Lenvatinib resistance remains a critical barrier in advanced hepatocellular carcinoma (HCC) therapy. However, the underlying mechanisms and strategies for reversing resistance remain incompletely understood. Integrated transcriptomics of lenvatinib-resistant patient tumors and an acquired-resistance murine model identified a novel macrophage subpopulation. Functional validation employed CRISPR-SAM screening, conditioned medium (CM) assays, subcutaneous/orthotopic xenografts, patient-derived organoids (PDOs), and xenografts (PDXs). Mechanistic studies included ChIP-qPCR, co-immunoprecipitation, and pharmacologic targeting. Clinical relevance was assessed in a retrospective cohort. Resistant HCC exhibited significant enrichment of a COLEC12high TAM subset, which correlated with poor survival and treatment response. These TAMs secreted neuregulin-1 (NRG1), activating HER2/HER3-AKT signaling in tumor cells to drive cancer stemness and lenvatinib resistance. Mechanistically, in TAMs COLEC12 sequestered STAT1 in the cytoplasm, preventing its phosphorylation, and thereby derepressing STAT3-mediated neuregulin-1 (NRG1) transcription. Depletion of NRG1 reversed the stemness phenotypes and resensitized tumors to lenvatinib both in vitro and in vivo. Clinically, high NRG1 expression predicted an inferior lenvatinib response and shorter survival. Crucially, the bispecific anti-HER2/HER3 antibody zenocutuzumab restored lenvatinib efficacy in PDOs, PDXs, and murine models. Our work establishes the COLEC12high TAM / NRG1 axis as a master regulator of therapeutic resistance and identifies NRG1 as a predictive biomarker, providing a clinically actionable strategy to overcome lenvatinib resistance in HCC.

BackgroundData on neoadjuvant treatment with trastuzumab biosimilars, particularly CT-P6, in combination with pertuzumab, are limited. This study evaluates the efficacy, tolerability, and immunogenicity of CT-P6 plus pertuzumab and chemotherapy, in routine clinical practice for HER2-positive early breast cancer, including translational biomarker analyses related to pathologic complete response (pCR).MethodsProspective, multicenter, observational study in 102 patients with HER2-positive early breast cancer. Patients received hospital-preferred neoadjuvant regimens protocols, with (scheme 1 and 3) or without anthracyclines (scheme 2). The primary endpoint was pCR, defined as the absence of invasive tumor in both the breast and axillary lymph nodes (ypT0/ypTis and ypN0). Translational endpoints included soluble HER2, anti-trastuzumab CT-P6 antibodies, and exploratory response-related modeling approaches supported by machine learning techniques.ResultsAmong patients who underwent surgery, pCR (ypT0/ypTis and ypN0) was achieved in 57.43% of cases, with no significant differences between anthracycline-based and non-anthracycline-based regimens. Soluble HER2 and anti-trastuzumab CT-P6 antibodies were not significantly associated with pCR. Treatment was well-tolerated; the most relevant Grade 3–4 treatment-related adverse events were diarrhea (2.25%) and asthenia (0.50%). No immunogenicity or clinically relevant cardiotoxicity was observed.ConclusionsTrastuzumab CT-P6 combined with pertuzumab and chemotherapy can be used in neoadjuvant treatment for HER2-positive early breast cancer, showing pCR rates comparable to the reference trastuzumab and without evidence of immunogenicity. Exploratory analyses of soluble HER2 and anti-trastuzumab CT-P6 antibodies did not demonstrate a significant association with pCR, although this possibility cannot be excluded. Their assessment contributes to the translational understanding of biosimilar integration into curative regimens.Trial registration: The study has been registered in Clinicaltrials.gov (https://clinicaltrials.gov/study/NCT06907082).Supplementary InformationThe online version contains supplementary material available at 10.1007/s10549-026-07895-8.

ABSTRACT Our laboratory specializes in the intratumoral (i.t.) delivery of virus nanoparticles (VNPs) that activate immune cells and modulate the tumor microenvironment (TME). For this study, we conjugated the CH401 peptide—containing B–cell, helper T-cell, and cytotoxic T–cell epitopes of HER2—to the plant cowpea mosaic virus (CPMV) and bacteriophage Qβ to develop vaccines for HER2+ cancer. In vivo studies confirmed that the CPMVCH401 i.t. therapeutic vaccine candidate was successful in CT26-HER2 and MC38-HER2 tumor models through the induction of IgG2a antibodies, upregulation of Th1 cytokines, and activation of Th1 cells. On the other hand, the QβCH401 i.t. therapeutic vaccine candidate induced IgG1 antibodies and a Th1/2 balanced response, which did not result in potent anti-tumor efficacy. This study highlights that different VNP carriers are differential in immune modulation and that CPMVCH401, but not QβCH401, is effective as an i.t. vaccine. We also learned that the subcutaneous prophylactic immunization route leads to stronger humoral immunity but fails to modulate the TME, limiting the anti-tumor efficacy. This study highlights that, combined with the HER2 epitope, the i.t. therapeutic approach has powerful anti-tumor capabilities. It induces therapeutic HER2 antibodies, immunomodulation of the TME, and T–cell responses against the tumor, making this a robust active immunotherapy regimen.

154eP HER2 and TROP2 antibody–drug conjugates in metastatic breast cancer: A systematic review of clinical outcomes

Abstract BACKGROUND Leptomeningeal disease (LMD) is a lethal complication of breast cancer (BC), occurring in ~5% of patients and characterized by poor prognosis. The cerebrospinal fluid (CSF) immune environment in LMD is predominantly innate, lacking adaptive immune responses. In murine LMD models, intrathecal (IT) administration of cDC1s was safe, elicited a CD4+ Th1-dominant response, eliminated HER2+ LMD, and prevented recurrence. These findings led us to hypothesize that in BC LMD patients, IT cDC1s could induce an adaptive Th1 immune shift in the CSF, as detected by transcriptomic analysis. METHODS We conducted a phase I, single-arm, dose-escalation trial (NCT05809752) in patients with triple-negative or HER2+ BC LMD. Participants received weekly IT cDC1s for up to 12 weeks across four dose levels (1 × 10⁶–5 × 10⁷ cells), using a BOIN design to determine the maximum tolerated dose (MTD). Primary endpoints were safety and MTD; secondary endpoints included clinical response (RANO-LM) and CSF immune profiling, including single-cell RNA sequencing (scRNA-seq). RESULTS (AS OF 03/15/25) Six patients received doses ranging from 1–2 × 10⁷ cells with no dose-limiting toxicities. Headaches were the most common adverse event (grade 3 in two patients). Five patients (83%) showed stable or improved performance status; median overall survival was 48 weeks. Cytokine analysis revealed elevated Th1 markers (IFN-γ, IL-6, IL-12, TNF-α) without increased IL-4. Transcriptomic analysis of CSF showed a remodeled immune landscape with expanded CD4+, CD8+, and B cell populations and reduced tumor cell burden. Anti-HER2/HER3 antibodies were detected in 80% of patients. CONCLUSION IT cDC1s are safe and induce robust Th1-polarized adaptive immune responses in BC LMD, as confirmed by transcriptomic and cytokine analyses. This approach offers a novel strategy to convert the immunologically cold CSF environment into one capable of mounting effective anti-tumor immunity.

HER2 expression in an endometrial cancer cohort.

Patients with HER2-low breast cancer (BC) may be eligible for trastuzumab-deruxtecan (T-DXd) treatment. However, studies have shown that different HER2 antibodies vary in their sensitivity for low HER2 expression, potentially impacting HER2-low BC diagnosis and patient selection for T-DXd. We investigated the frequency of HER2-low BC in relation to the HER2-antibody used across Dutchpathology laboratories. Patients with primary BC without neoadjuvant treatment, diagnosed between 2013 and 2024, were included. HER2-low frequencies from 34 laboratories were obtained from the Dutch Nationwide Pathology Databank (Palga). Additional information (e.g., type of HER2 antibody, staining protocol) was obtained through a questionnaire. A total of 88,713 patients were included, representing 103,505 tumors, of which 94,934 had a conclusive HER2 status. Among non-amplified cases, HER2-low frequencies varied widely across laboratories (33.4%-94.5%), with a gradual increase since 2022. The most commonly used antibody clones were 4B5 (n = 21), DG44 (n = 7), A0485 (n = 4), and SP3 (n = 2). HER2-low proportions were highest with A0485 (71.5%), followed by DG44 (66.7%), SP3 (60.1%), and 4B5 (59.1% with Ultraview, 57.0% with Optiview). Substantial inter-laboratory variation was observed even within the same antibody group (4B5/Ultraview: 40.5%-80.4%; 4B5/Optiview: 37.3%-68.4%; DG44: 40.6%-95.4%; A0485: 62.3%-94.7%; SP3: 31.6%-78.6%). Our data showed a notable variation in HER2-low BC frequency across Dutch pathology laboratories, even among those using the same antibody and detection system. These differences may influence patient eligibility for T-DXd.

Abstract Background: HER2-targeted therapy is broadly used in advanced breast cancer (aBC). For HER2+ aBC (HER2 IHC 3+ or 2+/ISH+), 1L standard of care includes unconjugated HER2 antibodies trastuzumab and pertuzumab (HP) in combination with chemotherapy. In the 2L+ setting, antibody drug conjugate trastuzumab deruxtecan (T-DXd) can be used for HER2 low (IHC 1+, 2+/ISH-) patients with significant benefit seen in DESTINY Breast 04 relative to physicians’ choice chemotherapy. For some targeted therapies, including MET inhibitor capmatinib, the magnitude of genomic copy number (CN) gains predict benefit. Here, we examined a real-world cohort of aBC patients (pts) treated with HER2 antibody therapies to determine if HER2 genomic CN ratio predicts outcomes. Methods: This study used the US nationwide de-identified Flatiron Health (FH) and Foundation Medicine Inc. (FMI) clinico-genomic database including patient-level structured and unstructured retrospective longitudinal clinical and genomic data from approximately 280 cancer clinics (approximately 800 care sites) in the United States and included patients who underwent tissue comprehensive genomic profiling (FoundationOne®/FoundationOne®CDx) after 2014. HER2 CN ratio was defined as ratio of modeled absolute CN to specimen ploidy. Results: 121 pts with HER2+ aBC were treated with HP + chemotherapy in 1L and received genomic profiling before start of 2L. HER2 CN ratio was generally high (median HER2 CN ratio of 10; IQR 2.8-17.8). Pts with a HER2 CN ratio of >5 (equivalent to CN 10 in a diploid tumor) had significantly better TTD (median 8.4 v 5.3 mo, HR = 0.55, 95% CI 0.36-0.84, p = 0.006), and OS (median 76 v 31mo, HR = 0.33 95% CI 0.18-0.61, p <0.001) than pts with a HER2 CN ratio of ≤5, and comparable trends were observed for PFS (median 13v 9mo, HR = 0.81 (95% CI 0.52-1.26), p = 0.35). Similar benefit was seen for pts with a HER2 CN ratio of 5-10, 10-15, 15-20, and 20+ relative to HER2 ratio ≤5. For 95 pts with HER2 low BC treated with T-DXd in the 2L or 3L setting, HER2 CN ratio was generally low (median ratio of 1, IQR 1-1) with a maximum HER2 CN ratio of 2.5 in the cohort. Across HER2 CN levels, outcomes were similar with the exception of pts with a HER2 CN ratio of ≤0.5 (11/95; 12%) who had significantly worse PFS (median 2.5 v 6.1 mo, HR = 0.37, 95% CI 0.17-0.79 p = 0.01) and OS (median 6.5 v 25.2 mo; HR =0.32, 95% CI 0.14-0.75; p=0.008) than pts with a HER2 CN ratio of >0.5, with similar trends for TTD (median 1.6 v 4.8 mo, HR = 0.52, 95% CI 0.26-1.06 p = 0.07). Conclusions: In a cohort of real-world aBC pts treated with HER2 antibody therapies, HER2 CN ratio was significantly associated with clinical outcomes. For unconjugated antibodies, pts with a HER2 CN ratio of <5 had significantly shorter TTD, PFS and OS. Future work should explore whether these pts may benefit from therapy escalation (e.g. use of conjugated HER2 antibodies) and/or increased surveillance. For the HER2 conjugated antibody T-DXd, pts with HER2 low BC with a hemizygous or deep deletion in HER2 (HER2 CN ratio ≤0.5) had significantly worse outcomes, consistent with partial or complete target loss. These results suggest potential additional predictive value of an NGS-based HER2 CN quantitative biomarker that could be deployed in conjunction with the currently employed qualitative IHC status biomarker to identify pts for whom an alternative therapy may be more efficacious. For pts with a ratio of >0.5, benefit was seen across the range of HER2 CN ratios suggesting that even low levels of HER2 are enough for activity of T-DXd, consistent with the high payload:antibody ratio and bystander effect for T-DXd. Prospective or observational studies should be performed to validate these findings. Citation Format: Nicole Odzer, Pusztai L, Chen KT, Jin DX, Sisoudiya S, Schrock A, Ross J, Sokol ES, Lustberg M. Targeting clinically advanced breast cancer with conjugated and unconjugated HER2 antibodies: Does copy number matter? [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2024; 2024 Dec 10-13; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2025;31(12 Suppl):Abstract nr PS8-04.

Abstract Introduction: Antibody-Drug Conjugates (ADCs) have had tremendous impact on patient outcomes in breast cancer and are now second-line therapy for stage IV HER2 high expressing metastatic breast cancer as well as HER2 low expressing tumors. However, many patients fail to respond or relapse after treatment with ADC therapies due to tumor heterogeneity and eventual resistance to the ADC payload. We are developing next generation Multi-Payload Conjugates™ (MPCs™) that deliver targeted combination chemotherapies within a single molecule. Method: CatenaBio has developed novel, highly stable single-molecule targeted combination therapies, MPCs™, with tunable drug-antibody ratio (DAR). Our selective conjugation platform allows the attachment of distinct payloads targeting different mechanisms of action at three unique sites on the antibody scaffold. Results: We screened combinations of different payloads targeting several different mechanisms of cell division attached to trastuzumab as well as sacituzumab at different DARs to optimize tumor cell killing. These targeted combination ADCs demonstrated robust killing in multiple cell lines containing high and low expression of HER2 and TROP2, as well as an Enhertu resistant cell line. Additionally, these novel MPCs show potent inhibition and excellent tolerability in mouse models of tumor growth in HER2 high and HER2 low/TROP2 expressing xenograft models including models of TNBC. Conclusion: While advances have been made in the design of ADCs to expand to previously unaddressed populations, high patient relapse and the failure of recent mono-payload ADCs in late-stage trials indicate a need for novel conjugate modalities. Multi-payload Conjugates™ offer the next step in ADC design and allow for the combination of multiple mechanisms of action in a single MPC™ that are highly effective across multiple breast cancer cell lines and target expression levels. Successfully constructed with both HER2 and TROP2 targeting antibodies, these molecules offer the potential to circumvent tumor resistance pathways and deliver deeper and more durable patient responses. Citation Format: Marco Lobba, Richard Kendall, Devin Trinter, Maxwell Nguyen, Daniel Gutierrez, Samantha Brady, Chanez Symister, Andrew Lau, Derek Garcia-Almedina, Saurabh Johri, Matthew Francis. Next generation antibody drug conjugates targeting HER2 and TROP2: Multi-Payload Conjugates™ targeting orthogonal mechanisms of cell killing [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2024; 2024 Dec 10-13; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2025;31(12 Suppl):Abstract nr P4-04-27.

Abstract Background: Anti-HER2 antibody-drug conjugate (HER2-ADC) therapies, such as trastuzumab emtansine (T-DM1) and trastuzumab deruxtecan (T-DXd), significantly prolong survival in patients with HER2-positive metastatic breast cancer (BC) compared to physician’s choice of chemotherapy with trastuzumab (The EMILIA, TH3RESA, and DESTINY-Breast clinical trials). Intrinsic or acquired resistance to HER2-ADC therapies remains a substantial clinical challenge since there is no established standard of care following progression on T-DXd. This study aimed to elucidate novel therapeutic targets that can overcome resistance to HER2-ADC therapies using unbiased approaches with synthetic lethal kinome RNA interference screening. Methods: We conducted targeted DNA sequencing (n=15) and whole transcriptome sequencing (n=11) to investigate genetic aberrations following anti-HER2 and/or HER2-ADC therapy (T-DM1 or T-DXd) in BC patient tissue samples. We generated HER2-positive human BC cell lines resistant to T-DM1 or T-DXd. To determine the effect of the T-DM1 and T-DXd on the ERBB2 gene and HER2 protein expression, we used fluorescence in situ hybridization, droplet digital PCR, and Western blotting assay. To discern resistance mechanisms of T-DXd and identify kinase targets whose inhibition might synergistically enhance the efficacy of T-DXd, we conducted whole-genome sequencing, cDNA microarray analysis, and synthetic lethal kinome RNA interference screening using T-DXd resistant HER2-positive BC cell lines. To determine the synergistic antitumor effect of T-DXd combined with targeted therapy, we conducted Bliss synergy analysis, colony formation, and xenograft assays by mammary fat pad injection of T-DM1- or T-DXd-resistant BC cells in nude mice. Results: Increased amplification of DNA repair-related genes (TOP2A, RAD21, RAD52, and CDK12) was found in patient tissue samples after progression to HER2 antibodies, T-DM1, or T-DXd. Significant enrichment of DNA damage repair-related gene sets was observed in the transcriptome of post-treatment human samples. Further, ERBB2 gene or HER2 protein expression was reduced compared to levels before such treatment. In HER2-ADC-resistant HER2-positive BC cell lines, we had similar upregulation of DNA repair-related genes (PCNA, ATM, RAD52) and reduction of ERBB2 gene and HER2 protein expression. By synthetic lethal kinome RNA interference screening, this non-bias screening identified the PI3 kinase, cell cycle, and DNA repair canonical pathways as potential targets to enhance the efficacy of T-DXd therapy in HER2-ADC–resistant HER2-positive BC. We confirmed that ectopic HER2 expression does not improve the efficacy of HER2 ADC in HER2-ADC–resistant HER2-positive BC. After further screening all potential targets, we found that the drugs targeting the DNA damage repair pathway were the most effective in enhancing the efficacy of T-DXd. Indeed, ATR inhibitor elimusertib led to significant HER2-ADC–resistant BC cell death in vitro (Bliss synergy score >5.0, P<0.01) and in four xenografts in vivo (P<0.01) compared to cell death with T-DXd alone. Importantly, we further observed the synergy of elimusertib and T-DXd in parental HER2-positive BC cell lines, ensuring the observed effects are not limited to resistant cell lines only. Conclusions: Our findings indicate that resistance to HER2-ADC therapies is associated with increased DNA repair-related genes. By non-biased screening, this study provides robust evidence that targeting DNA repair pathways can significantly enhance the efficacy of T-DXd in HER2-ADC-resistant HER2-positive BC. The potential of combining ATR inhibitors with T-DXd to overcome resistance and improve patient outcomes is a promising avenue that warrants further exploration in a clinical trial for patients with HER2-ADC-resistant- HER2-positive BC. Citation Format: Jangsoon Lee, Kumiko Kida, JiwonKoh, Huey Liu, GanirajuC. Manyam, Young JinGi, Dileep Reddy, Asha S. Multani, Jing Wang, Gitanjali Jayachandran, Dae-Won Lee, James M. Reuben, AysegulSahin, Lei Huo, DebuTripathy, Seock-Ah Im, and Naoto T. Ueno. Enhancing T-DXd Efficacy in HER2-positive Breast Cancer Resistant to HER2 ADC by Non-biased Kinase-related Target Screening [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2024; 2024 Dec 10-13; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2025;31(12 Suppl):Abstract nr PS13-07.

Abstract Biomarkers for specific cytotoxic chemotherapy sensitivity could better inform drug selection. Both CEP17 duplication and abnormal Topoisomerase 2 copy number appear associated with anthracycline sensitivity in the adjuvant setting (Bartlett et al 2015). This data is however not currently applied clinically. Taxanes are a routine component of adjuvant and neoadjuvant chemotherapy either in combination or sequenced with anthracyclines, or increasingly used in the absence of anthracyclines. ROSCO: ISRCTN15094808 was designed to prospectively evaluate the clinical utility of these two biomarkers for initial neoadjuvant chemotherapy selection. Between November 2015 and May 2023, 990 consenting patients with early breast cancer considered suitable for neoadjuvant chemotherapy were randomised to four cycles of either Epirubicin and Cyclophosphamide with optional 5 Fluorouracil ((F)EC), or Docetaxel and Cyclophosphamide (TC). Patients with Grade 1 or 2 ER Rich, PR Rich, HER-2 negative tumours and all T1 N0 tumours were excluded. All HER-2 positive cancers were treated with concurrent anti HER-2 antibodies. Participants were stratified by centrally assessed biomarker status as biomarker normal (BM normal) with both CEP17 and TOP2A normal, or biomarker abnormal (BM abnormal) with CEP17 duplication and/or TOP2A abnormal. Surgery was performed after 4 cycles of chemotherapy; where pathological complete response (pCR) was not achieved, crossover to the alternative treatment arm for a further 4 cycles was given in an adjuvant setting. Crossover before surgery was permitted where interim biopsy after 4 cycles confirmed viable residual disease. The primary endpoint of the study is pCR ypT0/Tis ypN0 after initial neoadjuvant chemotherapy. Of the 990 patients consented, 496 were randomised to TC and 494 to (F)EC. 24 patients with no cancer seen in interim biopsy received further neoadjuvant chemotherapy and also had pCR at final surgery, these are considered in the primary analysis as non pCR. Data from 950 patients are evaluable for the primary endpoint. Overall pCR was 245 (26%): the TC arm was 131 (27%) and (F)EC arm was 114(24%). Overall BM was normal in 233 (24%) with BM abnormal in 756 (76%). With TC the pCR percentages for BM normal and abnormal were very similar: 30% and 27% respectively. For (F)EC BM normal, pCR is 17% and for FEC BM abnormal, it is 26%. Final data cleaning is ongoing, testing for treatment by biomarker interaction will be presented. Higher response to (F)EC in the biomarker abnormal group was observed across all pathological subtypes tested. Sensitivity analysis excluding 35 TC and 22 (F)EC patients where crossover chemotherapy was given off protocol neoadjuvantly despite a negative core biopsy or where patients withdrew prior to the primary endpoint was conducted. The pCR proportions observed in the treatment by biomarker groups were not impacted. Preliminary analysis of this large prospective evaluation of CEP17 and TOP2A as potential predictors of anthracycline sensitivity conducted in a neoadjuvant context shows that evaluation of these biomarkers shows no predictive value for sensitivity to TC but demonstrates differential pCR to (F)EC. Suggesting women with BM abnormal cancers are likely to benefit more from inclusion of anthracyclines. Anthracycline-free chemotherapy may be considered as an option for women with BM normal cancers. This work was supported by CRUK [CRUK/12/046/ A15756] and Bristol Myers Squibb. Citation Format: Daniel Rea, S. Pirrie, L. Hayward, S. Chan, M. Varughese, S. Spensley, U. Barthakur, M. MacKenzie, S. J. Bowden, C. Gaunt, E. Southgate, N. Nicholson, P. Wetherell, M. Soden, L. Billingham, C. Brookes, D. Cameron, J. Starczynski, J. Dowds. H. Earl, R. Ste. ROSCO: Response to Optimal Selection of neoadjuvant Chemotherapy in Operable breast cancer: Randomised phase III, stratified biomarker trial of neoadjuvant 5-Fluorouracil,Epirubicin & Cyclophosphamide vs Docetaxel & Cyclophosphamide chemotherapy [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2024; 2024 Dec 10-13; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2025;31(12 Suppl):Abstract nr RF3-07.

With increasing use of human epithelial growth factor receptor two (HER2)-targeted therapies, outcomes for numerous breast cancer patients have improved. Nevertheless, patients with HER2-positive tumours face acomparatively heightened risk for developing brain metastases (BM), which are often treated with stereotactic radiosurgery (SRS). Radionecrosis represents one of the clinically most significant adverse events of SRS. However, aknowledge gap remains regarding the effects of concurrent use of HER2-targeted therapies with SRS on development of radionecrosis, given conflicting findings in existing studies. This systematic review was conducted in May 2024 through asearch across electronic databases PubMed/MEDLINE and Cochrane library and was supplemented by citation searching and an artificial intelligence (AI) search. The literature search yielded 194 articles. After applying eligibility criteria, atotal of 13studies with 3219 patients total were included, with approximately 270 patients in the topic-relevant subgroup. Investigated substances vary in different publications and include HER2 antibodies, antibody-drug conjugates (ADCs), such as trastuzumab emtansine (T-DM1), and kinase inhibitors. Four of six studies on ADCs demonstrated ahigher risk for radionecrosis with concurrent administration. Two studies on lapatinib found no significant effects, as did as most studies investigating mainly HER2 antibodies. One publication reported an even lower risk for radionecrosis (RN) with concurrent use of HER2/EGFR tyrosine kinase inhibitors (TKIs). While concurrent use of T‑DM1/ADCs seems associated to elevated radionecrosis risk, an ambiguous situation for other substances persists. Heterogenous study designs with varying substances, definitions of concurrent use, and radionecrosis parameters must be considered. Included studies are partly limited by sample size and retrospective study design. Therefore, clinical implications remain difficult to claim; further research on this topic is needed.

e13138 Background: TROP2 and HER3 are both highly expressed in TNBC and associated with worse survival. JSKN016 is a bispecific TROP2/HER3-targeting ADC developed with novel linker to conjugate the payload, a topoisomerase Ⅰ inhibitor with a drug-to-antibody (DAR) of 4. Methods: JSKN016-101 (NCT06592417) is an ongoing first-in-human (FIH), dose-escalation (0.5-11.5 mg/kg, BOIN design) and dose-expansion study of JSKN016 as monotherapy in patients (pts) with metastatic solid tumors conducted in China. Here, we report the preliminary antitumor activity in TNBC pts, the safety profile and PK characteristic of JSKN016. Results: As of data cutoff on Dec 23, 2024, 19 pts have been enrolled and been escalated to 8 mg/kg dose level, the maximal tolerated dose (MTD) was not reached. A total of 6 TNBC pts enrolled across 3 dose levels: 4 mg/kg Q3W (n = 3), 6 mg/kg Q3W (n = 1), 8 mg/kg Q3W (n = 2). The median age was 45.7 years (range 41-55), all pts were pretreated with standard of therapy (20% received ≥ 3 prior lines of systemic therapy). Among 5 efficacy-evaluable TNBC pts, the ORR was 80.0% with 4 partial response (PR) and 1 stable disease (SD) (showed 29.5% tumor shrinkage in target lesions from baseline at week 6). The PFS was not yet mature. There was one DLT occurred during dose-escalation, which was a grade 3 (G3) dermatitis acneiform at 8 mg/kg Q3W dose level. The most frequent treatment-related adverse events (TRAEs) were nausea, anemia, vomiting and oral mucositis. Notably, manageable hemotoxic was observed with ≥ G3 events only occurred in 2 (10.5%) pts, including 1 G3 neutropenia and 1 G3 lymphopenia. Oral mucositis occurred in 12 pts with G3 in 3 pts. No ≥ G3 oral mucositis was observed in breast cancer pts. No TEAEs led to treatment discontinuation or death. No interstitial lung disease (ILD) was observed. Following a single dose, Cmax of JSKN016 and free payload increased proportionally over a dose range of 0.5 to 6 mg/kg Q3W. The mean half-life of JSKN016 is approximately 3 days for 6 mg/kg. The exposure of released payload was significantly lower than JSKN016, demonstrating the stability of JSKN016 in circulation. Conclusions: JSKN016 demonstrated an excellent antitumor activity in heavily pretreated TNBC, with a manageable and predictable safety profile. Clinical trial information: NCT06592417 . Efficacy findings. Dose level 4 mg/kg(n=3) 6 mg/kg(n=1) 8 mg/kg(n=1) Total(n=5) ORR, n % 3 (100.0) 1 (100.0) 0 4 (80.0) CR, n (%) 0 0 0 0 PR, n (%) 3 (100.0) 1 (100.0) 0 4 (80.0) SD, n (%) 0 0 1 (100.0) 1 (20.0)

1028 Background: JSKN003 is a biparatopic HER2-targeting antibody-drug conjugate (ADC) conjugated to a topoisomerase I inhibitor (TOP1i) via a tetrapeptide linker, designed to enhance serum stability and anti-tumor activity. The efficacy and safety of JSKN003 in advanced ovarian cancer and other solid tumors have been highlighted in previous reports, and this analysis provides updated insights into its performance in HER2-positive breast cancer. Methods: JSKN003-101 is a dose-escalation and -expansion study in Australia, and JSKN003-102 is a phase I/II study in China, both involving patients with advanced solid tumors. A pooled analysis was performed to assess its efficacy and safety in HER2-positive advanced breast cancer. Results: As of November 29, 2024, the median follow-up duration was 3.52 months (range: 2.99-3.71). A total of 71 patients with HER2-positive breast cancer were enrolled, with the majority receiving 6.3 mg/kg or 8.4 mg/kg doses. The median age was 54 years (range: 32-79), with 78.9% ECOG 1. All patients had stage IV disease, with 76.1% having visceral metastases. All patients had prior anti-HER2 therapy, including 87.3% with prior ADCs or TKIs, and 56.3% having ≥3 prior lines. Among 62 evaluable patients, 56 were T-DXd naïve. In these 56 patients, the overall response rate (ORR) was 67.9% (95%CI: 54-79.7), and the disease control rate (DCR) was 94.6% (95%CI: 85.1-98.9). In the RP2D subgroup (6.3mg/kg, n = 30), the ORR was 70.0% (95%CI:50.6-85.3). Of 6 patients with prior T-DXd exposure, 1 achieved a partial response (PR), 3 had stable disease (SD), and tumor shrinkage was observed in 3. Both median progression-free survival (PFS) and median overall survival (OS) were immature. Treatment-related adverse events (TRAEs) ≥Grade 3 occurred in 11.3% of patients, and serious adverse events (SAEs) in 9.9%, with 2 drug-related. No TRAEs led to death or treatment discontinuation. The most common TRAEs (≥20%) included nausea, elevated liver enzymes, vomiting, decreased appetite, fatigue, diarrhea, and anemia. No≥Grade 3 neutropenia was observed. Grade ≥3 anemia and decreased platelet count were each reported in 1 patient (1.4%), both being Grade 3. Interstitial lung disease (ILD) occurred in three patients (4.2%), all Grade 1-2, with no Grade ≥3 events. Conclusions: JSKN003 demonstrated promising efficacy and manageable safety in heavily pretreated HER2-positive breast cancer, including T-DXd-experienced patients. The biparatopic HER2 antibody design likely enhanced its binding efficiency and contributed to the observed clinical benefit. These findings support the planned Phase 3 trial to further evaluate its therapeutic potential. Clinical trial information: NCT05494918 ; NCT05744427 .

Human epidermal growth factor receptor 2 (HER2) is highly expressed in various solid tumors, and its abnormal activation is closely associated with poor tumor prognosis, establishing it as a prominent target in contemporary research. The successful clinical treatment of multiple HER2-positive tumors with HER2 antibodies has prompted researchers to design chimeric antigen receptor T (CAR-T) cells targeting HER2 for solid tumor immunotherapy. To date, the development of CAR structures has progressed to the fifth generation, with most HER2-CAR-T cell structures being modified based on the second-generation CAR architecture. This review will delineate the structure and cytotoxic mechanism of HER2-CAR-T cells, elucidate the difficulties and optimization strategies for HER2-CAR-T cell therapy, and summarize recent clinical applications and advancements.

Abstract Many systemic therapeutic options for advanced metastatic HER2-positive breast cancer have been approved, including antibody-based therapies, chemotherapies, and tyrosine kinase inhibitors (TKIs). However, brain metastasis and acquired resistance are still major unmet medical needs for HER2-driven metastatic breast cancer and advanced solid cancers. VRN101099 is an orally available HER2 kinase inhibitor with excellent selectivity over other kinases, including EGFR. VRN101099 tightly occupies the ATP-binding pocket of HER2 forming a covalent bond with Cys805, resulting in high efficacy against various HER2 amplification and HER2 activating mutations. VRN101099 showed potent activity in HER2 antibody-resistant isoforms such as p95HER2 and exon16-deletion, lacking the binding epitope for HER2 antibodies including trastuzumab. It further demonstrated significant activity in T-DXd-acquired resistant cell lines (TR), including SK-BR-3 TR cell lines, which exhibits reduced HER2 expression and overexpression of BCRP (breast cancer resistance protein), suggesting that VRN101099 is not a substrate of BCRP. It also showed potency in N87-TR cell lines, which exhibits overexpression of AKR1C, an enzyme involved in the metabolism of chemical drugs, suggesting that VRN101099 can overcome the metabolism of chemical drugs such as DXd. VRN101099, a non-BCRP substrate, exhibits intracranial activity by efficiently crossing the blood-brain barrier(BBB). In intracranial xenograft models, VRN101099 demonstrated superior anti-tumor efficacy to tucatinib, neratinib, zongertinib, trastuzumab, and T-DXd. Key aspects including safety, intracranial activity, and potency against HER2 amplification, activating mutations, antibody-resistant isoforms, and T-DXd-acquired resistance in cancer cells will position VRN101099 as a valuable therapeutic option for various HER2-driven solid tumors, both as monotherapy and in combination with other treatments, including HER2 antibodies and antibody-drug conjugates (ADCs). The phase 1 dose-escalation study of VRN101099 in Korea and Australia has been approved and is evaluating its safety, tolerability, pharmacokinetics, and pharmacodynamics of VRN101099 to determine the recommended Phase 2 dose (RP2D). In conclusion, VRN101099 is a promising HER2-selective TKI with a favorable profile including potency, tolerability, and brain permeability. This identifies VRN101099 as a potential novel systemic treatment option for patients with HER2-driven cancers, particularly those with brain metastases or those who have progressed after or failed with T-DXd. Citation Format: Hong-Ryul Jung, Jihye Yoo, Yikyung Ko, Somi Lee, Yeonsil Kim, Byung-Woo Park, Donghyuk Seo, Hyerim Im, Ha Yeon Cho, Haelee Kim, DongGuk Shin, Hayeong Kim, Se-Hyuk Kim, Soochan Kim, Kyungah Seo, Daekwon Kim, Sunghwan Kim. VRN101099: A novel treatment option for HER2-driven cancer patients, overcoming T-DXd resistance and brain metastases [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_2):Abstract nr LB284.

Abstract The overexpression of HER3 in solid tumors is associated with poor prognosis. Therapies developed to target HER3-mediated signaling are often associated with resistance over time, highlighting the need to develop differentiated therapeutic strategies. To address the need for diversified targeted therapy options to treat HER3-positive solid tumors, we developed an antibody-drug conjugate, EO-1022, that selectively targets and eliminates HER3-expressing cancer cells. EO-1022 is comprised of the highly selective and clinically validated HER3 antibody, seribantumab, that is site-specifically glycan-conjugated with a cleavable valine-citrulline linker and monomethyl auristatin E (MMAE) payload to give a drug-to-antibody ratio (DAR) of 4. Methods: EO-1022 was evaluated in vitro and in vivo, with patritumab deruxtecan (patri-DXd) as a comparator. Binding to BT474 breast carcinoma cells (HER3 high; IHC 3+) was measured by flow cytometry. In vitro cytotoxicity was evaluated for EO-1022, isotype-MMAE and free MMAE as well as patri-DXd, isotype-DXd and free deruxtecan in BT474, SK-BR-3 breast adenocarcinoma (HER3 high; IHC 3+) and NCI-H446 lung carcinoma (HER3 low/absent; IHC 0-1+) cells. In vivo anti-tumor activity of EO-1022 vs. patri-DXd was assessed in a BT474 cell line-derived xenograft (CDX) model of HER3-expressing breast cancer. The stability of EO-1022 in human serum was evaluated using an LC-MS assay. Results: EO-1022 binding to cancer cells, endocytosis, MMAE release, and inhibition of proliferation were dependent on HER3 expression. In antibody internalization assays, seribantumab displayed comparable internalization capacity as patritumab. For both HER3 mAbs, internalization was greater when HER3 was high compared to HER3 low. In cytotoxicity assays, EO-1022 displayed HER3 expression level-dependent cell killing, whereas patri-DXd did not. Furthermore, EO-1022 outperformed patri-DXd in HER3 high cell lines based on in vitro cytotoxicity. In the BT474 CDX model (HER3 high), EO-1022 (5 mg/kg; DAR 4) induced a greater number of tumor regressions (7/9 mice, 78%) than patri-DXd (5 mg/kg; DAR 8; 1/9 mice, 11%). EO-1022 maintained a homogeneous DAR of 4 throughout a 14-day incubation in human serum. In contrast, patri-DXd and a proof-of-concept cysteine-conjugated seribantumab-vc-MMAE (Cancer Res. 2024, 84;6_Supp:3121) showed >50% decrease in the average DAR. Conclusions: EO-1022 demonstrated target-dependent in vitro cytotoxicity and in vivo anti-tumor activity in a HER3-expressing breast cancer CDX model. Additional results on the optimization and characterization of EO-1022 will be presented, including bystander effect assays and in vivo models of HER3-positive breast cancer and EGFR mutant NSCLC. Results from preclinical studies highlight the therapeutic potential of EO-1022 for patients living with HER3-expressing cancers. Citation Format: Thomas O'Hare, Jaya Srivastava, Jaclyn Cleveland, Valerie Malyvanh Jansen, David Dornan. Preclinical discovery and characterization of EO-1022, a site-specific glycan-conjugated anti-HER3 vc-MMAE ADC for treating solid tumors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_2):Abstract nr LB004.

Anti-HER2 antibodies are effective but often lead to resistance in patients with HER2+ breast cancer. Here, we report an epigenetic crosstalk with aberrant glycerophospholipid metabolism and inflammation as a key resistance mechanism of anti-HER2 therapies in HER2+ breast cancer. Histone reader ZMYND8 specifically confers resistance to cancer cells against trastuzumab and/or pertuzumab. Mechanistically, ZMYND8 enhances cPLA2α expression in resistant tumor cells through inducing c-Myc. cPLA2α inactivates phosphatidylcholine-specific phospholipase C to inhibit phosphatidylcholine breakdown into diacylglycerol, which diminishes protein kinase C activity leading to interleukin-27 secretion. Supplementation with interleukin-27 protein counteracts cPLA2α loss to reinforce trastuzumab resistance in HER2+ tumor cells and patient-derived organoids. Upregulation of ZMYND8, c-Myc, cPLA2α, and IL-27 is prevalent in HER2+ breast cancer patients following HER2-targeted therapies. Targeting c-Myc or cPLA2α effectively overcomes anti-HER2 therapy resistance in patient-derived xenografts. Collectively, this study uncovers a druggable signaling cascade that drives resistance to HER2-targeted therapies in HER2+ breast cancer.

Synergistic HER2-scFv mediated immunotherapy with targeted gemcitabine delivery harnessing biomimetic composite nanoparticles for enhanced metastatic tumor therapy

Abstract The PIK3CA gene is mutated in ∼40% of HR+/HER2- breast cancers. Alpelisib and inavolisib are currently FDA approved for treatment of PIK3CA-mutant breast cancers, although neither is selective for mutant PIK3CA. Inhibition of wild-type PI3K results in multiple adverse events, including hyperglycemia, which limit the dose and duration of treatment. Further, inhibition of PI3K/AKT relieves feedback suppression of receptor tyrosine kinase (RTK) expression and activity, thus limiting complete inhibition of PI3K/AKT signaling in cancer cells and resulting in adaptive drug resistance. Whether PI3K mutant-selective inhibitors can more potently block this compensatory adaptive resistance is unclear. In this study, we interrogated early adaptive resistance mechanisms to the PI3Kα mutant-selective inhibitor RLY-2608 in HR+ breast cancer cells with PIK3CA mutations. Phospho-RTK arrays of MCF7 and T47D cells treated with the IC75 of RLY-2608 for 0-48h revealed feedback activation of the ERBB family of RTKs (EGFR, HER2, HER3, HER4) at 24 h. These results were confirmed by immunoblot analysis with phospho-specific antibodies. After an initial inhibition of MAPK and PI3K/AKT signaling at 6 h, we found rebound activation of pERK1/2, pAKT1, and pS6 at 24 h, which was maintained for 48 h, suggesting a potential mechanism of adaptive resistance. To identify the ERBB RTKs causing this adaptation, we tested the TKIs erlotinib, neratinib, and tucatinib, and the HER3 antibodies patritumab and zenocutuzumab, each in combination with RLY-2608. RLY-2608 combined with the pan-HER TKI neratinib and the HER2-selective TKI tucatinib most effectively reduced pERK1/2, pAKT1 S473, and pS6 at 48 h, suggesting inhibition of HER2 could block this adaptive mechanism. In short and long-term proliferation assays, neratinib and tucatinib also synergized better with RLY-2608 compared to the other inhibitors. GSEA of RNA-seq data from MCF7 and T47D cells treated with RLY-2608 for 48 h revealed upregulation of pathways related to tyrosine kinase signaling, extracellular matrix, and redox signaling. FoxO signaling was one of the top upregulated KEGG and Reactome pathways, suggesting that activation of FoxO transcriptional activity may compensate for PI3K/AKT inhibition. The Hallmark Estrogen signaling pathway was also upregulated following RLY-2608 treatment; this was validated by immunoblot analysis, showing that ERα levels increased in T47D and MCF7 cells treated with RLY-2608 over a time course. Other upregulated Hallmark pathways in both cell lines included Myogenesis, Hypoxia, and KRAS signaling down. These findings suggest that, in addition to RTK activation, adaptive resistance to RLY-2608 may involve metabolic reprogramming, increased oxidative stress, and hypoxia signaling, highlighting the need for combination strategies to overcome resistance and enhance treatment efficacy. Citation Format: Fabiana Napolitano, Chang-Ching Lin, Khushi Ahuja, Dan Ye, María Rosario Chica-Parrado, Yasuaki Uemoto, Pamela Luna, Yuki Matsunaga, Saurabh Mendiratta, Nisha Unni, Jeon Lee, Ariella B. Hanker, Carlos L. Arteaga. Targeting mechanisms of adaptive resistance to the PI3Kαmutant selective inhibitor RLY-2608 in HR+/PIK3CA mutant breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 3003.