Javascript is disabled

Parallel Speakers

Cancer initiation and growth

Kristian Pietras, Division of Translational Cancer Research, Department of Laboratory Medicine, Lund University Cancer Centre, Lund University, Lund, Sweden
E-mail: Kristian.Pietras@med.lu.se

Exploring the organizational principles of the tumor microenvironment to reveal functional heterogeneity
Recent efforts analyzing tumor composition at single-cell resolution suggest a more intricate cellular complexity than previously recognized, indicating the need to study the context of the malignant microenvironment with improved detail. Conceivably, cellular microniches form functional units influenced by the origin and context of each constituent cell type. Our overarching aim is to functionally define the cellular elements of a tumor through an innovative set of integrated basic, pre-clinical and translational studies. Through the use of spatially resolved high-content technologies, we are mapping cellular interplays in microniches in order to define organizational principles of the tumor microenvironment. Ultimately, we are aiming to couple functional properties to tissue architecture. As a case-in-point we have detailed signaling pathways involving specific cancer-associated fibroblast subtypes with implications for tumor progression and response to therapy, for which we are now developing precision targeting strategies using both small molecule inhibitors, antibodies and antibody-drug conjugates.

The stromal compartment in general, and CAFs in particular, comprise a hitherto underexploited source for factors harboring prognostic, predictive or therapeutic potential. Taken together, our current work defines a sustainable framework for exploring and implementing cellular microniches as drug targets and tools for precision cancer medicines.

Volkan Sayin, Sahlgrenska Center for Cancer Research, Inst. Clinical Sciences, University of Gothenburg, SE
https://www.gu.se/en/research/volkan-sayin

Aging, Metabolism and KRAS-driven Lung Cancer: Do Tumors in the Elderly Refuse Growing Up?
Lung cancer is primarily a disease of the elderly. Although aging and cancer share key molecular alterations – such as deregulated protein homeostasis and permissive chromatin states – studies using physiologically aged models of lung cancer are few. Here, we show that aging alters the progression of KRAS-driven non-small cell lung cancer (NSCLC), promoting metastasis while suppressing primary lung tumor growth. Primary lung tumor cultures derived from older mice demonstrated an increased metastatic phenotype. Unbiased multiomics identified ATF4, the main effector of the integrated stress response, as a driver of aging-induced lung cancer metastasis. Furthermore, we found that the age-associated increase in ATF4 fuels metastatic dissemination through metabolic rewiring, which constitutes druggable vulnerabilities. Clinically, a multicenter analysis of all consecutively diagnosed NSCLC cases in Western Sweden over a 3-year period confirmed increased metastasis and smaller primary tumor size with increasing age in KRAS-driven NSCLC. Our findings suggest a novel adjuvant therapy for human lung cancer by targeting aging-induced metabolic plasticity, highlighting the need to consider the biology of aging in the development of cancer therapy.

Sophia Harild, Umeå University, Department of Diagnostics and Intervention
https://www.umu.se/en/research/groups/sophia-harlid/

Environmental exposures during sensitive time periods
– a key to understanding breast cancer risk?
Windows of susceptibility are time-periods when we are especially vulnerable to environmental exposures. For women, these occur during fetal life, puberty, pregnancy, and menopause. One aim in my research, and the focus of this lecture, is identification of exposures occurring during these time periods, that are associated with later breast cancer risk. The talk touches on several ongoing projects: In the first I will describe work aiming to identify pregnancy exposures that are associated with premenopausal breast cancer. This will include new results from a nationwide study on air pollution exposure during pregnancy where we compared risk estimates depending on exposure timing. In the second project we are focusing on risk factors that may have a mediating role between exposure and cancer development including lactation-length and breast density. This is done in a new cohort (NorthMom) consisting of women that were part of an ongoing mother-child study in Västerbotten and therefore have provided detailed information on pregnancy exposures. As part of NorthMom we are also hoping to identify biomarkers that can inform us about previous exposures. Finally, I will briefly discuss future studies of exposures occurring during the perimenopausal window of susceptibility and later risk of postmenopausal breast cancer.

Research field: Molecular Epidemiology

Oral presentations selected from abstracts

Aleksandra Wielento, Karolinska Institutet
https://ki.se/en/people/aleksandra-wielento

The impact of bacterial metabolites on colon cancer progression
Background Colorectal cancer, a leading cause of cancer-related mortality worldwide is closely associated with the gut microbiota, which plays a critical role in the tumor microenvironment. Gut microbes can directly or indirectly affect host cells, and disruption of the normal microbial community leads to dysbiosis, which can impact cancer progression. Recent studies have identified specific microbiome profiles associated with colorectal cancer. Currently, studies are examining the impact of metabolites produced by the gut microbiome on the progression of colorectal cancer.
Methods To assess cell viability, cancer cell lines from colorectal cancer and several other cancer types as well as normal cells (immortalized skin fibroblasts and primary gingival fibroblasts) were infected or treated with 10% bacterial supernatants collected from several pathogens and their various strains or commensal bacteria. After 48h cell viability was measured using resazurin assay. Metabolic fractions were isolated by filtering bacterial supernatants using Amicon 3kDa cutoff filters.
Results Colon cancer primary and metastatic cell lines surprisingly did not exhibit altered viability when infected with pathogens that have been associated with colorectal cancer. However, treatment with a specific microbial supernatant (hereafter referred to as "bacteria X" or "Bx" led to a significant decrease in the viability of SW480 and SW620 cells. By testing over 20 cancer cell lines from various origins, we found that this phenomenon occurred specifically in colon and medulloblastoma cell lines. Importantly, normal cells remained unaffected. Furthermore, supernatants from different Bx strains demonstrated differential killing potential, while commensal bacteria had no effect. Finally, using supernatant fractionation, we identified the active component responsible for the tumor-killing potential as a metabolite(s).
Conclusions By discovering novel metabolic vulnerabilities, this study highlights the importance of microbial metabolites as a potential source of therapeutic anticancer agents and opens new perspectives in drug discovery.

Cedric Patthey, Umeå University
https://www.umu.se/personal/cedric-patthey/

A rare fibroblast subtype supresses pancreatic cancer progression through signalling to neutrophils and tumour cells
Within the stroma of pancreatic ductal adenocarcinoma (PDAC), mesenchymal cells differentiate into cancer-associated fibroblast (CAF) subtypes that differentially mediate disease progression. Defining the regulatory mechanism and diversity of CAF subtypes could identify potential therapeutic strategies to harness the tumour suppressive activities of CAFs. Using single cell RNA sequencing on purified mesenchymal cells, we mapped the CAF subtype diversity in PDAC and identify a rare interferon-responsive subtype that we name ifCAF. After induction of ifCAFs in vivo through treatment with an interferon signalling agonist in a mouse model of PDAC, we used single cell RNA-sequencing to characterize the response in cell types of the tumour microenvironment. Predictions of cell-to-cell interactions reveal CAF-to-neutrophil signalling that promotes a tumour suppressive phenotype in neutrophils. In addition, our results suggest ifCAF-derived signals reduce proliferation and invasive characteristics on tumour cells.
In vitro co-cultures of CAF together with neutrophils or tumour cells confirm direct CAF-to-neutrophil and CAF-to-tumour cell interactions.

Radiotherapy and Theranostics

Vladimir Tolmachev, Department of Immunology, Genetics and Pathology (IGP), Uppsala University
Email: vladimir.tolmachev@igp.uu.se

Radionuclide molecular targeting of cancer using engineered scaffold proteins
Targeted cancer treatment can extend patients survival, with outcomes often depending on the level of molecular targets expression. Radionuclide molecular imaging allows us to visualize these targets in malignant tumors, addressing their heterogeneity and changes over time.  Commonly, therapeutic antibodies are labeled with 89Zr for immunoPET visualization, but their bulkiness results in slow tumor localization and poor specificity, making imaging possible only 4 to 7 days post-injection.

We develop alternative imaging probes called engineered scaffold proteins (ESPs). ESP affibody molecules, ADAPTs, and DARPins for visualization of HER2 and EpCAM were evaluated in clinical trials.  Clinical trials have shown that these ESP are safe, well-tolerated, and deliver low absorbed doses. These agents offer better imaging sensitivity and specificity two to four hours post-injection compared to antibodies after 5 days. ESP-based probes visualizing EGFR, IGF-1R, HER3, CAIX, CEACAM5, VEGFR2, and B7-H3 are in preclinical development. In a quest for optimal labels, ESPs were labeled with 15 different nuclides for SPECT, and PET imaging. We found that a combination of a radionuclide and a chelator/linker for its coupling to ESP has a strong impact on the targeting properties of an ESP.

Preclinical studies demonstrate that ESP-based therapeutics can extend the survival of tumor-bearing mice by delivering cytotoxic nuclides. They can also deliver anti-cancer drugs or block mitotic signaling. Co-developing imaging companion diagnostics enables patient selection and treatment monitoring.

In conclusion, ESP-based targeting agents offer high affinity and specificity in vivo, are cheaper to produce than monoclonal antibodies, and allow radionuclide imaging within hours of injection. ESPs can be engineered for multispecific constructs with adjustable targeting properties, aiding the development of anti-tumor therapeutics with different mechanisms of action.

Thuy Tran, Karolinska Institutet & Head of Theranostics Trial Center, Karolinska University Hospital
TRANslational Theranostics Group – Thuy Tran | Karolinska Institutet

Advances of Theranostics: Shaping the future for molecular radiotherapy
The developments of radiopharmaceuticals, radioactive drugs, for targeted therapy and diagnostics, so-called theranostics. It refers to a cutting-edge approach, “see what you treat and treat what you see”, in cancer medicine that combines the therapeutic and diagnostic processes to tailor treatment to individual patients.

Theranostics have gained rapid momentum  in the recent years thanks to precise tumor targeting and reduced off-target toxicity. At Karolinska’s Theranostics Trial Center, our multidisciplinary team encompassing nuclear medicine, oncology, radiopharmacy, and dosimetry accelerates translation from bench to bedside. Ongoing early-phase clinical trials employ image-guided radiopharmaceuticals against advanced targets such as GRPR, HER2, PD-L1, and FAP, will be showcased. As we explore new targets and develop next-generation radiopharmaceuticals and combination therapy with synergistic mechanisms, theranostics holds the promise to redefine molecular radiotherapy and deliver transformative patient outcomes across multiple cancers.

Martin Nilsson, Skåne University Hospital & Lund University, Sweden
martin.p.nilsson@skane.se

Individualized radiotherapy for anal cancer patients - moving away from one size fits all
The standard treatment of anal cancer is radiotherapy combined with chemotherapy, which leads to 80 % chance of cure but > 30 % risk of severe toxicity. In contrast to more common types of cancer, the radiotherapy of anal cancer has been based on a ‘one size fits all’ concept. Given the rarity of the disease, randomized phase 3 trials are difficult to perform. As an alternative way of generating “sufficient” level of evidence, the Nordic anal cancer group (NOAC) appointed a working group to first review the literature and then publish Nordic consensus guidelines. These guidelines are the first to include a risk-adapted strategy. With the aim of decreasing toxicity, patients with a low risk of recurrence are treated with smaller radiation volumes. In contrast, patients with a high risk of recurrence are treated with larger volumes, which might increase their chance of cure. Made possible through the centralization of anal cancer treatment in Sweden, all Swedish patients are treated according to the new guidelines since September 2023. Data on treatment and recurrences are registered in the Swedish Anal Cancer Registry, which will offer a simple and pragmatic way of evaluating the effectiveness of the new guidelines.

Oral presentations selected from abstract

Carlos F. D. Rodrigues, Karolinska Institutet
https://ki.se/en/people/carlos-rodrigues

Lithium treatment protects microglia and newly generated neuronal populations in a mouse model of cranial radiotherapy
Radiotherapy is a key treatment for high-grade brain tumors, significantly improving survival rates, especially in pediatric patients. However, it leads to long-term complications, including cognitive deficits, in 50-96 % of patients. No treatments currently exist to prevent these deficits. Lithium (Li), known for treating bipolar affective disorder, has been shown to reduce radiation-induced cognitive impairments in rodents by protecting the neuronal stem and progenitor cells in the hippocampus from apoptosis and promoting their proliferation. This study aimed to explore further the mechanisms underlying the protective and regenerative effects of Li in the irradiated young brain. To this end, postnatal day (PD) 21 C57BL6/J mice were injected intraperitoneally with Li chloride (4 mmol/kg) and kept on a Li carbonate-containing diet for 4 weeks. Control animals were injected with saline and administered an equivalent control diet. On PD 25, the animals were administered a single-dose whole-brain radiation of 8 Gy and were subsequently sacrificed at different time points, spanning from 2 weeks to 1 year. Hippocampi were collected for single-cell RNA sequencing using a novel protocol to capture viable cells, including neurons, and for electrophysiology analysis. The results showed that radiation induced the expression of the senescence genes in hippocampal microglia (e.g., Cdkn1a, Ccl12), which Li prevented. Additionally, Li also prevented the radiation-induced loss of hippocampal gamma oscillations and protected the newly generated hippocampal neurons, leading to the development of new neuronal subpopulations that prevailed in the hippocampus long after irradiation. Finally, a subpopulation of pruning microglia was shown to play a key role in Li-driven neuronal changes. This study advances our understanding of the effects of Li in the irradiated brain and supports its potential as the first pharmacological treatment for radiation-induced late complications.

Neuroimmunooncology

Mattias Belting, Professor of Oncology, Consultant Neuro-Oncologist, Lund University
https://www.tumor-microenvironment.lu.se/tumor-microenvironment-0

Tumor associated foam cells in brain tumors offer new therapeutic avenues
Immune cell metabolic rewiring and adaptation are common phenomena in major human conditions, including neurodegeneration and atherosclerosis. Based on our findings, we introduce tumor-associated foam cells (TAFs), a previously uncharacterized immune cell entity of lipid droplet (LD)-loaded macrophages in GBM, an incurable and common tumor of the brain. Through extensive analyses of patient tumors, together with in vitro and in vivo investigations, we reveal that TAFs exhibit distinct pro-tumorigenic characteristics related to hypoxia, mesenchymal transition, angiogenesis, impaired phagocytosis, and their presence correlates with worse patient outcome. Our mechanistic investigations demonstrate that TAF formation is facilitated by lipid scavenging from extracellular vesicles released by GBM cells. Importantly, we demonstrate that targeting key enzymes involved in LD formation, such as DGAT1 or ACSL, effectively disrupts TAF functionality. Our studies establish TAFs as a prominent immune cell entity in GBM and provide valuable insights into their interplay within the microenvironment. Disrupting LD formation to target TAFs presents an interesting avenue for future therapeutic development in GBM.

Sven Nelander, Uppsala University
https://www.uu.se/en/department/immunology-genetics-and-pathology/research/neuro-oncology-and-neurodegeneration/sven-nelander  

Leveraging Tumor Cell Plasticity for the Treatment of Neural Cancers
Tumor cell plasticity—the dynamic capacity of cancer cells to transition between distinct phenotypic states—underlies resistance, invasion, and recurrence in neural cancers. Using single-cell profiling, computational modeling, and advanced optical CRISPR-based live imaging, we identify and therapeutically target state-specific vulnerabilities. We highlight how engineered interventions against invasive mesenchymal states in glioblastoma and stem-like progenitor states in diffuse midline glioma offer novel treatment strategies, complementing conventional therapy and enhancing precision medicine approaches.

Sébastien Talbot, Karolinska Institutet, Sweden
sebastien.talbot@ki.se

Pain neurons control cancer immunosurveillance
Under the mentorship of Dr Rejean Couture, Dr Talbot obtained is PhD in Physiology from the Université de Montreal (2012). His thesis focused on the role of CNS immune cells in priming diabetic pain neuropathy. Next, he joined Dr. Clifford Woolf’s lab at the Neurobiology center of Harvard Medical School to study the crosstalk between pain neurons and adaptive immune cells in the context of allergic inflammation.

Dr Talbot started is independent work at Université de Montreal (2017-2022) and has since then obtained a Tenured Associate Professorship at Queen’s University (Canada). He also holds the Canada Research Chair in Neuro-Immunology. He is also cross-appointed as a group leader at Karolinska Institutet (Sweden) where he holds a Wallenberg Academy Fellowship. His work is supported by a CRS, CFI, CRC, CIHR, NSERC, NIH (R01, co-I), and NRFR

His group combines OMICs technique with optogenetics, tissue clearance and imaging and chemogenetics. They aim at defining a framework of the neuro-immune interplay at the system level, to decipher how and which sub-population of sensory neurons controls innate and adaptive responses, and to develop new targeted therapies for resolution of chronic inflammatory diseases.

Specifically, his investigation focusses on the role of nociceptor neurons in the control of cancer immunosurveillance. Dr Talbot will present data unraveling the transcriptome of tumor-innervating neurons as well as their ability to regulate cDC1 tumor antigen trafficking. Dr Talbot hope that blocking tumor innervating neurons will safeguard the host immune system anti-tumor immunity.

Oral presentations selected from abstract

Karolina Smolag-Klosowska, Lund University
https://portal.research.lu.se/en/persons/karolina-smolag-klosowska

Interplay between radiation-induced astrocytes and macrophages: implication for glioblastoma recurrence
Background Glioblastoma multiforme (GBM) is the most common primary intracranial tumor in adults, characterized by a high recurrence rate despite aggressive standard-of-care. Resistance to radiotherapy (RT) is a major driver of GBM recurrence. Previously, we demonstrated that RT induces a reactive phenotype in astrocytes, contributing to therapeutic resistance. Given the success of immunotherapies in other cancers, there is growing interest in targeting brain macrophages, which include resident microglia and infiltrating monocyte-derived macrophages (MDMs). However, current strategies targeting these immune cells have been largely ineffective in glioma treatment.
Methods We used single-cell RNA sequencing to analyze immune cell populations in primary and post-RT recurrent GBM in the RCAS/tva glioma mouse model. Co-culture experiments with RT-treated astrocytes and MDMs were conducted to assess their impact on macrophage survival and polarization, which was further evaluated using flow cytometry. Additionally, bulk RNA sequencing was performed to identify gene expression changes in macrophages influenced by RT-treated astrocytes.
Results Recurrent tumors exhibited an increased MDM-to-microglia ratio post-RT. Co-culture experiments revealed that RT-treated astrocytes enhanced MDM survival and promoted M2-like polarization, marked by upregulated CD163 and CD206 expression. Single-cell RNA sequencing showed downregulation of inflammatory immune pathways in MDMs from recurrent tumors. Furthermore, reactive astrocytes altered macrophage lipid metabolism, driving foam cell differentiation and lipid storage. Notably, when we projected the macrophage signature onto our single-cell data, the profile closely correlated with the lipid-associated and hypoxic macrophage cluster.
Conclusions These findings indicate a critical role of RT-induced astrocyte reactivity in shaping the tumor microenvironment by supporting MDM survival and polarization into tumor-promoting, lipid-loaded macrophages. Understanding these interactions may reveal novel therapeutic strategies to target astrocyte-macrophage crosstalk and mitigate GBM recurrence.

Sebastian Braun, Lund University
https://portal.research.lu.se/en/persons/sebastian-braun

Pericytes orchestrate a tumor-suppressive microenvironment by impinging on the crosstalk between macrophages and tumor-initiating cells in glioblastoma
Glioblastoma (GBM) is characterized by fast progression, infiltrative growth pattern and a high rate of relapse. A defining feature of GBM is the existence of spatially and functionally distinct cellular niches, i.e. a hypoxic niche, a leading-edge niche, and a peri-vascular niche, in which malignant cells engage in paracrine crosstalk with cell types comprising the tumor microenvironment, including immune cells, astrocytes, and vascular cells. Here, by analysis of single-cell transcriptomic data of human GBM and transgenic mouse models of GBM, we unexpectedly identified pericytes intimately associated with the endothelium as the most active paracrine signaling hub within the tumor parenchyma. Exclusive signaling axes emanating from pericytes were received by endothelial cells, malignant cells, astrocytes, and immune cells. Depletion of pericytes through genetic engineering in several different transgenic and orthotopic mouse models of GBM demonstrated accelerated tumor progression, a disrupted blood-brain-barrier, and premature death of pericyte-poor mice. Mechanistic studies revealed that pericyte deficiency altered the cellular composition of GBM, remodeled the endothelium, and impacted on the immune cell landscape. Specifically, endothelial cells deprived of pericyte association upregulated Colony Stimulating Factor (CSF)-1, which in turn attracted peri-vascular tumor-associated macrophages polarized towards an immune-suppressive phenotype. In the absence of pericytes, the recruited macrophages expressed Hepatocyte Growth Factor (HGF), which reinforced activation of its receptor tyrosine kinase c-Met on a subset of GBM cells harboring an extreme mesenchymal-like state. Indeed, orthotopic implantation of isolated c-Met-expressing GBM cells corroborated their superior tumor-initiating capability and invasive phenotype, compared to cells negative for c-Met. Taken together, we infer that the pericyte represents a critical modulator of GBM development by orchestrating a tumor-suppressive microenvironment; our findings thus highlight the importance of pericyte preservation in the face of current and future GBM therapies.

Early detection and surveillance

Marlene Malmström

Benny Björkblom, Department of Chemistry, Umeå University, Sweden.
Email: benny.bjorkblom@umu.se

Exploring Metabolic Markers for Early Detection and classification of Brain Tumors
Gliomas are highly complex and metabolically active brain tumors that are often diagnosed late, adversely affecting patient prognosis and quality of life. Our research aims to identify early metabolic changes in blood that could serve as biomarkers for glioma detection before clinical symptoms appear.
In our research, we have metabolically characterized pre-diagnostic blood samples from glioma cases and healthy controls in the Northern Sweden Health and Disease Study (NSHDS) and the European Prospective Investigation into Cancer and Nutrition (EPIC) study. Our findings revealed significant metabolic alterations two to eight years before diagnosis, with a distinct panel of metabolites strongly associated with glioma development. These metabolic changes were consistent across biological sexes and enabled the differentiation of future glioma cases from controls. Specifically, we observed elevated levels of TCA-cycle-related metabolites, including fumarate, malate, lactate, N-lactoyl-amino acids, and pyruvate, highlighting increased energy turnover in early tumor development. At the time of surgery, amino acid metabolism was markedly altered, with high levels of amino acids and their catabolic byproducts, particularly branched-chain alpha-keto acids. These metabolic signatures were detectable in blood, underscoring their potential as non-invasive biomarkers for early glioma detection.
Our longitudinal analysis, incorporating blood samples from the same individuals collected both years before diagnosis and at surgery, identified both general and glioma subtype-specific metabolic changes. Also, here a subset of TCA-cycle-related metabolites demonstrated strong predictive potential for glioma presence both pre-diagnostically and at surgery. We believe that these findings may provide insights into the metabolic evolution of glioma, possibly paving the way for improved early detection strategies

Qiaoli Wang, Lund University & Karolinska Institutet
Email:qiaoli.wang@med.lu.se 
https://ki.se/personer/qiaoli-wang

Extracellular Vesicles (EVs)-Based Liquid Biopsy for Pancreatic Cancer Detection
Liquid biopsy using extracellular vesicles (EVs) has emerged as a promising non-invasive tool for cancer detection. EVs isolated from blood carries molecular cargo such as proteins reflective of their origins including cancer cells. Compared with traditional biomarkers such CA19-9 alone, EV-based assays could improve panel sensitivity and specificity, enabling earlier diagnosis of pancreatic cancer. This talk will focus on the EV-based based mucin expressions in plasma and its discriminative ability in early detection of pancreatic cancer. Dr. Wang’s lab is highlighted by the Integrative Pancreatic Cancer Research (iPanCare), with focus on developing strategies for precision prevention, early diagnosis, individual risk stratification, and personalized medicine approaches for pancreatic cancer by leveraging integrative approaches of genetic and molecular epidemiology, pharmaco-epidemiology, and machine learning using large-scale population-based cohorts, national registries, electronic healthcare records, and multi-center clinical studies. As an emerging research group, we actively seek collaborative partnerships with investigators whose expertise intersects with our core research foci in early detection and precision medicine of pancreatic cancer.

Oral presentations selected from abstract

Pamela Buchwald, Skåne University Hospital, Malmö & Lund university
https://portal.research.lu.se/en/persons/pamela-buchwald

Tumor deposit count is a prognostic factor in colorectal cancer a population-based cohort study challenging the present TNM classification
Objective This study aimed to investigate the prognostic effect of tumor deposit (TD) count in colorectal (CRC) patients.
Summary Background Data TDs are a prognostic factor for overall survival (OS) and distant metastasis (DM) in lymph node-negative CRC. The current TNM-staging system does not account for the presence of TDs in lymph node-positive CRC or the presence of multiple TDs.
Methods Patients who underwent curative surgery for CRC between 2016 and 2019 were identified from the Swedish Colorectal Cancer Registry. Patients with undisclosed TD status/count, stage IV disease, non-radical resection, or early (<90 days) study outcomes were excluded. Uni- and multivariable Cox regression analyses were used to assess the prognostic total effect of TD-count on OS, DM, and local recurrence (LR), adjusted for age, sex, neoadjuvant treatment, and number of positive lymph nodes.
Results Of 18,913 patients assessed, 14,154 patients were analyzed with TDs present in 1,702 (12%) patients. TD-count was divided into 0, 1, 2, 3, 4, and ≥5 TDs. Increased TD-count was associated with decreased OS (88%, 81%, 73%, 76%, 61%, 64%) and increased 3-year risk for DM (10%, 24%, 31%, 39%, 46%, 53%) and LR (2%, 4%, 7%, 10%, 7%, 9%), respectively. TD-count remained an independent prognostic factor for reduced OS and increased risk of DM and LR after multivariable adjustments.
Conclusion TD-count is a negative prognostic predictor of OS and DM in CRC, independent of the number of positive lymph nodes or neoadjuvant treatment. These findings suggest that TD-count should be integrated into the TNM staging regardless of lymph nodes status.

Anna Martling, Karolinska Institutet
https://ki.se/en/people/anna-martling

Low-Dose Aspirin Reduces Recurrence Rate in Colorectal Cancer Patients with PI3K Pathway Alterations: 3-Year Results from a Randomized Placebo-Controlled Trial
Background Colorectal cancer (CRC) affects 1.9 million people globally each year, with 20–40% of stage II–III patients developing metastases. Aspirin reduces adenoma and CRC incidence in high-risk individuals, and observational studies suggest it may improve disease-free survival (DFS). Retrospective findings indicate that somatic PIK3CA mutations predict aspirin response, but validation in randomized trials is needed.
Methods The ALASCCA trial was a randomized, double-blind, placebo-controlled study across 33 Nordic hospitals. Patients with stage I–III rectal or stage II–III colon cancer and somatic PI3K pathway alterations were randomized to 160 mg aspirin daily or placebo for three years.
To detect a hazard ratio (HR) of 0.36 for time to recurrence (TTR) with 80% power, 150 patients with PIK3CA mutations in exon 9/20 (Group A) were required per arm. Another 300 patients with other PI3K pathway alterations (Group B) were included for secondary analysis.
Results Of 3508 screened patients, 1103 (37%) had PI3K pathway alterations (Group A: 515 [17.3%], Group B: 588 [19.7%]). 626 patients were randomized.
At three-year follow-up, HRs for TTR comparing aspirin to placebo were 0.49 (95% CI: 0.24–0.98; p=0.044) in Group A and 0.42 (95% CI: 0.21–0.83; p=0.013) in Group B. For DFS, HRs were 0.61 (95% CI: 0.34–1.08; p=0.091) in Group A and 0.51 (95% CI: 0.29–0.88; p=0.017) in Group B. Four patients experienced severe aspirin-related adverse events.
Conclusions The primary endpoint was met. Adjuvant aspirin (160 mg daily for three years) reduced recurrence in CRC patients with PI3K pathway alterations, potentially impacting clinical practice for one-third of CRC patients.

Advances in immunotherapy

Judit Svensson-Arvelund, Linköping University, SE
judit.arvelund@liu.se

Oncolytic virus-based immunotherapy for lymphoma: harnessing the power of dendritic cells
Advanced indolent non-Hodgkin’s lymphomas are incurable with standard therapy and are poorly responsive to immune checkpoint blockade. Since anti-tumor immunity depends on T cells that recognize tumor-antigens, immunotherapies have focused on increasing T cells at the effector phase, e.g., by blocking inhibitory signals through checkpoint blockade. While these strategies have been successful for some patients, they have limitations, such as the requirement for pre-existing tumor-reactive T cells. An alternative is to prime T cells at the tumor site by in situ vaccination (ISV) and harnessing the potential of dendritic cells (DCs) to cross-present tumor-antigens and induce anti-tumor immunity. An ISV can be generated through oncolytic viruses, which in addition to actively lysing tumor cells, release tumor-antigens and promote pro-inflammatory signals to induce anti-tumor immunity. We demonstrate that amplifying DCs in the context of an oncolytic virus-based ISV promotes durable tumor regressions and induction of tumor-reactive T cells that persist long-term in a murine lymphoma model. Our ongoing research aims to define the mechanisms of response versus resistance, and to determine the impacts of tumor heterogeneity and the tissue-specific microenvironments on therapy response. Our goal is to develop a safe and effective therapy that can overcome the limitations of current therapies.

Research field: Tumor immunology and Immunotherapy

Kristina Lundberg, Lund University
https://portal.research.lu.se/en/persons/kristina-lundberg

Mapping the Tumor Immune Microenvironment in Bladder Cancer: A Spatially Resolved Single-Cell Atlas Unveils Immune Crosstalk and Novel Therapeutic Opportunities
Checkpoint inhibitors that enable T cells to eliminate tumor cells have revolutionized cancer treatment, demonstrating the vast potential of immune-modulatory therapies. However, only a minority of patients respond to current treatments, underscoring the urgent need for more effective strategies. Advancing immunomodulatory therapies requires a comprehensive understanding of the immune cells within tumors, including their functional roles and the key cell-cell interactions that shape the immune response. We have constructed a spatially resolved single-cell atlas of immune cells in bladder tumors. Based on single-cell transcriptomics, we generated high-resolution transcriptional profiles of tumor-infiltrating immune population, which served as the foundation for subsequent spatial transcriptomics. This approach enabled us to define and characterize the populations with spatial precision. We identified immune cell aggregates and mapped their constituent subsets, providing valuable insights into immune cell interactions within the tumor microenvironment. Importantly, the results led to the discovery of a novel, promising immune cell drug target, and based on this we are currently developing a novel therapeutic strategy. Notably, this target appears relevant not only in bladder cancer but also other cancer forms. We welcome collaborations, especially to validate the target in different model systems and explore its therapeutic potential across additional cancer types.

Filipe Pereira, Lund University
https://pereiralab.com/team/carlos-filipe-pereira/

In vivo dendritic cell reprogramming for cancer immunotherapy
Immunotherapy can lead to long-term survival for some cancer patients, yet generalized success has been hampered by insufficient antigen presentation and exclusion of immunogenic cells from the tumor microenvironment. Here, we developed an approach to reprogram tumor cells in vivo by adenoviral delivery of the transcription factors PU.1, IRF8, and BATF3, which enabled them to present antigens as type 1 conventional dendritic cells. Reprogrammed tumor cells remodeled their tumor microenvironment, recruited, and expanded polyclonal cytotoxic T cells; induced tumor regressions; and established long-term systemic immunity in multiple mouse melanoma models. In human tumor spheroids and xenografts, reprogramming to immunogenic dendritic-like cells progressed independently of immunosuppression, which usually limits immunotherapy. Our study paves the way for human clinical trials of in vivo immune cell reprogramming for cancer immunotherapy.

Oral presentations selected from abstract

Jonas Nilsson, Sahlgrenska Center for Cancer Research, University of Gothenburg
https://www.gu.se/en/about/find-staff/jonasnilsson2

The hunt for TIL2.0, and improved cell therapy for solid cancers
Immune checkpoint inhibitor (ICI) therapy has revolutionized treatment of metastatic solid cancers including melanoma, lung cancer and renal cancer, resulting in cures of a sizable fraction of patients. Unfortunately not all patients respond or derive long-term benefit of ICI therapy. Intensive research is ongoing to find the next therapy that can either improve response rate or work in second line. Adoptive cell transfer with tumor-infiltrating lymphocytes (TILs) has emerged as a potential salvage treatment following recurrence after ICI treatment. With a 50% response rate and about 25% sustainable response, it is currently the best treatment available for patients with ICI refractory melanoma. But there are problems. Resistance develops in most of patients, side effects are seen because the need of preconditioning chemotherapy and treatment with interleukin-2 (IL-2), and the treatment is time-consuming, laborious and require access to clean rooms. Here I will outline our work on developing manufacturing procedures for TILs in Gothenburg and Perth , Australia. I will also show how we use mouse models to optimize TIL therapy, making it more tumor-reactive and more durable. The modified cell product is called TIL2.0 and is currently undergoing development.

Erik Yngve, Uppsala University
https://www.uu.se/en/contact-and-organisation/staff?query=N18-997

Oncolytic adenovirus therapy for solid tumors through induction of xenogeneic rejection
Background There is a great need to develop immunotherapies with capacity to stimulate effective T cell responses against solid tumors, in which today’s immunotherapies are ineffective. Xenogeneic rejection occurs when organs are transplanted from animals to humans, mainly due to the presence of α-gal epitopes in the organ that is recognised by natural α-gal antibodies present in humans. Due to its strong activation of the human immune system, α-gal have been suggested for use in immunotherapy of cancer and have been tested clinically with promising results. Neutrophil activating protein is a virulence factor of the common gut bacteria Helicobacter pylori and has a unique capacity to act as immunomodulator by creating an environment rich in IL-12 that skew the immune response toward Th1/T cytotoxic phenotype. We present Adf35(OGN), an oncolytic adenovirus intended to specially kill and express the immune stimulatory transgenes α-gal and NAP in solid tumors.
Methods Adf35(OGN)'s mechanisms of action were studied in human pancreatic cancer cell line Panc01. Infected cells were co-cultured with human serum and immune cells to analyse various effects on the immune system. A pancreatic cancer model in α-gal knockout mouse was established to mimic the human situation and study the effect of intratumoral injections of Adf35(OGN).
Results We show that infected tumor cells express α-gal and NAP (Fig. 1A-B) and subsequently undergo immunogenic cell death (Fig. 1C). When these cells are co-cultured with human serum and immune cells, this result in tumor cell opsonization by IgG (Fig. 1D), activation of complement (Fig. 1E) and upregulation of various immune cell activation markers (Fig.1F). In vivo, Adf35(OGN) inhibited tumor growth and improved mouse survival (Fig. 1G-H).
Conclusion Through its ability to infect tumor cells and express immune stimulatory molecules that activate a broad immune response, Adf35(OGN) shows promising potential as treatment for solid tumors.

AI, Data integration and multiomics

Kristina Lång, Lund University
https://portal.research.lu.se/en/persons/kristina-l%C3%A5ng

AI in mammography screening - unlocking the potential and navigating challenges
The use of artificial intelligence (AI) in mammography screening is rapidly evolving, with growing evidence supporting its potential to enhance accuracy and efficiency. Our team conducted the first randomised controlled trial, the Mammography Screening with Artificial Intelligence trial (MASAI), so far demonstrating clinical safety and improved cancer detection. Findings from the MASAI trial, together with other studies, are paving the way for its real-world implementation. As AI begins to be integrated into clinical practice, several challenges remain, including regulatory approval, integration into existing workflows, and trust among radiologists and policymakers. Drawing from my experience in the field, I will discuss the lessons learned, as well as the remaining barriers that must be addressed for widespread adoption.

Linda Bojmar, Linköping University
https://liu.se/en/research/bojmar-lab

Hepatic pre-metastatic niche in pancreatic cancer - Systemic effects of cancer as biomarkers for metastasis
Cancer is a systemic disease that distorts the homeostasis of the host, both locally in the microenvironment of the primary tumor and distantly within various organ-systems. For example, our group has recently shown that in patients with localized pancreatic cancer, the liver contains microscopic alterations that associate with future metastatic outcome. One way that such distant effects in the body can be mediated is through extracellular vesicles (EVs) and their bioactive cargo, which may originate from the primary tumor, or from other systemically affected organs. Besides their mechanistic significance through functionally modulating distant cells and organs, EVs also show great potential as biomarkers of cancer and its possible dissemination. As such, liquid biopsies obtained from plasma, or other bodily fluids, can be used for EV isolation and characterization of components such as proteins, RNA, DNA, lipids and metabolites. The goal of our team is to understand more about the systemic processes during cancer progression and to use this knowledge to tailor care in order to improve oncologic outcomes among patients with potentially metastasizing tumors.

Tuuli Lappalainen, PhD, Science for Life Laboratory, Department of Gene Technology, KTH Royal Institute of Technology, Stockholm, Sweden
New York Genome Center, New York, USA
https://tllab.org

Functional variation in the human genome: lessons from the transcriptome
Detailed characterization of molecular and cellular effects of genetic variants is essential for understanding biological processes that underlie genetic associations to disease. A particularly scalable approach has been linking genetic variants to effects in the transcriptome, which is amenable for scalable measurements in human populations and in model systems, including at the single cell level. Here, I will describe recent advances in our long-term work to characterize genetic associations to the transcriptome and other molecular traits, as well as our recent work on CRISPR-based perturbation of gene expression levels in cellular models. Altogether, integrating insights from these diverse approaches uncovers functional genetic architecture of human traits and the molecular and cellular mechanisms that mediate these effects.

Oral presentations selected from abstract

Iulian Emil Tampu, Linköping University
https://liu.se/en/employee/iulta54

Pediatric brain tumor classification using histopathology images and AI
Introduction Brain tumors are among the most prevalent solid tumors in children and young adults; however, the scarcity of histopathology datasets has hindered the implementation of computational methods in this population. AI has the potential to significantly improve diagnostic accuracy and consistency by enabling large-scale analysis of images and data. This study aimed to evaluate AI-based methods for analyzing pediatric brain tumors using whole slide images (WSIs) with hematoxylin and eosin (H&E) staining.
Material and methods A multi-institutional cohort was digitalized and curated in collaboration with Barntumörbanken at Karolinska, representing primary, metastatic, and recurrent pediatric tumors from six Swedish hospitals. WSIs were analyzed using weakly supervised multiple-instance learning (MIL) and state-of-the-art histology-specific foundation models for the hierarchical classification of pediatric brain tumors. Model generalization was evaluated by training and testing on data from the different hospitals. Balanced accuracy (BA) and attention maps were used to summarize model performance.
Results The best performance had a BA of 0.77±0.08, 0.62±0.05, and 0.48±0.05 for tumor category, family, and type classification, respectively. Generalization experiments revealed that model performance differed between test hospitals and training hospitals. The performance (BA) variations ranged from an increase of 0.01 for tumor category classification to a decrease of 0.17 for tumor type classification.
Conclusion In conclusion, this study demonstrates the ability of MIL and histology-specific foundation models to diagnose pediatric brain tumors in a multicenter Swedish cohort and their flexibility for new tasks and datasets.

Pontus Eriksson, Lund University
https://portal.research.lu.se/en/persons/pontus-eriksson

Real-World Application of the Lund Molecular Taxonomy for Urothelial Cancer.
Background The Lund Taxonomy for urothelial cancer has been established over the years. It includes five major molecular subtypes, Urothelial-like, Genomically Unstable, Basal/Squamous-like, Mesenchymal-like, and Small Cell/Neuroendocrine-like, with established clinical values. The molecular subtypes are defined at the cancer-cell level and do not include signals from infiltrating non-cancer cells as classifying features.
Methods/Results To translate this system into a clinical routine based on a single sample classification principle and real-time classification we established a pipeline that includes four different departments: Dept of Urology (Malmö), Dept of Pathology (Lund), Center for Translational Genomics (Lund) and the Dept of Oncology (Lund/Malmö). In data for 645 consecutive urothelial carcinomas, we show that real-time classification is feasible with a classification success of 90% on a routine basis. Classifications were validated by 14 gene expression signatures and by 16 protein markers by IHC. We develop algorithms for molecular grading based on both WHO1999 and WHO 2022 systems making tumor grading objective and not observer dependent. A gene expression signature strongly correlating with standard KI67 and CNNB1 IHC based assessments is derived making proliferation estimates standardized and independent on single markers. As the taxonomy is directed towards cancer cell phenotypes and does not consider the presence of infiltrating non-tumor cells as a classification feature, we derived/added four gene expression signatures for the stromal compartment and 14 for the immune compartment to the system. This makes it possible to treat cancer cell phenotypes and features of the microenvironment as separate variables. We show that four of these variables are strongly associated with tumor progression.
Conclusions We aim to fully classify a given tumor sample on 23 different features, molecular subtype, molecular grade (WHO1999 and WHO2022), proliferation, progression risk, stromal infiltration (4 signatures) and immune infiltration (14 signatures), based on one single RNA-seq analysis.