OVERALL: This is the 25th anniversary of the PDA/FDA conference and many of the presentations provided evidence of the significant progress made in research and medicine during this time. It was fascinating to stand back for a few moments to see just how far we’ve come. The overall conclusion I take from the two days of the conference that I attended are:

  1. FDA Centers continue to make significant efforts to coordinate their activities, and within centers they continue to work to provide consistency in both review and inspections. A work in progress.
  2. The current mantra seems to be Quality rather than simply Compliance. I agree with the philosophy of the sentiment but if we have regulations then we can’t just throw them out the window. Judicious, well-considered enforcement focused in patient safety is certainly needed. I question, however, whether outsourcing facilities, compounding pharmacies and pharma manufacturers in the developing world can appropriately interpret this approach.
  3. The long awaited and highly ‘advertised’ FDA programs including Quality Metrics, the New Inspection Protocol, and Guidance on Combination Product GMPs remain works in progress awaiting implementation, years after they were first announced.

The following does not represent all sessions that were conducted but only those which I attended.

PLENARY SESSION 1, A Patient Perspective. Prior to the ‘patient perspective’ presentation, Peter Marks MD, Director of CBER, addressed challenges faced by FDA in review of novel products, particularly those for very small patient populations. He focused on the CAR T-cell contingent of products under development. When his slides were prepared, the FDA had 105 INDs in place for engineered T-cells. Challenges for this type of product include those on both the manufacturing and medical side. Manufacturing concerns center on identifying the necessary manufacturing robustness to ensure the process is well controlled. On the medical side are the issues of side-effect management and development of appropriate clinical trial design for treatments intended for small patient populations. The concern about side effect management is meant to achieve the result where these products are NOT limited to only a handful of hospitals who can adequately manage side effect but rather than be used in a broad group of hospitals throughout the country. Certainly a challenge!

David Fajgenbaum M.D. spoke as a patient who has Castleman Disease, and is the cofounder and Director of the Castleman Disease Collaborative Network at the University of Pennsylvania. It truly brings home the important of work on orphan drugs, and the importance of evaluating off label use of existing approved drugs for some of these conditions. The challenge, with the very few patients with rare diseases, is to track which of the off label use regimens are actually successful that will encourage and enable firms to apply for labeling that includes treatment of such diseases.

PLENARY SESSION 2, Center Updates provided the perspective of FDA staff from: International Activities, CBER, CDER, CDRH, CVM and ORA. FDA continues their effort to work more collaboratively among the various centers, ever the challenge in any large complex organization. My take home is that a lot of the actions industry has been waiting for remain works in progress with unknown implementation dates. For example, the New Inspection Project won’t be implemented any time soon. It will require additional IT technology to allow FDA to incorporate data into its risk assessment model. FDA is also working to ensure consistency among reviewers and investigators. None of the issues they are undertaking are easy to address, but many have been promised for years and seem to fluctuate depending on who is the Commissioner and which political party resides in the White House and controls Congress.

FDA stressed their intent and need to embrace new technologies, the example given was continuous manufacture, and an afternoon session, below, went into detail on work they have done in concert with industry in this area, including original product approvals and supplement approvals.

CONTINUOUS MANUFACTURING:  Of interest was FDA’s statement and graphic at the beginning of the session that approximately 2/3 of the drug shortages were due to ‘quality issues.’ I don’t see this as new information but rather it confirms previous published data, and demonstrates that this ongoing cause of shortages has not been resolved. Quality issues are particularly acute for shortages of injectable oncology drugs.

FDA presented their engagement with the industry in the area of continuous manufacturing and made claims that this manufacturing approach cuts costs, provides for better assurance of drug quality and reflects application of new technology. An industry presentation for the use of continuous manufacturing presented similar conclusions. The example in this case was in the manufacture of solid oral dosage form drug products.

AUDITS: This session included presentations from Zena Kaufman (senior advisor to PriceWaterhouseCoopers) and Anil Sawant (Merck), and a Q&A session moderated by Thomas Arista of FDA. This was not so much a ‘how-to’ session, but rather addressed more deeply the attributes of an effective audit program with a focus on behavior of auditors.

Anil Sawant raised the very important issue that auditors need to be well grounded in regulatory intelligence and the analytics behind the actual ‘intelligence.’ Auditors must be aware of new regulations and guidance as well as enforcement trends to ensure that the internal audit program addresses topics on which regulatory authorities focus.

The Q&A session brought out about discussion on:

1) Why haven’t we reached to the point where inadequate investigations (21CFR 211.192) is no longer the #1 observations on forms 483 the way it has been for approximately the past 10 years and

2)  Why do we continue with data integrity problems after the generic drug scandal of the 1980 where both FDA and industry were found with problematic behavior.

  • One individual suggested that we have ‘forgotten’ about Part 11 and computer system validation, and no longer take that with the same level of seriousness that we did in the past.
  • Personally, I’m not sure that the substantial requirements of Part 11 were ever internalized and applied except by a subset of the industry.
  • No one mentioned that all firms are trying to do more with less and corners are likely being cut because there may not be enough trained staff to perform the work in a diligent manner.
  • Also, no one mentioned the serious cost involved in remediation including re-doing validation, purchasing new equipment and software and sometimes wholesale changes in data collection and review. Some firms may simply be taking the approach that they will wait until they are ‘caught’.

Everyone seemed to agree that improvement in this area must originate with executive management and is not something that can be effectively implemented from the operational staff.



  • Hal Baseman. ‘Evolution Of Validation / Verification Over Time, Lessons Learned In Relation To Aseptic Processing Validation’
    • Defining moment in the industry with regard to validation of aseptic processing was a process failure in 1970-1971, when LVP bottles were found with microbial contamination between septum and overcap. cloacae and erwinia were the contaminating organisms. The drugs passed sterility tests but this demonstrated that testing alone isn’t adequate to establish process control and ensure product quality and safety.
    • The 1987 FDA guidance on Process Validation. Emphasis focused on documentation, reams of it, to demonstrate control. Replicate runs, usually three (3) were assumed to demonstrate process control.  But the industry still had process failures because not all process variables are addressed and controlled.
    • In 2011, FDA published a new approach to process validation that is more science based: 1) Lifecycle approach and 2) evaluation of scientific evidence.
    • Where do we go from here? Process Design to Process Qualification to Continuous or Ongoing Process Verification. So when is the process ‘validated’? A good question that must be justified by the manufacturer. 

Key Take Home message for success is the need is to design the process to meet control objectives.

  • Grace McNally (FDA), ‘Process Validation and Pre-Approval Inspection’
    • It is important to identify sources of process variability and then consider the range of variability for each input. Use DOE and multivariate analysis, simulations and modeling to investigate these sources of variability.
    • PAI program purpose: to establish readiness for manufacture, evaluation of process validation (either protocol or data), and data integrity analysis. Commercial parameters need to be included in the filing.
    • Selected examples of form 483 observations from PAI:
      • Acceptance criteria for mixing was not adequate, protocol was silent. Sampling plan was not adequate.
      • 90% yield criteria was not met even though firm declared that the process was validated. Outcome suggests that process was not robust and that there might be underlying equipment or process design issues
      • Inconsistencies between sampling number and the raw data and process validation sampling records.


  • David Anderson, Legacy (1997 definition and 2013 rule) Combination Products, Preparing a Design History File
    • The company started with FluMist as a legacy product and retrospectively determined where they stood relative to design controls. The biggest gap identified was the user focused risk assessment, so they performed one. Ultimately they assembled a package of data / knowledge that could be shared upon inspection that constituted a Design History File. While it would certainly be different that a DHF developed de novo for a new product, it met the intent of the regulations.
    • David proposed that drugs and device are developed separately and then a cover piece brings them together with risk and design for the two constituents. Bottom line, you don’t need to really re-do everything.
  • Melissa Burns (FDA) addressed recurring themes within the Office
    • Combination product manufacturer has overall responsibility for CGMP
    • Exempt devices are ‘consistent with the Class I use’, and therein is the complication
    • For legacy products; do not necessarily go back and reinvent, FDA is not trying to add unnecessary documentation and work.
    • Reserve samples for stability testing for device part may require alternative approaches to compliance
    • Convenience kit (such as first aid kits) just putting parts in a box.
    • What information needs to go into NDA/PMAs etc., and where do you put it? There are clearly differences among reviewers and this frustrates the industry.
    • Training being given to FDA investigators is a work in progress.

It is important to be able to map what data are available to demonstrate meeting the CGMP requirements.

  • INTERcenter consult process: Improvements are being currently piloted.  They are working to:
    • Get right staff involved early so that timelines for interactions set expectations
    • Tiered approach to streamline interactions
    • Defined roles and responsibilities
    • Revised forms for tracking consults.
  • Combination product initiatives:
    • Pre-FRD process
    • Role of Human Factors Studies
    • Post market safety reporting
    • Combination product CGMPs
    • Importing combination products and constituent parts
    • Combination Product Council (leaders from all centers to address cross cutting issues)
    • How will combination products fit into the functional area assignments with the implementation of the New Inspection Protocol. They are aware and working on it.