Akorn nets 22-item 483 for validation and quality deviations.

Akorn, a drug manufacturer in Decatur, IL, received a 22-item 483 following a 2007 GMP audit because procedures designed to prevent microbiological contamination of drug products purporting to be sterile did not have adequate validation of the sterilization process, and for other quality assurance problems, according to FDA records.

The audit was performed by investigators Susan Yusicus and Thomas Arista from the Chicago, IL, District Office.

The firm's website states that Acorn develops, manufactures and markets branded and generic diagnostics and therapeutics for niche segments such as ophthalmology, rheumatology and anesthesia.

According to the 483, "the validation protocol supporting the aseptic filling on line AH and lyophilization of products in area AJ was not followed and the established acceptance criteria were not met.

"This data was used to support the aseptic filling and lyophilization of IC Green. Lyophilization Process Media Fill Validation Lyophilizer #2, 5 mL, 20mL, 100mL Vials Lyophilization Area. Protocol was executed for lots 111206, 111246, and 111436. Quality assurance approved the summary report on May 7, 2007, without all of the protocol requirements and acceptance criteria being met."

For example:

* None of the media fills executed met the protocol objective to validate the full lyophilized chamber which consisted of several shelves. The batch records did not document the number of shelves used;

* All required interventions were not documented as required for any of the media fill lots. The protocol acceptance criteria indicated that all planned interventions must be performed as scheduled;

* The positive unit found in the integral containers of lot 111206 was not adequately investigated. The acceptance criterion indicated that any positive will be investigated.

Next, the FDA report noted that validation protocols supporting the aseptic filling of products on line AH were not followed and the established acceptance criteria were not met. This filling line was used to fill aseptic products including Lorazepam Injection USP, Myochrysine (Gold Sodium Thiomalate) Injection USP, and Diltiazem Hydrochloride Injection.

For example, the 483 stated that the Aseptic Process Validation Qualification 2 mL Vials Lyophilization Area protocol was executed for lots 51426, 61046, 61056, and 61286 in May-June 2006. Quality assurance approved the summary report on Dec. 20, 2006, without all of the protocol requirements and acceptance criteria being met.

"Only one filling speed was documented for lots produced. No fill speed was documented for lot 61286. The filling line speeds for lots 51426, 61046 and 61056 were reported as [undisclosed], which represented the maximum fill speed. The protocol required for the fill to be performed at both minimum and maximum fill speeds," according to FDA records.

Further, all required interventions were not documented as required for any of the executed media fill lots. The protocol acceptance criteria indicated that all planned interventions must be performed as scheduled.

Regarding quality control, FDA investigators determined that the responsibilities and procedures applicable to the quality control unit were not in writing and fully followed.

"Specifically, there is no tracking system established for management reviews which are performed to provide quality oversight and document management evaluations of the quality system. There is no SOP describing the management reviews," according to the report.

Also, on July 23, 2007, one of the FDAers requested a list of management reviews be provided and the firm indicated that no list was available. On Aug. 3, 2007, a non-controlled binder located in the office of the Executive Director for Decatur Quality Assurance and Quality Control was provided. It contained copies of approximately 49 management reviews performed on various subjects since June 17, 2005.

According to the 483, the binder was not intended to include copies of all management reviews performed.

Next, batch production and control records did not include the results of any investigation made into any unexplained discrepancy and whether or not the batch of drug product had already been distributed.

Specifically, the investigation was initiated after numerous complaints were received since 2004 regarding the use of AK-Fluorhas. The investigation identified the presence of thin, flat flakes that are consistent with glass delamination flakes and transient particles consistent with Fluorescein Monosodium precipitate. The firm did not implement corrective action to prevent this particulate formation from occurring in future lots of AK-Fluor, the 483 stated.

Also, the firm did not change testing specifications or retain sample examination procedures in response to these findings.

Pertaining to the same violation, there was no documentation in the batch record and no QA investigation regarding the source or cause of the high number of defective glass rejects found during the inspection of AK-Fluor lot 51177. The batch record indicated that [undisclosed] units of 17,143 filled units were rejected due to defective glass. The product was released by QA for shipment on May 30, 2007. During the inspection the Director of Quality Assurance inquired about the rejects and indicated that the rejects were caused by a defective sealing rail on the Westcap capper.

Next, procedures designed to prevent microbiological contamination of drug products purporting to be sterile were not established, written, and followed.

Specifically, master batch production records for aseptically filled and lyophilized products produced in areas AH and AJ did not include information regarding the filling line speeds used for production. This affected all production lots produced in this area including IC Green USP, Lorazepam Injection USP, Myochrysine (Gold Sodium Thiomalate) Injection USP, and Diltiazem Hydrochloride Injection.

"Master batch production records for aseptically filled and lyophilized products produced in areas AH and AJ did not include reject accountability forms to categorize the reasons for filled but rejected units. The SOP AA 259 'Filling Reject Accountability' had not been revised to include a form for the new areas," FDA documents noted.

In addition, equipment and utensils were not cleaned, maintained, and sanitized at appropriate intervals to prevent contamination that would alter the safety, identity, strength, quality or purity of the drug product.

For example, there were two open floor drains located in the personnel corridor (used to access the gowning rooms that lead into the Class 10,000 and Class 100 areas) in the new AH manufacturing area. There were no backflow preventers connected to, or part of, the drain pipe to preclude the back siphonage from the plant's so-called "pits" that lead to the municipal waste drain system.

Also, "the environmental monitoring program does not include obtaining samples from the surfaces of the open floor drain to assure the absence of objectionable microorganisms and to prevent the ingress of objectionable microorganisms into the personnel corridor," the 483 added.

Also in the same observation, a mobile cart was used to transfer material/equipment into and out of the Class 10,000 areas and aseptic fill room AH. The normal production flow and transfer operations included movement of the cart via the personnel corridor that contain the open floor drains.

"However, there is no record to document that the wheels of the mobile cart are cleaned and sanitized, to control and prevent objectionable microorganisms from entering into the controlled environments and the aseptic fill room AH," the document noted.

Next, the firm's drug product operations and procedures were not performed in specifically defined areas of adequate size to prevent contamination or mix-ups, the report stated.

For example, personnel entry into the aseptic filling rooms was via gowning room J, which consisted of Class 100,000 and Class 10,000 gowning areas. The Class 100,000 is the area nearest the entry door leading from the common personnel corridor and the Class 10,000 area is adjacent the entry door that lead to the aseptic fill rooms.

"However, there is no data available to distinguish or delineate where the Class 100,000 ends and where the Class 10,000 area begins in gowning room J. In addition, the gown room that is used by laboratory personnel to enter into and work in the sterility test suite is also set up in a similar manner as gown room J," the report stated.

Also, FDA stated that the firm's testing and release of drug product for distribution did not include appropriate laboratory determination of satisfactory conformance to the final specifications prior to release.

"Specifically, finished product samples of refrigerated finished products (Diltiazem Hydrochloride Injection, Lorazepam Injection, Proparacaine Ophthalmic Solution, Tetracaine Injection, are collected at the Grand Avenue manufacturing facility after filling prior to transfer to the Wyckles Road location for inspection, labeling and packaging under ambient conditions. These finished samples do not represent distributed product that is subject to increased exposure at ambient temperatures," the 483 noted.

Further, "the samples also do not represent distributed product that may have been exposed to extreme temperatures in a non-climate controlled truck during transfer of the product between the Grand Avenue and Wyckles Road facilities. The firm justifies this practice based on process validation data which includes analysis of packaged product. However, the firm's own data demonstrates that increase degradation occurs."

Next, Akorn's employees were not given training in the particular operations they perform as part of their function, current good manufacturing practices, and written procedures required by current good manufacturing practice regulations, the document stated.

For example, "the visual inspection of media fill containers performed by the Wyckles personnel is via the Media Fill Inspection By Non-Microbiology Personnel document #ML201, stated that training would demonstrate to non-Microbiology personnel how to identify microbial growth in media filled units. For example, the procedure establishes to 'open the unit by unscrewing the cap and removing the tip. Observe the interior of the tip to assure no residual growth or growth media remains.

"However, as confirmed by the Microbiology Supervisor, the training did not include the use of the white color LDPE bottles that are used for ophthalmic finished products. Rather, the inspection training used colorless glass vials to observe visual microbial growth." Further, the firm's procedure "is silent with respect to visually inspect the interior surfaces of the white color LDPE bottles to assure there is no evidence of microbial growth attached to the interior surfaces. The number of containers to be sampled is not based upon appropriate criteria."

In addition, the 483 noted the number of containers to be sampled is not based upon appropriate criteria. "Specifically, there are so-called "tailgate" samples, which Akorn requests that 750 stoppers, pulled randomly to be representative of the lot, be packaged separately for use in testing of the stoppers that is forwarded by the contract supplier. The 750 random sampling is taken from a typical stopper lot size.

"The Sampling, Inspection and Release of Incoming Closures procedure, document #WQ 1 03, define the process for sampling and inspecting elastomeric closures for injections. A 315-sample size is taken from the 750 vendor random sample to perform the various requisite analyses, such as, dimensional measurement tests, microbiological and analytical tests. However, there is no document to describe the rationale to support the statistical validity for the number of tests performed."

Akorn, Decatur, IL, 7/23-8/7/07, Doc. 110055M, $16 plus retrieval.

* The Checklist--Akorn

* Inadequate validation for sterility

* Validation protocols not followed