FDA inspections of biologic manufacturing facilities are identifying deficiencies that include potential microbiological contamination in aseptic processing along with data integrity and process control and variability concerns.

FDA Center for Drug Evaluation and Research (CDER) Office of Pharmaceutical Quality (OPQ) Office of Pharmaceutical Manufacturing Assessment (OPMA) Senior Pharmaceutical Quality Assessor Dr. Madu Dharmasena provided manufacturing facility inspection case studies as part of her presentation at the PDA/FDA Joint Regulatory Conference 2024 held in Washington, DC, September 9-11 2024.

The six case studies illustrate issues her office is finding on inspections of biologic products covering both drug substance (active pharmaceutical ingredient or API) and drug product inspections. The case studies supplemented insights she provided on how CDER regulates and inspects biologic products (see Facility Issues Top Inspection Deficiencies for Biologics).

Dharmasena explained that the manufacturers of biological therapeutic products are subject to 21 CFR parts 600-680 as well as Parts 210 and 211.

“Specifically, these are parenterally administered biotech products manufactured using aseptic and sterile fill finished products in which the drug product container and closure system are first sterilized and then brought together.”

The major operations or steps in a drug product process may include formulation, sterile filtration, aseptic filling, packaging, labeling, and testing.

Manufacturers of biologic APIs are inspected based on the principles of ICH Q7.

The case studies are presented below with the CDER regulator’s comments on what investigators observed and the deficiencies documented in the 483 drawn from the observations document.

CASE STUDY 1 (API)

What Did the Investigators Observe?

In this case, the drug substance process pools were contaminated with over-the-limit microbial excursions and below the action limit bioburden.

The company released 18 drug substance batches, and 14 of them had microbial excursions or recoveries in one or more in-process pools.

There were ten bioburden action limit excursions in the in-process pools and each of them went through subsequent filtration, which then brought the bioburden within the limit. However, “these filtrations do not remove impurities from the contaminating microorganisms,” Dharmasena pointed out. “They also had contaminations from an endotoxin excursion toward the end of their purification step, which led to rejection of this pool.”

The most probable root cause, according to the company, is that a contamination event occurred during the manufacturing process due to manual working procedures. There was also contamination at the bioreactor level that was rejected that harvest.

“There were 41 bioburden recoveries that were within limit, and they didn’t speciate any of those contaminants.”

What Deficiencies Were Noted?

Because of these observations, the company was cited that GMP requirements per ICH Q7 were not met. The process was not appropriately designed to ensure that the in-process intermediates did not become contaminated. “Also, the process lacked consistency because of all these contaminations,” she said.

In its response to this observation, the company provided some Corrective and Preventive Action plans (CAPAs) and did a risk assessment. In its risk assessment, the company said it would be performing an assessment based on a methodology published in a paper where the CFU result and the identified bacteria are converted into mass-based exposure.

“They are proposing that the in-process, bioburden excursions are deemed low-risk and acceptable according to CFU to patient mass exposure conversion methodology. Also, they went ahead and changed their IPC [in-process control] document and SOP to allow justification of out-of-limit batches using this CFU to mass exposure conversion method.”

The firm’s impact assessment in response to this observation “is inadequate and does not meet the ICH Q6B requirements,” Dharmasena explained. “Contamination should be strictly avoided and suitably controlled with appropriate in-process acceptance criteria or action limits. Contaminants in the product may impact the quality of the product via degradation and modifications as they may release byproducts.”

In the post-action letter, FDA stated that the adoption of the flawed microbial risk assessment methodology should not be applied to out-of-limit in-process testing results, as stated in the referenced article. Also, this methodology should not be used for and it is not suited for assessment of the drug substance in process pools.

Case Study 2 (Drug Product)

What Did the Investigators Observe?

The critical operations were not performed with sterile tools. For instance, “the aseptic connection with 18 filling needles to sterile tubing was performed by the operator wearing non-sterile gloves. The installation of the filling needle into its clamp required holding the upper part of the filling needles with the non-sterile gloves. In addition, after setup, the top of the filling needles was adjusted multiple times with RABS gloves just prior to production. The needle tips were exposed throughout these manipulations.”

Part two of this same observation states that “during manufacturing operations, including intervention, RABS gloves were observed to block first air on sterile containers.”

Also, “a small amount of liquid is collected from an air filter in the Grade A during setup. The liquid is collected in an open container that is subsequently removed by opening the RABS door. This operation is performed after the installation of the filling needles.”

“RABS gloves for fill lines 1 and 2 are reused for one year (approximately 100 batches) with no periodic sterilization.”

“Personnel in the Grade A/B during filling setup and production did not always move slowly and deliberately to minimize disruption of unidirectional air. Operators were observed with their head, shoulder, upper arms, and stomach inside Grade A near critical surface areas. Operators blocked first air by placing gloved hands and forearms over the filling needles and sterile path areas.”

What Deficiencies Were Noted?

The primary issue, according to the 483, was “the procedures designed to prevent microbial contamination of drug products purporting to be sterile are not established or followed.”

Because of these deficiencies, the requirements of 21 CFR 211.213(b) to establish and follow procedures to prevent microbial contamination of sterile drug products were not met.

Case Study 3 (Drug Product)

This is a case study where the firm failed to have adequate document control of forms that are used for GMP activity, Dharmasena pointed out.

What Did the Investigators Observe?

“We observed shredded material in the building, and the material contained references to the SOPs,” she said. “So, there is no assurance that this shredded material did not contain any quality documents. In this case, the document control requirements were not met.”

What Deficiencies Were Noted?

The following are taken from the 483 issued after the inspection.

“Your Quality Unit failed to establish adequate document control of forms that are used for GMP activities.”

“Uncontrolled blank forms have been used for in-process sample logs that are attached to the production master records.”

“Uncontrolled facility cleaning logs are used to record cleaning activities of classified areas.”

“Uncontrolled form xx used by the Qualified Person for batch release.”

Case Study 4: (API and Drug Product)

What Did the Investigators Observe?

This is a case study regarding environmental monitoring. “Lately, we have been seeing environmental monitoring observations at a very high frequency,” Dharmasena stressed. “In this case, the drug substance manufacturing areas are not under control for mold.”

What Deficiencies Were Noted?

The following are taken from the 483 issued after the inspection.

“There is an unacceptably high number of mold recoveries in classified rooms for the manufacture of xx drug product. Specifically, there were 696 mold recoveries since January 2021 (157, 372, and 167 mold recoveries in 2021, 2022, and 2023). The mold action limits (Grade C >5 CFU and Grade D >10 CFU) are inadequate to initiate additional cleaning, environmental monitoring investigation, or deviation.”

“The mold contaminants recovered via the environmental monitoring program have not been trended in a meaningful manner to identify trending patterns. There have been 166 settle plate mold recoveries and 521 active sampling recoveries compared to only nine surface sample monitoring recoveries since 2021. No investigation was performed to identify the trending mold recovery pattern, and corrective and preventive actions (CAPAs) have not been implemented due the mold recoveries.”

“Environmental controls are inadequate to minimize the risk of contamination.”

Case Study 5 (API)

What Did the Investigators Observe?

“Case study five is another case study where the process is not consistent,” Dharmasena emphasized. “This particular drug substance had a cancel or rejection rate of about 50% out of 16 batches. They had three batches that were terminated due to contamination in bioreactors, two batches were rejected due to out-of-specification test results, and three batches were terminated due to their changing manufacturing plan.”

What Deficiencies Were Noted?

This does not meet GMP requirements as per ICH Q7 guidance, as it lacked process consistency.

Case Study 6 (Drug Product)

“In this last case study, we observed another drug manufacturing process that is not under adequate control,” the CDER regulator pointed out.

What Did the Investigators Observe?

Specifically, “during the filling of xx drug product in the restrictive access barrier systems (RABS) of xx vial fill line, droplets of product were frequently observed at the fill needle tips before dispensing the product into the vials. Occasionally the droplets formed at the tip of the filling needle dripped on to the tray table below.”

Dharmasena commented, “surprisingly, the operators told us they have never seen this before, but when we were there for just five minutes, we saw many droplets dropping on the tray below.”

What Deficiencies Were Noted?

This practice did not meet the requirements of 21 CFR 211.61, as the filling needle equipment is not adequately maintained to prevent malfunction and contamination.

Conclusions

In addition to its responsibility for drugs, CDER is responsible for regulating and inspecting certain biologics. CDER-regulated biologic products include enzymes, monoclonal antibody, antibody conjugates, fusion proteins, growth factors, cytokines and botulinum toxins.

Case studies prepared by CDER’s Dharmasena to illustrate predominant concerns her office is finding during inspections of both APIs and finished products show a heavy emphasis on sterile manufacturing, including the following deficiencies: