Harvard Law
School Harvard Law
School
THE REGULATION OF DRUG MANUFACTURING CHANGES:
by Peter J. Schildkraut
for Peter Barton Hutt
Food & Drug Law
January 24, 1997
INTRODUCTION
The manufacturing processes and facilities used to produce a new drug or biological product are under change throughout the investigation of the product and after marketing approval is obtained from the FDA. Innovations are sought to reduce impurities, increase yield, reduce the complexity and time required for manufacture, eliminate equipment, automate procedures, increase stability, and otherwise to improve the drug and reduce its cost. The benefits of these innovations are passed on to the consumer in the form of improved products and lower prices.[1]
With these words, the Senate Committee on Labor and Human Resources introduced its efforts to reform the Food and Drug Administration’s (FDA) treatment of proposed changes in the manufacturing of pharmaceuticals. The committee echoed the industry’s complaints that the FDA undertakes (and forces companies to await the outcomes of) more thorough reviews of such proposals than are necessary to preserve drugs’ safety and efficacy. [1]
According to manufacturers and the committee, these requirements and delays have some unwelcome effects on public health, consumers, industry, and the U.S. economy. First, since foreign countries impose lesser regulatory burdens, “many companies have established manufacturing facilities abroad, where they can use a modern process to supply a drug to the rest of the world long before they can use the same process to supply the United States.” [1] Unfortunately, it is extremely difficult to come by data to back up this claim. For political reasons, manufacturers do not publicize their moves to offshore production, and the industry association (the Pharmaceutical Research and Manufacturers of America or PhRMA) does not survey its members on the subject. [1] Second, the cost and length of the regulatory approval process induces some companies not to attempt to introduce “important drug manufacturing improvements.” [1] Third, drug prices remain “unnecessarily high” whether or not manufacturers institute changes either because they retain the older, more expensive technology or because they invest resources in unwarranted regulatory reviews. [1]
Although FDA reform legislation died in the 104th Congress, congressional leaders will continue to push the proposals in the 105th Congress. [1] This paper examines the history of FDA regulation of manufacturing supplements for drugs, [1] summarizes the current equivalent regulations of the European Union (EU), and concludes by discussing some current proposals for reforming the U.S. rules.
THE HISTORY OF FDA REGULATION OF MANUFACTURING SUPPLEMENTS
Over the last three and one-half decades, the Food and Drug Administration has gradually loosened its controls on manufacturing changes to approved new drug applications (NDAs). The relaxations in the 1960s appear to have been simple, common-sense reactions to presumably unanticipated and completely indefensible rigidities in the regulations, and I would guess that they were noncontroversial. [1] However, the history since the 1970s has reflected the ebb and flow of the power of the agency’s chemists.
The chemists’ institutional culture has reflected an extreme risk aversion in approving drugs and changes to approved drugs. [1] As the McMahon Commission noted in 1982:
Difficulties in satisfying NDA requirements concerning chemistry and manufacturing controls have significantly contributed to delays in the approval of new drugs. [A 1980 General Accounting Office report] concluded that the FDA’s review of the chemistry portion of the NDA takes longer than do the pharmacological and medical reviews. Subsequently, the FDA reported that the majority of the deficiencies in NDAs are related to chemistry and manufacturing. ... [T]he Commission believes that [the] significance [of the chemistry requirements] to the public health is rarely of the same magnitude as the kinds of issues regarding clinical evidence of safety and effectiveness discussed elsewhere in this report. The chemistry problems stem from a different set of reasons; they often appear overly technical and trivial.[1]
This cultural conservatism stems from the criticism the FDA received after one drug disaster in the early 1940s and another in the early 1960s. [1] In both cases, manufacturers created generic drugs through different modes of synthesis from the pioneers, and these new modes of synthesis unexpectedly left behind harmful residues. [1] As a result, agency chemists have sought to review thoroughly all potential chemical changes in drugs. [1]
Historically, FDA managers have had great difficulty in limiting these reviews because, ultimately, they have had to depend upon the chemists’ expertise to determine what information FDA needed to judge the safety and efficacy of a drug. As the McMahon Commission noted, “The senior line managers usually have backgrounds in medicine or biology and may not fully understand chemistry problems. As a result, requests by chemists for more information, additional specifications, or special analyses cannot routinely be reviewed by these managers for reasonableness or justification.” [1]
Over the last few years, however, two changes have altered this entrenched culture of risk aversion. First, Dr. Roger Williams, a Deputy Director of the FDA’s Center for Drug Evaluation and Research (CDER), has had the chemistry background to supervise the Center’s chemists meaningfully. [1] Under his leadership, the agency has liberalized the rules governing manufacturing changes through the Scale-Up and Postapproval Changes (SUPAC) guidance described below. [1] Second, a new generation of chemists with a less conservative attitude towards risk has entered the agency. [1] Together, these changes have created an environment in which the industry can more easily make manufacturing changes that have no significant risk of affecting safety or efficacy.
1963: The Initial Regulation
Following the enactment of the Drug Amendments of 1962, [1] the Food and Drug Administration substantially expanded its regulations enforcing the Federal Food, Drug, and Cosmetic Act in 1963. [1] This massive rewrite included a new section governing changes to approved NDAs. [1] Section 130.9(a) provided that “[a]fter an application is approved, a supplemental application may propose changes” [1] and stipulated that a supplement
should be submitted for any change beyond the variations provided for in the application (including changes in the scale of production, such as from pilot-plant to production batch), that may alter the conditions of use, the labeling, the safety, effectiveness, identity, strength, quality, or purity of the drug, or the adequacy of the manufacturing methods, facilities, or controls to preserve them. [1]
Subsection (c) gave manufacturers a huge incentive to wait for FDA approval of such supplements before implementing any changes; this provision permitted the agency to suspend or withdraw the approval of an NDA if the manufacturer marketed a “changed” drug prior to FDA approval of the supplement. [1] However, the regulation did offer a safe harbor for “changes ... that are not significant from the standpoint of safety or effectiveness.”[1] But, the value of this safe harbor to manufacturers was limited by a further provision that the agency would determine “whether the approval of a supplemental application is required for such change or changes.” [1] Thus, a manufacturer that believed its proposal was insignificant and wished to avoid filing a supplement still had to “submit ..., in writing, full details of any proposed change or changes”[1] and wait for the FDA to concur or to request a formal supplement. In effect, manufacturers lacked the authority to implement a manufacturing change without some form of agency preclearance.
1965: The Erosion Begins
The Food and Drug Administration quickly realized that these regulations were too restrictive. In 1965 the agency amended the safe harbor provisions to enable manufacturers to effect certain improvements in safety and effectiveness without any form of premarketing approval. [1] Among the three listed changes that “should be placed into effect at the earliest possible time” were
[c]hanges in the methods, facilities, or controls used for the manufacture, processing, packing, or holding of the drug (other than utilization of establishments not covered by the approval that is in effect) that give increased assurance that the drug will have the characteristics of identity, strength, quality, and purity which it purports or is represented to possess. [1]
Under this revision, manufacturers had to submit a supplemental application for such changes, [1] but the FDA committed itself “to take no action against a drug or applicant solely because” the manufacturer has implemented the changes prior to FDA approval. [1]
1968: The FDA Relaxes Further
Three years later, the agency first permitted manufacturers to notify it of manufacturing changes that are not expected to have a material negative effect on safety or effectiveness subsequent to the changes’ implementation. [1] Instead of filing a supplement, the companies could merely notice certain specified changes in the next periodic report on the drug. [1] The specified nonmaterial changes included “A different container size for solid oral dosage forms where container and closure are of the same materials as those provided for in the approved application”; [1] a “[c]hange in personnel not involving new facilities”; [1] “[c]hange in equipment that does not alter the method of manufacture of a new drug substance or dosage form of a new drug”; [1] a “[c]hange from one commercial batch size to another without any change in manufacturing procedure”; [1] a “[c]hange to [a] more stringent specification without altering the method described in the approved application”; [1] the “[i]nclusion of additional specifications and methods without deletion of those described in the approved application”; [1] and the “[a]lteration of specifications or methods for inactive ingredients to bring them into compliance with new or revised specifications or methods in an official compendium.” [1]
The 1968 amendments did reduce the reporting requirements facing pharmaceutical companies. Nevertheless, the exceptions were quite narrow, so the regulations continued to force manufacturers to file supplements and incur the costs of awaiting agency approval in order to incorporate improved production technologies.
1985: The Mid-80s Rewrite of the IND [1] /NDA Regulations
Over the next seventeen years a public perception gradually grew that the Food and Drug Administration’s approval processes for investigational new drug applications and new drug applications (including supplements) were too rigorous. As the FDA described this history in 1982:
This country’s drug approval process has been the subject of extensive debate over the past decade. Critics contend that FDA’s regulations and testing requirements are excessive, thereby stifling innovation and depriving the American public of beneficial new therapies. They further contend that FDA’s review of new drug applications ... takes much too long and that, as a result, new therapies are often available in foreign countries before they are available in this country.[1]
Leaders of the House Committee on Science and Technology like James H. Scheuer (D - N.Y.); Albert Gore, Jr. (D - Tenn.); and George E. Brown, Jr. (D - Cal.) became concerned that the agency’s attitude towards risk was unbalanced. In essence, they feared that the FDA was so focused on maximizing its certainty of the safety and efficacy of approved drugs that it ignored the public health (and other) risks associated with delaying the approval of new drugs or alterations to existing ones. [1] Although it couched its language more diplomatically, the McMahon Commission strongly implied that “[t]he present NDA requirements for information about chemistry and manufacturing controls [were] excessive.” [1] In fact, the President’s Task Force on Regulatory Relief, chaired by then-Vice President George Bush, named the drug approval process as one of its top twenty priorities for reform. [1]
In the face of this pressure, the chemists had to back down. During the course of the FDA’s own review of the IND and NDA procedures, the Pharmaceutical Manufacturers Association and Parke-Davis petitioned the agency “to expand the kinds of changes an applicant can make under an approved application and place in effect before receiving agency approval of the change.” [1] When the FDA proposed its NDA rewrite in October 1982, it included a substantial amendment to the regulations for supplemental NDAs, [1] which was largely retained when the agency promulgated the final rule in February 1985. [1]
Although the 1985 rewrite preserved the prior three-tier system (filing a supplement for premarketing approval, filing a changes being effected supplement, and annual reporting), [1] it established a new baseline for reporting changes to approved NDAs. Instead of providing that a change would require premarketing approval of a supplement unless it fell within one of the listed exceptions, the regulation made post hoc annual reporting the default. [1] Changes that still required premarketing approval included “establish[ing] a new regulatory analytical method” for the drug substance or drug product; [1] “delet[ing] a specification or regulatory analytical method” for a drug substance or drug product; [1] “chang[ing] the synthesis of the drug substance”; [1] “us[ing] a different facility or establishment to manufacture the drug substance” unless (a) the manufacturing process does not undergo a “material[]” change and “(b) the new facility or establishment has ... received a satisfactory current good manufacturing process (CGMP) inspection within the previous 2 years covering that manufacturing process”;[1] “chang[ing] the method of manufacture of the drug product, including changing or relaxing an in-process control”; [1] and “us[ing] a different facility or establishment, including a different contract laboratory or labeler, to manufacture, process, or pack the drug product.” [1] As in the 1968 revision, “[a]dd[ing] a new specification or test method or changes in the methods, facilities (except a change to a new facility), or controls to provide increased assurance that the drug will have the [purported or represented] characteristics of identity, strength, quality, and purity” was to be covered by a changes being effected supplement. [1] In addition, new facilities or establishments not needing premarketing approval under 21 C.F.R. § 314.70(b)(1)(v) could be brought on-line at the time of the supplement’s filing. [1]
Interestingly, the chemists reasserted themselves in the FDA’s internal battles between the proposed and final regulations. While the pharmaceutical industry believes the proposal contained “very good language,” it regards the final rule as “a step backwards.” [1] Prior approval was restored for new contract laboratories and labelers, changes in the synthesis of drug substances, and new facilities and establishments for manufacturing drug substances in certain circumstances. [1] Nevertheless, the industry had high hopes that the 1985 rewrite would free them to incorporate new manufacturing technologies efficiently, but these hopes were never fully realized as the chemists retained control over the implementation of the new regulations. [1]
1990: The Lasagna Commission [1] Report
The Lasagna Commission was formed to study possible statutory, regulatory, and administrative “changes that, while preserving protection for patients, would accelerate the” development and marketing of new treatments to cancer and AIDS patients. [1] The Commission strongly criticized the FDA’s handling of manufacturing supplements: “[S]upplemental NDAs relat[ing] to changes in manufacturing and other technical modifications to an approved NDA ... consume a great deal of FDA resources to no great social purpose.”[1] It proposed a default rule of approval after 180 days, placing the burden on the agency to provide “a detailed an specific reason why the change would result in a safety hazard” to block approval. [1] While neither Congress nor the FDA adopted this proposal, one may presume that it influenced congressional consideration of the Prescription Drug User Fee Act of 1992 [1] (PDUFA) (described below) and placed pressure on the FDA to reduce the burden of its approval process on manufacturers.
1992: The Prescription Drug User Fee Act of 1992
By the early 1990s a consensus emerged that the FDA lacked the resources to review product applications (including manufacturing supplements) expeditiously. Since “[s]upplemental applications for manufacturing changes ha[d] ... traditionally been given a very low priority within the FDA,” it frequently took years for manufacturers to receive premarketing approval.[1] Recognizing that Congress would not provide the necessary additional resources given concerns about the federal deficit, the pharmaceutical industry agreed to beef up the agency through user fees. [1] In exchange, the industry persuaded Congress to require the FDA to report annually on its progress in meeting statutory goals for timely approvals. [1]
For manufacturing supplements, PDUFA has worked incredibly well. In fiscal year 1995 the Centers for Drug Evaluation and Research and Biologics Evaluation and Research combined to review eighty-nine percent of manufacturing supplements within six months of submission (exceeding the statutory goal of seventy percent). [1] During the first six months of fiscal year 1996, the two centers had a combined on-time performance of ninety-six percent, and the FDA projected they would ultimately exceed the eighty percent goal. [1] As a result of this legislation, the premarketing approval process causes much less delay (albeit at the cost of the user fee) for companies that want to upgrade their manufacturing technology. No longer will FDA supervisors permit
the chemists to sit on supplements for a matter of years as they used to do. [1]
1995: Reinventing Government and the SUPAC
In its effort to demonstrate that New Democrats are business-friendly and opposed to regulation for regulation’s sake, the Clinton Administration launched its Reinventing Government initiative headed by the National Performance Review (NPR) under Vice President Gore. Perhaps reflecting the Vice President’s continued interest in the subject from his service on the House Science Committee in the early 1980s, the NPR included the regulation of drugs and medical devices among its focuses. In the spring of 1995, it issued a report on the reinvention of FDA’s operations in this area. [1] Complaining that “[m]anufacturers must often wait six to twelve months to receive FDA approval [of a manufacturing supplement], during which time the manufacturers are prevented from making changes to the product or production facility that they believe are more efficient or otherwise necessary,” the report stated, “FDA will reduce the number of changes that require pre-approval.” [1] The report then explained that the agency’s CDER was “developing a guidance document for drugs in tablet and capsule form (other than those for controlled release) ... [which] would distinguish changes that are unlikely to have any detectable impact on a drug product’s quality and performance from those that could have a significant impact.”[1] (In addition, the FDA was to issue “similar guidance document[s] for other dosage forms” by the end of 1996. [1] )
As promised, the FDA released the guidance document for immediate release drugs (the SUPAC Guidance) in November 1995. [1] The regulations on supplements direct applicants to follow “a guideline, notice, or regulation published in the Federal Register that provides for a less burdensome notification of the change” than required by the regulations themselves. [1] The SUPAC Guidance invoked this provision to reduce the reporting requirements for changes including those in manufacturing sites, changes in batch size, and manufacturing processes. [1]
The SUPAC Guidance divided site changes into three “levels” and emphasized that such changes should ordinarily not demand premarketing approval of a supplement. [1] Level 1 changes, which manufacturers may merely notice in their annual reports,
consist of site changes within a single facility where the same equipment, standard operating procedures ..., environmental conditions ... and controls, and personnel common to both manufacturing sites are used, and where no changes are made to the manufacturing batch records [with two exceptions]. Common is defined as employees already working on the campus who have suitable experience with the manufacturing process.[1]
Level 2 changes, which require a changes being effected supplement as well as annual reporting of long-term stability test data, “consist of site changes within a contiguous campus, or between facilities in adjacent city blocks.” [1] Otherwise, they have the same limits as Level 1 changes. [1] Level 3 changes, which require the same reporting as Level 2 changes, “consist of a change in manufacturing site to a ... campus ... that is not on the same original continuous site or where the facilities are not in adjacent city blocks.”[1] Otherwise, they also have substantially the same limits as Level 1 changes. [1] While not substantially changing the requirements for drug substances, [1] the SUPAC Guidance did significantly liberalize the rules for drug products, for which the 1985 Rewrite insisted upon premarketing approval for new manufacturing facilities and establishments. [1]
The SUPAC Guidance also clarified that changes in batch size do not require premarketing approval, provided that the batch does not fall below 100,000 dosage units;
[t]he equipment used to produce the test batch(es) is of the same design and operating principles; ... the batch(es) is (are) manufactured in full compliance with CGMP’s; and ... the same SOP’s and controls, as well as the same formulation and manufacturing procedures, are used on the test batch(es) and on the full-scale production batch(es).[1]
Changes of up to a factor of 10 in the approved batch size (Level 1) need only be listed in the annual report [1] while manufacturers must file a changes being effected supplement for all other changes (Level 2). [1] Manufacturers must also provide long-term stability data for Level 2 changes in the annual report. [1]
Instead of requiring premarketing approval for all changes in equipment type, the SUPAC Guidance also permits annual reporting of (1) the automation of ingredient-moving equipment and (2) equipment changes that do not alter the equipment’s design or operating principles (Level 1). [1] Manufacturers still must receive premarketing approval for Level 2 equipment changes (“to a different design and different operating principles”). [1]
Likewise, the SUPAC Guidance reduced the premarketing approval requirement for certain manufacturing process changes. Level 1 process changes (those that retain the same basic process and remain “within application/validation ranges”) are to be noted in annual reports. [1] When manufacturers retain the same basic process but move outside these ranges (Level 2), they must file a changes being effected supplement. [1] Finally, “change[s] in the type of process used” (Level 3) cause the company to have to file a supplement and wait for premarketing approval. [1]
The SUPAC Guidance was “‘a sea change in FDA operations’ [since it was] the first time the agency has internally generated serious deregulation.” [1] It should eliminate about twenty percent of the roughly 1,200 manufacturing supplements the FDA used to receive annually. [1] Moreover, the guidance has clarified what data and documentation the agency expects from manufacturers for each type of change — a marked contrast from the frequent uncertainty of the past. [1]
EUROPEAN UNION PROCEDURES
The European Union requires premarketing approval for all “variations” (a supplement) to “marketing authorizations” (approved NDAs). While formally this procedure is more restrictive than the American practice, the EU commits to providing approvals or disapprovals much more rapidly than the FDA, so companies can make many manufacturing changes in Europe earlier than in the United States. On the other hand, they may have to devote greater resources to applying for European variations since the U.S. annual reporting procedure for insignificant alterations is less burdensome than filing a supplemental NDA. [1]
The EU has two different methods of approving marketing authorization applications, a decentralized (but concerted) procedure conducted by the various national regulatory authorities and a centralized procedure run by the European Medicines Evaluation Agency (EMEA). [1] Separate regulations provide for variations to marketing authorizations granted under each of the procedures. [1] However, the basic principles underlying both procedures are identical. The EU divides changes to marketing authorizations into three categories: minor (type I) variations, major (type II) variations, and changes that are so fundamental as to require separate marketing authorizations. [1]
The Commission regulations give the relevant regulatory authorities sixty days to act on a major variation [1] application (although they may ask the applicant for additional information one time which adds another sixty days to the clock). [1] But, this guideline appears to be hortatory since major variations do not take effect without actual (as opposed to constructive) regulatory approval. [1]
For minor variations, the relevant regulatory authority or authorities have thirty days to disapprove the application and “stat[e] the objective grounds on which its opinion is based.” [1] If the application is not disapproved within thirty days, it is deemed to be approved. [1] Minor variations are, thus, roughly equivalent to changes being effected supplements in the United States in that manufacturers may implement the changes relatively quickly. However, minor variations offer companies greater certainty since they will be approved within thirty days one way or another while the FDA can always disapprove a changes being effected supplement up to six months (or even later) after it is filed. [1]
In their respective first annexes, Commission Regulations 542/95 and 541/95 set out the list of possible minor variations along with the conditions that the marketing authorization holders must demonstrate to win acceptance of the variation. Changes in the manufacturer(s) of active substances are minor variations provided that “the specifications, synthetic route and quality control procedures are the same as those already approved or a European Pharmacopoeia Certificate of suitability covering the active substance is submitted.” [1] , Minor changes of the manufacturing processes of active substances are minor variations if the substance’s “[s]pecifications are not adversely affected; [and there is] no change in the physical properties [and] no impurities or change in level of impurities which would require further qualifications in safety studies.” [1] An alteration of the batch size of the active substance is a minor variation so long as “the change does not affect consistency of production ... or physical properties.”[1] A change in the specifications of the active substance or of the medicinal product that tightens the specifications or adds new tests or limits is a minor variation. [1] Changes in the manufacture of the medicinal product are minor variations when “[m]edicinal product specifications are not adversely affected; [and] the new process ... lead[s] to an identical product regarding all aspects of quality, safety and efficacy.”[1] Changes in the batch size of the finished product are minor variations provided they “do[] not affect consistency of production.” [1] Changes in the test procedures of the active substance or the medicinal product are minor variations if they do not reduce the potency of the test or (for medicinal products) adversely affect the product’s specifications. [1]
This list of minor variations includes a number of changes that would require premarketing approval of a supplement in the United States. For example, consider a change in the basic type of manufacturing process of the medicinal product that does not adversely affect the product’s specifications or produce some difference in the drug relevant to quality, safety, or efficacy. It would be a Level 3 manufacturing change under the SUPAC Guidance, requiring premarketing approval of a supplement, in the United States. [1] In Europe it would be a minor variation and would be permitted unless disapproved within thirty days. [1] Likewise, changes in test procedures must provide enhanced accuracy to qualify for a changes being effected supplement in the United States but must merely be equivalent to be a minor variation in the EU. [1] Indeed, in the 1985 Rewrite, the FDA specifically rejected equivalence as the touchstone for not requiring premarketing approval of the supplement: Changes that
giv[e] increased assurance that the drug will possess its represented characteristics ... can be made without prior approval because, by assuring to a greater degree that the drug will possess its represented characteristics, the change provides a public health benefit. A change that provides only the same level of assurance, however, does not provide such a benefit and, thus, the agency finds no basis for
allowing manufacturers to report such alterations in a changes being effected supplement.[1]
Thus, the EU’s procedures provide companies with more flexibility than the FDA allows. This conclusion leads to three key questions. First, is more flexibility desirable? Second, will current legislative proposals provide this flexibility? Third, is the FDA institutionally capable of meeting such a command?
CURRENT PROPOSALS TO AMEND THE AMERICAN REGULATIONS
In some cases, the FDA’s rules governing manufacturing changes probably should be loosened. [1] For instance, manufacturers should only have to notice changes in test procedures that provide equivalent or enhanced accuracy in their annual reports rather than in premarketing and changes being effected supplements, respectively. If companies can deliver the same (let alone better) levels of quality assurance with less expensive methods, we should encourage, not discourage, them from doing so. Likewise, the FDA should probably adopt a version of the EU’s minor variation procedure for reviewing changes in the basic types of drug manufacturing processes which do not adversely affect the products’ specifications or produce some difference in the drugs’ identity, strength, quality, purity or potency relating to the medicines’ safety or efficacy. While some form of premarketing approval seems appropriate given a wholesale revamping of the process, full-scale reviews would probably not deliver benefits commensurate with their costs since the products (by definition) remain within specifications and do not change materially.
Last summer, FDA, PhRMA, and congressional negotiators reached a tentative “meeting of the minds” on new legislative language on manufacturing changes. [1] This agreement would preserve the three tiers but alter the middle one (changes being effected supplements) to require the applicant to wait for thirty days before distributing the drug prior to approval. [1] It would permit annual reporting of changes “of a type determined by the Secretary to have minimal potential to adversely affect the identity, strength, quality, purity, or potency as they may relate to the safety or effectiveness of a drug.” [1] Such changes would include “add[ing] a new specification(s) or test method(s) or control(s) to provide increased assurance that the drug will have the [represented or purported] characteristics of identify (sic), strength, quality, and purity” [1] — improvements that now require a changes being effected supplement. [1] The agreement would retain a premarketing approval supplement for changes “of a type determined by the Secretary to have a substantial potential to adversely affect the identity, strength, quality, purity, or potency as they may relate to the safety or effectiveness of a drug.” [1] Such
[m]ajor manufacturing changes are (i) changes in the qualitative or quantitative formulation or the specifications in the approved marketing application (unless exempted by the Secretary); (ii) changes which the Secretary determines by regulation or guidance require completion of an appropriate human study demonstrating equivalence to the drug manufactured before such changes; and (iii) other changes which the Secretary determines by regulation or guidance have a substantial potential to adversely affect the safety or effectiveness of the drug. [1]
All other changes would fall in the new mid-level tier, requiring a thirty day delay during consideration of the supplement before implementation of the change unless the change falls in a class determined by the Secretary to be implementable at the time of submission. [1]
With the exception of permitting annual reporting of specification, testing, or control improvements, [1] this legislation will not ensure that the FDA loosens its grip on manufacturing changes. It reasonably permits the agency to define the classes of variations that “have a substantial potential to adversely affect the safety or effectiveness of the drug.” [1] Under Supreme Court precedents such definitions will be virtually unreviewable by courts, [1] so the FDA will retain the legal power to do as it chooses. Whether the FDA chooses to relax further will depend upon changes to its institutional culture. As discussed above, the agency has recently begun to shed its conservatism about risk; [1] if this trend continues, it will promote manufacturing changes more than any legislative fix.
Finally, the proposed legislation does not address manufacturers’ fears that they will implement a change under a changes being affected supplement only to have the FDA disapprove the supplement down the road. It does not follow the European approach for minor variations and deem applications not disapproved within a fixed period to be approved. [1] In any event, it is unclear that moving to the European approach would actually provide any greater security for manufacturers that their investments in change will not be wasted. Currently,
[i]f FDA refuses to approve a supplement for a change that the applicant has already placed into effect, the agency must consider all the factors surrounding its refusal to approve the supplement, including the applicant’s reasons for making the change and the alternatives available to the applicant to resolve the problem. ... [I]f circumstances warrant, [applicants] may be required to discontinue the change immediately. Nonetheless, if circumstances permit, FDA agrees that applicants should be able to correct a problem at minimal expense and without unnecessary waste.[1]
Thus, present policy is not absolutely rigid, and the European approach would not provide an absolute safe harbor since the FDA would still have the authority to withdraw approval of the NDA (as amended by the supplement) if it determines that “the methods used in, or the facilities and controls used for, the manufacture, processing, and packing of such drug are inadequate to preserve its identity, strength, quality, and purity.” [1]
It should not surprise one that enacting partial legislative changes will have no assured effect on agency behavior. The Federal Food, Drug, and Cosmetic Act is amazingly open-textured because Congress cannot and does not dare micromanage food and drug regulation. It lacks the expertise and the time to legislate in the detail necessary to force the agency to act in a certain way. Just as importantly, members of Congress know that such statutory micromanagement would deprive the FDA of the ability to react to food and drug emergencies; our public health depends on the statute’s open texture. As a result, the agency’s treatment
of manufacturing changes will depend, as it always has, on the answer to one question: Whither the chemists?
S. REP. NO. 104-284, at 39 (1996) (accompanying the Food and Drug Administration Performance and Accountability Act of 1995, S. 1477, 104th Cong. (1996) (as reported by the Senate Committee on Labor and Human Resources)).
See id.
Id.
See Telephone Interview with Matthew B. Van Hook, Assistant General Counsel, PhRMA (Jan. 10, 1997) [hereinafter Van Hook Interview]. Mr. Van Hook suggested a comparing the age of manufacturing plant and equipment in the United States with other countries as a proxy, but I was unable to locate such data in the time frame I had to write this paper.
S. REP. NO. 104-284, at 39
Id.
See John Schwartz, In Swan Song, Kessler Croons of FDA Progress , WASH. POST , Dec. 11, 1996, at A23, available in LEXIS, News Library, WPost File; John Schwartz, With Leadership in Transition, FDA to Face Many Challenges in New Congress , WASH. POST , Dec. 1, 1996, at A30, available in LEXIS, News Library, WPost File.
Although closely related and frequently treated identically in current reform proposals, the regulation of manufacturing supplements for biological products has a separate history and is, thus, outside the scope of this paper.
I attempted to confirm this hunch during a conversation with William Goodrich, the FDA’s Chief Counsel during this period, but was unable to elicit a direct answer. See Telephone Interview with William Goodrich, former Chief Counsel, FDA (Jan. 22, 1997) [hereinafter Goodrich Interview].
See Telephone Interview with William Vodra, Partner, Arnold & Porter (Jan. 9, 1997) [hereinafter Vodra Interview].
SUBCOMMITTEE ON NATURAL RESOURCES, AGRICULTURE RESEARCH AND ENVIRONMENT & SUBCOMMITTEE ON INVESTIGATIONS AND OVERSIGHT, 97TH CONG., COMMISSION ON THE FEDERAL DRUG APPROVAL PROCESS: FINAL REPORT 76-77 (Comm. Print 1982) [hereinafter MCMAHON COMMISSION REPORT ].
See Vodra Interview, supra note 10 (based on Mr. Vodra’s recollection of a conversation with William Goodrich, former Chief Counsel of the FDA). Although Mr. Vodra could not recall the name of the 1940s drug, he explained that the 1960s problem involved a generic copy of Smith, Kline & French’s chlorpromazine. Mr. Goodrich no longer remembers the incidents. See Goodrich Interview, supra note 9.
See id.
Whatever one thinks of the social value of this caution, it reflects the institutional incentives facing the FDA’s employees. The American public and the Congress have tended to perceive, and thus react, much more strongly to erroneous (in hindsight) approvals of medicines that lead to disasters than to delays in approvals that postpone or even prevent new therapies. The Congress has gone so far as to call FDA line employees at a GS-9 level to testify (i.e., be harangued by Representatives or Senators) about their mistakes. See Telephone Interview with Jerome Halperin, Executive Director, The United States Pharmacopeial Convention, Inc., and former Deputy Director, FDA Center for Drug Evaluation and Research (Jan. 15, 1997) [hereinafter Halperin Interview].
MCMAHON COMMISSION REPORT , supra note 11 at 77.
See Halperin Interview, supra note 14.
Immediate Release Solid Oral Dosage Forms; Scale-Up and Postapproval Changes: Chemistry, Manufacturing, and Controls; In Vitro Dissolution Testing; In Vivo Bioequivalence Documentation; Guidance, 60 Fed. Reg. 61638 (1995) [hereinafter SUPAC Guidance]; see Jill Wechsler, Washington Report: Tug-of-War with FDA , PHARMACEUTICAL EXECUTIVE , May 1996, at 18, 20, 22 (describing the leadership of Williams).
Halperin Interview, supra note 14.
Pub. L. No. 87-781, 76 Stat. 780 (amending the Federal Food, Drug, and Cosmetic Act, 21 U.S.C. § 301 (1994)).
28 Fed. Reg. 6375 (1963).
Id. at 6381 (codified at 21 C.F.R. § 130.9, recodified as amended at 21 C.F.R. § 314.8 by 39 Fed. Reg. 11680 (1974), recodified as amended at 21 C.F.R. § 314.70 by 50 Fed. Reg. 7452, 7498 (1985)).
Id. (codified at 21 C.F.R. § 130.9(a)).
Id.
Id. (codified at 21 C.F.R. § 130.9(c)).
Id. (codified at 21 C.F.R. § 130.9(d)).
Id.
Id.
30 Fed. Reg. 993, 993-94 (1965) (amending 21 C.F.R. § 130.9).
Id. (amending 21 C.F.R. § 130.9(d)(3)) (emphasis added).
I will refer to such supplements as “changes-being-effected supplements.”
Id. at 994 (amending 21 C.F.R. § 130.9(e)).
33 Fed. Reg. 9954, 9954-55 (1968) (amending 21 C.F.R. § 130.9).
See id. at 9955. Currently, such reports are required annually. See 21 C.F.R. § 314.81(b)(2) (1996).
33 Fed. Reg. at 9955 (amending 21 C.F.R. § 130.9(a)(5)(i)).
Id. (amending 21 C.F.R. § 130.9(a)(5)(ii)).
Id. (amending 21 C.F.R. § 130.9(a)(5)(iii)).
Id. (amending 21 C.F.R. § 130.9(a)(5)(iv)).
Id. (amending 21 C.F.R. § 130.9(a)(5)(v)).
Id. (amending 21 C.F.R. § 130.9(a)(5)(vi)).
Id. (amending 21 C.F.R. § 130.9(a)(5)(vii)).
Investigational new drug. See generally 21 C.F.R. pt. 312 (1996).
New Drug and Antibiotic Regulations, 47 Fed. Reg. 46622, 46622 (1982) (to be codified at 21 C.F.R. pt. 314) (proposed Oct. 19, 1982) [hereinafter 1982 Proposal]. Although this discussion refers to the availability of new drugs in this country, it applies similarly to the use of newer manufacturing technologies that improve quality and reduce cost.
See MCMAHON COMMISSION REPORT , supra note 11, at vi-viii.
Id. at 77. The bracketed “were” replaces the actual “may be” in order to convey the tone of the discussion more accurately.
1982 Proposal, 47 Fed. Reg. at 46622.
Id. at 46634-35.
See id. at 46634-36, 46650-51 (to be codified at 21 C.F.R. § 314.70).
See New Drug and Antibiotic Regulations 50 Fed. Reg. 7452, 7466-70, 7498-99 (1985) (codified at 21 C.F.R. § 314.70) [hereinafter 1985 Rewrite].
Id. at 7498 (codified at 21 C.F.R. § 314.70(a)).
Id. at 7499 (codified at 21 C.F.R. § 314.70(d)).
Id. at 7498 (codified at 21 C.F.R. § 314.70(b)(1)(ii)); id. at 7499 (codified at 21 C.F.R. § 314.70(b)(2)(iii)).
Id. at 7498 (codified at 21 C.F.R. § 314.70(b)(1)(iii)); id. at 7499 (codified at 21 C.F.R. § 314.70(b)(2)(iii))..
Id. at 7498 (codified at 21 C.F.R. § 314.70(b)(1)(iv)).
Id. at 7499 (codified at 21 C.F.R. § 314.70(b)(1)(v)).
Id. at 7499 (codified at 21 C.F.R. § 314.70(b)(2)(v)).
Id. at 7499 (codified at 21 C.F.R. § 314.70(b)(2)(vi)).
Id. at 7499 (codified at 21 C.F.R. § 314.70(c)(1).
Id. at 7499 (codified at 21 C.F.R. § 314.70(c)(3)).
Telephone Interview with Thomas X. White, Associate Vice President for Manufacturing and Quality Control, PhRMA (Jan. 10, 1997) [hereinafter White Interview].
See 1985 Rewrite, 50 Fed Reg. at 7467.
Vodra Interview, supra note 10.
Formally, the National Committee to Review Current Procedures for Approval of New Drugs for Cancer and AIDS.
NATIONAL COMMITTEE TO REVIEW CURRENT PROCEDURES FOR APPROVAL OF NEW DRUGS FOR CANCER AND AIDS, FINAL REPORT OF THE NATIONAL COMMITTEE TO REVIEW CURRENT PROCEDURES FOR APPROVAL OF NEW DRUGS FOR CANCER AND AIDS at viii (1990).
Id. at 12.
Id. at 12-13.
Pub. L. No. 102-571, 106 Stat. 4491 (codified at 21 U.S.C. §§ 379g-h (1994)).
S. REP. NO. 104-284 , at 39.
Cf. 21 U.S.C. §§ 379g-h (1994).
Cf. 21 U.S.C. § 379g Note (1994).
OFFICE OF PLANNING AND EVALUATION, FOOD AND DRUG ADMINISTRATION, FOURTH ANNUAL PERFORMANCE REPORT: PRESCRIPTION DRUG USER FEE ACT OF 1992: FISCAL YEAR 1996 REPORT TO CONGRESS (Dec. 1, 1996) <http://www.fda.gov/ope/96pdufa.htm>.
Id.
See White Interview, supra note 59.
PRESIDENT BILL CLINTON & VICE PRESIDENT AL GORE, NATIONAL PERFORMANCE REVIEW, REINVENTING REGULATION OF DRUGS AND MEDICAL DEVICES (1995). I will only describe the report’s discussion of drug supplements and not the parallel reforms for biological products since these have a different regulatory history which is outside the scope of this paper.
Id. at 7.
Id.
Id. at 9.
SUPAC Guidance, 60 Fed. Reg. at 61638. However, as of January 15, 1997, the agency had not published guidance for other dosage forms. On the other hand, as of May 1996, the FDA was reportedly finalizing these documents. Wechsler, supra note 17, at 22.
21 C.F.R. § 314.70(a) (1996).
See SUPAC Guidance, 60 Fed. Reg. at 61638, 61641-42. Interestingly, however, the FDA also insists that the “guidance is an informal communication under 21 CFR 10.90(b)(9) that reflects the best judgment of CDER employees at this time. It does not create or confer any rights, privileges, or benefits for or on behalf of any person, nor does it operate to bind or obligate FDA in any way. ” Id. at 61638.
However, site changes “do not include ... changes in manufacturing (including process and/or equipment),” and “[n]ew manufacturing locations [still] should have a satisfactory current good manufacturing practice (CGMP) inspection.” Id. at 61641. Of course, these two exceptions were the ones that triggered premarketing approval requirements for drug substances under the 1985 Rewrite, 50 Fed. Reg. at 7499 (codified at 21 C.F.R. § 314.70(b)(1)(v)).
SUPAC Guidance, 60 Fed. Reg. at 61641.
Id.
Id.
Id.
Id.
See supra note 54 and accompanying text.
See 1985 Rewrite, 50 Fed. Reg. at 7499 (codified at 21 C.F.R. § 314.70(b)(2)(vi)).
SUPAC Guidance, 60 Fed. Reg. at 61641.
Id.
Id. at 61642.
Id.
Id.
Id.
Id.
Id.
Id.
Wechsler, supra note 17, at 20 (quoting Gary Hollenbeck, Associate Dean of Maryland’s School of Pharmacy).
Id. at 22.
Id.
Cf. CLINTON & GORE , supra note 73, at 8 (noting that the SUPAC and other guidance documents will enable companies to “sav[e] resources that would have been spent on preparing supplemental applications” by switching to annual reporting). The comparative paperwork burden would actually depend upon the relative complexity of the notification and justification that the EU and the U.S. FDA require companies to provide in supplements/variations or annual reports. As this question is beyond the scope of this paper, I am merely pointing out that the European procedures may not be uniformly more favorable to industry.
See Council Regulation 2309/93, arts. 1-18, 1993 O.J. (L 214), available in LEXIS, Intlaw Library, ECLaw File (establishing the centralized procedure); Council Directive 93/39/EEC, 1993 O.J. (L 214), available in LEXIS, Intlaw Library, ECLaw File (establishing the current decentralized procedure); see generally Richard F. Kingham et al., The New European Medicines Agency , 49 FOOD & DRUG L.J. 301, 304-13 (1994) (describing both); Robin Madell, Tour Through a Work in Progress , PHARMACEUTICAL EXECUTIVE, Jan. 1996, at 38, 39 (briefly summarizing both processes).
See Commission Regulation 542/95, 1995 O.J. (L 55), available in LEXIS, Intlaw Library, ECLaw File (establishing the centralized procedure); Commission Regulation 541/95, 1995 O.J. (L 55), available in LEXIS, Intlaw Library, ECLaw File (establishing the decentralized procedure); DIRECTORATE-GENERAL III, EUROPEAN COMMISSION, DOC. NO. III/5371/96, DRAFT NOTICE TO APPLICANTS FOR MARKETING AUTHORISATIONS FOR MEDICINAL PRODUCTS FOR HUMAN USE IN THE EUROPEAN UNION , ch. v, available in IHS Regulatory Products, Inc., FDL/EUPL database, issue 96-12 [hereinafter DRAFT NOTICE TO APPLICANTS ], replaced by DOC. NO. III/5429/96 (which I was unable to obtain) (providing detailed guidelines for both procedures).
See Commission Regulation 542/95, art. 3, Annex II; Commission Regulation 541/95, art. 3, Annex II.
Major variations are defined as variations not listed in the annexes. See Commission Regulation 542/95, art. 3, ¶ 1(b); Commission Regulation 541/95, art. 3, ¶ 1(b).
See Commission Regulation 542/95, art. 7; Commission Regulation 541/95, art. 7.
See Commission Regulation 542/95, arts. 6-8; Commission Regulation 541/95, arts. 6-7.
Commission Regulation 542/95, art. 5; Commission Regulation 541/95, art. 5. The 30-day clock does not start running immediately. Under the centralized procedure, the EMEA Secretariat has five working days to validate the application’s completeness; only at this point does the clock start. DRAFT NOTICE TO APPLICANTS , supra note 102, ch. v, § 4.4(d). Under the decentralized procedure, the “reference” Member State (the country principally responsible for the authorization) starts the clock. Id. § 3.1.1.
See Commission Regulation 542/95, art. 5; Commission Regulation 541/95, art. 5.
Cf. supra text accompanying notes 67-72.
Commission Regulation 542/95, Annex I, ¶ 11; Commission Regulation 541/95, Annex I, ¶ 11.
Commission Regulation 542/95, Annex I, ¶ 12; Commission Regulation 541/95, Annex I, ¶ 12.
Commission Regulation 542/95, Annex I, ¶ 13; Commission Regulation 541/95, Annex I, ¶ 13.
See Commission Regulation 542/95, Annex I, ¶¶ 14, 17; Commission Regulation 541/95, Annex I, ¶¶ 14, 17.
Commission Regulation 542/95, Annex I, ¶ 15; Commission Regulation 541/95, Annex I, ¶ 15.
Commission Regulation 542/95, Annex I, ¶ 16; Commission Regulation 541/95, Annex I, ¶ 16.
See Commission Regulation 542/95, Annex I, ¶¶ 24-25; Commission Regulation 541/95, Annex I, ¶¶ 24-25. Annex I also lists further minor variations that are less relevant to this paper.
See supra note 96 and accompanying text.
See supra note 114 and accompanying text.
See supra notes 57, 116 and accompanying text.
1985 Rewrite, 50 Fed. Reg. at 7468-69.
I am deliberately making these conclusions tentative. I do not yet feel that I understand all the ramifications of making manufacturing changes, so I cannot make confident pronouncements about the appropriate level of review for any particular one.
White Interview, supra note 59. Mr. White emphasized that PhRMA did not consider this agreement to be permanently binding and that the industry would pursue further changes in the 105th Congress. See id. ; see generally Text of PhRMA Draft FDA Reform Proposal , INSIDE WASHINGTON’S FDA WEEK , Jan. 10, 1997, at S1, S8 [hereinafter PhRMA Draft ] (text of proposed sec. 19, § 505(n)).
See Amendment to H.R. 3199, 104th Cong. sec. 12, § 741(b)(2) (staff discussion draft of Jul. 10, 1996) [hereinafter Staff Discussion Draft]. The Secretary could block implementation after 30 days by informing the applicant that premarketing approval is required. See sec. 12, § 741(b)(2)(C). The more recent PhRMA proposal lengthens the waiting period to 60 days. See PhRMA Draft , sec. 19, § 505(n)(2)(C).
Staff Discussion Draft, sec. 12, § 741(b)(2)(A); see also PhRMA Draft , sec. 19, § 505(n)(2)(A) (removing the Secretary’s determination: “of a type that have minimal potential”).
Manufacturing Changes Report Language: Consensus Report Language, Sec. 604 of S. 1477, at 1, 1 (July 8, 1996 draft) (obtained from Matthew B. Van Hook, Assistant General Counsel, PhRMA).
See 21 C.F.R. § 314.70(c)(1) (1996).
Staff Discussion Draft, sec. 12, § 741(b)(2)(B); see also PhRMA Draft , sec. 19, § 505(n)(2)(B) (removing the Secretary’s determination: “of a type that have a substantial potential”).
Staff Discussion Draft, sec. 12, § 741(b)(2)(B); see also PhRMA Draft , sec. 19, § 505(n)(2)(B) (requiring determinations by regulation rather than by regulation or guidance).
See Staff Discussion Draft. sec. 12, § 741(b)(2)(C); see also PhRMA Draft , sec. 19, § 505(n)(2)(C) (requiring a 60 day delay).
See supra notes 124-25 and accompanying text.
Staff Discussion Draft, sec. 12, § 741(b)(2)(B)(iii); see also PhRMA Draft , sec. 19, § 505(n)(2)(B)(iii) (requiring determinations by regulation rather than by regulation or guidance).
See Sullivan v. Zebley , 493 U.S. 521, 528 (1990) (holding that, when Congress expressly delegates legislative authority, “our review is limited to determining whether the regulations promulgated exceeded the Secretary’s statutory authority and whether they are arbitrary and capricious” (internal quotation marks omitted)); Chevron, U.S.A., Inc. v. Natural Resources Defense Council, Inc. , 467 U.S. 837, 842-43 (1984) (holding that, in the face of statutory ambiguity, courts should defer to any “permissible construction of the statute” by the agency).
See supra text accompanying notes 16-18.
Compare Commission Regulation 542/95, art. 5, and Commission Regulation 541/95, art. 5, with Staff Discussion Draft. sec. 12, § 741(b)(2)(C), and PhRMA Draft , sec. 19, § 505(n)(2)(C). Even if the proposed legislation did follow the European approach, it is far from clear that the FDA has the institutional capacity to make meaningful decisions within one or two months. I have not investigated this question, but any legislative consideration of matching the European standards would be meaningless without an answer.
1985 Rewrite, 50 Fed. Reg. at 7470.
21 U.S.C. § 355(e).