Many of the 14 warning letters FDA issued during the second quarter highlighted problems with microbial contamination.

Other common themes included mix-ups and data shortfalls as well as lax out-of-specification reporting procedures.

Laboratoire Atlas. A June 25 warning letter emphasized contamination-related issues that FDA investigators found when they inspected the firm's drug manufacturing facility in Ville d'Anjou, Quebec, Canada, on Sept. 2-5, 2008.

The facility was producing a product that is required to be sterile and was labeling the product as complying with a U.S. Pharmacopeia sterility test, USP 30. But upon inspection, FDA found that the facility only makes non-sterile products and did not actually perform the sterility test.

Furthermore, there were problems with microbial contamination that the company did not properly address. It failed to identify microbial contaminants or find, correct and prevent contamination of equipment. Nor did it perform root cause investigations when its contract laboratory repeatedly found that certain samples exceeded aerobic microbial counts.

In its written response, the company included a retrospective post-inspection investigation of the problem. But the investigation failed to evaluate a distribution line that had been involved in all five microbial test failures.

Also, the company should not have invalidated failing microbial test results from its contract laboratory without any investigation or scientific justification. Nor should it have concluded on the basis of finished product testing that product was free of microbial contamination without first conducting thorough root cause investigations and implementing corrective and preventive actions, FDA said.

FDA said Laboratoire Atlas should have audited its contract laboratories to qualify them and that it should establish quality agreements with them.

The agency noted a need for cGMP training. Even the company's director of quality control acknowledged a lack of familiarity with U.S. cGMP requirements.

Apotex. FDA also on June 25 issued a warning letter to another Canadian firm, Apotex, due mainly to lax investigations and belated field alert reports uncovered during a Dec. 10-19, 2008 inspection of its Etobicoke, Ontario, non-sterile oral solid dosage drug product manufacturing facility.

FDA found that during the previous two years, the company had rejected 554 batches of products such as cyclosporine, gabapentin, topiramate, divalproex and carbidopa-levodopa at the plant, and yet there was no record of any resulting investigations.

"This unusual high number of rejected batches demonstrates a lack of adequate process controls and raises significant concerns regarding the capability and reliability of your processes to consistently manufacture drug products meeting predetermined specifications," FDA said.

Investigations of post-encapsulation assay failures of hydrochlorothiazide scale-up batches that occurred in March 2008 still had not been completed in December.

Also an unknown peak found during three-month stability interval testing of ketoconazole on Oct. 15, 2005, was found to result from cross-contamination of the tablets, but the investigation never mentioned a root cause, whether other products were affected or mentioned any corrective actions that might have been taken.

Apotex was not submitting field alert reports to FDA within the required three days, FDA said. In the ketoconazole case, the report was 15 months late. The report on an out-of-spec stability testing result for glipizide tablets was four months late. The report on an investigation confirming a complaint that two bottles each contained a few overlarge lovastatin tablets was more than five months late.

In its response, the company said the problem was that only senior staff was trained in the field alert reporting procedure and that it would start training all staff.

AccuMed. A June 24 warning letter to AccuMed Inc., Lawrenceville, N.J., a maker of store brand drug products sold over the counter, emphasized widespread failures of the quality control unit to follow written responsibilities and procedures.

The quality control unit failed to: ensure cough formulas went into the correct packaging, conduct thorough out-of-spec investigations, make sure the stability program was followed, review reprocessing procedures for nonconforming batches, ensure the accuracy and completeness of batch records, timely notify FDA of a recall, and reject drug products that failed to meet stability specifications.

In its response to FDA's original Form 483 report, AccuMed said it had decided to change the leadership of the quality unit and the quality control laboratory, prompting FDA to ask in the warning letter how the new leadership will do better.

Two incidents drew special attention from FDA. One was a mix-up in which AccuMed packaged Cough Formula DM into cartons for Tussin Cough Congestion Formula. FDA raised numerous issues in connection with the mix-up, including the company's failure to notify FDA's recall coordinator until four months later.

The other incident was the addition of citric acid in an ad hoc reprocessing of orange-flavored laxative powder that had failed the pH test.

Xttrium Laboratories. A May 20 warning letter to Xttrium Laboratories Inc., Chicago, focused on contamination at the facility where it manufactures chlorhexidine gluconate (CHG) solutions for use as antimicrobial skin cleansers in surgical settings and for oral rinses.

The agency's inspection of Xttrium and the resulting warning letter also reflect FDA's growing attentiveness to supply chain concerns.

A Jan. 22 warning letter to Sage Products of Cary, Ind., for making CHG-impregnated cloths that were contaminated with Burkholderia cepacia, was based on findings from a June 2 to July 10, 2008, inspection. The cloths are used to disinfect skin prior to surgical incisions (Also see "Drug Recall Totals for 2008 Highlight Global Outsourcing and Other Risks" - Pink Sheet, 1 March, 2009., p. 4).

FDA then inspected Xttrium from July 2 to Aug. 27, 2008. During this inspection, the agency sub-sampled Xttrium's retention samples. Two of the four sub-sampled turned out to be contaminated with Burkholderia cepacia.

The agency further inspected Xttrium from Oct. 27 to Dec. 31, 2008.

FDA also found cases where lots failed microbial limits tests and were even found to be positive for Pseudomonas aeruginosa, and yet were released for marketing because retained samples were within specifications - and even though the root cause of the out-of-spec findings was never determined.

The problem, FDA said, was that Xttrium's standard operating procedure of overruling out-of-spec results on the basis of duplicate testing of retain samples was inadequate.

Similarly, Xttrium would quarantine water on the basis of a high total plate count, but then would release it if retain water samples tested clean, an approach FDA found unacceptable "because a negative re-test alone is not enough evidence to release a batch from quarantine."

The agency also had problems with the company's investigations into out-of-limits microbial results for purified water. Typically, numerous OOL findings led to root cause investigations that attributed the findings to either biofilm in the production hose or sampling error.

FDA advised the company that "finished product testing, regardless if it is performed in duplicate, is not a substitute for adequate cGMP controls."

FDA found fault with the company's responses to various customer complaints as well.

In one case, Xttrium analyzed a sample from a customer who complained about orange precipitate at the bottom of 15 drums. However, Xttrium's lab filtered the sample so that none of the precipitate actually got tested.

In a similar case, the company rejected two lots because of cloudiness and white precipitate. A root cause investigation found that the inferior chlorhexidine base used didn't fully precipitate. However, the investigation was not extended to other lots that might have been - and in fact were - affected, according to later customer complaints.

In yet another case, a nurse found an insect inside a CHG solution bottle when she opened it in the operating room. In a subsequent investigation, Xttrium identified 14 cases of bottles that contained insects and missed another lot of insect-containing bottles.

Xttrium has switched to a different bottle vendor, but FDA stressed that it still must tighten up its acceptance procedures so that it can reject contaminated bottles upon receipt.

FDA raised additional concerns about complaint handling procedures. In one case there was a paint chip in an unknown lot. In another, six months went by before the company investigated a complaint about adverse reactions two premature infants experienced.

FDA strongly encouraged Xttrium to follow through on plans to possibly hire a consultant.

The agency also recommended "that the final formulation of these products be produced as sterile to avoid introducing exogenous microorganisms."

FDA had also urged Sage Products to switch to sterile products.

Additionally, FDA recently asked its pharmaceutical science advisory committee to endorse a possible requirement for such products to be produced in a sterile manner. However, committee members raised so many questions that FDA withdrew the proposal for further development.

Lupin. Data-related concerns were most prominent of a wide range of concerns cited in a May 7 warning letter to Lupin Ltd., Mumbai, India, regarding its manufacturing facility in Bhopal, India.

FDA found batch records for the sterile and non-sterile manufacturing facility that needed to be maintained but were slated for destruction.

The company gave FDA a smoke study summary report but the agency was unable to open the CD ROM to review it.

The facility's enterprise resource planning system allows rejected batches to be released, FDA said.

Certain calibration data was only available in a spreadsheet, not in raw form. Also, there were verified and signed calibration certificates but no raw data to support them.

There were a number of bulk glass vials stored without any labeling in a seemingly ad hoc fashion.

FDA also was left wondering who actually inspected media fill vials.

There were questions about water with black residue in one area and a greenish film on the bottom of a water drain channel that could be either paint or biofilm.

Synkem. A May 6 warning letter to Synkem SAS of Chenove, France, emphasized concerns that the non-sterile active pharmaceutical ingredient manufacturer often disregarded out-of-specification findings on the basis of unspecified laboratory error without performing any investigations.

Cilclopirox and Fenofibrate API were mentioned specifically in this regard.

FDA saw similar problems with raw materials. In the several cases, the company asserted the supplier was not responsible for failing OOS test results even though it had not performed an investigation.

There were also problems involving the use of unauthorized copies of batch production and batch cleaning records and failure to investigate a process deviation.

Hill Dermaceuticals. An April 27 warning letter emphasized evidence that sometimes when workers for Sanford, Fla., prescription dermatological manufacturer Hill Dermaceuticals didn't get the right answer they simply tried again.

There was evidence that someone didn't follow procedures and used a pencil instead of the required pen with blue or black ink when making entries in the batch record book. The penciled data would be erased and overwritten in ink, sometimes with different data.

In at least one case when the contract laboratory reported out-of-specification results, Hill Dermaceuticals re-submitted samples for testing without conducting a root cause investigation.

A laboratory notebook showed assay results but not the underlying spreadsheet calculations that generated them.

Also the high-pressure liquid chromatography system that Hill Dermaceuticals used to analyze API and preservatives in its Tri-Luma Cream was different than the Perkin-Elmer HPLC system that it previously used.

Nostrum Laboratories. An April 27 warning letter to Nostrum Pharmaceuticals, Edison, N.J., highlighted concerns with stability testing at the company's Nostrum Laboratories Inc. unit's facility in Kansas City, Mo.

FDA said accelerated stability studies that Nostrum performed on sucralfate tablets made with API from a new supplier did not support the assigned five-year expiration date.

In response to the Form 483 report stemming from the Oct. 22 to Nov. 14, 2008 inspection, Nostrum provided a root cause investigation of an apparent failure of the accelerated stability studies to meet specifications at the six-month time point.

However, FDA said the investigation apparently came after FDA's inspection, which in turn followed the failure by seven months.

FDA also noted that it took five months for the lab to determine what type of laboratory error caused erroneous out-of-spec test results during accelerated stability testing of theophylline tablets.

FDA also warned Nostrum about cleanliness. Some rooms were listed as clean but were obviously dirty or contained dirty equipment. Investigators observed workers traversing from dirty to clean areas without first donning appropriate gowning and footwear.

There were also questions about the impurity and dissolution test methods the facility was using.

Dabur Oncology. In an April 22 warning letter, FDA raised concerns about microbial contamination, particulate matter and other problems investigators found during an Aug. 21-29, 2008 inspection at Dabur Oncology PLC in Borden, UK.

FDA focused a lot of attention on out-of-limit results from environmental monitoring collected during paclitaxel production. In one case a failure investigation report noted contamination by gram-positive cocci, possibly a Staphylococcus spp, but asserted that the impact on product quality was low.

In another case there was mention of a gram-positive rod, possibly a Bacillus spp, in a settling plate sample, but the organism was not identified.

FDA also found rusty, deteriorating equipment in isolators used in making sterile drug products, which it found to be unacceptable.

There were recurring incidents of particulate matter contamination, which raised questions at FDA about the company's procedures for visual inspection of sterile drug products.

The agency found cases where the company attributed the root cause of out-of-spec findings to analytical failure without fully documenting the problem.

FDA concluded that the quality control unit was doing a poor job in general, given that the unit was regularly signing off on incomplete or inadequate deviation reports and investigations, inadequately reviewed testing documents or log books, and was failing to ensure proper equipment maintenance.

FDA noted with concern that it had cited very similar problems during its previous inspection of the facility in December 2005.

First Quarter Warning Letters

FDA also raised a number of concerns in warning letters issued in April to Qingdao Jiulong Biopharmaceuticals Co. LTD in Qingdao, China, and Shanghai No. 1 Biochemical & Pharmaceutical Co. Ltd., in Shanghai (Also see "'Show' and 'Shadow' Heparin Facilities Targeted in Pair of FDA Warning Letters" - Pink Sheet, 1 May, 2009., p. 15).

Two more warning letters went out in April, one to Pharma Pac LLC, De Kalb, Miss., and the other to Procter & Gamble concerning issues at its Olay LLC facility in Cayey, Puerto Rico.

Those four April letters were inadvertently included in a listing in the May issue with three warning letters that FDA issued in the first quarter. The three first-quarter warning letters went to: Genzyme Corp., Cambridge, Mass.; Taro Pharmaceuticals, Brampton, Canada; and Prime Enterprises, Miami Lakes, Fla. ('The Gold Sheet,' May 2009, p. 17).

A fourth first quarter warning letter went to Sage Products, Cary, Ill., due to microbial contamination of its pre-operative skin preparation cloths that are soaked with a topical microbicide, chlorhexidine gluconate (Also see "Drug Recall Totals for 2008 Highlight Global Outsourcing and Other Risks" - Pink Sheet, 1 March, 2009., p. 4).

[note: Q2 warning letter box; revised Q1 box.]

- Bowman Cox (b.cox@elsevier.com )

Contamination Issues Featured in Second Quarter Warning Letters

More from Archive

More from Pink Sheet