Part 2: What You Should Expect from A High-Functioning CDMO
By Mark Roache, VP of Cell and Gene Therapies for Quality Executive Partners, Inc. and President of BioQPartners, Inc. and Sarah Boynton, owner of Biopharma Staffing and Consulting Solutions, LLC
For the first time in history, medical science has advanced enough to offer cures for diseases that were previously treated only with maintenance therapies at best. The advanced technologies of cell and gene (C&G) therapies are driving this promise of a functional cure. This revolution is giving rise to large numbers of small companies (Sponsors) who are working as fast as possible to deliver that potential to patients.
This is the continuation of a series of articles aimed at discussing the issues and challenges faced by sponsors as they attempt to successfully navigate a development program to approval. This is Part 2 on the topic of working with Contract Development and Manufacturing Organizations (CDMOs). The first part of this topic (published earlier) discussed the initial decision to engage with a CDMO and some points for how to optimize selection of the right CDMO partner. This second part picks up immediately after selection and covers how to make the partnership between client company and CDMO mutually successful.
Working with Sponsors – Advice for CDMOs
While we may have touched on aspects of the CDMO world up to this point that are potentially problematic, not all CDMOs are made the same and in any case, we think that problems arise not because of sponsor or CDMO characteristics per se, but because there is a poor match and inadequate alignment. A well-rounded CDMO understands most sponsors, or clients they serve, are limited in their understanding of GxP and regulatory aspects associated with manufacturing complex C&G therapies. As a CDMO, your role in the relationship is to get clarity on what the sponsor values.
The CDMO can take a leading role here, as advisors and experts in Chemistry, Manufacturing and Controls (CMC) for the product. This is especially true if the project has decided to leverage as fully as possible the CDMO’s standard processes, analytics, and procedures for manufacture. In our experience, this works best when the CDMO does not just tell the client the outcome, but explains how this endpoint, or procedure, was derived. As with our math homework, points will be awarded for showing your work. This is both an opportunity and obligation and when it is done well, creates a foundation of trust and credibility. This is a strong client-retention mechanism for when times get tough (here comes Murphy, with his deviations again!)
On the topic of deviations, one aspect of CDMO behavior that leaves the authors frustrated is the propensity of some CDMOs to not communicate with their client when something goes awry, and a deviation is recorded. With such occurrences, CDMOs often wait to inform clients, especially if the deviation can be classed as minor, or no product impact (illustrating why this is important to specify clearly in the Quality Agreement). However, we have seen many times that the client has product or process knowledge that would help greatly with investigation and impact assessment for that same deviation, and in any case, they should know as soon as possible. While no client enjoys an error or failure, unlike fine wine, such things do not improve when allowed to age. Bottom line – communicate frequently, that way disclosure will not be the only communication and both parties can cooperate on a constructive path forward.
Ironically and in contrast to the communication on deviations, there is a temptation for CDMOs to agree to client requests, even when they may not support the longer-term goals of the project or be difficult to deliver. We strongly recommend that CDMOs feel comfortable saying a polite “no, and then offering the best path forward that has a high probability of addressing the initial driver for the request.
To act as a clearing house for these discussions, and to facilitate the decisions that will keep the project on track, it is best practice, and in the CDMO’s strong interest, to establish a governing committee for a project, made up of experts and decision makers from both organizations. This meeting should have a charter, agreed by both organizations, some aspects of fixed agenda, and should refer to the approved contract and quality agreement. Participants should have sufficient authority to make most decisions, and a clear escalation path for communication and to more senior decision makers on both sides.
Working with CDMOs – Advice for Sponsors
After selecting the right CDMO, then the real work starts. Sponsor companies can help both parties in the relationship by thinking ahead and establishing provisions during times of calm that will help when there’s a challenge in hand. Sponsors should fully support and embrace the CDMO’s introduction of their program management team and should adopt as much standard operating and technical practice as possible. Joint workshops early on will be beneficial here and should have the goal of outlining the commercial agreement and the quality agreement (QAG) that governs the joint program. This up-front work has the great benefit of relieving cognitive load from the team when dealing with problems later on.
We will not go into all the content necessary in both commercial contracts and a robust QAG. However, the following are some common issues and challenges that we have observed:
- Milestones related to project start are often very optimistic. The effort required to set up for a first GMP run, including material and component procurement, testing and reporting, batch record and other document authoring and approval and the training of staff for novel processes invariably creates a crunch, sometimes getting resolved just hours before that first batch starts. Challenge the scope and timing and allow for contingency.
- Responsibility for key decisions, especially approval of deviations, changes and material disposition needs to be explicitly clear. We discuss the specific disposition topic later in more detail.
- When things go wrong and a batch is in jeopardy, behavior can easily be influenced by concerns over financial responsibility. Do as much up-front work as possible to remove this influence from time-sensitive efforts to properly resolve a deviation.
Just as we can see when the work starts, it is equally important to know when the work should stop. This is easy to pinpoint on a per-batch level, since the disposition decision is so well defined. However, it is much less easy to define for programs that are dominated by development services from the CDMO and by the typical development program milestones. There is a strong temptation to keep aiming for perfection, or at least the next big thing, often a yield improvement or a process refinement. We strongly recommend that the program set up a joint Target Product Profile (TPP) early on which defines, in as much quantitation as possible, success for the development program. This reference document will allow the partnership (and specifically the joint committee) to make decisions on whether continued work is needed or can be deferred before a license application is compiled and submitted.
CDMO business model:
As noted above, each CDMO is a little different in their relative focus and blend of expertise across the development and manufacturing continuum. Nevertheless, both the CDMO themselves and prospective partners will benefit from clarity of focus and purpose of that CDMO. We will consider the two extreme cases of a CDMO that focuses almost entirely on development (which we will refer to as the cDmo model) and a CDMO focused almost entirely on manufacturing services (similarly referred to as cdMo). As we consider the extremes of this spectrum, we acknowledge the reality that strategy and culture are strongly related and that both will drive daily operation, communication style, hiring preferences, organization structure and allocation of responsibility and most other aspects of company identity. It is worth repeating that there is no right or wrong answer here, but knowing what one wants and needs from a partnership, and honest objective assessment of the candidates, will greatly improve chances of long-term success.
cDmo – Focused on Science, Technology and Intellectual Property (IP)
Such a CDMO will focus on scientific and academic capability and will display heightened ability to solve technical problems. Culturally, cDmo’s will show elevated levels of data transparency. Challenges might include naivety in areas of GMP, especially impact assessment of deviations (but not root cause analysis due to the high level of technical content in RCA). There might also be a strong drive for the client to adopt relatively novel technology with high projected value but less demonstrated success.
Contracts and commercial terms will be focused on the delivery of process and analytical sciences (i.e. the IP). Clients should pay attention in the commercial contract and the QAG to set clear expectations for quality systems follow up and ownership responsibility.
cdMo – Operational Effectiveness and Cost Management
Such a CDMO will focus on typical larger-scale manufacturing operations. There will be a focus on maximizing short- and mid-term return on CapEx and maintaining low idle times. Consistency is emphasized and valued, and so error avoidance and elimination is a priority. This might mean that there is a reluctance to support implementation of changes on necessary timelines and there might be a limited capability for some of the more basic science required for root cause investigation for deviations (but the need for impact assessment will be well understood since this is a product-centric quality issue in GMP).
Audit and Oversight:
Over the course of the relationship between CDMO and sponsor, there will be a need to perform routine audits and ensure oversight is keeping the process in check. Let’s begin the discussion with audits. A comprehensive audit typically involves a thorough examination of the CDMOs systems, processes, and procedures to ensure compliance. The main goal is to catch any gaps that may be found by regulatory agencies and create a plan to close them. Bottom line: are the applicable regulations and standards being met? Usually included as part of the Quality Agreement, sponsor and CDMO should agree upon how often and to what extent audits will be conducted. This can depend on the stage of development, the complexity of the process, and the risk involved. Taking a risk based approach prevents mundane audits that don’t provide much value. This can help prioritize areas for review and ensure that resources are allocated appropriately. As part of the agreement on timing, sponsor and CDMO will also need to define the audit process. Most CDMO’s have a process or procedure in place, which should be reviewed by the sponsor. Information about who will be conducting the audit, the areas to be covered, and how the findings will be reported must be defined so there is no confusion as to who is responsible for what. Here are some specific areas to cover when conducting audits:
- Quality management systems: This may include a review of the company's quality policy, quality objectives, and the processes in place to ensure that products meet quality standards.
- Manufacturing processes: This may include an examination of the procedures in place to manufacture the product, including raw material handling, equipment maintenance, and cleaning procedures. Other critical aspects of the manufacturing process include personnel and material flow. This is especially true for viral vector therapies, as cross-contamination events can easily spiral out of control.
- Testing and validation: This may include a review of the testing procedures in place to ensure that the product meets specifications, as well as the validation processes used to ensure that the manufacturing processes are reliable and consistent.
- Documentation and recordkeeping: This may include a review of the company's documentation and recordkeeping practices, including SOPs, processing records, and other documents related to the manufacturing process.
- Training and competence: This may include an assessment of the training provided to employees and their level of competence in performing their roles. It is imperative that those performing the actual manufacturing work day-to-day have evidence of training for the processes they are supporting. Keep in mind that reading an SOP and hands-on practice are very different types of training, with the former being self-guided learning. This means you are relying on people to not only read the SOP in entirety but be able to apply what they read during the manufacturing process. At the very least, SOP read-only training should include a quiz or other type of assessment to gauge understanding. All too often, “human error” is seen as a common root cause to deviations, when in reality the initial training provided was inadequate.
- Risk management: This may include a review of the company's risk management processes, including the identification of potential risks and the steps taken to mitigate them.
- Facilities and equipment: This may include a review of the company's facilities and equipment to ensure that they are suitable for the manufacturing process and are maintained appropriately.
A comprehensive audit may be conducted over a period of several days or even weeks, depending on the size and complexity of the CDMO and the scope of the audit. Therefore, sponsor companies with limited quality auditing and compliance capability should strongly consider engaging with experienced consultants to provide objective and measured assessment and manpower to conduct the audits. The resulting audit report should provide a detailed assessment of the CDMOs compliance with expected standards, as well as any recommendations for improvement.
The best way sponsors and CDMOs can limit the findings during an audit is by implementing consistent oversight. While auditing is performed at various times throughout the manufacturing agreement window, oversight is happening real-time. This allows companies to quickly course-correct any behavior that is not up to the agreed upon GxP standards. The primary parties involved in oversight are the CDMO's quality assurance (QA), quality control (QC) departments, and manufacturing management, as well as the client's own representatives. Oversight should involve coaching and correction by the QA and manufacturing management teams. This may involve regular training sessions, feedback on performance, and coaching on specific aspects of the process where improvements are needed. The goal is to ensure that everyone involved in the process is working towards a common goal of producing a high-quality product that meets all regulatory requirements and the sponsor's specifications. Coaching and feedback is often overlooked when the speed of getting to market takes over. When poor behaviors are not corrected in a timely manner, they often become the norm. Often times, even when new employees are provided a robust onboarding, they will quickly fall into poor behavior once they become wrapped up in the manufacturing process, not questioning those around them.
There is a clear opportunity in this real-time oversight to use a Person-In-Plant (PIP) arrangement. CDMOs often display reluctance to agree to having a Sponsor employee present as a PIP during important operations. To be fair, this may be due to bad experiences in the past, where every action becomes a de facto audit. To be effective, the PIP should work as a collaborating expert, decision maker and communication conduit, with the shared goal of successful operations. Sponsors should motivate and reward PIP staff for avoiding or correcting problems, not for discovering relatively minor compliance issues. We strongly recommend the PIP arrangement as beneficial to both Sponsor and CDMO, but leaders on both sides need to make sure that this is a collaborative relationship not a competitive, or even confrontational setup.
Readers will note that we do not refer to the decision point on the fate of a batch as the “release” decision. This is deliberate, for two reasons: Firstly, we want to avoid setting a preconception of release, when batches may also be rejected, of course, always on consideration of quality, compliance, and merit. Second, we want to make clear that there are differing responsibilities for the sponsor and CDMO quality units in the disposition process.
We strongly recommend considering disposition as the culmination of two decision steps:
Step 1: Technical release. This step can, and should be performed by the manufacturing organization. In this case, the manufacturing organization means the CDMO (and specifically the CDMO Quality unit) and perhaps the Sponsor’s QA oversight function for the CDMO. The intent of technical release is to state, at the highest level, that the batch was made in accordance with all approved processes and instructions, has been tested according to defined requirements and specifications and the results of those test data are compliant with specifications, is compliant with all applicable regulations and that all aspects of the quality system have been followed (including especially the proper handling of deviations and changes).
Step 2: Release to Purpose. This step can only be performed by the sponsor, since it is the sponsor who will hold the authority on any conceivable purpose of administering drug product from this batch. In general, purpose is either commercial distribution in accordance with an approved license (which the sponsor will eventually hold, but not the CDMO) or the purpose may be supply of a clinical trial (in accordance with a clinical trial protocol and quality system, which the sponsor has at least approved and may own fully).
The benefits of this two-step disposition include:
- Responsibilities are made clear and maximally leveraged for both Sponsor and CDMO
- CDMO is relieved of the unfair burden of appearing to release to purpose (about which they are very under-informed)
- Facilitation of specific and selective release to different purposes, such as two separate regional markets, if the requirements and specifications are not fully aligned
- Facilitation of management of release to specific markets based on history of approved changes in each market (the same change may be approved by different health authorities at different times)
It is worth noting that in a case of highly converging evolution, most major, international pharma companies have adopted a similar two-step disposition process for some or all the above reasons.
In the explosion of novel ideas and technologies in the C&G space, there is great opportunity for sponsors to partner with CDMOs, as we jointly seek to serve patients. Deliberate decision making is key, and this is particularly true at the partner selection phase. The success of the subsequent partnership depends on there being an aligned and carefully managed set of goals and expectations, with both parties owning the relationship jointly. Objective assessment of strengths and focus (sponsors focus on medical and biological science in treating disease, CDMO focuses on development and operation of manufacturing technology) allows both parties to excel. Sponsors and CDMOs that navigate this definition, selection and partnering process effectively will be rewarded with faster approvals and time to market as the C&G development timelines accelerate.