FDA Complete Response Letter Analysis
The gap between a nearly complete filing and a CRL is still usually the unfinished execution layer around the science.
This brief consolidates 291 unapproved FDA CRLs into a share-ready narrative, then connects that narrative to a curated 500-check operating file and the underlying raw exports.
291
unapproved CRLs
1,748
deficiency rows
500
ranked Top 500 checks
77%
major or critical findings
Report frame
Share-first artifact
Leadership should start with the executive report, not the raw CSVs. The ranked checklist only makes sense once the pattern and burden distribution are clear.
Operating file
The Top 500 checklist is ranked by distinct letter recurrence, total mentions, and dominant severity so teams can work recurring risk before submission.
Evidence retained
The full checklist and full deficiency exports remain in the package to preserve traceability back to the extracted dataset.
Executive takeaways
What this report says in plain terms
Execution gaps still dominate
The recurring pattern is not just scientific disagreement. The letters repeatedly surface unfinished safety, labeling, translation, facility, and cross-module execution work.
The working file must be curated
2,802 unique extracted requirement statements were reduced into the 500 most recurring checks so a team can work the highest-signal items first.
Recent years carry the heaviest load
The 2024-2025 slice alone contributes 168 letters and 3,078 extracted checklist actions in the current repository set.
Package assets
Top 500 checklist
The ranked operational file for pre-submission work allocation and internal review.
Download CSV
Full checklist export
All extracted checklist lines retained for QA, RA, and regulatory operations teams.
Download CSV
Full deficiency export
One row per deficiency with module, regulation, root-cause tags, and response action language.
Download CSV
Repeated triggers
The same avoidable patterns keep showing up
213
Safety update findings
Safety update wording still appears at a scale that should force earlier ownership and closer resubmission discipline.
163
Proprietary name findings
Proprietary name work is still treated as late-stage admin too often, even though it repeatedly triggers delay and rework.
110
Manufacturing facility findings
Facility readiness remains a hard gate. Once unresolved inspection issues enter the letter, document quality no longer offsets them.
70
Foreign labeling translation findings
Translation misses are low-complexity defects that should be eliminated by process, not discovered in review.
Category spread
Where the remediation burden actually lives
Clinical
753
Clinical remains the largest category, but most of the recurring load is still packaging, completeness, interpretation, and documentation discipline.
Labeling
488
Labeling is not a side stream. PI structure, carton/container text, translations, and proprietary name loops keep surfacing.
Data Integrity
342
Cross-module mismatch continues to create late credibility problems that are expensive to remediate once FDA is already inside the file.
Administrative
338
Administrative misses are still common enough to prove that process design, not just science, drives CRL risk.
Manufacturing
299
Manufacturing pressure rises sharply in 505(b)(2) work and remains inseparable from facility readiness.
Quality Systems
294
When quality-system defects appear in the letter, the remediation footprint usually expands beyond one isolated item.
Cited regulations
Safety and labeling still dominate the citation surface
21 CFR 314.50(d)(5)(vi)(b)
199
21 CFR 314.50(l)(1)(i)
90
21 CFR 201.56(a)
37
21 CFR 201.56(d)
35
21 CFR 201.57
34
Pathway comparison
The burden moves by filing path
The average load changes meaningfully across NDA 505(b)(1), NDA 505(b)(2), and BLA work. The operating mistake is assuming a lighter clinical bridge also means a lighter execution burden.
NDA 505(b)(1)
105 letters
5.5
avg findings
Standard NDAs still skew toward clinical packaging discipline, safety work, and cross-module consistency rather than one isolated science issue.
Avg checklist
18.5
Primary modules
M5 (314), M1 (204), M2 (68)
Top issue clusters
Clinical (323), Labeling (171), Data Integrity (150)
NDA 505(b)(2)
106 letters
6.2
avg findings
505(b)(2) work continues to underweight M3 and labeling burden relative to the clinical bridge assumptions many teams start with.
Avg checklist
20.1
Primary modules
M1 (247), M3 (221), M5 (198)
Top issue clusters
Clinical (205), Labeling (183), Manufacturing (159)
BLA
76 letters
6.6
avg findings
BLAs carry the heaviest average deficiency load in the set and show deeper data-integrity spillover when issues compound.
Avg checklist
19.2
Primary modules
M5 (217), M1 (163), M3 (135)
Top issue clusters
Clinical (218), Labeling (130), Data Integrity (118)
Top 500 preview
The ranked checklist starts with recurring requirements, not generic advice
| Rank | Requirement | Owner | Letters |
|---|---|---|---|
| 1 | MajorM1 Provide English translations of current approved foreign labeling not previously submitted. | Regulatory / Medical Writing | 176 |
| 2 | CriticalM5 Include a safety update as described at 21 CFR 314.50(d)(5)(vi)(b) when responding to deficiencies. | Clinical Safety / Pharmacovigilance | 98 |
| 3 | MajorM5 Provide updated exposure information for the clinical studies/trials (e.g., number of subjects, person time). | Clinical / Biostatistics | 89 |
| 4 | MajorM5 Describe any information that suggests a substantial change in the incidence of common, but less serious, adverse events between the new data and the original application data. | Clinical Safety / Pharmacovigilance | 72 |
| 5 | MajorM5 For indications other than the proposed indication, provide separate tables for the frequencies of adverse events occurring in clinical trials. | Clinical Safety / Pharmacovigilance | 71 |
| 6 | CriticalM5 Present new safety data from studies/clinical trials for the proposed indication using the same format as in the original submission. | Clinical / Biostatistics | 70 |
| 7 | MajorM5 Include tables that compare frequencies of adverse events in the original application with the retabulated frequencies. | Clinical Safety / Pharmacovigilance | 66 |
| 8 | CriticalM5 Provide narrative summaries for serious adverse events. | Clinical Safety / Pharmacovigilance | 63 |
Severity mix
Most findings are not advisory cleanup
Major
1036 (59%)
Critical
310 (18%)
Minor
282 (16%)
Informational
120 (7%)
Methodology notes
The report and downloads share one source of truth
Source letters: data-letters/letters/unapproved_CRLs
Requirement extraction field: metadata_overlay.summary.key_requirements_detail
Finding extraction field: findings.items
Curated ranking order: Distinct letter recurrence -> Total extracted mentions -> Dominant severity -> Requirement type priority -> Requirement text
Interpretation notes
What to do with this report
Use the executive report to align leadership on where the actual CRL exposure concentrates. Use the ranked Top 500 checklist to assign operational ownership. Use the raw exports to validate, challenge, or extend any finding back to the extracted evidence.
The numbers here are generated from the same dataset and summary file that feed the landing page and the downloads. The intent is to reduce packaging risk without losing traceability.
Primary share route
Send the PDF when the recipient needs a compact, executive-ready artifact. Send the web report when they are likely to browse the methodology and downloads directly.
Recommended operating sequence
1. Review the executive pattern. 2. Work the ranked Top 500. 3. Escalate into the raw checklist and deficiency exports only where the evidence needs a closer read.