serve to reduce them.

However, the revision process, he stressed, "really should focus on the conventional submissions with the realization that we have thousands of approved products, both NDAs and ANDAs that are out there."

Noting that it will be very difficult for companies to go back and invest in quality by design for those products, what the revision should target, Lucisano asserted, is rewarding "manufacturers for taking steps in that direction … and reward the application of prior knowledge."

PhRMA advocates that, "rather than just looking at a change in a vacuum and looking at a prescription in a PAC [postapproval change] guidance, you actually reflect on the product history, maybe the product line that you manufacture, and apply that thinking to how that impacts change - and also that you are willing to invest in risk-based approaches. Because as we found, if you are going to do a valid risk assessment, you need special skill sets, you need to invest additional time, energy and initiative."

In turn, Lucisano proposed that if 314.70 is revised in such a manner as to reward the application of prior knowledge and risk-based approaches, "it will really build a bridge to quality by design and almost accelerate efforts for companies to start embracing that as a normal piece of business in developing their new drug or new chemical entities."

PhRMA Recommends Reducing Reporting Categories

Lucisano went on to present a set of recommendations from PhRMA's pharmaceutical quality and technical leadership committees to help achieve the intended result.

The recommendations include: ? reducing or removing reporting categories that are not necessary ? removing change categories considered low risk ? revising statements not consistent with a risk-based approach ? reevaluating the format and content of annual reports ? borrowing from the risk evaluation experience of the agency's review and compliance offices ? considering a different approach to classifying manufacturing sites ? utilizing research outcomes ? continuing to change the emphasis of guidance documents from prescriptive to conceptual, and ? focusing on the conventional to lay the groundwork for QbD.

Commenting on the recommendation to simplify the reporting categories, potentially eliminating the changes-being-effected (CBE) category, Lucisano explained that the complexities of the current system create a lot of gray areas, resulting in an expansion in supplement filings since pharmaceutical companies generally default to the conservative side.

In separate written comments "strongly" endorsing the PhRMA perspective, Pfizer advocated further a two-tier prior approval/annual report approach. "Collaboration between the agency and industry, the firm said, "should result in a limited number of significant changes or critical situations that would require prior approval by FDA; all other changes should be managed by the firm's quality system and reported in the annual report."

Pfizer also reinforced the general point that FDA's new approach "must be applicable to existing products with 'conventional' applications" in order for industry or the agency to gain any significant benefit with regard to reducing the number of manufacturing supplements. "With robust internal quality systems in place, a manufacturer should be allowed to implement post-approval changes that do not impact safety or efficacy without submission of a supplement," the company stressed.

Echoing other respondents to the FDA inquiry, Pfizer suggested that modifications to the existing process for post-approval changes to an approved application "could be expeditiously addressed by issuance of a new guidance document," and the 314.70 regs "revised at a later date if necessary for consistency."

For newer QbD applications that contain substantially increased product knowledge, Pfizer proposed that "the expectations for post-approval supplements should be based on the availability of enhanced process and product understanding that is included in the registration dossier." For these QbD-based submissions, "the process for how the sponsor will manage post-approval changes should be described in a 'regulatory agreement' between the agency and the sponsor."

PhRMA proposed that the comparability protocol could serve a parallel role to the regulatory agreement for already approved products, and help lay the groundwork for QbD.

The comparability protocol "allows companies the opportunity to go to the agency and say, here is my plan for changes. And if I can convince you that I have a sound plan in place that is science-based and risk-based I can make other changes without filing supplements." At the same time, PhRMA spokesman Lucisano continued, "if the regulations are changed to also reward companies for taking a risk-based approach, it also will reduce the number of supplements that are required. And these two buckets really can be applied to the currently approved conventional NDAs and ANDAs that are out there."

FDA Questions On Changing 314.70

In a January Federal Register notice, FDA asked for input on the following questions related to revising its manu-facturing change regulations and policies:

? Is it valuable for the agency to move toward a more risk-based and quality systems oriented strategy for regulating post approval CMC changes outside of the formal application review process? What are the advantages and/or disadvantages?

? Would revising § 314.70 as described in this notice provide the same level of protection to the public as the current regulatory scheme with respect to ensuring the safety and efficacy of human drugs? What inspectional approaches might the agency consider to evaluate manufacturing changes while ensuring public safety?

? Would revising § 314.70 as described in this notice change the regulatory burden on the pharmaceutical industry? If so, how would the burden change?

? Would reducing the prescriptiveness of § 314.70 provide manufacturers with greater regulatory flexibility? Would it encourage manufacturers to adopt CMC-related risk management strategies? Would there be disadvantages?

AstraZeneca also suggested that "generic" comparability protocols, which set out in advance the acceptance criteria for changes, "should be explored in tandem with the regulatory agreement. This would allow a firm to effect a change(s) in a timelier manner."

Backing a regulation change to reduce or eliminate unnecessary supplement categories such as the CBE-30 and CBE-0, AstraZeneca proposed a "do and tell" notification approach for changes assessed by the firm as having a lower risk of impact on the key product quality attributes. This could involve either immediate notification to FDA or periodic notification, for example through the annual report. In either case, FDA approval would not be required, unlike for the current CBE supplement category.

The company added that the current prior-approval route "should be limited to changes with a higher risk of impact upon the key product quality attributes that are not covered by the regulatory agreement/compara-bility protocol."

Many Supplements Are Low Risk

The associations and companies responding to the FDA inquiry concurred that a lot of the current supplements do not meet the higher-risk criteria. A variety of examples were cited to support the point.

PhRMA pointed to a site change for packaging. In the usual CBE supplement, firms indicate that they are not making any changes to the container/closure system, are putting a batch on stability and verifying that the new packaging site has a satisfactory cGMP approval status for that particular packaging operation. "That is a very low-risk scenario, and we should consider not having a supplement for a scenario such as that," Lucisano asserted at the FDA meeting.

ä In submitted comments, Schering-Plough explained that it conducted a historical review of recent changes filed as either prior-approval or CBE supplements. Based on this review, the firm included with its comments a list of the types of changes which appeared to pose a low risk in terms of impact to product quality and/or patient safety, recommending that these should only need to be filed in an annual report.

The list included: ? release specification changes that improve quality control; for example, the addition of a test to improve product QC ? removal of a manufacturing, packaging, or testing site from an application when an alternative qualified site is already included ? changes in equipment design for a non-critical step in the process ? changes in the order of addition of ingredients for solution dosage forms ? removal of the intermediate stability interval for a product with a well-established stability history ? introduction of a new product or line extension into a multi-use facility that has been producing similar products, and ? changing an analytical testing facility to one with a positive FDA inspection history.

Schering-Plough also said in its comments that it conducted a similar review for biopharmaceutical products, and that the same changes should be considered for reduced reporting requirements in this context as well.

Noting that historically, changes normally covered under GMPs have required supplemental applications for biological products, Schering-Plough maintained that these types of changes "also present a low risk and should be considered for reduced reporting requirements." The firm cited as examples: ? introduction of new products into an approved manufacturing area ? updates to environmental monitoring programs, and ? minor facility and equipment modifications.

Schering-Plough affirmed that in general using the risk-based approach for evaluation of changes for biopharmaceuticals would harmonize the approach with that for small molecules.

The company further recommended that the regulations be revised to eliminate a reporting requirement for compendial changes not relevant to the information in the filing, pointing out that a firm's compliance with the current compendia may be confirmed during routine inspections.

ä Commenting for the Generic Pharmaceutical Association (GPhA) at the FDA meeting, Hospira Regulatory Affairs VP Richard Stec also cited examples of types of changes not warranting supplemental filings.

Examples on the GPhA list were: ? manufacturing changes to companion applications after approval of a lead supplement ? a change to a drug substance or a drug manufacturing process that reduces levels of byproducts or impurities, and ? a move to an alternative testing lab or, for solid dosage forms, an alternate packaging site.

Other examples of changes that could be qualified through a firm's risk-based quality system cited by GPhA were: ? the addition of a new drug substance supplier previously approved in an existing application with the same dosage form ? minor changes in size and shape of the container for a sterile product ? adjustment of in-process specifications based on prior manufacturing history of the firm, and ? deletion of non-compendial tests after an appropriate product history has been collected.

GPhA Highlights Size of ANDA Supplement Workload

In the GPhA presentation, Stec pointed out that the framework for qualifying a change via a quality systems approach already exists within the medical device regulations. In turn, he said, "most elements of the CMC quality system structure are already in place within the pharmaceutical industry to qualify CMC changes. For example, generic manufacturers operate under an integrated quality system structure and set up procedures. Systems are in place for documentation control, IQ, OQ, PQ, equipment, process and method validation, change control and CAPA procedures."

Stec stressed that guidance documents such as the NDA/ANDA changes guidance would continue to be an important element to a risk-based quality system approach. However, he maintained that the content could be restructured to provide greater specificity on major changes requiring FDA prior approval.

GPhA also proposed that decision tree tools could be incorporated as "an effective means to determine if a change could be qualified via a firm's quality systems." Changes thus qualified would be annual reportable.

ä In giving an affirmative response to FDA's general question on the need for a new approach to reviewing and implementing post-approval changes, Stec stressed the current dimension of the regulatory workload involved.

During the product lifecyle, he pointed out, generic firms may submit upwards of 20 or more post-approval supplements to keep an application current. "I think we all agree that the [number of supplements is] very large and contributes to an overwhelming workload, both in the Office of Generic Drugs [OGD] and in ONDQA [the Office of New Drug Quality Assessment]."

Stec then cited the affect of the process on the ability to implement change and reviewed the main drivers of change for the generics industry.

ä A typical CMC post-approval review time for a generic drug application may range from nine to 18 months, extending to 24 months if additional data is required such as impurity qualification. In turn, the timeline for development to approval of a change may range from one to four years.

Stec cited as a typical example replacing a piece of manufacturing equipment in a process line. "The timeline would extend from facility design to build out, equipment qualification, process or analytical development and validation, manufacture of stability batches, and regulatory submission, review and approval."

Changes may be initiated by raw material suppliers, which may discontinue manufacturing the drug substance, "often with little warning," move manufacturing sites, or implement process changes to increase production efficiency.

Application holders, Stec continued, also have to submit a number of manufacturing changes involving: process improvements to improve product quality; installing new equipment or replacing obsolete equipment; consolidating manufacturing facilities; expanding and relocating lines to increase capacity; and providing alternate suppliers for manufacturing ingredients. They must also respond to compendial changes and upgrades to analytical methodology. Firms may also opt to outsource select manufacturing processes or analytical services.

ä Under a new risk-based post-approval CMC change process, major changes such as bringing online a new facility or a new API supplier that may not have been previously inspected by FDA should continue to require prior FDA approval, Stec said.

However, he proposed that changes that may have fallen into the "moderate" category under the old system could be allowed to be qualified by a firm's quality system without FDA clearance, with reporting done either at the time of implementation or within the annual report.

Shifting the burden to industry to qualify moderate changes would "allow the agency to focus its resources on changes that have the greatest potential to impact product quality," the GPhA spokesman maintained. Also, "a quality system approach is anticipated to only minimally increase the scope of GMP inspections, and would provide for faster implementation of change," as well as incorporate quality-by-design principles.

Generic manufacturers, Stec pointed out, "generally hold a broad production experience across multiple products, rather than a single product that could be leveraged to qualify change. A quality system approach is adaptive and responsive to changes in manufacturing technology equipment and practices whereas a prescriptive approach is not." (Continued on p. 11)

Recommendations From PhRMA On 21 CFR 314.70

Providing the PhRMA perspective at the CDER public meeting in February on "Supplements and Other Changes to an Approved Application," GlaxoSmithKline CMC Regulatory Affairs, Post-Approval, Regional Director Leo Lucisano discussed developments in FDA's post-approval change policy and presented some considerations and recommendations for better alignment of the 21 CFR 314.70 regulations with the goals of the agency's 21^st century quality initiative.

I just want to preface my remarks by saying that in the profession of regulatory affairs for chemistry manufacturing and controls, a great deal of attention is placed on working with pharmaceutical development and chemical development in developing new chemical entities, filing investigational new drugs and gaining approval of new drug applications. But if a product is approved, it typically spends the majority of its lifetime in the post-approval phase. It can go on for years and even decades. And it is a very dynamic phase because of changing regulations, changing technologies and changing market forces. So I am delighted to be here at a public meeting today that focuses attention on that phase of the product lifecycle.

I have had the opportunity to specialize in this field for the last 13 years, so I want to spend a few minutes reflecting on the amount of change that I have seen during that interval, provide some recommendations, concepts and considerations that underpin changes to 314.70, talk about the attraction, the importance and the timing of global harmonization - because PhRMA manufacturing companies supply a global marketplace - mention some of the other parallel activities that are ongoing and that could perhaps be integrated in any revision to 314.70, and provide some summary comments.

HISTORICAL PERSPECTIVE

Back in the early '90s with 314.70, the wording was vague, expectations unclear - the vast majority of manufacturing changes being done via prior approval supplement. Due to concerns from industry and a request for more clarity about changes in this area, there was the issuance of the SUPAC-IR Guidance in 1995 ('Scale-up and Post-approval Changes for Immediate Release Solid Dosage Forms'). That was really a hallmark guidance for four reasons:

? One, it was based on research - FDA collaborated with industry to run some biostudies, to look at the impact of formulation and process variables on the bioequivalence of drug products.

? It provided now a new vocabulary, a common language that industry could talk to FDA about with respect to manufacturing, design and operating principles of equipment, dissolution similarity.

? It also provided very clear expectations about the filing category and the data and information package required to progress a specific change.

? But fourth, and maybe the most important aspect for the discussions today, it introduced the concept of risk. It talked about the risk potential of a change affecting the identity, strength, quality and purity of the product. I think that was significant, because we wouldn't be at a juncture here today to talk about quality by design, unless we have been at least living with the idea of the importance of risk assessment for manufacturing change for the last 10 or 12 years.

Between 1995 and 1999, when 314.70 expired, FDA issued a number of other guidance documents, many of them product-specific or topic-specific - for example, about equipment or about dissolution specifications. 314.70 expired in 1999 and then was reissued in 2004. CANA [FDA's guidance on 'Changes to an Approved NDA or ANDA'] was revised also to be aligned with 314.70.

So what you had really was about a 12 year-period where the agency was issuing many guidance documents, so that it came down to a very prescriptive approach. You defined what change you wanted to do, go to the particular guidance document, it would tell you exactly how to progress that change.

But at the same time, around 2002, the agency challenged industry with a new way of thinking, highlighted by 'CGMPs for the 21^st Century - A Risk-based Approach.' And now, we started to see guidances that were more conceptual - the PAT guidance, ICH Q9 for quality risk management - that didn't talk about specific dosage forms, but talked about concepts and ways to approach the assessment of change.

So we are at a juncture today where one can take one of two paths in either assessing change for your currently approved products or how you want to develop your new chemical entities: the prescriptive approach that is represented by the PAC [post-approval change] guidances; or the QbD approach that is highlighted by cGMPs for the 21st century.

This table just shows some of the metrics that were reported to Congress with respect to manufacturing supplements during the six-year interval from 1999 to 2004, when really we were managing change under the Changes Guidance for New Drug Applications and Abbreviated New Drug Applications.

Two important points here, you see that the percentage of prior approvals went from about two-thirds in 1999 to about one-third of the total supplements in 2004. From a manufacturer's perspective that's a positive thing, because changes-being-effected supplements allow you to implement change faster than a prior-approval supplement. The other highlight here - and I think it was also reflected in some of the comments by [Office of New Drug Quality Assessment Division Director Eric Duffy and Office of Generic Drugs Division of Chemistry III Director Vilayat Sayeed] - that we really haven't seen a change in the number of supplements that are filed. So even though the number of prior approvals are significantly reduced, we are still seeing most of the changes being progressed as supplemental applications.

PHRMA RECOMMENDATIONS

So PhRMA supports revision of 21 CFR 314.70 if essentially it reduces the number of manufacturing supplements - and by manufacturing, I also mean changes to analytical testing and also to packaging. I think we are all aware of, and it has been highlighted in some of the previous presentations, that a lot of the submissions that we do are fairly low-risk, and supplemental applications really don't add a lot of value and drain resources.

But in looking to revise 314.70, it should really focus on the conventional submissions with the realization that we have thousands of approved products, both NDAs and ANDAs that are out there. It will be very difficult for companies to go back and invest in quality by design for those products.

But what it should do in any revision is reward manufacturers for taking steps in that direction for quality by design and reward the application of prior knowledge. Rather than just looking at a change in a vacuum and looking at a prescription in a PAC guidance, that you actually reflect on the product history, maybe the product line that you manufacture, and apply that thinking to how that impacts change. And also that you are willing to invest in risk-based approaches. Because as we found, if you are going to do a valid risk assessment, you need special skill sets, you need to invest additional time, energy, and initiative.

And if 314.70 is revised in such a manner as to reward the application of prior knowledge and risk-based approaches, I think it will really build a bridge to quality by design and almost accelerate efforts for companies to start embracing that as a normal piece of business in developing their new drug or new chemical entities.

So, what are some recommendations?

? One, reduce or remove reporting categories that aren't necessary.

Right now, as has been highlighted before, we have two different types of changes-being-effected supplements. There is really not any material difference between the two. We should look into consolidating them or maybe even thinking about eliminating them altogether. Because in practice, if you have a choice between one reporting category or another - whether it is prior approval and CBE or whether it is a CBE and annual-reportable - you are always going to have a gray area of interpretation. And I think pharmaceutical companies in general always err to the conservative side, and that results in a greater number of supplements being submitted.

? Remove change categories that are considered low-risk.

I very much agree with some of the points made by [Hospira Regulatory Affairs VP Richard Stec representing GPHA] with respect to specific changes that are really low-risk. I will highlight a site change for a packaging site. The CBE supplement has three elements to it: Most people indicate we are not making any changes to the container closure system, we are making a commitment to put a batch up on stability, and we are verifying that this new packaging site has a satisfactory cGMP approval status for that particular packaging operation. That is a very low-risk scenario, and we should consider not having a supplement for a scenario such as that.

? In crafting any new wording for 314.70, we have to be very careful about the wording that is used to make sure it is consistent with a risk-based approach.

Any change has a certain amount of risk associated with it. And the job of a team who is conducting a risk assessment of a change is to identify all those risks and to make determinations as to whether or not those risks are acceptable or can be mitigated, or the risk is simply unacceptable and we can't progress that change. So wording it such as ['that may affect'] will [incline] companies to always file supplements, because any change always has risks. So a wording may be that, upon completion of a risk-assessment exercise, if the risks are appropriately identified and if they are appropriately mitigated, then a supplement is not required. So we have to be thinking about a language in 314.70 that is in parallel with the mindset of people who conduct risk assessments.

? If we are going to decrease the number of supplements, we probably have to take another look at annual reports. Because if we are shifting more to annual reports, we have to give some consideration about their role.

One thought is to streamline the requirements by including only an index of changes, and the supporting data available upon an FDA inspection. We see annual reports going in with hundreds of pages; stability data on multiple batches, very detailed description about very minor changes being made to analytical methods. Maybe one way to streamline the review process is to just have the index of changes and have it be incumbent on the field to go to the manufacturing site and make sure that supporting data is available.

And maybe we need to go a little bit further. Again, following up on [Rich Stec's] comments about the importance of quality systems, if we are going to be looking at annual reports, we also need to be looking at the annual product review.

So the NDA annual report: We file it yearly. It is reviewed by Dr. Duffy's staff in New Drug Quality Assessment. It is done on an annual basis, and essentially the annual report talks about the changes that were made in that year to the NDA-registered detail. It also provides the stability profile and the stability data of all of the batches there are in the routine stability testing program.

Now, part 211, cGMPs, is also a requirement. So a manufacturing site has that information available during a site inspection by a representative from the Office of Compliance. It is done annually. But in a way it is a misnomer, because a manufacturing facility that has a modern-day quality system is really doing this product review periodically and almost continuously. The annual product review also has a summary of the changes. In fact, it has a summary of changes that not only affect the NDA, but also that are transparent to the NDA and cGMP. It has a stability profile. And if it is done well, it can be used as a tool for continuous improvement.

So when you look at these two and the content of both of these documents, the intent is really still the same. And that is, you are providing documentation to the regulator to show that your process is under control and that the product that you make at that site meets its regulatory specifications throughout its shelf life. So there is certainly an opportunity here to decrease the number of supplements by putting more of an emphasis on leveraging the amount of work that goes into annual reports and periodic process reviews.

? [Borrow from the experience of FDA's Office of Compliance].

I am pleased to see that as FDA challenges industry to think about quality by design, gaining a greater level of [understanding of] their processes, adopting risk-based approaches, they have been walking the talk. And since 2004, the Office of Compliance has adopted a risk-based approach to determining where to expend resources to conduct site inspections.

They use the three categories of product, process and facility. So for example, a facility that may be considered high-risk or maybe where the FDA should expend their resources for the product: a facility that makes multiple products that are high volume; the products there are narrow therapeutic index, so it is very important that those products are well-controlled and have a very tight drug release. A high-risk facility may be one that has recently undergone ownership [change]. So Compliance needs to go out and make sure that the quality system there still is being maintained to current standards.

At the same time, the Office of New Drug Quality Assessment since its reorganization in November 2005 has been applying a risk-based approach to review, as Dr. Duffy indicated in his earlier remarks. And what we have been seeing is that they have prioritized their review based on high-risk change scenarios and also to assure that there is no disruption of product supply.

So I was delighted to receive a letter several months ago that was an action letter to a supplement that essentially said, 'we looked at your supplement and the change scenario. We consider it low-risk. A supplement is not necessary. Please file it as an annual report.' Now, I was delighted to receive this letter. I took it to my management because I was so excited, never thought I would see the day to see a letter like this. And where I thought I was the great facilitator, my manager was convinced now that regulatory affairs represents the 'Division of Manufacturing Hindrance.' And if you would have told me this was an annual several months ago, we could have implemented it already.

? So we encourage FDA to continue to translate this experience with risk-based review and also risk-based inspections as they consider revising 314.70.

What are some other concepts that should be considered?

? A different approach to classifying manufacturing sites.

Right now, sites are classified according to the particular dosage form that they manufacture and their experience in passing a cGMP inspection. But rewards should be given, maybe, to sites that adopt a truly modern quality system. So that they conduct risk assessments - they have the right personnel to do that. They do real-time trend analysis. They have a change control system in place and corrective and preventive actions policies also in place. And perhaps it is these sites that should be allowed the additional leverage to have these non-reportable changes, because they have demonstrated that they have their product under control and the systems to manage risk.

? As SUPAC-IR was based on research, there is a lot of other research, good research, that has been done since then and should be considered and industry encouraged really to utilize this research in progressing change.

An example being the Product Quality Research Institute, their container/closure group, which is looking at a different way to assess the impact of packaging on product stability, rather than going through the task of actually generating some real-time stability data before the application can be progressed.

? We also encourage this increased emphasis on conceptual guidance documents - from prescriptive to conceptual.

So if you look at the PAT guidance if you read ICH Q9 on quality risk management or the FDA guidance on quality systems, it more or less provides guidelines for teams at manufacturing sites and also in development to embrace and to apply these risk-based approaches and to gain a greater level of process understanding, and to be encouraged and rewarded for applying prior knowledge.

But if the intent of 314.70 and revising it is to build a bridge from the current scenario to where we want to be with quality by design, I think the agency needs to move very carefully in withdrawing any of the guidances that are currently out there and do serve a real purpose for the products that are already approved. The reality [is] that in the majority of cases companies will not go back and invest in those products, but would rather focus resources on quality by design into future new chemical entities.

? But in doing that, if we focus on the conventional, I think it is possible to lay the groundwork for quality by design.

How that would work is like this: We had the draft comparability protocol out there that allows companies the opportunity to go to the agency and say, here is my plan for changes. And if I can convince you that I have a sound plan in place - it is science-based and risk-based, I can make other changes without filing supplements. At the same time if the regulations are changed to also reward companies for taking a risk-based approach, it also will reduce the number of supplements that are required. And these two buckets really can be applied to the currently approved conventional NDAs and ANDAs that are out there.

At the same time, if companies see a reward for taking this approach, they will be more encouraged to apply the concepts of quality by design, establishing design space and the sources of variability, so as part of their new drug application approval, they already have a regulatory agreement in place that will significantly reduce the number of supplements in the future. So by dealing with the present and laying the groundwork for the future, at the end result we have a reduced number of supplements.

Now, I like to kid Dr. Duffy that his end game and mine is that we work ourselves out of a job, because I work in post-approval CMC regulatory affairs. I think it will take some years to get there, but I think it is doable ….

NEED FOR GLOBAL ALIGNMENT

A few notes about global alignment. Pharmaceutical companies supply a global marketplace. And the global regulatory environment that has different philosophies, different systems, really represents a hurdle to continuous improvement and technical innovation.

A couple of weeks ago I visited a manufacturing site with some of my regulatory counterparts from Europe. It was a manufacturing site that supplies a product to over 60 different markets. We were there to talk about redesigning the manufacturing process. And we indicated that even though the FDA regulations were an impetus to change, that long term to gain approval in all 60 of those markets would probably take somewhere between three to five years.

Essentially he had two choices: He could run two different manufacturing processes and test the same product according to two different specs for that five-year period of time, or do a stock build to five years and drain off that stock build until he got approval in all 60 markets. Either scenario is not very appealing. Either scenario is really not a motivator for change.

So really we have a responsibility both in industry and in the agency to promote a more global approach to post approval changes. And maybe the time is just right to progress serious discussion about revising 314.70.

Last year, EFPIA, which is the European Federation of Pharmaceutical Industries and Associations, provided a proposal to the European regulators that was very much aligned with some of the thinking over here in the U.S. with respect to a risk-and science-based approach, the application of conceptual guidances like quality risk management, pharmaceutical development and quality systems, and were suggesting that there just be two buckets of categories except only in the rare exceptions: So essentially minor changes, which could now be done via annual report. Annual report is not a known concept in Europe. But the idea is now being floated. And only major changes really requiring the resources of the regulator to assess and to approve. And also introducing the concept of a regulatory agreement, which has undergone a lot of discussion here between FDA and industry. So the opportunity is probably very good at this time now to engage in discussion with our European colleagues to have a more aligned approach....

SUMMARY

I talked about some of the other activities that are ongoing - a risk-based review, risk-based inspections. FDA has also initiated two other programs, the CMC pilot program and the collaborative research agreement with Conformia, where they have engaged pharmaceutical companies to talk about the challenges of adopting quality by design and how we translate those concepts into regulatory submissions and work toward the day when we will have very few post-approval supplements because we have a fundamental knowledge of how we manufacture our products and the sources of variability.

PhRMA would like to applaud, and as a private citizen I applaud FDA for your initiative, your energy, your investment and your courage in challenging industry and the international regulatory arena to have a new way of thinking about our products.

Should we revise 314.70 at this point in time? Well, it is worthy of consideration if from a resource standpoint it could be done to reduce the number of manufacturing supplements, if it is done from a realistic standpoint that the vast majority of NDAs will not be redesigned according to quality by design. But there should be rewards out there so that from a philosophical standpoint, if a company is willing to invest in prior knowledge and risk analysis, they would have some sort of regulatory downsizing in their applications. From a philosophical standpoint, if it could be done in a manner that it sets the foundation and almost accelerates the adoption of quality by design for our future products. And it is also done from a synergistic standpoint that the learnings that are coming out from the CMC Pilot Program and risk-based review are incorporated into any revisions of 314.70.

So it really should be done if it can represent a step change toward achieving the balance. And what does that balance look like?

From the manufacturer's standpoint: predictability and control of the timeline; that we can be rewarded for process understanding and risk management but still have the flexibility to use different systems, both the prescriptive approach as well as the quality by design and risk-based approach; that we have harmonization across regions, so that very disappointed manufacturing site director a couple of weeks ago has hope for a brighter future; and also that we really maximize the use of our quality systems, if they truly are modern-day quality systems. And I mentioned before, if you have a good quality system in place, perhaps we don't have to report as much information in the annual reports and supplements.

From the agency standpoint: not so much a decrease in review workload as a prioritization, and that those resources are only expended on those changes that represent real risk; that the agency can be seen as encouraging innovation, but still have the ability to exercise their regulatory authority - so when they come to the manufacturing site, they make sure that all the work has been done, they can meet the folks, gain a good understanding about the expertise that was applied to a risk-based approach; and lastly to assure no impact to patient safety. And certainly hearing Mrs. Ritter's comments, I think it drove home the importance and the obligation that we have that we appropriately regulate the post-approval arena to make sure our products are of sufficient quality.

In closing, I would like to thank my colleagues on PhRMA's Pharmaceutical Quality Steering Committee and Technical Leadership Committee for helping me put together this program today.

Industry Input Will Impact Agency Decisions

FDA's public meeting on the manufacturing change issue began with presentations by key policy officials from the review and compliance offices, who explained the dimension of the current problem driving the need for regulatory revision.

ä Office of Pharmaceutical Science (OPS) Director Helen Winkle emphasized that while the agency has given a lot of thought to the issue, "we have not made any final decisions, and the discussion here today as well as the information submitted to the docket will be very influential on us in making our final decision."

FDA's overall quality initiative, she said, is aimed at relying on improved process and product knowledge and a firm's internal change control and quality systems "to really be able to understand the risk associated with the changes and make the changes without FDA approval."

For both industry and FDA, the goal is to reduce the number of post-market supplements, Winkle affirmed. "We are inundated with supplements …. They are time consuming and many of them probably unnecessary, because there is little risk associated with the change."

ä Center for Drug Evaluation and Research (CDER) Deputy Director Doug Throckmorton linked the resolution of the manufacturing change issues with the agency's "Critical Path" initiative, which is focused on encouraging innovation and "consensus development between industry, academia and government in a very effective and efficient way."

The heart of the initiative, he said, is remaining open to new paradigms of manufacturing - "being willing to question our assumptions, being willing to think of new ways to approach things that continue to provide assurance of quality." By contrast, 314.70 is "somewhat prescriptive and rules-based," and does not fully recognize the recent developments in manufacturing and the values of risk management activities and internal quality systems.

Citing Deputy Commissioner and Chief Medical Officer Janet Woodcock's vision of the 'desired state,' Throckmorton summarized that "the evolving manufacturing science promises a new approach to ensuring product quality, with the goal of efficient and agile manufacturing and regulation of pharmaceuticals. Achieving that goal requires that industry, FDA, academia, and the American public confront the assumptions that have guided manufacturing assessments to date and be prepared to change if those assumptions can not be supported."

ä OPS Associate Director for Policy Development Jon Clark explained how the goals of the 21st century quality initiative encompass a regulatory approach that encourages rather than stymies manufacturing change and innovation.

He then analyzed the specific language in 314.70 to show the internal discrepancies - for example in the application of the concepts of "major," "moderate" and "minor" potential for adverse effect, which in turn are correlated with filing categories.

A revised regulation and supporting guidance was issued in 2004 to incorporate these concepts. Later in 2004, FDA issued its final guidances on GMPs for the 21^st Century and PAT, which provide an alternative approach and framework for industry centered on manufacturing science and quality management.

"When applied, these terms [in 314.70] don't really play out," Clark said. Examples listed in the reg under "moderate potential" include an increase or decrease in production scale and certain manufacturing aspects that do not affect the process methodology or process operating parameters, as well as changes in specifications, methods or controls to provide increased assurance of product quality. The terms "collide," he stressed, querying how "increased assurance" and "moderate potential to cause harm" correlate.

Eliminating CBEs Would Cut OGD Workload By 75%

Office of Generic Drugs Division of Chemistry III Director Vilayat Sayeed reviewed the regulatory background for manufacturing changes, provided statistics on supplement submissions to OGD for the last few years, and discussed the impact of the new question-based review (QbR) process.

In fiscal year 2006, OGD received 1,164 supplements in the "major," prior-approval category. An additional 3,514 CBE supplements came in during the year, about two-thirds of which were CBE-30s (2,335) and one-third CBE-0 (1,179). Altogether, the 4,678 supplements filed during FY 2006 represented a 10% increase from the 4,247 supplements filed in FY '05.

Looked at as a whole, about one-quarter of the total supplements filed during FY 2006 were prior approval, another quarter CBE-0s and half CBE-30s. Thus if the CBE category were eliminated as is being proposed and all of these changes reduced to an annual report filing status, OGD's supplement workload would decrease by about 75% without other changes to the process.

ä Sayeed broke down the FY '06 CBE filings by the types of changes involved.

Adding a new manufacturing or testing facility to the existing application was the dominant reason, making up well over half of the CBE-30s filed (1,481).

Control revision, a "catch-all" category, was the next most frequent reason (718), followed by labeling revisions (616), most of which were triggered by the CMC changes made, Sayeed explained.

Fewer in number were microbiology (216), packaging (163) and manufacturing (160) CBEs. Most of the microbiology submissions fell into the CBE-30 category.

Reducing or extending expiration dating (29) and formulation revisions (14), most of which fell under SUPAC level 1, Sayeed explained, made up the bulk of the remainder.

"As you can see," he pointed out, "in the last three years, the Office of Generic Drugs has received close to 10,000 supplements in this CBE filing category as defined under the current CFR and changes guidance." The work of reviewing these changes made to the legacy products, the OGD official emphasized, "continues to pose a tremendous challenge to our review resource management and review resource allocations."

The stress of the growing workload pushed OGD to put a process in place in 2004 to allocate more resources to the supplemental submissions. ("The Gold Sheet" January 2005). As they come in, the supplements are routed through the team leaders and a determination made on assigning review resources based on the product, type of change that is being proposed, and risk associated with that change.

Sayeed pointed out that this is an internal process that allows OGD to manage its review resources and "has worked quite well." However, he stressed, "it does not address the core issue of providing regulatory relief for post-approval changes."

The approach that is currently available for industry for regulatory relief for existing products is use of the comparability protocol, although it is not being used very often to date, Sayeed said.

For incoming ANDA submissions, the effort to address post-approval supplement relief is focused on the new question-based review program ("The Gold Sheet" August 2005).

"In this process, the sponsor can use the knowledge gained in product development, and where applicable, leverage in-house knowledge they have for similar dosage forms and processes in providing a scientific basis for post-approval change management," Sayeed explained. He added that OGD "has positioned itself by implementing the QbR initiative to meet the expectations of CFR revisions."

At the conclusion of Sayeed's remarks, OPS Director Winkle underscored his point that "very clearly the number of supplements coming into OGD is almost overwhelming, and we really do need to look at more flexibility in the regulations to help with some of that burden."

Major Supplements Follow On Heels Of NDA Approval

Office of New Drug Quality Assessment (ONDQA) Division of Post-Marketing Evaluation Director Eric Duffy followed with the new drug perspective on the manufacturing change problem. He reviewed the implications of the new quality-by-design paradigm and pharmaceutical quality assessment system (PQAS), and addressed the mission and risk-based approach of his new post-marketing evaluation division.

ä The Office of New Drug Chemistry was reorganized into the Office of New Drug Quality Assessment in November of 2005 with the objective of implementing the pharmaceutical quality assessment system ("The Gold Sheet" November 2004 and August 2005).

Within ONDQA, separate divisions were formed to focus on pre-market and post-market review activities, respectively. A Manufacturing Science Branch was also established to bring in pharmaceutical scientists, chemical engineers and industrial pharmacists to complement the current review staff and help drive forward the new paradigm.

The mission of his post-marketing division is "very clear," Duffy said: ? "to foster implementation of continuous improvement, innovation and effective manufacturing changes within a knowledge-based framework ? to develop a streamlined review process within that risk-based framework and to capture the knowledge from the evaluation and review, and ? to develop strategies to streamline the review process and to downgrade where possible or eliminate certain types of supplements based upon a risk analysis."

Recognizing that "assigning risk can be in the eye of the beholder," Duffy explained that the guiding principle for his office is to use an approach that "is based upon the impact of a proposed change on product performance to meet patient need" as well as "the extent of product and process knowledge and understanding." Supplements, in turn, "would be triaged based upon a risk assessment, and appropriate resources applied based upon that analysis" - an approach that the division is now following.

ä Duffy reviewed data on NDA supplemental submissions "to illustrate the magnitude of the program." In FY 2006 there were 1,809 filed.

Unlike for generics, he commented, when an NDA product is introduced into the marketplace, there is relatively slim manufacturing experience. "So as a consequence we have seen - and this is statistically derived - between two and three prior-approval supplements for major changes submitted immediately within a year or two after approval."

Duffy noted that the percentages of prior-approval and CBE supplements are similar to the OGD experience, with about a third prior approval, a fifth CBE-0s and a half CBE-30s.

ä The catch-all category of "control revision" was the most frequent reason for CMC supplements in FY '06, accounting for 46% of the filings, followed by manufacturing changes/additions at about 40%. Most of the latter are prior approval.

Packaging changes represented about 11% of the total. Duffy explained that many of these are an outgrowth of a merger to standardize packaging across the new product line and are not of great significance.

Expiration dating and formulation changes added about 2% each to the total. The changes in formulation, Duffy explained, frequently involve multidisciplinary review and potentially a bioequivalence study. The reason the expiration date filings are relatively small, he said, is that in most cases change or extension of expiry can be accomplished according to an established protocol and reported in an annual report.

ä The post-marketing division director concluded by pointing to both opportunities and challenges in the "great task in front of us."

The opportunities "would derive in many respects from the quality-by-design initiative and the risk-based approach to making changes." The challenges, in turn, lie in determining "how does one actually apply quality-by-design principles to approved or legacy products" and also in transitioning between "the current way of doing business and the new way, which is based upon risk." In the meantime, there will be a dual system in place allowing firms to opt to continue with the current system of making post-approval manufacturing changes, he said.

GMPs Accommodate Management-Based Regulation

Duffy was followed at the podium by Office of Compliance Division of Manufacturing and Product Quality Director Richard Friedman who explained the role his office and the inspection program are playing in creating a new regulatory system that will be more amenable to manufacturing changes and "promote continuous improvement and implementation of technological advancement."

"While the CMC review program would be expected to continue with needed oversight of changes that directly impact product safety or efficacy," Friedman commented, "many of the changes that occurred over the product life cycle would be handled by the FDA cGMP program." As a result, "it will be far less common for FDA to ask a firm to delay a change while awaiting FDA review of the modification to their operations."

In the new regulatory paradigm, Friedman said, the CMC review function and GMP programs "will work more synergistically to create an environment conducive to continuous improvement." It will emphasize "the responsibility of the firm to implement effective change control practices and of FDA in its routine surveillance inspection program to verify that changes are adequately implemented." He pointed out that this desired state rests on two fundamentals of cGMP - science-based change control procedures and sound quality risk management.

ä In explaining the shifting paradigm, Friedman drew from a paper published in the Law & Society Review in 2003 (Volume 37, No. 4), which delineated three basic types of government regulation - technology-based, performance-based and management-based.

The review and approval of manufacturing process steps or the associated equipment used for such a process would fall under the technology-based regulatory strategy, Friedman explained - the most onerous of the three.

As stated in the paper, he said, "technology-based approaches intervene in the … production stage, specifying technologies to be used, or the steps to be followed, to achieve a social goal. This type of approach includes regulatory approval of the details of the firm's manufacturing approach, and regulatory permission when a firm would like to change one or more steps in the process or introduce a new technology."

A somewhat lower level of regulatory scrutiny is the review and approval of product specifications. "This is akin to a performance-based regulatory strategy as defined by the authors, and allows a firm to identify the approaches used to meet these specifications, and then holds the firms accountable to do so consistently," Friedman said. The regulator establishes requirements for measuring the product output that must be tested to conform to those criteria.

The third or management-based regulatory approach provides the most latitude to the manufacturer to innovate and improve, Friedman explained. "It is defined as one which requires firms to produce plans that comply with general criteria designed to promote the targeted social goal, and places responsibility on the manufacturer to routinely evaluate and refine their management of issues to reach the stated social objective on a daily basis."

The authors of the paper "clearly encourage" management-based approaches for industries such as pharmaceuticals where there is diversity and rapid change in technology. The advantages are that "they place responsibility for decision-making with those who possess the most information about risks and potential control methods. Thus the actions that firms take under a management-based approach may prove to be not only less costly, but more effective. By giving firms flexibility to create their own regulatory approaches, management-based regulation enables firms to experiment and seek out better and more innovative solutions."

Friedman noted that, in contrast, the authors caution that technology-based regulatory regimes can be problematic for such industries in eliminating incentives for the regulated firms to seek out new technologies that would achieve public goals at a lower cost.

FDA's Friedman On Change Control

At the FDA meeting on 314.70 in February, Compliance Office Division of Manufacturing and Product Quality Director Richard Friedman explained that a quality system should be responsive in identifying the need for changes and then managing the change control program. A GMP-compliant change control procedure will do four basic things, he said:

? The first thing it will do is reliably estimate the risk posed by the proposed change. And just to note that as we move to this paradigm, there is a responsibility of manufactures to handle changes in a way that the right questions are being asked before the change is implemented. A vigorous open discussion of what the issues might be associated with the change, and that means the right scientific disciplines from your company, need to be at the table to estimate the risk accurately.

? The second thing in this change-control procedure is the determination of how much scrutiny should be applied to the change; how much scrutiny is needed. For example, what type of data needs to be generated; is validation or revalidation necessary, who needs to be involved with the internal sign off of the change, et cetera?

? The third is documenting the change and any relevant data or information that is generated.

? And of course, the fourth, could science and quality risk management call for analysis of the data, subsequent to the change in order to ensure its effectiveness. So the final major feature of change control would be to evaluate the actual impact of the change.

The compliance official pointed out that the rules-based, prescriptive approach to regulating post-approval CMC change in 314.70 is an example of a technology-based scheme. In turn, he stressed, converting to an appropriate management-based regulatory approach "would greatly serve to achieve the desired state we have outlined over the last few years."

ä Friedman noted that the 314.70 working group on which he serves has recognized that the agency's cGMP program and its quality system approach do not prohibit or require specific equipment or process steps and "provide an existing platform to institute continual improvement."

"The management-based regulatory paradigm of the cGMPs," Friedman concluded, "provides a foundation to allow for many post-approval manufacturing changes to be properly implemented by firms without prior regulatory oversight." In turn, "FDA's quality systems guidance and the ICH Q10 initiative provide the needed framework to accomplish this goal."

Annual Report Vs. Annual Product Review

The FDA presentations were followed by those from PhRMA and GPhA. Also presenting was Consumer Healthcare Products Association Technical Affairs Director Fred Razzaghi. CHPA represents manufacturers of over-the-counter (OTC) drugs, many of which were originally cleared through the new drug approval route at FDA and thus fall under its manufacturing change provisions.

Following the association representatives. SST Technical Affairs Director Arthur Fabian discussed considerations relevant to active pharmaceutical ingredients (API) and intermediates, and Genentech Regulatory Policy and Liaison Director Earl Dye commented from the biopharmaceutical point of view. Industry consultant Calvin Koerner also took the podium to share his perspective as a former senior CMC reviewer in both CDER and CBER and a lead inspector for prior approval inspections and having worked in industry in quality assurance for a number of years before joining FDA.

Among the PhRMA recommendations espoused by GlaxoSmithKline's Lucisano was that care be taken in overhauling 314.70 and the related guidance to make sure that the language "is in parallel with the mindset of people who conduct risk assessments."

There is some risk in any change, Lucisano explained. Reflecting the way teams operate, appropriate wording, he proposed, "might be that upon completion of a risk-assessment exercise, if the risks are appropriately identified and if they are appropriately mitigated, then a supplement is not required."

ä A keynote in PhRMA's comments, echoed by several individual companies, was the need as supplement requirements are reduced to reevaluate the structure and role of annual reports and their relationship to the annual product review requirements of the cGMPs.

PhRMA along with Pfizer, AstraZeneca, Novartis and Wyeth, suggested that the annual report requirements could be streamlined by requiring only an index of changes, with the supporting data available upon an FDA inspection.

"We see annual reports going in with hundreds of pages; stability data on multiple batches, very detailed descriptions about very minor changes being made to analytical methods," PhRMA spokesman Lucisano told the FDA meeting. He proposed that "maybe one way to streamline the review process is to just have the index of changes and have it be incumbent on the field to go to the manufacturing site and make sure that supporting data is available."

ä Lucisano went on to compare the annual report with the annual product review, noting their duplicative content and function.

The latter also has a summary of changes - not only those that affect the NDA, he explained, "but also that are transparent to the NDA and cGMP. It has a stability profile. And if it is done well, it can be used as a tool for continuous improvement."

The intent of the two, he pointed out, "is really the same. And that is, you are providing documentation to the regulator to show that your process is under control and that the product that you make at that site meets its regulatory specifications throughout its shelf life. So there is certainly an opportunity here to decrease the number of supplements by putting more of an emphasis on leveraging the amount of work that goes into annual reports and periodic process reviews."

AstraZeneca suggested that consideration be given to integration of the NDA annual report and annual product review, "with a preference to eliminate one of these and revise the other." The firm further recommended considering "global coordination of timing of the annual report/product review" and placing emphasis on "inspection of the quality system, rather than filing and review of lists."

The respective roles of the annual report and annual product review received significant attention at an earlier formative conference on FDA's quality initiative held in October 2005. Support was expressed by participants for potentially increasing the role of annual product reviews in the change regulatory process ("The Gold Sheet" November 2005).

Annual Report Guidance Urged Under New Paradigm

FDA in conjunction with the American Association of Pharmaceutical Scientists (AAPS) and the International Society for Pharmaceutical Engineering (ISPE) held another conference to assess the progress of the initiative and discuss the emerging issues in Bethesda, Md., in March, a month after the FDA meeting on 314.70.

ä A series of breakout groups at the conference included one addressing "the post-marketing regulatory process" at which many of the same themes emerged as were sounded in the 314.70 industry comments (see box on pp. 21-24).

Key questions presented for breakout discussion were: ? What would be the requirements of an internal quality system to review and approve changes? ? How should the current process be revised? ? What would be the effect on annual reports? ? What would be the impact on the inspection process? And ? how would design space concepts apply to legacy products?

The objective stated in the breakout session was to "identify opportunities in streamlining the current post-marketing regulatory process in order to make better use of resources by both FDA and industry."

On the issue of what kinds of changes could be handled internally through a firm's quality system rather than through a supplemental filing, participants concurred that the criteria should be the safety impact on the patient rather than what is deemed "critical or non-critical" in the manufacturing setting. However, they also recognized the challenge of determining this safety risk and defining assessment standards given all the product and use differences.

The breakout group expressed support for the concept of a two-tiered, prior approval/annual report system, and echoed the 314.70 comments in proposing that the annual report be limited to an index so that FDA could quickly deal with it upon submission. The agency could then follow up by asking questions on specific changes or review the data on site during inspections.

ä Concern was expressed that FDA investigators might then extensively review each change for compliance with the filing requirement.

The discussion produced the consensus that the inspection process should remain a review of the overall quality system and not become a comprehensive product review.

Presenting the summary of the breakout discussion at the conference, Pfizer Consumer Healthcare Global Manufacturing Compliance Senior Director Richard Norgard explained that the expectation was not that the inspection would be longer. The inspector "would just monitor the system" and make sure that the "science used behind making decisions as to what can be managed internally, and that the information that the firm has to support the change that it made, makes sense … and there would be no further need to look into this."

ä The breakout participants suggested the shift in procedures would need to be supported by revised guidelines on annual report requirements. "If the annual report is going to change, we need some guidelines as to what that means," Norgard summarized.

More generally, the group recognized that "we can't get legislation or regulations to be changed quickly, so we talked about developing some kind of a guidance document on what the new post-approval regulatory process might look like, and let the regulations and legislation follow in its normal order of business."

Wyeth also pointed to the need for such guidance in its written comments on the 314.70 inquiry, recommending that it include specific examples of appropriate areas for regulatory flexibility and how to evaluate risk and document a risk assessment when using this type of approach. The firm added that the recommendations should be generated with input from industry and FDA and implemented with the help of public workshops.

In general, Wyeth recognized that reducing the prescriptiveness of 314.70 would provide manufacturers with greater regulatory flexibility and encourage companies to adopt risk-management strategies for process improvements. However, the firm cautioned, "the impact may not be fully realized by smaller companies with fewer resources to adopt and implement these strategies."

To better clarify the relationship between the degree of process understanding and regulatory flexibility, Wyeth recommended that "the agency, in consultation with industry, develop a SUPAC-like guidance on this topic."

Importance Of International Harmonization Cited

Another very significant dimension of the manufacturing change regulatory problem is the lack of international harmonization. The theme is being stressed by both regulators and industry in trying to craft a less burdensome system.

The idea of the International Conference on Harmonization (ICH) developing a "Q11" guideline that would provide a standard for management of global post-approval changes has been under discussion and was again proposed at the FDA/AAPS/ISPE breakout session.

"We are operating in a global environment, and the thoughts are that someday it would be nice to have one package that industry could file for all the changes for all the countries around the world," Norgard said wistfully.

ä PhRMA underscored the importance of the concern in its comments at the 314.70 meeting.

"Pharmaceutical companies supply a global marketplace, and the global regulatory environment that has different philosophies, different systems, really represents a hurdle to continuous improvement and technical innovation," Lucisano stated.

He cited his experience in visiting a GlaxoSmithKline site in Europe to talk about redesigning a manufacturing process as a case in point. Even though FDA may be encouraging such changes, gaining approval in all of the 60 markets supplied by the plant would probably take three to five years.

Essentially there were two choices, Lucisano explained: run two different manufacturing processes and test the same product according to two different specs for that five-year period, or create a stockpile that would last for the duration.

"Either scenario is not very appealing [and] not really a motivator for change," the GSK official stressed. "So really we have a responsibility both in industry and in the agency to promote a more global approach to post-approval changes."

ä Lucisano pointed to the dialogue going on between industry and regulators in Europe as an indication that "maybe the time is just right to progress serious discussion about revising 314.70" in this context.

He noted that the European Federation of Pharmaceutical Industries and Associations (EFPIA) a year ago submitted a proposal to the European regulators "that was very much in line" with the proposals being made in the U.S. with respect to a simplified risk-based system and increased reliance on the annual report and the regulatory agreement. "So the opportunity is probably very good at this time to engage in discussion with our European colleagues to have a more aligned approach," the PhRMA spokesman said.

In its written comments, Pfizer underscored the point that "the maximum benefit for industry would be in a post-approval change system which is globally aligned." However, the company said, "we also recognize the importance of progress in the near term, and therefore we support proceeding with appropriate revision to the current U.S. post-approval process at this time, while continuing the agency's efforts, both within and outside of the ICH process, to gain global alignment."

ä OPS Director Winkle commented that FDA shares the industry concern about global alignment.

She noted that representatives from other regulatory authorities were invited to present at the meeting but were not able to attend. However, she stated, "I want to assure you as we are looking at 314.70, we will consider this because we agree that it is a very important aspect of what we are doing here."

The importance of global harmonization was also underscored by Deputy Commissioner Woodcock in her presentation on the quality initiative at the AAPS/ISPE/FDA conference.

Speaking of the initiative generally, she stressed that "we have heard repeatedly and we do have this message - and I think the presence of regulators from other parts of the world at this meeting show that we have heard this - we have to harmonize. This isn't going to work if it is just done in the U.S."

The different regulatory requirements are a "set-up for errors" and work against manufacturing control, Woodcock said. "That is something the international regulators have to take under advisement and have to do something about. That is our responsibility." However, she urged those in the manufacturing sector to "keep reminding us, keep pressing us, because this is something we must do."

Finishing the work of the ICH guidances, implementing them and seeing how they are interpreted in the various regions is a key component. However, she noted that "we also need to work outside the ICH" to "bring the world along with us." Industry and professional societies need to join with regulators in putting a lot of effort into this, "because even if we reached agreement within three regions in harmonization and all the other regions of the world have different requirements, we haven't really progressed that much."

Woodcock stressed the importance of developing a global approach to post-approval changes, in particular. Continuous improvement is a "bedrock" of FDA's initiative, she said, "and if you can't improve because you have X, Y and Z countries who have locked you into some paradigm, then we haven't advanced in the way we need to."

Disharmony Particularly Problematic For APIs

The manufacturing constraints and change problems caused by international regulatory disharmony are particularly troubling in the API context, due to the complexities of the distribution chain and the significant differences in the oversight systems.

The Active Pharmaceutical Ingredients Committee (APIC) of Europe's chemical industry association CEFIC has been a strong voice for the problems created by the lack of harmonization and the need for a new regulatory paradigm for APIs that will better accommodate manufacturing improvements and innovation ("The Gold Sheet" January 2006).

"The current post-approval change authorization system is in many situations unworkable, especially within 'multi-customer systems' in which an API is supplied to a multitude of dosage form manufacturers," APIC explained in its comments to FDA on revising 314.70.

The committee expressed strong support for a new risk-based and quality systems-oriented approach that would "provide flexibility for API manufacturers to assume their own responsibility, especially in cooperation with their customers (the dosage form manufacturer) for many of their post-approval changes…b ased on the commitment to adhere to PQAS principles (Q8-Q9-Q10-PAT)."

The system should allow for submission of non-major changes in annual updates/annual reports or "entirely remove the need for submission of those changes and instead require that all information on the appropriate management of such changes will be available for on-site inspections." In situations where the API and dosage form manufacturer are different companies, APIC proposed, "the inspection focus should include the proper functioning of the management of change at the interface(s) between the companies."

A post-approval change system that removes the barriers to innovation in API manufacturing and is workable for all parties involved "will be a crucial step for increasing the safety of medicines," the committee concluded.

APIC/CEFIC member DSM Anti-infectives commented further on problems created by the current change regulatory process for API manufacturers, recommending that if a separate drug master file (DMF) approval process is not forthcoming, "the only way forward lies in a large shift from post-approval change oversight through review of submitted paperwork to oversight through on-site inspections" (see box below).

DSM Comments To FDA On API Manufacturing Change Considerations

The following are comments submitted by API manufacturer DSM Anti-infectives in response to FDA's request for input on revising the 314.70 regulations.

The current regulatory system forms a high barrier to continuous improvement and innovation in particular [for] the dedicated manufacture of APIs. The barrier to improvement is highest with regard to dedicated manufacturers of APIs. The barrier to improvement is highest with regard to dedicated manufacturers of the older, mainstay APIs that are being supplied to multi-customer environments. We understand that a key objective of the FDA Pharmaceutical Quality Assessment System (PQAS) in the 21^st Century is to encourage continuous improvement and innovation in pharmaceutical manufacture, including APIs. Another FDA objective is to drastically bring down the total number of manufacturing supplements [by] up to 80%. Both objectives can probably only be achieved if the regulatory system for APIs will be drastically adapted.

The revision of 21 CFR. 314.70 will be a key step in reaching these objectives. The current DMF-based structure for submitting information on dedicated API manufacture implies that one change in an API DMF may trigger up to hundreds of ANDA supplements. This causes extensive duplication of submissions and review or - much more often - it results in the decision by the manufacturer to refrain from implementing the improvement at all. Even worse: If an involved company would be one with insufficiently high ethics it may also result in implementation of the change without any notification at all - neither to FDA nor to the ANDA holders.

FDA's current review approaches normally require that the impact of each change to API manufacture will be assessed separately for each final drug product, i.e. in relation to each affected ANDA. For this reason the approval of DMFs is until now not being regarded by the FDA as a fully realistic option (even though such option has existed for antibiotic APIs via bulk AADAs, until FDAMA was implemented).

The challenge is to define a new approach with regard to APIs that on the one hand will foster improvement and innovation in API manufacture and on the other hand will continually secure (or even improve!) the safety of medicines. If from a 21^st Century PQAS point of view neither DMF approval, nor limiting the level of detail in DMFs down to [around] 10% of what is the usual approach today would be acceptable options, then the only way forward lies in a large shift from post-approval change oversight through review of submitted paperwork to oversight through on-site inspections.

For situations in which the producer of an API and the producer of the dosage form are two different companies, important focus of inspections will then need to be on the adequate functioning of the interface between these companies. If both companies adhere to modern quality management systems and philosophies, the management of change at such interface will normally receive high priority from both sides.

In multi-customer systems the shift to an inspectional approach will of course still imply the existence of multiple interfaces between an API producer and its customer companies. The appropriate management of change through such multiple interfaces is a difficult task but - as opposed to operating via fully DMF/ANDA-based oversight - a feasible one.

Filling in the detail of the paradigm shift from reviewing paperwork to including change oversight firmly within on-site inspections goes beyond the scope of these comments.

We believe that the above approach will be fully in line with the basic philosophy for the 21^st Century PQAS. It would imply that drastic relaxation of regulatory oversight on information in DMFs should only be available to companies that have committed to the 21^st Century PQAS principles (Q8-Q9-Q10-PAT) and have proven to comply with those.

We also believe that by removing barriers to continuous improvement and innovation in API manufacture and by thus creating a workable system, the dilemmas for industry that may come down to the choice of going "Out of Business vs. Out of Compliance" will be resolved. The chance of companies deciding to choose for un-notified implementation of changes in API manufacture would thus decrease strongly. Therefore, the proposed new approach will not only consolidate the current level of safety of our medicines, it will most probably significantly improve it. The current post-approval change authorization system is inadequate and is malfunctioning with regard to protecting the patient. The need to create something better is evident.

ä SST's Fabian has also been a strong proponent of the need for reshaping the current API regulatory process to allow more flexibility for manufacturing improvements and technological innovation ("The Gold Sheet" May 2003, August 2004 and January 2006).

SST represents API and intermediate manufacturers from around the world, marketing their products to the brand and generic industry in the U.S. This gives his firm "a unique regulatory vantage point" and the ability to "assess the impact of FDA guidance and regulations on these companies, how understandable the regulation actually is, and in fact in some cases how effective that regulation has been." Its suppliers hold Type II DMFs and its customers are either sponsors of ANDAs or NDAs, "and SST is there in the middle to create hopefully a win-win situation," Fabian explained.

SST's role involves maintaining the competitiveness of its suppliers by encouraging innovation, which matches FDA's role in the quality initiative, he pointed out.

Fabian proposed five recommendations that he views as lining up with the principles of FDA's quality initiative: ? revise the changes guidance prior to 314.70 as a more timely approach ? separate the drug substance from the drug product sections in the revisions to better address the different API situation ? address DMF holders in the new guidance/regs and extend the DMF annual update as a change reporting mechanism ? recognize that the final synthesis steps are a continuum where changes are not necessarily of prior approval-level significance if no change to the final API can be demonstrated, and ? redefine "major changes" accordingly.

Explaining his recommendation to extend the use of the annual DMF update to include the reporting of minor changes, Fabian said that "the great advantage of doing this is that we now would have a way to file changes without any additional paperwork going to FDA." The agency already gets annual reports from sponsors and annual updates from DMF holders, he pointed out. "So here we have a way with no additional paper to be filed to report certain types of changes, minor of course."

ä Fabian echoed other respondents to FDA in maintaining that the "real way" to solve the problem of API supplier changes is to approve drug master files.

Noting that idea had been discussed during the 1990s, Fabian "implored" FDA, in the "spirit of the quality initiative for the 21^st century … to reopen that discussion. Because I believe there are many valid responses to the agency's very valid concerns about approving drug master files."

Having a DMF amendment for minor changes with no link to a sponsor filing would be a bridge, Fabian said. "By doing that, the change is looked at, it is approved, and then the DMF holder simply notifies the 15 customers that this in fact has been accomplished."

Biotech, Combo And OTC Firms Also On Board

Manufacturers of biotech, combination and OTC products also expressed support for FDA moving forward in developing a more risk- and quality systems-based regulatory paradigm for manufacturing changes in accord with the ICH Q8-10 guidances.

ä Genentech pointed out that the Federal Register notice, however, was silent with regard to the intent to revise 601.12, which prescribes the requirements for reporting changes to approved biological drug products regulated under the PHS Act.

Noting that many natural and recombinant DNA-derived protein products are regulated as drugs under the FD&C Act, Genentech maintained that "there is no scientific and technical reason that biotechnology and other protein products regulated under 601.12 should be treated differently.

"The increased regulatory flexibility, afforded by the use of risk-based approaches to facilitate innovation and improvements in manufacturing processes to reliably produce pharmaceuticals of high quality, can and should apply to manufacturers of protein drugs and specified biotechnology products. Therefore, to ensure consistency in handling manufacturing changes for this class of products, any revisions proposed to 314.70 should also be applicable to 601.12, and the specified biotechnology products."

Genentech asserted that this common approach would be particularly beneficial to sponsors who manufacture biotech products in both categories. The firm went on to point out that when the manufacturing change regulations were last revised, both 314.70 and 601.12 were implicated.

Critical to the success of this approach, Genentech added, will be coordination between field investigators and center reviewers to assure "clear communication, uniform expectations and a shared understanding of a manufacturer's design space and regulatory agreements, which support a reduced reporting requirement for manufacturing changes."

The biotech manufacturer further encouraged FDA "to work closely with other international regulatory agencies to harmonize their respective variation regulations with any revisions made by the agency to 314.70 and 601.12 so that innovations and improvements in manufacturing processes can be implemented globally without disparate supplement submission."

ä The Combination Products Coalition (CPC) requested that, in making changes to 314.70, the agency "be mindful" of the impact on marketing applications for certain combination products that have a drug or biological as one component.

The coalition recommended that CDER involve the agency's Office of Combination Products in discussions "in order to ensure consistency among all relevant FDA regulations and policies."

In applauding the more flexible risk-based approach being proposed and its relevance to combination products, CPC pointed out that "the regulatory landscape for combination products continues to evolve in response to the rapid technological advances that are occurring with these products. We therefore believe that the regulation should set forth general parameters for post-approval modifications, and that more specific approaches should evolve through guidance documents developed under the agency's Good Guidance Practices. Such an approach will allow the combination product regulatory scheme to evolve appropriately."

ä Offering equally strong support from the OTC perspective, CHPA's Razzaghi pointed to the indication and dosage form as potentially primary considerations for a risk-based regulatory scheme. Secondary considerations, he commented, may include "length of time on the market for an OTC product, the safety profile and from a compliance perspective, the risk profile of the firm."

Razzaghi pointed out that the existing OTC monograph system "provides a framework for regulation of drugs outside the application review process that we are talking about here today. This new approach may include changes from an NDA to an OTC monograph status as well as enabling quality by design," the CHPA official said. ¨ ¨

AAPS/ISPE/FDA Conference On Post-Marketing Regulatory Process

At the AAPS/ISPE/FDA conference on FDA's pharmaceutical quality initiative in early March, a breakout session was held on the "post-marketing regulatory process." The following is a summary of the breakout discussion presented at the final day of the conference by Pfizer Consumer Healthcare Global Manufacturing Compliance Senior Director Richard Norgard.

I think it goes without saying that the number of supplements that are developed by industry and issued to FDA over the course of a year is really a burden to both industry and the agency. So we accept that, and we have the five points below that really summarize the five questions that we had which kind of got to that burden:

? The first was requirements of the internal quality system to review and approve changes. What would an internal quality system look like, what would it be composed of, to manage changes internally?

? The second point: if we were to make a recommendation to revise the current process, what would it look like? What would it be made out of? How would it be handled?

? If the recommendation were accepted, what would be the effect on annual reports?

? What would be the impact on the inspection process? Would it make the inspection process longer or shorter?

? And what would be the considerations for the design [space] concepts for the old legacy products? We are going to explore that as we move forward here.

The objective stated in the breakout session was to gather information and identify opportunities in streamlining the current post-marketing regulatory process in order to make better use of both industry and FDA resources.

Shared Understanding and Agreements

I think we all recognize that the post-approval regulatory process really needs revision based on risk analysis. And to do so you have to have a system in place that will help us to decide what that is. So, the suggestion of the team really was to change the current regulations and eliminate all CBE supplements. This would reduce the options to two: That there would be prior approval supplements, with everything else falling into the annual report.

So we talked about the firm's quality system. We wanted to know what kinds of changes could be handled through a firm's quality system. Could they be critical or non-critical? And we really thought that they could [be both]. We did not think that critical vs. non-critical would really be the designation between what is prior approval and what is not. We think that that should be more of what would have a negative impact on the patient, or potentially an effect on a patient. You will see that in the slides to come.

We thought that there should be a single robust global change management system for those companies that have multiple sites. We felt that all manufacturing sites of a firm should have one system, that they should operate in a similar manner with similar standards, similar processes, so that all changes would be handled in a consistent manner.

The recommendation was that there be a multi-disciplinary review during the change management system. Then of course, any good change management system should be supported by a knowledge management system to capture data and to maintain it and disseminate it when needed.

When we talked about then what might be held internally vs. what might be sent to the FDA for prior approval we started to develop a list. The participants quickly came to the conclusion that the list probably would not be so useful, because it is almost a certainty that if you decided to try to list everything that should or should not you would miss something. There is such a wide range of products and processes it is hard to anticipate everything. So, rather than create a list of what might be submitted, we thought that it would be helpful to understand that the only things that should be submitted were those that have a potential negative impact on safety to the patient. And this kind of is universally accepted. [However, determining this] remains one of the challenges. Again, recognizing that there are different products and different processes and that risk is different for different product categories, it is kind of hard to imagine, at least in the short period of time that we had to kind of distill this down.

We talked about annual reports. If you accept the fact that there are only two categories - prior approval supplement and that there is everything else in the annual report - the question came to then what is not in the annual report? We certainly did not want to hit everything that would be in all of the CBEs over the course of the year and send it in an annual report that used to be, oh I don't know, perhaps a half an inch thick to maybe a couple of feet thick. The idea was not to supplant one with the other.

So, the thought was that the annual report really should be limited to a summary or an index report [of changes], so that the FDA could quickly deal with this upon submission. And frankly if they had questions they could come back and ask about an individual item. And that this information would be available to review of course during inspections on a routine basis …. There was a recommendation that there is a lot of information in the annual product review that might be added to the annual report that might be helpful for the center organization to understand some of the changes that have occurred. So that concern was on the inspection process.

Again, to accept that many changes are handled internally and that these changes are kept by the manufacturing site, people were concerned well then the FDA investigators will show up and they will go through each and every one of the changes that perhaps you should have sent in. And I think we wanted to make it clear that the inspection process remains a review of the overall quality system and really is not a comprehensive product review.

There was concern that the investigation would last longer. But I think [participants felt it] really should not be any longer - that the inspector again would just monitor the system and see if the science used behind making decisions as to what can be managed internally, and that the information that the firm has to support the change that it made, makes sense … and there would be no further need to look into this.

We certainly did not want to … have it thought that there would be a shift of the burden from the center to the field, because it is not going to be that. But there certainly would have to be coordination between the field and the center - that the field would understand that they would have to be monitoring this as they do their routine inspections, that they monitor it as part of the quality system. And that quite candidly I think the term is a handshake between the center and the field to understand who is responsible for what.

We talked about design space concepts for some legacy products. I think it was a pretty good recommendation not to go out and try to develop all new information. I don't think any company that has some of these older products ought to spend the time and effort. It doesn't make sense to spend the time and effort to develop design space around these older products. However, it is universally accepted that there is a huge body of information, of data, that really can be leveraged into product knowledge that then through some sort of risk assessment could be used to redefine or to define perhaps design space as we know it today. There was some consideration for updating or preparing your filing to update the design space, so then there would be no question as to if changes were made later on within the design space that that would not require a filing to FDA.

Remaining Challenges and Recommendations

So what are the remaining challenges? I think first of all, on both the part of the agency and the industry, there really is an urgency to develop some short term improvements here. As we started out, I think [Office of New Drug Quality Assessment Director] Moheb [Nasr] yesterday when he introduced this session talked about what changes could be made right now to the post-approval system for improvement. We recognize that this is going to take changes to regulations and legislation and guidances, and that that is a real challenge and a real lengthy process from what I understand. And we are looking for something certainly faster than that.

There is also a need to identify the risk analysis. Not so much critical vs. non-critical, but quite candidly how does one determine if a change that you are considering potentially [has] a negative affect on the patients' safety?

So what are the recommendations? Here it is stated again: the recommendation simply is to eliminate all CBE supplements [CBE-0 and CBE-30] to reduce the options to prior approval supplements [PAS] and annual reports. And once again, that only changes that have the potential for adverse consequences to the patient should require prior approval from the FDA, and that all others should be reported in the annual report. To do this we believe that we need revised guidelines on requirements for the annual report. If the annual report is going to change we need some guidelines as to what that means.

We have to make sure that there is an understanding, as I mentioned earlier, between the field and the center for roles and responsibilities in the new paradigm.

And as soon as I put the slide up I saw people giggling about the concept of "ICH Q11". This came up yesterday that perhaps there would be a need for an ICH standard for management of global post-approval changes - not only for changes within a firm with multiple sites. But we are operating in a global environment, and the thoughts are someday it would be nice to have one filing package that industry could file for all the changes for all the countries around the world. What a wonderful place this would be.

So how are we resolving remaining challenges? Well, we recognize that we can't get legislation/regulations to be changed quickly, so we talked about developing some kind of a guidance document on what the new post-approval regulatory process might look like. And let the regulations and legislation follow in its normal order of business.

And then of course we urge FDA to work with industry representatives to define what 'a potential negative impact on safety to a patient' is.