• Decrease font size
  • Return font size to normal
  • Increase font size
U.S. Department of Health and Human Services

Science & Research

  • Print
  • Share
  • E-mail

New Drug, Antibiotic, and Biologic Drug Product Regulations: part 2

21 CFR Parts 312, 314, 511, and 514
New Drug, Antibiotic, and Biologic Drug Product Regulations
[Docket No. 82N-0394]
52 FR 8798

IND Content and Format -- General Investigational Plan (ß 312.23(a)(3))

39. Many comments opposed the proposed requirements for a general investigational plan (proposed ß 312.23(a)(4); final ß 312.23(a)(3)(iv)). Several comments suggested that the information submitted in the plan would also be available elsewhere in the IND application. On the other hand, other comments criticized the requirements for the plan as being too vague. One comment strongly disputed the need to provide the required information in the plan, arguing that the clinical development plan of a drug product is not within the realm of information needed for FDA, either to decide whether it is safe to proceed with a clinical study, or to evaluate the scientific merit of a particular clinical study. Additionally, the comment contended that the information requested for the plan is often not available at the time of a new IND submission. The comment concluded that the requirement may force sponsors to formulate plans prematurely at the time of IND submission rather than at a later stage, when sufficient data are available upon which a more concrete plan may be based.

FDA believes that many of these comments misunderstood the limited purpose of the general investigational plan, which is to give agency reviewers a very brief overview of the scale and kind of clinical studies to be conducted during the following year. This overview, which is general should be no more than two or three pages in length, will provide the necessary context for FDA reviewers to assess the sufficiency of technical information to support future studies and to provide advice and assistance to the sponsor.

FDA does not agree with those comments that suggest that the requirements for the general investigational plan are either too vaguely expressed or are redundant with respect to other requirements in the IND regulations. In general, the information submitted in the general investigational plan regarding the sponsor's short-term plans for clinical studies -- the indications to be studied, the rationale for the study, the number of subjects to be involved -- will not be available in the clinical protocols or elsewhere in the application.

FDA does view this requirement as forcing the sponsor to formulate plans prematurely. When development plans are not yet crystallized, the sponsor should simply so indicate in the appropriate place in the plan.

Finally, the agency has clarified the regulation to state that the general investigational plan is to be limited to the plans for the following year. As noted in the comments, it would be unreasonable to require a sponsor to formulate and describe its plans for a 4-and 5-year study on "day 1" of the initial trials.

40. One comment asked whether the brief description of the overall plan for investigating the drug would include plans for nonclinical investigations, or whether it would be confined to plans for clinical studies only. Another comment asked whether a sponsor would be required to adhere to the general investigational plan, or would be permitted to make adjustments during the course of the investigation.

The general investigational plan is intended to be limited to plans for clinical studies in the coming year. It is not the appropriate place to discuss plans for animal or other nonclinical tests.

FDA neither insists that a sponsor adhere to the general investigational plan nor does it necessarily require that the sponsor inform FDA of a deviation at the time the deviation is made. The sponsor is free to make changes in the plan during the course of the year as the need may arise, subject to the reporting requirements for protocol amendments and information amendments (ß ß 312.30 and 312.31).

41. One comment recommended that the reference in the general investigational plan (proposed ß 312.23(a)(4)(vi); final ß 312.23(a)(3)(iv)(f)) to "special risks anticipated" should be made consistent with similar references with respect to information in the investigator's brochure (proposed ß 312.23(a)(5)(v)) and information the sponsor is required to submit with respect to previous experience with the drug (proposed ß 312.23(a)(9)(i)). The comments suggested that all three sections use the wording of proposed ß 312.23(a)(9)(i): "Information that is relevant to the safety of the proposed investigation." Alternatively, the comment suggested that the three sections incorporate the wording in the current IND regulations: "All relevant hazards, contraindications, side effects, and precautions suggested by prior experiences."

Although FDA favors consistency whenever appropriate, the comment erroneously assumes that the information to be submitted in the three sections would be identical. In fact, each of the sections calls for somewhat different information, and different requirements are therefore warranted.

The distinction between the "special risks" section of the general investigational plan and the "possible risks" provision in the investigator brochure is primarily one of scope and detail. Although both sections should contain safety information that may be relevant to precautions and special monitoring to be done during the clinical investigation, the agency expects the general investigational plan to be a more selective document than the investigator brochure. Accordingly, FDA believes that the "special risks" section of the general investigational plan should be limited to those risks that most concern the sponsor -- the most serious and significant risks that can be anticipated on the basis of previous experience. FDA has revised the final rule to reflect this distinction.

Finally, the information to be reported in ß 312.23(a)(9)(i) is limited to previous human experience with the investigational drug, in contrast to the information expected in the general investigational plan and the investigator brochure, both of which should include animal test data as well.

Protocols (ß 312.23(a)(6))

42. This section would require a protocol for each planned study. Two comments asked whether "planned study" meant a study definitely planned, or a study to be conducted in the future if the investigation followed the desired course. One of these comments noted that protocols may not yet have been completed for some studies to be conducted at later stages of the investigation.

As noted above, the sponsor may limit the IND submission to the study or studies to be conducted at the end of the 30-day review period, or may also include some or all of the studies to be conducted subsequently. To the extent that protocols for future studies have not yet been developed, the sponsor is under no obligation to submit them in the initial submission.

43. Several comments criticized the provision in proposed ß 312.23(a)(6)(i), which requests that protocols for Phase 2 and Phase 3 investigations "be designed in such a way that, if the sponsor anticipates that some deviation from the study design may become necessary as the investigation progresses, alternatives or contingencies to provide for such deviation are built into the protocols at the outset." One comment contended that in some cases it may not be possible to anticipate deviation at all, or it may not be possible to anticipate deviations in sufficient detail to provide for an alternative course of conduct. Comments suggested that inclusion of such contingency plans should be at the discretion of the sponsor and that such information should only be required "where feasible."

The final regulation, like the proposal, puts the inclusion in the protocol of contingency plans at the sponsor's discretion. Nevertheless, the agency strongly encourages submission of such plans as it believes there is much to be gained in thinking about and planning for possible alternative courses of action early in the protocol development process. Providing in the initial protocol for possible departures from the study design enhance the value and reviewability of study results. Such advance planning also permits both FDA and the sponsor to raise useful questions about study design and supporting information at the earliest possible time. Moreover, to the extent FDA is aware in advance of how a sponsor may need to depart from a planned protocol, misunderstandings between FDA and sponsors over such departures may be minimized.

The agency agrees with the comment that in some cases it may not be possible to anticipate the need to depart from the planned protocol and, in such cases, the sponsor would not be expected to submit plans for alternative or contingent courses of action.

44. Several comments objected to the requirement that the sponsor submit a curriculum vitae for each investigator. One comment suggested that instead of the curriculum vitae, which can extend to 30 or 40 pages, a sponsor should be able to submit a shorter data sheet on each investigator.

Under section 505(i) of the act (21 U.S.C. 355(i)), the agency is required to assure that the investigational drug will be provided to "experts qualified by training and experience to investigate" a new drug. To discharge that responsibility, FDA must have sufficient information about an investigator to show that he or she is qualified by reason of training and experience to conduct the proposed study. While this information is ordinarily most conveniently provided through a conventional curriculum vitae, the agency will accept any other statement of qualification that demonstrates the investigator's fitness to conduct the study. FDA has revised the final regulation accordingly.

45. Several comments contended that the names of each investigator, subinvestigator, and IRB should not be included in the protocol for the investigation, but should be included as a separate part of the study documentation. One comment claimed that when multicenter studies are conducted, it is more efficient for all investigators to conduct their studies using a master protocol that is individualized only for investigator name and address. The comment observed that, in multicenter trials, investigators are frequently added or changed during the course of the study.

To promote efficient review of an IND, all information pertaining to the protocol, including the names and qualifications of the investigators and identification of participating IRB's, should be presented together. However, whether the information pertaining to the investigators and IRB's is part of the protocol itself, or is an addendum to the protocol or accompanying document, is a matter on which the sponsor may use its discretion. When this issue arises, FDA will be willing to discuss such alternative ways of presenting the information.

When a multicenter study is conducted under a single "master" protocol, the sponsor is not required to resubmit the protocol for every new investigator added, but under ß 312.30(d) may simply reference the protocol in an appropriate protocol amendment submission containing information on the new investigator, subinvestigators, or IRB.

46. Another comment suggested that information in the protocol on investigator qualifications redundantly repeated information on investigators provided in the investigator statement (Form FDA-1572).

While it is true that the investigator statement, including information on the investigator's qualifications, is provided by the investigator to the sponsor of the clinical investigation, the sponsor is not required to submit that statement to FDA. Therefore, it cannot take the place of information contained in the sponsor's submission to the agency. At the same time, FDA acknowledges that in the past some sponsors have submitted information to FDA on their investigators by simply attaching copies of the investigator statements from the investigators. The agency believes that such a practice is appropriate provided the investigator statements submitted by the sponsor contain sufficient information to demonstrate the investigators' qualifications to undertake the proposed studies.

47. One comment asked that the term "subinvestigator" be defined. Specifically, the comment questioned whether the term included nonphysicians, nurses, technicians, and other assistance to the clinical investigator.

Studies frequently are conducted by a team of individuals who share responsibility for designing and conducting the investigation. The principal investigator is the responsible leader of that team. Subinvestigators include all other professionals who assist the principal investigator in the design and conduct of the investigation. Subinvestigators would not include those technicians and other assistance who assume no responsibility for the conduct of the study. FDA has revised the rule to reflect this concept of "subinvestigator."

48. Several comments objected to requiring the sponsor to identify the reviewing IRB for each participating investigator. One comment argued that information on IRB's may not be available at the time that an IND is filed. Another comment argued that it is inappropriate and impracticable to include the name and address of each reviewing IRB, contending that normally the investigator is the IRB contact. The comment asked whether, in requiring the identification of the IRB in the protocol, FDA intended that IRB approval be obtained before the pertinent protocol is submitted to FDA. Several comments concluded that FDA can always obtain the identification of IRB's if a need exists, but that such information should not be part of the protocol or sponsor's responsibility.

This requirement is based on FDA's regulatory responsibility to ensure that the safety, rights, and welfare of human test subjects are adequately protected. To carry out this responsibility, FDA conducts on-site inspections of both clinical investigators and IRB's. By identifying the reviewing IRB in the protocol submission, FDA is assured of having an up-to-date record of active IRB's, together with studies under their purview. FDA believes that requiring sponsors to include this information in their submissions constitutes a minimal burden and will substantially aid the agency in carrying out its mandate to monitor subject safety.

As noted in response to paragraph 67, the final rule requires that IRB approval precede the start of a clinical study but does not require that IRB approval be obtained before the IND is submitted to the agency. If information on the IRB is not available at the time the protocol is submitted, the sponsor may later add the information to the protocol through a protocol amendment.

49. One comment suggested that the protocol provisions be revised to include a requirement that the sponsor state the criteria by which effectiveness of the investigational drug will be judged. Another comment argued that the protocol should include a proposed method of analysis of results of the study.

The protocal section lists the essential elements that protocols for all studies possess in common. As not every protocol contemplates a specific method of analyzing study results or is intended to examine a drug's effectiveness, it would not be appropriate to list them in this section. The essential elements of a protocol for a study intended to demonstrate effectiveness are described in the regulation outlining the characteristics of an adequate and well-controlled investigation (21 CFR 314.126).

Chemistry, Manufacturing, and Control Information (ß 312.23(a)(7))

50. A comment agreed with FDA that the amount of chemistry, manufacturing, and control information should be less in the clinical pharmacology stage (Phase 1) than in later stages of drug development, but suggested that the proposed chemistry requirements for Phase 1 would still require more information than is necessary to assure subject safety in early research. Specifically, the comment urged that, rather than provide information on the "general method of preparation of the drug substance" for Phase 1 studies, sponsors should only be required to provide a brief outline in the form of a schematic diagram outlining the manufacturing process. Additionally, the comment recommended that sponsors should not be required during Phase 1 studies to provide detailed information on raw materials used in investigational products.

FDA does not agree that it is asking for more information than is actually needed to assure human subject safety in Phase 1 studies. In general, a schematic diagram of the process by which the drug substance is synthesized, while a useful symbolic representation of the method of drug synthesis, will not provide adequate information about the manufacturing process -- including, for example, information on equipment used, work-up and isolation procedures, purification steps, tests for completion of reaction and yields -- to permit FDA to make a number of key safety determinations, including determinations about the presence of contaminants and byproducts.

51. Several comments urged that the proposal be revised to indicate that complete stability data are not required prior to beginning clinical studies. These comments urged that the regulations permit the development of stability data concurrently with the conduct of the clinical investigations. One comment argued that in the closely controlled distribution system that is required for investigational drug accountability, corrective action for materials that no longer meet the appropriate standards for use is easily undertaken. The comment contended that the concurrent development of stability data is consistent with current good manufacturing practice. Two comments suggested that if data developed concurrently indicate that a drug product does not meet its acceptance standards during the entire period of the investigation, appropriate corrective action can easily be undertaken to replace the material. Several comments maintained that permitting concurrent stability testing would further the regulatory objective to speed up the drug testing and approval process.

The regulation does not preclude a sponsor from conducting stability tests on an investigational drug product concurrently with clinical investigations of the product. However, the agency does expect that, by the time a clinical study is begun, the sponsor will have submitted to FDA at least preliminary evidence (obtained from accelerated studies) to show that the product is likely to remain stable for the duration of the study. The applicable requirements for stability testing are set forth in 21 CFR 211.166 of the regulations describing current good manufacturing practice for finished pharmaceuticals. At the same time, sponsors should be aware that a decision not to complete stability tests before commencing a clinical study may jeopardize the value of study results should the tests ultimately show problems in the drug product's degradation or bioavailability.

While the regulations thus permit concurrent testing of investigational drug products, the agency believes that testing of the stability of the drug substance should be substantially completed before initiation of human clinical studies of the drug. This should not present significant difficulties to sponsors, as these tests are usually conducted while preclinical animal studies of pharmacology and toxicology are underway.

52. The proposed rule indicated that, as drug development proceeds, and as the scale of production of the investigational drug is changed from the limited pilot production appropriate for the initial clinical studies to larger scale production necessary for expanded clinical investigations, the sponsor should submit information amendments to supplement the initial information submitted on the manufacturing and control processes. One comment argued that information amendments should only be required if the manufacturing formula changes, not every time the scale of production changes, since the scale does not change the compositional formula of the clinical supplies.

FDA does not agree. Although it is true that changes in scale may not affect a drug's composition, such changes may affect a drug's physical or biochemical characteristics and thus possibly affect the safety of proposed studies. Specifically, changes in scale may involve use of new kinds of production equipment or use of the same equipment in different ways to accommodate larger batch processing. These changes may significantly affect important chemical and physical properties of the drug, including the drug's content uniformity, hardness, moisture content, and dissolution. Utimately these sorts of changes can affect a drug's bioavailability and be of clinical significance.

53. One comment recommended that sponsors be required to provide information on the composition, manufacture, and control of any placebo used in a controlled clinical trial, including information demonstrating that the placebo is identical to the drug under study in all respects other than the presence of the active drug substance. The comment contended that the validity of a blinded study depends in part on the placebo being perceived as identical to the drug under study.

FDA agrees that information on placebos is needed to assure that the blinded nature of a study is not compromised by the failure of a placebo to mimic the odor, taste, texture, and other physical characteristics of the investigational drug. FDA has requested such information for a number of years.

In response to this comment, the agency has revised the final rule to require a brief, general description of the composition, manufacture, and control of any placebo used in a controlled clinical trial. The agency, however, is not requiring a demonstration that, but for the presence of the active drug substance, the placebo is "identical in all respects" to the drug under study. This is because exact duplication of the investigational drug may not be possible. For example, the use of a coloring agent or an inactive bitter flavoring may be required to mimic characteristics of the drug substance so that the placebo will be perceived as identical to the drug under study.

54. As proposed, the rule would permit reference to the United States Pharmacopeia -- National Formulary to satisfy relevant portions of the chemistry section. One comment noted that compendial requirements may in some cases not meet FDA's requirements. The comment urged that the rule make clear that in some circumstances reference to the formularies may not satisfy relevant requirements of the chemistry provisions.

As noted in the preamble to the NDA Rewrite final rule (50 FR 7459), although the agency believes that references to the official compendia may be relied on under proper circumstances to provide the required information, new developments in drug synthesis and advances in analytical technology may introduce new concerns about the chemistry of drug substances that are not adequately addressed by current compendial monographs. In those cases, FDA may need additional information about a drug substance to ensure that additives or byproducts of the synthetic process are properly controlled. Although a reference to official compendia will often satisfy the requirements, the final rule has been revised to indicate that FDA may require additional information to permit proper review of the application.

55. One comment claimed that the information on manufacturing facilities submitted in the IND in accordance with proposed ß 312.23(a)(6) would be inadequate to determine whether the applicable IND termination provisions should be invoked, i.e., whether the facilities used for the manufacturing, processing, and packing of the investigational drug are adequate to establish and maintain appropriate standards of identity, strength, quality, and purity as needed for subject safety.

FDA believes that the information required in ß 312.23(a)(6) should ordinarily be adequate to determine whether the drug's manufacture and control may compromise subject safety. The required information includes descriptions both of the general method of preparation of the drug substance (ß 312.23(a)(7)(iv)(a)) and of the method of manufacturing and packaging of the drug product (ß 312.23(a)(7)(iv)(b)). If additional information is needed on the manufacture and control of the drug, FDA can either request the sponsor to submit the information, or under certain circumstances, can inspect the manufacturing site to determine compliance with applicable current good manufacturing practice (21 CFR Part 211).

56. One comment suggested using the word "strength" instead of "potency" in ß 312.23(a)(7)(iv)(a) to be consistent with the language of 21 CFR Part 211.

FDA agrees and has revised the final rule accordingly.

Pharmacology and Toxicology Information (ß 312.23(a)(8))

57. The proposed pharmacology and drug disposition section would require information describing the pharmacological effects and mechanisms of action of the drug in animals and information on the absorption, distribution, metabolism, and excretion of the drug. One comment asked whether the information on the absorption, distribution, metabolism, and excretion of the drug required under proposed ß 312.23(a)(8)(i) should, like the information on pharmacological effects, also be based on animal studies.

The pharmacology and toxicology section (ß 312.23(a)(8)) refers principally to data derived from animal studies, but could include human data for comparison, if available. Therefore, FDA expects that any information in the initial IND submission on the absorption, distribution, metabolism, and excretion of the drug will be derived from animal studies of the drug. As information is obtained on the pharmacokinetics of the drug in humans, the agency would expect such information to be reflected in the investigator brochure (ß 312.23(a)(5) (ii) and (iii)) and reported, as appropriate, in information amendments and annual reports.

58. Proposed ß 312.23(a)(8)(ii)(b) would require the submission of full tabulations of data suitable for detailed review for each toxicology study that is intended primarily to support the safety of the proposed clinical investigation. Many comments objected to the requirement that full toxicological data be submitted. Several comments contended that this requirement is inconsistent with the principle that data in the IND should normally be submitted in summary form. Other comments expressed the belief that full data are not necessary in order to achieve the objective of assuring subject safety, arguing that the safety of subjects can be adequately protected by requiring submission of a summary in sufficient detail to permit scientific review, and allowing FDA access to full data when necessary.

FDA believes that, as a general rule, full tabulations of data from subacute and chronic studies and other studies intended primarily to support safety are crucial in permitting their scientific review. The agency has found that summaries, by heavy reliance on statistical averaging of data, may not reveal the actual magnitude of response in some animals and do not provide for comparing the spectrum of responses in any one animal. Summaries, while helpful adjuncts to the pharmacological and toxicological review process, cannot substitute for full tabulations in providing adequate insight into the extent and course of drug effects in individual test animals.

Because FDA believes that full tabulations of such data are necessary in every case for an adequate review of the safety of proposed studies, it sees no merit in a procedure that would permit reviewers to obtain detailed information only on request. Such a procedure would likely only serve to delay the review process.

59. One comment asked for clarification of the meaning of "full tabulations" of toxicological data. The comment stated that it would object to a requirement that every data point collected in a study be tabulated, suggesting that such a requirement would create an unwarranted burden.

This section is intended to continue current practices with respect to submission of individual animal data. Thus, applicants are not required to submit laboratory notebooks, worksheets, and other documents relating to individual animals. However, the agency does expect the full tabulations to include every significant recorded observation, pathology finding, and laboratory measurement that relates to a scientific evaluation of the drug's safety for its proposed investigational use. This would ordinarily include all notable periodically measured toxic signs, as well as blood values, electrocardiograms, and any other measurements or observations that would contribute to the evaluation of the drug's toxic potential.

60. Several comments urged the adoption of specific pharmacological and toxicological testing requirements. Thus, for example, one comment urged that the regulation clearly require sponsors to test drugs that have not previously been tested in human subjects in at least two animal species prior to the commencement of clinical trials.

As noted above, the regulation is intended to describe in general terms an appropriate format and content for the initial IND submission. Because of the dynamism and complexity of the scientific issues involved, the agency does not believe that it would be either feasible or wise to specify in the regulation detailed, substantive pharmacology and toxicology testing requirements. Current FDA guidelines do specify the type of animal studies needed for new drug substances before commencement of human studies, and the agency is developing, and will soon make publicly available, an updated guideline that will outline the scope of animal testing submissions for the more common and expected circumstances.

Previous Human Experience with the Investigational Drug (ß 312.23(a)(9))

61. Section 312.23(a)(9)(iii) requires the sponsor to list the foreign countries in which the investigational drug has been marketed as well as those countries in which the drug has been withdrawn from the market for any reason relating to safety or effectiveness. One comment urged that this responsibility be limited to experience with the sponsor's own drug as "it may not be feasible for a sponsor to be fully aware of all actions taken by all firms worldwide."

FDA believes that it is not unreasonable to expect a commercial drug firm to make a good faith effort to determine the foreign marketing experience of a drug it seeks to market in the United States. Given the potential hazardous consequences that may follow from the use of unsafe or ineffective drugs, FDA would expect commercial sponsors to obtain the information for their own benefit, apart from regulatory requirements. Much of this information should already be available to the sponsor as a result of patent searches or other routine business practices. Because additional information on foreign marketing is readily obtainable through trade journals available in the United States, a comprehensive review of the pertinent information should not be unduly burdensome to the sponsor.

62. Proposed ß 312.23(a)(9)(i) would require that any published material relevant to an assessment of the drug's safety and effectiveness be provided in full. One comment claimed that this requirement is inconsistent with the general principle that the sponsor should provide a summary of previous human experience. The comment argued that it would be possible to provide relevant information on a number of similar studies in a single narrative summary and that such a summary of the available literature would provide all the information the agency would need. One comment claimed, moreover, that a requirement for all literature could result in voluminous submissions under certain circumstances, especially if a sponsor were testing a combination product in which one component is a well-established drug.

The agency believes that some of these comments may have misinterpreted the proposed provision, the purpose of which is to give agency reviewers easy access to those reports in the scientific literature that are most directly relevant to the safety and effectiveness of the drug for its proposed use. Reports of greatest relevance would include, for example, reports of the most serious or frequent drug-associated adverse reactions, reports of critical dose-response information, as well as reports of the results of controlled clinical trials. Publications from the scientific literature less directly relevant or exclusively relevant to other indications for use need not be submitted, although they should be included in the sponsor's bibliography. Thus, for example, a sponsor studying aspirin to reduce the risk of stroke would not be expected to submit to FDA studies relevant only to the drug's analgesic effects.

If a sponsor were testing a combination drug in which one of the components had already been lawfully marketed in the United States, the sponsor would not need to submit all the literature on the component's marketed use, but only publications of direct relevance to the proposed use (including publications relevant to component-component interactions). FDA has revised the final rule in ß 312.23(a)(9)(ii) to make this requirement explicit.

For the reasons given, the agency does not believe that the provision should ordinarily be unduly burdensome or result in the submission of excessive numbers of publications from the scientific literature. Of course, if a sponsor is concerned about the extent of published literature to be submitted in a particular instance, the agency would be willing to discuss the issue with the sponsor in advance of the submission.

63. One comment stated that providing information for each component of a combination product, the components of which have been previously investigated or marketed, is reasonable only if the requirement is understood to relate to the active drug components.

FDA agrees and has revised ß 312.23(a)(9)(ii) accordingly.

Drug Dependence and Abuse Potential (ß 312.23(a)(10)(i))

64. Proposed ß 312.23(a)(10) would require a sponsor of a drug with abuse potential to provide a description of "relevant clinical studies and experience and studies in test animals." One comment asked that this section be clarified to require that such information be supplied only if it is available and only for the later phases of a clinical investigation.

The comment misunderstands the intended function of this section, which is simply to establish a place in the IND for a sponsor to compile and present available information on the dependence or abuse potential of its drug. The provision does not establish substantive requirements with respect to clinical studies. Guidance on these substantive matters can be obtained from the published clinical guidelines issued by FDA and from the agency's scientific review divisions.

Material in a Foreign Language (ß 312.23(c))

65. One comment objected to the requirement that the sponsor submit a copy of each original literature publication for which an English translation is also submitted. The comment claimed that this requirement is of questionable value and is inconsistent with the principles of the Paperwork Reduction Act. The comment suggested deleting the requirement for routine submission and replacing it with a requirement that foreign language materials be made available to FDA on request.

FDA believes that it is reasonable to ask an applicant who relies upon a publication in a foreign language to submit both the foreign publication and an English translation of it. FDA believes it is under some obligation to verify the bases of documents it receives only in translation, and views sponsors' furnishing to FDA of the non-English original as the least burdensome method by which verification can be accomplished.

Protocol Amendments (ß 312.30)

66. One comment suggested that, to speed early clinical research, sponsors should not have to submit protocol amendments: (1) For modifications of a clinical pharmacology research protocol made on the basis of experience gained in the investigation; (2) for continuation of a human subject from Phase 1 to the subsequent phases of the investigation; or (3) in situations where the investigator concludes that immediate action is necessary to reduce or eliminate an apparent immediate hazard to a subject.

The final rule, like the proposal, does not require a sponsor to submit a protocol amendment for a change in a Phase I protocol that may affect the scope or scientific quality of an investigation, if it does not significantly affect the safety of subjects. Therefore, to the extent that a modification to a clinical pharmacology (Phase 1) protocol does not raise significant safety issues, it would not have to be reported in a protocol amendment. In addition, the protocol amendment provisions do not require a sponsor to file an amendment to continue a subject from one phase of the study to the next, assuming a protocol is in effect for the subsequent phase that covers administration of the investigational drug to that subject.

Finally, as noted in response to paragraph 69 below, FDA has revised the final rule to state that a sponsor may change a protocol to eliminate an apparent immediate hazard to a subject, provided FDA is subsequently notified of the action. The agency believes this clarification of the requirement meets the concerns of the comment.

67. As proposed, protocol changes that would require a protocol amendment under ß 312.30(b) may only be implemented after the sponsor has submitted "the amendment to the IND following completion of review of the change by the IRB that is responsible for review and approval of the study." Several comments read this requirement as obliging the sponsor to ensure that an IRB reviewed and approved the change before submitting it in a protocol amendment to FDA.

FDA has revised ß 312.30(b) to clarify that IRB review and approval may be obtained before or after submission of the protocol amendment to FDA, provided the sponsor and investigator ensure that the change that is the subject of the amendment is not begun until IRB review and approval has been obtained.

68. One comment argued that the rationale for requiring submission of a protocol amendment to report the addition of a new test or procedure to monitor for side effects or adverse events is unclear, since any effect of such action would be a positive one, increasing the safety precautions afforded the subject. The comment suggested that comparable considerations dictated that changes to enhance the scientific quality of studies should also not require a protocol amendment.

The purpose of a protocol amendment is to give the agency timely notice concerning the kinds of changes that bear directly on its review and monitoring responsibilities. The agency believes it is responsible for making an independent review of significant protocol changes even when their intended effect is to increase the safety of subjects. In this context, it should be noted that submission of a protocol amendment to FDA does not delay implementation of the change described in the amendment.

69. One comment noted that, for protocol changes designed to reduce the risks of injury, any delay in undertaking the change caused by the need to submit the change to FDA or to obtain IRB approval might jeopardize subject safety. The comments suggested that prior notification to FDA and prior approval by IRB's not be required for changes designed to eliminate hazards to study subjects.

The agency agrees that a protocol change intended to eliminate an apparent immediate hazard to human subjects should not be delayed because of a need to notify FDA or the reviewing IRB. FDA has revised final ß 312.30(b)(2)(ii) to permit such changes, provided FDA and the reviewing IRB are subsequently notified.

70. One comment asked the agency to clarify a sponsor's responsibilities with respect to protocol changes that would not require submission of a protocol amendment, including, for example, a Phase 1 change that does not significantly affect subject safety.

The sponsor's responsibility depends on the nature of the change. Changes that are not required to be reported in a protocol amendment may still be reportable under another section of these regulations, or under the regulations governing review of marketing applications (Part 314). Thus, for example, a change in the scope of a Phase 1 investigation may not require a protocol amendment but should be reported in the next annual report in accordance with ß 312.33(b). Other changes -- minor modifications of a study design, for example -- may not be reportable until the study is submitted in a marketing application, where it would be reported as part of the application under ß 314.50(d)(5).

Finally, it should be noted that investigators may be required under ß 56.109 to report to reviewing IRB's some protocol changes that are made during Phase 1 even though such changes need not be reported to FDA under these protocol amendment requirements.

71. Several comments expressed support for the provision in ß 312.30(c) that would require a sponsor to notify FDA within 30 days of adding an investigator, but asked that the final rule make clear that a sponsor may ship an investigational drug to a new investigator at the time that the investigator is added by the sponsor to the study, and that the newly added investigator may begin his or her participation in the study prior to submission of the protocol amendment, so long as the amendment is submitted within 30 days of the commencement of the investigator's participation.

FDA has revised ß 312.30(c) to make clear that, once the sponsor has added an investigator to a previously submitted study, the investigator may begin participation in the study. Notification to FDA is required within the next 30 days.

72. Several comments read ß 312.30(d)(1)(i) as requiring a sponsor to describe in detail the differences between the new and the old protocols. The comments claimed that this provision would impose significant burdens without corresponding benefits. One comment claimed there are many cases where detailed explanations would not be needed. Finally, one comment suggested that, if a requirement to explain protocol changes is retained at all, a sponsor should only be required to highlight significant changes from previous protocols.

Proposed ß 312.30(d)(1)(i) was not intended to require detailed explanations of the differences between old and new protocols. In fact, a detailed and undiscriminating enumeration of the differences would defeat the purpose of this requirement, which is to identify the most important differences between the old and new protocols and to alert FDA reviewers to major changes that may require additional supportive data, such as changes in dose, route of administration, or indication. What is expected is that the sponsor will briefly highlight the most clinically significant features of the new protocol, such as an increase in dose or duration of treatment, or a change in patient population. To clarify agency intent, FDA has revised ß 312.30(d)(1)(i) to require, for each new protocol, a brief description of the most clinically significant differences between it and previous protocols. As so modified, the highlighting of the changes should not be a significant burden on sponsors, and will be of considerable help to FDA in directing reviewers' attention to the parts of a protocol most in need of scrutiny.

73. One comment asked for clarification of a sponsor's responsibilities when a new investigator is added to conduct a previously submitted protocol. The comment stated its assumption that FDA would still want all the information that is currently required, including a copy of the protocol itself.

In this situation, the agency does not believe a copy of the previously submitted protocol is necessary, if the protocol is adequately identified in the protocol amendment. However, FDA would expect the sponsor to submit the same information about the individual investigator that would be required if the investigator had been named at the time the protocol was initially submitted. These items of information are listed in ß 312.23(a)(6)(iii)(b) and include, in addition to the investigator's name and qualifications, the name of each subinvestigator, the name and address of the research facilities, and the name and address of the reviewing IRB. FDA has revised ß 312.30(d)(1)(iii) to clarify this requirement.

74. Several comments addressed the requirement in ß 312.30(d)(2) that the sponsor reference in the protocol amendment the specific information relied upon to support the new protocol or protocol change. The comments claimed that the provision is overly broad and may be read to require extensive cross-referencing to virtually all data in support of every new protocol or protocol change. One comment urged that the section be deleted, claiming that, under most circumstances, new protocols or protocol changes rely not on specific information but rather are based upon the totality of the experience derived from earlier or ongoing clinical trials. Another comment suggested that the final rule be revised to require references only to specific information in support of significant differences in new protocols or in support of significant protocol changes.

In one sense, FDA agrees that every new protocol and protocol change relies ultimately upon the totality of previously submitted information. The intent of the provision, however, is to elicit reference to the specific technical information supporting the clinically significant aspects of the proposed change. Thus, if a sponsor intends to change the dosage form of the investigational drug, appropriate chemistry and manufacturing information supporting this change should be referenced. Or, if a sponsor proposes to increase significantly the duration of patient exposure to the drug, the sponsor should reference the appropriate animal tests that would support this increased human exposure. To the extent that FDA is apprised of the basis for a change in a protocol, it can more quickly and comprehensively review the change. Of course, if the change is one that plainly does not require specific technical support, the sponsor would not be expected to reference any supporting technical information. FDA has revised ß 312.30(d)(2) accordingly.

75. Several comments complained that the agency had not justified the use of serial numbering of protocol amendments and expressed doubt about the utility of this policy. One comment claimed that serial numbering will make tracking more difficult than under the current system, and recommended that an attempt be made to develop a numbering system that will provide for easy access to individual protocols and investigators. The comment suggested that it may be possible, for example, to use prefixes or suffixes to identify the protocol of an investigator to which a specific protocol amendment applies.

As noted in the preamble to the proposal, the formatting requirements for amendments, including the requirements for serial numbering of amendments, are intended to make these submissions easier for FDA to process and review. Deficiencies in formatting have frequently produced a disorderly and sometimes unintelligible flow of amendments and other documents to the IND file. The changes are designed to rationalize this flow of information to permit agency reviewers to gain an understanding of the significance of amendments and their relationship to one another. The changes should also help reviewers determine the completeness of amendments to an IND.

While the agency does not believe that sequential numbering of amendments will somehow make tracking more difficult, FDA agrees with the comment that an identification method that separates protocol changes and new protocols from new investigators might be preferable to a simpler system. The final rule does not mandate the use of a particular method of serial numbering, and so individual sponsors are not precluded from adopting a more complex system. FDA would be happy to work with sponsors in developing a system of maximum usefulness.

76. Section 312.30(d)(3) specifies that if a sponsor desires FDA to comment on a protocol amendment submission, the protocol amendment should so indicate and should include the specific questions FDA should address. One comment suggested that FDA should be obliged to respond within 15 days to a sponsor's request to avoid impeding the progress of the investigation and to avoid the imposition of clinical holds due to deficiencies in a proposed protocol.

Because protocol amendments do not require prior agency approval before implementation, the lack of an agency response should not, in itself, impede the progress of an investigation. Nevertheless, FDA understands the importance of conscientiously reviewing and responding to sponsor requests for assistance, and will respond as quickly as is allowed by the complexity of the questions, the availability of agency reviewers, and the demands of other priority matters.

Information Amendments (ß 312.31)

77. One comment recommended that proposed ß 312.31(b) be revised to define what discipline categories should be used when information amendments are "numbered serially" by discipline.

FDA has revised ß 312.31(b) to add examples of appropriate headings for information amendments.

IND Safety Reports (ß 312.32)

What is Reportable? 

78. A number of comments expressed confusion about what would be reportable in an IND safety report. As proposed, ß 312.32 would require a sponsor to report "Any serious adverse experiences or other information * * * not previously reported (in nature, severity, or incidence) that may suggest significant hazards, contraindications, side-effects, or precautions." One comment asked that the agency define the term "serious adverse experience." Other comments asked that the agency clarify the meaning of "not previously reported (in nature, severity, or incidence)." One comment suggested that the parenthetical phrase was unnecessary and should be deleted. Another comment argued that the severity or incidence of an adverse experience is more appropriately the basis for adverse reaction reporting of marketed drugs than it is for investigational drugs. That comment contended that the appropriate basis for adverse reaction reporting in the IND process should be whether an event is "alarming," alarming being defined as any event requiring the discontinuation of an IND.

FDA agrees that it will be useful to provide additional guidance on what information should be reported in an IND safety report. The goal of the safety report section is to ensure timely communication of the most important new information about experiences with the investigational drug. To better achieve that goal, FDA has revised the safety report requirement to require reporting of any "adverse experience associated with the use of the drug that is both serious and unexpected." To clarify further a sponsor's reporting obligation, the agency has further revised the final rule to codify definitions of "serious adverse experience" and "unexpected adverse experience."

Serious Adverse Experience 

Under this final rule, "serious adverse experience" is defined to mean any experience that "suggests a significant hazard, contraindication, side effect, or precaution." The definition distinguishes human (clinical) experiences from drug-related experiences in laboratory animals. "Serious adverse experience," as it applies to human experience, is defined as any experience that is fatal or life-threatening, is permanently disabling, requires inpatient hospitalization, or is a congenital anomaly, cancer, or overdose. (This definition is identical to a proposed revision of the definition of "serious" adverse drug experience for purposes of postmarketing reporting of adverse drug experiences published in the Federal Register of December 30, 1986 (51 FR 47028).) In contrast, with regard to results obtained from tests in laboratory test animals, a serious adverse drug experience includes any experience suggesting a significant risk for human subjects, including any finding of mutagenicity, teratogenicity, or carcinogenicity.

The language "suggests a significant hazard, contraindication, side effect, or precaution" is taken directly from the current IND regulations, ß 312.1(a)(6), and the IND proposal, ß 312.32(b). Thus, the underlying standard for determining what is a serious adverse drug experience has remained constant over time. The additional examples of serious human adverse drug experience are taken from the NDA Rewrite final rule, ß 314.80 (50 FR 7500), pertaining to reports for marketed drugs. These examples have been added in order to clarify the underlying standard and to provide continuity in reporting between the investigational and marketing stages.

Unexpected Adverse Experience 

In proposing to require a sponsor to report adverse experiences "not previously reported (in nature, severity, or incidence)," FDA intended to ensure the timely communication of the most important new information about the drug. However, the agency has concluded that the "not previously reported" language is unsatisfactory in that it might be read as limiting safety reports to the first case of a particular adverse experience. This was not intended. While the report of the first case of an adverse experience may indeed be the most useful one in terms of alerting FDA to a potential safety problem, reports on the first case are usually not adequate to determine whether or not an experience is truly drug related, to evaluate its likely frequency, and otherwise to assess the significance of the risk posed. To avoid any possible misinterpretation of agency intent and to ensure continued reporting of cases of a serious adverse experience until the risk posed is reasonalby well characterized and understood, the agency has deleted the "not previously reported" phrase, and substituted for it a requirement that sponsors report "unexpected" experiences.

The final rule defines an "unexpected adverse experience" to mean any adverse experience that is not identified in nature, severity, or frequency in the current investigator brochure for the study. For those IND's for which an investigator brochure is not required (i.e., IND's conducted by sponsor-investigators), an unexpected adverse experience would include any experience that is not identified in nature, severity, or frequency in the "risks" information contained in the current application. This definition is analogous to that used in the NDA Rewrite for reporting adverse drug experiences on marketed drugs, where the term "unexpected" is defined as any serious adverse experience outside of the drug's approved labeling (see ß 314.80(a); 50 FR 7500; February 22, 1985). Under the final rule, therefore, a serious adverse reaction that had been reported previously could still be unexpected.

To increase assurance that the significance of safety information will be placed in proper context, the final rule also requires that the sponsor identify all safety reports previously submitted by the sponsor concerning a similar adverse experience, and analyze the significance of the adverse experience in light of all previous similar safety reports.

Finally, FDA disagrees with those comments that suggest that it is inappropriate to base IND safety reporting on adverse experience severity or incidence. A study under an IND is allowed to proceed in part on the basis of the sponsor's characterization of the risks posed by the investigational drug. This characterization, usually described in some detail in the investigator brochure and elsewhere in the IND, is based primarily on the clinical and nonclinical experiences with the drug available at the time studies under the IND begin. As additional experience with the drug is obtained that either suggests new risks or casts previously identified risks in a new light, it is essential to FDA's safety monitoring responsibilities that the agency be promptly apprised of such information.

79. Under the final rule, a sponsor would be required to report each successive case of a serious and unexpected adverse experience until the risk posed by the experience is sufficiently well understood to be described in the investigator brochure or until an equally satisfactory resolution of the issue is reached (for example, a determination that the experience is not drug related). Ordinarily, reports of succeeding cases would, like the report of the first case, be submitted in IND safety reports as soon as possible and in no event later than 10 working days after the sponsor's initial receipt of the information. However, in some situations it may be desirable for the sponsor to "group" such cases at some different frequency, or to report such cases in a format not conventionally used for reporting a single case. Therefore, the agency has revised the final rule to authorize FDA to require a sponsor to submt safety reports in a format or at a frequency different than that normally required (ß 312.32(c)(3)). Section 312.32(c)(3) also permits the sponsor to propose and adopt an alternative reporting arrangement, if the alternative is agreed to by the director of the division of FDA's Center for Drugs and Biologics responsible for review of the IND.

Review Requirements 

80. Several comments objected to the proposed requirement in proposed ß 312.32(a) (now ß 312.32(b)) that a sponsor "immediately" review all information relevant to the safety of the drug. One comment contended that the immediate review requirement was unrealistic, given that a sponsor may receive hundreds of medical journals within a short period of time, not all of which can be reviewed "immediately," and some of which may require translation into English.

FDA expects a sponsor to review all information it receives that may be relevant to the safety of its investigational drug in sufficient time to meet its reporting obligations. However, as noted below, FDA has deleted the proposed 3-working-day time frame for written reports of fatal or life-threatening experiences and has established a uniform, 10-working-day time frame for all written reports of serious and unexpected adverse experiences. (As noted below, sponsors are still required to give FDA an "early warning" by telephone of any information obtained from the sponsor's own clinical studies suggesting an unexpected fatal or life-threatening experience no later than 3 working days after receipt of the information.) This means, in effect, that sponsors who would have had no more than 3 working days to review and report safety information under the proposal will now have up to 10 days to complete their review and submit required reports. In light of this change, the agency believes a "prompt review" requirement is a more accurate characterization of a sponsor's reporting obligation and has revised ß 312.32(b) accordingly.

81. One comment asked that FDA clarify when it would impute to a large, multi-national corporation knowledge of an adverse event gained by one of its employees. In particular, the comment wanted to know when FDA would deem a parent company to have "received" a report in a medical journal obtained by an employee of one of the parent company's subsidiaries.

FDA expects drug companies to review those reports that come to its attention in the normal course of business. Whether an employee's knowledge of a report of an adverse experience would be imputed to the sponsor will depend upon the factors surrounding the employee's knowledge of the report. As a general rule, however, FDA will consider a drug firm responsible for information known to its employees (including the employees of a division or separately incorporated subsidiary of the firm), and companies should adopt procedures to ensure that employees will expeditiously bring important information to the attention of company officials.

82. Several comments objected to the requirement that sponsors review and report in safety reports information about "related drugs." Suggesting that the term might be variously construed to include drugs with related chemical structures, drugs of the same pharmacological class, and drugs with the same intended therapeutic use, the comments criticized the term as vague and potentially subject to an overbroad interpretation. One comment complained that the provision would impose a greater reporting burden on IND sponsors than that imposed on holders of approved marketing applications.

The agency agrees that the category of "related drugs" may be overbroad, and that a requirement based on that category might well elicit much information of little relevance or value to FDA's safety evaluation of a particular investigational drug. Therefore, the agency has deleted the requirement that expressly calls for sponsors to report in safety reports information about related drugs. As revised, the regulation limits safety reports to those experiences that are associated with use of the particular investigational drug under study. This revision is not intended to suggest that safety information about related drugs is never important to evaluating the safety of an investigational drug. Indeed, a drug firm developing a new member of a structurally related class of drugs should monitor clinical reports on other members of that class. FDA's experience is that sponsors frequently do report to FDA significant and relevant safety information about such related drugs, and FDA strongly encourages continued reporting of this information to FDA in information amendments or annual reports.

Reporting Time Frames 

83. FDA received a considerable number of comments concerning the proposed time frames for reporting IND safety reports. The proposal would have required the sponsor to submit a safety report to FDA no later than 3 working days after receiving information on a fatal or life-threatening experience, and no later than 10 working days after receiving information on any other serious adverse experience. Although most comments agreed on the need for timely reporting of adverse experiences, many contended that the reporting provisions, especially the 3-working-day time frame for fatal and life-threatening experiences, would not give sponsors enough time to review and assess the significance of safety information and would result in the submission of incomplete or misleading information. To remedy these perceived problems, several comments suggested giving sponsors up to 15 days from date of receipt of the initial safety information to make a safety report. Alternatively, other comments recommended that the reporting obligation run from the time that a sponsor received the "essential information" on the experience, or from the time the sponsor determined an event was drug related, rather than from the date of receipt of the initial, perhaps fragmentary, report of the experience. Finally, several comments suggested that safety information derived from foreign experience with investigational drugs should be reported less frequently than other safety information -- one comment recommended 3-month intervals -- because of the need for translation and because of the greater problems in investigating such experiences.

FDA has carefully considered these comments and has concluded that 3 working days may not be sufficient time to determine whether a death or life-threatening experience should be reported in a written IND safety report under ß 312.32(c). Therefore, FDA has revised the final rule to require that all serious, unexpected adverse drug experiences be reported in a written IND safety report to FDA as soon as possible and in no event later than 10 working days after the sponsor's initial receipt of the information. FDA believes that this change will ensure timely communication of the most important safety information, while giving sponsors a reasonable amount of time to review incoming safety information, to identify reportable information, and to prepare and transmit to FDA complete and accurate safety reports. Although FDA has changed the proposed 3-working-day reporting time for written reports of fatal and life-threatening experiences to 10 working days in order to improve the quality of the reports received (and therefore the likelihood that FDA would have sufficient data to take action, if necessary), FDA emphasizes that such information is required to be submitted "as soon as possible" in order to protect patient safety. Moreover, as described below, sponsors are also required to notify FDA by telephone of an unexpected fatal and life-threatening adverse experience, in advance of the written notification, to provide an early warning that a potential problem exists.

FDA does not agree with those comments that suggest that the reporting obligation should run from the time that the "essential" information on the event is collected as such a provision might unduly delay reporting of vital information. However, FDA understands that 10 working days may not be sufficient time in every case to determine conclusively whether the factors triggering a report under ß 312.32(c) are present, i.e., for determining that an adverse event reported to the sponsor is associated with use of the drug and that the event may suggest a significant hazard, contraindication, side effect, or precaution. In those cases in which the sponsor's initial information may not be conclusive, FDA advises the sponsor to err on the side of caution, to submit the preliminary information, and to follow up this initial report with whatever more definitive information is subsequently obtained.

Finally, FDA declines to adopt a different time frame for reporting foreign safety information than the 10-working-day time frame adopted for all other safety information. As the relevance and importance of safety information should usually not depend on the source of the information, FDA concludes that an exception should not be made for foreign experiences.

84. The final rule requires the sponsor to report a serious and unexpected adverse experience if the experience is "associated with the use of the drug." The proposal defined this phrase to mean that "there is a reasonable possibility that the event may have been caused by the drug." One comment suggested that requiring a sponsor to determine whether an event was possibly caused by the drug introduced a new concept to adverse reaction reporting. Although supporting the concept, the comment suggested that a determination about diverse event causality would require more time than the proposal allowed.

FDA rejects this comment as it believes that 10 working days allowed under this final rule should generally be adequate time to make the required determination. Moreover, FDA does not regard the cited requirement as representing a significant departure from current requirements. Under the current regulation, the sponsor is required to report "any finding associated with the use of the drug that may suggest significant hazards, contraindications, side effects, and precautions pertinent to safety * * *." Implicit in this requirement is an expectation that the sponsor will report events that the sponsor believes may have been caused by the drug. The revision simply makes this expectation explicit. To the extent that assessing the causation of an adverse experience is considered a close call, FDA advises the sponsor to err on the side of reporting.

Telephone Call Requirement 

85. A number of comments objected to the proposed requirement that sponsors transmit each IND safety report by telephone at the same time as a written safety report is submitted. Comments criticized the requirements as being unnecessarily burdensome on both FDA and sponsors. Several objected specifically to the proposed requirement that the sponsor contact each investigator by phone, one comment noting that there may be over 100 investigators in a single investigation. Another comment argued that telephone notification served no useful purpose because the same information conveyed by telephone would also be concurrently submitted in a written report prominently identified as an "IND safety report."

Those comments that did not urge rescinding the requirements in its entirety recommended scaling it back significantly. One comment suggested that telephone reporting should be required only when an adverse event is so alarming that the sponsor elects to discontinue the study. Other comments recommended that sponsors not be required to telephone investigators at all, or that the sponsor only be required to telephone investigators concerning the most significant new safety information.

FDA has revised ß 312.32(c)(2) to limit the telephone call requirement to adverse experiences that are obtained from the sponsor's own clinical studies that suggest an unexpected fatal or life-threatening experience associated with use of the drug. The information would be required to be relayed by telephone only to the agency unless FDA also requests the sponsor to telephone all investigators. The change should ensure that a sponsor's reporting obligations are no greater than necessary for the timely communication of the most urgent safety information. Because drug-related deaths or life-threatening experiences are relatively rare occurrences during a clinical trial, the change should also keep the amount of information transmitted by telephone to a manageable level.

FDA emphasizes that the 3-day telephone alert is reserved for the most urgent circumstances. Thus, for purposes of this section, the term "life-threatening" means that the patient was, in the view of the investigator, at immediate risk of death from the reaction as it occurred, i.e., it does not include a reaction that, had it occurred in a more serious form, might have caused death. For example, drug-induced hepatitis that resolved without evidence of hepatic failure would not be considered life-threatening in this context even though drug-induced hepatitis can be fatal.

86. Two comments mistakenly interpreted the proposal as requiring IND safety reports for experiences that occur due to the natural course of the disease being treated.

As noted in the proposal, only adverse experiences "associated with the use of the drug" need be reported, i.e., those events for which there is a reasonable possibility that the event may have been caused by the drug. A death due to the natural course of a disease would not meet this criterion and thus would not have to be reported in an IND safety report. Such deaths, however, would be reported in the annual report. See ß 312.33(b)(3).

Safety Report Format 

87. One comment noted that the proposal did not specify a reporting format for IND safety reports and suggested that the form used to report experiences with marketed drugs -- the Form FDA-1639 -- be used.

This final rule does not prescribe the use of any specific format for reporting safety information. However, the agency notes that the one page form FDA-1639 is designed primarily as a means to permit individual physicians to make "spontaneous" reports concerning adverse drug reactions in patients under their care. The form is clearly inappropriate for reporting in a safety report information about animal tests. It is also in most cases not an appropriate means of transmitting information about human clinical experience during a clinical investigation, as more extensive information on individual adverse experiences is needed than can ordinarily be included in a one page report. Generally, while the kinds of data entries required in a FDA-1639 report for a marketed drug are also appropriate for reporting adverse experiences in IND safety reports, more detailed reporting is desirable, particularly for reporting clinical adverse experiences from Phase 1 and 2 studies. While FDA does not encourage use of the form, FDA believes the Form FDA-1639 may in some cases be acceptable for submitting IND safety reports about human clinical experiences during Phase 3 studies and would be happy to discuss use of the form with individual sponsors.

IND Study of a Marketed Drug 

88. One comment urged that the regulations specify whether, when a marketed drug is used in a clinical study under and IND, an adverse experience associated with use of that drug product should be reported to the division in FDA's Center for Drugs and Biologics which is responsible for monitoring adverse reactions for marketed drugs or to the IND. The comment suggested that the adverse experience should be reported to only one application with appropriate cross-reference filed in the other application.

As a general rule, FDA agrees that adverse experiences associated with use of a drug that is subject to both an investigational new drug application and a marketing application need not be reported to both. Accordingly, the agency has in this final rule limited IND safety reporting for clinical studies of marketed drugs to those adverse experiences associated with the clinical study itself. Adverse experiences that originate from outside the clinical study (including, for example, "spontaneous" reports submitted to the drug firm by individual practitioners) need not be reported to the IND file provided such experiences are reported to the marketing application file in accordance with the applicable NDA regulations (21 CFR 314.80).

Followup Reports 

89. Several comments addressed the issue of followup reports. One comment, noting that the regulation would require the sponsor to investigate all safety information received by it, asked FDA to clarify what level or degree of investigation would be required for various sources of safety information including, for example, clinical experiences in studies conducted under the IND, reports from the scientific literature, and reports on foreign experiences with the drug.

Regardless of the source of the safety information, FDA expects a sponsor to conduct as thorough an investigation as is feasible to interpret the adverse experience that is the basis for the initial safety report. Of course, some initial reports will require more followup than others. For example, reports of clinical experience in the sponsor's own IND studies or reports of formal clinical trials from the scientific literature might be sufficiently complete in themselves to require little, if any, followup. In contrast, a literature report of an adverse experience that does not tie the experience to an individual patient may require substantial followup. Reports of adverse experiences from foreign marketing experience may be sketchy or even uninterpretable, and a sponsor may be unable to obtain further information. Thus, the extent of followup will depend on the source of the safety information, on the amount of information already reported, and on the potential for obtaining additional useful information through diligent effort.

90. One comment urged FDA to require followup reports to be submitted within 60 days of the initial report (unless a shorter period is required by the agency for a specific adverse experience on grounds of safety), rather than "promptly" as had been proposed. Two comments suggested that the final rule be amended to require the submission of followup reports to IND safety reports "if needed."

Under the final rule, a sponsor is required to investigate all safety information received by it. Ordinarily, these investigations will not be completed within the time limit prescribed for the IND safety report. However, if such investigations are completed within the time frame prescribed, the sponsor should indicate this fact in the IND safety report. No further followup report would then be required. With respect to the suggested time period for the submission of followup reports to an IND safety report, FDA does not believe it should prescribe any specific time period, given the variety of experiences that may require followup. Therefore, the final rule will remain as proposed.

91. With respect to the provision requiring prompt reporting of "relevant information" in followup reports to an IND safety report, one comment asked FDA to clarify the term "relevant."

Determining the relevance of information is invariably a matter of judgment. In this case, relevant information is information that explains or clarifies the circumstances of the reported adverse experience. For example, each followup might include reports of autopsy findings or reports of the results of additional blood tests. FDA will provide additional guidance on followup on request.

91a. Finally, FDA has added a provision stating that a safety report submitted in accordance with these regulations does not necessarily reflect a conclusion by either the sponsor or FDA that the report constitutes an admission that the drug caused or contributed to an adverse experience. This "disclaimer" provision parallels similar provisions adopted in the NDA Rewrite (50 FR 7452; February 22, 1985) and in the medical device reporting regulation (49 FR 48272; December 12, 1984). The disclaimer provision was adopted in response to comments expressing concern about the legal liability consequences of reporting possible adverse experiences. FDA advises, as it has done previously, that although FDA does not intend for such a report to be viewed as an admission of liability, whether a court will treat a submission to FDA as an admission will depend on factors outside of the agency's control, such as the contents of the report.

Annual Reports (ß 312.33)

92. A number of comments asked that the agency give more detailed guidance on what information should be included in the annual report of an investigation's progress. In addition, comments were interested in knowing whether specifically identified items of information should be submitted in the annual report or in some other submission to the agency. For example, one comment, noting that there was no explicit mention in the annual report section regarding submission of reports from the scientific literature, asked whether such information should be reported in annual reports, safety reports, or both. Along similar lines, several comments asked whether information on animal studies should be submitted in information amendments or in annual reports. Finally, one comment contended that the proposed annual report requirement for summaries of the previous year's clinical and nonclinical investigations was "unnecessary and burdensome," arguing that by the time the annual report was due such information would already have been reported to FDA in information amendments or other submissions.

FDA has carefully considered these comments and concludes that the submission requirements for annual reports, which are expressed in very general terms in the proposed rule, should be identified in the final rule in more detail. These changes, which are outlined below, should significantly increase the usefulness of these reports in providing both sponsors and FDA with insight into the status and progress of studies. The changes will also provide guidance to sponsors in determining whether information obtained during the course of the investigation should be submitted in information amendments or safety reports rather than in the annual report.

As proposed, ß 312.33(a) called for a brief summary of the status of each of the clinical studies conducted (both those in progress and completed) during the past year, but did not specify the contents of such reports. To clarify this requirement, FDA has revised ß 312.33 to require the sponsor to submit: (1) Brief "identifier" information for each study, and (2) a brief numerical analysis of patient exposure to the investigational drug in that study, i.e., the number of subjects planned for inclusion in the study, the number whose participation in the study was completed as planned, and the number who dropped out. The final rule also requires a brief description of the study outcome or interim results for each study -- on a study-by-study basis -- for which results are available. This should be a concise, one or two sentence statement of study results. For example, if the study made some important finding about pharmacokinetics, that should be so indicated. Likewise, if a placebo-controlled study distinguished or failed to distinguish between the investigational drug and the placebo, that too should be so stated.

In addition to these clarifications of the "status report" elements of the annual report, the final rule also elaborates on the proposed provision in ß 312.33(b) requiring a brief summary of information obtained during the previous year's investigations. This section serves as a means to bring together data from individual studies and briefly communicate what was learned during the past year about the investigational drug's safety and effectiveness. The final rule specifically identifies five pieces of information to be included in the annual report relating to the clinical experience with the drug: (1) A summary showing the most frequent and most serious adverse experiences by body system; (2) a summary of the past year's safety reports; (3) a list of subjects who died during the past year; (4) a list of subjects who dropped out of clinical investigations during the past year; and (5) a brief description of what, if anything, was obtained that is pertinent to an understanding of the drug's actions. Also, this provision requires a list of preclinical studies (including animal studies) completed or in progress during the past year and a summary of major preclinical findings. Finally, the sponsor is expected to submit a summary of any significant manufacturing or microbiological changes made during the past year.

The agency believes that there should be little overlap between the information submitted by the sponsor in information amendments, protocol amendments, or safety reports and information submitted in the annual report. As noted above, FDA expects annual reports in general to contain brief information summing up what was learned about the investigational drug during the past year. Annual reports thus provide a periodic overview of the investigation's progress. In contrast, amendment and safety reports contain specific information needed by the agency in determining whether to continue to allow the study to proceed.

93. Several comments recommended that the required lists of deaths and drop-outs include only deaths and drop-outs related to the safety of the investigational drug. One comment contended that to include nonsafety related deaths and drop-outs would require FDA's reviewers to sort through potentially long lists of subjects and extract from those the cases related to safety, and that this sorting process would create considerable work for both the sponsor and the agency without offsetting benefits. The comment recommended confining the lists of deaths and drop-outs to those subjects who suffered a drug-related adverse reaction and who were not previously identified in safety reports to FDA. Thus, according to the comment, every drug-related reaction would be submitted to FDA in either a safety report or an annual report.

Because of the difficulty of assessing the meaning of single adverse experiences -- of determining, for example, whether the death of a subject in a study of a cardiovascular drug is due to the drug itself or to the natural course of the disease being treated -- FDA believes it is important periodically to aggregate all such experiences, whether or not the individual events are thought to be drug related, for review and analysis. Such grouping may show an increased incidence of an adverse experience or other problem that would not be readily ascertainable in a review of single, discrete adverse experiences. Therefore, FDA believes that the list of deaths and drop-outs in the annual report should include all deaths or drop-outs, whether or not thought by the sponsor to be drug related.

94. Several comments objected to the requirement that the annual report contain an updated general investigational plan for the following year. One comment questioned the value of submitting in each annual report a wholly new description of the general investigational plan for the coming year. The comment claimed that it is difficult, if not impossible, to schedule clinical trials with precision and that artificial time frames like the "coming year" are, therefore, inappropriate. Another comment suggested that the requirement would increase the sponsor's burden in preparing annual reports and increase the amount of material that FDA must review. The comment recommended that the provision be revised to require only a description of significant changes in the investigational plan not covered by previously submitted amendments or other sections of the annual report.

As noted in the discussion of comments on the general investigational plan in paragraphs 39 through 41, FDA does not expect the general investigational plan to be a detailed description of future clinical studies, but rather a very brief summary of plans for clinical studies for the following year. As noted earlier, the purpose of the plan is simply to place individual studies within a larger context so that FDA reviewers are not operating in a vacumm. Moreover, the provision does not obligate the sponsor to invest resources into formulating plans that are not otherwise available; if at the time the sponsor submits the annual report, plans for the following year are not yet formulated, the sponsor need only so state in the submission. For these reasons, FDA believes the requirement does not represent a significant burden and should be retained as proposed.

95. One comment asked whether, if the general investigational plan for the coming year is unchanged, the plan must be resubmitted.

FDA advises that if the plan for the following year is unchanged, the sponsor may simply refer to the previously submitted plan.

96. One comment asked whether the requirement for a brief summary of significant foreign marketing developments with the drug during the past year applied to the experience of other firms that may be marketing the same dosage form of the drug, or applied only to the sponsor's experience.

Reports of regulatory actions taken by foreign drug licensing authorities -- such as license refusals, or withdrawals from the market for safety reasons -- frequently signal potential problems with an investigational drug. As the relevance and significance of this information would usually not depend on the identity of the company marketing the drug, FDA believes a sponsor should report to FDA all significant foreign regulatory actions taken, whether or not the action was taken with respect to the sponsor's own drug. Finally, FDA would expect a sponsor to report significant actions taken not only with respect to the specific dosage form under study, but also with respect to other dosage forms of the drug, since such information may also be extremely valuable.

97. On its own initiative, FDA has revised ß 312.33 to specify that the sponsor is required to submit the first annual report no later than 60 days after the anniversary date of the initial IND submission. FDA believes this provides needed guidance on the proper timing of annual reports and represents a reasonable time frame within which to prepare and submit such reports.

Identification of Patients

98. One comment expressed concern that the proposed regulation would not protect subjects' rights of privacy. The comment noted that several sections of the proposal would require the submission of a list of "deaths and drop-outs" and interpreted this to require the sponsor to identify subjects by name. The comment contended that the sponsor should not be expected to have this information in its files, and that, moreover, many sponsors go to great lengths to delete subject names from files. The comment contended that the regulation should permit the use of other identifiers, which could then be used in conjunction with the investigator's records to identify a subject by name.

As noted in the NDA Rewrite final rule (ß 312.80(h)), FDA does not expect a sponsor to maintain in its records the names and addresses of individual subjects. In reporting deaths and drop-outs, moreover, to protect subject confidentiality, sponsors should identify subjects by initials or some other sort of coding, rather than listing subjects by name and address. However, sponsors and/or participating investigators are still required to retain sufficient information about subjects to permit FDA to find the name and address of a subject should the need to do so arise.

Treatment Use of an Investigational New Drug (Proposed ß 312.34)

98a. Elsewhere in this issue of the Federal Register, FDA is reproposing new rules governing treatment use of investigational new drugs. Comments received on this issue are addressed in that reproposal. Because there are no existing regulations governing treatment use, ß 312.34 has been held in reserve.

Emergency Use of an Investigational New Drug (ß 312.36)

99. One comment from a professional medical association complained that the proposal did not specify the appropriate procedures for obtaining an investigational drug in an emergency and urged that the final rule include detailed guidance for the benefit of individual physicians. Another comment contended that if an emergency need for an investigational drug arises after normal working hours or on a week-end or holiday, a requirement that no emergency shipment may be made without FDA authorization could possibly delay initiation of vitally important therapy. This comment recommended that the provision be revised to permit emergency shipment of a drug without FDA authorization if: (1) No responsible agency official can be reached by telephone; (2) the sponsor obtains the authorization of the chairman of an appropriate reviewing IRB; and (3) FDA is notified of the shipment by telephone as soon thereafter as is practicable.

FDA advises that, even when a situation arises that in the judgment of a treating physician calls for the emergency use of an investigational drug, an IND is still necessary. The physician's first step should be to contact the manufacturer of the drug and determine whether the physician may be added as an investigator under the manufacturer's IND. Should the company elect not to add the physician to its IND, the physician should then contact the agency directly.

When contacting the agency, the physician will be placed in contact with an FDA medical officer familiar with the drug who will review the proposed circumstances for use. If the medical officer is satisfied that emergency use of the drug is justified, the medical officer may authorize its shipment and use in advance of any formal written submission to the agency.

Because the procedures governing "emergency IND's" may change from time to time, FDA has not codified the details of current practices into this final rule. However, the final rule does identify the specific review office to contact to get the process in motion. Also, FDA has prepared an informational sheet that describes in some detail the procedures to be followed to obtain emergency authorization to use an investigational drug. The informational sheet also describes an investigator's responsibilities in an emergency with respect to informed consent and IRB review requirements. This informational sheet is available from the Food and Drug Administration, Office of the Associate Commissioner for Health Affairs (HFY-20), 5600 Fishers Lane, Rockville, MD 20857 (301-443-6143).

Requests for emergency authorization that are received after normal duty hours are handled like those received during the working day, except that in such cases the initial contact will be FDA's duty officer or the agency answering service rather than the appropriate review office. Because this procedure has worked well, and has not, to the agency's knowledge, materially delayed shipment of urgently needed drugs, FDA does not believe there is a need for an alternative procedure.

100. One comment perceived an inconsistency between the emergency use provisions of the IND regulations and the provisions in the IRB regulations (21 CFR 56.104) governing the proper role of the IRB with respect to the emergency use of test articles. The IRB regulations permit the emergency use of an investigational drug without prior IRB review and approval provided that the IRB is notified of such use within 5 working days. This comment concluded that this provision of the IRB regulation represented sound policy and should override any conflicting sections of the IND regulations.

This comment erroneously characterized this provision of the IRB regulations as exhausting all FDA regulation of emergency use of investigational drugs. However, IRB review requirements are supplemental to the requirements for an IND. As noted above, when a physician wants to obtain authorization to use an unmarketed investigational drug in an emergency, an IND is still required.

Withdrawal of an IND (ß 312.38)

101. Proposed ß 312.38 sets forth procedures for withdrawing an IND by a sponsor. Two comments recommended that the final rule provide that IND withdrawal not result in the public availability of confidential data submitted by the sponsor. One comment contended that the act of withdrawing an IND, in itself, should not trigger the release of confidential data, since that data may have proprietary value with respect to related compounds or other possible indications for that same compound.

The public availability of all data and information in an IND for a new drug or antibiotic drug will be governed by ß 314.430, which describes the rules for disclosing information submitted in a marketing application under Part 314. The rules for public disclosure of information for biological investigational drugs are set forth in 21 CFR 601.50 and 601.51. In general, these rules hinge public disclosure on whether the information requested is trade secret or confidential commercial or financial information, and not on whether the application is formally pending with the agency. Thus, the fact that an IND has been withdrawn is not, in itself, determinative of the public availability of information in the IND file.

102. One comment suggested revising ß 312.38(b) to make clear the sponsor's responsibilities for disposing of an investigational drug once the investigation is ended.

The agency has added a new section (ß 312.59) to describe a sponsor's responsibilities for disposing of an investigational drug. Under that section, sponsors are required to assure the return or other authorized disposition of all unused supplies of an investigational drug whenever an investigator ends his or her participation in the investigation, or the investigation is terminated.

Administrative Actions
General Requirements for Use of an Investigational New Drug In a Clinical Investigation (ß 312.40)

103. One comment objected to retaining the system under which an IND goes into effect 30 days after FDA receives the IND unless FDA notifies the sponsors that the investigations covered by the IND may not begin. The comment acknowledged the need for expeditious review of investigational applications, but expressed concern that during substantial portions of the 30-day period the application may not actually be available for review by FDA's scientific reviewers because of the time needed to route the IND to the scientific reviewers. The comment recommended that, in light of this "delay," the 30-day period should be deemed to begin from the time that the application has actually been transmitted to the responsible reviewing officials.

Under longstanding practice, agency reviewers have had 30 days from date of receipt of the IND to review the submission. Agency reviewers are asked during this period to decide whether the information submitted in the IND supports initiation of the proposed clinical investigations. Only rarely has the agency found the 30-day period, which period includes the administrative time taken up in routing a submission from the file room that initially receives the IND to the designated scientific review team members, insufficient to conduct an adequate initial review. In those rare cases, the agency has invariably obtained the sponsor's agreement to delay its proposed studies pending completion of the agency review. FDA believes this system has worked satisfactorily and should not be changed.

Return to Part 1 | Continue to Part 3