Mylan Laboratories, India Receives Warning Letter (08/06/2015)

Significant Microbiological Issues were Observed

The U.S. Food and Drug Administration (FDA) inspected the following three pharmaceutical manufacturing facilities between August 2014 and February 2015.

1. February 6-13, 2015: Mylan Laboratories Limited OTL, Plot No. 284-B (19A) Bommasandra Jigani Link Road, Ind. Area, Anekal Taluk, Bangalore, 560 105
2. September 23, 2014 through October 3, 2014: Agila Specialties Private Ltd., Specialty Formulation Facility (SFF) 19A, Plot No. 284-B/1 Bommasandra Jigani Link Road, Anekal Taluk, Bangalore 560 105
3. August 1-8, 2014: Agila Specialties Private Ltd., Sterile Product Division, Opp II M, Bilekahalli, Bannerghatta Road, Bangalore, Karnataka, 560 076

At all three sites, we identified significant violations of current good manufacturing practice (CGMP) regulations for finished pharmaceuticals, Title 21, Code of Federal Regulations, Parts 210 and 211.

Specific violations were observed during the inspections, including, but not limited to, the following. This Blog addresses only the Mylan site.

Mylan Laboratories Limited OTL (FEI: 3007512701)

1. Your firm failed to establish and follow appropriate written procedures that are designed to prevent microbiological contamination of drug products purporting to be sterile, and that include validation of all aseptic and sterilization processes (21 CFR 211.113(b)).
2. Non-integral (b)(4) gloves were used in Suites (b)(4) and (b)(4) for conducting aseptic processing operations.

For example, on February 12, 2015, we found that 15 of (b)(4) gloves in Suite (b)(4), and 4 of (b)(4) gloves from Suite (b)(4), were non-integral. (b)(4) gloves used for aseptic processing had tears and pin holes. Glove S2C8 had large cuts in two different fingers. Your firm was aware that non-integral (b)(4) gloves were being used in Suite (b)(4).

Additionally, certain records indicated that you were testing (b)(4) gloves for integrity, but the integrity data indicated testing for (b)(4) gloves. You did not follow your procedure PDN/039/R10, “Leak Testing of (b)(4) Glove” for testing glove integrity. You did not test each glove represented in your firm’s analytical data, as required by the SOP. Instead, you repeatedly used the results for your (b)(4) gloves to falsely represent the results of your (b)(4) gloves.

The same SOP, PDN/039/R10 “Leak Testing of (b)(4) Glove,” states that (b)(4) gloves are to be replaced after (b)(4) cycles. However, according to your “(b)(4) Log Sheet,” the(b)(4) gloves in use when we inspected Suite (b)(4) had been (b)(4) replaced after no more than (b)(4) cycles.

During the inspection, we reviewed environmental monitoring (EM) data that showed excursions in your ISO 5 area, which you attributed to gloves. Finally, during the inspection, we observed unidentified white particles on (b)(4) gloves exposed to critical areas inside the RABS.

(b)(4) gloves are worn during critical interventions such as making aseptic connections, clearing jams, clearing fallen vials, (b)(4) sterile primary and secondary closures, purging filling needles, adjusting equipment, and changing environmental monitoring plates. Because (b)(4) gloves are worn during these critical interventions, using non-integral gloves for aseptic processing is an unacceptable practice. It is a direct risk to product sterility. To minimize risks to sterile products, you should implement an adequate monitoring and maintenance program to identify and eliminate non-integral gloves.

COMMENT 

The above Observation speaks for itself. Why someone would use gloves that contain tears and pinholes suggests a complete lack of understanding of sterility.

1. There is a lack of assurance that you maintain your manufacturing environment in a state of control suitable for aseptic processing.

Environmental Monitoring

For example, you did not utilize environmental monitoring data to identify environmental control issues and identify appropriate follow-up actions. There were repeated out-of-action-level (OAL) results from microbial testing, but you did not examine the data for trends or take appropriate follow-up action.  Your SOP “No. MIP/047/R7 Microbiological Evaluation of Clean Rooms and Other Controlled Environments of Suite (b)(4) Area” describes OALs as (b)(4) CFU for setting plates inside the RABS (ISO 5) and (b)(4) CFUs for your ISO 6 area.

In 2014, you reported 375 OAL results (b)(4) or more CFUs in your ISO 6 area. Then, on January 31, 2015, you obtained OAL results of (b)(4) or more CFUs during the manufacture of (b)(4) Injection ((b)(4)) in Suite (b)(4). You also obtained OAL levels from the settle plates inside the RABS ((b)(4) CFU near the (b)(4)), and (b)(4) CFUs were recovered from the air sampling point near the (b)(4). From February 6-7, 2015, you obtained OAL results of (b)(4) or more colony-forming unit (CFU) in three critical ISO 5 areas of Suite (b)(4) during the manufacture of (b)(4) Injection ((b)(4)):

• (b)(4)—(b)(4) CFUs;
• (b)(4)—(b)(4) CFUs (fungi);
• (b)(4)—(b)(4) CFUs

COMMENT 

Microbiological OOS in an ISO 5 (RABS) environment is one CFU. Not only did the facility observe numbers greater than zero, but they did not identify environmental control issues and identify appropriate follow-up actions.

Personnel Monitoring

Deficiencies in your operators’ practices indicate that your manufacturing personnel monitoring program is deficient. For example, in your video of (b)(4) batch (b)(4)manufacturing, we observed an operator entering the RABS and in contact with (b)(4)gloves without sanitizing his gloved hands. These (b)(4) gloves were later worn for aseptic connections, purging filling needles, and interventions on the filling machine.  Furthermore, you do not monitor these operators when they exit the area, so you have no way to determine whether the operators who enter RABS without sanitizing compromise the aseptic environment.

Additionally, on video and in person, we observed employees with (b)(4) on their hands before EM checks. Sanitizing gloved hands just before sampling is unacceptable because it can prevent recovery of microorganisms. This undermines the reliability of personnel monitoring data.

We are also concerned about your failure to review the results of microbial tests to identify possible trending problems in environmental control in aseptic processing areas. Your response in this regard was inadequate because it did not include a retrospective assessment to ensure that microbial test results reported are reliable. In your response to this letter, please describe your evaluation of the potential effects of the OAL results and personnel monitoring deficiencies discussed above on the quality of products you have already manufactured and distributed. Also provide the improvements you have made to your written procedures to address these violations and describe any other actions you have taken or plan to take to correct the problems and prevent their recurrence.

COMMENT 

Monitoring of gloves upon leaving a classified environment is essential to assuring that personnel are operating properly in an ISO 5 environment. Using alcohol as a sanitizer just prior to an EM check defeats the reason for the check and suggests inadequate training on the part of the operators.

1. Aseptic garments worn in the filling area were also non-integral. We observed 7 of (b)(4) sterile gowns with tears or holes; 8 of (b)(4) had loose threads. We observed 2 of(b)(4) sterile hoods with tears or holes; 12 of (b)(4) had loose threads. We observed 8 of (b)(4) sterile booties with tears or holes; 11 of (b)(4) had loose threads.

Procedure PDN/013/R8 “Handling of Aseptic Area Garments” required production personnel to examine the garments for tears, holes, and loose threads, but our investigator found that these checks were not being performed.

In response to this letter, explain how your production management will ensure that gowns are suitable for aseptic processing in the future.

COMMENT 

Wearing aseptic garments in the ISO 5 environment containing tears or holes is unacceptable and totally defeats the rationale for wearing sterile gowns. Again, proper training remains essential.

2. Your firm failed to establish laboratory controls that include scientifically sound and appropriate specifications, standards, sampling plans, and test procedures designed to assure that components, drug product containers, closures, in-process materials, labeling, and drug products conform to appropriate standards of identity, strength, quality, and purity (21 CFR 211.160(b)).
3. You failed to follow your SOP MIA/007/R11 “Monitoring of Water for Microbiological Quality” for collecting water samples. During our inspection, we documented that the scheduled water drop points for (b)(4) and (b)(4) water were not sampled.

However, according to the analytical raw data work sheet, microbiological testing was performed for water sampling points (b)(4). We confirmed during the inspection that these samples were not, in fact, collected by your microbiologist.

COMMENT 

Falsification of data is a very serious violation.

1. Your environmental monitoring data is not reliable because of the materials you use to conduct EM tests.

On February 6, 2015, our investigator observed (b)(4) environmental monitoring plates previously incubated at (b)(4) C being used for surface and personnel monitoring. Three of (b)(4) plates showed signs of desiccation. Media was shrinking away from the edge of microbial plates.

On February 13, 2015, our investigator observed signs of drying on three of (b)(4) plates used for water samples and four of (b)(4) plates used for bio-burden.

These observations indicate that your media’s growth promotion potential may be compromised. In your response to this letter, inform us if you will be discontinuing this practice of pre-incubating plates.

COMMENT 

Previously incubating plates prior to use is a desired practice to assure plates are not contaminated prior to use. However, using overly dry plates will lead to compromised results.

Your firm failed to establish an adequate system for monitoring environmental conditions in aseptic processing areas (21 CFR 211.42(c)(10)(iv)). You do not have a scientific rationale for the environmental monitoring sampling locations in aseptic filling Suites (b)(4). You did not include factors such as smoke study findings, number and location of operators, and historical microbial data in your assessment of hazardous points.

For example, we found that settling plates are not appropriately placed in critical areas. Your smoke study showed that during set-up and filling, air flows toward the front (when (b)(4) open) or back of the RABS. However, two relevant sampling points were recently eliminated. As a result, these points of increased risk are not monitored.

COMMENT 

Performing a Risk Management analysis for determining the environmental monitoring locations is critical to assuring the proper placement of plates.

1. During our inspection, we noted that you have no justification for two different action levels for finger dab results. While you have an ISO 5 action level of (b)(4) CFU for set-up personnel, you use an ISO 6 action level of (b)(4) CFU for operators who do not routinely participate in aseptic processing operations using the RABS.

However, the inspection found that these “ISO 6 operators” made ISO 5 interventions, including within the (b)(4) laminar airflow hood (LAF) and the RABS. Notably, when >(b)(4) CFU was recovered from an “ISO 6 operator” who had accessed the RABS during an intervention, your firm did not consider the result to be outside the action limit.

Detecting sources of contamination during aseptic processing operations is critical to safeguard product sterility. In response to our inspection, you followed a Product Quality Assessment (PQA) protocol and found visible foreign particulate matter within your examined lots. In March and April, 2015, you voluntarily recalled seven “for Injection” lots of Gemcitabine 200 mg/vial, Gemcitabine 2g, Gemcitabine 1 g, Carboplatin 10 mg/mL, Methotrexate 25 mg/mL, and Cytarabine 20 mg/mL. In June, 2015, you expanded your recall to an additional eight lots of Gemcitabine and Methotrexate. However, other lots released into distribution may have been compromised by this manufacturing issue.

In response to this letter, send a progress report on your search for the root cause of this particulate contamination problem.  Describe your product quality assessment protocol, including analysis of retain samples, corrective actions and preventive actions, and risk assessments to evaluate the product quality effects of your inadequate aseptic processing activities and inadequate environmental monitoring program.

COMMENT 

When the FDA becomes the Quality Assurance Unit (QAU) as noted above, the firm is in “deep trouble”. The QAU should thoroughly examine why they did not observe visible foreign particulates on multiple lots. Was this a training issue?