AACR Annual Meeting 2025

Gene therapeutical intervention using viral vectors is becoming increasingly relevant for the treatment of various diseases, including cancer. A critical factor for a successful viral therapy is that the patient has not previously developed neutralizing antibodies to the respective viral serotype in the context of a naturally occurring anti-infection immune response. Such antibodies would impose the risk to neutralize and thus jeopardize the intended therapeutic response. This is especially relevant in therapies with adeno-associated virus (AAV), as immunity against AAV-serotypes is very common. We here report on the development and validation of a straightforward method to functionally detect neutralizing antibodies in primate sera, employing a transduction method on a human cell line using AAVs of different serotype coding for an easily detectable transgene.

The reported methodology will be helpful for the stratification of primates and patients exploiting their sera for the prevalence of neutralizing antibodies prior to the initiation of the AAV treatment. In addition, our assay has the potential to be applied to other viral treatment strategies also.

Often referred to as the “silent killer”, ovarian cancer poses significant challenges due to its asymptomatic nature and lack of reliable early detection methods, leading to late-stage diagnoses in over 70% of cases. As the eighth most common cancer among women worldwide, robust translational models for preclinical testing of ovarian cancer are essential. Unfortunately, the limited availability of orthotopic in vivo models presents a significant challenge in accurately replicating ovarian cancer for research purposes. By harnessing the ascitic nature of two well-established ovarian adenocarcinoma cell lines, we have successfully generated robust orthotopic in vivo models, which we can now offer as a valuable platform for preclinical research. Both our intra-peritoneal and intra-bursal models exhibit predictable growth patterns, thus contributing to the advancement of ovarian cancer treatments by enabling the exploration of tumor biology, drug resistance mechanisms, and therapeutic efficacy. The innovative models presented here are powerful tools for investigating tumor biology and advancing preclinical drug development, with the ultimate goal of improving therapeutic outcomes for patients with ovarian cancer.

The clinical potential of therapies on tumor metastases can be specifically tested on metastases that have naturally developed through the migration of the tumor cells from the primary tumor. Removing the primary tumor allows for a clearer assessment of a substance’s efficacy in targeting metastatic sites alone, free from the influence of the primary tumor. If the primary tumor is still present during treatment, it is difficult to determine whether the reduction in metastatic spread is due to the anti-metastatic effect of the treatment or merely to the reduced growth of the primary tumor, which could indirectly limit the progression of metastases. In addition, the rapid growth of the primary tumor reduces the therapeutic window. The intra-mammary implantation approach (also referred to by us as subQperior for non-breast tumor cells) allows for clean surgical removal of the primary tumor, as the tumor remains isolated without adhering to the skin or peritoneum. This uncomplicated removal minimizes local trauma and residual primary tumor and provides a consistent basis for studying metastatic development after tumor resection.

The study of metastasis in mouse models following primary tumor removal provides essential insights into the progression and treatment of metastatic disease. In this context, the luciferase-labelled 4T1 tumor cell line, known for its high metastatic potential, offers a robust model, particularly for investigating lung metastasis. After subQperior implantation of mice, we show that 4T1 cells even at small tumor sizes metastasize rapidly, predominantly to the lungs, making this model highly suitable for evaluating therapeutic strategies against metastatic spread.

Furthermore, experiments with LL-2 tumor cells have shown that subQperior injection and removal of the primary tumor provides significantly better metastasis compared to traditional subcutaneous application. The approach of subQperior application not only increases the homogeneity and growth of the tumor and simplifies the removal of the primary tumor, but also comes closer to an orthotopic implantation with regard to the natural tumor environment and the progression of tumor growth. The advantages of subQperior injection suggest it may be a preferable method for achieving more consistent and representative metastasis in preclinical studies.

Together, the methodological improvement of primary tumor removal and refined injection techniques enable a more precise evaluation of drugs targeting metastasis, avoiding confounding effects associated with the presence of the primary tumor. By isolating the impact of therapies on metastatic sites, researchers can more accurately assess the effectiveness of anti-metastatic agents, advancing our understanding and treatment of metastatic disease.

HER2-positive breast cancer is characterized by overexpression of the HER2 receptor, driving aggressive tumor growth. Antibody-drug conjugates (ADCs), like trastuzumab emtansine (Kadcyla, T-DM1) and trastuzumab deruxtecan (Enhertu, T-DXd), are effective in targeting HER2-positive tumors, though their efficacy across varying HER2 levels remains under study. This study compares Kadcyla and Enhertu regarding cell proliferation, ADC internalization, and bystander cytotoxicity in cancer cell lines with high (SK-BR-3/SKOV-3), moderate (JIMT-1), and low (MDA-MB-231/MDA-MB-435) HER2 expression. It also examines off-site ADC toxicity in primary non-tumor cells caused by off-target payload delivery.

The xCELLigence system, which works via label-free impedance measurements on living cells, was employed to monitor real-time cell proliferation and viability. Both, Kadcyla and Enhertu, eliminate tumor cells, effectively at nanomolar concentrations, correlating with HER2 expression levels. Interestingly, Kadcyla's effect appears at least 24 hours earlier than Enhertu, whereas the unconjugated antibody Trastuzumab itself has little to no effect on tumor cell growth. Low HER2-expressing cells were killed at 1 µM, the highest dose tested. ADC internalization was analyzed using pH-sensitive, fluorescently labeled Trastuzumab in flow cytometry and microscopy, assessing uptake by mean fluorescence intensity (MFI) in viable cells. This analysis allowed the comparison of the rate and extent of Trastuzumab uptake across the tested cell lines. A co-culture system evaluated bystander effects, with high HER2-expressing SKOV-3/SK-BR-3 cells co-cultured with luciferase-labeled HER2-negative LN-229 cells. After treatment with Kadcyla or Enhertu, luciferase assays were conducted to assess whether the cytotoxic payload released by the ADCs in HER2-positive cells could affect the viability of neighboring HER2-negative cells. Enhertu demonstrated a bystander effect, whereas Kadcyla, on the other hand, showed a more limited activity, reflecting the non-cleavable linker and the membrane-impermeable nature of its payload DM1.

Toxicity analyses of Kadcyla and Enhertu on non-tumor cells, and especially on bone marrow, complement on-target investigations by revealing critical results about their safety profiles. While these ADCs are designed to selectively target tumor cells, off-site toxicity can affect healthy tissues and impair immune system function. Studies on hematopoietic stem cells are essential to assess potential adverse effects on hematopoietic regeneration and differentiation, aiming to reduce unintended immunosuppression. These findings support the optimization of ADC design, to enhance both selectivity and therapeutic windows.

Inflammasomes have recently emerged as the exciting and promising drug targets for inflammatory disease therapy. The best characterized is the NLRP3 inflammasome, The NLRP3 inflammasome is associated with onset and progression of various diseases, including metabolic disorders, multiple sclerosis, inflammatory bowel disease, cryopyrin-associated periodic fever syndrome, as well as other auto-immune and auto-inflammatory diseases. The NLRP3 inflammasome is a cytosolic multiprotein complex composed of the innate immune receptor protein NLRP3, adapter protein ASC, and inflammatory protease caspase-1 that responds to microbial infection, endogenous danger signals, and environmental stimuli. The assembled NLRP3 inflammasome can activate the protease caspase-1 to induce gasdermin D-dependent pyroptosis and facilitate the release of IL-1β and IL-18, which contribute to innate immune defense and homeostatic maintenance. We have established cell-based Caspase-Glo 1Inflammasome Assay and NanoBRET target engagement intracellular NLRP3 assay for high throughput screening of inflammasome inhibitors. MCC950 is a specific NLRP3 inflammasome inhibitor. Here, we demonstrate that MCC950 inhibits LPS/Nigericin-induced caspase-1 activation and pyroptotic cell death in dose response mode in PMA primed THP-1 cells in Caspase-Glo 1 inflammasome and CellTiter-Glo viability assays. Our NanoBRET target engagement NLRP3 assay results show that MCC950 binds to cellular NLRP3 within 2 hours of incubation. Interestingly, NanoBRET NLRP3 residence time analysis in living cells indicates that a slow dissociation kinetics (long residence time) is observed for MCC950. Furthermore, Western blot analysis confirms that MCC950 blocks LPS/Nigericin-induced caspase-1 cleavage and activation in PMA primed THP-1 cells. Taken together, our results indicate Caspase-Glo 1 inflammasome and NanoBRET target engagement NLRP3 assay serve as great cell-based assay platforms to facilitate inflammasome drug discovery against inflammatory diseases.

Anticipation of the in vivo activity of an investigational new drug (IND) is challenging. In the context of the living body, compounds are subjected to degradation, metabolic modification or binding to plasma proteins – mechanisms that are well described to potentially compromise the in vivo activity seen in the patient compared to its predicted potential to inhibit its target in specific in vitro assays. As such assays are performed in artificial cell-culture media, we had previously modified our proprietary cellular FLT3 kinase phosphorylation assay to measure the activity of FLT3 inhibitors directly in and from human plasma, similar as described by Levis et al. 1. We hence established a method to analyze the activity of kinase inhibitors in human or rodent plasma (plasma inhibitory activity assay, PIA) in ELISA-based cellular kinase assays. We have successfully applied this methodology to determine the pharmaco-kinetics of a newly developed inhibitor of FLT3 in the context of a clinical trial targeting FLT3. We have now expanded this methodology for the detection of further kinase targets of potential clinical relevance. We were able to compile in vitro PIA assay data for further kinases such as VEGFR2 and BCR-Abl which will allow us to better predict the in vivo effect of specific inhibitors of these kinases and to determine the amount of freely available active test substance in the blood/sera of treated cancer patients.

1: Levis et al. Blood. 2006; 108 (10):3477-3483

The identification and quantification of biomarkers in clinical blood samples play a pivotal role in personalized medicine, enabling disease monitoring and therapeutic assessment. Protein phosphorylation, a key post-translational modification, reflects cellular signaling in immune or tumor cells. To exploit this, we developed a platform detecting phosphorylation states of key signaling proteins from clinical blood samples via flow cytometry. Our approach analyzes phosphorylated STAT5 (pSTAT5), ERK1/2 (pERK1/2), and p38 (p-p38), vital components in immune regulation and cancer progression.

A major challenge in the analysis of phosphorylated proteins is the rapid degradation or change in phosphorylation state ex vivo, which requires immediate stabilization of samples after collection. To solve this problem, we use Smart Tubes, which enable rapid fixation of clinical blood samples and preserve the phosphorylation state upon collection, ensuring signal integrity and minimizing variability. Sample collection with Smart Tubes requires no technical equipment and can be easily implemented and standardized at any clinical trial site.

Our test system employs flow cytometry, enabling the analysis of linage-specific markers (e.g., T/B cells, monocytes) and phospho-specific antibodies to quantitatively detect pSTAT5, pERK1/2, and p-p38 in peripheral blood mononuclear cell (PBMCs) subpopulations. Validation experiments with specific inhibitors demonstrate sensitivity and specificity in detecting these phosphorylated targets, even in heterogeneous cell populations within blood samples. The method allows for its application across a range of clinical conditions, from cancer immunotherapy monitoring to autoimmune disease characterization.

By integrating rapid sample fixation with advanced analytical techniques, this platform addresses key challenges in analyzing phosphorylated biomarkers in clinical samples. The use of flow cytometry as the central analytical tool enables single-cell resolution, allowing detailed investigation of individual cell populations within complex samples by differential marker expression. This precision makes it possible to differentiate and study specific subpopulations of immune cells or circulating tumor cells, within a heterogeneous sample. The ability to preserve and accurately measure phosphorylation states directly in patient-derived samples provides invaluable insights into real-time cellular signaling and facilitates the optimization of therapeutic strategies targeting dysregulated signaling pathways.

In summary, our system can detect phosphorylated proteins in clinical samples, with Smart Tubes ensuring sample stability and reliable biomarker analysis, advancing translational research and precision medicine.

Macrophages play a critical role in tumor development by shaping the tumor microenvironment. In particular, the M2-polarized macrophage phenotype is associated with pro-tumoral functions such as promoting angiogenesis, immune suppression, and tumor progression. Consequently, numerous therapeutic approaches aim to repolarize M2 macrophages to mitigate their tumor-supportive behavior. While these effects are consistently and effectively demonstrated in vitro in our laboratory, translating these findings into in vivo models remains a significant challenge.

Here, we present an in vivo co-injection model that allows the detailed study of macrophage-tumor cell interactions and the evaluation of potential therapeutic agents targeting macrophage polarization. First, we isolated monocytes from heathy donors and differentiated them in vitro into various phenotypes (M0, M1, and M2) according to our established protocols. We selected a tumor cell line that secretes high levels of macrophage survival promoting factors and has good in vivo growth characteristics. In this model we co-injected the tumor cells together with macrophages in NXG animals. Our results demonstrate that the co-injected M2 macrophages remains stable for over a week in the tumor microenvironment, closely mimicking the persistence observed in human tumors. Additionally, M0 macrophages undergo differentiation under the influence of the tumor microenvironment, reflecting the dynamic and plastic nature of macrophage polarization in vivo. Marker expression as well as functional properties, such as phagocytotic activity of pHrodo Zymosan particles ex vivo, was analyzed by flow cytometry.

This model provides a biologically relevant system for studying macrophage-targeting therapies, complementing the robust and reproducible in vitro experiments routinely conducted in our laboratory. The presence of the M0 and M2 macrophages over a significant period of time enhance the model’s utility for evaluating substances designed to repolarize macrophages in vivo.

In conclusion, this in vivo co-injection model represents a valuable tool for preclinical research into macrophage polarization and its impact on tumor progression. Its potential ability to replicate key aspects of macrophage-tumor cell interactions enables comprehensive investigations of therapeutic strategies targeting the tumor microenvironment. This approach has the potential to accelerate the development of novel treatments aimed at disrupting the pro-tumoral functions of M2 macrophages and enhancing anti-tumoral immune responses.

Protein kinases are essential for the regulation of many biological processes, e.g. proliferation, differentiation, migration, and apoptosis. Deregulated kinase activity is observed in tumor cells and altered activity of specific kinases is essential for development and progression of cancer and other diseases. Consequently, regulation of protein kinase activity became a target for therapeutic intervention. Intensive research resulted in the development of many small molecule kinase inhibitors (SMI) being clinically approved as targeted therapies. Functional biochemical in-vitro assays are essential requirements to assess SMIs effects on target kinase activity after validation of specific kinases being molecular drivers of a pathological condition. Only limited information is gained by monitoring the interaction of kinase and compound in pure binding assays, but more conclusive data can be generated by in-vitro kinase activity assays. However, the relevance of results from both approaches depends on how well the assay setup responds to the mode-of-action of the compounds. Recombinant proteins and generic substrates are well established in development of protein kinase assays due to cost and technical feasibility considerations. Especially some generic substrates are highly artificial e.g. the commonly used Tyrosine-kinase substrate Poly(Glu/Tyr)4:1 and may be considered as phosphate group acceptors only. However, such generic substrates have successfully been used for development of numerous clinically approved SMIs. Other kinases however are not compatible with generic substrates and can only be established using less artificial substrates or even require their respective physiological in-vivo substrate for in-vitro activity. These kinases include several members of the RAF and MAPK family. Functional in-vitro assay using a generic substrate are rarely replaced by physiological substrate due to more complex assay conditions and increased cost for drug screening.

We compared biochemical in-vitro kinase activity assays for the LIM kinases, LIMK1 and LIMK2, using either generic substrates or the published physiological substrates, CFL1 and CFL2. We further evaluated these results by comparing them in different assays, using radiometric and luminescent read-out technologies. Several published LIMK inhibitors displayed significant differences in the relative potency when tested with generic or physiological substrates.

Generic substrates for biochemical in-vitro kinase activity assays in preclinical drug development have been well established and successfully applied in the past. However, our data indicate, that the choice of a more physiological substrate should be considered carefully, as it might significantly increase the relevance and therefore value of the results obtained from early-stage compound screening.