Investigational New Drug (IND) applications: what FDA expects, and what has triggered a clinical hold

An Investigational New Drug application (IND) is the submission a sponsor makes to FDA to begin administering an investigational drug or biologic to humans. It is not an approval and not an application to market; it is the mechanism that lets a sponsor lawfully ship an unapproved drug across state lines and dose it in a clinical trial. In regulatory terms, an IND is required whenever a sponsor intends to conduct a clinical investigation with a drug that is subject to 21 CFR 312.2(a), and the sponsor may not begin that investigation until it is subject to an IND that is "in effect" under 21 CFR 312.40 ⁹.

The timing rule is the part every clinical team plans around. Under 312.40, an IND goes into effect 30 days after FDA receives it, unless FDA notifies the sponsor that the investigation is subject to a clinical hold, or notifies the sponsor earlier that the study may begin ¹¹⁴. The 30-day window is therefore a default safe harbor: submit, wait 30 days, and dose, provided FDA has not intervened. A clinical hold is the intervention that removes that safe harbor.

The governing principle: safety first, and requirements that scale

FDA's review of an IND is anchored in one idea. Under 21 CFR 312.22, the agency's primary objective in all phases is to protect the safety and rights of subjects; in Phase 2 and Phase 3, FDA additionally focuses on whether the studies are of sufficient scientific quality to evaluate effectiveness and safety ⁸. Just as important for planning, the amount of information an IND must contain is not fixed. It is graded, scaling with the drug's novelty, how much it has already been studied, the known or suspected risks, and the phase of development ⁸. An early Phase 1 IND for a first-in-human molecule is judged mainly on whether it is reasonably safe to start; a Phase 3 IND is judged on that plus whether the trial design can actually support a marketing decision ⁸.

The phase definitions FDA uses for this purpose are set out in 21 CFR 312.21 ¹:

Phase	Purpose	Typical size
Phase 1	First introduction into humans; determine metabolism, pharmacologic actions, side effects with increasing dose, and, if possible, early effectiveness; inform Phase 2 design	~20 to 80 subjects 1
Phase 2	Controlled studies to evaluate effectiveness for a specific indication and to determine common short-term risks	up to several hundred patients 1
Phase 3	Expanded controlled and uncontrolled trials to characterize benefit-risk and support physician labeling	several hundred to several thousand 1

What an IND must actually contain

The content and format requirements live in 21 CFR 312.23. FDA expects the submission in a defined order, with nine components ².

1. Cover sheet (Form FDA 1571). The administrative and legal spine of the IND. It carries sponsor identification, the phase(s) to be conducted, and a set of binding commitments: not to begin until the IND is in effect, that an IRB compliant with Part 56 will review and approve each study, and that the investigation will follow all other applicable requirements. It names the person responsible for monitoring the study and the person(s) responsible under 312.32 for safety review, discloses any obligations transferred to a CRO, and must be signed by the sponsor. If the signer is outside the United States, a U.S.-resident attorney or agent must countersign ².

2. Table of contents ².

3. Introductory statement and general investigational plan. A brief description of the drug (name, all active ingredients, pharmacologic class, structural formula if known, dosage form, route) and the broad objectives and planned duration of the investigations, a summary of prior human experience, disclosure of any withdrawal from investigation or marketing in any country for safety or effectiveness reasons, and the plan for the coming year, including indications, the general evaluation approach, the kinds of trials, the estimated patient numbers, and any anticipated serious risks ²⁴.

4. Investigator's brochure (when required under 312.55): a description of the drug substance and formulation, a summary of pharmacologic and toxicologic effects in animals and, to the extent known, humans, pharmacokinetics and disposition, prior clinical safety and effectiveness, and the anticipated risks, side effects, and special monitoring ⁴.

5. Protocols. A protocol for each planned study, with detail scaled to phase: objectives, investigator and site information and the reviewing IRB(s), patient selection and exclusion criteria and estimated enrollment, study design including any control group and bias-minimization methods, the method for determining dose and the planned maximum dose and duration of exposure, and the observations, measurements, laboratory tests, and clinical procedures used to monitor effects and minimize risk ⁴³.

6. Chemistry, manufacturing, and controls (CMC). A description of the composition, manufacture, and control of both drug substance and drug product: physical/chemical/biological characteristics, manufacturer, method of preparation, acceptable limits and analytical methods for identity, strength, quality, and purity, and stability data adequate to support the toxicology and planned clinical studies. It also covers any placebo, the labeling to be given to investigators, and an environmental assessment or categorical exclusion claim ⁷. CMC expectations are explicitly phase-dependent: early Phase 1 emphasis is on raw materials and identification and control of the new drug substance, and final specifications are not expected until the end of development ³.

7. Pharmacology and toxicology. The nonclinical package that supports the sponsor's conclusion that it is reasonably safe to begin: pharmacologic effects and mechanism of action, ADME where known, and an integrated summary of toxicologic effects (acute, subacute, and chronic toxicity as appropriate, reproductive and developmental toxicity, and route- or use-specific toxicity), with a full data tabulation for each safety-supporting study. It must identify and qualify the people who evaluated the studies and state where they were conducted, and it must include a GLP compliance statement for each study subject to Part 58 (or a brief explanation of any noncompliance) ⁷⁶.

8. Previous human experience. A summary of prior human experience, with detailed safety and effectiveness information from earlier U.S. or foreign investigation or marketing, full copies of the most relevant published literature, component-level detail for combination products, and a list of countries where the drug is marketed or was withdrawn for possible safety or effectiveness reasons ⁶.

9. Additional information. Special-topic content when relevant: drug dependence and abuse-potential data, radioactive-drug dosimetry (required for Phase 1 studies of radioactive drugs), pediatric assessment plans, and any other information FDA requests to complete its review ⁵.

Clinical holds: the grounds FDA can invoke

A clinical hold is FDA's order to delay a proposed study or suspend an ongoing one. The grounds are enumerated in 21 CFR 312.42, and they map directly onto the safety-first principle. For a Phase 1 study, FDA may impose a hold if any of the following applies ¹⁰:

1. Human subjects are or would be exposed to an unreasonable and significant risk of illness or injury.
2. The clinical investigators named in the IND are not qualified by scientific training and experience to conduct the described investigation.
3. The investigator brochure is misleading, erroneous, or materially incomplete.
4. The IND does not contain sufficient information required under 312.23 to assess the risks to subjects.
5. The IND is for a drug intended to treat a life-threatening disease or condition affecting both sexes, and men or women of reproductive potential are excluded because of reproductive or developmental toxicity risk ¹⁰.

For a Phase 2 or Phase 3 study, all five Phase 1 grounds still apply, plus one additional ground: the plan or protocol is clearly deficient in design to meet its stated objectives ¹⁰. That extra ground is the regulatory expression of FDA's added Phase 2/3 focus on scientific quality ⁸.

Two of these grounds deserve emphasis because they are the ones sponsors most often trip over. Ground 1 (unreasonable and significant risk) is the toxicity-signal trigger. Ground 4 (insufficient information to assess risk) is the completeness trigger, and it links the hold power back to the content requirements of 312.23: an IND that is thin on nonclinical data, CMC, or safety characterization can be held not because a specific hazard is proven, but because FDA cannot rule one out.

How a hold works, and how it is lifted

FDA may impose a hold by telephone or other rapid communication and confirm it in writing. The order identifies the studies under the IND to which the hold applies and briefly explains the basis; FDA provides the written explanation as soon as possible and no later than 30 days after imposing the hold ¹².

To get the hold lifted, the sponsor submits a written request for removal together with a complete response to each issue in the hold order. FDA must respond in writing within 30 calendar days of receiving the request and complete response, either removing or maintaining the hold and stating its reasons. Critically, the sponsor may not resume the trial until FDA affirmatively notifies it that the hold has been lifted; the 30-day clock is FDA's response deadline, not a self-executing release ¹². If all investigations under an IND remain on hold for a year or more, FDA may place the IND on inactive status ¹².

Safety reporting keeps the IND under continuous review

An IND is not a one-time filing. Under 21 CFR 312.32, the sponsor must submit IND safety reports for serious and unexpected suspected adverse reactions, for findings from other studies (epidemiologic, pooled analyses, or clinical studies) that suggest a significant human risk, for findings from animal or in vitro testing that suggest a significant human risk, and for a clinically important increase in the rate of a serious suspected adverse reaction over what the protocol or investigator brochure anticipated ⁶⁸. The default timeframe is as soon as possible but no later than 15 calendar days; for an unexpected fatal or life-threatening suspected adverse reaction, the deadline compresses to 7 calendar days ⁶⁸⁶⁹. These reports are a principal way emerging risk reaches FDA on an active program, and they are frequently the information that precedes a hold on an ongoing study.

What has actually triggered clinical holds

FDA review documents in Drugs@FDA record the real-world basis for holds across a wide range of programs. In practice the triggers cluster into a handful of recurring categories.

Nonclinical toxicology signals and data gaps. These are among the most common Phase 1 and early-development triggers. FDA placed rilzabrutinib (later Wayrilz) on full clinical hold citing neurotoxicity findings in animal models and unknown long-term effects for chronic ITP use, invoking both unreasonable risk and insufficient information to assess risk ⁹³. Clesrovimab (Enflonsia) went on full hold over potential cytokine release syndrome flagged by an in vitro cytokine release assay ⁶⁵. Vigabatrin (Sabril) was held over intramyelinic edema seen in animals until reversibility could be monitored ¹⁴⁹. Mycapssa (oral octreotide) drew a partial hold for incomplete monkey histopathology ⁶³, and Sohonos (palovarotene) a partial hold in children under 14 over premature growth-plate closure risk ⁹⁸.

Hepatotoxicity and deaths. Pretomanid-containing regimens were placed on partial hold after hepatotoxicity-related deaths ¹³⁵¹⁴². The elbasvir/grazoprevir program (Zepatier) was held over late transaminase elevations at higher grazoprevir doses, and the dose was capped at 100 mg on release ⁹⁰. Pexidartinib (Turalio) went on partial hold for hepatotoxicity concerns, with further enrollment halted ⁹⁵.

Serious adverse-event signals in humans. Pacritinib (Vonjo) drew a full hold across all protocols under its IND over detrimental overall survival together with fatal and life-threatening hemorrhage, cardiac failure, and arrhythmias, including sudden death ¹⁴⁸. Concizumab (Alhemo) protocols went on full hold after non-fatal treatment-emergent thrombosis in three patients ¹⁴⁰. Cenobamate (Xcopri) drew a partial hold over DRESS (drug reaction with eosinophilia and systemic symptoms) risk ¹¹. The fingolimod (Gilenya) MS program was held over macular edema, AV block, and pulmonary toxicity concerns ¹¹¹.

Carcinogenicity findings. Pregabalin (Lyrica) was placed on partial hold (IND 53,763) over tumorigenic hemangiosarcoma in mice and the concern that the risk was unacceptable for the neuropathic pain population ⁷⁴. The amikacin liposome inhalation program (Arikayce) was held over increased squamous cell carcinoma in rats ¹¹³.

Insufficient information to assess risk. Lenacapavir (Yeztugo) INDs were placed on full hold for insufficient information to assess risk and unreasonable and significant risk of illness or injury, tied to the injection container-closure system and injection-site reaction reports ¹³⁶¹³⁷. Dimethyl fumarate (Tecfidera) was held for insufficient safety information, including on prior BG-12-exposed patients who had cardiac and liver toxicity and inadequate safety support for the Phase 2 dose-finding trial ⁸⁷.

CMC deficiencies. Methylphenidate extended-release (Quillivant XR) was initially held for inadequate CMC information ¹⁰⁰, a direct example of ground 4 operating on the manufacturing section rather than on a safety signal.

How sponsors get holds lifted

The review record also shows what a "complete response" looks like in practice, and the common thread is that sponsors resolve holds by generating the specific data or protocol changes FDA asked for, not by argument alone.

• Additional nonclinical studies. Otiprio (ciprofloxacin otic) cleared a full hold, imposed over cochlear hair-cell damage seen in guinea pigs, after the sponsor ran additional animal studies with a modified vehicle showing reduced hearing loss and changed the processing method ⁴⁶¹⁰⁵. Elelyso (taliglucerase alfa) cleared a partial hold after a 9-month chronic toxicity study in monkeys ¹⁶²⁰. Arikayce's hold was removed in January 2012 after additional toxicology, including an absence of pre-neoplastic findings in a chronic dog study ¹¹³.
• Protocol and dose changes plus risk mitigation. Turalio's hold was addressed with a revised protocol adding lower dose-modification thresholds for ALT/AST/bilirubin ⁹⁵. Cobenfy (xanomeline/trospium) cleared its hold with additional toxicology reports, a new Phase 1 protocol, and PK modeling, with a commitment not to escalate dose until specified reports were reviewed ⁴⁹. Cenobamate's DRESS partial hold was met with reduced dose-escalation, expanded monitoring, and investigator education ¹⁴.
• CMC amendments. Quillivant XR's hold was removed after the sponsor submitted an amendment addressing the CMC deficiencies ¹⁰⁰.

The pattern is consistent with the regulation's logic. A hold is lifted when the sponsor corrects the cited deficiency or otherwise satisfies FDA that subjects will not face unreasonable risk, and only once FDA says so in writing ¹². For teams planning a first-in-human program, the practical lesson is that most holds are foreseeable: they follow from an incomplete nonclinical or CMC package, a dosing plan that outruns the safety data, or a specific toxicology signal that has not been adequately characterized or mitigated, which are exactly the elements 312.23 asks the IND to establish before dosing begins ²¹⁰.