Sterile manufacturing3 min readBy QLogixCMC

Annex 1 three years on: the contamination control strategy inspectors now read first

A contamination control strategy (CCS) is the document, required by the revised EU GMP Annex 1 since August 2023, that sets out how a sterile manufacturing site keeps microbial, pyrogen and particulate contamination out of its products: every control, from facility design and gowning to sterilization, aseptic process simulation and environmental monitoring, with the reasoning that connects them and the data that shows they work. Three years on, inspectors ask for it first, and the gap they find most is not a missing control but a missing connection between the controls and the data.

What the revised annex changed

  • A documented contamination control strategy for the site, maintained and reviewed, that links the controls to the risks they address
  • Quality risk management as the basis for the design of the facility, the process and the monitoring, with the assessments on file
  • Environmental and personnel monitoring with alert and action levels, trended, and reviewed as part of the CCS rather than as a stand-alone log
  • Aseptic process simulation, sterilization and barrier technology expectations written out in more detail, and the same expectations applied to advanced therapies made aseptically
  • Explicit attention to the people in the cleanroom: qualification, gowning, behaviour and the frequency of their monitoring

Where inspections find the gaps

  • A CCS that lists procedures without saying which contamination risk each one controls, or what evidence shows it works
  • Environmental monitoring reviewed sample by sample against action levels, with no trend that would show a slow drift
  • Alert and action levels copied from the annex's tables rather than set from the site's own data
  • Interventions in the aseptic process that are performed but not represented in the process simulation
  • A CCS written once for the inspection and not updated after changes, deviations or monitoring excursions

Trending is the part that pays

Action levels catch a bad day. A contamination recovery rate, the share of samples showing growth in a period, trended against a limit set from the site's own baseline, catches a bad month: a gowning change that did not take, a disinfectant rotation that stopped, an air handling unit losing capacity. The annex asks for both, and the second is where a site learns something before a batch is at risk.

Interactive

Catching a contamination shift: action levels or the trend?

6weeks earlier from the trend

The trend crosses its limit in week 23; the first action-level excursion comes in week 29. Every batch in between was made in an area already drifting.

0%6%13%week 11020304050Week 1: 2 of 40 samples with growth (5.0%)Week 2: 0 of 40 samples with growth (0.0%)Week 3: 2 of 40 samples with growth (5.0%)Week 4: 1 of 40 samples with growth (2.5%)Week 5: 1 of 40 samples with growth (2.5%)Week 6: 3 of 40 samples with growth (7.5%)Week 7: 1 of 40 samples with growth (2.5%)Week 8: 2 of 40 samples with growth (5.0%)Week 9: 0 of 40 samples with growth (0.0%)Week 10: 2 of 40 samples with growth (5.0%)Week 11: 1 of 40 samples with growth (2.5%)Week 12: 0 of 40 samples with growth (0.0%)Week 13: 2 of 40 samples with growth (5.0%)Week 14: 1 of 40 samples with growth (2.5%)Week 15: 2 of 40 samples with growth (5.0%)Week 16: 0 of 40 samples with growth (0.0%)Week 17: 2 of 40 samples with growth (5.0%)Week 18: 1 of 40 samples with growth (2.5%)Week 19: 0 of 40 samples with growth (0.0%)Week 20: 2 of 40 samples with growth (5.0%)Week 21: 3 of 40 samples with growth (7.5%)Week 22: 2 of 40 samples with growth (5.0%)Week 23: 4 of 40 samples with growth (10.0%)Week 24: 1 of 40 samples with growth (2.5%)Week 25: 3 of 40 samples with growth (7.5%)Week 26: 4 of 40 samples with growth (10.0%)Week 27: 2 of 40 samples with growth (5.0%)Week 28: 3 of 40 samples with growth (7.5%)Week 29: 5 of 40 samples with growth (12.5%)Week 30: 2 of 40 samples with growth (5.0%)Week 31: 1 of 40 samples with growth (2.5%)Week 32: 4 of 40 samples with growth (10.0%)Week 33: 3 of 40 samples with growth (7.5%)Week 34: 1 of 40 samples with growth (2.5%)Week 35: 4 of 40 samples with growth (10.0%)Week 36: 4 of 40 samples with growth (10.0%)Week 37: 3 of 40 samples with growth (7.5%)Week 38: 5 of 40 samples with growth (12.5%)Week 39: 2 of 40 samples with growth (5.0%)Week 40: 3 of 40 samples with growth (7.5%)Week 41: 2 of 40 samples with growth (5.0%)Week 42: 4 of 40 samples with growth (10.0%)Week 43: 0 of 40 samples with growth (0.0%)Week 44: 3 of 40 samples with growth (7.5%)Week 45: 4 of 40 samples with growth (10.0%)Week 46: 2 of 40 samples with growth (5.0%)Week 47: 4 of 40 samples with growth (10.0%)Week 48: 2 of 40 samples with growth (5.0%)Week 49: 3 of 40 samples with growth (7.5%)Week 50: 5 of 40 samples with growth (12.5%)Week 51: 1 of 40 samples with growth (2.5%)Week 52: 3 of 40 samples with growth (7.5%)shift beginstrend limit 4.5%Trend alert, week 23trend alertFirst action-level excursion, week 29action level
  • Weekly recovery rate
  • 6-week trend
  • Trend limit from baseline
  • First action-level excursion
Illustrative counts. Action levels are set per sample and catch a bad day; the contamination recovery rate, trended against a limit set from the site's own baseline, catches a worse month. Annex 1 asks for both, with the trend reviewed as part of the contamination control strategy. A tighter window reacts faster and alarms more often; the site picks the window from its baseline data.

How QLogixCMC helps

QLogixCMC writes or remediates the contamination control strategy from the site's risk assessments and data, sets alert and action levels and trend limits from the baseline, designs the environmental monitoring program and its review, prepares the aseptic process simulation protocol, and runs mock inspections that read the CCS the way an inspector does. The site's qualified people own the strategy and sign every limit.

Questions this post answers

What has to be in a contamination control strategy?

The controls the site relies on to keep microbial, pyrogen and particulate contamination out of its products, the risks each control addresses, the monitoring that shows the controls work, and the process for reviewing and updating the strategy. Annex 1 lists the elements it expects to see covered, from facility design and utilities to personnel, process, sterilization and monitoring.

How are alert and action levels set?

From the site's own monitoring data, reviewed periodically, within the maximum limits the annex sets per grade. A level copied from the annex's table is a ceiling, not a site's alert level; a level set from the baseline is what makes a trend meaningful.

Does Annex 1 apply to US sites?

Directly only to products made for the EU market and to sites inspected by EU authorities. In practice its expectations, in particular the contamination control strategy, are read alongside FDA's aseptic processing guidance, and FDA inspectors ask similar questions of the same controls.

Talk it through with someone who has run a plant

Bring your backlog, your submission timeline or the Form 483. We will tell you plainly what we would do first, what it would take, and what we would leave alone.

In the first meeting we cover

  • What is on your mind: a backlog, a submission, a validation program, an inspection date or a system that is not working
  • What we would assess first, and what we would leave alone
  • Whether an engineer on your site, a fixed-scope assessment or a defined deliverable is the right way to start