Institute of Cancer Research

The systemic treatment landscape for advanced and metastatic chondrosarcoma, a malignancy with limited responsiveness to conventional therapies, has always been notoriously challenging. While standard chemotherapy offers minimal benefits, certain subtypes, such as mesenchymal and dedifferentiated chondrosarcomas, have shown some response to systemic therapies initially developed for other sarcomas. Investigational strategies are focusing on molecular targets, including mutations in the isocitrate dehydrogenase gene (IDH), signaling pathways, such as hedgehog and death receptor 5 (DR5) and immune modulation. IDH mutations, notably found in conventional and dedifferentiated chondrosarcomas, have prompted the evaluation of IDH inhibitors, which have demonstrated promising efficacy in preclinical and early clinical trials, despite limited data in chondrosarcoma. Additionally, the hedgehog pathway, implicated in chondrosarcoma progression, has been targeted with inhibitors, although clinical translation has shown mixed results. Immunotherapy, including programmed cell death 1 (PD-1) checkpoint inhibitors and chimeric antigen receptor-T (CAR-T) cells, is also being investigated but faces challenges due to the immunosuppressive tumour microenvironment. Among new approaches, DR5 agonists such as INBRX-109 have shown single-agent efficacy, with minimal toxicity, opening possibilities for use in combination therapies to improve outcomes. Given the heterogenous and treatment-resistant nature of chondrosarcoma, we highlight the need for multi-omics and genetic profiling to guide personalized, combination therapies that target multiple carcinogenic pathways. The integration of multi-targeted approaches could enhance efficacy, address tumour heterogeneity, and overcome resistance, presenting a hopeful direction for systemic therapy in this challenging cancer. The investigation of combination regimens with IDH inhibitors, immunotherapy and DR5 agonists hold promise for transforming the management of advanced chondrosarcoma.

Metastatic melanoma remains a major clinical challenge. Large-scale genomic sequencing of melanoma has identified bona fide activating mutations in RAC1, which are associated with resistance to BRAF-targeting therapies. Targeting the RAC1-GTPase pathway, including the upstream activator PREX2 and the downstream effector PI3Kβ, could be a potential strategy for overcoming therapeutic resistance, limiting melanoma recurrence, and suppressing metastatic progression. Here, we used genetically engineered mouse models and patient-derived BRAFV600E-driven melanoma cell lines to dissect the role of PREX2 in melanomagenesis and response to therapy. While PREX2 was dispensable for the initiation and progression of melanoma, its loss conferred sensitivity to clinically relevant therapeutics targeting the MAPK pathway. Importantly, genetic and pharmacological targeting of PI3Kβ phenocopied PREX2 deficiency, sensitizing model systems to therapy. These data reveal a druggable PREX2/RAC1/PI3Kβ signaling axis in BRAF-mutant melanoma that could be exploited clinically.

Venous thromboembolism (VTE) is a rare event in children and does not usually trigger investigation for malignancy. We report the case of a previously healthy female teenager presenting with unilateral leg swelling. Colour-Doppler ultrasound confirmed deep vein thrombosis (DVT), and the thrombophilia workup was negative. Cross-sectional imaging identified multiple liver and lung lesions, diagnosed as epithelioid haemangioendothelioma (EHE) at biopsy. Lifelong anticoagulation was commenced; the patient was initially observed and then started on sirolimus at disease progression. We describe the first published case of EHE presenting with DVT in a child. Clinicians need to be mindful of the association between cancer and thrombosis, even in paediatric patients.

Background Common genetic variation at 11q23.1 is associated with colorectal cancer (CRC) risk, exerting local expression quantitative trait locus (cis-eQTL) effects on POU2AF2 , COLCA1 and POU2AF3 genes. However, complex linkage disequilibrium and correlated expression has hindered elucidation of the mechanisms by which genetic variants impart underlying CRC risk. Objective Undertake an interdisciplinary approach to understand how variation at 11q23.1 locus imparts CRC risk. Design We employ analysis of RNA sequencing, single-cell RNA sequencing, chromatin immunoprecipitation sequencing and single-cell ATAC sequencing data to identify, prioritise and characterise the genes that contribute to CRC risk. We further validate these findings using mouse models and demonstrate parallel effects in human colonic mucosa. Results We establish rs3087967 as a prime eQTL variant at 11q23.1, colocalising with CRC risk. Furthermore, rs3087967 influences expression of 21 distant genes, thereby acting as a trans-eQTL hub for a gene-set highly enriched for tuft cell markers. Epigenomic analysis implicates POU2AF2 as controlling the tuft cell-specific trans-genes, through POU2F3-correlated genomic regulation. Immunofluorescence confirms rs3087967 risk genotype (T) to be associated with a tuft cell deficit in the human colon. CRISPR-mediated deletion of the 11q23.1 risk locus genes in the mouse germline exacerbated the Apc Min/+ mouse phenotype on abrogation of Pou2af2 expression specifically. Conclusion We demonstrate that genotype at rs3087967 controls a portfolio of genes through misregulation of POU2AF2. POU2AF2 is the primary transcriptional activator of tuft cells with a tumour suppressive role in mouse models. We therefore implicate tuft cells as having a key tumour-protective role in the large bowel epithelium.

Protein–inhibitor crystal structures aid medicinal chemists in efficiently improving the potency and selectivity of small-molecule inhibitors. It is estimated that a quarter of lead molecules in drug discovery projects are halogenated. Protein–inhibitor crystal structures have shed light on the role of halogen atoms in ligand binding. They form halogen bonds with protein atoms and improve shape complementarity of inhibitors with protein binding sites. However, specific radiation damage (SRD) can cause cleavage of carbon–halogen (C– X ) bonds during X-ray diffraction data collection. This study shows significant C– X bond cleavage in protein–ligand structures of the therapeutic cancer targets B-cell lymphoma 6 (BCL6) and heat shock protein 72 (HSP72) complexed with halogenated ligands, which is dependent on the type of halogen and chemical structure of the ligand. The study found that metrics used to evaluate the fit of the ligand to the electron density deteriorated with increasing X-ray dose, and that SRD eliminated the anomalous signal from brominated ligands. A point of diminishing returns is identified, where collecting highly redundant data reduces the anomalous signal that may be used to identify binding sites of low-affinity ligands or for experimental phasing. Straightforward steps are proposed to mitigate the effects of C– X bond cleavage on structures of proteins bound to halogenated ligands and to improve the success of anomalous scattering experiments.

Background Prolactin, a hormone produced by the pituitary gland, regulates breast development and may contribute to breast cancer etiology. However, most epidemiologic studies of prolactin and breast cancer have been restricted to single, often small, study samples with limited exploration of effect modification. Methods The Biomarkers in Breast Cancer Risk Prediction consortium includes 8,279 postmenopausal women sampled from four prospective cohort studies, of whom 3,441 were diagnosed with invasive breast cancer after enrollment. Prolactin concentrations were measured for all study participants on plasma samples collected when all women were postmenopausal and before any breast cancer diagnosis using ELISA assays. Pooled, unconditional logistic regression models, adjusted for confounders, estimated odd ratios (OR) for associations of prolactin and postmenopausal breast cancer risk overall and stratified by breast cancer risk factors. Results Higher plasma prolactin concentrations were positively associated with postmenopausal breast cancer risk (> 13.2 ng/mL vs. < 7.9 ng/mL, OR: 1.20, 95% CI: 1.06, 1.36; P-trend < 0.001). Although associations did not appear to vary by time since blood draw or most breast cancer risk factors, associations were primarily observed in current users of postmenopausal hormones at blood draw (> 13.2 ng/mL vs. < 7.9 ng/mL, current users, OR: 1.58, 95% CI: 1.27, 1.96, P-trend < 0.001; non-current users, OR: 1.08, 95% CI: 0.93, 1.27, P-trend = 0.11; P-heterogeneity = 0.06). Conclusion Prolactin may be a risk factor for postmenopausal breast cancer, particularly in the context of postmenopausal hormone use. Investigations of prolactin interactions with other hormonal factors may further inform breast cancer etiology.

Objective This study defined a retrospective cohort of patients in England with primary advanced or recurrent (A/R) endometrial cancer (EC) who may have been eligible for clinical trials evaluating immune checkpoint inhibitors (ICIs) in the first-line (1L) setting within a real-world dataset, and described the characteristics, treatment patterns and outcomes within this cohort. Design This was a retrospective, population-based study. Setting Routine population-level data from the National Cancer Registration and Analysis Service in England were used. Patients diagnosed with A/R EC between 1 January 2013 and 31 December 2019 were included (follow-up until 23 August 2021). ICI-eligible patients who received any 1L therapy (defined as first systemic treatment for A/R EC with or without radiotherapy) and met key eligibility criteria for the RUBY trial ( NCT03981796 ; 1L cohort) were included. A subpopulation who solely received carboplatin–paclitaxel at 1L (carboplatin–paclitaxel subcohort) was identified. Methods Demographics, characteristics and therapy received were reported. Overall survival (OS), time to next treatment (TTNT) and time to treatment discontinuation (TTD) from 1L chemotherapy initiation were assessed using Kaplan-Meier methodology. Results Of 13 954 patients identified, 2376 ICI-eligible patients were included in the 1L cohort (median [range] age: 67.9 [26.7–94.0] years); 902 patients received solely carboplatin–paclitaxel at 1L. Demographics and disease characteristics were generally similar between cohorts. Median (95% CI) OS, TTNT and TTD from 1L chemotherapy were 27.2 (24.7, 30.2), 16.9 (15.8, 18.5) and 3.4 (3.4, 3.4) months, respectively, in the 1L cohort, and 17.2 (15.5, 19.0), 12.4 (11.6, 13.5) and 3.4 (3.4, 3.4) months, respectively, in the carboplatin–paclitaxel subcohort. Conclusion Long-term outcomes were poor for both cohorts, particularly the carboplatin–paclitaxel subcohort, where patients did not receive radiotherapy and had predominantly metastatic disease. This reflects the unmet need for more durable treatment options to prevent relapse and prolong survival in this patient population. This real-world study will help contextualise outcomes from ongoing phase III clinical trials investigating 1L ICI treatments.

Purpose: To evaluate safety, tolerability, and anti-tumor response of lete-cel, genetically modified autologous T-cells expressing a T-cell receptor specific for NY-ESO-1/LAGE-1a shared epitope, alone or in combination with pembrolizumab, in human leukocyte antigen HLA-A*02-positive (HLA-A*02:01-, HLA-A*02:05-, and/or HLA-A*02:06-) patients with New York esophageal squamous cell carcinoma 1 (NY-ESO-1)- and/or LAGE-1a-positive non-small cell lung cancer (NSCLC). Experimental design: Study 208749 was a single-arm study of lete-cel alone. Study 208471 was a multi-arm study of lete-cel alone or in combination with pembrolizumab in patients with advanced or recurrent NSCLC. Results: Over 2500 patients were screened for target expression. In the multi-arm study, 738 (45%) of 1638 tested patients were HLA-A*02-positive. NY-ESO-1 and LAGE-1a testing was positive in 12% (62/525) and 4% (15/348) of tested patients, respectively. Forty-one patients positive for HLA-A*02 and antigen expression were screened in the single-arm study. Overall, 43 patients underwent leukapheresis and 18 received lete-cel across studies. Lete-cel demonstrated a manageable safety profile. No fatal treatment-related serious adverse events (AEs) were reported in either study. Cytopenias and cytokine release syndrome were the most common treatment-emergent AEs. Combining pembrolizumab with lete-cel did not appear to increase toxicity over lete-cel alone. Limited anti-tumor activity was observed; one of 18 patients had a durable response persisting for 18 months. Pharmacokinetic data showed similar T-cell expansion in all patients. Conclusions: Extensive HLA-A*02 and antigen expression testing was performed to identify potential participants. Lete-cel was generally well tolerated and had no unexpected AEs. Anti-tumor activity was observed in a limited number of patients.

Bile acids are trans-genomic molecules arising from the concerted metabolism of the human host and the intestinal microbiota and are important for digestion, energy homeostasis and metabolic regulation. While diurnal variation has been demonstrated in the enterohepatic circulation and the gut microbiota, existing human data are poorly resolved, and the influence of the host circadian system has not been determined. Using entrained laboratory protocols, we demonstrate robust daily rhythms in the circulating bile acid pool in healthy male participants. We identify temporal relationships between bile acids and plasma lipids and show that these relationships are lost following sleep deprivation. We also highlight that bile acid rhythmicity is predominantly lost when environmental timing cues are held constant. Here we show that the environment is a stronger determinant of these temporal dynamics than the intrinsic circadian system of the host. This has significance for the intimate relationship between circadian timing and metabolism.

Background MEDI5395 is a recombinant attenuated Newcastle disease virus engineered to express a human granulocyte-macrophage colony-stimulating factor transgene. Preclinically, MEDI5395 demonstrated broad oncolytic activity, augmented by concomitant programmed cell death-1/programmed cell death ligand-1 (PD-L1) axis blockade. Durvalumab is an anti-PD-L1 immune checkpoint inhibitor approved for the treatment of various solid tumors. We describe the results of the first-in-human study combining intravenous MEDI5395 with durvalumab in patients with advanced solid tumors. Methods This phase I, open-label, multicenter, dose-escalation, dose-expansion study recruited adult patients with advanced solid tumors, who had relapsed or were refractory or intolerant to ≥1 prior line of standard treatment. MEDI5395 was administered intravenously as six doses over 15–18 days. The dose-escalation phase assessed four-dose levels (10 ⁸ , 10 ⁹ , 10 ¹⁰ , 10 ¹¹ focus forming units (FFU)) of MEDI5395, with sequential or delayed durvalumab. Durvalumab 1500 mg was administered intravenously every 4 weeks up to 2 years. The dose-expansion phase was not initiated. The primary objectives were to evaluate safety and tolerability, dose-limiting toxicities (DLTs) and the dose and schedule of MEDI5395 plus durvalumab administration. Secondary objectives included the assessment of the efficacy, pharmacokinetics, pharmacodynamics, and immunogenicity of MEDI5395. Results 39 patients were treated with MEDI5395; 36 patients also received durvalumab. All 39 patients experienced ≥1 treatment-emergent adverse event (TEAE), most commonly fatigue (61.5%), nausea (53.8%) and chills (51.3%). Grade 3–4 TEAEs occurred in 27 (69.2%) patients; these were deemed MEDI5395-related in 12 (30.8%) patients. Two patients experienced a DLT, and the maximum tolerated dose of MEDI5395 with sequential and delayed durvalumab at study termination was 10 ¹¹ and 10 ¹⁰ FFU, respectively. Four patients (10.3%) achieved a partial response (PR). Patients with PR or stable disease tended to have higher baseline PD-L1 and CD8+ levels in their tumor tissue. A tendency to dose-dependent pharmacokinetics of the viral genome was observed in whole blood and a tendency to dose-dependent viral shedding was observed in saliva and urine. Neutralizing antibodies were observed in all patients but did not appear to impact efficacy negatively. Conclusion This study demonstrates the feasibility, safety and preliminary efficacy of MEDI5395 with durvalumab in patients with advanced solid tumors. Trial registration number NCT03889275

Background Parameningeal (PM) site is an unfavorable characteristic in rhabdomyosarcoma (RMS). We described the treatment and outcome for patients with PM RMS and investigated the prognostic value of risk factors. We scored PM site by originating site and by highest risk extension. Methods Patients with PM RMS were treated within the European pediatric Soft tissue sarcoma Study Group (EpSSG) RMS 2005 study with risk‐adapted, multi‐modal treatment. Results Three‐hundred‐eighty‐one patients with PM RMS were included. Radiotherapy was administered in 359 patients (77 with surgery). After a median follow‐up of 75 months, 5‐year event‐free survival was 60% (95% confidence interval (CI) 55%–65%), 5‐year overall survival was 65% (95% CI 60%–70%). Conclusions The outcome for patients with PM RMS has not improved in comparison to previous historical studies, despite the more rigorous application of radiotherapy (94% of patients). Signs of meningeal involvement, PM site, and age at diagnosis remained prognostic risk factors. Trial Registration EudraCT number 2005‐000217‐35

Objective. Intensity modulated proton therapy (IMPT) is susceptible to uncertainties in patient setup and proton range. Robust optimization is employed in IMPT treatment planning to ensure sufficient coverage of the clinical target volume (CTV) in predefined scenarios, albeit at a price of increased planning times. We investigated a deep learning (DL) strategy for dose predictions in individual error scenarios in head and neck cancer IMPT treatment planning, enabling direct evaluation of plan robustness. The model is able to differentiate between scenarios by using embeddings of the scenario index. Approach. To accommodate resolution disparities in planning CT-scans and accommodate the setup error scenarios, we introduced scenario-specific isocentric distance maps as inputs to the DL models. For 392 H&N cancer patients, high-quality 9-scenario ground truth (GT) robust plans were generated with wish-list driven fully automated multi-criteria optimization. The scenario index is converted to one-hot-vector that is used to derive the scenarios embeddings through the training of the DL model, aiding the model to predict a scenario specific dose distribution. Main results. The model achieved within 1%-point of agreement with the GT the predicted V95% of the voxelwise minimum dose for CTV Low and CTV High for 96% and 75% respectively of the test patients. Considering all robustness scenarios, median differences were 0.035%-point for CTV High V95%, 0.11%-point for CTV Low V95%, 0.29 GyE for parotids Dmean, 0.7 GyE for submandibular glands Dmean and 0.9 GyE for oral cavity Dmean. Prediction of full 3D dose distributions for all scenarios took around 14 s. Significance. Predicting individual scenarios for robust proton therapy using DL dose prediction is feasible, enabling direct robustness evaluation of the predicted scenario doses.

Background Patients with PAX3/7‐FOXO1 fusion‐negative rhabdomyosarcomas (fnRMS) harbouring the rare L122R MYOD1 mutation have significantly poorer prognosis than other fnRMS. We undertook a detailed clinicopathological evaluation of a cohort of patients with MYOD1 mutat ed fnRMS in order to improve risk stratification and treatment options. Procedure Histological, mutational and clinical data from a cohort of patients with MYOD1 mutant RMS treated in Europe were analysed. Results Thirty‐two cases with mutant MYOD1 RMS were identified from patients enrolled in sequential European rhabdomyosarcoma clinical trials from 1992 to 2022 ( n = 22) and non‐trial cohorts ( n = 10). Thirty cases had the recurrent L122R missense mutation, one case harboured a K124E mutation and one case had a truncating mutation (S63X). Increased MyoD1 and reduced MYF4 immunostaining were consistent features of MYOD1 L122R ‐mutated RMS. Applying the risk stratification of the European paediatric Soft tissue sarcoma Study Group (E p SSG) RMS2005 trial, among 20 localised RMS cases that could be assigned a risk category, one was Very High Risk, 13 were High Risk and six were Standard Risk. Eight patients had distant metastases at diagnosis. Of the 25 patients with adequate clinical follow‐up data, 15/25 (60%) patients had an event at a median time of 9 months (12/15 included failure of local control) and 13/25 (52%) died of disease. Conclusion This MYOD1 mutant cohort demonstrates increased MYOD and reduced MYF4 immunostaining, high risk of local failure and poor survival in agreement with other studies. Increased treatment intensity and improved local control should be considered for these patients.

Objectives MR fingerprinting (MRF) has the potential to quantify treatment response. This study evaluated the repeatability of MRF-derived T 1 and T 2 relaxation times in bone metastasis, bone, and muscle in patients with metastatic prostate cancer. Materials and methods This prospective single-centre study included same-day repeated MRF acquisitions from 20 patients (August 2019–October 2020). Phantom and human data were acquired on a 1.5-T MR scanner using a research MRF sequence outputting T 1 and T 2 maps. Regions of interest (ROIs) across three tissue types (bone metastasis, bone, muscle) were drawn on two separate acquisitions. Repeatability of T 1 and T 2 was assessed using Bland-Altman plots, together with repeatability (r) and intraclass correlation (ICC) coefficients. Mean T 1 and T 2 were reported per tissue type. Results Twenty patients with metastatic prostate cancer (mean age, 70 years ± 8 (standard deviation)) were evaluated and bone metastasis ( n = 44), normal-appearing bone ( n = 14), and muscle ( n = 20) ROIs were delineated. Relative repeatability of T 1 measurements was 6.9% (bone metastasis), 32.6% (bone), 5.8% (muscle) and 21.8%, 32.2%, 16.1% for T 2 measurements. The ICC of T 1 was 0.97 (bone metastasis), 0.94 (bone), 0.96 (muscle); ICC of T 2 was 0.94 (bone metastasis), 0.94 (bone), 0.91 (muscle). T 1 values in bone metastasis were higher than in bone ( p < 0.001). T 2 values showed no difference between bone metastasis and bone ( p = 0.5), but could separate active versus treated metastasis ( p < 0.001). Conclusion MRF allows repeatable T 1 and T 2 measurements in bone metastasis, bone, and muscle in patients with primary prostate cancer. Such measurements may help quantify the treatment response of bone metastasis. Key Points Question MR fingerprinting has the potential to characterise bone metastasis and its response to treatment. Findings Repeatability of MRF-based T 1 measurements in bone metastasis and muscle was better than for T 2 . Clinical relevance MR fingerprinting allows repeatable T 1 and T 2 quantitative measurements in bone metastasis, bone, and muscle in patients with primary prostate cancer, which makes it potentially applicable for disease characterisation and assessment of treatment response.