Is PAT Leaving Quality Behind?

The recent emphasis on continuous improvement, operational excellence and Process Analytical Technology (PAT) within the pharmaceutical and biotech industries has driven us to evaluate the basic tenets of our approach to quality. Historically, the ability to ensure that a drug will meet its intended form, fit and function has been achieved through the combined application of quality infrastructure (SOP’s, policies, specifications), qualification or validation (commissioning, IQ, OQ, PQ process Validation) and Testing (in-process and final release). Despite this rigid environmental approach, the number of drug recalls continues to rise, escalating from 176 in 1998 to 354 in 2002 [US CDER website]. In a break from tradition, the FDA recognized that the methods that lead to this problem could not be part of the solution. In 2002 the FDA issued its revised cGMP guidance document entitled Pharmaceutical cGMP’s for the 21st Century- A Risk Based Approach. This guidance advocated a shift in industry reliance on a rigid punitive quality structure to achieve regulatory compliance (with increasing failures) toward a more rigorously scientific argument for product development, quality and production. The revised guidance for the 21st century is included as part of Presidential Executive Order 13329; an order that is designed to encourage innovation in manufacturing and includes PAT. As inventing, developing and bringing new therapies to the marketplace has long been considered a core strength of the pharmaceutical and biotech industry, and given the increasing pressure and scrutiny from patient, shareholder, government and regulatory entities, the industry has never been more driven to look at their core business practices for solutions.

Quality Systems
The current model for ensuring product quality is based upon the establishment of six major quality systems1.

The integration of these systems is designed to ensure through a combination of design, sampling, testing and documentation that product manufactured under these systems will meet their regulatory commitment for form, fit and function. The challenge within the industry has been that due to the realities of drug discovery in the current marketplace, too little emphasis is placed on the design and characterization elements of the product development lifecycle and too much reliance on inspection and verification/validation. With this weak foundation, regardless of the strength of the downstream cGMP quality system and checks, there is a failure mode that is inherent due to the lack of characterization and thus understanding of a product’s underlying chemical/physical/biopharmaceutical properties.

PAT
The intent of PAT was to advocate a more scientific and methodical approach to product development, scale-up and production. The impact of PAT will be felt in all sectors of the organization, and if applied correctly, will increase granularity in the quality and quantity of data being created throughout the product development lifecycle. This data becomes the basis for process understanding and ultimately ensuring product quality. In the PAT guidance document issued by FDA the agency discusses several of the key elements required to be successful in deploying PAT. Specifically, the guidance document discusses PAT Tools, the need for Process Understanding, Risk-Based Management, Integrated Systems Thinking and Real-Time Product Release. The reduced time and operational cost of taking a product from the end of manufacturing to the marketplace has captured both the imagination and interest of the industry.

The ideal PAT program will develop and archive data that can be used to understand the process from every aspect of the drug development lifecycle. The concept of a data warehouse has been proposed for organizations that would like to deploy PAT enterprise-wide2.

To view chart larger, click here.

Many pharmaceutical and biotech organizations capture this information as a byproduct of their manufacturing and operational systems but have no clear plan for utilizing the data. PAT will require the planned generation and effective utilization of all of this data in order to be able to reduce the risk associated with the manufacturing process variation.

The current industry development approach requires that a process be “locked down” once it is validated. This means any and all improvements to equipment, procedures and processes must go through change control and, in cases where a significant change to the original validated state has occurred, may require re-submission to the agency. It also means that when processes vary unexpectedly there is a reluctance to address the variation for fear of impacting the validated state. In the current system, deviations from the process require an investigation and corrective action plan (CAPA). The objective of a CAPA is to identify the root cause of the deviation before implementing the corrective action.

Real-Time Product Release
The FDA guidance document3 describes PAT as:
“A system for continuous analysis and control of manufacturing processes based on real time measurements or rapid measurements during processing, of quality or performance attributes of raw and in-process materials and processes to assure end product quality at the completion of the process.”

The lure of Real-Time Product Release (RTPR) has moved the industry forward faster technologically than in any period in recent memory. It was not that long ago (the 1980’s) that the industry was dabbling in the possibilities of Near Infrared (NIR) Spectroscopy as a surrogate to HPLC only to have the agency rebuff any attempts to demonstrate comparability. Now FT-NIR, FT-IR, Raman and Mass Spectroscopy are being integrated in-line with process and production equipment to accelerate release testing.

The unexpected byproduct of this has been the shift to technology that ensures process compliance. In and of itself, this is not a bad development. Enhancing the sophistication of control schemes brings greater light to the stability of the process, can manage risk and increase the probability of success. However, to be fully effective, it does require a thorough understanding of key process variables and their influences. Without this understanding, there can be no managing of the risk of product failure.

The challenges in integrating PAT with manufacturing hardware are significant. Sampling technology, cleanability, assay specificity, accuracy and linearity and comparability to bench top methods and previous production methods must all be addressed before proceeding to implementation.

Blending Case Study
In a recent PAT deployment a blender was equipped with in-line hardware in an effort to determine Content Uniformity of the API and key rate controlling excipient.
The project was driven by intermittent dissolution failures in the product. Data revealed that 1/3 of the tablets failed S1 testing, while 1 out of 12 tablets failed S2. Subsequent analysis found that the tablets that failed dissolution had significantly different amounts of lubricant required for the formulation. The PAT study attempted to integrate in-line FT-NIR analytical technology in the final blending stage. The data were correlated against the bench top method that utilized ICP-MS.

Once in-line correlation was established a series of DOE’s were initiated to determine the op