FDA Warning letter to Dupont cites Inadequate OOS Investigations

Failure to thoroughly investigate OOS results in a timely manner, appropriately identify root causes, expand investigations to all potentially affected lots, and implement adequate CAPA. 

Firm had failing results for conductivity in Avicel lots. Laboratory OOS results were only investigated by the original analyst using a checklist. The supervisory review did not include an evaluation of the records and test data; failed to expand the investigation to production and other potentially affected lots. Investigation also lacked sufficient evidence to determine the root cause and identify CAPAs.

Firm did not perform a timely and thorough investigation into an inaccurate conductivity meter reading. Investigation revealed a probe “encrusted with grime/resin” caused lower conductivity values. Although OOS results were identified in November 2020, firm did not expand investigation in a timely manner to determine the scope of potentially impacted lots tested using this meter.

It was determined the root cause of the conductivity OOS values was related to elevated levels of ammonium chloride in the Avicel. FDA is concerned as elevated levels of ammonium chloride in excipients has the potential to lead to impurity formation in finished drug products. Of note, such impurity formation could include nitrosamines. For FDA’s current thinking on this topic, see FDA’s guidance, Control of NitrosamineImpurities in Human Drugs at https://www.fda.gov/media/141720/download, (https://www.fda.gov/media/141720/download)

Firms response stated, firm conducted a review of laboratory OOS investigations for the past year; however, did not provide a rationale to support this limited time period versus the standard expiration period of drugs. In addition, firm did not expand this review to manufacturing investigations. Your response also lacks a commitment or details on how to address released excipient associated with process deviations or OOS investigations.

In response to Warning letter, firm should provide:

• A retrospective, independent review of all invalidated OOS (including in-process and release / stability testing) results for excipients currently in the U.S. market and within expiry as of the date of this letter and a report summarizing the findings of the analysis, including the following for each OOS:
  
  • Determine whether the scientific justification and evidence relating to the invalidated OOS result conclusively or inconclusively demonstrates causative laboratory error.
  • For investigations that conclusively establish laboratory root cause, provide the rationale, and ensure that all other laboratory methods vulnerable to the same or similar root cause are identified for remediation.
  • For all OOS results found by the retrospective review to have an inconclusive or no root cause identified in the laboratory, include a thorough review of production (e.g., batch manufacturing records, adequacy of the manufacturing steps, suitability of equipment/facilities, variability of raw materials, process capability, deviation history, complaint history, batch failure history). Provide a summary of potential manufacturing root causes for each investigation, and any manufacturing operation improvements.
• A comprehensive review and remediation plan for your OOS result investigation systems. The CAPA should include, but not be limited to addressing the following:
  
  • Quality unit oversight of laboratory investigations
  • Identification of adverse laboratory control trends
  • Resolution of causes of laboratory variation
  • Initiation of thorough investigations of potential manufacturing causes whenever a laboratory cause cannot be conclusively identified
  • Adequately scoping of each investigation and its CAPA
  • Revised OOS investigation procedures with these and other remediations

Failure to thoroughly investigate OOS results in a timely manner, appropriately identify root causes, expand investigations to all potentially affected lots, and implement adequate CAPA. 

Firm had failing results for conductivity in Avicel lots. Laboratory OOS results were only investigated by the original analyst using a checklist. The supervisory review did not include an evaluation of the records and test data; failed to expand the investigation to production and other potentially affected lots. Investigation also lacked sufficient evidence to determine the root cause and identify CAPAs.

Firm did not perform a timely and thorough investigation into an inaccurate conductivity meter reading. Investigation revealed a probe “encrusted with grime/resin” caused lower conductivity values. Although OOS results were identified in November 2020, firm did not expand investigation in a timely manner to determine the scope of potentially impacted lots tested using this meter.

It was determined the root cause of the conductivity OOS values was related to elevated levels of ammonium chloride in the Avicel. FDA is concerned as elevated levels of ammonium chloride in excipients has the potential to lead to impurity formation in finished drug products. Of note, such impurity formation could include nitrosamines. For FDA’s current thinking on this topic, see FDA’s guidance, Control of NitrosamineImpurities in Human Drugs at https://www.fda.gov/media/141720/download, (https://www.fda.gov/media/141720/download)

Firms response stated, firm conducted a review of laboratory OOS investigations for the past year; however, did not provide a rationale to support this limited time period versus the standard expiration period of drugs. In addition, firm did not expand this review to manufacturing investigations. Your response also lacks a commitment or details on how to address released excipient associated with process deviations or OOS investigations.

In response to Warning letter, firm should provide:

• A retrospective, independent review of all invalidated OOS (including in-process and release / stability testing) results for excipients currently in the U.S. market and within expiry as of the date of this letter and a report summarizing the findings of the analysis, including the following for each OOS:
  
  • Determine whether the scientific justification and evidence relating to the invalidated OOS result conclusively or inconclusively demonstrates causative laboratory error.
  • For investigations that conclusively establish laboratory root cause, provide the rationale, and ensure that all other laboratory methods vulnerable to the same or similar root cause are identified for remediation.
  • For all OOS results found by the retrospective review to have an inconclusive or no root cause identified in the laboratory, include a thorough review of production (e.g., batch manufacturing records, adequacy of the manufacturing steps, suitability of equipment/facilities, variability of raw materials, process capability, deviation history, complaint history, batch failure history). Provide a summary of potential manufacturing root causes for each investigation, and any manufacturing operation improvements.
• A comprehensive review and remediation plan for your OOS result investigation systems. The CAPA should include, but not be limited to addressing the following:
  
  • Quality unit oversight of laboratory investigations
  • Identification of adverse laboratory control trends
  • Resolution of causes of laboratory variation
  • Initiation of thorough investigations of potential manufacturing causes whenever a laboratory cause cannot be conclusively identified
  • Adequately scoping of each investigation and its CAPA
  • Revised OOS investigation procedures with these and other remediations

The FDA observation 2 in the Dupont warning letter is very much a continuation of the Observation 1, but focusing on OOS investigations and lapses. The observations has several pointers.

Inadequate OOS investigation and investigation procedure, inadequate depth of investigation – lack of review of OOS investigation and reports by lab supervisor, lack of timely investigation, failure to extend investigation into manufacturing process and potential failures where conclusive laboratory error is not identified, lack of Quality unit oversight into OOS investigations,

Inadequate impact assessment and remediation measures due to the OOS and identified causes for the OOS; The investigation did not look whether other batches, products could have impacted similarly and also actions to be taken on impacted batches.

Inadequate assessment and response of the FDA inspectional observation of lapses in OOS investigation  –  definition of scope, lack of rationale for the time period considered for assessment (assessment considered one year instead of defining time period based on expiry of all distributed batches), commitment to assess all OOS and other similar deviations which could have an impact all distributed batches and take remediation actions.

As can be seen from the FDA warning letter manufacture of Excipients need to comply with applicable cGMPs as per 21 CFR 210&211 and other applicable regulations just as the drug product and drug substance manufacturers.

Adequate procedure for investigation of OOS should be in place. The OOS investigation procedure should be comprehensive covering laboratory errors & lapses in manufacturing and controls. The OOS investigation procedure should be detailed and in compliance to the USFDA (Investigating Out-of-Specification (OOS) Test Results for Pharmaceutical Production) & MHRA ( Out-of-specification investigations) guidelines. The procedure should address in a systematic manner:

• Phase 1A investigation – obvious Laboratory errors
• Phase 1B investigation – extended investigation into potential laboratory errors
• Phase 2 – full scale investigation covering production process, sampling methods, additional laboratory investigation
• When laboratory investigation is not conclusive, there should always be a production process investigation to determine whether the OOS is due to lapses/concerns in production process.
• Clearly defined responsibilities of laboratory analyst (following the test methods and procedures fully, reporting the OOS promptly, documenting obvious errors), laboratory supervisor (review of the OOS results, test solutions, instruments, data and documents, analyst interview to check knowledge and procedures followed), Quality unit (review of laboratory investigation, extend OOS investigation to production process and controls, evaluation of root cause and CAPA identified and their adequacy, impact assessment and remediation action, conclusion of investigation, batch release decision) in OOS investigation. There should be appropriate checklists to document all aspects of investigation to ensure no areas are missed.
• Criteria for retesting and resampling.
• Extension of investigation to other potential areas which could be similarly impacted (e.g if the root cause was in inadequate calibration or maintenance of specific laboratory instrument, whether similar situation can occur in other instruments ad methods)
• Impact assessment (whether other batches, products could be similarly impacted) and remediation actions
• Review of implementation of all remediation measures and CAPAs
• Tracking of OOS trends, periodic review of trends to see recurrence of OOS incidents
• Effectiveness evaluation of CAPA. (If CAPAs are effective, the incidence should not recur).

Investigations should be performed by cross functional teams. Appropriate investigation tools should be used for the investigation (For e.g – Fish bone analysis, Fault tree analysis, Why-Why techniques). The investigations should focus on identifying root cause(s) and corrective actions which will address all the root cause. For example, if the cause for the conductivity failure complaints was erroneous readings from conductivity meter, this is still not the root cause. Root cause should look at why the product batches failed in conductivity parameter, for e.g the batches had higher levels of ammonium chloride – but this still is not the root cause; the investigation should look at why the batches had higher ammonium chloride. And the CAPAs should address all the root causes – how failures of conductivity meter will be prevented in future, what manufacturing process improvements or control measures will be introduced to prevent elevated levels of ammonium chloride in batches.

When an investigation identifies the OOS is valid, and root cause is identified, potential impact on other batches and even other products should be assessed. The scope of such assessments should cover all batches which are in distribution and live inventory.

The investigations should be well documented. If the investigations conclude that there are no other impacted batches this should be done supported with sound scientific rationale.

When deficiencies are identified in regulatory inspections, the observations should be comprehensively assessed and responded addressing all concerns raised by the inspector. Such assessments should be comprehensive, performed against a defined protocol, with well-defined scope. The time period for retrospective review of batches and impact should be well defined, and should at least cover all distributed batches within expiry (For e.g. if the oldest batch under distribution was manufactured 3 years before, the time period for assessment should be at least 3 years, not arbitrary time periods like 1 year or 2 years).

The investigations should be performed with objective of identifying extent of deviations, root causes, impact of the deviations and remedial actions, corrective and preventive actions (CAPA) and should also be well documented. Such investigations and conclusions should be with sound scientific rationale, shall not contradict existing information and observations – most importantly the conclusions shall not be preconceived. 

Engage competent independent cGMP expert consultant to review and assess the overall system for failure investigations and OOS events. The assessment should also cover all invalidated OOS results of batches within expiry (distributed as well as inventory). Based on the assessment remediation actions on impacted batches shall be taken (customer intimation, recall).

The comprehensive assessment of past invalid OOS should be performed against a defined protocol and take remediation measures based on assessment (customer intimation, recall). A comprehensive matrix for assessment and report shall be developed, addressing:

• Stage of the OOS – Inprocess, release, stability
• Whether a laboratory error is conclusively identified.
• Extension of investigation into production process and controls (where conclusive laboratory error is not identified) covering:
  
  • Review of batch manufacturing record to identify any gaps / abnormalities
  • Adequacy of manufacturing steps
  • Equipment suitability, facility suitability
  • Input materials quality / concerns, variability that can contribute to failures or deviations
  • Process deviations
  • Previous history of similar OOS, batch failure
  • Complaint history (Similar incidents, issues reported)

Where production process concerns are identified, appropriate CAPAs to be established including improvement in manufacturing process.

Make a list of all laboratory failures identified during the above assessment and review whether similar failure could have impacted other laboratory methods, procedures, instruments. Document the assessment; Any conclusion that there is no impact on other laboratory methods or procedures, should be with sound scientific rationale. Take remediation measures for all the methods which could be similarly impacted (measures could include but not limited to – improvement of the test methods, calibration procedures, maintenance procedures).

Identify and implement all the corrective actions (CAPAs) for the sited OOS – address laboratory instrument/ conductivity meter issues, manufacturing process issues 

Comprehensively review the OOS procedure addressing all OOS investigation requirements in line with USFDA / MHRA guidance (As detailed under What companies should have..).

Train and coach the laboratory personnel, other concerned operating personnel (production, engineering, R&D), Quality unit personnel in cGMPs , regulatory guidances, and OOS procedure and investigation. 

Enhance Quality oversight – both by Quality Unit as well as Senior / Leadership Management. Periodic Management review meetings (ideally monthly) with defined agenda where Quality events, Open issues, Effectiveness of CAPAs should get reviewed; further actions identified and followed up through subsequent reviews.