An FDA Warning Letter to Indian API manufacturer Micro Orgo Chem describes non-dedicated equipment that had never been cleaned under a validated procedure, cleaning SOPs that were not followed, and no evidence that the rinse and swab samples required by the firm's own procedure were ever collected or tested.
What happened
The Warning Letter identifies five significant CGMP deviations for active pharmaceutical ingredients, recorded during an FDA inspection of Micro Orgo Chem's facility at 57, C-1/B, LIC sector, GIDC, Vapi, 396195, Gujarat (FEI 3004513977), conducted 11 to 18 April 2024. It carries Reference # 320-25-21, is dated 3 December 2024, and was issued to Managing Director Mr. Vinod Ranka. The first deviation is headed "Failure to adequately validate written procedures for the cleaning and maintenance of equipment." The Warning Letter identifies that the firm "failed to adequately validate cleaning procedures for your non-dedicated manufacturing equipment to ensure it is fit for its intended use." It also records that quality assurance personnel told investigators: "your firm has not carried out cleaning validation for non-dedicated manufacturing equipment routinely used for the manufacture of intermediates and APIs for the U.S. market."
The gap was not only the absent validation study. The letter states the firm "failed to follow established cleaning procedures" and was "unable to demonstrate rinse and swab samples were collected and tested as required by your procedure." Equipment identified as (b)(4)-03 "appeared dirty and not appropriately maintained." FDA drew the consequence explicitly: "A failure to adequately clean and maintain non-dedicated equipment may lead to cross-contamination between drug products."
The firm's May 2024 response did not close the observation: it contained neither a cleaning validation protocol nor a revised maintenance plan showing the APIs would be free from cross-contamination. The firm committed to a change control and to training, but supplied no underlying science. The other four observations compound the picture — no process performance qualification records for any API process, inadequately qualified equipment, unvalidated in-house test methods, no batch of any API placed on stability in 2021–2023, and a missing batch record for a batch already shipped to the U.S. FDA recommended a qualified CGMP consultant and placed the firm on Import Alert 66-40 on 22 October 2024.
Why it matters
A cleaning failure is usually framed as an execution problem: the procedure existed but was not followed. This letter sits a level deeper. Routine manufacture of intermediates and APIs for the U.S. market ran on equipment for which no cleaning validation had ever been performed, and the rinse and swab data the firm's own procedure required could not be produced. Without either the validation or the documentary trail, no one — firm or regulator — can bound the cross-contamination risk of batches already released. That is why FDA demanded a retrospective assessment of cleaning effectiveness, rather than a promise to write a protocol going forward.
What is particularly interesting
The firm had a cleaning procedure requiring rinse and swab sampling. The failure was not that nobody thought sampling mattered; the requirement existed on paper and could not be shown to have been executed. A documented sampling requirement silently not performed invites reliance on records that do not exist.
Equally notable is the remedy. FDA did not simply ask for "cleaning validation" — it asked for a programme incorporating worst-case conditions, and enumerated them: higher toxicities, higher potencies, lower solubility in the cleaning solvent, characteristics making a drug difficult to clean, swabbing locations for areas most difficult to clean, and maximum hold times before cleaning. Hold time before cleaning appears as a worst-case dimension in its own right, not an afterthought. FDA also asked for a comprehensive evaluation of the facility cleaning and maintenance programme with CAPA.
The letter says nothing about acceptance criteria, MACO calculation or analytical recovery — instructive, because the observation never reached the quantitative residue-limit discussion. The basic validation and sampling evidence did not exist.
How such a gap can realistically develop
A finding of "never validated" rarely begins as a decision to skip validation. More common routes: shared equipment is commissioned for a new product without extending the cleaning validation programme to the new equipment train or residue; a site pursues product approvals under time pressure and treats cleaning as campaign execution rather than a validated process; a facility anticipates a relocation and defers remediation to "the new facility" while the current one keeps running on unqualified procedures. The letter shows traces of the last pattern — stability chambers "recently moved," deregistration pending qualification. Deferring validation to a future facility does not neutralise the obligation for product already distributed from the present one.
Questions worth asking in our own programme
- Can we produce the cleaning validation report — not the procedure — for every non-dedicated equipment train in routine use, naming the equipment, residues and worst-case product selected?
- If our procedure requires rinse and/or swab sampling, can we show actual results by cleaning event for the last twelve months, tracing one cleaning record to its supporting sample data?
- Is our worst-case rationale documented against toxicity, potency, solubility in the cleaning solvent, difficulty of cleaning, hardest-to-clean swab locations and maximum hold time before cleaning?
- Do we know our maximum hold time before cleaning, and do we have data supporting the microbial control claim for that period?
- Has every piece of equipment used for more than one product been included in the cleaning validation scope, including equipment repurposed since the last validation?
- If equipment had never been validated, could we reconstruct the exposure of already-distributed batches, and would our quality unit know how to scope that?
Wider context
FDA's requested corrective actions are a recognisable template applied to API and finished-dose manufacturers for years. The value of such letters is the checklist they force onto other sites. Any facility with shared, product-contact equipment and more than one product in its portfolio can run the same questions against its own documentation, without waiting for an inspection to surface the answer.
Sources
- U.S. Food and Drug Administration — Warning Letter 320-25-21, Micro Orgo Chem, Reference # 686458, 3 December 2024 (primary source; accessed via full-text mirror below).
- EIN Presswire — full reprint of Warning Letter 320-25-21, 17 December 2024 (complete letter text).
- ECA Academy (GMP Compliance) — "Warning Letter issued for an Indian API facility," 15 January 2025.
- IGMPI — "FDA Flags Cleaning Validation Issues at Micro Orgo Chem," 17 March 2025.
Validation takeaway
The lesson of Warning Letter 320-25-21 is that a cleaning procedure never tested against a validated, worst-case-based programme leaves no defensible answer for product already shipped. A sampling requirement written into an SOP that cannot be shown to have been executed is not a control — it is an unverified claim. Sites with shared, product-contact equipment should be able to produce, on demand, the cleaning validation report for each equipment train, the sample data behind each routine cleaning event, and a scientific rationale for the worst-case residue, the hardest-to-clean swab locations and the maximum hold time before cleaning. Where that evidence does not exist, the honest position is a retrospective cleaning effectiveness assessment covering all equipment used for more than one product — the same remedy FDA requested here.
Questions worth asking in our own programme
- Can we produce the cleaning validation report, not just the cleaning procedure, for every non-dedicated equipment train currently in routine use, naming the equipment, the residues and the worst-case product selected?
- If our cleaning procedure requires rinse and/or swab sampling, can we show the actual results per cleaning event for the last twelve months, and can an inspector trace a single cleaning record to its supporting analytical data?
- Is our worst-case product rationale documented against higher toxicity, higher potency, lower solubility in the cleaning solvent, difficulty of cleaning, hardest-to-clean swab locations and maximum hold time before cleaning?
- Do we know and have data for our maximum hold time before cleaning, including the microbial control justification for that period?
- Has every piece of equipment used to manufacture more than one product been included in the cleaning validation scope, including equipment added, repurposed or transferred since the last validation?
- If equipment had historically never been cleaning-validated, could we reconstruct the exposure of already-distributed batches, and does our quality unit have a defined method for scoping that assessment?
