The EACR’s ‘Highlights in Cancer Research’ is a regular summary of the most interesting and impactful recent papers in cancer research, curated by the Board of the European Association for Cancer Research (EACR).
The list below appears in no particular order, and the summary information has been provided by the authors unless otherwise indicated.
Explore the links or scroll down and use the ‘Next’ and ‘Previous’ buttons to navigate
-
Immune evasion through mitochondrial transfer in the tumour microenvironment
-
Concurrent SOS1 and MEK suppression inhibits signaling and growth of NF1-null melanoma
- Controlling intracellular protein delivery, tumor colonization and tissue distribution using flhDC in clinically relevant ΔsseJ Salmonella
- Chemotherapy induces myeloid-driven spatially confined T cell exhaustion in ovarian cancer
-
Survivin Promotes Stem Cell Competence for Skin Cancer Initiation
- Multiparameter imaging reveals clinically relevant cancer cell-stroma interaction dynamics in head and neck cancer
1Immune evasion through mitochondrial transfer in the tumour microenvironment
Ikeda, H. et al. Nature. 638: 225-236. (2025).
doi: 10.1038/s41586-024-08439-0.
Summary of the findings
Cancer cells in the tumor microenvironment (TME) evade immune attack through various mechanisms, including metabolic reprogramming and mitochondrial dysfunction in tumor-infiltrating lymphocytes (TILs). This study reveals a novel immune evasion mechanism where cancer cells transfer mitochondria with mutated mitochondrial DNA (mtDNA) to TILs, leading to metabolic abnormalities and immune dysfunction.
.
Clinical samples showed that TILs share mtDNA mutations with cancer cells, suggesting mitochondrial transfer. Using fluorescence-labeled mitochondria, we confirmed that mitochondria move from cancer cells to TILs via tunneling nanotubes (TNTs) and small extracellular vesicles (EVs). These transferred mitochondria resist mitophagy due to inhibitory molecules, resulting in homoplasmic replacement.
.
TILs that acquire mutated mitochondria exhibit increased reactive oxygen species (ROS) production, impaired ATP generation, senescence, and reduced memory formation. In mouse models, tumors with an mtDNA mutation transferred mitochondria to TILs, leading to immune dysfunction and resistance to PD-1 blockade therapy. Blocking mitochondrial transfer with an EV inhibitor (GW4869) partially restored TIL function and improved PD-1 blockade efficacy. Accordingly, the presence of mtDNA mutations in tumor tissues was a poor prognostic factor for immune checkpoint inhibitors in patients with melanoma or non-small-cell lung cancer. Particularly, durable response was impaired.
.
Future impact
This study highlights mitochondrial transfer (“swap”) as a critical immune evasion strategy, suggesting new therapeutic targets and biomarkers. Targeting EV-mediated mitochondrial transfer or enhancing mitophagy could restore T cell function and improve immunotherapy outcomes. The findings could lead to novel cancer treatments, particularly for patients resistant to immune checkpoint inhibitors combined with predictive biomarkers.
.
2Concurrent SOS1 and MEK suppression inhibits signaling and growth of NF1-null melanoma
Marasco, M. et al. Cell Rep Med. 5 (11): 101818. (2024).
doi: 10.1016/j.xcrm.2024.101818.
Summary of the findings
Loss of neurofibromin (NF1), a negative regulator of RAS signaling, occurs in ~20% of melanoma cases. Unlike BRAF V600E melanoma, no effective targeted therapies exist for these patients if they fail or cannot tolerate immunotherapy.
.
In this paper, Marasco et al. provide a detailed biochemical characterization of how NF1 loss affects oncogenic signaling in melanoma, revealing unique vulnerabilities that can be pharmacologically targeted. The increase in RAS signaling due to NF1 loss is associated not only with increased proliferation but also with compensatory mechanisms that reduce the ability of RAS to be reactivated by external stimuli – an example of a negative feedback loop.
.
Therefore, when both RAS signaling and the ability to reactivate RAS are inhibited, NF1-null melanoma cells experience greater growth inhibition and apoptosis than wild-type cells, as a lower drug concentration is sufficient to prevent RAS reactivation in this tumor type. This study shows that this dual inhibition can be achieved with a combination of avutometinib (a MEK inhibitor) and BI-3406 (a SOS1 inhibitor), which effectively suppresses NF1-null melanoma growth both in vitro and in vivo.
.
Future impact
NF1-null melanoma represents a major unmet clinical need. Despite its higher RAS activity, this tumor is not sensitive to MEK inhibition. Analysis of signaling dynamics revealed a specific dependency of NF1-null melanoma cells on the activity of SOS1 in mediating RAS reactivation following MEK inhibition. Concurrent inhibition of MEK and SOS1 effectively inhibits NF1-null melanoma cell growth and proliferation.
.
The findings presented in this paper provide a strong rationale for the clinical evaluation of the avutometinib + BI-3406 combination as a potential treatment strategy for this tumor type.
.
3Controlling intracellular protein delivery, tumor colonization and tissue distribution using flhDC in clinically relevant ΔsseJ Salmonella
Raman, V. et al. Molecular Therapy. 33 (2): 649-669. (2025).
doi: 10.1016/j.ymthe.2024.12.038.
Summary of the findings
There is no off-the-shelf approach to selectively treat cancer without harming healthy tissue. Existing therapies target cancer cells that overexpress a cell-surface receptor, which occur in a minority of patients. Each patient must also be screened for cancer cell receptor
overexpression in order to ensure therapeutic compatibility. Conversely, conventional
chemotherapies are broadly cytotoxic towards both cancer cells as well as healthy tissue.
Treatment with chemotherapy creates unwanted systemic toxicity that can cause permanent damage. We created a bacteria-based delivery system, BacID, to address these
delivery challenges associated with existing cancer therapies. The BacID delivery strain
colonizes and delivers therapies selectively inside tumor cells without affecting healthy
tissue. Intracellular delivery occurs in all solid tumors regardless of the genetic and receptor expression heterogeneity. This is because the BacID platform utilizes the Type 3 secretion systems of Salmonella to invade and deliver cytotoxic and immunotherapies inside cancer cells in a receptor/biomarker independent manner. The universal targeting and delivery mechanism of BacID makes the platform ideally suited for off-the-shelf treatment of genetically diverse cancers.
.
In this publication, we employed genetic engineering strategies that allowed BacID to deliver therapeutic protein inside both primary and metastatic breast cancer cells as well as
demonstrate efficacy against cancers with a high unmet need without causing any damage
to healthy tissue. One of these critical engineering steps enabled BacID to activate and
deliver therapies in tumors simply after ingestion of an over-the-counter dose of Aspirin. The Aspirin-controlled delivery mechanism was specifically implemented to enhance BacID
safety and efficacy in a manner that would be convenient in a clinical setting. Using this
system, we demonstrated that delivering pre-activated caspase 3 reduced the growth of
aggressive triple negative breast cancer in vivo. The therapy was also strongly effective
against other hard-to-treat cancers, like pancreatic and human liver cancer. These results
demonstrate the applicability of the BacID platform in treating hard-to-treat cancers without requiring any knowledge of the patient-specific tumor characteristics.
.
4Chemotherapy induces myeloid-driven spatially confined T cell exhaustion in ovarian cancer
Launonen, I.M. et al. Cancer Cell. 42 (12): p2045-2063.e10. (2024).
doi: 10.1016/j.ccell.2024.11.005.
Summary of the findings
High-grade serous ovarian cancer (HGSC), the most common and aggressive form of ovarian cancer, is marked by genomic instability and the development of resistance to treatments. Anti-tumor immunity plays a crucial role in therapy response and clinical outcomes, but how it is modulated by chemotherapy in HGSC remains poorly understood. To explore this, we used multi-omics profiling, including genomics, single-cell RNA sequencing (scRNA-seq), tissue imaging, and spatial transcriptomics on 117 patient samples containing over 25 million single cells. We discovered Myelonets—networks of connected immune cells—acting as previously unrecognized compartments that contribute to CD8+ T cell exhaustion following chemotherapy. Our data show that the M1/M2 polarization at the tumor-stroma interface is linked to T cell exhaustion and exclusion, which is associated with poor chemotherapy response. Additionally, we found that myeloid-T cell interactions through NECTIN2-TIGIT signaling, which is induced by chemotherapy, are crucial to this process. We successfully targeted these interactions using a patient-derived immuno-oncology platform, showing that high NECTIN2-TIGIT signaling in tumor samples can predict responses to immune checkpoint therapies. Our findings highlight the role of myeloid-driven, spatially confined T cell exhaustion, offering new targets for immunotherapy and strategies for patient stratification in ovarian cancer.
Future impact
We identified TIGIT-NECTIN2 interactions as a potential therapeutic target for overcoming T cell exhaustion particularly within myeloid-driven spatial hubs (Myelonets). We suggest that patients with limited chemoresponse may benefit from combination therapies targeting M1-driven T cell exhaustion and M2-driven inhibition of T cell interactions. These findings support testing anti-TIGIT and PD1 therapies post-chemotherapy and in specific patient subsets, offering insights into patient stratification to enhance CD8+ T cell anti-tumor immunity in HGSC. Importantly, detailed analysis of spatial tumor-immune dynamics in clinical samples can uncover new therapeutic targets, providing strategies for personalized treatments and ultimately improving patient outcomes in ovarian cancer.
.
5Ultrasensitive ctDNA detection for preoperative disease stratification in early-stage lung adenocarcinoma
Black, J.R.M., Bartha, G. et al. Nature Medicine. 31: 70-76. (2025).
doi: 10.1038/s41591-024-03216-y.
Summary of the findings
Tumour-node-metastasis (TNM) staging is an imperfect mechanism for stratifying patients with non-small cell lung cancer (NSCLC). Ultimately, this means that a subset of patients who will ultimately go onto relapse are not offered potentially curative adjuvant therapy, whilst unnecessary therapy is given to other patients who have in effect been cured by surgery. We used an ultra-sensitive circulating tumour DNA (ctDNA) detection assay to understand the degree to which pre-operative ctDNA detection might further stratify patients for clinical outcome. We studied ctDNA within 171 patients with early-stage NSCLC from the TRACERx study using a whole-genome sequencing-based tumour-informed assay that leveraged a median of 1,800 personalised variants per patient. This assay was capable of detections at between 1 and 3 parts per million (ppm) of ctDNA, and confirmed that pre-operative ctDNA detection in lung adenocarcinomas was strongly predictive of poor clinical outcome. Importantly, this effect was observed when analysis was restricted to tumours harbouring ctDNA at levels lower than could be reliably detected by the previous most sensitive assay studied in TRACERx. There, patients with ctDNA detected at below 80 ppm had worse progression-free and overall survival when compared to those in whom it was not detected.
.
.
These findings suggest ctDNA detection assays might have an important future role when applied more widely to stratify patients for escalation or de-escalation of post-operative treatment. There is a need for prospective studies to study whether this technology, integrated into routine clinical care, can improve outcomes for patients identified for escalation of treatment as a result of ctDNA detection. In addition, further studies, including work currently being done in TRACERx, should seek establish the utility of post-operative monitoring and peri-treatment ctDNA kinetics using ultra-senstive approaches in understanding clinical outcomes and the dynamics of relapse.
.
6Survivin Promotes Stem Cell Competence for Skin Cancer Initiation
Canato, S. et al. Cancer Discovery. 15(2): 427–443. (2025).
doi: 10.1158/2159-8290.CD-24-0263.
Summary of the findings
Basal cell carcinoma (BCC) is the most frequent cancer in humans. Stem cells (SCs) but not progenitors are competent to initiate BCC formation following oncogenic hedgehog signaling activation. However, the mechanisms conferring the competence to BCC formation in SCs and restricting tumour formation in progenitors were not known. This study uncovered the mechanisms allowing SCs and not progenitors to initiate BCC formation.
.
Canato and colleagues uncovered that Survivin, a protein regulating proliferation and apoptosis, is expressed at a higher level in SCs compared to progenitors. Using genetic gain and loss of function of Survivin in mouse models, the authors showed that Survivin deletion in oncogene-expressing SCs prevents BCC formation, whereas Survivin overexpression renders oncogene-expressing progenitors competent to BCC formation. Survivin expression enhances cell survival and renewing divisions, while restricting apoptosis and differentiation. The authors showed that pharmacological inhibition of Survivin or of SGK1, a downstream factor regulated by Survivin, prevents the conversion of preneoplastic lesions into invasive BCCs.
.
Altogether, this study identifies Survivin as a key licensing factor expressed by SCs that regulate BCC formation by promoting self-renewal and survival, while preventing cell death and differentiation in oncogene-expressing cells. In addition, the demonstration that Survivin expression is required for BCC formation suggests that Survivin or SGK1 inhibitors can be used to prevent the progression of preneoplastic lesions into invasive skin cancer.
7Multiparameter imaging reveals clinically relevant cancer cell-stroma interaction dynamics in head and neck cancer
Punovuori, K. et al. Cell. 25: p7267-7284.e20. (2024).
doi: 10.1016/j.cell.2024.09.046.
Summary of the findings
Cancer heterogeneity drives disease aggression and therapy resistance, but understanding its origins, mechanisms, and diagnostic potential remains challenging. Using an innovative strategy that combines high-resolution, high-throughput analyses of cell states, position and morphology (millions of cells with subcellular resolution) in clinical tumor biopsies, Punovuori et al established a quantitative image analyiss paradigm that revealed a highly aggressive disease subtype for human head and neck cancers that currently lack any clinical biomarkers. Partial epithelial-to-mesenchymal transition (pEMT) phenotypes have been identified in a number of tumor types, including HNSCC, and are associated with aggressive behavior and metastatic potential. This study reveals previously unreported subgroups of patients, with a particularly relevant poor survival patient group characterized by pEMT status of the tumor and a cancer associated fibroblast (CAF)-enriched fibrotic stroma. Further mechanistic studies using spatial transcriptomics on patient biopsies and patient-derived three-dimensional cancer and stromal cell co-cultures, Punovuori and colleagues identify highest signaling potential between these cell states through cancer-related extracellular matrix (onco-ECM) and EGF signaling pathways. Interestingly, the pEMT state of cancer cells was found to be a required for CAF-mediated reprogramming into an invasive phenotype through amphiregulin (AREG)/EGF signaling crosstalk.
.
This study highlights fibroblast-epithelial signaling as critical for HNSCC invasiveness. These findings concerning the mechanisms of cross-compartment crosstalk provide an important framework for better understanding of cancer heterogeneity. Further, the new computational tool for multiplex analyses of tissue biopsies might be applicable as a diagnostic tool for treatment selection in head and neck and other cancers.
8Development of patient-derived lymphomoids with preserved tumor architecture for lymphoma therapy screening
Santamaria-Martínez, A. et al. Nature Communications. 15: 10650. (2024).
doi: 10.1038/s41467-024-55098-w.
Summary of the findings
The advent of immunochemotherapy has significantly improved the survival rates of non-Hodgkin B-cell lymphoma patients, yet refractory disease and relapses persist, often resulting in poor prognoses. Developing a lymphoma model that can predict sensitivity to targeted therapies in human patients would be a valuable tool for personalized treatment decisions. However, modelling lymphoma ex vivo presents significant challenges, primarily due their inherent heterogeneous cellular composition.
.
In this manuscript, Santamaria-Martínez et al. generated an ex vivo culture system of lymphoma explants. First, they demonstrated using molecular, cytological, and high-plex spatial proteomic analyses that this ex vivo culture system can maintain the tissue architecture and cellular complexity of the original tumours. Next, the authors tested the response to therapies on 27 human primary tissue biopsies obtained from multiple lymphoma subtypes with a panel of drugs that included targeted therapies and immune modulators used in the clinic to treat lymphomas. Interestingly, using this approach not only is it possible to study the effects of anti-cancer therapies on the lymphoma cells, but also the contribution of the tumour microenvironment. Overall, lymphomoids can represent a valuable tool to predict drug sensitivity ex vivo. Such a system could potentially spare patients from unnecessary treatment toxicities and side-effects, and guide treatment choices in a personalized manner.
.
Future impact
The development of a tool for testing ex vivo drug sensitivity in lymphoma patients marks a significant leap toward personalized haemato-oncology. This innovation may enable tailored treatment plans by identifying the most effective drugs for individual patients, potentially improving response rates and reducing unnecessary toxicity. As a result, it could lead to more successful, personalized therapeutic strategies, enhancing patient outcomes and quality of life, particularly in cases of relapse and refractory disease. Additionally, it paves the way for studying the tumour immune microenvironment ex vivo, sparking a broader transformation in tumour modelling and fostering the era of tumour avatars.
.
9Blocking IL1RAP on cancer-associated fibroblasts in pancreatic ductal adenocarcinoma suppresses IL-1-induced neutrophil recruitment
Hansen, N. et al. JITC. 12: e009523. (2024).
doi: 10.1136/jitc-2024-009523.
Summary of the findings
Pancreatic ductal adenocarcinoma (PDAC) is an aggressive cancer characterized by an immune-suppressive tumor microenvironment. This study investigates the role of the IL-1 receptor accessory protein (IL1RAP), which is highly expressed on both tumor cells and stroma, and correlates with poor prognosis. Targeting IL1RAP with the antibody nadunolimab significantly inhibited tumor growth in xenograft models, with effects dependent on cancer-associated fibroblasts (CAFs). Specifically, inhibition of IL1RAP on CAFs reduced the secretion of critical chemokines preventing the migration of neutrophils and monocytes, immune cells that contribute to the tumor’s immune-suppressive environment. Notably, nadunolimab treatment not only suppressed neutrophil migration but significantly reduced their viability. Furthermore, nadunolimab enhanced monocyte-mediated antibody-dependent cellular cytotoxicity against PDAC cells. In a clinical trial, high IL1RAP expression was associated with longer progression-free survival in late-stage metastatic PDAC patients treated with nadunolimab, validating IL1RAP as a promising therapeutic target in PDAC. These findings underscore the pivotal role of IL1RAP on CAFs in immune suppression within PDAC.
.
Future impact
Targeting IL1RAP could disrupt CAF-mediated immune suppression, particularly the recruitment of neutrophils, offering a promising new therapeutic opportunity for PDAC patients.
.
10Estrogen-dependent activation of TRX2 reverses oxidative stress and metabolic dysfunction associated with steatotic disease
Smiriglia, A. et al. Cell Death & Disease. 16: 57. (2025).
doi: 10.1038/s41419-025-07331-7.
Summary of the findings
Metabolic-dysfunction-associated steatotic liver disease (MASLD) is a leading cause of chronic liver disease and hepatocellular carcinoma (HCC), often associated with metabolic dysfunctions like obesity and insulin resistance. The progression from MASLD to HCC involves chronic inflammation, oxidative stress, and altered growth factor signaling, with a notable male predominance. In this context, estrogens play a protective role by reducing oxidative stress and inflammation, key drivers of MASLD progression to HCC.
Smiriglia et al. demonstrated that estrogens improve redox balance, metabolic status, and mitochondrial function in preclinical liver steatosis models, including hepatocyte-like cells derived from human embryonic stem cells exposed to clinically relevant steatotic-inducing agents. Estrogen treatment reversed the steatotic phenotype, reducing lipid droplet accumulation and oxidative stress via upregulation of mitochondrial thioredoxin 2 (TRX2), an antioxidant player regulated by the estrogen receptor. TRX2 silencing abolished estrogen’s protective effects. Importantly, these findings were validated in a retrospective RNA sequencing analysis of MASLD patient cohorts, confirming the clinical relevance of the molecular mechanisms identified. This study highlights estrogen’s potential role in mitigating MASLD progression and its transition to HCC, providing new insights into sex-related differences in liver cancer risk.
.
Future impact
This study suggests that targeting TRX2 activation may provide a novel strategy to manage or even prevent MASLD progression to HCC. Therapeutic approaches that mimic estrogen’s activation of TRX2 could be especially beneficial for men and postmenopausal women, reducing the likelihood of cancer development. Furthermore, these findings open the door to personalized therapies that consider each patient’s hormonal and metabolic profile—such as gender and hormone levels—to optimize protection against liver cancer.
.
.
