Pharmaceutical data integrity: from chromatography PDFs to LIMS

Pharmaceutical data integrity is a focus of GMP inspections, and chromatography data — high-volume, complex format, tied to critical quality attributes — has been a high-risk item for years.

This article approaches the problem from the compliance angle: how to migrate chromatography data from PDF to LIMS while preserving data integrity end-to-end.

1. Regulatory background

1.1 FDA 21 CFR Part 11

Part 11 sets explicit requirements for electronic records and signatures:

• Audit trail — every change must be traceable
• Access control — only authorized personnel may create, modify, or delete electronic records
• Tamper resistance — electronic records must be protected against tampering

1.2 WHO ALCOA+ principles

The ALCOA+ framework from WHO Technical Reports is the bedrock of data integrity:

Letter	Meaning	Requirement
A	Attributable	Each record is tied to a specific person
L	Legible	Records are readable and durable
C	Contemporaneous	Recorded at the time of activity
O	Original	Original record (or verified copy) is preserved
A	Accurate	Records are correct
+	Complete / Consistent / Enduring / Available	Round out ALCOA

1.3 China NMPA “Drug Records and Data Management Requirements” (2024)

Chinese regulators require truthful, accurate, complete, and traceable data management. Notably, manually transcribed data requires two-person review — which underscores the value of automation.

2. Risks across the chromatography data flow

The data passes through multiple stages from creation to archive. The manual entry from PDF to Excel is the highest-risk step:

Chromatograph → Workstation → PDF report → Manual entry to Excel → LIMS import
                                                     ↑
                                       Risk: typos, omissions, no traceability

3. Risk assessment — three approaches

3.1 Manual entry

Risk dimension	Rating	Notes
Accuracy	⚠️ High	Human eye + keystroke; errors are inevitable
Completeness	⚠️ Medium	Multi-page tables get rows missed
Traceability	❌ Very high	Cannot prove “this Excel cell came from that PDF cell”
Audit trail	❌ None	Entry process is undocumented

Manual entry is the worst option in a GMP environment — yet it’s the most common in practice.

3.2 Generic OCR

Risk dimension	Rating	Notes
Accuracy	⚠️ Medium	Decimals and scientific notation often misread
Completeness	⚠️ Medium	Cross-page tables may lose rows
Traceability	⚠️ Low	No precise linkage to PDF source position
Audit trail	❌ None	No structured audit of the extraction

3.3 ChromaParse (purpose-built)

Risk dimension	Rating	Notes
Accuracy	✅ Low	Tuned for chromatography tables, 99.9%+ accuracy
Completeness	✅ Low	Handles cross-page and merged-cell layouts
Traceability	✅ Low	Source-trace: click an Excel value → highlight in PDF
Audit trail	✅ Yes	Extraction process is logged; mapping is verifiable

4. Best practices from PDF to LIMS

1. Step 1 — Preserve the original. Whatever extraction method you pick, keep the original PDF. ALCOA+ “Original” requires it.

2. Step 2 — Extract with a validated tool. Use a validated tool to produce structured data from the PDF, with values matching the PDF and source-trace available.

3. Step 3 — Verify a sample. Spot-check 5–10% of records against the PDF source, focusing on retention times and peak areas.

4. Step 4 — Load into LIMS. Import the validated data as Excel/CSV, ensuring format and field mapping are correct.

5. Step 5 — Establish an audit chain. Complete chain: PDF → ChromaParse extraction → Excel intermediate → LIMS record. Every link traceable, every step verifiable.

5. Implementation tips

1. Phased rollout — pilot with non-GMP-critical data first; expand to GMP data after accuracy is validated
2. SOPs — write standard operating procedures for extraction, verification, and import
3. Training — ensure operators understand both data integrity requirements and the tool
4. Periodic review — sample-check extracted results even after the tool is in use
5. Recordkeeping — retain extraction and verification records per GMP retention rules

Don’t just evaluate the tool — evaluate the whole data flow’s compliance. Purpose-built chromatography extraction tools (like ChromaParse) have clear advantages in accuracy, completeness, and traceability. Combined with proper validation and SOPs, they hit the best balance between GMP compliance and efficiency.

---

References

• FDA 21 CFR Part 11 — Electronic Records, Electronic Signatures
• WHO Technical Report Series 1033 Annex 4 — Data integrity
• NMPA Drug Records and Data Management Requirements
• EU GMP Annex 11 — Computerised Systems