Selecting a CDMO Part TWO: Writing the RFP

At this stage, there is a rare opportunity to ‘multi-task” or more properly, put a couple of key workstreams on parallel paths. Given that the RFP writing process will likely take some time, now is a good time to connect with candidate CDMOs and put nondisclosure agreements (NDA) in place. The RFP will contain confidential information that should not be publicly disclosed. It is necessary then to constrain the RFP recipients so that they maintain confidentiality during and after the Proposal generation process.

The list of candidate CDMOs can and should be generated in multiple ways. The first and most straightforward is to connect with organizations by reputation. This can be based on previous first-hand knowledge within the sponsor organization or through their professional networks. Keep in mind it’s critical to take a cross-functional approach and seek input from stakeholders throughout the organization. Regulatory Affairs and Quality are important stakeholders, as they may have access to regulatory actions such as 483 observations, Warning Letters and Consent Decrees as well as recent Approvals, that should greatly inform the selection process.

Another way to identify candidate organizations is to review recent press releases available in industry publications. These often highlight news, both good and bad, generated by CDMO organizations. This is an excellent way to ensure that assessments are based on current information and not ‘yesterday’s news’.

It is important to focus efforts on CDMOs that have core competencies that align with the product at the current or future phase of production. For example, there is little use in approaching a CDMO that only works on commercial programs when the molecule is still in the early stages of clinical development. Likewise, if the RFP scope includes late-stage process development and commercial manufacturing, the sponsor should avoid CDMOs that have no commercial product track record.

While the NDA process is underway, the organization should begin writing the RFP. This task usually falls to the CMC or Manufacturing team leader or project manager for the program. I’ll now outline the major sections of the RFP, with an eye toward making the document comprehensive and actionable. A well-crafted RFP can be used directly in the subsequent assessment process to facilitate a fair, comprehensive review of proposals.

Introduction: This opening section describes the purpose of the RFP. Specify the main scope elements that should be covered by the proposal. These may include manufacturing services (specify the stage of manufacturing, i.e. cell lines, bulk drug substance, drug product, packaged and labeled finished drug product etc.), the development scope (i.e. process development, cell line development, analytical development) and any testing (release, stability, characterization). Avoid too many details, as these will be outlined comprehensively in the ‘Scope’ section later.

Product Overview: This section outlines both the sponsor organization and the product to be manufactured. It is a mistake to assume the CDMO candidates will know anything about the sponsor or their product, so its important to ensure this section provides them with enough knowledge to ensure alignment with their capabilities. For example, the product attributes should be clearly described:

• Company profile in brief: This can often be cut and pasted from the sponsor website. Provide a link to the sponsor website.

• Product designation, include USAN/INN designation if known and a short description of the target indication(s).

• Type of product, i.e. small molecule or biologic and a description of the final dosage form (even if the RFP is targeted to an API or drug substance CDMO).

• The stage of development, including a brief, high-level summary of non-clinical data (for a pre-IND molecule) or human clinical data. It is important not to overshare at this point, especially if the clinical data has not been shared publicly. Be specific enough so the CDMO can provide a phase-appropriate proposal without being so specific that important clinical results are shared prematurely.

Scope: This is the real meat and potatoes of the RFP. The section should follow the basic outline described in the Introduction, augmented with additional scope elements and details. This section will form the basis for the resulting Proposals, so its important not to leave anything out. Doing so may result in costly Change Orders, to add back the missing deliverables. I will outline as many elements of scope as possible here, knowing that in the real world, not all of these will be needed in every situation. I’ll generate lists for products in varying stages of development. I will also base these lists on the assumption the product is a recombinant protein such as a monoclonal antibody. The deliverables for a small molecule or a cell or gene therapy product would be substantially different.

Pre-IND Programs: This assumes no previous GMP or GLP-enabling manufacturing and only a limited amount of purified drug substance is available (likely generated in a non-GxP laboratory).

• Cell line development and qualification; non-GMP (used for initial process and analytical development) and GMP cell banks (for GMP manufacturing)

• Drug substance and/or drug product process development at bench scale. Products from this scale may be used for IND-enabling toxicology studies in some cases, so careful assessment of capabilities is necessary. For first generation processes, additional characterization studies are needed to meet regulatory expectations. Examples include viral clearance studies, extractable/leachable studies and impurity clearance studies for single use systems. Process development raw data and unit operation process development reports.

• Scale up to non-GMP pilot scale and GMP production scale (define these if you have a reasonable estimate of product quantity requirements. Often, this may not be known with certainty until cell line and process yields are determined. Reasonable estimates can be made using conservative industry predecessors such as 2-5 g/L gross cell line yield and 60% overall purification yield for bulk drug substance).

• Clinical production, testing and release

• Development and qualification of analytical methods for in-process, release and stability testing. All raw data generated during method development, as well as method qualification protocols and reports.

• Stability studies. Keep in mind these studies run for years.

• Product Storage (short and/or long-term if required)

• Regulatory support

Mid to Late-Stage Programs: Here the requirements may vary considerably, depending on how close the program is to filing for marketing approval (BLA, MAA for example). For our purposes, we will assume this is a program preparing to initiate registrational clinical trials. As for earlier stage programs, the sponsor should specify where raw data and final reports are expected during process and analytical development.

• Tech transfer, if moving the program to a new CDMO, with associated TT reports

• Scale up to commercial scale if necessary

• Clinical production using the proposed commercial, testing and release processes

• Process optimization, if needed

• Product comparability studies (if major process or scale changes have been made or the site of manufacturing has changed)

• Process characterization studies, including additional viral clearance studies and studies to justify the proposed commercial specifications. A control strategy based on final or provisional CQAs that justifies Critical Process Controls.

• Validation of analytical methods

• PPQ campaign, with associated process validation studies needed to support commercial production (resin reuse studies, membrane cleaning studies, intermediate product hold studies etc.)

It is not uncommon for the sponsor to divide complex projects up into phases and to request that proposals follow this structure when writing up their proposals. This facilitates a good governance process, where the key deliverables for each phase can be reviewed by a joint steering committee to assess completion and delivery. The assessment of deliverables can then be used as the basis for milestone payments as well as a means to keeping the project on track.

Quality Considerations: Another parallel workstream that can be undertaken during the RFP process is defining the terms of the Quality Agreement (QAA). This important document is used to hold the CDMO accountable for meeting the Quality standards of the sponsor, which in term are based on established regulatory guidance and statute.

The QAA is negotiated separately from the MSA, or other contractual terms and this task is led by the Quality unit. Ideally, the Quality unit has specified internal requirements for QAAs in an internal SOP. As such, the Quality unit may desire to include high level elements of scope in the RFP to ensure the CDMO candidates understand sponsor expectations. This is particularly important when it comes to the implementation of shared Quality systems (systems that overlap the two organizational quality units). These systems include batch release, change control, deviation management, CAPA, shipping authorization and many others. For example, if it is the requirement that all deviations must be approved by the sponsor Quality unit, this should be specified to ensure the CDMO Quality unit can accommodate this requirement without violating internal SOPs.  In addition, Quality expectations regarding on-site interactions (PIP, PAI prep, etc.) and regulatory support can be outlined here as well.

Ideally, the final selection decision should include the inputs from the Quality unit. Often however, formal vendor qualification often occurs after the selection has been made, limiting the impact of Quality on the decision. I do recommend the Quality unit be involved as much as possible during the RFP process to mitigate this situation.

Program Governance:  Often times, the notion of a formal program governance structure is not considered until after the program has been initiated and problems begin to emerge. I always suggest the sponsor consider how they would like to approach join oversight of the program in advance. Points to consider include:

• Overall program success: Regular review of timelines and deliverables is essential to keep the CDMO accountable. Nothing gets attention like those things that get regular scrutiny.

• Assessment of milestone completion: This goes beyond hitting a target date. The quality of the deliverables needs to be assessed as well.

• Forward planning: Typically, detailed project planning occurs for the near-term deliverables at the inception of the program, leaving later deliverables ‘up in the air’. A joint forum allows for a mechanism to align sponsor requirements with CDMO constraints. This also creates a mechanism to raise new issues as the business context of the sponsor changes (i.e. unexpected clinical results, regulatory actions or changes in finances).

• Dispute resolution: Unfortunately, differences of opinion will occur. This forum provides for escalation of issues to the right people who are empowered to address the issue(s). This is also the forum to review and approve/deny proposed changes to the agreement (by either party).

• Team building: It is not a cliché to characterize the CDMO relationship as a true partnership. Although the interests of each party never align perfectly, it is always true that program success is in everyone’s best interest. Close relationships require care and nurturing and steering committee members should use this opportunity to get to know each other, share important program and company developments from both sides and to improve a relationship that is likely to persist for years.

Terms and Conditions: Specify sponsor organization expectations regarding the form and substance of the business terms proposed by the CDMO candidate. It is recommended that costs be based on the delivery of key deliverables and meeting milestones, rather than on labor hours or other effort-based criteria. It is not the sponsor’s obligation to manage the workforce of the CDMO, so it’s best to base the contract on fulfillment of key obligations rather than the work it takes to achieve them. Keep in mind that incentives and penalties can also be applied to the achievement of deliverables as an inducement to maintain adherence to the approved project plan.

Selection Criteria: In this section, the sponsor should outline the basis on which the selection will be made. These can include criteria such as project management systems, cost, speed and adherence to scope as outlined in the proposal and quality/regulatory factors. It may also include the outcomes of any interactions such as proposal review meetings (virtual or F2F), site visits, GxP audits etc.

Timeline: The sponsor should include an outline of the planned timeline for making the selection. Allow yourself sufficient time to receive proposals and review the leading candidates in considerable depth. Also allow time for obtaining internal executive and financial approval.

Submission Process: Who, when, how and where to send proposals.

The draft RFP should be subject to comprehensive cross-functional review before distribution to CDMO candidates. Obviously, this should include business, financial and legal reviews to ensure internal alignment.

In conclusion, it’s important to use the RFP process to drive accountability and to align expectations down to a finite level. The sponsor should make their needs known and use the willingness of the candidate companies to comply as one gauge in the selection process. We will take up this final aspect, evaluation and selection, as we conclude this series next time.

Next Time: Part THREE: CDMO Evaluation and Selection