10-K

UNITED STATES

SECURITIES AND EXCHANGE COMMISSION

Washington, D.C. 20549

FORM 10-K

(Mark One)

For the fiscal year ended December 31, 2024

Commission File Number 001-41551

Acrivon Therapeutics, Inc.

(Exact name of Registrant as specified in its Charter)

480 Arsenal Way, Suite 100Watertown, Massachusetts 02472

(Address of principal executive offices) (Zip Code)

Registrant’s telephone number, including area code: (617) 207-8979

Securities registered pursuant to Section 12(b) of the Act:

Title of each class TradingSymbol(s) Name of each exchange on which registered

Common Stock, par value $0.001 per share ACRV Nasdaq Global Market

Securities registered pursuant to Section 12(g) of the Act: None

Indicate by check mark if the Registrant is a well-known seasoned issuer, as defined in Rule 405 of the Securities Act. Yes ☐ No ☒

Indicate by check mark if the Registrant is not required to file reports pursuant to Section 13 or 15(d) of the Act. Yes ☐ No ☒

Indicate by check mark whether the Registrant: (1) has filed all reports required to be filed by Section 13 or 15(d) of the Securities Exchange Act of 1934 during the preceding 12 months (or for such shorter period that the Registrant was required to file such reports), and (2) has been subject to such filing requirements for the past 90 days. Yes ☒ No ☐

Indicate by check mark whether the Registrant has submitted electronically every Interactive Data File required to be submitted pursuant to Rule 405 of Regulation S-T (§232.405 of this chapter) during the preceding 12 months (or for such shorter period that the Registrant was required to submit such files). Yes ☒ No ☐

Indicate by check mark whether the registrant is a large accelerated filer, an accelerated filer, a non-accelerated filer, a smaller reporting company, or an emerging growth company. See the definitions of “large accelerated filer,” “accelerated filer,” “smaller reporting company,” and “emerging growth company” in Rule 12b-2 of the Exchange Act.

Large accelerated filer ☐ Accelerated filer ☐

Non-accelerated filer ☒ Smaller reporting company ☒

Emerging growth company ☒

If an emerging growth company, indicate by check mark if the registrant has elected not to use the extended transition period for complying with any new or revised financial accounting standards provided pursuant to Section 13(a) of the Exchange Act. ☐

Indicate by check mark whether the registrant has filed a report on and attestation to its management’s assessment of the effectiveness of its internal control over financial reporting under Section 404(b) of the Sarbanes-Oxley Act (15 U.S.C. 7262(b)) by the registered public accounting firm that prepared or issued its audit report. ☐

If securities are registered pursuant to Section 12(b) of the Act, indicate by check mark whether the financial statements of the registrant included in the filing reflect the correction of an error to previously issued financial statements.☐

Indicate by check mark whether any of those error corrections are restatements that required a recovery analysis of incentive-based compensation received by any of the registrant’s executive officers during the relevant recovery period pursuant to §240.10D-1(b). ☐

Indicate by check mark whether the Registrant is a shell company (as defined in Rule 12b-2 of the Exchange Act). Yes ☐ No ☒

The aggregate market value of the voting and non-voting common stock held by non-affiliates of the registrant based on the closing price of the registrant’s common stock as reported on the Nasdaq Global Market on June 30, 2024, was approximately $99.1 million. For purposes of this disclosure, shares of common stock held by each executive officer, director and stockholder known by the registrant to be affiliated with such individuals based on public filings and other information known to the registrant have been excluded since such persons may be deemed affiliates. This determination of executive officer or affiliate status is not necessarily a conclusive determination for other purposes.

The number of shares of Registrant’s Common Stock outstanding as of March 24, 2025 was 31,351,480.

DOCUMENTS INCORPORATED BY REFERENCE

Portions of the Registrant's definitive Proxy Statement, to be filed pursuant to Regulation 14A under the Securities Exchange Act of 1934, for its 2025 Annual Meeting of Stockholders are incorporated herein by reference in Part III.

Table of Contents

Page

PART I

Item 1. Business 4

Item 1A. Risk Factors 64

Item 1B. Unresolved Staff Comments 111

Item 1C. Cybersecurity 111

Item 2. Properties 112

Item 3. Legal Proceedings 112

Item 4. Mine Safety Disclosures 112

PART II

Item 7A. Quantitative and Qualitative Disclosures About Market Risk 126

Item 8. Financial Statements and Supplementary Data 126

Item 9A. Controls and Procedures 126

Item 9B. Other Information 126

Item 9C. Disclosure Regarding Foreign Jurisdictions that Prevent Inspections 127

PART III

Item 10. Directors, Executive Officers and Corporate Governance 128

Item 11. Executive Compensation 128

Item 14. Principal Accounting Fees and Services 128

PART IV

Item 15. Exhibits, Financial Statement Schedules 129

Special Note Regarding Forward-Looking Statements

This Annual Report on Form 10-K, or the Annual Report, contains forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995 about us and our industry that involve substantial risks and uncertainties. All statements other than statements of historical facts contained in this Annual Report, including statements regarding our future results of operations or financial condition, business strategy and plans and objectives of management for future operations, are forward-looking statements. In some cases, you can identify forward-looking statements because they contain words such as “anticipate,” “believe,” “contemplate,” “continue,” “could,” “estimate,” “expect,” “intend,” “may,” “plan,” “potential,” “predict,” “project,” “should,” “target,” “will,” or “would” or the negative of these words or other similar terms or expressions. These forward-looking statements include, but are not limited to, statements about the following:

•

the timing, progress and results of our preclinical studies and clinical trials of our drug candidates, including statements regarding the timing of initiation and completion of studies or trials and related preparatory work, the period during which the results of the trials will become available and our research and development programs;

•

the timing of any Investigational New Drug, or IND, submissions, initiation of clinical trials and timing of expected clinical results for our lead drug candidate, ACR-368, ACR-2316, and our other future drug candidates;

•

the timing of any submission of filings for regulatory approval of, and our ability to obtain and maintain regulatory approvals for, ACR-368, ACR-2316, and any other drug candidates for any indication;

•

our ability to identify patients with the cancers treated by our drug candidates, and to enroll patients in trials;

•

our expectations regarding the size of the patient populations, market acceptance and opportunity for and clinical utility of our drug candidates, if approved for commercial use;

•

our manufacturing capabilities and strategy, including the scalability and commercial viability of our manufacturing methods and processes;

•

our expectations regarding the scope of any approved indication for ACR-368, ACR-2316, or any other drug candidate;

•

our ability to successfully commercialize our drug candidates;

•

our ability to leverage our proprietary precision medicine platform, Acrivon Predictive Precision Proteomics, or AP3TM, to identify and develop future drug candidates;

•

our estimates of our expenses, ongoing losses, future revenue, capital requirements and our need for or ability to obtain additional funding before we can expect to generate any revenue from drug sales;

•

our ability to establish or maintain collaborations or strategic relationships;

•

our ability to identify, recruit and retain key personnel;

•

our reliance upon intellectual property licensed from third parties and our ability to obtain such licenses on commercially reasonable terms or at all;

•

our ability to protect and enforce our intellectual property position for our drug candidates, and the scope of such protection;

•

our financial performance;

•

our competitive position and the development of and projections relating to our competitors or our industry;

•

our estimates regarding future revenue, expenses and needs for additional financing;

•

the impact of laws and regulations; and

•

our expectations regarding the time during which we will be an emerging growth company under the JOBS Act.

You should not rely on forward-looking statements as predictions of future events. We have based the forward-looking statements contained in this Annual Report primarily on our current expectations and projections about future events and trends that we believe may affect our business, financial condition, and results of operations. The outcome of the events described in these forward-looking statements is subject to risks and uncertainties, including the factors described in “Part I, Item 1A. Risk Factors” and elsewhere in this Annual Report. Moreover, we operate in a very competitive and rapidly changing environment. New risks and uncertainties emerge from time to time, and it is not possible for us to predict all risks and uncertainties that could have an impact on the forward-looking statements contained in this Annual Report. The results, events, and circumstances reflected in the forward-looking statements may not be achieved or occur, and actual results, events, or circumstances could differ materially from those described in the forward-looking statements.

In addition, statements that “we believe” and similar statements reflect our beliefs and opinions on the relevant subject. These statements are based on information available to us as of the date of this Annual Report. While we believe that such information provides a reasonable basis for these statements, that information may be limited or incomplete. Our statements should not be read to indicate that we have conducted an exhaustive inquiry into, or review of, all relevant information. These statements are inherently uncertain, and investors are cautioned not to unduly rely on these statements.

The forward-looking statements contained in this Annual Report relate only to events as of the date on which the statements are made. We undertake no obligation to update any forward-looking statements made in this Annual Report or to reflect new information or the occurrence of unanticipated events, except as required by law. We may not actually achieve the plans, intentions or expectations disclosed in or expressed by, and you should not place undue reliance on, our forward-looking statements. Our forward-looking statements do not reflect the potential impact of any future acquisitions, mergers, dispositions, joint ventures, or investments.

Unless the context otherwise requires, all references in this Annual Report to “we,” “us,” “our,” “our company,” and “Acrivon” refer to Acrivon Therapeutics, Inc. and its subsidiaries.

Summary Risk Factors

Investing in our common stock involves a high degree of risk because our business is subject to numerous risks and uncertainties, as more fully described in “Part I, Item 1A. Risk Factors” of this Annual Report. Below are some of these risks, any one of which could materially adversely affect our business, financial condition, results of operations, and prospects:

•

We are a clinical stage biopharmaceutical company and have incurred significant losses since our inception. We expect to incur losses over the next several years and may never achieve or maintain profitability.

•

We have a limited operating history and no history of commercializing products, which may make it difficult for an investor to evaluate the success of our business to date and to assess our future viability.

•

We will need additional funding to meet our financial obligations and to pursue our business objectives. If we are unable to raise capital when needed, we could be forced to curtail our planned longer-term operations and the pursuit of our growth strategy.

•

Our business substantially depends upon the successful clinical development of drug candidates using our AP3 platform and OncoSignatureTM, or OncoSignature companion diagnostics. If we are unable to obtain regulatory approval for, and successfully commercialize, drugs developed through the application of our AP3 platform and OncoSignature tests, our business may be materially harmed.

•

We are highly dependent on the success of ACR-368 and/or ACR-2316, as these are our first drug candidates being developed for clinical development and regulatory approval. We may never obtain approval for ACR-368, ACR-2316, or any other drug candidate.

•

The regulatory approval processes of the U.S. Food and Drug Administration, or FDA, and comparable foreign regulatory authorities are lengthy, time consuming and inherently unpredictable, and if we are ultimately unable to obtain regulatory approval for our drug candidates, on a timely basis or at all, our business will be substantially harmed.

•

The successful clinical development of some of our drug candidates may depend on the co-approval of an OncoSignature test as a companion diagnostic test. If we or our companion diagnostic collaborator are unable to obtain regulatory approval for our OncoSignature companion diagnostic tests for such drug candidates, we may not obtain regulatory approval and realize the commercial potential of certain drug candidates.

•

Our relationships with customers, healthcare providers, including physicians, and third-party payors are subject, directly or indirectly, to federal and state healthcare fraud and abuse laws, false claims laws, health information privacy and security laws and other healthcare laws and regulations. If we are unable to comply, or have not fully complied, with such laws, we could face substantial penalties.

•

Enacted and future legislation may increase the difficulty and cost for us, and any collaborators, to progress our clinical programs and obtain marketing approval or licensure of and commercialize our drug candidates and may affect the prices we, or they, may obtain.

•

Even if we are able to commercialize any drug candidates, the products may become subject to unfavorable pricing regulations, third-party reimbursement practices or healthcare reform initiatives, which would harm our business.

•

We rely, and expect to continue to rely, on third parties, including independent clinical investigators, contracted laboratories and contract research organizations, or CROs, to conduct our preclinical studies and clinical trials. If these third parties do not successfully carry out their contractual duties or meet expected deadlines, we may not be able to obtain regulatory approval for or commercialize our drug candidates and our business could be substantially harmed.

•

The precision oncology space is competitive, which may result in others discovering, developing or commercializing products before or more successfully than we do.

•

Our future success depends on our ability to retain key executives and to attract, retain and motivate qualified personnel.

•

Our success depends in part on our ability to obtain intellectual property rights for our proprietary technologies and drug candidates, as well as our ability to protect our intellectual property. It is difficult and costly to protect our proprietary rights and technology, and we may not be able to ensure their protection.

•

We depend on intellectual property licensed from a third party and termination of this license could result in the loss of significant rights, which would harm our business.

PART I

Item 1. Business.

Overview

We are a clinical stage biopharmaceutical company discovering and developing precision oncology medicines for patients whose tumors are predicted to be sensitive to each specific medicine by utilizing our proprietary Generative Phosphoproteomics platform, Acrivon Predictive Precision Proteomics, or AP3. Recently approved precision oncology treatments, such as kinase inhibitors, have transformed the cancer treatment landscape, and while the therapeutic benefit of these agents has provided significant benefit to patients, these precision oncology treatments unfortunately only address the less than 10% of patients with cancers that harbor certain easily-identifiable genetic mutations. Our approach is designed to overcome the limitations of genomics-based patient selection methods. We do this by using our precision medicine platform, AP3, to develop our pipeline of oncology drug candidates. AP3 is engineered to measure compound-specific effects on the entire tumor cell protein signaling network and drug-induced resistance mechanisms in an unbiased manner. These distinctive capabilities enable AP3’s direct application for drug design optimization for monotherapy activity, the identification of rational drug combinations, and the creation of drug-specific proprietary OncoSignature companion diagnostics that are used to identify the patients most likely to benefit from Acrivon’s drug candidates, which we refer to as patient responders. We are currently advancing our lead candidate, ACR-368, a selective small molecule inhibitor which targets CHK1 and CHK2 at sub single-digit nM and single-digit nM potency in intact cells, respectively, in a potentially registrational Phase 2 trial focusing on patients with endometrial cancer. We are continuing enrollment and dosing of patients in this multi-center trial based on AP3-predicted sensitivity to ACR-368 in patients with endometrial adenocarcinoma, a tumor type that was predicted to be sensitive to ACR-368 prior to clinical entry through preclinical AP3-based indication finding, and not previously evaluated in past clinical trials. Our ACR-368 OncoSignature test, which has not yet obtained regulatory approval, has been extensively evaluated in preclinical studies, including in two separate, blinded, prospectively-designed studies on pretreatment tumor biopsies collected from patients with ovarian cancer treated with ACR-368 in past Phase 2 clinical trials conducted by Eli Lilly and Company, or Lilly, and at the National Cancer Institute, or NCI, providing evidence of robust enrichment of responders through our method. Moreover, clinical data from the ongoing registrational intent trial in endometrial cancer showed initial validation of the ACR-368 OncoSignature for prospective patient selection (see further details below). Based on these sets of data, the FDA has granted Breakthrough Device designations for the ACR-368 OncoSignature assay for the identification of endometrial cancer patients and ovarian cancer patients who may benefit from ACR-368 treatment.

Our AP3 approach is proteomics-based and designed to enable identification and treatment of the patients whose tumors are sensitive to a specific drug or drug candidate based on direct protein measurement of critical tumor-driving mechanisms and independent of underlying genetic alterations. We believe our approach is applicable across stages of drug development and across therapeutic modalities. Accordingly, the AP3 method is not limited to the typically very small subset of cancers driven by single gene driver mutations or susceptible to a synthetic lethal approach. Rather, we believe our method is broadly applicable to the vast majority of cancers, in particular the majority of solid tumors, for which genetics-based approaches have proven insufficient to identify patient responders in many cases. In principle, we believe a much larger percentage of tumors can be addressed therapeutically using agents attuned to the specific biochemical signaling pathways found in these tumors, which our AP3 platform was purposefully designed to enable.

By applying our highly specific patient selection approach to drug development, we seek to both accelerate clinical development and significantly increase the probability of successful treatment outcomes for patients. Our pipeline includes the Phase 2 lead program, ACR-368, also known as prexasertib, a precision oncology asset that targets CHK1 and CHK2, or CHK1/2. In past trials, ACR-368 was dosed in more than 400 patients at the recommended Phase 2 dose, or RP2D, with reported deep, durable responses, including complete responses, or CRs, in a proportion of patients with solid tumors in past single center and multi-center Phase 2 clinical trials in tumor indications with high unmet need. ACR-368 also demonstrated a generally favorable safety and tolerability profile with primarily reversible hematological toxicity and very limited non-hematological adverse events. We received clearance from the FDA for an IND application to advance ACR-368 in Phase 2 single arm clinical trials conducted under the FDA program known as the master protocol, which was developed to help expedite drug development in multiple tumor types for drugs with an established RP2D within the same overall trial structure. Patients are stratified for treatment based on OncoSignature-predicted sensitivity to ACR-368 across multiple sites in the United States in this trial with registrational intent. Through the use of our OncoSignature test, we believe we can significantly increase the overall response rate, or ORR, observed in previous trials that were conducted without a prospective patient responder identification method. In September 2024 we reported positive clinical data for endometrial cancer, including a confirmed ORR of 62.5% (95% CI, 30.4 - 86.5). The data further validated our AP3-based ACR-368 OncoSignature assay, which is used for prospective patient selection in the registrational intent trial, showing a segregation of responders in the OncoSignature-positive, or BM+, versus OncoSignature-negative, or BM-, arms (p = 0.009). The median duration of treatment was not yet reached, but the duration on study was six months at the time of the data cut. With the encouraging maturing data in endometrial cancer combined with the competitive positioning given limited treatment options in second line and the potential market opportunity, we are now prioritizing this tumor type.

An interim data extract from the EDC clinical database was done on February 25, 2025, including 20 BM+ endometrial cancer patients treated with ACR-368 monotherapy and 38 BM- treated with ACR-368 plus ultra low dose gemcitabine (LDG) that were efficacy-evaluable by RECIST (2 BM- had treatment discontinued without scan). All BM+ patients had progressed after prior platinum-based chemotherapy and prior anti-PD-1, and the median and mean prior lines of therapy for these patients were 2 and 2.6, respectively. A majority of these BM+ patients were refractory to the last prior line of therapy, with aggressive, generally heavily pre-treated tumors: 12 had refractory disease (best overall response of PD in the last prior line of therapy), 6 had relapsed disease, and 2 unknown. Amongst these 20 BM+ patients, 15 were either serous or carcinosarcomas, 13 were pMMR (2 dMMR, 5 not tested), and 11 p53 mutated (3 wild-type; 6 unknown). In patients that had relapsed after the prior line of therapy (N=6), the confirmed ORR was 50% and the disease control rate (DCR) was 100%. Amongst the 12 patients with tumors refractory to the last prior line of therapy (ORR = 0%) we observed meaningful ACR-368 clinical activity with a confirmed ORR of 33% and DCR of 75%. The ACR-368 OncoSignature accurately identified patients whose tumors are sensitive to ACR-368, with 80% of BM+ patients demonstrating tumor shrinkage. Among all 20 BM+ patients the confirmed ORR was 35% and the DCR was 80%. Overall, we observed significant anti-tumor activity and disease control in BM+ patients with aggressive, refractory tumors that did not respond at all (0 % ORR) to the last line of prior therapy, and with a confirmed ORR more than double (35%) the best ORR observed in the last prior line of therapy (15%) for all BM+ patients.

ACR-368 is also being studied in combination with low dose gemcitabine (LDG) in additional indications, such as squamous cell carcinomas, including squamous cell cancer, or SCC, of head and neck (H&N), or SCCHN, in an Investigator-Initiated Trial (IIT). Given broad anti-tumor activity observed in past trials in other tumor types, we will potentially study ACR-368 in additional tumor types where there is high unmet need and competitive positioning opportunity. For example, we are assessing trial initiation in myelodysplastic syndrome/myeloproliferative neoplasms (MDS/MPN), diseases with high unmet need, based on transcription factor gene mutations rendering these malignancies sensitive to CHK1/2 as observed in various preclinical studies, including studies using ACR-368.

We have previously confirmed in preclinical studies that LDG sensitizes both OncoSignature-negative and BM+ tumors to ACR-368, as predicted by the AP3 platform, and this is consistent with an upregulation of the ACR-368 OncoSignature biomarkers in both human tumor cell lines and in human tumor xenograft mouse models after LDG treatment. We now have obtained further evidence of such OncoSignature biomarker upregulation in human patient tumors based on serial pre- and post- LDG biopsies in an ongoing Investigator-Initiated Trial at the Moffitt Cancer Center in patients with H&N cancer. Consistent with this preclinical and now clinical evidence of sensitization by LDG in BM- patients, we are continuing to explore the combination of ACR-368 with LDG in our ongoing endometrial cancer trial. Preliminary analyses of the 38 BM- patients, who are heavily pretreated (median of 3 prior lines of therapy) show a confirmed ORR of ~13% with the ACR-368 + LDG combination, which is comparable to the best ORR in the last prior line of therapy (median = 3) in these patients, which was 17%. Based on the totality of the preclinical and observed clinical data, we believe this supports significant LDG sensitization to ACR-368 in BM- patients. We expect a similar sensitization in BM+ patients which could be explored in a future all-comer study of ACR-368 + LDG.

We are also leveraging our proprietary Generative Phosphoproteomics AP3 precision medicine platform for streamlined drug discovery through AP3-based drug optimization in intact cells and co-crystallography and to develop our internally-discovered pipeline programs. These include ACR-2316, our second clinical stage asset, a novel, selective, dual WEE1/PKMYT1 inhibitor designed specifically for enhanced therapeutic index by achieving superior single-agent activity through strong activation of not only CDK1 and CDK2 but also of PLK1 to drive pro-apoptotic cell death, as observed in preclinical studies against benchmark inhibitors, combined with exquisite selectivity. Accordingly, ACR-2316 has demonstrated complete tumor regression at different dosing regimens across cell line-based and patient derived human tumor xenograft models with only limited, reversible, short-lived mechanism-based hematological adverse events. Utilizing AP3 we were able to advance ACR-2316 from being discovered as an initial lead to being dosed in a Phase 1 trial in only 15 months. This trial was initiated during the third quarter of 2024, two quarters ahead of our original timelines, and we are currently completing Dose Level (DL) 3 after clearing of DL1 and DL2 by the Safety Review Committee, without safety concerns or dose-limiting toxicities. Based on pharmacokinetic (PK) analysis in the first two dosing level cohorts, we have observed encouraging approximate dose proportionality. Moreover, using our internal MS-based AP3 profiling to support the clinical trial of ACR-2316, we are already detecting drug target engagement in peripheral blood mononuclear cells (PBMCs) in dose level 1. In addition, initial clinical activity has been observed in a patient in DL3, with significant decrease in size of metastatic lesions throughout the chest, abdomen and pelvis. This patient (who had received 3 prior lines of therapy including chemotherapy and anti-PD-1) remains on therapy. Using AP3-based Indication Finding and AP3-based analyses of in-house and publicly available data, we are enrolling selected high unmet need solid tumor types predicted sensitive to ACR-2316 in our Phase 1 trial. In addition to ACR-2316, we also have a preclinical stage, internal AP3-and co-crystallography-based preclinical cell cycle program against a critical, undisclosed target.

Furthermore, we have developed the AP3 Interactome, a proprietary, computational analytics platform driven by Generative Phosphoproteomics for integrated comprehensive analyses across all large, in-house AP3 phosphoproteomic drug profiling data sets to advance our in-house research programs.

We were founded and are led by pioneers in oncogenic signaling, oncology precision medicine and the use of proteomic technology to uncover intracellular biochemical signaling pathways with the goal of applying this knowledge to develop drug candidates and clinical diagnostics. Our founders have established proof-of-concept, including clinical implementation, for the underlying technologies in our AP3 platform. Our scientific advisors are thought leaders from leading global cancer and academic centers and are actively involved in our drug development process.

Our AP3 Platform

Our proprietary AP3 platform is engineered to measure compound-specific effects on the entire tumor cell protein signaling network and drug-induced resistance mechanisms in an unbiased manner at very high resolution and throughput. As such, all drug-regulated effects on the disease-driving, upregulated pathways and active proteins are revealed for each compound that we profile. We apply these distinctive capabilities of AP3’s for drug design optimization for monotherapy activity, the identification of rational drug combinations, and the creation of drug-specific proprietary OncoSignature companion diagnostics that are used to identify the patients most likely to benefit from Acrivon’s drug candidates.

One of the applications of our AP3 platform are our proprietary response-predictive clinical tests that we refer to as OncoSignature tests. These are drug-tailored, automated, quantitative proteomic tissue imaging tests applied to pretreatment tumor biopsies as a companion diagnostic, or CDx, to select and treat the patients predicted to benefit from the drug candidate. Our OncoSignature test, which has not yet obtained regulatory approval, is being developed with Akoya Biosciences, Inc., or Akoya, pursuant to a companion diagnostic agreement. Our OncoSignature tests encompass a signature of three classes of functionally-defined protein biomarkers assembled into a single signature assay. The quantitative levels for each of the three biomarkers are defined to determine whether a patient’s individual tumor has upregulated the biochemical signaling mechanisms that the drug modulates, and that the tumor depends on for growth and/or survival. Our company name, Acrivon, is derived from Greek for “accurate.” We chose it to embody how our AP3 platform can accurately match our therapies with patients who will benefit.

The tumor-agnostic application of OncoSignature tests enables us to identify and focus on tumor types for which a high unmet need for a treatment exists and that are predicted to be highly sensitive to our drug candidates. We achieve this by deploying our OncoSignature screening of human cancer samples across various tumor types. Through this process, we can preclinically identify new tumor types predicted to be sensitive to a drug candidate and even estimate the percentage of predicted responders before entering clinical trials. For example, we have identified endometrial cancer as a highly sensitive cancer type for ACR-368. Moreover, we have found through this approach that a proportion of patients with HPV+ cancers are predicted to be responsive to ACR-368, consistent with previously observed clinical activity in a proportion of patients with SCCHN and anal cancer. Furthermore, we predicted that patients with squamous non-small cell lung cancer, or sqNSCLC, would not respond to ACR-368, consistent with an observed ORR of 0% in patients with this tumor type in a past trial with ACR-368. Hence, through our preclinical AP3 based indication finding approach, we can specifically avoid running clinical trials in cancer types predicted to have limited sensitivity to our drug candidate.

We are using our AP3 platform not only to generate drug-tailored, response-predictive clinical OncoSignature tests, but also to provide unbiased, quantitative analyses of off-target effects on intracellular signaling using phosphoproteomic profiling, enabling us to discover inhibitors that are both highly potent and highly selective, as demonstrated by ACR-2316. Finally, by enabling an unbiased, high-resolution measurement of effects of drugs on intracellular signaling pathways, we are leveraging AP3 to predict which indications are likely sensitive to a particular drug candidate prior to clinical entry. Using AP3-based Indication Finding and AP3-based analyses of in-house and publicly available data, we are enrolling selected high unmet need solid tumor types predicted sensitive to ACR-2316 in our Phase 1 trial.

Our AP3 Interactome, a proprietary, computational analytics platform driven by Generative Phosphoproteomics machine learning has been developed by us for integrated comprehensive analyses across all large, in-house AP3 phosphoproteomic drug profiling data sets to advance our in-house research programs.

We believe that by leveraging our AP3 platform, we will profoundly alter precision oncology drug development and the treatment landscape of patients suffering from cancer.

Our Pipeline

Figure 1. Acrivon’s internal pipeline including the clinically advanced ACR-368, ACR-2316, and preclinical programs.

Our Lead Clinical Candidate ACR-368

ACR-368 is a selective small molecule inhibitor targeting CHK1/2. CHK1/2 are key regulators of the cell cycle and of DDR and inhibition of CHK1/2 has been demonstrated to have anti-tumor activity in multiple preclinical models as well as in clinical trials in humans. Several CHK1/2 inhibitors including ACR-368, also known as prexasertib, have been investigated in the clinic; however, none have been approved by the FDA. ACR-368 has shown deep, durable single agent clinical activity, including CRs and partial responses, or PRs, in a proportion of patients with solid tumors with high unmet need for a treatment, such as platinum-resistant ovarian cancer, and SCCs, including SCCHN and anal cancer. More than 400 patients with these tumors have been treated with ACR-368 monotherapy at the RP2D in advanced single- and multi-center clinical trials conducted by Lilly, NCI, and at MD Anderson Cancer Center, or MDACC. The confirmed ORR in these trials without a predictive biomarker was 29% at the single center Phase 2 ovarian cancer trial at NCI in the intent to treat, or ITT, population, and approximately 12% across the platinum-resistant ovarian cancer cohorts in the large Phase 2 multi-center international trial sponsored by Lilly. The median duration of response, or mDoR, at the RP2D across trials to date have ranged from almost six months to 12 months, and ACR-368 monotherapy demonstrated a generally favorable safety and tolerability profile with primarily reversible hematological toxicity and very limited non-hematological toxicity.

Using our AP3 platform, we have developed a predictive OncoSignature test for ACR-368, called ACR-368 OncoSignature, that we believe can predict patient response to ACR-368 monotherapy and therefore substantially improve the clinical ORR and, furthermore, that we believe, has the potential to enable expedited drug development. Predicted patient responders are referred to as ACR-368 OncoSignature-positive and predicted non-responders are referred to as ACR-368 OncoSignature-negative. The ACR-368 OncoSignature test has been extensively evaluated in preclinical studies in both patient-derived xenograft, or PDX, mouse tumor models as well as in two separate blinded, prospectively designed preclinical studies of pre-treatment tumor biopsies collected from patients with ovarian cancer that received ACR-368 in previous clinical trials.

By applying our AP3 platform for preclinical indication finding and expansion across human cancer types, as described below, we have found that approximately 30% of samples from patients with ovarian cancer are ACR-368 OncoSignature-positive. Moreover, we observed that between 30% and 40% of patients with endometrial and bladder cancer are predicted to be highly sensitive to ACR-368. Patients with these two types of cancer were not previously treated in ACR-368 clinical trials. All three tumor types were therefore initially included in our ongoing Phase 2 clinical trial, which now, based on the encouraging maturing data, competitive positioning, and potential market opportunity, is focused on endometrial cancer.

We have also used our AP3 platform to identify resistance mechanisms to ACR-368. Through phosphoproteomic profiling of human tumor cell lines that are either highly sensitive or highly resistant to ACR-368, we uncovered key resistance mechanisms and found that very low dose gemcitabine, or LDG, could be used to overcome resistance and further sensitize human tumor cells to ACR-368 through inducing increased DDR stress. Moreover, the use of LDG was observed to enhance sensitivity to ACR-368 in the already sensitive cells. We expect this may enable ACR-368 in combination with LDG to be an important treatment for ACR-368 OncoSignature-negative patients who would otherwise be excluded from ACR-368 treatment.

Based on these results, we are conducting a Phase 2 clinical trial focusing on endometrial cancer. ACR-368 OncoSignature-positive patients receive ACR-368 monotherapy in a single arm Phase 2b trial and ACR-368 OncoSignature-negative patients receive ACR-368 combined with LDG at the RP2D of ACR-368 and RP2D established in the study which is 10 mg/m2 in the exploratory Phase 2 dose expansion portion of the trial. As a result, all patients biopsied are eligible to receive therapy. Akoya procures and manufactures the necessary supplies to perform the OncoSignature tests. Based on our communications with the FDA to date, we believe the monotherapy trial, if successful, has the potential to be registrational for ACR-368. At ESMO 2024 (September 14, 2024 R&D event and press release), we reported that endometrial cancer was our prioritized indication, as it represents the first potential registrational opportunity for ACR-368. We remain confident in this strategy based on emerging clinical data, competitive positioning given limited treatment options, and the strong commercial opportunity in both second- and front-line settings. Our blinded KOL market research estimates that there are approximately 27,000 U.S. patients annually in the second-line setting alone for endometrial cancer. Due to increased competition and a smaller market opportunity, we set a high internal clinical bar for ovarian cancer, which preliminary data suggests is unlikely to be met. Bladder cancer is also being deprioritized due to lower than preclinically predicted BM+ rate, leading to challenging enrollment with single digit BM+ patient enrollment to date. We have now officially deprioritized ovarian and bladder cancers, reallocating all clinical resources to ACR-368 in endometrial cancer and ACR-2316.

We are carrying out our trial under the auspices of the master protocol guidance issued by the FDA in March 2022 to enable expedited drug development. This guidance provides sponsors of drugs or biologics for the treatment of cancer and for which the RP2D has been established in prior studies, the opportunity to simultaneously evaluate more than one investigational drug and/or multiple cancer subpopulations within the same overall trial structure under master protocol in adult and pediatric cancers.

In September 2024, we reported positive clinical data at the European Society of Medical Oncology conference from the ongoing registrational intent, multicenter Phase 2 trial of ACR-368 in patients with locally advanced or metastatic, recurrent endometrial cancer who had progressed on prior anti-PD-1 therapy, unless ineligible. Patients of all histopathologies (endometrioid, serous, clear cell, and carcinosarcoma) and regardless p53, MMR and other molecular status are eligible. Endometrial cancer had not been previously studied in prior ACR-368 Lilly-sponsored trials. Using AP3 for indication screening, this tumor type was predicted to be particularly sensitive to ACR-368 prior to the current ongoing Phase 2 study. The data were based on 35 safety-evaluable patients, of which 23 (8 BM+ and 15 BM- patients) were efficacy-evaluable with at least one on-treatment scan (data cut off July 25, 2024). This data included a confirmed ORR of 62.5% (95% CI, 30.4-86.5) observed in prospectively-selected ACR-368 OncoSignature-positive patients with endometrial cancer. The data further validated our AP3-based ACR-368 OncoSignature assay, which is used for prospective patient selection in this registrational intent trial, achieving a clear segregation of responders in the OncoSignature-positive versus OncoSignature-negative arms (p-value = 0.009). Consistent with previous Phase 2 clinical trials conducted with ACR-368, the adverse events, or AEs, were primarily reversible, transient, mechanism-based hematological, including neutropenia and thrombocytopenia. Based on these results, which demonstrated single agent RECIST activity in patients who all had progressed on anti-PD-1 and included patients with the most difficult to treat forms of endometrial cancer, including p53 mutated, pMMR serous and carcinosarcomas, Acrivon is planning a confirmatory trial combining ACR-368 with anti-PD-1 vs anti-PD-1 in the maintenance phase of the recently approved first-line therapy for endometrial cancer. This is further supported by the strong rationale and synergy observed in preclinical studies combining ACR-368 with anti-PD-1.

An interim data extract from the EDC clinical database was done on February 25, 2025, including 20 BM+ endometrial cancer patients treated with ACR-368 monotherapy and 38 BM- treated with ACR-368 plus LDG that were efficacy-evaluable by RECIST (2 BM- had treatment discontinued without scan). All BM+ patients had progressed after prior platinum-based chemotherapy and prior anti-PD-1, and the median and mean prior lines of therapy for these patients were 2 and 2.6, respectively. A majority of these BM+ patients were refractory to the last prior line of therapy, with aggressive, generally heavily pre-treated tumors: 12 had refractory disease (best overall response of PD in the last prior line of therapy), 6 had relapsed disease, and 2 unknown. Amongst these 20 BM+ patients, 15 were either serous or carcinosarcomas, 13 were pMMR (2 dMMR, 5 not tested), and 11 p53 mutated (3 wild-type; 6 unknown). In patients that had relapsed after the prior line of therapy (N=6), the confirmed ORR was 50% and the DCR was 100%. Amongst the 12 patients with tumors refractory to the last prior line of therapy (ORR = 0%) we observed meaningful ACR-368 clinical activity with a confirmed ORR of 33% and DCR of 75%. The ACR-368 OncoSignature accurately identified patients whose tumors are sensitive to ACR-368, with 80% of BM+ patients demonstrating tumor shrinkage. Among all 20 BM+ patients the confirmed ORR was 35% and the DCR was 80%. Overall, we observed significant anti-tumor activity and disease control in BM+ patients with aggressive, refractory tumors that did not respond at all (0 % ORR) to the last line of prior therapy, and with a confirmed ORR more than double (35%) the best ORR observed in the last prior line of therapy (15%) for all BM+ patients.

We have previously confirmed in preclinical studies that LDG sensitizes both BM- and BM+ tumors to ACR-368, as predicted by the AP3 platform, and this is consistent with an upregulation of the ACR-368 OncoSignature biomarkers in both human tumor cell lines and in human tumor xenograft mouse models after LDG treatment. We now have obtained further evidence of such OncoSignature biomarker upregulation in human patient tumors based on serial pre- and post- LDG biopsies in an ongoing Investigator-Initiated Trial at the Moffitt Cancer Center in patients with H&N cancer. Consistent with this preclinical and now clinical evidence of sensitization by LDG in BM- patients, we are continuing to explore the combination of ACR-368 with LDG in our ongoing endometrial cancer trial. Preliminary analyses of the 38 BM- patients, who are heavily pretreated (median of 3 prior lines of therapy) show a confirmed ORR of ~13% with the ACR-368 + LDG combination, which is comparable to the best ORR in the last prior line of therapy (median = 3) in these patients, which was 17%. Based on the totality of the preclinical and observed clinical data, we believe this supports significant LDG sensitization to ACR-368 in BM- patients. We expect a similar sensitization in BM+ patients which could be explored in a future all-comer study of ACR-368 + LDG.

We believe that use of our ACR-368 OncoSignature test to select patients predicted to be sensitive to ACR-368 for treatment may significantly increase the ORR, which has the potential to lead to accelerated approval for endometrial cancer while avoiding treatment of patients with tumors that are not likely to respond. However, we cannot guarantee that the FDA will permit us to utilize an accelerated approval process or that our intended approach will be sufficient for regulatory approval. We are planning to file one or more IND application amendments to add one or more additional cancer types under the same or a similar trial protocol design at a later time, including head and neck cancer, anal cancer, and cervical cancer.

Our Internally Discovered and Developed Programs

We also have wholly-owned, internally developed preclinical and clinical drug programs uniquely enabled by our AP3 platform and its ability to rationally design compounds with optimal target selectivity properties aiming to achieve potent single agent activity through elimination of dominant resistance mechanisms. These programs are also structure-guided with rational medicinal chemistry efforts based on co-crystallography of lead series with their respective targets to ensure high selectivity.

ACR-2316 is our first internally discovered clinical stage asset. It is a novel, dual WEE1 and PKMYT1 inhibitor small molecule development candidate, rationally designed through advanced co-crystallography and the AP3 platform to achieve optimal target potency and selectivity through strong activation of not only CDK1 and CDK2 but also of PLK1 to drive pro-apoptotic cell death, as observed in preclinical studies against benchmark inhibitors, delivering potent single agent anti-tumor activity across in vitro and in vivo preclinical studies, compared to benchmark WEE1 and PKMYT1 inhibitors. Clinical WEE1 inhibitors have demonstrated promising anti-tumor activity in early clinical trials conducted by competitors; however, their clinical activity has been hindered by a narrow therapeutic index and WEE1 inhibitor-induced resistance mechanisms. ACR-2316 was specifically designed using AP3 to address these limitations through very high selectivity to limit adverse events to mechanism-based, on-target, and to simultaneously inhibit PKMYT1, a closely related protein serine/threonine kinase also serving critical functions in the cell cycle and DDR pathways, that accounts for a major part of WEE1 inhibitor-induced resistance, as revealed by AP3. Based on mechanism of action and confirmed in our preclinical studies, balanced inhibition of PKMYT1 results in more potent single agent activity.

We believe there is a need for novel patient selection methods to overcome the challenges with genetics-based patient selection methods, and that using AP3 will enable us to identify drug-sensitive indications and individual patients predicted sensitive to WEE1 and PKMYT1 inhibitors. Using AP3 for unbiased quantitative high-resolution measurement of the effects of ACR-2316 on the human tumor cell phosphoproteome, this compound has been optimized for potent induction of mitotic catastrophe, which is key to its strong single agent activity in preclinical models and potentially favorable clinical profile for monotherapy development. ACR-2316 was discovered by AP3-based SAR, facilitated by co-crystallography, and designed by AP3 to overcome WEE1-induced resistance mechanisms. The preclinical data generated demonstrate a very potent single agent activity so we believe patient selection might not be needed in the most sensitive tumor indications. Using AP3-based Indication Finding and AP3-based analyses of in-house and publicly available data, we are enrolling selected high unmet need solid tumor types predicted sensitive to ACR-2316 in our Phase 1 trial. We plan to generate an OncoSignature to be used for drug target engagement-based dose optimization in the Phase 1 and for potential individual patient responder prediction, if needed.

ACR-2316 entered clinical development in the third quarter of 2024, two quarters ahead of original timelines, and the Phase 1 monotherapy clinical trial of ACR-2316 is currently in the dose escalation portion of this trial. The Phase 1 study will assess the safety and tolerability of ACR-2316. Additionally, the study will seek to establish the pharmacokinetic profile, evaluate preliminary anti-tumor activity and determine the recommended Phase 2 monotherapy dose. Dose optimization is being guided by drug target engagement in alignment with the FDA’s Project Optimus. We anticipate providing a clinical data update in the second half of 2025. Based on pharmacokinetic (PK) analysis in the first two dosing level cohorts, we have observed encouraging approximate dose proportionality. Moreover, using our internal MS-based AP3 profiling to support the clinical trial of ACR-2316, we are already detecting drug target engagement in peripheral blood mononuclear cells (PBMCs) in dose level 1. In addition, initial clinical activity has been observed in a

patient in DL3, with significant decrease in size of metastatic lesions throughout the chest, abdomen and pelvis. This patient (who had received 3 prior lines of therapy including chemotherapy and anti-PD-1) remains on therapy.

Preclinical Program

We also have a wholly-owned, internally developed preclinical cell cycle program with an undisclosed target being developed uniquely using AP3. Consistent with our approach to ACR-2316 development, this program is also based on rational medicinal chemistry efforts informed by the intracellular pathway effects of compounds revealed by AP3 combined with co-crystallography of lead series with the undisclosed target and off-targets to ensure high selectivity. We anticipate nominating a development candidate in 2025. We also anticipate initiating a new program in autoimmune/inflammatory diseases in 2025, leveraging AP3.

AP3 Potential for Broad Clinical Impact

Our AP3 platform is based on two integrated technology pillars, mass spectrometry-based proteomic profiling and our automated tumor imaging biomarker platform. Mass spectrometry, or MS, enables a systematic, unbiased quantitative analysis of the proteins inside a cell or entire tissues and is used to identify our biomarker candidates. These are validated using our biomarker platform which is also used to run our OncoSignature tests. AP3 is designed to generate multiple clinically-actionable, valuable outputs:

•

Predictive biomarkers and patient responder identification: Our AP3 platform enables identification and treatment of patients predicted to be sensitive to the drug candidate, while avoiding treatment of patients predicted not to benefit.

•

Indication finding and expansion: AP3 screening of human patient tumor samples is used to preclinically predict what proportion of various tumor types are expected to be highly sensitive to our drug candidates. This enables indication expansion and could potentially increase the response rates in clinical trials.

•

Identification of resistance mechanisms: AP3 is a powerful technology to identify either pre-existing (intrinsic) resistance or acquired (therapy-induced) resistance to drugs demonstrated in prior studies. We intend to apply this technology to develop combination therapy candidates that target the druggable resistance mechanisms and re-sensitize tumors and to prevent resistance development.

•

Identification of rational drug combinations: Through our AP3 platform, we uncover the entire protein signaling pathways underlying resistance. The druggable targets on such pathways are a basis for rational drug combinations and we believe can efficiently overcome resistance demonstrated in multiple prior studies. We intend to apply this for indication expansion and confirmatory trials for our drug candidate pipeline.

•

Unbiased drug target engagement and pharmacodynamic biomarker discovery: Through our high resolution phosphoproteomic drug profiling, we uncover thousands of on- and off-target interactions and drug-regulated pharmacodynamic, or PD, biomarkers for each drug candidate. These can be used to guide selectivity optimization of preclinical lead series and to measure drug target engagement in patient tumor tissues during clinical trials, and hence guide dose optimization.

Our AP3 platform deploys high resolution, high throughput MS resulting in large datasets reflecting differentially drug-regulated phosphorylation sites and signaling pathways inside sensitive and resistant cells for each drug candidate we profile. The data are highly structured and amenable to machine learning, which has enabled us to create a streamlined process and to integrate all the analytical

steps into a single workflow. We intend to apply our AP3 platform to both our existing and future pipeline of drug candidates addressing prevalent, high unmet need cancers and where patient responder identification has proven challenging, as further described below.

Figure 2. AP3 has potential for broad impact across the drug discovery and development process. Figure illustrates proven deliverables with the AP3 platform.

Our Team

We were founded in 2018 and are led by pioneers in oncogenic signaling, oncology precision medicine, and the use of proteomic technology to uncover intracellular biochemical signaling pathways and to apply this knowledge to develop drug candidates and clinical diagnostics. Peter Blume-Jensen, MD, PhD, our co-founder, President and Chief Executive Officer, is the inventor of our AP3 platform and OncoSignature patient selection method. He has extensive experience in oncology drug discovery and development at leading pharmaceutical companies including Serono, Merck & Co. and Daiichi Sanyo. While Chief Scientific Officer at Metamark Genetics, Dr. Blume-Jensen led the development of an automated, proteomics-based predictive clinical diagnostic for prostate cancer which was validated through blinded clinical trials and included as the only stand-alone test under National Comprehensive Cancer Network, or NCCN, guidelines and reimbursement in 2015. Kristina Masson, PhD, MBA, co-founder, Executive Vice President, Business Operations and head of our discovery research site in Sweden, previously founded and operated OncoSignature AB, a biotech company which established the phosphoproteomics and drug discovery infrastructure and which we subsequently acquired. Jesper Olsen, PhD, our academic co-founder, is Professor of Quantitative Proteomics at the University of Copenhagen and Vice Director of the Novo Nordisk Foundation for Protein Research and a recognized pioneer of MS-based quantitative phosphoproteomics. Rasmus Holm-Jorgensen, our Chief Financial Officer through April 1, 2025, has over 20 years of experience in the biopharmaceutical industry, most recently as Chief Strategy & Portfolio Officer and part of the founding team at Kiniksa Pharmaceuticals. Effective April 1, 2025, Adam Levy, PhD, MBA, has been appointed to the position of Chief Financial Officer. Dr. Levy has been serving as Senior Vice President and Head, Corporate Affairs and Investor Relations of the Company since July 2023. Dr. Levy has more than 25 years of finance and investor relations experience in the biopharma industry. Erick Gamelin, MD, PhD, our Chief Development Officer, has led over 100 Phase 1 through Phase 3 oncology clinical trials and most recently served as Chief Medical Officer of Step Pharma. Eric Devroe, PhD, our Chief Operating Officer, has extensive experience in operations and business development leadership from his time at Metamark Genetics, MDACC, and several start-up companies. Jean-Marie Cuillerot, MD, our Chief Medical Officer, was most recently Chief Medical Officer at Dragonfly Therapeutics and Agenus. He has extensive experience in leading clinical programs in immune-oncology from pre-IND through global submissions, including development of Avelumab at EMD-Serono, the first drug approved to treat Merkel cell carcinoma and second-line bladder cancer, as well as multiple Phase 2 and Phase 3 clinical trials at Bristol Myers Squibb for ipilimumab across indications. Mary-Alice Miller, our Chief Legal Officer, has over 20 years of experience as a corporate attorney. She most recently served as General Counsel of Butterfly Network, a publicly listed medical device company.

Our founders have pioneered and established proof-of-concept, including clinical implementation, for the underlying technologies in our AP3 platform. Our scientific advisors are thought leaders from leading global cancer and academic centers and are actively involved in our drug development process.

Our Strategy

Our goal is to be the leading biopharmaceutical company leveraging proteomic and phosphoproteomic data, which we access through our proprietary AP3 platform, to unlock insights superior to those from traditional genomic-based approaches and discover and efficiently develop medicines to benefit patients with cancer.

While our AP3 approach is broadly applicable across disease areas, we are initially focused on oncology. Our goal is to treat patients with cancer with clinically active therapeutics that have a high likelihood of success based on predicted sensitivity to our drug candidates. Oncology is an area of high unmet clinical need, in which only a small fraction of patients currently benefits from existing predictive biomarkers, such as next-generation sequencing, or NGS. We are currently applying the AP3 technology to both in-licensed clinical stage and to internally developed drug candidates for tumors that do not harbor single gene driver mutations, which is estimated to be more than 90% of all human cancers. The relevant drug target classes in these tumors that we believe are well-suited for our AP3 approach include but are not limited to DDR pathways, DNA replication stress, super enhancers, and cell cycle and transcriptional regulators. We are currently focused on expedited clinical development of our clinically advanced asset ACR-368, in our ongoing Phase 2 registrational intent trial focusing on patients with endometrial cancer, with an option for further clinical development of ACR-368 in HPV+ and other cancers, e.g. MDS/MPN. The Phase 2 trial is based on OncoSignature-predicted sensitivity to ACR-368 and has been cleared by the FDA to be conducted under a master protocol. In addition to ACR-368, we are also leveraging our proprietary AP3 precision medicine platform to develop our co-crystallography-driven, internally-discovered pipeline programs. These include ACR-2316, our second clinical stage asset which is now advancing in Phase 1, a novel, potent, selective WEE1/PKMYT1 inhibitor designed using AP3 for superior single-agent activity through strong activation of not only CDK1 and CDK2 but also of PLK1 to drive pro-apoptotic cell death, as observed in preclinical studies against benchmark inhibitors. In addition, we have a preclinical cell cycle program with an undisclosed target.

The key elements of our strategy summarized below are to:

•

Advance ACR-368, our CHK1/2 inhibitor, through clinical development in endometrial cancer by enrolling ACR-368 OncoSignature-positive patients. Our lead program, ACR-368, has already demonstrated deep, durable anti-tumor activity, including CRs, in patients with ovarian cancer in past clinical trials. Based on preclinical AP3-based indication finding screening we predict that patients with other solid tumor types of high clinical unmet need, including 30% to 40% of patients with endometrial cancer, could benefit from ACR-368 monotherapy. We have further confirmed this prediction in preclinical studies on PDX models where we observed that endometrial tumors were highly sensitive to ACR-368, and that our ACR-368 OncoSignature test was able to prospectively identify which models are the most sensitive. In September 2024, we reported positive clinical data for endometrial cancer, including a confirmed ORR of 62.5% (95% CI, 30.4 - 86.5). The data further validated our AP3-based ACR-368 OncoSignature assay, which is used for prospective patient selection in the registrational intent trial, showing a segregation of responders in the OncoSignature-positive versus OncoSignature-negative arms (p = 0.009). The median duration of treatment was not yet reached, but the duration on study was six months at the time of the data cut. With the encouraging maturing data in endometrial cancer combined with the competitive positioning given limited treatment options in second line and the potential market opportunity, we are now prioritizing this tumor type.

•

Selectively pursue AP3 identified rational drug combinations with our drugcandidates in OncoSignature-negative patients, initially ACR-368 with LDG. Our AP3 platform is able to elucidate pathways of underlying tumor resistance mechanisms, both pre-existing (intrinsic) and acquired (therapy-induced). This allows us to identify rational drug combinations that can re-sensitize ACR-368 OncoSignature-negative patients to our drug candidates in resistant tumors. For example, we have shown that LDG was highly synergistic with ACR-368 in resistant human tumor cell lines and was able to re-sensitize ACR-368 resistant tumors to ACR-368, in ovarian, bladder, and endometrial cancers. Based on these findings, we are conducting a Phase 1b/2 clinical trial with ACR-368 in combination with LDG for endometrial cancer patients whose tumors are ACR-368 OncoSignature-negative. Work is now proceeding in the exploratory Phase 2 dose expansion portion of the study utilizing the newly established RP2D for low dose gemcitabine and the previously established RP2D for ACR-368. In April and September 2024, we reported that evidence of sensitization to ACR-368 by ultra-low dose gemcitabine (ULDG) was observed in a proportion of BM- subjects, with an initial disease control in eight out of 15 subjects, including one confirmed complete response (CR).

•

Discover and develop a pipeline of proprietary drug candidates by leveraging our AP3 platform. We are applying our AP3 platform in multiple ways to build and advance a pipeline of structure-guided, wholly owned precision oncology drug candidates. The first of these, ACR-2316, is a novel, dual WEE1 and PKMYT1 inhibitor small molecule, rationally designed through advanced co-crystallography and the AP3 platform to achieve optimal target potency and selectivity, delivering potent single agent anti-tumor activity across in vitro and in vivo preclinical studies, compared to benchmark WEE1 and PKMYT1 inhibitors. While WEE1 inhibitors have shown single agent clinical activity across patients with solid tumors of high unmet need, the ORR so far has been insufficient for approval and, despite significant efforts in identifying patient responders, these efforts have not been fruitful to date. The ACR-2316 program is currently being

investigated in a Phase 1 clinical trial designed to assess the safety and tolerability of ACR-2316. Additional objectives of this trial include the determination of the maximal tolerated dose and recommended Phase 2 monotherapy dose, characterization of the pharmacokinetic profile, and preliminary evaluation of anti-tumor activity. We also have a cell cycle program for which we anticipate nominating a development candidate in 2025. All of our internally derived drug candidates will leverage AP3 phosphoproteomic drug candidate profiling to guide and optimize drug potency and selectivity. We believe that this approach will help ensure that our drug candidates directly affect the pathways of interest while minimizing off-target effects, an approach that is highly differentiated from traditional drug discovery programs. Additionally, by utilizing AP3, we believe we can identify patients with highly sensitive tumor types of high unmet clinical need for treatment before initiation of our clinical trials.

•

Acquire rights to drugcandidates for which we believe our OncoSignature tests can increase the likelihood of clinical success. We in-licensed ACR-368 after successfully developing a predictive ACR-368 OncoSignature test to increase the probability of clinical success. We intend to take a similar approach and in-license other attractive drug candidates where genetics-based patient selection is challenging or impossible, and develop drug-tailored OncoSignature tests for these drug candidates. We intend to pursue only the opportunities that, similar to ACR-368, have high clinical potential and where we believe we can successfully select patients who are likely to respond to such specific drug candidates, based on our proprietary OncoSignature tests.

•

Opportunistically enter into strategic co-development partnerships to maximize the full potential of our AP3 platform. We believe that there are opportunities to partner with organizations that have approved drugs or drug candidates in development under competitive pressure and where the availability of a highly predictive OncoSignature test to achieve high ORRs can potentially provide an advantage in obtaining regulatory approval and market share. Moreover, we believe that identification of rational drug combinations for such drugs to improve ORR and clinical benefit are of high value to prospective partners. We intend to pursue such partnerships where we can realize the value that our AP3 platform can bring to the drug candidate through early co-development.

Urgent Need for Precision Oncology Approaches that Transcend the Limitations of Genomics

Cancer is a disease of dysregulated protein activity, which occurs as a result of underlying genetic changes. The majority of precision medicine efforts in oncology have been focused on identifying patients who are most likely to respond based either on genetic changes in their tumors, such as specific mutations, gene amplifications, and gene translocations, or on the patient’s own genetic background. The availability of genomic sequences from tens of thousands of tumors has begun to transform oncology treatment away from the use of broad cytotoxic drugs approved based on tumor location towards precision medicines that address tumors with specific genetic alterations. However, while this approach has led to the recent approval of a number of targeted therapies, their use is limited to a very small fraction of patients with these mutations. It is estimated that only 9% of all patients with cancer have tumors with genetic profiles that make them eligible for an available precision oncology medicine, so-called genetically-defined cancers, and only 5% of all patients with cancer are likely to benefit from available therapies.

Figure 3. Proteomic biomarkers have the potential to be broadly applicable across the vast majority of cancers.

Proteomic biomarkers have the potential to be broadly applicable for the vast majority of cancers where more traditional genetics-based approaches have proven challenging. In this small subset of genetically defined cancers, most often the alterations in the gene lead to drug target protein dysregulation that drives the cancer, which are potential targets of cancer therapies. There are three main types of such recurrent single driver gain-of-function, or GOF, gene alterations known in human cancer: point mutations, gene fusions, and amplifications, which represents less than 10% of all cancers. These most easily addressable GOF mutation-driven cancers have been the obvious focus of drug discovery and development for more than two decades. Examples of such approved drugs include Vemurafenib for B-RAF-V600E-mutant melanoma, Imatinib for KIT and PDGFR-alpha mutant GIST, Crizotinib for EML4-ALK+ lung cancer, Trastuzumab for HER2 amplified breast cancer, Larotrectinib for solid tumors with N-TRK fusions, and most recently, Retevmo for patients with RET-mutated tumors. However, more than 90% of cancers have tumor-driving targets that do not harbor underlying single genetic alterations. Such tumor-driving drug targets are activated through post-translational modifications, including phosphorylation, due to complex genetic alterations elsewhere in the genome of tumor cells, rather than in the drug target itself. Successful clinical development of inhibitors for these targets is highly challenging as prevailing predictive methods such as NGS, polymerase chain reaction, or PCR, fluorescent in-situ hybridization, or FISH, immunohistochemistry, or IHC, and transcriptomics have not been successful in identifying patients that would significantly benefit from the drug.

Accordingly, while a powerful tool to uncover underlying mechanisms of disease, the utility of genomics for patient selection is limited when it comes to drug response prediction in oncology. Additionally, the lack of therapeutic efficacy for a given drug, due to inability to identify patient responders, is still a top attrition factor in drug development. The vast majority of cancers contain multiple, complex genomic alterations resulting in the dysregulated, tumor-driving protein activity. Relatively few genetic alterations are common to a broad percentage of patients with cancer, such as mutations in the K-RAS or p53 genes. However, precision medicines against these targets have been difficult to develop and, because of the complex genetic alterations often co-existing in tumors, treatment often does not elicit expected clinical benefit.

The AP3 Solution: Matching Drug Action to the Disease-driving Mechanisms in Patients’ Tumors

Our AP3 platform has been developed over the last decade to be an efficient process and workflow to determine sensitivity to drugs based on the biological signaling pathways that are activated in diseased cells and are required for their survival. Our AP3 platform leverages proteomic biomarkers which enable direct measurement of disease-driving mechanisms independent of target gene alterations, and allow for accurate matching with the mechanism of action of a particular drug. For example, in the case of ACR-368, which is a selective CHK1/2 inhibitor, the three biomarkers we quantify with our ACR-368 OncoSignature assay measure the level of activated DNA repair downstream of the activated drug targets and whether the tumor likely depends on it. One biomarker is a specific phosphorylation site in the drug target and another is a specific phosphorylation site in a key DNA repair protein, which together inform us about their functional activity. The third is a protein that drives premature DNA replication, which implies that the tumor is dependent on the upregulated DNA repair measured by the two other biomarkers. We have designed our proprietary AP3 platform to be agnostic to the underlying genetic alterations in the genome and enable identification and treatment of patients based on direct measurement of the disease-driving mechanisms that are regulated by and sensitive to the drug. Hence, in contrast to measuring genetic alterations in a patient’s tumor, which is only a surrogate read-out for protein dysregulation, and having to infer whether the drug will act on the inferred protein dysregulation, the AP3 method directly reveals the dysregulated proteins and pathways driving the tumor that the drug acts on. The AP3 method is drug-tailored, and we believe enables an accurate match (Acrivon is derived from Greek for “accurate”)between the mechanism of the drug action with the disease-driving mechanisms in the patient’s tumor.

Our AP3 platform is fundamentally different from genetics-based methods to identify patient responders and we believe it is particularly applicable to the majority of cancers without genetic alterations in the drug target itself. It specifically focuses on the proteins and pathways that drive tumor growth and survival and enable drug action, rather than exploring complex biology and accumulated genetic alterations that have proven very difficult to connect to drug response.

While the principles and technology behind AP3 are not limited to cancer, we are initially committed to oncology, where we are applying AP3 to develop drug candidates with the potential to transform the treatment of solid tumors of high unmet clinical need. Strategically, we are applying AP3 to drug classes where genetics has proven difficult or insufficient for response prediction, and that are active in major fractions of solid tumors, but where the ORR is insufficient for approval without a prospective patient responder identification method. In addition to DDR pathway inhibitors such as ATR, ATM, WEE1, and CHK1/2, examples of drug classes that we believe would benefit from our AP3 platform include cell cycle regulators (such as CDK2, 4, 6), mitotic regulators (such as Aurora kinases), transcriptional regulators, DNA replication modulators, such as CDC7, super enhancer kinases (such as CDK7, 9, 12), and inhibitors of mutated forms of K-RAS. We believe our ability to apply AP3 to these drug classes allows us to open up the potential of precision medicine approaches to a much larger fraction of patients than has been possible using exclusively genetics-based approaches. We are initially progressing a pipeline of DDR and cell cycle drug candidates, but intend to broaden our pipeline to some of these other drug classes and targets through preclinical AP3-based indication finding. We anticipate conducting target/compound validation and indication finding for a new program in autoimmune/inflammatory diseases leveraging AP3.

Our AP3 platform is based on two underlying technology pillars typically executed in two sequential steps: the first step, a high-resolution MS for biomarker identification which is integrated with, the second step, our automated tumor biopsy-imaging biomarker platform that enables biomarker validation and which is also used to run our OncoSignature tests.

Figure 4. Our AP3 platform is based on unbiased biomarker identification using global phosphoproteomic profiling by MS and an automated biomarker platform.

MS enables a systematic, unbiased quantitative measurement and analysis of the proteins inside a cell or entire tissues. We specifically use it to identify and measure in an unbiased manner the effects of any given drug or drug candidate on the activity state of the protein signaling networks inside a cell through analysis of the phosphorylation state and levels of proteins inside a tumor cell. Phosphorylation is the best-studied, allosteric on-off switch regulatory mechanism for protein activity involved in all forms of intracellular signaling. Analysis of the entire phospho-proteome before and after drug treatment, so-called phosphoproteomic drug profiling, enables us to objectively identify the global effect of any drug on the activity state of the protein signaling network.

Our MS efforts allow us to identify attractive drug-regulated biomarker candidates, which include identifying changes in overall protein levels as well as in post-translational modifications of proteins, such as those that involve phosphorylation and are involved in activation or inhibition of protein function in biological signaling pathways. Our data-independent acquisition, label-free phosphoproteomic methods provide for very high resolution. Starting with lists of thousands of potential biomarker candidates that correlate with drug sensitivity and resistance, our proprietary algorithms and workflows distill biomarker candidates into three functionally defined classes. The biomarkers are further validated in tumor models and through quantitative measurements on PDX models as well as on patient tumor samples and, when available, clinical trial biopsies, as we have done with ACR-368.

Use of our AP3 platform to develop drug-tailored, predictive OncoSignature tests

One of the key outputs of our AP3 platform are our drug-tailored OncoSignature tests, which are based on an assembly of biomarkers from each of the three classes selected by the process described above, resulting in a single, quantitative signature test. They are automated, quantitative protein imaging tests designed to be applied to pretreatment tumor biopsies as a CDx to select and treat the patients predicted to benefit from the specific drug candidate for which they are developed. The tests are developed for routine-processed, paraffin-fixed biopsy tissue and stained with fluorescently labeled antibodies against the OncoSignature biomarkers. Digital images of these stained tissues are then processed by proprietary software that identifies both tumor cells and tumor cell nuclei. They are quantitatively measured in only defined tumor tissue regions of a patient biopsy where they function, called the “region-of-interest,” or ROI. A proprietary algorithm assesses the quantitative level of each biomarker and combines them to predict the likely response to a drug or drug candidate.

The AP3 approach is designed to provide a streamlined, rationale-driven workflow to identify and validate biomarkers. Every OncoSignature test is drug-tailored. Our process to generate an OncoSignature test, including technical biomarker validation, can be

completed in approximately two to three months. It measures three functionally defined classes of biomarkers that in combination are predictive of sensitivity to the particular drug. Each biomarker class can contain more than one biomarker, but we typically measure only one in each class for a total of three biomarkers. A key rationale is that patients whose tumors do not harbor the specific protein disease-driving mechanisms that are sensitive to the drug are predicted to be unlikely to respond to a particular drug or drug candidate and hence can be excluded from treatment.

Figure 5. AP3 approach for streamlined development and validation of predictive OncoSignature tests.

In order to create an OncoSignature test that can be readily performed on clinical samples, we qualify monoclonal antibodies for the prioritized set of three biomarkers. These antibodies are chosen based on our systematic evaluation of their specificity and sensitivity including correlation in changes in biomarker levels with drug sensitivity in cell lines and, most importantly, their technical performance on human intended use FFPE-processed cancer tissues as well. This technical validation ensures specificity (that it only recognizes the biomarker of interest), dynamic range (the fold changes of the biomarker level across tumor samples), and proper intensity. The technically qualified antibodies are then assembled into a final drug-tailored predictive OncoSignature test that is functionally validated in a blinded, prospectively designed manner in various preclinical studies. These include prediction of drug sensitivity across human tumor cell lines, in PDX models, and across human tumor samples, and, when available, on pretreatment tumor biopsies collected from past trials with the drug or drug candidate. Using our AP3 platform workflow, we have developed and evaluated in preclinical studies an OncoSignature predictive test for ACR-368, as further described below. We have also developed and done preliminary validation for two prototype OncoSignature tests for two other clinical stage assets, a CDK7 and a CDC7 inhibitor, for which genetics-based patient selection has also proven challenging.

The ACR-368 OncoSignature test has been transferred as a clinical trial assay under an exclusive license to our external companion diagnostic partner, who is running our test during our ongoing clinical trial, and developing it into a companion diagnostic test which they will also commercialize, pending regulatory approval. The tests are performed on a standard, routine processed pre-treatment tumor biopsy with an anticipated turnaround time of five to seven business days. We intend to protect all our drug-tailored OncoSignature tests via patents for their tumor-agnostic usage across cancers.

Figure 6. Our OncoSignature tests are applied to pretreatment tumor biopsies and will be offered by our CDx partner with an anticipated turn-around time of five to seven business days.

Enablement of AP3 through our team’s expertise

The enablement of the AP3 approach as a means to realize the potential of proteomic drug profiling and protein signature tests in precision medicine is the result of the vision of our founders and their long-standing expertise in the field, including pioneering the underlying AP3 technologies and implementation experience. Three critical aspects behind AP3 are:

•

Founding concept and vision. Our founders are leaders and respected authorities in the understanding of oncogenic kinase signaling, protein dysregulation through tyrosine phosphorylation, and the relationship of each to human cancer. Our vision was embedded in the 2001 Nature review article, “Oncogenic kinase signaling,” by our founder Peter Blume-Jensen, which became a citation classic in the field of medicine. It described how cancer and other diseases are inevitably driven by dysregulated protein signaling resulting from either very simple or complex underlying genetic alterations. The paper linked simple GOF mutations in a class of proteins called tyrosine kinases with their disease-driving dysregulation and involvement in a certain small subset of human cancers. Our founding vision is that proteomic biomarkers enable direct measurement of the disease-driving mechanisms and allow for accurate matching with drug action, independent of underlying genetic alterations.

•

Technical expertise and implementation experience. The two underlying technologies used in a stepwise manner in our AP3 platform, (1) high resolution MS for quantitative protein and protein phosphorylation analysis and (2) the automated biomarker platform, have been pioneered and established by our founders and team and integrated into a content system and approach. Jesper Olsen, our academic co-founder, is a recognized world leader in the use of MS-based phosphoproteomics, or the study of protein phosphorylation and its impact on biology. Dr. Olsen is one of the most highly cited authors in this field. Our co-founder, Kristina Masson, has established the entire infrastructure for phosphoproteomics at our subsidiary in Medicon Village, Lund, Sweden in close proximity with Dr. Olsen’s laboratory in Copenhagen, Denmark. Our OncoSignature technology is enabled by this comprehensive proteomics infrastructure and demonstrated proof-of-concept for the first unbiased MS step in the AP3 approach, resulting in identification of resistance mechanisms and rational drug combinations with the potential to be tested in controlled clinical trials with the drug selinexor in acute myeloid leukemia. This work was published in Cell Reports on August 9, 2022.

Peter Blume-Jensen led the first proof-of-concept for unbiased identification of drug-regulated PD biomarkers for PI3’K pathway-targeted agents through an MS-based phosphoproteomics approach. Under his leadership, our team also led the establishment of our automated biomarker platform and the research and development of ProMark, a proteomics eight biomarker imaging test for prostate cancer outcome prediction launched by Metamark. That test was validated in a blinded trial and was subsequently included in the NCCN Clinical Practice Guidelines. Through this experience, we understand the technical and regulatory challenges involved in developing

and implementing a clinically meaningful proteomic test, and we fully leverage and factor these insights into the design of our OncoSignature tests.

ACR-368, Our Phase 2 Lead Candidate

Our lead drug candidate, ACR-368, also known as prexasertib, is a selective inhibitor with sub single-digit potency against CHK1 and single-digit potency against CHK2. ACR-368 was originally discovered by Array BioPharma and acquired by Lilly, who evaluated the compound in over 1,000 patients across 18 clinical trials, where it demonstrated deep, durable single agent activity, including CRs, in a proportion of patients across several Phase 2 studies of platinum-resistant ovarian cancer and other solid tumors. Despite the demonstrated clinical activity in a proportion of patients, there was no obvious patient selection strategy to improve responses sufficient for approval. We chose to in-license ACR-368, prioritizing it over other carefully evaluated candidates, based on multiple criteria, including its proven clinical single agent activity, extensive safety data set and extensive comparison work and in-house AP3 profiling.

We are enrolling and dosing patients in our Phase 2 trial of ACR-368 focusing on patients with endometrial cancer based on OncoSignature-predicted sensitivity to ACR-368, due to the encouraging maturing data, competitive positioning, and potential market opportunity.

An interim data extract from the EDC clinical database was done on February 25, 2025, including 20 BM+ endometrial cancer patients treated with ACR-368 monotherapy and 38 BM- treated with ACR-368 plus LDG that were efficacy-evaluable by RECIST (2 BM- had treatment discontinued without scan). All BM+ patients had progressed after prior platinum-based chemotherapy and prior anti-PD-1, and the median and mean prior lines of therapy for these patients were 2 and 2.6, respectively. A majority of these BM+ patients were refractory to the last prior line of therapy, with aggressive, generally heavily pre-treated tumors: 12 had refractory disease (best overall response of PD in the last prior line of therapy), 6 had relapsed disease, and 2 unknown. Amongst these 20 BM+ patients, 15 were either serous or carcinosarcomas, 13 were pMMR (2 dMMR, 5 not tested), and 11 p53 mutated (3 wild-type; 6 unknown). In patients that had relapsed after the prior line of therapy (N=6), the confirmed ORR was 50% and the DCR was 100%. Amongst the 12 patients with tumors refractory to the last prior line of therapy (ORR = 0%) we observed meaningful ACR-368 clinical activity with a confirmed ORR of 33% and DCR of 75%. The ACR-368 OncoSignature accurately identified patients whose tumors are sensitive to ACR-368, with 80% of BM+ patients demonstrating tumor shrinkage. Among all 20 BM+ patients the confirmed ORR was 35% and the DCR was 80%. Overall, we observed significant anti-tumor activity and disease control in BM+ patients with aggressive, refractory tumors that did not respond at all (0 % ORR) to the last line of prior therapy, and with a confirmed ORR more than double (35%) the best ORR observed in the last prior line of therapy (15%) for all BM+ patients.

We also plan to study ACR-368 in one or more additional indications, such as HPV+ squamous cell carcinomas, including SCCHN, anal, and cervical cancer, based on demonstrated clinical single agent activity in SCCHN and anal cancer and OncoSignature-based prediction of sensitivity to ACR-368 in a proportion of patients. In addition, we are assessing trial initiation in MDS/MPN, diseases with high unmet need, based on transcription factor gene mutations rendering these malignancies sensitive to CHK1 and CHK2 as observed

in various preclinical studies, including studies using ACR-368. Akoya is procuring and manufacturing the necessary supplies to perform the OncoSignature tests used in our Phase 2 clinical trial.

ACR-368, a selective inhibitor of CHK1 and CHK2, key DDR regulators

CHK1 and CHK2 are checkpoint proteins that prevent cell replication when DNA damage is present. In the absence of DNA damage, CHK1 and CHK2 are largely inactive. Most normal tissues, other than certain dividing cells such as those in bone marrow, are not reliant on DDR mechanisms such as CHK1 and CHK2, and hence not subject to the negative side effects from such inhibitors. In contrast, inhibition of the kinase activity of these proteins or knockdown of their expression by RNA interference in certain G1/S checkpoint-deficient tumor cells has been shown to prevent repair of double-strand DNA breaks resulting in cell death. Treatment of cells with DNA damaging agents or inhibitors of other proteins involved in the DDR, sensitizes them to cell killing by CHK1 and CHK2 inhibitors.

Figure 7. CHK1 functions as a cell cycle checkpoint to inhibit DNA replication when DNA damage is present.

ACR-368 is a selective CHK1/2 inhibitor with a potency of less than 1 nM against CHK1 and 8 nM against CHK2. In preclinical studies, ACR-368 inhibited growth with a potency of less than 100 nM in over 75% of 600 cancer cell lines screened, including a potency of less than 50 nM in 16 of 23 tested ovarian cancer cell lines. ACR-368 as a single agent led to complete tumor regression in approximately 40% of 38 ovarian cancer PDX models tested. Significant anti-tumor activity was also observed in other tumor models such as sarcomas and neuroblastoma. The anti-tumor activity of ACR-368 was enhanced in preclinical models when it was combined with DNA damaging agents such as cisplatin and gemcitabine.

Clinical development of ACR-368 for patients with endometrial cancer

We are developing ACR-368 for the treatment of patients with advanced solid tumors including endometrial cancers. ACR-368 has demonstrated deep, durable single agent anti-tumor clinical activity, including CRs, in a proportion of more than 400 solid tumor patients treated at RP2D in past clinical trials conducted by Lilly, its previous sponsor, and in several investigator-initiated trials, including at the NCI and at MDACC. Importantly, ACR-368 was well-tolerated in these trials, exhibiting primarily reversible, manageable hematological adverse events and limited dose-limiting non-hematological adverse events. Accordingly, there have been no clinical or regulatory holds reported and less than 2% drug-related discontinuations across all trials to date. By pairing ACR-368 with our compound-specific OncoSignature test, we believe we may significantly increase the ORR by targeting treatment to the patients that are predicted to be most dependent on CHK1/2, and therefore more likely to respond.

Based on our preclinical studies, we expect 30% to 40% of patients in our lead indication, endometrial cancer, will be ACR-368 OncoSignature-positive. We expect the ORR to exceed that of current standard of care and, if the data are sufficient, we will aim for single-agent, single-arm approval. These patients are treated with ACR-368 in a Phase 2 trial at the RP2D. The remaining 60% to 70% of ACR-368 OncoSignature-negative patients receive ACR-368 at the RP2D with LDG, which we have found to be highly synergistic with ACR-368 using our AP3 platform in preclinical studies.

of 62.5% (95% CI, 30.4-86.5) observed in prospectively-selected BM+ patients with endometrial cancer. The data further validated our AP3-based ACR-368 OncoSignature assay, which is used for prospective patient selection in this registrational intent trial, achieving a clear segregation of responders in the OncoSignature-positive versus OncoSignature-negative arms (p-value = 0.009).

An interim data extract from the EDC clinical database was done on February 25, 2025, including 20 BM+ endometrial cancer patients treated with ACR-368 monotherapy and 38 BM- treated with ACR-368 plus LDG that were efficacy-evaluable by RECIST (2 BM- had treatment discontinued without scan). All BM+ patients had progressed after prior platinum-based chemotherapy and prior anti-PD-1, and the median and mean prior lines of therapy for these patients were 2 and 2.6, respectively. A majority of these BM+ patients were refractory to the last prior line of therapy, with aggressive, generally heavily pre-treated tumors: 12 had refractory disease (best overall response of PD in the last prior line of therapy), 6 had relapsed disease, and 2 unknown. Amongst these 20 BM+ patients, 15 were either serous or carcinosarcomas, 13 were pMMR (2 dMMR, 5 not tested), and 11 p53 mutated (3 wild-type; 6 unknown). In patients that had relapsed after the prior line of therapy (N=6), the confirmed ORR was 50% and the DCR was 100%. Amongst the 12 patients with tumors refractory to the last prior line of therapy (ORR = 0%) we observed meaningful ACR-368 clinical activity with a confirmed ORR of 33% and DCR of 75%. The ACR-368 OncoSignature accurately identified patients whose tumors are sensitive to ACR-368, with 80% of BM+ patients demonstrating tumor shrinkage. Among all 20 BM+ patients the confirmed ORR was 35% and the DCR was 80%. Overall, we observed significant anti-tumor activity and disease control in BM+ patients with aggressive, refractory tumors that did not respond at all (0 % ORR) to the last line of prior therapy, and with a confirmed ORR more than double (35%) the best ORR observed in the last prior line of therapy (15%) for all BM+ patients.

Based on strong biological rationale and internally generated preclinical data showing strong synergy of ACR-368 with immune checkpoint inhibitors, we are currently evaluating the opportunity for a confirmatory label expansion trial in first line combing ACR-368 with anti-PD-1 in the maintenance phase of the recently approved new frontline therapy of platinum-based chemotherapy combined with anti-PD-1, in a pMMR all-comer setting.

Figure 8. Potent synergy of ACR-368 with anti-PD-L1 displays immune memory in 100% of mice.

Endometrial cancer background

Endometrial cancer is a cancer of the lining of the uterus that primarily affects post-menopausal women. The American Cancer Society estimates that in the United States there will be 66,200 new cases of endometrial cancer and approximately 13,030 patients will die of this disease in 2023. First-line treatment for patients with localized, early-stage disease is surgery with radiation therapy. Patients with more advanced disease, stages III or IV, especially high-grade, are treated with the recently approved frontline therapy consisting of platinum/taxane backbone chemotherapy in combination with anti-PD-1, followed by a long maintenance phase of anti-PD-1 (Eskander et al, NEJM 2024; Raza-Mirza et al, NEJM, 2024). In second line, the ~15% of patients with HER2 (IHC3+) are offered trastuzumab-deruxtecan, an ADC, when eligible. Apart from these, there is a huge unmet need in second line, as the previously used standard-of-care therapy with anti-PD-1 combined with Lenvatinib is not an option, at least for the ~75-80% of patients that are pMMR. There is no standard of care in later lines or therapy. Five-year survival for patients with metastatic endometrial cancer is approximately 20%.

In August 2024, company-sponsored, blinded third-party KOL market research showed strong interest in the emerging clinical profile of ACR-368 (product name blinded) as an important potential therapy in the rapidly evolving treatment landscape of high-grade, recurrent endometrial cancer where second-line options are now limited due to the recent approval of anti-PD-1 and chemotherapy as front-line therapy. There is a limited number of new emerging therapies in the second line setting, where the SOC for patents that are no longer eligible for lenvatinib plus pembrolizumab due to the new frontline therapy with chemotherapy and anti-PD-1, is very limited with an estimated ~12% ORR and mPFS of ~3 months. The market research also showed a significant market potential with an estimated ~30K new cases of high-grade, locally advanced or metastatic, recurrent (progressed on anti-PD-1 and chemotherapy) endometrial cancer per year in the U.S. and that ~90% of these patients will progress to second line.

HPV+ squamous cell carcinoma background

Squamous cell carcinomas are cancers that develop in the squamous cells that make up the outermost layer of the mucosa. More than 90% of anal cancers and cervical cancers and about 70% of SCCHN (the oral/oropharyngeal sub-group) are linked to infections with HPV. There are over 46,000 HPV+-associated cancers diagnosed in the United States each year and up to 5% of cancers worldwide are potentially caused by HPV+ infections.

Unlike many cancers, HPV+ cancers are not typically driven by high levels of genomic instability but rather by alterations in cell cycle regulation, including upregulation of DDR pathways. Certain HPV+ cancers, primarily SCCHN and cervical cancer, respond to PD-1 or PD-L1 immune checkpoint inhibitor therapy with ORR of approximately 20%, as single agent or in combination with chemotherapy, depending on the line of therapy and the level of PD-L1 expression in the tumor. Several PD-1/L1 inhibitors have received FDA approval for use in patients with metastatic or unresectable recurrent head and neck squamous cell carcinoma, in combination with platinum and fluorouracil or as a single agent for patients whose tumors express PD‐L1 as determined by an FDA‐approved test. On October 13, 2021, the FDA approved pembrolizumab in combination with chemotherapy, with or without bevacizumab, for patients with persistent, recurrent or metastatic cervical cancer whose tumors express PD-L1, as determined by an FDA-approved test. FDA also granted regular approval to pembrolizumab as a single agent for patients with recurrent or metastatic cervical cancer with disease progression on or after chemotherapy whose tumors express PD-L1 as determined by an FDA-approved test.

Sarcoma background

In addition to previously observed clinical activity in the above tumor types as monotherapy, ACR-368 has also shown clinical activity in patients with certain sarcoma subtypes in combination with various chemotherapeutic agents. Patients with sarcomas have very limited treatment options, primarily surgery, chemotherapy, and radiation, depending on the subtype. The five-year survival for patients with metastatic soft tissue sarcomas is approximately 17%.

MDS/MPN background

Myelodysplastic syndrome and myeloproliferative neoplasms (MDS/MPN) are hematological malignancies. MDS is a malignant clonal disorder of hematopoietic stem cells, leading to ineffective hematopoiesis, dysplasia, and cytopenias. It has a risk of infection and overall progression to acute myeloid leukemia (AML). MPNs (including polycythemia vera (PV), essential thrombocythemia (ET), and primary myelofibrosis (PMF)) are also clonal hematologic malignancies, characterized by excessive blood cell production. Some MPNs have an indolent course, but PMF and advanced PV/ET can transform into secondary AML, which is aggressive and difficult to treat. Preclinical research has identified frequent mutations in certain transcription factor genes that play a crucial role in the pathogenesis of MDS/MPNs. These mutations can disrupt normal cell cycle regulation and DNA repair mechanisms, making the cancer cells more

reliant on alternative pathways for survival. This dependency renders them potentially sensitive to inhibitors of DNA damage checkpoint kinases, such as CHK1 and CHK2, as observed in various preclinical studies, including with ACR-368.

Previous clinical trials of ACR-368 have demonstrated compelling, durable single agent activity in a proportion of patients with various tumor types

In previous clinical trials, conducted prior to the development of the ACR-368 OncoSignature test, ACR-368 has demonstrated deep, durable single agent activity, including CRs, in a proportion of more than 400 patients with high-grade serous, primarily platinum-resistant, ovarian cancer and SCC treated at RP2D. Overall, ACR-368 has been tested in 18 clinical trials as monotherapy or in combination with both targeted agents and chemotherapy in over 1,000 patients across primarily solid tumor types and has shown a generally favorable safety profile.

Phase 1a/b trial in squamous cell carcinoma established single agent clinical activity and the RP2D

A 146-patient Phase 1 multicenter trial was conducted in patients with refractory or recurrent squamous cell carcinoma and led by Dr. David Hong at MDACC. The trial included patients with SCCHN, sqNSCLC, and anal cancer. The primary objective of the Phase 1b expansion cohorts was to determine the safety, toxicity, and RP2D of ACR-368. In addition, the ORR according to Response Evaluation Criteria in Solid Tumors, or RECIST, version 1.1 for patients with specific types of SCC was recorded.

The RP2D was established at 105 mg/m2 given as an intravenous infusion every 14 days, and used in the expansion phase for 101 patients. The study demonstrated clinical monotherapy activity of ACR-368, with a 5% ORR in SCCHN and 15% ORR in anal cancer. The mDoR was seven months and over 12 months, respectively, including a CR in anal cancer. Based on these results and the lack of highly effective treatments, ACR-368 has been granted FDA Orphan Drug Designation, or ODD, for the treatment of anal cancer.

Of note, approximately half of the patients with SCCHN were HPV+ and showed a significantly higher ORR of 19% in response to ACR-368—a similar finding to the ORR recorded in patients with anal cancer, which is almost obligate HPV+. This was reflected in a markedly longer progression-free survival, or PFS, in HPV+ compared to HPV-negative, or HPV-, patients, with some HPV+ patients benefiting from therapy well over 12 months while no HPV- patients had benefit beyond five months (Fig. 9.).

Figure 9. ACR-368 treatment resulted in a significant improvement in progression-free survival in patients with HPV+ SCCHN compared to patients with HPV- SCCHN.

In this trial, as in most of the other clinical trials with ACR-368, an attempt was made to identify biomarkers predictive for response to ACR-368 in pretreatment tissue samples by NGS. In an analysis of genetic changes in 24 genes involved in DDR or increased replication stress, no obvious correlation with clinical response was observed. This lack of correlation between genetic changes and clinical response underscores the need for an alternative patient responder identification method, such as AP3.

Figure 10. Maximal percentage change in tumor size from baseline by best ORR across all expansion cohorts.

Phase 2, single center NCI trial in patients with high-grade serous ovarian cancer

A Phase 2 trial of ACR-368, led by Dr. Lee at NCI, enrolled 28 women with high-grade serous ovarian cancer. ACR-368 was administered at the RP2D every 14 days until disease progression, an event of unacceptable toxicity, or withdrawal of consent. Twenty-four women had evaluable responses after three withdrew consent because of travel inconvenience and one developed an intervening illness that prevented radiological evaluation of tumor progression. All patients in this trial had failed at least one round of prior cytotoxic chemotherapy and three quarters of the patients had failed three or more prior lines of therapy. The primary endpoint in this single center trial was investigator assessed tumor response based on RECIST v1.1.

In the analysis of the ITT population of 28 patients, an ORR of 29% was achieved. For the 21 patients in the ITT population with platinum-resistant disease, an ORR of 29% was achieved. The mean duration of response in patients with platinum-resistant ovarian cancer was over ten months, with some patients remaining on ACR-368 therapy for over 16 months (Fig. 11).

Figure 11. Duration of response with ACR-368 in a 28-patient ovarian cancer Phase 2 trial.

Similar to the findings reported for ACR-368 in SCC, there was no correlation observed between clinical response and alterations or the expression of potential biomarker genes (Fig. 12).

Figure 12. No correlation was observed between ACR-368 response and genetic alterations or potential biomarker expression in patients with ovarian cancer.

Phase 2 multicenter trial in advanced, high-grade serous ovarian cancer by Lilly

A large Phase 2 trial of ACR-368 in patients with platinum-resistant and platinum-refractory ovarian cancer sponsored by Lilly was conducted in 46 centers across eight countries. The 169 patients enrolled in this trial had failed two to four prior systemic therapies and 90% of patients had stage III or stage IV disease. The trial included patients with either an altered BRCA1 or BRCA2 gene, or BRCA-positive, or unaltered BRCA1 or BRCA2 gene, or BRCA-negative, ovarian cancer and was divided into four cohorts.

•

Cohort 1: patients with platinum-resistant, BRCA negative ovarian cancer with at least three lines of prior therapy

•

Cohort 2: patients with BRCA negative platinum-resistant ovarian cancer with no more than two lines of prior therapy

•

Cohort 3: patients with platinum-resistant BRCA mutant ovarian cancer with any line of prior therapy, but with obligatory prior PARP inhibitor therapy

•

Cohort 4: patients with platinum-refractory BRCA negative or BRCA mutant ovarian cancer and any line of prior therapy.

Figure 13. ORR and DCR in each of the four cohorts of patients with platinum-resistant and platinum-refractory ovarian cancer.

The primary outcome in this study was ORR. Results from this trial showed that a subset of patients with ovarian cancer treated with ACR-368 across all four cohorts achieved durable PRs. The ORR in the 140 patients with platinum-resistant ovarian cancer was 12.1%, not including unconfirmed responders.

Secondary outcomes included DCR, which is the percentage of patients with a best overall response of CR, PR, or stable disease, or SD, for at least four months. The DCR was over 30% across all four cohorts, varying from 31% in patients with platinum-refractory disease to 45% in patients with platinum-resistant disease with at least three lines of prior failed therapies. In the three cohorts of patients with platinum-resistant disease, the median duration of response was 5.6 months (95% confidence interval: 3.9 months; 7.6 months), and the median duration of overall survival was 11.9 months (95% confidence interval: 9.9 months; 14 months).

Consistent with previous observations, retrospective analyses of patient pretreatment tumor samples by both NGS and by IHC failed to identify biomarkers that strongly correlated with clinical response. Despite the demonstrated clinical activity, these data underscore the need for an effective patient responder enrichment method.

ACR-368 has been generally well-tolerated with manageable side effects

There have been eight Lilly-sponsored clinical trials with ACR-368. In these trials, ACR-368 was administered to a total of 681 subjects, to 479 subjects as monotherapy and to 202 subjects in combination with other treatments. In addition, there have been 10 Investigator-Initiated Trials, or IITs, where ACR-368 was administered to a total of 283 patients as either monotherapy or in combination. The primary adverse events observed in these trials were hematological, including transient neutropenia and thrombocytopenia, both of which were generally reversible and manageable. The neutropenia and thrombocytopenia are thought to be part of the mechanism-based suppression of cells in the bone marrow, or myelosuppression, which is also seen with other DDR inhibitors. However, by dosing ACR-368 at the established RP2D once every 14 days it was found that in most patients who experienced drug-related hematologic toxicities such as neutropenia had already begun to recover by the 14th day after dosing. Hence, granulocyte colony-stimulating factor and platelet infusions to correct for neutropenia and thrombocytopenia, respectively, were not mandated but were used at the discretion of the treating physicians in these trials. Nonhematologic toxicities deemed related to ACR-368 treatment occurred at a much lower frequency and severity as summarized below, with fatigue, nausea, and diarrhea being the mostly commonly observed events. In addition, in a few patients, an association was identified between increasing ACR-368 plasma concentration following monotherapy and transient QTcF prolongation. None of these episodes led to clinical manifestations. Accordingly, drug-related discontinuations were between only 1% to 2% across all patients. A proportion of patients experienced very durable responses, and in a few cases remained on therapy for several years.

Summary of adverse events from published reports on clinical trials with ACR-368 monotherapy dosed at RP2D

Ovarian Carcinoma

Other Cancer Types

* Adverse events greater than or equal to Grade 3 are considered serious adverse events

Using Our ACR-368 OncoSignature Test For Prediction of Sensitivity to ACR-368 in Our Ongoing Phase 2 Trial

Using the AP3 streamlined process as described above, we have developed a predictive OncoSignature test for ACR-368, called ACR-368 OncoSignature. We are using this in our ongoing Phase 2 trial to treat patients with endometrial cancer based on predicted sensitivity to ACR-368. We have extensively evaluated our ACR-368 OncoSignature test in various preclinical studies and models demonstrating the ability to predict sensitivity to ACR-368.

Prediction of sensitivity to ACR-368 across multiple human ovarian tumor samples

Two key questions facing companies entering clinical trials is whether the chosen tumor types in a particular trial will be sensitive to the drug candidate and, if so, what percentage of patients with each of these tumor types are expected to be sensitive to the drug candidate. To acquire this important information, we use our OncoSignature tests to screen across human patient tumor samples and multiple tumor types to predict not only which tumors are sensitive to our drug candidates, but also what percentage of patients with these tumor types are predicted to respond. We have used our ACR-368 OncoSignature test in this manner to screen across commercially available human patient tumor samples and across tumors that have been routine-processed by formalin-fixation and paraffin embedding, or FFPE, just like the pretreatment tumor biopsies collected from patient tumors are being processed in our ongoing clinical trial.

Using automated image acquisition software, the biomarkers in our ACR-368 OncoSignature tests are measured quantitatively within the ROI, which is where they are informative and exert their biological function. Patient tumor samples with a minimal predictive threshold of each of the three biomarkers present predicts sensitivity to ACR-368. Conversely, patients without presence of any of the three biomarkers are predicted to not benefit from ACR-368 and are being excluded from the monotherapy arm in our ongoing clinical trial.

Figure 14. Screening with our ACR-368 OncoSignature across human patient tumor samples is used to predict which patients are believed to be sensitive and resistant to ACR-368, in this example using human ovarian and anal tumor samples.

Through our analysis of patient tumor samples acquired from biorepositories, we have found that in high-grade serous ovarian cancer approximately 30% of all patient tumor samples have each of the three biomarkers present above the minimal predictive threshold. This result, combined with the results described below, suggests that approximately 30% of patients could potentially benefit from treatment with ACR-368 monotherapy.

Figure 15. Our ACR-368 OncoSignature provides quantitative scores that we use to objectively predict tumor response. Patient tumor samples with all three biomarkers above a certain minimum level on the heatmap are predicted to benefit from ACR-368 therapy.

Prediction of sensitivity to ACR-368 in human tumor cell lines

Human tumor cell lines are very different from human intact tumor tissue, but are still widely used to assess anti-tumor efficacy. To date, it has been very challenging to predict sensitivity to DDR inhibitors with prevailing genetics-based methods in human tumor cell lines. However, by applying our ACR-368 OncoSignature to a small panel of human tumor cell lines, we demonstrated our ability to predict sensitivity to ACR-368 with a high degree of certainty. The presence of all three biomarkers above a minimal level predicted sensitivity to ACR-368 in all cells that are highly sensitive to ACR-368 in viability assays, except for one.

Figure 16. Prediction of ACR-368 sensitivity across human tumor cell lines.

EC50: concentration of ACR-368 resulting in 50% inhibition of tumor cell survival.

R = predicted responder and NR = predicted non-responder.

Prediction of sensitivity to ACR-368 in ovarian PDX models

To demonstrate that we can also predict responders to ACR-368 in PDX models, we obtained untreated tumor tissue samples from 20 PDX models of ovarian cancer and generated quantitative biomarker scores with our ACR-368 OncoSignature test. Using the same approach, we assessed whether the tumor samples with a minimal level of each of the three biomarkers would predict sensitivity to ACR-368. We found that our ACR-368 OncoSignature was able to capture 80% of responders in PDX models while improving the ORR to approximately 55% compared to an approximated 20% baseline response rate.

Figure 17. Our ACR-368 OncoSignature accurately distinguished responders from non-responders in PDX models; sensitivity and specificity plotted as an area under receiver operator curve, or AUC.

Blinded, prospectively designed prediction of ACR-368 sensitivity in two separate studies of pretreatment tumor biopsies from past Phase 2 trials with ACR-368 in patients with high grade serous ovarian cancer

Our OncoSignature tests are developed using only tumor cells independent of any input from clinical results. The tests are dictated by a mechanistic, functional definition of each of the three classes of biomarkers based on a strong scientific and clinical rationale as well as on our insights into biological signaling. Based on our approach, we believe we can predict that if all three classes of biomarkers are present at a minimal level in a tumor sample, the tumor depends on upregulation of the drug target signaling axis for its growth and survival. Moreover, from our phosphoproteomic drug profiling of tumor cells, we have found that this upregulated signaling axis is modulated by the drug candidate.

To test our ACR-368 OncoSignature for its ability to identify the patients that benefit from monotherapy with ACR-368, we conducted two separate studies on pretreatment tumor biopsy samples collected from patients treated with ACR-368 in past trials. Importantly, the studies were blinded to any treatment outcome annotation, the analyses were prospectively defined, and results were analyzed by an independent third-party statistician.

We were able to obtain pre-treatment biopsy samples from a subset of patients with ovarian cancer treated with ACR-368 in the prior clinical Phase 2 trials: patients treated at NCI and in the multi-center trial sponsored by Lilly. We generated OncoSignature scores on these biopsy samples blinded to treatment outcome and handed these over to the third-party biostatistician, who received the treatment outcome annotation separately. The results of these studies showed that use of our tumor-agnostic ACR-368 OncoSignature test was able to significantly improve the response rate, to 47% and 58%, respectively. Moreover, the results also demonstrated that a negative ACR-368 OncoSignature largely eliminated patients who are less responsive to ACR-368, hence sparing these patients from ACR-368 single-agent treatment from which they would not benefit.

Figure 18. Blinded OncoSignature scoring of pre-treatment tumor biopsies from prior clinical trials of ACR-368 was able to segregate responders from non-responders.

Patients predicted to be sensitive to ACR-368 had a median PFS, or mPFS, of 7.9 months compared to 2.2 months for those predicted to be non-responders. This reflects the fact that not only the patients with PR or CR, but also with SD predicted by ACR-368 OncoSignature to be responders to ACR-368 treatment did indeed benefit for longer periods of time than the predicted non-responders. This could be valuable for confirmatory trials where mPFS is typically a primary endpoint.

Figure 19. Patients with OncoSignature positive scores had improved PFS compared to OncoSignature negative patients.

Prediction of ACR-368 clinical activity in additional cancer indications

To identify tumor types predicted to be sensitive to ACR-368, we used our ACR-368 OncoSignature test to screen across large numbers of human patient tumor samples across tumor types obtained from biorepositories. Through this tumor-agnostic usage of ACR-368 OncoSignature we found that between 30% and 40% of samples from patients with endometrial cancer and bladder cancer were predicted to be sensitive to ACR-368. In addition to confirming the positive predictive value of our ACR-368 OncoSignature test, we have also demonstrated the high negative predictive value of our ACR-368 OncoSignature test. For example, in sqNSCLC our ACR-368 OncoSignature test predicted that none of the patient samples would be sensitive to ACR-368, which is consistent with the Phase 1 trial that was conducted in SCC types and described above, which showed an ORR of 0% in sqNSCLC. Based on these findings, which were further confirmed in PDX models of endometrial and bladder cancer, as described below, we predicted that a significant proportion of patients with endometrial and bladder cancer will also be sensitive to ACR-368 monotherapy, and these two tumor types were therefore initially included together with ovarian cancer in our ongoing Phase 2 trial.

Figure 20. ACR-368 OncoSignature screening across human routine-processed FFPE patient tumor samples predicts which tumor types and what proportion thereof are sensitive to ACR-368. Each line in the four heat maps for each of the four tumor types—ovarian cancer, sqNSCLC, endometrial cancer, and bladder cancer—represents an individual tumor sample and the three columns from left to right represent the quantitative level of each of the three biomarkers in the OncoSignature test, BM1, BM2, and BM3.

Confirmation of activity in PDX models of predicted tumor types

In order to confirm our prediction based on screening of human patient tumor samples that a proportion of patients with bladder and endometrial cancer are sensitive to ACR-368 monotherapy, we generated PDX models of these two tumor types and assessed anti-tumor activity of ACR-368 in these tumors. Fresh tumor tissues from mice bearing established primary human endometrial and bladder cancer tissues from 20 and 18 patients, respectively, were harvested and small pieces inoculated into mice randomized into two groups, receiving vehicle control and ACR-368, respectively, as well as a PD group used to predict ACR-368 sensitivity on the tumor tissue prior to treatment.

Mice were treated in a three-days-on, four-days-off weekly schedule for four weeks at 10 mg/kg. Mice were sacrificed either four days after last dosing or when the tumor volume in one of the arms reached 2,000 mm, whichever came first. ACR-368 demonstrated anti-tumor single agent activity in a proportion of models while others were less sensitive, consistent with the prediction obtained from screening of human patient tumor samples. This result in these preclinical studies confirmed the predicted single agent activity of ACR-368 in endometrial and bladder cancer.

Figure 21.Assessment of anti-tumor activity of ACR-368 in PDX models of endometrial cancer (left two columns) and bladder cancer (right two columns) confirm that a proportion are indeed highly sensitive to ACR-368.

Blinded, prospectively designed prediction of sensitivity to ACR-368 in endometrial PDX models

To further demonstrate the predictive power of our ACR-368 OncoSignature test, we were able to obtain de-identified FFPE tissue samples from the PD arm of the endometrial cancer PDX model study. ACR-368 OncoSignature biomarker scores were generated for 18 out of 20 PDX models, as two PDX models lacked cytokeratin expression.

Using the same minimal biomarker levels established and evaluated in all our other studies summarized above, we found that eight PDX models were ACR-368 OncoSignature-positive and predicted to be sensitive to ACR-368. After unblinding of the data and analysis by a third-party biostatistician, we showed that these models all were sensitive and experienced tumor growth inhibition, or TGI, in response to treatment with ACR-368. The ACR-368 OncoSignature-negative models, which are predicted less sensitive to ACR-368, contained all the non-responsive PDX models as well as some models with overall less pronounced TGI. The segregation of non-responders from responders was statistically significant, and a sensitivity and specificity analysis demonstrated an AUC of 0.88. Despite the well-known observation that PDX models in general tend to show a much higher percentage of responders compared to human patients, as also demonstrated in our ovarian PDX model study above, this result nevertheless confirmed the ability of our ACR-368 OncoSignature test to segregate the most sensitive from non-sensitive PDX models in a blinded, prospectively designed manner.

Figure 22.Blinded, prospectively designed prediction of ACR-368 sensitivity with our OncoSignature test demonstrates segregation of responders and non-responders with a p-value = 0.003 and an AUC of 0.88.

AP3 Platform Prediction of LDG as a Rational Combination to Circumvent ACR-368 Resistance

Not all tumors are sensitive to ACR-368, and those that are sensitive can develop resistance to treatment. We used our AP3 platform to identify pathways that drive resistance to ACR-368 and to propose potential combination therapies to circumvent these resistance pathways.

As an example, we generated ACR-368-resistant ovarian cancer cell lines by growing five different human tumor cell lines, including OVCAR3, that are normally sensitive to ACR-368 in the presence of a clinically relevant dose (50 nM) of ACR-368 for over ten weeks. While most cells died, a few cells developed resistance to ACR-368 and were able to grow in the presence of the drug candidate. In general, resistant cells were at least 1,000-fold less sensitive to ACR-368 than the parental cell lines. Removal of ACR-368 for up to two months in the cell lines did not alter this level of resistance, and resistance was maintained in the presence of drug efflux inhibitors, suggesting that the resistance was not due to drug efflux from the cells, but rather permanent change in cell signaling in these cell lines drove the development of resistance.

Using AP3, we conducted global proteomic analyses comparing ACR-368 sensitive and resistant OVCAR3 cells, identifying thousands of differentially expressed proteins and phosphoproteins in these cells. Pathway mapping and analyses of these proteins and phosphoproteins showed that the activity state of proteins involved in DNA damage repair were significantly downregulated, with a compensatory upregulation of proteins involved in cell cycle progression. These changes demonstrate a low level of active DNA damage repair and hence we believe that they allowed these ACR-368 resistant cells to continue to progress through the cell cycle regardless of the presence of the drug. Furthermore, we found that cells treated with low doses of gemcitabine led to reversal of these changes, upregulating the activity of the core DNA damage repair pathways, consistent with potentially identifying a means of reversing ACR-368 resistance. This was in line with our quantitative phosphoproteomic data, which showed that treatment of ACR-368 resistant cells with LDG resulted in an upregulation of the three OncoSignature biomarkers, rendering the tumor cells more ACR-368 OncoSignature-positive after treatment with LDG.

Figure 23. Proteomic analyses of ACR-368 sensitive and resistant ovarian cell lines identified activation of proteins that regulate cell cycle progression and inactivation of proteins in the DNA damage repair pathways.

These findings suggested that tumor cells that are resistant to ACR-368 should be sensitized by treatments such as gemcitabine that function by disrupting cell cycle progression. We tested this hypothesis in cell-killing assays. The five parental human ovarian tumor cell lines were highly sensitive to ACR-368 killing with a concentration required for 50% inhibition, or EC50 between ten to 30 nM. The EC50 for OVCAR3 was 15 nM ACR-368. By contrast, the resistant OVCAR3 cells had an EC50 of over 10 μM, which means they were over 1,000-fold less sensitive to ACR-368. Treatment of these cells with 0.53 nM gemcitabine, lowered the EC50 for ACR-368 to 100 nM. A further increase in gemcitabine concentration to 2.7 nM lowered the EC50 for ACR-368 to 6 nM. Likewise, treatment of the parental cells with the same low doses of gemcitabine increased the sensitivity to ACR-368. Treatment of these cells with 0.53 nM gemcitabine lowered the EC50 for ACR-368 to 2.7 nM. A further increase in gemcitabine concentration to 2.7 nM lowered the EC50 for ACR-368 to 0.2 nM. These findings of synergy between ACR-368 and LDG were extended into other human tumor cell lines, including endometrial and bladder.

Support for the synergistic action of ACR-368 and gemcitabine comes from the observation that gemcitabine alone does not induce potent cell death in OVCAR3 cells: at 0.53 and at 2.7 nM of gemcitabine there was no effect on cell survival, and more than half of treated cells survived at concentrations exceeding 30 μM.

Figure 24. Low concentrations of gemcitabine sensitize a highly resistant ovarian cancer cell line to ACR-368.

Based on these results, we are treating patients who are predicted to be resistant to ACR-368 in our clinical trials with LDG in combination with ACR-368 to potentially overcome resistance to ACR-368.

Our Ongoing Phase 2 Clinical Trials of ACR-368 Based on ACR-368 OncoSignature-Predicted Drug Sensitivity

Based on these preclinical results, we are conducting a Phase 2 clinical trial where we were initially enrolling and treating patients with all three tumor types: platinum-resistant ovarian, endometrial, and bladder cancer. ACR-368 OncoSignature-positive patients, which we predicted would represent 30% to 40% of patients of each tumor type, received ACR-368 monotherapy in a single arm Phase 2b trial for each of the three tumor types. The ACR-368 OncoSignature-negative patients with one of these three tumor types receive ACR-368 combined with LDG at the RP2D of ACR-368 and RP2D established in the study which is 10 mg/m2 in the exploratory Phase 2 dose expansion portion of the trial. As a result, all patients with these tumor types that have been biopsied were eligible to receive therapy. Akoya procures and manufactures the necessary supplies to perform the OncoSignature tests. Based on our communications with the FDA to date, we believe the monotherapy trial, if successful, has the potential to be registrational for ACR-368. At ESMO 2024 (September 14, 2024 R&D event and press release), we reported that endometrial cancer was our prioritized indication, as it represents the first potential registrational opportunity for ACR-368. We remain confident in this strategy based on emerging clinical data, competitive positioning given limited treatment options, and the strong commercial opportunity in both second- and front-line settings. Our blinded KOL market research estimates that there are approximately 27,000 U.S. patients annually in the second-line setting alone for endometrial cancer. Due to increased competition and a smaller market opportunity, we set a high internal clinical bar for ovarian cancer, which preliminary data suggests is unlikely to be met. Bladder cancer is also being deprioritized due to lower than preclinically predicted BM+ rate, leading to challenging enrollment with single digit BM+ patient enrollment to date. We have now officially deprioritized ovarian and bladder cancers, reallocating all clinical resources to ACR-368 in endometrial cancer and ACR-2316.

who had progressed on prior anti-PD-1 therapy, unless ineligible. Patients of all histopathologies (endometrioid, serous, clear cell, and carcinosarcoma) and regardless p53, MMR and other molecular status are eligible. Endometrial cancer had not been previously studied in prior ACR-368 Lilly-sponsored trials. Using AP3 for indication screening, this tumor type was predicted to be particularly sensitive to ACR-368 prior to the current ongoing Phase 2 study. The data were based on 35 safety-evaluable patients, of which 23 (8 BM+ and 15 BM- patients) were efficacy-evaluable with at least one on-treatment scan (data cut off July 25, 2024). This data included a confirmed ORR of 62.5% (95% CI, 30.4-86.5) observed in prospectively-selected BM+ patients with endometrial cancer. Among subjects who had a confirmed partial response, 3 out of 5 had shown a previous best overall response as progressive disease on last prior line of therapy. The data further validated our AP3-based ACR-368 OncoSignature assay, which is used for prospective patient selection in this registrational intent trial, achieving a clear segregation of responders in the OncoSignature-positive versus OncoSignature-negative arms (p-value = 0.009). Consistent with previous Phase 2 clinical trials conducted with ACR-368, the AEs were primarily reversible, transient, mechanism-based hematological, including neutropenia and thrombocytopenia. Based on these results, which demonstrated single agent RECIST activity in patients who all had progressed on anti-PD-1 and included patients with the most difficult to treat forms of endometrial cancer, including p53 mutated, pMMR serous and carcinosarcomas, Acrivon is planning a confirmatory trial combining ACR-368 with anti-PD-1 vs anti-PD-1 in the maintenance phase of the recently approved first-line therapy for endometrial cancer. This is further supported by the strong rationale and synergy observed in preclinical studies combining ACR-368 with anti-PD-1.

An interim data extract from the EDC clinical database was done on February 25, 2025, including 20 BM+ endometrial cancer patients treated with ACR-368 monotherapy and 38 BM- treated with ACR-368 plus LDG that were efficacy-evaluable by RECIST (2 BM- had treatment discontinued without scan). All BM+ patients had progressed after prior platinum-based chemotherapy and prior anti-PD-1, and the median and mean prior lines of therapy for these patients were 2 and 2.6, respectively. A majority of these BM+ patients were refractory to the last prior line of therapy, with aggressive, generally heavily pre-treated tumors: 12 had refractory disease (best overall response of PD in the last prior line of therapy), 6 had relapsed disease, and 2 unknown. Amongst these 20 BM+ patients, 15 were either serous or carcinosarcomas, 13 were pMMR (2 dMMR, 5 not tested), and 11 p53 mutated (3 wild-type; 6 unknown). In patients that had relapsed after the prior line of therapy (N=6), the confirmed ORR was 50% and the DCR was 100%. Amongst the 12 patients with tumors refractory to the last prior line of therapy (ORR = 0%) we observed meaningful ACR-368 clinical activity with a confirmed ORR of 33% and DCR of 75%. The ACR-368 OncoSignature accurately identified patients whose tumors are sensitive to ACR-368, with 80% of BM+ patients demonstrating tumor shrinkage. Among all 20 BM+ patients the confirmed ORR was 35% and the DCR was 80%. Overall, we observed significant anti-tumor activity and disease control in BM+ patients with aggressive, refractory tumors that did not respond at all (0 % ORR) to the last line of prior therapy, and with a confirmed ORR more than double (35%) the best ORR observed in the last prior line of therapy (15%) for all BM+ patients.

In the ACR-368 OncoSignature-positive arm, up to 23 endometrial cancer patients will receive ACR-368 monotherapy at RP2D. The trial incorporated an opportunity to refine the OncoSignature biomarker patient selection threshold based on the first 12 patients treated with ACR-368 monotherapy. An interim futility analysis will be used to exclude the non-interesting response rate and assess the ORR. Based on this result, the study is designed to enroll up to an additional 48 patients with a registrational intent. We have completed the Phase 1b portion of the study exploring ACR-368 and LDG combination in OncoSignature-negative patients and have progressed into Phase 2 of this exploratory component in OncoSignature-negative patients dosing ACR-368 at 105 mg/m2 with the LDG dose of 10 mg/m2.

Figure 25. Design of the single arm Phase 2 ACR-368 monotherapy and single arm Phase 1b/2 ACR-368 with LDG combination trials.

The prior studies in SCCHN and anal cancer, described above, have demonstrated an unenriched ORR of 19% in patients with HPV+ SCCHN, and 15% in patients with anal cancer. Moreover, the mDoR was seven months for SCCHN and above 12 months in anal cancer. The FDA has granted ODD for ACR-368 for anal cancer. Preclinical screening on human patient tumor samples suggests that approximately 25% of cases of these cancers have activated biochemical signaling pathways that are consistent with sensitivity to ACR-368. Dr. Christine Chung at Moffitt Cancer Center has initiated an Investigator-Initiated Trial in both HPV+ and HPV- SCCHN to assess the combination of ACR-368 with LDG.

Figure 26. Clinical results reported in September 2024 in ongoing registrational-intent trial of ACR-368.

Figure 27. Confirmed responses in endometrial subjects who all progressed on prior anti-PD-1.

Several past investigator-initiated trials, or IITs, have demonstrated clinical activity of ACR-368 in combination with various chemotherapeutic agents in patients with different types of sarcomas. These are of high unmet need for improved treatments, and only 17% of patients with metastatic soft tissue sarcomas survive more than five years. Importantly, these IITs have not only demonstrated clinical activity of ACR-368, but have also demonstrated that combination with chemotherapy is generally well-tolerated in these patients. For example, in a Phase 1/2 trial in patients with relapsed/refractory desmoplastic small round cell tumor and rhabdomyosarcoma conducted at Memorial Sloan Kettering Cancer Center, it was reported that ACR-368 in combination with irinotecan resulted in a 32% ORR and mPFS of over 5.5 months. The combination was generally well-tolerated, leading to primarily hematological adverse events, which were manageable. We intend to initiate certain carefully selected trials in patients with sarcomas that have demonstrated promising preliminary clinical results in past trials at a later date.

Our Internally Discovered and Developed Programs

ACR-2316 is our first internally discovered clinical stage asset. It is a novel, dual WEE1 and PKMYT1 inhibitor small molecule development candidate, rationally designed through advanced co-crystallography and the AP3 platform to achieve optimal target potency and selectivity through strong activation of not only CDK1 and CDK2, but also of PLK1 to drive pro-apoptotic cell death, as observed in preclinical studies against benchmark inhibitors, delivering potent single agent anti-tumor activity across in vitro and in vivo preclinical studies, compared to benchmark WEE1 and PKMYT1 inhibitors. Clinical WEE1 inhibitors have demonstrated promising anti-tumor activity in early clinical trials conducted by competitors; however, their clinical activity has been hindered by a narrow therapeutic index and WEE1 inhibitor-induced resistance mechanisms. ACR-2316 was specifically designed using AP3 to address these limitations through very high selectivity to limit adverse events to mechanism-based, on-target, and to simultaneously inhibit PKMYT1, a closely related protein serine/threonine kinase also serving critical functions in the cell cycle and DDR pathways, that accounts for a major part of WEE1 inhibitor-induced resistance, as revealed by AP3. Based on mechanism of action and confirmed in our preclinical studies, balanced inhibition of PKMYT1 results in more potent single agent activity.

We believe there is a need for novel patient selection methods to overcome the challenges with genetics-based patient selection methods, and that using AP3 will enable us to identify drug-sensitive indications and individual patients predicted sensitive to WEE1 and PKMYT1 inhibitors. Using AP3 for unbiased quantitative high-resolution measurement of the effects of ACR-2316 on the human tumor cell phosphoproteome, this compound has been optimized for potent induction of mitotic catastrophe, which is key to its strong single agent activity in preclinical models and potentially favorable clinical profile for monotherapy development. ACR-2316 was discovered by AP3-based SAR, facilitated by co-crystallography, and designed by AP3 to overcome WEE1-induced resistance mechanisms. The preclinical data generated demonstrate a very potent single agent activity so we believe patient selection may not be needed in the most sensitive tumor indications. Nevertheless, we plan to generate an OncoSignature to be used for identifying the sensitive tumor indicationsto enable drug target engagement-based dose optimization using the OncoSignature assay. Using AP3-based Indication Finding and AP3-based analyses of in-house and publicly available data, we are enrolling selected high unmet need solid tumor types predicted sensitive to ACR-2316 in our Phase 1 trial.

Preclinical Program

Expansion of Our Pipeline Through Application of AP3

We have shown that our AP3 platform is capable of generating OncoSignature tests that can predict preclinical sensitivity to a number of potential cancer therapies. We are applying the power of this technology to expand our pipeline in several ways:

•

Selectively pursue carefully selected in-licensing candidates for which we believe a genetics-based patient selection method is challenging and where we believe an OncoSignature predictive test can be developed that will significantly improve response rates, similar to how we identified ACR-368.

•

In addition to ACR-368 and ACR-2316, Acrivon is also leveraging its proprietary AP3 precision medicine platform for developing its co-crystallography-driven, internally-discovered preclinical cell cycle program with an undisclosed target.

•

Establish carefully selected co-development partnerships with leading biopharmaceutical organizations that either have approved products or attractive drug candidates under competitive pressure where the availability of an OncoSignature test could significantly increase response rates, leading to new drug approvals, label expansions and the ability to deliver effective therapies to the right patients.

Broad Utility and Applications of Our AP3 Platform

Based on our extensive studies, we have demonstrated that our AP3 platform has many high impact applications, including:

•

Predictive biomarkers and patient responder identification. Our AP3 platform enables identification of predictive biomarkers that are assembled into OncoSignature tests used to select patients to be treated that are predicted to be sensitive to a drug or drug candidate, so-called patient responders. This capability has been demonstrated in the studies described above. Using this approach, we have also developed predictive OncoSignature tests for a clinical stage CDK7 inhibitor and a clinical stage CDC7 inhibitor. The goal is to only treat patients most likely to benefit from the drug and avoid overtreatment of patients that do not benefit from it with the potential for side effects.

•

ACRV US Equity

ACRV · 10-K · period ended 2024-12-31

Item 1. Business.

ACRV US Equity☆

ACRV · 10-K · period ended 2024-12-31

Item 1. Business.

ACRV US Equity