Merck (known as MSD outside of North America) is invested in both vaccines and personalised cancers vaccines, the latter through a collaboration with Moderna which is in phase III. Merck has provided Keytruda for free in the phase II study of EVX-01. Furthermore, MSD participated in both the recent private placement and public offering. It is now the largest owner with around 15% of all shares. One can make the case for investing in Evaxion as an acquisition case, or there is at least a potential upside to the share price because of this. Merck could make great use of Evaxion’s AI platform in both bacterial vaccines and personal cancer vaccines. We believe it is unlikely that Moderna, being focused on mRNA technology, has the same capability to select the best antigens for their personalised vaccine platform, so it could make sense to engage PIONEER for this. The investment Merck has made in Evaxion together with the collaboration on EVX-B3 could be the seed for a later takeover bid. The EVX-B3 collaboration could be a way for Merck to get an insight into Evaxion and establish relations that could make a takeover transition more smooth. Management and insiders only hold 25% of all shares according to the latest prospectus, and less when warrants are included, so they could probably not block an offer if they wanted to.

Evaxion anticipates several important milestones in 2024. The company expects to have a negative cash burn of USD14m in the year. The goal is to cover this with income from licensing deals. EVX-B3 should bring in some money from MSD/Merck if preclinical development and evaluation is successful this year. The indication for EVX-B3 has not been disclosed, but it is against “a pathogen associated with repeated infections, increasing incidence and often serious medical complications, and for which no vaccines are currently available”. EVX-B1 is also available for out-licensing and the partner with which pre-clinical experiments were conducted this year could potentially take it over this year through a license. ERV-based precision vaccines could be ready for out-licensing after proof-of-principle is established in pre-clinical models in H2. These are not individualised vaccines, so making a deal will be straightforward. Management believes partnering discussions make sense upon strong pre-clinical data in H2, but further development may also be necessary. Licensing deals at such an early stage are not that common and would likely require strong data in several indications and a platform deal involving several EVR candidates.

It will be more complicated to make a deal with the personalised programmes EVX-01, EVX-02 and EVX-03 since they require the use of the PIONEER platform. Furthermore, oncology is becoming a less active deal space, especially for early-stage clinical programmes, with a rather saturated market (except for some niches like antibody-drug conjugates or CAR-T therapies). Infectious diseases is becoming a more active deal space, in part thanks to COVID. We believe it is not a coincidence that Evaxion entered into collaborations in this area last year. Merck/MDS already has a collaboration with Moderna, so we think it would probably be more interested in acquiring PIONEER than licensing one of the programmes, such as EVX-01. In theory, however, a deal with another company is possible upon positive results from the ongoing phase II study.

We believe there is great potential in the checkpoint response prediction tool (AI-DeeP). Considering Evaxion maintains it is superior to other biomarkers such as PD-L1 expression, this could change the way such drugs are used. Currently, several checkpoint inhibitors are indicated with certain PD-L1 expression levels based on biopsies, which is not the most efficient way of deciding whether to use checkpoint treatment. The prediction checkpoint tool could be used as a companion diagnostic for new drugs in clinical trials, which would reduce costs and increase the likelihood of approval. It could also be developed in the clinical setting for more general use, which would be more complicated and expensive but with a significantly larger market. A collaboration with an important actor in this latter space could change the investment case for Evaxion.

On March 19, Evaxion hosted an R&D day. New information on the AI technology employed was presented, among other things. It has been termed AI-Immunology and is based on several modules that can be combined. The combination of different modules creates the various models Evaxions has, such as PIONEER, EDEN, etc. An example of a module is EvaxMHC, which calculates whether antigens fit in the MHC molecule. If it does not, an immune response cannot occur. Another example is the module Personalized Design, which is plugged into PIONEER to render personalised vaccines. New modules can be added expanding the functionality of AI-Immunology. AI is very hyped. It would be positive for the share if the company could find a way to ride the AI-wave.

As shown below, operating expenses have declined significantly in Q4. The new management has been very effective in implementing cost control. it has guided for a cash burn of USD14m in 2024, so we should see a further decline in the coming quarters.

The cash position at the end of 2023 was USD 5.6m. Together with the proceeds from the public offering in February 2024 (USD12.7m) and with potential funding in the future under the ATM program, this will fund operating expenses and capital expenditures into April 2025 (under the assumption that pre-funded warrants from the public offering are exercised before February 2025). USD4m related to the pre-funded warrants from the rights issues, which were issued instead of shares because of technical reasons, are held in an escrow account and expected to be released gradually as the warrants are subscribed.

We have included a full terminal value for the Pioneer programmes (EVX-01, EVX-02 and EVX-03), since personalised drugs cannot be patented (meaning that there is value left in the candidates after our forecasting period). The main protection comes from owning data and AI-models trained on this data. The data is proprietary and not easily available to third parties even if patents related to the software lapse. We see Pioneer as a software that can be continuously improved, leading to a network effect with more patients treated, i.e., the more customers, the better the product becomes. Generics or biosimilars would, therefore, never substitute the Pioneer vaccines. Even if this is the case, we believe it will be difficult to maintain high prices due to reimbursement policy, especially with the Inflation Reduction Act in the US. The same goes for Europe. We have, therefore, reduced the price in the final year of our estimates (2042) to 30% since we believe the authorities would negotiate a much lower price (after 13 years of sales for biologics, as per the Inflation Reduction Act).  

We increased the likelihood of approval of EVX-B1 to 12% (10%) but also added one year of pre-clinical development.

We have not yet included the checkpoint response prediction tool (AI-DeeP) to our valuation because the market opportunity has not been clearly defined and a partner must be secured to develop it.

We have calculated the fully diluted value using 13 million shares (with 7.8m shares outstanding including the prefunded warrants). Our bull case is USD48m and our bear case is USD0.