Part 2: Challenges and Solutions for University Technology Commercialization - License Terms

In Part 1 of this series, I touched on some of the unique considerations for advancing early-stage innovation beyond the academic setting. For this article, I’d like to dig deeper into the topic of university expectations for monetizing these assets and how they may or may not align with reality outside the ivory tower. I realize that this is often a difficult topic and that there are numerous perspectives, but my overarching goal is to have a constructive discussion that presents the viewpoint of one of the few groups that gets involved at the very beginning of the “valley of death,” the no-man’s land between basic research and it becoming an investment- and/or partnering-ready asset.

First, some background on my firm’s model. VIC Tech identifies life science technology (therapeutics, vaccines, medical devices and diagnostics) from universities, national labs and institutes in the US and abroad. We form new companies (newcos) with a seed investment, license the asset into the startup, and provide business and research expertise in the form of an interim management team during the first 2-3 years of operations. During this time, we typically allocate several follow-on rounds of funding, teeing the company up for a Series A. Our primary objective is to advance the technology as far along as possible, thus de-risking it in the process and enhancing its viability as a future, life-saving product with significant commercial opportunity.

As such, one of the initial go/no-go decision points during our due diligence are the terms of the license to the underlying intellectual property (IP). For this, I will provide a case study, doing my best to give a balanced presentation of the licensor’s (university) and licensee’s (newco) rationale for each of the primary components. In this example, the IP is in the form of several submitted US provisional patents. The PCT- and national-stages are 6-12 months in the future. Our hypothetical technology is an optimized cancer therapeutic, with strong in vitro and in vivo data in accepted, disease-specific animal models. The inventor is a top-notch, well-respected thought-leader in their field with publications in high-impact, peer-reviewed journals. They are also highly motivated to move their life’s work beyond the lab and on the path (albeit long and costly) towards the clinic. Equally important is that they are willing to participate as a part time CSO in the startup while maintaining their faculty position, which often entails assisting with writing SBIR/STTR proposals, and serving as the Principal Investigator (PI) on subawards allocating a significant portion of the grant to back to their lab. With these and other components of our diligence checklist in place, let’s turn our attention to license expectations.

Field of Use

This is a restriction in a patent license that limits the scope of the licensed patent rights to a defined application or geography and is dependent on the claims of the IP and in which countries it will be filed. Ideally, the claims are sufficiently broad to cover multiple applications of the technology without risking non-allowance by the examiner. Much of this is based on depth and quality of the proof-of-concept data included in the filings.

In this example, the inventor and their team have results for several different types of cancer, including breast and prostate, and have claimed utility across all solid tumors. From the university’s perspective, it could be optimal if they license rights to individual applications to more than one entity, diversifying their risk and potentially increasing the revenue stream by having multiple licensees each having similar terms. As the investor and creator of the newco, I want to avoid at all costs having another group working on the same technology due to the fact that: 1. If they fail, it casts doubts on what my company is working on, regardless of whether or not it is because the science didn’t work or they made significant operational missteps; and 2. They will likely be competing for the same investors, partners and grant funding. The critical consideration here is striking a balance between control of the primary uses of the IP with the requirements of the university that specific amounts be allocated or development milestones be met for too many applications. If the expectations of the TTO are realistic, efforts by the licensee to pick one or two primary applications to pursue in earnest, while allocating modest resources to validate several secondary opportunities within a defined timeframe (2-3 years), is usually mutually satisfactory.  

Exclusivity

As is the case with the Field of Use, the university could increase patent-derived deal flow by licensing non-exclusively to more than one entity, which may be a potential non-starter for any interested party. There may be specific situations where this makes sense for both groups, especially if the IP is not central to the licensee’s overall strategy. With our example oncology asset, we do require exclusivity to ensure that no other party is working with the asset and possibly muddying the waters regarding who has control.

License Fee 

Now we are getting into quantifiable, monetary components of the negotiation and this is when the type of licensee becomes extremely relevant. For a larger organization—such as a pharmaceutical company—interested in obtaining the technology, the desired balance of upfront versus longer-term payments is quite different than what a startup like our hypothetical company will want. The big pharma player has the wherewithal to pay more, earlier in the license timeline (license and maintenance fees, milestones, IP costs) while wanting to limit the later components (royalties, sublicense revenue). Alternatively, our scrappy newco very much prefers to retain as much of the initial seed investment as possible by backend-loading the deal structure. Ideally, we try to limit the upfront license fee to 1% of the formational funding, i.e. $5K on a $500K round, with some degree of flexibility, while the larger entity would probably be able to pay many multiples more. [Point of emphasis: View all terms in aggregate, not in isolation; case in point, some licenses include annual maintenance fees, minimum annual royalty, and milestone payments which may also be due in the first year.] What the startup can offer to the university that the pharma company cannot, however, is equity with significant potential upside (see below).

Equity Stake

As one can imagine, license terms vary widely from university to university. Expectations for which terms are included and the expected range for each can be quite different from one institution to another. Equity in the newly formed company is not always requested but is increasingly common. We have seen values of 5% up to 15% in the term sheets we receive. Our position is that 5% is a reasonable equity stake to compensate for the reduced upfront payments compared to licensing to a large pharma company (if such a license to a larger company is even possible which, in the case of most IP at the stage available for license from universities, is not the case). Sometimes, the university will request anti-dilution protection up to a specified dollar value, often $2M. Providing one stakeholder anti-dilution can misalign incentives. None of the founding shareholders and investors in the newco have anti-dilution protection. Our approach is that—by building the worth of the company through reaching significant milestones—future rounds of investment will be at substantially higher pre-money valuations, thus limiting the dilution for holders of common stock. The underlying philosophy is that we all benefit from the growth and positive performance of the newco, and conversely, no one entity should be shielded from the risks associated with developing early-stage, science-driven innovation. In certain situations (more advanced technology, more mature IP, multiple large market opportunities) we may exceed the 5% target, but—by and large—this is the number we’re looking for.

IP Cost

In all but the most well-off private institutions (those with large IP budgets), there is significant pressure to limit IP spending and to recoup sunk costs as quickly as possible, as well as off-load future expenses. As a result, most US-based universities will not absorb the cost of pursuing international patent protection and rely on the licensee to bear this expense, which often exceeds $50K initially for the top pharmaceutical markets. If—in some cases—the national stage deadline has passed, then US rights are the only option and it is up to the licensee to find creative ways to create new US IP which then opens the door for future international filings.

For our example technology, the IP outlay by the university to date is $25K, which is roughly the mid-point of the range we typically see ($10K to as high as $100K). This wide variation is primarily dependent on the type of IP counsel used to prepare and submit the patent applications. An in-house IP resource usually keeps costs down, while a big-name external firm may bill at an hourly rate of $1K and higher. Given that the initial seed investment into the startup is a modest $500K, we prefer to pay down the $25K over the course of 1-2 years, with the expectation that new expenses will be reimbursed within 30 days of invoicing. Top of mind for the startup will be to identify more cost-effective IP counsel should this be a concern.

Another consideration is who owns future IP. This depends on whether or not the new innovation was created by the company, the university, or both. If it is both, then the mechanism through which the research was funded becomes important: Was it though sponsored research or a fee-for-service agreement? Typically, the former entails IP rights to the university, while the latter does not. Co-owned IP should include first right of refusal for an exclusive license to the startup.

Royalty

Simply put, royalty is the share of revenue the licensor (university) receives from sales of the product based on the licensed IP. Setting a realistic rate is essential for several reasons. Our startup’s exit strategy is acquisition by a large pharmaceutical company, likely with initial efficacy results from a Phase II clinical study. Potential suitors will not accept high royalty payments, and would want the license to be modified if needed. Another consideration is the impact on sublicensing since the royalty rate will pass through to sublicensee. For example, if the royalty is 2% and we sublicense at 6%, then 1/3 of the royalty goes to university and 2/3 to the company as the sublicensor. Early-stage therapeutic technologies are typically licensed for low-single digit royalty rates. As such, our startup proposes 1% to 2% for future marketed therapeutics. If this were a medical device, diagnostic, or non-FDA-regulated product—with shorter timelines and lower development costs—then higher rates are usually justifiable: 3% to 4% for medical devices and 4% to 5% for diagnostics are reasonable boundaries.

Minimum Annual Royalty

I’m not a fan of licenses that include minimum annual royalty (MAR), maintenance fees and milestone payments. When combined, all three of these can quickly drain the licensee of cash and limit its chances of success. In many ways, one can view these elements as a tax or penalty on progress: You launched your product but sales uptake has been slow (as is often the case for pharmaceuticals), you still owe us X amount; Your company hasn’t gone bankrupt and still exists, please pay X amount; You started your Phase I clinical study, congrats…that will be X dollars. Our preference is to not have all three of these onerous components in the agreement, or lessen the financial exposure if they are included. For our case study, the university has asked for MAR beginning year 5. Our response is that—in a best-case scenario—it will take 10 years to take this early preclinical asset through clinical development and approval before any sales can be realized. Therefore, we counter that MAR should begin year 11 at $100K, and increase to a maximum of $1M in year 15 and after. We are comfortable with this amount given our assessment of the market opportunity of the product due to the significant unmet needs and competitive advantages. Note that MARs are waived if royalties on net sales meet or exceed the prescribed annual minimums.

Sublicense Revenue

In the event that the licensee sub-licenses rights to the IP to another entity—regardless of the nature of the relationship (development, commercialization, etc.)—a portion of any revenue generated will be given to the licensor. Universities often propose 20% to as high as 40%, while our goal is the lower value. Again—given the specific costs and duration of subsequent advancement of the asset—there may be some wiggle room. One approach I advocate for is staging or tiering the rates to reflect how long down the road the sublicence is executed. If it is relatively soon after the effective date of original license and the startup has not invested a tremendous amount in the technology, then a higher rate makes sense. If instead, the sublicence is inked 5 years later after millions of dollars of investment, a lower rate is equitable. With our oncology asset, we received and agreed to the following: 20% if sublicensed prior to completion of Phase I clinical trials; 15% if before Phase II; 10% if before Phase III; 5% any time after completing Phase III. 

Maintenance Fee

As mentioned above, the inclusion of yearly maintenance fees—especially if combined with MAR and/or milestone payments—is not ideal. Luckily, the licensor has opted not to require this term in the agreement.

Milestone Payments

Following a fair amount of back and forth, the following milestone payments were agreed to for the oncology asset: $125K upon successful completion of Phase I human clinical trial for every indication; $250K upon successful completion of human Phase III human clinical trial for every indication; $1.5M due upon first regulatory approval for commercial use for each indication in either US or EU; Upon cumulative net sales of $250M of each product, $1.25M is due; Upon cumulative net sales of $500M of each product, $3M is due. As you can see, the sales-stage milestone payment overlaps with the royalty/MAR and sublicence terms, and could be interpreted as double-dipping, which it is. All things considered, the absence of an annual license maintenance fee was the primary motivation for agreeing to this component. 

Additional Considerations

Defining the inventor(s) role in the newco is a separate, but equally important piece of the puzzle. We want them to be involved while still retaining their academic positions, which allows the startup to fund continued work in their lab through sponsored research and/or grant (e.g. SBIR/STTR) subawards. In addition, if their university conflict of interest policy allows them to serve as a part-time (i.e. 5%-10%) CSO of the company, we very much want them to be involved in this capacity. In consideration for this, they receive an equity stake in the newco, which is separate from the revenue share they have via their invention assignment agreement with the university. Most universities have a revenue share between 25% and 50% of all IP-derived proceeds. The inventor can also be a powerful advocate for getting to “yes.” In other words, they often recognize how difficult it is to find investors or partners at the interface between basic research and IND-enabling development, and want to move the technology forward without getting stuck in endless license negotiations. We always need to remind ourselves that the end goal is getting life-saving innovation through the challenging, decade-long development process.

Conclusion

Not to oversimplify, but the university perspective is typically (and understandably) one of immediate gratification, and not of shared risk and reward. There is often a perception that the technology is more mature than it actually is. Most TTO personnel do not have first-hand experience taking early-stage assets through preclinical and clinical development, and—as a result—do not have a reference point for the associated cost (tens to hundreds of millions of dollars depending on the point of exit) and time (years), and so the expectation for near-term renumeration can be noticeably off target. Our emphasis is that a rising tide lifts all boats, and is the reason why we try to structure the license to reflect longer-term gains rather than short-term payments.