After the IND: from first in human to the marketing application.

What the regulator expects

The amount of CMC information "will vary with the phase of the investigation" and final specifications "are not expected until the end of the investigational process" (21 CFR 312.23(a)(7)(i)). Phase 1 production is exempt from the drug CGMP regulations in Part 211, but the exemption ends once the investigational drug has been made available for a Phase 2 or Phase 3 study (21 CFR 210.2(c)). For small-molecule drugs, FDA’s Phase 2 and 3 CMC guidance (May 2003) separates CMC safety information, sent in information amendments under 21 CFR 312.31, from corroborating information that can wait for the annual report under 21 CFR 312.33. ICH E8(R1) §4.1 adds that changes to the product during development should be supported by comparability data.

Key documents

• CMC information amendments, labelled as such (21 CFR 312.31(b))
• The CMC summary and corroborating information in each annual report (21 CFR 312.33)
• Comparability data between material made before and after a manufacturing change
• Development records linking clinical material to the toxicology batches and to the product you intend to market

Where it goes wrong

A manufacturing change that cannot be bridged. FDA’s Phase 1 IND guidance (1995) expects sponsors "to relate the drug product being proposed for use in a clinical study to the drug product used in the animal toxicology studies"; its Phase 2 and 3 CMC guidance (2003) says that when material made before and after a change is not comparable, sponsors should perform "additional qualification and/or bridging studies".

What this means — SPEQ interpretation

Treat each scale-up as a regulatory event, not only a manufacturing one. The 2003 guidance does not apply to biologics or biotechnology-derived proteins; their CMC expectations come from other FDA guidance and are not summarised here.

What the regulator expects

Phase 1 protocols may be less detailed than later ones, but must specify in detail "those elements of the study that are critical to safety" (21 CFR 312.23(a)(6)(i)). A new protocol may begin once it has been submitted to FDA and approved by the IRB (21 CFR 312.30(a)). IND safety reports are due within 15 calendar days, and unexpected fatal or life-threatening suspected adverse reactions within 7 calendar days (21 CFR 312.32(c)). An annual report is due within 60 days of the anniversary of the IND going into effect (21 CFR 312.33).

Key documents

• Phase 1 protocol, and protocol amendments for new studies (21 CFR 312.30)
• Investigator’s brochure, kept current
• IND safety reports (21 CFR 312.32)
• IND annual report (21 CFR 312.33)

Where it goes wrong

A clinical hold. FDA may hold a Phase 1 study if, among other grounds, the investigator brochure is "misleading, erroneous, or materially incomplete" or the IND lacks "sufficient information required under § 312.23 to assess the risks to subjects" (21 CFR 312.42(b)(1)). On the CMC side, FDA’s Phase 1 guidance says a safety concern or insufficient data to evaluate safety is the only basis for a hold based on the CMC section.

What this means — SPEQ interpretation

Phase 1 is also where the evidence for Phase 2 is gathered: 21 CFR 312.21(a)(1) expects enough pharmacokinetic and pharmacological information to design well-controlled Phase 2 studies. Plan those measurements now.

What the regulator expects

Phase 2 and 3 protocols must describe all aspects of the study in detail (21 CFR 312.23(a)(6)(ii)). Clinical supplies now fall under Part 211 (21 CFR 210.2(c)). FDA’s Phase 2 and 3 CMC guidance (2003) recommends that the CMC safety information for Phase 2 is submitted before Phase 2 starts. ICH M3(R2) §9 says a complete genotoxicity battery should be completed before Phase II trials, and §5 that repeated-dose toxicity studies should be at least as long as the clinical trials they support. ICH E8(R1) §4.3.2 describes these exploratory studies as refining the dose, the target population and the endpoints for later confirmatory studies.

Key documents

• Detailed Phase 2 protocols (21 CFR 312.23(a)(6)(ii))
• CMC safety information amendment before Phase 2 begins
• Nonclinical reports covering the planned trial duration (ICH M3(R2) §5)
• Dose-response and summary efficacy and safety data, which the End-of-Phase 2 meeting will need

Where it goes wrong

A Phase 2 or 3 study can also be held if "the plan or protocol for the investigation is clearly deficient in design to meet its stated objectives" (21 CFR 312.42(b)(2)(ii)), a ground that does not apply to Phase 1.

What this means — SPEQ interpretation

M3(R2) is written for pharmaceuticals; for biotechnology-derived products it points to ICH S6 for the nonclinical programme (M3(R2) §1.3), so check the timings against S6 before relying on them.

What the regulator expects

The regulation asks for background information at least one month before the meeting (21 CFR 312.47(b)(1)(iv)). FDA’s Formal Meetings guidance (August 2026) classes it as a Type B (EOP) meeting: FDA schedules it within 70 days of receiving the request, the meeting package is due no later than 50 days before the meeting, and FDA sends preliminary responses no later than 5 calendar days before. For an End-of-Phase 2 meeting the package should cover dose-response results, summary Phase 2 efficacy and safety data, the planned Phase 3 trial designs, major safety concerns and major analyses. Agreements are recorded in FDA’s minutes, and studies run in accordance with them are presumed sufficient for approval "barring a significant scientific development that requires otherwise" (21 CFR 312.47(b)(1)(v)).

Key documents

• Meeting request and meeting package (Formal Meetings guidance, §V and §VII)
• Phase 3 protocols and the plan for additional nonclinical studies (21 CFR 312.47(b)(1)(iv))
• Pediatric study plans, or the basis for a waiver or deferral
• FDA’s meeting minutes, kept as the record of what was agreed

Where it goes wrong

Asking too early. FDA’s Formal Meetings guidance says that if the request states the Phase 2 summary efficacy and safety data will not be in the briefing document, "the EOP2 meeting request will likely be denied", and that FDA generally grants no more than one of each Type B meeting per potential application.

What this means — SPEQ interpretation

The meeting is worth the most "before major commitments of effort and resources to specific Phase 3 tests are made" (21 CFR 312.47(b)(1)(iii)). Book it before committing money to Phase 3 sites and drug supply.

What the regulator expects

ICH E8(R1) §4.3.2 describes confirmatory studies as "often intended to provide an adequate basis for marketing approval". ICH M3(R2) expects pharmacokinetic data in test species to be available "generally before Phase III" (§3) and a male fertility study before large-scale or long trials such as Phase III (§11.1). FDA’s Phase 2 and 3 CMC guidance (2003) recommends that the Phase 3 CMC safety information is submitted before Phase 3 starts, and encourages a meeting with the CMC review team before or during Phase 3 on issues that might affect the approvability of the NDA. Near the end of Phase 3, a pre-NDA or pre-BLA meeting needs a summary of the clinical studies and a proposed format at least one month ahead (21 CFR 312.47(b)(2)).

Key documents

• Phase 3 protocols and amendments
• CMC safety information amendment before Phase 3 begins
• Completed nonclinical studies the marketing application will rely on
• Pre-NDA or pre-BLA meeting package (21 CFR 312.47(b)(2))

Where it goes wrong

Leaving a required nonclinical study until the application. ICH M3(R2) §10 says that where carcinogenicity studies are recommended for the indication, they "should be conducted to support the marketing application". They are long studies, so a late start delays the submission.

What this means — SPEQ interpretation

The process that makes Phase 3 supplies is, in practice, the one the application will describe. Changes after this point need the comparability argument the CMC stage describes, under more scrutiny.

What the regulator expects

An NDA carries technical sections for CMC, nonclinical pharmacology and toxicology, and clinical data (21 CFR 314.50(d)), including a GLP compliance statement for each study subject to Part 58 (21 CFR 314.50(d)(2)(v)). A BLA must show the product "meets prescribed requirements of safety, purity, and potency", with the same kind of GLP statement and IRB and informed consent statements (21 CFR 601.2(a)). ICH Q8(R2) sets out the Pharmaceutical Development section (3.2.P.2) and ICH Q11 the drug substance development and manufacture sections (3.2.S.2.2 to 3.2.S.2.6) of the Common Technical Document. Under the PDUFA goals for fiscal years 2023 to 2027, FDA aims to act on 90 percent of standard new molecular entity NDAs and original BLAs within 10 months of the 60-day filing date, and priority ones within 6 months.

Key documents

• CTD Module 3: drug substance and drug product, including 3.2.S.2 and 3.2.P.2 (ICH Q11, ICH Q8(R2))
• Nonclinical section with a GLP statement for each Part 58 study (21 CFR 314.50(d)(2)(v))
• Clinical section with integrated summaries of effectiveness and safety (21 CFR 314.50(d)(5))
• IRB and informed consent compliance statements for each clinical study

Where it goes wrong

A refuse-to-file decision. FDA may refuse to file an NDA that is "incomplete because it does not on its face contain information required", or that lacks the GLP statement for each nonclinical laboratory study, or the IRB and informed consent statement for each clinical study (21 CFR 314.101(d)(3), (d)(6) and (d)(7)).

What this means — SPEQ interpretation

Each of those statements is about a study finished years earlier. If the GLP and GCP status of a study was not recorded when it ran, it cannot be written truthfully at submission.