Failure to maintain complete data derived from all laboratory tests conducted to ensure compliance with established specifications and standards
The firm inspected lacked accurate raw laboratory data records for batches of API that were shipped. The inspection revealed that batch samples were retested until acceptable results were obtained. In addition, Quality Control (QC) laboratory failed to include complete data on QC testing sheets. Failing or otherwise atypical results were not included in the official laboratory control records, not reported, and not investigated.
According to laboratory analysts interviewed by FDA during the inspection, the common practice followed was to complete the analysis and to record the sample preparation data only if the results were acceptable. If the results obtained were atypical, a fresh sample was to be prepared and analyzed. The original sample testing was not recorded.
Investigations performed by FDA during the inspection conveyed a general lack of reliability and accuracy of data generated by the firm's laboratory, which is a serious cGMP deficiency that raised concern on the integrity of all data generated.
Outcome - FDA was very concerned that the firm's laboratory allowed the practice of retesting for several methods without appropriate documentation, justification, and investigation. Hence FDA strongly advised investigating these data integrity issues and identify the extent these practices were followed in the laboratory and manufacturing operations as part of a comprehensive data integrity audit. In the response to follow, FDA asked to provide copies of the procedures in place that set forth the requirements to review and preserve complete data generated from these operations.
Failure to investigate and document out-of-specifications results
The firm had failed to investigate unknown peaks found during HPLC testing for related compounds of API. A reviewer of the raw data had reported on the “finished product report review data” worksheet that unknown peaks were observed due to vial contamination. The electronic records indicate that the first analysis performed failed the specification limit for both any single unknown impurity. These OOS results were not reported or adequately investigated, and the raw data was discarded. The sample was reanalyzed at which time it met specifications and the results were recorded. According to a laboratory analyst, the sample preparation printout corresponding to the initial testing was destroyed.
Investigation of the data from one of the firm's software (LIMS) identified instances where additional testing was performed but not properly documented in laboratory records. This investigation was limited in scope to only a short timeframe and to only one type of laboratory instrumentation, HPLC. During FDA’s inspection, the QC manager explained that the scope of the risk assessment was limited only to that particular month because he was busy with other laboratory responsibilities. According to the firm's response to the Form FDA 483, some of the corrective actions implemented as a result of this investigation include: retraining of QC analysts, revision of the laboratory incident investigations SOP, and enhancements to the documentation and sample handling practices. As failures to investigate and document OOS results have persisted, it is clear that the corrective actions were not sufficient.
The firm's management had failed to prevent the practices of product sample retesting without investigation, and rewriting and/or omission of original CGMP records persisted without the implementation of controls to prevent data manipulation. The firm’s response to the Form FDA 483 acknowledged the deficiencies regarding data integrity observed during this inspection. Nevertheless, the firm’s health hazard evaluation “Drug Safety Analysis” conducted in response to the Form FDA 483 concluded that there was no effect on product quality or patient safety. However, this evaluation was based on unreliable and incomplete data, as undesired records appear to be excluded.
Outcome - In response to the warning letter, the firm was asked to provide documentation of all corrective actions taken to address these failures to initiate investigations as required by the firm's procedures and determine root cause(s) of OOS results. The failure to perform adequate investigations was awarded as a repeat observation reported to the facility.
Failure to include adequate documentation during compliant investigation
The firm had failed to investigate failing assay results. The original investigation, approved on 29 March 2013 indicated that the complaint was received on February 26, 2013. During the review of that investigation, the investigators found a test record from January 8, 2013 (prior to the date of documented receiving the complaint) that reported failing HPLC assay results of the retain sample of batch (X). This record was not included in the investigation report.
The firm’s response acknowledges that the assay result for the retain sample was omitted, but also claims additional investigations are ongoing and that preventive and corrective actions will be implemented as appropriate. However, the response to the previous 2010 inspectional findings included a similar corrective action plan regarding OOS, deviations, stability, product complaint and CAPA investigations. In this plan, the firm committed “to address this omission and provide assurances that the scope and detail related to future investigations is appropriately documented … to ensure consistent consideration for failure investigations.
Outcome : Evidence should be provided supporting the additional investigation conducted and the corrective actions implemented to prevent the omission of data. Provide records of all complaints relating to the APIs, including returned API and the disposition of each returned batch. Discuss the expansion of your investigation to other batches and APIs that could be affected by failing assay results. Furthermore, explain the failure of firm’s complaint system and how it going to implement proper management oversight to ensure adequate corrections to this deficiency.
Failure to record activities at the time they performed
The firm had failed to record the data on cGMP document at the time the activity is performed. Data related to laboratory testing had been recorded days after the testing was performed. The secondary reviewer of these data points had marked the listings as corrected and this in turn implied that the data was not recorded contemporaneously.
Though the firm had procedures in place to control quality and despite the fact that they had re-trained personnel on it, US FDA had expressed concern over the credibility and capability of the document. This is because US FDA had observed instances of data that was back-dated.
The inspection revealed that the firm selectively omitted cGMP records directly related to testing and manufacturing of products. The observations in this context directly question the accuracy and completeness of the data recorded.
Outcome: US FDA advised implementing adequate controls and systems to prevent omission, manipulation of laboratory data as critical responsibility. They also advised that the plan should also ensure that controls are put in place to prevent unauthorized changes to existing data. Any changes to the data should occur with strict accordance with approved established procedures, with date of change and identity of the person who made the change and the explanation or the reason for change should also recorded. Also as a corrective action, training is imparted to all the managers, supervisors, quality unit personnel in detecting lacunae related to doubts in integrity and/or manipulation of data.
The firm had failed to maintain written production, control, or distribution records specifically associated with a batch of a drug product for at least one year after the expiration date of the batch
During the inspection, the investigators found approximately 10 waste bags in the facility that contained torn or partially destroyed raw data CGMP records related to a variety of manufacturing activities. Some of the records found in these waste bags included the following:
A. A calibration check record for balance was torn and partially destroyed. According to associate interviewed by FDA during the inspection, wrong weights were used for calibration. Recalibration of the balance was performed with new documentation, and subsequently discarded the original record. Furthermore, auditors learned that additional original calibration records of other balances had similarly been discarded.
B. Six corrective action and preventive action (CAPA) records) were torn. According to Senior QA officer interviewed by FDA during the inspection, these forms were used for extending the due date of an ongoing CAPA. Inspection team compared the discarded records to the official records and identified corresponding official copies of only three of the records. The three other discarded records did not have an official corresponding copy. During the inspection, the firm could not produce official records of the corrective actions described in these three partially destroyed documents.
C. Five completed preventive maintenance forms were torn. According to staff member interviewed by FDA during the inspection he mistakenly tore and destroyed these original records.
D. Investigations performed by FDA during the inspection conveyed a general lack of basic oversight by operations, quality unit, and site managers, as rewriting and destruction of original CGMP records was allowed to persist over a significant period without implementation of systems and controls to prevent data manipulation, which is a serious CGMP deficiency that raised concern on the integrity of all records generated.
OUTCOME: FDA was very concerned that the firm allowed the practice of destruction of original records without appropriate documentation, justification, and investigation. In the response to follow, FDA asked to provide:
• Third party auditor’s report of the investigation of the data integrity practices associated with Firm’s CGMP records along with a list of all records that Firm’s employees rewrote, destroyed, or altered in any way.
• The root cause of the firm’s failure to control and detect the manipulation, alteration, or premature destruction of CGMP records and describe systemic actions to prevent recurrence.
• Copies of procedures to manage and retain all CGMP records.
• The list of all the batches of drug products shipped to the U.S. market that relied upon missing, inaccurate, or unreliable records.
Failure to prepare batch production and control records for each batch of drug product that include documentation of each significant step in the manufacture, processing, packaging, or holding of the batch
A. The inspection revealed “unofficial” visual inspection records, signed by production personnel, with data that is different from the official batch records reviewed by the firm’s quality unit. The unofficial record completed by production personnel showed 200 units failing to meet specifications. Production personnel later completed the official batch record for this batch, showing only 18 units as having been rejected. During the inspection, the firm was unable to demonstrate that all units with quality defects were in fact rejected.
The firm’s response states that, in many cases, production personnel add extra units and lower the number of rejected units on the official paperwork to account for these extra units. This explanation is unacceptable for several reasons, including that this practice does not accurately represent the number of units with quality defects present in each batch in the official batch records, it obscures the number of rejected units in any given batch, and it misrepresents the number of units sterilized during each batch. The firm’s response states that they have audited a random selection of 848 batch records and found a difference between unofficial and official records in 2.5% of instances. However, during the inspection, investigators found discrepancies between official and unofficial records in 76 of 156 batches reviewed. The Firm’s response does not include an explanation as to the differences between internal audit results and FDA inspectional findings.
B. The inspection revealed use of scratch paper containing critical manufacturing data. The data on these scratch paper records did not always match the data on the corresponding official batch records, as in the case for the amount of raw materials added to Batch. Although the firm stated that this batch was destroyed on October 18, 2013, the investigators observed that firm’s records showed that the batch was removed from quarantine on October 25, 2013.
The firm’s response to this finding does not explain why production personnel used scratch paper for documenting CGMP-relevant data. In addition, the discrepancy between destruction records and the quarantine records provide further evidence that the firm’s documentation is not accurate and reliable. The use of unofficial and scratch paper records is not acceptable with CGMP.
C. Employees interviewed during the inspection admitted that the firm recorded activities in batch records that were not performed. Specifically, the firm’s head of production reported to our investigator that he completes “in process quality assurance check” fields in the batch record but does not actually perform the listed operations.
Failure to maintain adequate written records of major equipment maintenance
FDA investigators identified two maintenance logbooks that included multiple entries describing significant equipment malfunctions, but for which no investigation into the potential effect on product quality was performed. In addition, the firm’s records do not always include information on repairs following these malfunctions. For instance, no maintenance actions or product impact investigations were recorded for out-of-limit findings during equipment calibration. In other cases, the firm determined that there was no product impact without justification. In addition, auditors noted that ten serialized entries had been torn out of the logbooks. The firm’s staff could not locate these records during the inspection and reported to FDA investigator that the entries had likely been destroyed.
OUTCOME : FDA requested the firm to hire a third party auditor, with experience in detecting data integrity problems, to assist the firm with this evaluation and assist in overall compliance with CGMP. It is the responsibility of the firm to ensure that data generated during operations is accurate and that the results reported are a true representation of the quality of the firm’s drug products. In addition FDA requested,
(a) To provide an updated assessment of the extent of the differences between unofficial and official records used at the facility.
• To describe controls in place to prevent data manipulation by your operators and supervisors.
• To Interview current and former employees to identify activities, systems, procedures, and management behaviors that may have resulted in or contributed to inaccurate data reporting in CGMP records.
(b) To provide assurance that the use of unofficial and scratch paper records has been discontinued and describe how the firm will prevent this practice in the future.
• Also describe the procedure to assure that all CGMP-related operations are documented at the time of occurrence.
• Identify all instances in which unofficial and scratch paper records have been used in the manufacturing and laboratory facilities and assess the possibility of misrepresentation of data on official records in each case.
(c) To describe firm’s investigation for batch record entries, outlining the efforts to determine the scope of data falsification within batch records and also the corrective and preventive actions.