PIC/S Issues PI 006-4 on Qualification and Validation, Effective 1 October 2026

PIC/S has adopted PI 006-4, Recommendations on Qualification and Validation, which enters into force on 1 October 2026. It replaces PI 006-3 from September 2007 and was revised by a dedicated PIC/S working group.

The change of title reflects a change of construction. PI 006-3 was four separate recommendations covering validation master plan, installation and operational qualification, non-sterile process validation and cleaning validation. 

PI 006-4 is a single document covering qualification and validation as a whole, updated following the 2015 revision of Annex 15, with quality risk management under ICH Q9 running through every section. It now adds prequalification stages, performance qualification, ongoing process verification, transport verification, packaging validation, utility qualification and test method validation. The principles apply to active pharmaceutical ingredients, intermediates and finished dosage forms. Computerised systems remain under Annex 11 and aseptic processes under PI 007.

The document serves as a training resource for GMP inspectors as well as guidance for industry. Its content should therefore be read as a statement of current inspection expectations rather than as optional good practice.

Process Validation: The Three Positions That Matter

Retrospective validation is confirmed as no longer acceptable. More significantly for sites carrying older products, legacy processes validated to previous standards are expected to meet current process validation standards, with any additional testing identified through gap analysis and risk assessment and delivered through ongoing process verification.

The second position concerns batch numbers. PI 006-4 requires the number of consecutive validation batches to be scientifically justified, driven by the risk assessment and by existing product and process understanding. Where intra-batch and inter-batch variability at full scale is well characterised, fewer replicate batches may be supported. 

Where it is not, such flexibility is stated as inappropriate. Footnote 3 records that the current Annex 15 wording on a minimum of three consecutive batches is regarded as transitory, with the next revision expected to reflect a risk-based approach. Sites applying three batches by default, without variability data behind the number, should read that footnote as a direct signal of what will be asked at inspection.

The third is that ongoing process verification has formally replaced periodic revalidation, is required for every process irrespective of how initial validation was performed, and is expected to be reported on a cycle that normally does not exceed one year.

Before commercial process validation begins, PI 006-4 expects the following to be in place at the production site:

• the quality target product profile (QTPP) with associated critical quality attributes (CQA) and critical process parameters (CPP), and the critical material attributes (CMA);
• development, scale-up and technology transfer reports, and a full description of the process to be validated;
• a fully defined initial control strategy covering test stages, sampling schemes and acceptance criteria;
• qualified facilities, utilities, equipment and systems, and validated test methods;
• qualified suppliers and materials meeting predefined specifications;
• trained personnel for both manufacture and testing of the validation batches;
• a risk assessment of the process and control strategy, and an approved protocol reflecting its outcome.

The protocol content required by Annex 15 is supplemented with a reference to that risk assessment, a justification for the number of batches, a description of how routine variability will be simulated, and details of the statistical analysis to be applied. 

Planned variation is expected to cover different API and raw material lots, alternative equipment, hold time and environmental variation, and manufacture across shifts including night shifts and shift changes. Variation that cannot be planned is to be recorded and assessed in the report against expected routine ranges.

Continuous process verification and hybrid approaches are both accommodated. CPV may require prior approval by the competent authority for the marketing authorisation, and any multivariate mathematical models applied require validation in their own right.

Related Article: Process Validation in GMP

Qualification and the PQ Sampling Expectation

Section 4 retains URS, DQ, IQ, OQ and PQ while acknowledging that approaches based on good engineering practice, with ASTM E2500 cited, may also be applied. The points most likely to require changes to existing protocols are these:

• URS as a controlled document. Versions are managed through document change control as the selection process progresses, with essential and flexible requirements distinguished so the supplier pool is not narrowed unnecessarily.
• FAT and SAT reduce IQ and OQ scope. This is permitted with justification that shipment and installation will not affect the result. Both require an approved protocol and report.
• Conditional release between stages. Permitted where acceptance criteria or deviations remain open, subject to a documented assessment of no significant impact on the next activity.
• PQ sampling. Additional samples above routine in-process control are expected. Sampling only at the beginning, middle and end of the batch is stated as not appropriate. Where IPQC runs hourly, a fifteen-minute sampling plan may be required, with additional tests such as uniformity or particle size.
• Statistical treatment. Cpk, Ppk and multivariate statistical process control are identified as more powerful than simple trend charts.
• Requalification. Like-for-like changes may sit in the preventive maintenance system if documented, justified and approved by the Quality Unit. Periodic review must consider whether a series of minor changes amounts to a major change over time.

Deviations, Data Integrity and Change Control

Two requirements here will affect how validation is executed rather than how it is documented.

On deviations, PI 006-4 confirms that a result failing predefined acceptance criteria is itself a deviation, not only a failure to execute the protocol as written. It then goes further than Annex 15 by requiring that validation deviations be visible within the site deviation system and not recorded solely in the qualification or validation report, so that later production investigations can find them.

On data integrity, the VMP is expected to define how integrity of qualification and validation data will be confirmed, and that confirmation should be performed independently by personnel who are not executing the work. This is as much a resourcing question as a documentation one.

Change control is expanded considerably against PI 006-3, with cross-reference to PI 054. The additions include change control boards for complex changes, staged approval covering authorization to proceed, completion of actions and effectiveness checks, mandatory Quality Unit involvement in all change controls, tracked timelines, quantifiable effectiveness measures, and documented justification where an effectiveness check is not performed. 

See Also: Deviation Management Process in GMP

Cleaning Validation

Cleaning validation has been rewritten from the ground up and is built around health-based exposure limits (HBELs) throughout. The program is defined as three elements: development of the cleaning process with critical cleaning parameters identified, a validation study made up of a predetermined number of cleaning verification assessments, and ongoing evaluation once the study closes.

The requirements most likely to generate findings:

• Low HBEL products and manual cleaning. The required consistency may not be achievable manually. Dedicated parts, automation, or verification at each changeover should be considered.
• Verification at every changeover. Required where maximum allowable carryover limits fall below the limit of visual detection, incorporating a safety factor.
• Visual inspection. Visually clean is the minimum standard, not the standard. Inspection is performed after the equipment has dried, by qualified personnel, under defined lighting and viewing conditions, with procedures covering disassembly and use of a light source, mirror or borescope.
• Non-contact parts. Seals, flanges, mixing shafts, oven fans and heating elements are named as examples where residue may migrate, to be assessed through a practical risk-based study using the HBEL value.
• Hold times and campaign length. Dirty hold time, clean hold time and maximum campaign length all need supporting evidence, with campaign length normally defined by both time and number of batches.
• Sampling. Swab sampling is considered more effective than rinse sampling and should be used where feasible. The averaging effect of rinse sampling is explicitly noted.
• Analytical methods. Non-specific methods such as TOC and conductivity are permitted with justification, on the assumption that the whole result is attributable to the target residue.
• Limits. Acceptance criteria must address cumulative carryover across shared equipment trains, not equipment in isolation, and must not assume uniform distribution of contamination. PI 046 is referenced for HBEL, with the caveat that limits are not to be set at the calculated HBEL level.
• Failures. Continued cleaning failures and testing until clean are not acceptable. Intermittent failure requires remediation, which may extend to equipment or facility dedication.

Topics Not Present in PI 006-3

These topics are newly presented in PI 006-4:

Transport Verification

Routes documented in detail from despatch to arrival and verified before use by execution of a test protocol, with a report covering issues, delays and improvements. Where data loggers are used, the protocol must describe device location, traceability to the data, download procedures, calibration and battery life. 

One logger per pallet is recommended for air shipments where pallets may be separated, and separately shipped samples require a risk assessment on logger inclusion. Where advanced devices such as RFID are used, the system itself requires qualification.

Packaging Validation for Solid Dose

Sealing temperature, sealing pressure, and machine speed are mapped in a pre-validation exercise until an operating point in the center of the passing region is identified rather than one near the point of failure. Pressure with vacuum is identified as more sensitive than vacuum alone for integrity testing.

Supporting Utilities

Purified water PQ in three phases across a full year to capture seasonal variation. Where there is no mains supply and water arrives by tanker, the document is explicit that qualification is difficult and that procedural controls carry the weight: technical agreements with the tanker company or water supplier, a specified source such as a named borehole rather than river water, dedicated tankers, defined cleaning frequency for tankers and storage tanks, specified internal coatings and materials of construction, and holding times. 

Steam for sterilisation is tested at user points for non-condensable gases, superheat and dryness, plus chemical testing against the WFI monograph. Lower quality steam in non-contact use requires assessment of barrier failure modes. Capacity of a steam system supplying multiple equipment must be confirmed under concurrent operation and reassessed when equipment is added.

Test Method Validation

Covered at a high level with reference to ICH Q2(R2), FDA and WHO guidance, plus suitability testing for antimicrobial effect and neutralisation of sanitising agents in surface monitoring.

How PI 006-4 Sits Against PI 006-3, Annex 15 and the Annex 15 Revision

PI 006-4 does not amend Annex 15. It sits alongside the current Annex 15 of the PIC/S and EU GMP Guide, which came into operation on 1 October 2015 and remains the applicable text in the EU. 

A separate revision of Annex 15 is already underway through a joint EMA and PIC/S drafting group. Its concept paper, EMA/INS/GMP/163466/2026, was agreed by the EMA GMP/GDP IWG on 30 November 2025 and by PIC/S on 14 January 2026, and went to public consultation in February 2026. 

That revision is targeted rather than comprehensive: it implements the recommendations of the 2020 sartans lessons learned report and aligns the annex with ICH Q9(R1), principally by making the annex mandatory for chemical and biological active substance manufacturers. The indicative timetable runs to publication and PIC/S adoption in April 2028.