Background Clear cell renal cell carcinoma (ccRCC) is a highly heterogeneous cancer requiring a large number of biopsies to correctly characterize the tumor. Multiple biopsies are rarely feasible in the clinical setting, and therefore imaging methods offer the potential to evaluate the whole tumor non-invasively. For example, metabolic imaging offers the potential to probe the altered metabolism and metabolic heterogeneity that is characteristic of ccRCC. In this study we have explored and validated the use of hyperpolarized carbon-13 MRI (HP- 13 C-MRI) as a non-invasive clinical tool to probe metabolic heterogeneity in ccRCC patients and to more accurately identify which metabolic pathways are altered in vivo . Methods 58 tumor and healthy tissues biopsies were acquired postoperatively from 6 ccRCC patients imaged following injection of hyperpolarized [1- 13 C]pyruvate. MRI parameters were correlated with the metabolomic (146 metabolites) and transcriptomic (2523 metabolic genes) data obtained from these biopsies, split across 34 metabolic pathways. The results were used to generate metabologram projections as a visual representation of these metabolic differences. For each metabolic pathway, we generated a novel metabolic consensus scoring system for the identification of key altered metabolic pathways in ccRCC and their relationship to the imaging parameters. Results We show that the apparent exchange constant between pyruvate and lactate ( k PL ) and the lactate to pyruvate ratio (LP r ) on MRI can be used to measure differential metabolic pathways: they correlated positively with glycolysis and the pentose phosphate pathway, negatively with the TCA cycle, while also correlating with other pathways. Dichotomizing the imaged signal based on high and low k PL measurements was sufficient to discriminate metabolic distinct regions on biopsy and this could be a simple tool to assess metabolism clinically. Furthermore, metabolic heterogeneity increased in regions with a higher k PL and could be used to assess metabolic divergence. Conclusion This work validated the role of HP- 13 C-MRI to measure not only glycolysis, but also a range of other altered metabolic pathways in ccRCC. This could improve tumor stratification and provide novel methods to monitor treatment response. Metabolic imaging can also be used to guide biopsy acquisition based on metabolic alterations, and therefore could improve tumour characterization.

OBJECTIVES:The aim of the study was to translate abdominal deuterium metabolic imaging (DMI) to clinical field strength by optimizing the radiofrequency coil setup, the administered dose of deuterium ( 2 H)-labeled glucose, and the data processing pipeline for quantitative characterization of DMI signals over time. This was assessed in the kidney and liver to establish a basis for routine clinical studies in the future. MATERIALS AND METHODS:5 healthy volunteers were recruited and imaged on 2 or 3 separate occasions, with varying doses of 2 H-glucose: 0.75 g/kg (high dose), 0.50 g/kg (medium dose), and 0.25 g/kg (low dose), resulting in a total of 13 DMI scan sessions. DMI was performed at 3 T using a flexible 20 × 30 cm 2 2 H-tuned transmit-receive surface coil. For quantitative comparisons across scans, the 2 H-glucose signal was normalized against the sum of 2 H-glucose and 2 H-water (GGW ratio). To quantify the time course of GGW, 3 novel metrics of metabolism were defined and compared between doses and organs: the maximum value across the time course (GGW max ), the sum over the whole time course (GGW AUC ), and the average signal across a defined plateau (GGW mean plateau ). The 2 H-lipid signal overlaps with 2 H-lactate; hence, the 2 signals were measured as the combined 2 H-lipid+lactate signal. RESULTS:The careful positioning of a dedicated surface coil minimized unwanted gastric signals while maintaining excellent hepatic and renal measurements. The time courses derived from the liver and kidney were reproducible and comparable across different doses, showing the potential for dose reduction. The signal from the liver plateaued at approximately 30 minutes, and that from the kidney at approximately 40 minutes. The liver exhibited higher quantitative values for 2 H-glucose uptake compared to the kidney, a trend consistent across all 3 quantitative metrics and doses, for example, for the highest dose: GGW AUC liver  = 31 ± 3; GGW AUC kidney  = 27 ± 3; P  = 0.05. A trend toward lower quantitative measurements with decreasing dose was observed: this was significant between the high and the low dose for all 3 parameters and between the medium and low dose for GGW mean plateau and GGW AUC , but was not significant between the high and the medium dose for any of the 3 parameters. The hepatic 2 H-lipid+lactate signal increased over 70-90 minutes in 12/13 cases (mean: 39 ± 24%), while the renal lipid+lactate signal increased in only 8/13 cases (mean: 5 ± 17%). The hepatic 2 H-water signal increased in all 13 cases (mean: 18 ± 10%), and the renal 2 H-water signal increased in only 10/13 cases (mean: 10 ± 13%). CONCLUSIONS:DMI of the human abdomen is feasible using a clinical magnetic resonance imaging system and the signal changes measured in the kidney and liver can serve as a reference for future clinical studies. The 2 H-glucose dose can be reduced from 0.75 to 0.50 g/kg to minimize gastric signal without substantially affecting the reliability of organ quantification. The increase in 2 H-lipid+lactate or 2 H-water signal over time could serve as direct and indirect measures of metabolism, respectively.

Hyperpolarized 13C Magnetic Resonance Imaging (HP-MRI) enables real-time, non-invasive assessment of metabolism in diseases including cancer and neurodegeneration. Broader adoption has been limited by the complexity, duration, and lack of standardization of current hyperpolarization methods. This study evaluated POLARIS Preclinical, a parahydrogen-induced polarization (PHIP) hyperpolarizer designed to streamline production of hyperpolarized 13C agents. Four POLARIS systems were deployed across eight international research centers to produce hyperpolarized [1-13C]pyruvate doses within 90 seconds. In vitro and in vivo imaging was conducted in multiple animal models using MRI systems operating at 1.4, 3, 7, and 9.4 Tesla. Metabolic conversion of pyruvate to lactate and bicarbonate was successfully measured across all sites. POLARIS enabled rapid, reproducible production of hyperpolarized [1-13C]pyruvate and demonstrated consistent performance across instruments, institutions, and field strengths. These results support standardized, high-throughput metabolic MRI for multicenter studies and translational research in oncology, neurology, and cardiovascular disease.

Fumarate hydratase-deficient renal cell carcinoma (FHd-RCC) is a rare and aggressive renal cancer subtype characterised by increased fumarate accumulation and upregulated lactate production. Renal tumours demonstrate significant intratumoral metabolic heterogeneity, which may contribute to treatment failure. Emerging non-invasive metabolic imaging techniques have clinical potential to more accurately phenotype tumour metabolism and its heterogeneity. In this case study we have used hyperpolarised 13C-pyruvate MRI (HP 13C-MRI) to assess 13C-lactate generation in a patient with an organ-confined FHd-RCC. Post-operative tissue samples were co-registered with imaging and underwent sequencing, IHC staining, and mass spectrometry imaging (MSI). HP 13C-MRI reveals two metabolically distinct tumour regions. The 13C-lactate-rich region shows a high lactate/pyruvate ratio and slightly lower fumarate on MSI compared to the other tumour region, as well as increased CD8 + T cell infiltration, and genetic dedifferentiation. Compared to the normal kidney, the vascularity in the tumour is decreased, while immune cell fraction is markedly higher. This study shows the potential of metabolic HP 13C-MRI to characterise FHd-RCC and how targeting of biopsies to regions of metabolic dysregulation could be used to obtain the tumour samples of greatest clinical significance, which in turn can inform on early and successful response to treatment. Horvat-Menih et al. use hyperpolarised 13C-pyruvate MRI (HP 13C-MRI) to assess 13C-lactate generation in a patient with a fumarate hydratase-deficient renal cell carcinoma (FHd-RCC), and correlate imaging findings with genetic and metabolic analysis on post-operative tissue samples. HP 13C-MRI reveals two metabolically distinct tumour regions. Kidney cancer resulting from a genetic mutation of a protein responsible for sugar metabolism, called “fumarate hydratase” is a rare and aggressive cancer. To understand this complex cancer better, our study aimed to assess the sugar metabolism directly in a patient bearing this tumour, using a specialised imaging technique. In addition, we performed biological analyses of the tumour. The results revealed two regions with different biology and aggressiveness within the same tumour. This shows the potential of the imaging technique to detect regions of varying aggressiveness within the same tumour, which could guide biopsies to obtain tumour samples and inform upon early treatment decisions.

PURPOSE:This study aims to address the limitation of long acquisition times in luminal water imaging (LWI), a promising noninvasive MRI technique for prostate cancer detection and grading, by implementing an accelerated multi-echo spin-echo method termed T2 mapping using echo merging plus k-t undersampling with reduced flip angles (TEMPURA). METHODS:TEMPURA enables faster acquisition through echo merging, k-t undersampling, and reduced refocusing flip angles. A prospective study was conducted on 24 patients (age 59-75 years) with biopsy-proven prostate cancer. Imaging was performed on a 3 T MRI system, comparing two TEMPURA-based LWI sequences-Fast (standard resolution) and High-Resolution (HR, doubled spatial resolution)-against a standard LWI sequence (Standard). Luminal water fraction and five additional parameters were compared across the methods. Statistical analyses included Spearman rank correlation, Wilcoxon rank sum test, receiver operating characteristic analysis, Bland-Altman plots, and Delong tests. RESULTS:Compared to Standard, Fast reduced the acquisition time from 8.3 to 2.8 min, whereas HR reduced it to 5.4 min and doubled spatial resolution. Both Fast and HR showed high correlation with Standard for luminal water fraction (r = 0.97/0.90 in peripheral zone, r = 0.91/0.93 in transition zone), with low bias (0.014/0.018 in peripheral zone, 0.024/0.024 in transition zone) and no significant differences (p = 0.05-0.84/0.08-0.51). No significant difference was observed in area under the curve values between Fast/HR and Standard among all parameters (p = 0.05-0.87). CONCLUSION:The acceleration method greatly reduced the acquisition time and increased the spatial resolution of LWI. Compared with the Standard acquisition, both the Fast and HR methods showed a high correlation for LWI measurements and consistent diagnostic performance in detecting malignant lesions.

New therapies are needed for patients who experience primary or secondary resistance to renal cell cancer (RCC) treatment. Pre-clinical data suggests PARP inhibitors may be effective for RCC in combination with VEGF inhibitors. Neoadjuvant clinical trials offer an opportunity to understand the mechanisms of new therapies by comparing tumour and blood before and after treatment. WIRE is a window of opportunity, phase II, multi-centre, multi-arm, non-randomised, neoadjuvant clinical trial platform (NCT03741426). Arms 1-3 comprised: 1. cediranib (VEGF inhibitor), 2. cediranib + olaparib (PARP inhibitor), 3. olaparib. Eligible patients (pts) have cT1b+, cN0/1, cM0/1 clear cell RCC, planned for surgery, with no contraindication to IMP. A Bayesian adaptive design optimises recruitment to arms based on interim analyses of treatment effects. Pts receive 14-28 days of IMP to fit with the planned surgery date. Primary endpoint for arms 1-3 is a ≥30% reduction in DCE-MRI assessed capillary permeability (median Ktrans) post-treatment compared to baseline. Secondary endpoints include RECIST v1.1 primary tumour response and adverse events. Translational analysis will include blood profiling and multi-region tissue analysis by transcriptomics and digital pathology. 29 pts were recruited (arm 1=6, arm 2=16, arm 3=7), 28/29 were male, median age 61y (range 48-75y). All pts were treatment naïve with ECOG PS of 0 or 1. 8/29 pts had M1 disease. All surgeries were completed within the planned window. 3/29 pts were not evaluable for the primary endpoint due to inadequate dose of IMP. The numbers of evaluable pts which met the primary endpoint were arm 1: 4/6 (67%), arm 2: 4/14 (29%), arm 3: 0/6 (0%). Table 1 shows changes in Ktrans, RECIST response and adverse events. Plasma angiogenic factors were significantly induced on treatment in the combination therapy arm 2. There was no correlation between angiogenic factor induction and Ktrans change. Positive responses in median Ktrans were observed in both the cediranib monotherapy and cediranib and olaparib combination therapy arms. Therapy was well tolerated, with no substantial toxicity or delays to surgery. Greater induction of angiogenic factors in the combination arm indicates possible synergy between cediranib and olaparib. Ongoing translational analysis of tissue and blood samples will investigate the mechanisms of response to these agents. James O. Jones, Ines Horvat Menih, Martin Thomas, Helen Mossop, Rebecca Wray, Maria Aquino, James Armitage, Harriet Baker, Carley Batley, James Blackmur, Sarah Burge, Anita Chhabra, Farhana Easita, Tim Eisen, Kate Fife, Angela Godoy, Richard Goodwin, Will Ince, Rose John, Alexander Laird, Natalia Lukashchuk, Athena Matakidou, Thomas J. Mitchell, Andrew N. Priest, Andrew Protheroe, Sreenidhi Ranjit, Anthony Riddick, Sulekha Said, Jamal Sipple, Amy Strong, Helen Su, Mark Sullivan, Silvia Tarantino, Gemma Tsang-Pells, Stephan Ursprung, Balaji Venugopal, Lauren Wallis, Anne Y. Warren, James Wason, Sarah J. Welsh, Younghwa Kim, John Stone, Mireia Crispin-Ortuzar, Ferdia A. Gallagher, Brent O'Carrigan, Grant D. Stewart, on behalf of the WIRE Trial Group. Neoadjuvant olaparib and cediranib in renal cancer: Outcomes of the WIndow-of-opportunity in REnal cancer (WIRE) Trial [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_2):Abstract nr CT175.

Magnetic resonance imaging of hyperpolarized (HP) [1-13C]pyruvate allows in-vivo assessment of metabolism and has translated into human studies across diseases at 15 centers worldwide. Consensus on best practice for multi-center studies is required to develop clinical applications. This paper presents the results of a 2-round formal consensus building exercise carried out by experts with HP [1-13C]pyruvate human study experience. Twenty-nine participants from 13 sites brought together expertise in pharmacy methods, MR physics, translational imaging, and data-analysis; with the goal of providing recommendations and best practice statements on conduct of multi-center human studies of HP [1-13C]pyruvate MRI. Overall, the group reached consensus on approximately two-thirds of 246 statements in the questionnaire, covering 'HP 13C-Pyruvate Preparation', 'MRI System Setup, Calibration, and Phantoms', 'Acquisition and Reconstruction', and 'Data Analysis and Quantification'. Consensus was present across categories, examples include that: (i) different HP pyruvate preparation methods could be used in human studies, but that the same release criteria have to be followed; (ii) site qualification and quality assurance must be performed with phantoms and that the same field strength must be used, but that the rest of the system setup and calibration methods could be determined by individual sites; (iii) the same pulse sequence and reconstruction methods were preferable, but the exact choice should be governed by the anatomical target; (iv) normalized metabolite area-under-curve (AUC) values and metabolite AUC were the preferred metabolism metrics. The work confirmed areas of consensus for multi-center study conduct and identified where further research is required to ascertain best practice.

The chemical shift of many molecules changes with temperature, which enables non-invasive magnetic resonance imaging (MRI) thermometry. Hyperpolarization methods increase the inherently low 13C MR signal. The commonly-used hyperpolarized probe [1-13C]pyruvate, and its metabolic product [1-13C]lactate, exhibit temperature and concentration dependent chemical shift changes that have not previously been reported. These effects were characterized at 7 T and 11.7 T in vitro and applied for in vivo thermometry both preclinically at 7 T and to human data at 3 T. Apparent temperature values from mouse abdomen and brain were similar to rectally measured temperature. Human brain and kidney apparent temperatures from 13C MRSI were lower than known physiological temperatures, suggesting that additional effects may currently limit the use of this method for determining absolute temperature in humans. The temperature dependent chemical shift changes also have implications for sequence design and for in vitro studies with hyperpolarized pyruvate.

LBA9508 Background: Optimal first line therapy for patients with metastatic melanoma is an immunotherapy regimen containing an anti-PD1 antibody, regardless of tumour BRAF mutation status. Anti-PD1 antibodies are licensed for use until disease progression. Recurrence rarely occurs in responding patients after 2 years of treatment. Optimal duration of anti-PD1-based immunotherapy has not been established. Reduced treatment duration may reduce the risk of long-term side-effects and generate cost savings for healthcare systems. Methods: DANTE (ISRCTN15837212) was a UK academic multi-centre parallel group non inferiority trial. Adults with unresectable stage III/IV melanoma receiving first line anti-PD1 +/- anti-CTLA-4 antibody immunotherapy were eligible. Patients who were progression-free after 1 year of treatment were randomised (1:1) to stop treatment (with the option of restarting on progression) or to continue treatment to at least 2 years in the absence of disease progression / unacceptable toxicity (control). The primary endpoint was progression-free survival (PFS) at 1 year post-randomization. Secondary endpoints included quality of life, best objective response, overall survival, toxicity and cost-effectiveness. A qualitative study explored patient acceptance of randomization. Follow-up to 4-years was planned for PFS with secondary outcomes collected up to 18-months post-randomization. Assuming a 2-year PFS rate in the control arm of 86% and defining non-inferiority (NI) as a reduction in PFS of no more than 6%, a sample size of 1208 patients (604 per arm) was required (80% power, 5% significance, 5% drop-out). DANTE closed early due to slow patient enrolment. PFS was compared between arms using Cox’s proportional hazards model, adjusting for stratification factors. Results: Between September 2018 and March 2023, 415 patients were registered from 36 UK hospitals and 166 patients (65.6% male, median age 74, BRAF mutant 25.9%) were randomised. Patient characteristics were broadly balanced. As of 27 th January 2025, with a median follow-up of 29.1 (IQR 17.9-39.3) months, there were 53 PFS events in total: 18 in the control arm (15 progressions+3 deaths) versus 35 in the stop arm (29 progressions+6 deaths). PFS rates at 1-year were 87.6% in the control arm and 80.2% in the stop arm (HR 1.76; 90% CI 1.03-3.03), with an absolute difference of -7.4% and 90% two-sided CI -17.1-2.32, which is within the pre-defined NI margin of 6%. Analyses are ongoing, results for secondary endpoints will be presented. Conclusions: DANTE is the largest prospective melanoma trial evaluating immunotherapy duration completed to date. Although results suggest stopping immunotherapy at 1 year was non-inferior compared to at least 2 years of treatment, the trial was underpowered due to early closure. Continuing immunotherapy for at least 2 years should remain as standard treatment. Clinical trial information: 15837212 .

To evaluate the capability of hyperpolarized [1-13C] pyruvate MRI to predict pathologic response to neoadjuvant treatment in multi-site abdominopelvic disease of high-grade serous ovarian cancer (HGSOC) patients and to compare 13C MRI and [18F]-FDG PET/CT measurements for detecting early treatment response. We recruited eight patients with HGSOC in this prospective study who underwent 13C MRI and [18F]-FDG PET/CT before and after the first cycle of neoadjuvant chemotherapy treatment (NACT). Imaging parameters were compared with clinical and histophatologic parameters. We demonstrate here that 13C MRI of hyperpolarized [1-13C]pyruvate metabolism in multiple abdominal metastases resulted in rapid labeling of the endogenous tumor lactate pool. The rate of labeling was similar between the different anatomical disease sites and independent of tumor volume. The apparent rate constant describing exchange of 13C label between pyruvate and lactate (kPL) was positively correlated with PET standard uptake values (SUVmax) for [18F]-FDG in metastatic tumor deposits in the ovary/pelvis (R = 0.471, P = 0.02). Decreased lactate labeling could be detected after the first cycle of neoadjuvant chemotherapy and was associated with pathological response. There was no overall decrease in lactate labeling in a single patient who lacked a complete histopathologic response. kPL was associated with cancer tissue LDHA concentration (rho = 0.641; P = 0.02). This exploratory study demonstrates the potential of 13C MRI measurements for assessing early response to neoadjuvant chemotherapy in patients with HGSOC.

Venous tumour thrombus (VTT), where the primary tumour invades the renal vein and inferior vena cava, affects 10-15% of renal cell carcinoma (RCC) patients. Curative surgery for VTT is high-risk, but neoadjuvant therapy may improve outcomes. The NAXIVA trial demonstrated a 35% VTT response rate after 8 weeks of neoadjuvant axitinib, a VEGFR-directed therapy. However, understanding non-response is critical for better treatment. Here we show that response to axitinib in this setting is characterised by a distinct and predictable set of features. We conduct a multiparametric investigation of samples collected during NAXIVA using digital pathology, flow cytometry, plasma cytokine profiling and RNA sequencing. Responders have higher baseline microvessel density and increased induction of VEGF-A and PlGF during treatment. A multi-modal machine learning model integrating features predict response with an AUC of 0.868, improving to 0.945 when using features from week 3. Key predictive features include plasma CCL17 and IL-12. These findings may guide future treatment strategies for VTT, improving the clinical management of this challenging scenario.

Cribriform prostate cancer (crPCa) is associated with poor clinical outcomes, yet its accurate detection remains challenging due to the poor sensitivity of standard-of-care diagnostic tools. Here, we use untargeted spatial metabolomics to identify fatty acid biosynthesis as a key metabolic pathway enriched in crPCa epithelium. We also show that imaging tumor lipid metabolism using [1-11C]acetate PET/CT and proton magnetic resonance spectroscopy differentiates cribriform from noncribriform intermediate-risk prostate cancers in two prospective patient cohorts. These findings support the feasibility of using clinical metabolic imaging techniques as adjunctive tools for improving crPCa detection in clinical practice, with prospective studies in larger cohorts warranted to obtain definitive results.

PURPOSE:To establish and optimize abdominal deuterium MR spectroscopic imaging in conjunction with orally administered 2H-labeled molecules. METHODS:A flexible transmit-receive surface coil was used to image naturally abundant deuterium signal in phantoms and healthy volunteers and after orally administered 2H2O in a patient with a benign renal tumor (oncocytoma). RESULTS:Water and lipid peaks were fitted with high confidence from both unlocalized spectra and from voxels within the liver, kidney, and spleen on spectroscopic imaging. Artifacts were minimal, despite the high 2H2O concentration in the stomach immediately after ingestion, which can be problematic with the use of a volume coil. CONCLUSION:We have shown the feasibility of abdominal deuterium MR spectroscopic imaging at 3 T using a flexible surface coil. Water measurements were obtained in healthy volunteers, and images were acquired in a patient with a renal tumor after drinking 2H2O. The limited depth penetration of the surface coil may have advantages in characterizing early uptake of orally administered agents in abdominal organs, despite the high concentrations in the stomach which can pose challenges with other coil combinations.

Venous tumor thrombus (VTT) is present in 10-15% patients with renal cell carcinoma (RCC), where the primary tumor invades the renal vein and inferior vena cava, increasing the morbidity and mortality of curative surgery. Neoadjuvant anti-angiogenic therapies have shown potential to reduce VTT length and improve nephrectomy success rates. The NAXIVA trial, a phase II study of neoadjuvant axitinib in 20 RCC patients with VTT, demonstrated that 35% of patients experienced significant downstaging of VTT. However, the biological mechanisms driving response or resistance in the remaining 65% are unclear. Machine learning has the potential to uncover mechanistic relationships between features in these patients and guide biomarker discovery for early response.In this study, we present a predictive machine learning framework based on multi-modal integration of digital pathology, flow cytometry and plasma cytokine profiling, using tissue and blood samples from the NAXIVA trial. A machine learning model incorporating recursive feature elimination and logistic regression with stochastic gradient descent was developed to predict response. To avoid overfitting, a leave-one-out nested cross validation approach was used for 20 iterations, leaving one patient out at a time and training the model on the remaining 19 patients. Two separate model ensembles were produced, using baseline features (N=62) and features from both baseline and two weeks post-treatment initiation (N=69) respectively. We used a statistical consensus approach to identify key predictive factors.The baseline model achieved an AUC of 0.868, identifying tissue microvessel density, plasma IL-12p70 and plasma CCL17 as key predictive factors. Responding patients had lower plasma IL-12p70, lower CCL17 and higher microvessel density than non-responding patients. Bulk RNA-seq data from pre-treatment tumor biopsies revealed elevated IL-12R expression in non-responders. Incorporating features post-treatment initiation improved model performance (AUC = 0.945), with fold changes in PlGF and sTie-2, alongside baseline IL-12p70 and CCL17, emerging as key predictors. Responders showed greater PlGF and sTie-2 induction following treatment onset.In conclusion, our machine learning framework accurately predicts response to neoadjuvant axitinib in RCC patients with VTT. The high microvessel density and growth factor induction in responders suggests they exhibit an “angiogenic” phenotype (clusters 1/2 from IMmotion151), while higher cytokine levels in non-responders suggest an “immunogenic” phenotype (cluster 4). Early changes in PlGF and sTie-2 may serve as actionable biomarkers for patient stratification. Further validation in independent datasets is essential to facilitate clinical integration and improve personalized treatment strategies for RCC patients with VTT. Rebecca Wray, Hania Paverd, Ines Machado, Johanna Barbieri, Farhana Easita, Abigail R. Edwards, Ferdia A. Gallagher, Iosif A. Mendichovszky, Thomas J. Mitchell, Maike De La Roche, Jacqueline D. Shields, Stephan Ursprung, Lauren Wallis, Anne Y. Warren, Sarah J. Welsh, Mireia Crispin-Ortuzar, Grant D. Stewart, James O. Jones. Multi-modal machine learning identifies predictive biomarkers of response to neoadjuvant axitinib in renal cell carcinoma with venous tumor thrombus [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 1103.

Background: Early and accurate grading of renal cell carcinoma (RCC) improves patient risk stratification and has implications for clinical management and mortality. However, current diagnostic approaches using imaging and renal mass biopsy have limited specificity and may lead to undergrading. Methods: This study explored the use of hyperpolarised [1-13C]pyruvate MRI (HP 13C-MRI) to identify the most aggressive areas within the tumour of patients with clear cell renal cell carcinoma (ccRCC) as a method to guide biopsy targeting and to reduce undergrading. Six patients with ccRCC underwent presurgical HP 13C-MRI and conventional contrast-enhanced MRI. From the imaging data, three k-means clusters were computed by combining the kPL as a marker of metabolic activity, and the 13C-pyruvate signal-to-noise ratio (SNRPyr) as a perfusion surrogate. The combined clusters were compared to those derived from individual parameters and to those derived from the percentage of enhancement on the nephrographic phase (%NG). The diagnostic performance of each cluster was assessed based on its ability to predict the highest histological tumour grade in postsurgical tissue samples. The postsurgical tissue samples underwent immunohistochemical staining for the pyruvate transporter (monocarboxylate transporter 1, MCT1), as well as RNA and whole-exome sequencing. Results: The clustering approach combining SNRPyr and kPL demonstrated the best performance for predicting the highest tumour grade: specificity 85%; sensitivity 64%; positive predictive value 82%; and negative predictive value 68%. Epithelial MCT1 was identified as the major determinant of the HP 13C-MRI signal. The perfusion/metabolism mismatch cluster showed an increased expression of metabolic genes and markers of aggressiveness. Conclusions: This study demonstrates the potential of using HP 13C-MRI-derived metabolic clusters to identify intratumoral variations in tumour grade with high specificity. This work supports the use of metabolic imaging to guide biopsies to the most aggressive tumour regions and could potentially reduce sampling error.

Abstract Background Tumour vascular density assessed from CD-31 immunohistochemistry (IHC) images has previously been shown to have prognostic value in breast cancer. Current methods to measure vascular density, however, are time-consuming, suffer from high inter-observer variability and are limited in describing the complex tumour vasculature morphometry. Methods We propose a method for automatically measuring a range of vascular parameters from CD-31 IHC images, which together provide a detailed description of the vasculature morphology. We first used a U-Net based convolutional neural network, trained and validated using 36 partially annotated whole slide images from 27 patients, to segment vessel structures and tumour regions from which the measurements are taken. The model also segments the vascular smooth muscle, benign epithelium, adipose tissue, stroma, lymphocyte clusters, nerves and CD-31 positive leukocytes, and we applied it to an additional 21 images from 15 patients. Using these segmentations, we investigated the relationship between the various tissue types and the vasculature and studied the relationship of various vascular parameters with clinical parameters. We also performed a 3D histology analysis on a separate tumour sample as a proof of principle, providing a more comprehensive visualization of vasculature morphology compared to the standard 2D cross-section of a tissue sample. Results Using two-way cross-validation, we show that vessels were accurately segmented, with Dice scores of 0.875 and 0.856, and were accurately identified, with F1 scores of 0.777 and 0.748. All vascular parameters exhibit strong ( $$r>0.7$$ r > 0.7 ) and significant (p<0.001) correlations with measurements taken from the manual ground truth vessel segmentations. A significant relationship between the major/minor axis ratio, a measure of elongation, and the tumour grade was found. Conclusion Our proposed method shows promise as a tool for studying the tumour vasculature and its relationship with surrounding cells and tissue types. Furthermore, the correlation with tumour grade highlights the clinical relevance of our approach. These findings suggest that our method could have substantial implications for improving prognostic assessments and personalizing therapeutic strategies in breast cancer treatment.

In renal histopathology, the routine clinical use of several histological stains presents challenges for the direct application of stain-specific deep learning-based analysis tools to whole-slide images. We present an approach to the in silico histological staining of kidney tissue where samples stained with hematoxylin and eosin (H&E) are virtually restained with periodic acid-Schiff (PAS). Our approach is underpinned by cycle-consistent generative adversarial neural networks trained on the National Unified Renal Translational Research Enterprise data set-the first UK-wide Biobank for chronic kidney disease-which features diverse data from 16 nephrology centers. Our work is divided into the following 4 main components: (1) we developed a virtual staining model, which infers PAS staining from H&E; (2) 2 board-certified pathologists assessed the virtual staining by attempting to distinguish it from real examples; (3) we trained a glomerular segmentation model using 3 independent renal segmentation data sets (Kidney Precision Medicine Project, Human BioMolecular Atlas Program [Kidney], and data by Jayapandian et al); and (4) we demonstrated the utility of virtual staining by inferring PAS staining from previously unseen H&E test images and applying our PAS-specific glomerular segmentation model. Each pathologist was able to identify 52.5% and 75.8% of the virtually stained images, respectively, showing an overlap in the variability of the authentic and synthetic staining. We discussed the utility of virtual staining in digital pathology, the need for pathology-specific testing with respect to chronic damage, and minimal changes and steps for incorporating more stains. Furthermore, alongside this article, we included complete glomerular annotations for 20 Kidney Precision Medicine Project H&E-stained slides.

BACKGROUND:Succinate dehydrogenase (SDH) deficient wild-type Gastrointestinal Stromal Tumours (wtGIST) are a rare GIST subtype with limited treatment options. Gallium-68 labelled Gastrin Releasing Peptide Receptor (GRPR) antagonist NeoB has shown promise in PET imaging for multiple primary malignancies. This investigation sought to assess the biodistribution of [68Ga]NeoB via PET/CT imaging in metastatic wtGIST patients and aimed to evaluate GRPR expression in lesions to determine the ligand's potential for patient selection in future therapeutic trials. RESULTS:Twelve patients with histologically confirmed metastatic wtGIST were enrolled. [68Ga]NeoB PET/CT imaging was conducted for lesion segmentation and analysis of uptake characteristics. 8 of 12 (66.7%) patients exhibited intense but heterogeneous [68Ga]NeoB uptake in lesions, with variable tracer uptake both within and between lesions. Physiological uptake was highest in the pancreas, liver, and spleen. Four patients (33.3%) displayed minimal or no uptake in tumour lesions. CONCLUSIONS:The majority of wtGIST patients in this small cohort show lesions with intense [68Ga]NeoB uptake. Heterogeneity of uptake indicates GRPR has highly variable inter- and intralesional expression. NeoB has potential for theranostic application in wtGIST, with limited effective standard of care treatments available. Ongoing trials are investigating the therapeutic use of [177Lu]NeoB in this setting. CLINICAL TRIAL NUMBER:Not applicable.