We include forms 483 issued to four sites in the US, one in the UK and one in India. One of the sites where FDA cited observations in the area of potential cross contamination manufactures solid oral dosage forms, four manufacture sterile parenteral products and one manufactures APIs. One of the sites is a compounding pharmacy, a facility type to which FDA has issued more than 50% of their drug warning letters in the past three years. The forms 483 below are provided in chronological order.
Bayer Health Care LLC of Shawnee, Kansas received a 10-page form 483 at the end of an inspection on June 17, 2009. Thomas Arista was one of the investigators. FDA focused on deficiencies in media simulations, cleaning, environmental monitoring and airflow patterns. These include but are not limited to:
· The firm failed to follow exiting procedures regarding: time and temperature of incubation of media fill vials; discrepancies in the batch record when three media filled vials could not be accounted for.
· The design of the personnel entry into the facility does not protect against microbial contamination because: the locker rooms do not adequately segregate the areas of street clothes and facility uniforms and shoes; shelving is made of wood and is not cleaned or sanitized and further a buildup of dirt and dust was present on the shelves; sanitization / cleaning of factory shoes is not conducted and recorded on a routine schedule.
· Gram positive Bacillus species were identified in the facility EM program but no investigation was conducted nor was root cause for this recurrence identified. Gowning rooms are not included in the EM program. EM alert actions are based on historical data though action limits are not based on historical data. The non-viable particle monitoring probe is positioned above the work area without justification for this positioning.
· “Smoke studies” do not adequately demonstrate unidirectional airflow. Further “…the Microbiology Supervisor confirmed that they have not performed an assessment of the smoke studies, which are used to demonstrate unidirectional airflow within the Class 100 and Class 10,000 aseptic fill areas.”
· EM sampling sites are not scientifically justified.
· Aseptic processing areas are not adequately cleaned or disinfected. For example, the top areas of the curtain in Class 100 filling areas are not sprayed or wiped, and a scientific justification for this practice was not available.
· Sterile filling room cleaning was left at rest in excess of the duration identified in the company SOP.
· Records do not include the disposition of rejected components.
Ranbaxy Laboratories, Ltd site in Mohali, Punjab, India received an 8-page form 483 at the close of inspection on September 26, 2012. The site was placed on import alert in September, 2013. The observations focus on the manufacture of solid oral dosage forms. Most of the observations address events outside the scope of the topic of this article. But regarding the potential for cross contamination the following observations are relevant:
· The written procedures of cleaning of non-dedicated equipment are not written in adequate detail to ensure ‘consistent / reproducible cleaning results.’ Visible residue was visible in the air inlet and exhaust areas of cleaned and inspected process equipment.
· In the tablet bottling operations, the air filter equipment, Air Displacement Unit did not include HEPA filters ‘…to prevent the release and recirculation of dust created during the bottling operation, whereby the potential for cross-contamination may exist.”
Hospira, Inc (now Pfizer), in Rocky Mount, NC received a 22-page form 483 at the close of an inspection on March 1, 2013. A total of 20, often multi-part, observations were identified. Thomas Arista, Penny McCarver, and Jason Chancey were among the five investigators. The EIR from this inspection is available from FDAzilla also. The firm received a warning letter for an inspection at this site conducted later in the year, ending November 15, 2013. We provide only a selected number of items; it is not meant to be full summary of the form 483 content. The form 483 focuses on observations in the area of aseptic processing, including but not limited to:
· Anaerobic media fills are not performed under anaerobic conditions to mimic the drug product fills that are overlaid with [what I assume is] nitrogen.
· Media fills do not require that repeat manual interventions are performed the same number of times and for the same duration of time during media fills as they would be expected to be performed during routine manufacture.
· Not all manual interventions are included in media fill operations.
· Routine autoclave load configurations are not document to ensure they do not exceed the validated load configuration.
· There are no records to document the life cycle (number of laundry cycles) of gowning attire (scrubs, clean room gowning / coverall etc.) to ensure they are fit for use.
· Smoke studies to demonstrate unidirectional airflow are inadequate, seven examples are provided.
· The Quality Unit failed to ensure that all procedures and test methods comply with CGMPs.
· Environmental monitoring conditions were deficient, five detailed examples provided.
· Visual and audio alarms in the manufacturing area are not addressed appropriately.
· Humidity controls in the aseptic processing areas are deficient.
· Not all critical surfaces are adequately evaluated in the aseptic processing areas. Further, no justification exists for the selection of surfaces included.
· Staff who perform visual inspection of in process and finished drug products are not qualified to detect defects under the conditions normally used during routine operation. FDA provided a laundry list of examples in support of this observation.
· Equipment require “ad hoc’ modifications that include adhesive tape to secure items.
Americare Compounding LLC of Garden City South, NY compounds sterile drug products and received a 5-page form 483 at the close of inspection on June 19, 2013. This form 483 includes observations regarding aseptic manufacture and the potential for cross contamination of products with penicillin. Observations include but are not limited to:
· The firm does not have written procedures governing operations conducted at the facility.
· The firm failed to test drug product for assay or identification, only one of the parenteral products is tested for sterility, and no endotoxin testing data is available for any lot of sterile product manufactured.
· The ‘beyond use date’ given to the products is not supported by stability studies conducted by the firm. They rely on the scientific literature or vendor information to establish these dates that do not exceed 30 days.
· Sterilization processes have not been validated.
· Penicillin products are not processed under a separate air handling system, or in a separate facility from other sterile products. Included in this lack of separation are beta-lactam non-penicillin drugs such as ceftazidime.
· The firm does not have adequate environmental monitoring data to demonstrate the ability to maintain the necessary ISO room classifications.
· Gowning is inadequate: gowning in the ISO 7 areas can be reused during the week; exposed skin was visible on staff working in the ISO 5 work station areas.
· The firm does not use sporicidal cleaning agents inside the ISO 5 hoods.
· Environmental monitoring is not performed on a frequent basis under dynamic conditions. Further, environmental monitoring of personnel is not performed each day of production, there is no sampling of operator’s gloves.
· The firm also received a form 483 at the close of inspections in June, 2015 and a warning letter in April, 2014.
Luitpold Pharmaceuticals, Inc. in Shirley, NY received a form 483 at the close of inspection on October 29, 2014. This form 483 addresses the always challenging issue of mold identified in aseptic processing areas. Observations include but are not limited to:
· Mold was repeatedly recovered in both the aseptic processing areas and support areas. Several roof leaks in 2013 and 2014 were not considered in the investigation. The firm could not determine how long the roof damage existed prior to the leaks documented in 2013 and 2014.
· The firm conducted ‘surveillance sampling’ of a variety of facility areas, including interstitial facility spaces to identify possible entry points of mold spores. The FDA identifies four specific deficiencies in the process and methods used in this analysis.
· Investigations associated with mold recovery were open for extended periods of time without resolution.
· The firm does not have procedures to evaluate potential for adsorption of compounds from contact filling components such as silicone tubing used in peristaltic pumps.
· The firm does not have documentation to ensure that they reviewed the disinfectant effectiveness studies performed by a contractor.
And finally, SmithKline Beecham Limited in West Sussex UK received a 9-page form 483 at the end of inspection on July 10, 2015. The firm manufactures APIs, and the inspection focused on potential for cross contamination with penicillin. This inspection resulted in a warning letter issued on June 30, 2016. Observations include but are not limited to:
- Penicillin was found on the surface of non-penicillin processing areas 187 times between 2012 and the time of conduct of this inspection. The corrective actions did not include a comprehensive cleaning and decontamination of the associated areas and equipment.
· The cleaning methods used to remove penicillin on surfaces have not been adequately validated.
· Methods used to test for the presence of penicillin are not adequate.
· A variety of possible causes for the presence of penicillin in non-penicillin areas were identified but the corrective actions did not fully correct the findings.
· Investigations into microbial alert and action level findings in the water system did not establish a root cause. FDA identified 18 examples. Conclusions regarding ‘sampling errors’ were not supported by data