"With This MSA, I Thee Contract:" The "Make-or-Breaks" In CGT Outsourcing Negotiations

By Anna Rose Welch, Editorial & Community Director, Advancing RNA

In the last three installments of this four-part article, I’ve shared a wealth of knowledge from two CGT outsourcing “goldmines,” Adam Haskett, head of external manufacturing for cell therapy company CARGO Therapeutics, and Joseph Graskemper, the director of external network strategy at Biogen. Our journey together has been vast, starting with an overview of the past, present and future of outsourcing capacity and capabilities. This evolved into a discussion on the current strengths, weaknesses, and much-needed innovations in the development of CGT manufacturing platforms, and how biotechs and CDMOs can define their risk-sharing and IP expectations in a Master Service Agreement (MSA).

By the time our conversation reached its conclusion, we’d moved even more deeply into the inner workings of the MSA, with Haskett and Graskemper offering their (non-IP-related) make-or-break factors and what to look out for when contracting with a CDMO. After all, as Graskemper nicely explained, “Over the course of my career, I’ve seen developers struggle the most with getting the right terms in the MSA, because once that MSA is executed, that’s it. Trying to go through a renegotiation amendment is difficult, especially when you’re not in a position of strength.”

To help us all feel a bit stronger in the face of a negotiation, Haskett and Graskemper did not shy away from providing their go-to contracting best practices and tips for tackling an MSA and Quality Technical Agreement (QTA). While our discussion was far from exhaustive, they singled out several potentially showstopping areas that demand greater prescriptiveness in our MSAs for the sake of much stronger and long-term outsourcing partnerships.

Defining Your Value Proposition Upfront

As Haskett explained in our previous article (part 3), it behooves each biotech to align around the most important priorities in their product development — whether it be increasing titer or reducing impurities. But I appreciated Haskett’s reminder to look beyond the “limited” scope of the product and think about how our outsourcing relationships can or will fit into long-term corporate objectives. After all, what happens to our company and its pipeline in the short- and long term will impact what we may need — or no longer need — from our outsourcing partners.

For those of us working in small biotechs, there are a few unique, long-term considerations that deserve to be (and should be) voiced in negotiations. For example, as Haskett went on to define, we may be faced with acquisitions or co-development deals that could result in manufacturing being brought in-house. Obviously, these potential growth milestones demand upfront conversations about how such flexibility can be baked into an outsourcing partnership and contract.

In particular, he pointed to the cell therapy side of the industry, in which we’re seeing a greater number of companies transitioning from mid-stage to late-stage development. Such a transition demands much closer attention and goal setting around, for example, decreasing cell therapy turn-around-times and reducing manufacturing failure rates for commercialization.  

“These factors are not commonly defined in scopes of work and MSAs — at least not from what I’ve seen,” Haskett added. “But it’s critical you’re expressing your expectations upfront because the CDMO may have very different expectations on how they can and are willing to support a growth-stage company. For example, I’ve often seen CMOs specifying they will need anywhere from 30 to 45 days to release the product. It’s on you as the biotech to be prescriptive up front and emphasize when that won’t work for you and, perhaps more importantly, why.”

Aligning around a common vision — both internally and externally — for a company’s journey can also help a biotech stand out from the crowd. In a previous article that touched on supply-chain sustainability, one SME emphasized the importance of having a “good story” around how that particular CDMO’s services or supplier’s materials will enhance the product. Haskett reaffirmed that this level of relationship management is an essential component of successfully negotiating an MSA — and it starts by engaging your CDMO at its highest levels.

“You have to share your value proposition with the CDMO’s executive level and define why your product is valuable and how and why that CDMO will be valuable to your product,” Haskett said. “This creates a lot of momentum behind your mission and ensures your CDMO is clear on your expectations, which paves the way toward more tactical conversations on key terms thereafter.”   

Defining Batch Failure & Responsibility

It should come as no surprise that batch failures were high on both Haskett’s and Graskemper’s lists of key terms demanding greater contracting due diligence. Not only will batch failures happen because, to state the obvious, manufacturing is hard, but as Graskemper acknowledged, batch failures are difficult enough to sufficiently define let alone in a way with which both parties can agree.

For example, as Graskemper explained, an innovator might consider not achieving a certain yield a batch failure early in the manufacturing process (i.e., an engineering batch or first GMP batch).

“Good luck trying to get that into an MSA — it’s going to be incredibly difficult because the processes are typically not mature or robust enough to set appropriate yields,” Graskemper admitted.

In turn, he recommends companies start first by examining how their definitions of terms surrounding “batch failure” differ from those of the CDMO’s own definition of “batch failure,” as this will ultimately lay the groundwork for how an innovator identifies a failure under a CDMO’s watch.

Now, should the worst happen and a CDMO and innovator find themselves grappling with a batch failure, everyone’s attention will inevitably be trained on answering two big questions: “Who is responsible?” and “Who is going to pay for this?” If a biotech isn’t careful in their contracting, they may find they don’t have the necessary leverage to get the batch replaced or refunded. Though there are many ways to address this challenge in an MSA, for Graskemper, one way is to eliminate a word most of us don’t like to hear in any context, let alone within an MSA: “Gross.”

“When it comes to determining responsibility for a failure, I always look to see if the CDMO is held to a standard of ‘negligence’ or ‘gross negligence,’” Graskemper explained. “This is a very important differentiator; if they are held to a standard of ‘gross negligence,’ you lose leverage against the CDMO because proving ‘gross negligence’ can be very difficult.”

As he went on to explain, removing “gross” within an MSA makes it easier for an innovator to prove that, for example, an operator at the CDMO did not follow a batch record properly, and it’s now the CDMO’s responsibility to replace or refund the batch. Depending on the liabilities agreed to between CDMO and innovator, the CDMO could cover the batch price itself or the cumulative worth of all the materials that went into the failed batch.

Building on this, Haskett added that it is important to consider not only the CDMO’s liability, but also the point in the value chain where the liability occurs — for example, incoming receipt, storage, or during process development or manufacturing. “They each may have their own flavor of liability,” said Haskett.

It’s also critical to consider who owns the relationship with subcontractors and suppliers. “Typically, the gross negligence qualifier is specific to the CDMO unless ‘and representatives’ (or similar) is included,” he added.

Defining “Information Sharing”

As you can imagine, there are a lot of directions in which we can take “information sharing.” But throughout the course of our discussion, we homed in on how much access to its manufacturing batch and testing records a CDMO is willing to offer its clients. On the one hand, it’s not unusual to find CDMOs who limit transparency into this data for a variety of business reasons. For example, they may offer “view-only” access solely to quality team members and withhold certain highly proprietary records to maintain trade secrets. But for many of the CDMOs that do offer greater transparency, there’s an additional amount of scrutiny required, especially because paper and electronic batch records are, unfortunately, not created equal.

As Haskett pointed out, you’ll most likely see CDMOs include a statement in their QTA about giving the client access to their batch record. But if it’s an electronic batch record, in particular, this raises a few additional questions for the innovator. For example, does access to an electronic batch record equate to transparency into all fields of information, or is it limited to a smaller number of fields chosen by the CDMO? “Transparency into an electronic batch record is not as straightforward as a paper-based batch record,” Haskett added.

As we discussed in the previous article segment, IP considerations can also muddy these waters. CDMOs may have platform-esque processes, some of which, Haskett qualified, may be highly similar to the innovator’s own processes and/or involve similar technologies. In such cases, these similarities could seemingly promote more seamless access to the CDMO’s standard operating procedures and batch records. However, the biotech’s desired level of transparency into the CDMO’s SOPs and batch records may conflict with what the CDMO defines or considers its own background IP. Combining platform-esque IP and manufacturing know-how with electronic batch records and Manufacturing Execution Systems may exacerbate this issue, as it may become more difficult and time consuming to detangle.

There’s no easy answers or solutions here, Haskett admits. “It ultimately comes down to being very prescriptive and detailed upfront with your expectations for information access. For example, how do you define ‘disposition?’ Do you want the raw data for flow plots? A lot of clients will want that kind of information, but a lot of CDMOs will not be willing to provide this; some of the data packages comprise a substantial volume of information, or the CDMO may have other concerns related to data integrity and the file format in which it is provided. If it’s an electronic batch record, do you want access to the full end-to-end batch record with all fields expanded?”  

Additionally, it is important to understand not only what access rights the sponsor may have but with whom they can share that information. Regulatory bodies are typically included, but existing and potential strategic partners and investors, consultants, and contractors may or may not be included or permitted. For growth companies reliant on further investment and a contract workforce, it may be important to be able to share key aspects of the manufacturing process.

Additional “Bite-Sized” Tips

In addition to the three previous areas of concern within an MSA, both Haskett and Graskemper shared a few additional areas in which we as innovators may have a bit more control than it seems.  

• Mind The Relationship Between The QTA & MSA

Though we didn’t spend a huge amount of time delving into the QTA throughout our discussion, both Haskett and Graskemper emphasized that we should consider the QTA and MSA as a “dynamic duo” that need to be joined together “at the hip,” if you will.

It’s not unheard of that, in certain situations, the CDMO may claim that the terms in the MSA supersede the terms defined in the QTA. The same can be said about the QTA over the MSA — in fact, the MSA may even refer parties back to the QTA.

As such, Haskett strongly advised negotiating and signing these two agreements at the same time.

“You don’t want to sign the MSA and then try to negotiate the quality agreement at a later period of time,” he explained. “By that point, you’ll already be working with the CDMO, and you won’t have the same leverage to exert over the terms and conditions in the QTA that you had prior to signing the MSA.”

• Don’t Accept Cancellation Clauses At Face Value

In any MSA, you’ll most likely encounter a table depicting the different time-dependent percentages of pass-through costs owed in the case of a cancellation. However, as Graskemper admitted, there may be more flexibility here than it seems — especially given today’s capacity-rich environment.

“We may no longer need to ‘bite the bullet’ and shell over a large percentage to the CDMO if we are, for instance, 90-120 days out and cancel,” Graskemper added. “Now that there’s much more available capacity, you actually have a lot of leverage to get more favorable cancellation percentages.”

That said, however, he did urge innovators to be mindful of the termination events that are covered within an MSA. Termination for an innovator’s convenience, for example, is not always found in an MSA. 

• Do The Math In Advance Of Accepting A CDMO’s Pass-Through Costs Terms

When starting negotiations, it’s possible pass-through costs (e.g., materials, shipping, etc.) will start as high as 15 or 20 percent. Given that our understanding of material costs will most likely be hazy, at best, when entering development, Graskemper urged us all to build COGS models to ensure we don’t find ourselves on the hook for potentially millions in pass-through costs when all is said and done. By building such a model, we can then arrive at some assumptions around what pass-through costs would be tolerable and use this information to negotiate a lower pass-through percentage.

As Haskett added, such modeling can also be helpful in determining if you want the CDMO to manage material procurement, or if it would be more advantageous for the sponsor to procure from the material vendor directly, especially if you have increased purchasing power or a strategic relationship.

“You’ll usually have a starting point of around 15 percent,” Graskemper said, “but if you can get to 5-10 percent, you’re doing pretty well.” 

If you missed the previous parts, check them out here!

Part 1: From Capacity To Comparability: The Shifting Onus In CGT Outsourcing

Part 2: “Once Upon A Platform:” A Realistic Look At “Plug & Play” CGT Development

Part 3: What’s Mine Is (Maybe) Yours: Risk-Sharing In CGT Outsourcing

Like what you’re reading? Sign up to continue following ARW’s writings on the CGT + RNA manufacturing paradigm.