Alpharma gets blasted with 59-item 483 citing quality system failures.

Alpharma USHP, Baltimore, MD, Baltimore District

Serious quality deficiencies for a number of products landed human-drug maker Alpharma USHP a whopping 59-item, 24-page 483.

Deficiencies uncovered by a team dispatched from FDA's Baltimore District included process validation, lack of qualification data, computer system not validated, numerous quality assurance failures, product testing, failure to investigate complaints, documentation, production and processing controls, cleaning validation, and failures with container closure systems. The Baltimore team conducting the investigation consisted of Melissa Garcia, Suzanne Muchene, Alicia Mozzachio, Jean Blackston-Hill and Kenneth Nieves. The EIR was not available at press time.

According to the 483, process validation for Phenytoin Oral Suspension bottles was deficient because there was "no scientific justification" that sampling during the 1998 process validation studies for scale up was adequate for the batch size. Specifically, the "validation studies do not adequately address product uniformity in the surge tank." Also, 10 batches were rejected due to viscosity out-of-specification (OOS) results since 1999.

Further, there was no validation data to support the allowance of "hold time" in transfer overhead lines during filling of suspensions and lotions, according to the 483.

Qualification had not been completed for a new pint line, which was initiated in order to evaluate the interaction between the silicone diaphragms and products. The 483 noted that qualification was deficient because there was "no scientific justification/rationale for the [unnamed] difference between the test sample and reference sample assay results defined as the acceptance criterion." Also, data from the study was not used to take appropriate corrective actions for products that failed to meet established acceptance criterion.

Additional highlights from the 483 follow:

* Portions of three different lots of three different products (Phenytoin Oral Suspension, Auroto Ear Drops, Cromolyn Sodium Nasal Solution) were distributed, although at least one product sample failed to meet release specifications.

* Products that failed in-process testing were released based on finished product testing, although the exact cause of the OOS was never determined.

* QA staff failed to adequately review and approve pertinent documentation. For example, change request assessments, variance OOS results, annual product review and positive microbial water testing results were not reviewed.

* A variance that occurred during packaging of Phenytoin Oral Suspension was "erroneously voided by QA.... This variance was voided on 11/27/01 by the QA Manager [at that time], and the lot was released for distribution on 12/11/01. The variance reports a deviation from the validated manufacturing process and should not have been voided."

* The Change Control tracking system does not accurately reflect the current status of change requests. The 483 listed four examples where changes were not closed in the tracking system, yet were approved by QA and were in effect.

* Failure to adequately investigate and evaluate an OOS specification viscosity result for Phenytoin Oral Suspension reported by a customer.

* Customer complaints are not always investigated in a timely manner. For example, packaging related complaints for missing lot numbers and expiration dates for Metoclopramide Oral Solution and for Promethazine with Codeine Cough Syrup received in November 2001, are still open. According to the 483, a complaint meeting held in April 2002 concluded a packaging problem due to human error was the problem, yet a packaging investigation was not initiated. Similar examples were listed on the 483.

* "There is no conclusive evidence to explain the cause and potential effects of precipitation observed in various lots of Guiatuss AC. Your firm has failed to adequately investigate hundreds of complaints of precipitation in this product."

* The returned complaint samples are not always tested.

* Numerous instances of failure to follow SOPs for documentation procedures were listed on the 483. For example, during a walk-though of the packaging area, one FDAer observed that the number of "damaged" units of Guiatus CF "on Line 3 being recorded on a paper towel."

* Regarding production and processing controls, the 483 stated that "in-process weights are not always maintained within the acceptable ranges." For example, "there are no procedures to assure that Phenytoin Oral Suspension bottles exceeding the maximum fill weight are discarded."

* There is no data to support the effectiveness of running the filling machine for a single cycle every four hours when a line is down, in order to avoid purging the filler and overhead lines. Examples were listed whereby the overhead line and filler were not purged as required for stoppages.

* Cleaning validation studies and cleaning verification practices were deficient. For example, there was no raw data for the method validation used to test rinse water samples after every cleaning, and recovery studies were not performed for all products included in the cleaning validation studies. Additionally, there were multiple cleaning verification test failures on various pieces of equipment.

* Information concerning the failure of a distributed Phenytoin Oral Suspension lot to meet all established specifications was not reported to FDA in a Field Alert. A complaint from a unit-dose repackager was received concerning OOS viscosity results, and "Alpharma confirmed this OOS on 4/9/01 using an unopened bottle from this lot. Additional testing was performed, and an investigation summary dated 8/7/01 was written. A Filed Report was never submitted to the FDA."

* "Failure to reject any lot of components that did not meet the appropriate written specification for identity, strength, quality and purity." Nine examples were listed on the 483 that failed microbiology laboratory testing.

* "Laboratory controls do not include the establishment of scientifically sound and appropriate specification and test procedures designed to assure that components and drug products conform to appropriate standards of identity, strength, quality and purity." Specifically, raw materials with OOS results were used by the firm during the manufacturing of products. Additionally, "the firm relied in finished product testing to release these products while using contaminated raw materials."

* Laboratory records lack specificity in that they do not include the quantity of the sample.

* The firm's use of membrane filtration technique, the USP method that is recommended for WFI testing, has not been validated to show that the method is adequate for testing purified water.

* "There was no assurance that products analyzed for stability remained within specification during their shelf life. There was either insufficient data or no data at all to demonstrate that analytical methods were stability indicting and capable of separating the active ingredient from its impurities and degradation products."

* There are no SOPs or protocols that contained the firm's procedures and acceptance criteria for conducting "Operation and Performance Qualification studies for each instrument including, but not limited to the following instruments: UV/VIS spectrophotometers, infrared spectrophotometer, particle size analyzer, stability chambers, and the Chromatographics Data Acquisition System."

According to an Oct. 15, 2002 letter to FDA, Alpharma President Michael Nestor wrote: "In light of the broad-ranging and serious concerns raised by the Investigators' observations, we recognize that more is called for than the proffering of a typical 483 reply. The concerns presented in the observations underscore for us the need to develop a comprehensive, workable corrective action plan."

The corrective action plan outlined by Nestor included voluntary recall of products related to specific product observations, and a scaling down of the facility to "dramatically reduce production at the Baltimore facility."

Nestor said his firm was committed to "spend substantial sums, currently estimated to be $30 million dollars to fund the corrective action plan, to accomplish the goals detailed in this response, and to ensure that the Baltimore facility is adequately staffed in both numbers and requisite expertise."

The plan outlined by Nestor included a complete reorganization of the Baltimore facility into six functional units that will report directly to the director of Quality. Those units would cover:

1. Laboratory Operations: chemical testing, stability testing, microbiological testing, environmental monitoring.

2. Validation: process validation, cleaning validation, utilities/equipment validation, computer/software validation.

3. Documentation and change control: documentation control, change control.

4. Quality assurance: investigations/material review board, CAPA, in-process QA, product complaints, final product release, incoming inspection.

5. Compliance: quality systems implementation, management review, internal assessments, regulatory commitment tracking, annual product reviews.

6. Training: training and qualification.

Additionally, Nestor wrote in his letter that the firm recently hired a new vice president of Quality and a new vice president of Operations, as well as several new managers, and would continue to fill additional positions as required by the firm's compliance plan.

Finally, aside from retaining compliance and quality consultants, the firm also retained a second group of consultants to fill the role of independent, third-party auditors and certifying consultants.

Steven Barber, FDA compliance officer, requested in a July 2, 2003 letter, that the firm visit the Baltimore District Office in person because the firm was not living up to the 18-20 month time frame to which it had committed.

"Since the April 12, 2003 update, we have observed the commitment dates for corrective actions are being delayed. In the June 30, 2003 update, the commitment dates for corrective action have been delayed to September 2005. We find this timeline unacceptable. The original corrective action you committed to are a result of an inspection ending on August 2, 2002, which was conducted under a 1992 Consent Decree of Permanent Injunction," FDA wrote.

A comment from Alpharma was not forthcoming by deadline.

Alpharma USHP, Baltimore, MD, 6/18-28, 7/1-3, 8-23, 25-26, 8/1-2/02, Doc. 109493M, $136.50 plus retrieval.