A whirl of radiomics-based biomarkers in cancer immunotherapy, why is large scale validation still lacking?

The search for understanding immunotherapy response has sparked interest in diverse areas of oncology, with artificial intelligence (AI) and radiomics emerging as promising tools, capable of gathering large amounts of information to identify suitable patients for treatment. The application of AI in radiology has grown, driven by the hypothesis that radiology images capture tumor phenotypes and thus could provide valuable insights into immunotherapy response likelihood. However, despite the rapid growth of studies, no algorithms in the field have reached clinical implementation, mainly due to the lack of standardized methods, hampering study comparisons and reproducibility across different datasets. In this review, we performed a comprehensive assessment of published data to identify sources of variability in radiomics study design that hinder the comparison of the different model performance and, therefore, clinical implementation. Subsequently, we conducted a use-case meta-analysis using homogenous studies to assess the overall performance of radiomics in estimating programmed death-ligand 1 (PD-L1) expression. Our findings indicate that, despite numerous attempts to predict immunotherapy response, only a limited number of studies share comparable methodologies and report sufficient data about cohorts and methods to be suitable for meta-analysis. Nevertheless, although only a few studies meet these criteria, their promising results underscore the importance of ongoing standardization and benchmarking efforts. This review highlights the importance of uniformity in study design and reporting. Such standardization is crucial to enable meaningful comparisons and demonstrate the validity of biomarkers across diverse populations, facilitating their implementation into the immunotherapy patient selection process.

Clinical value of liquid biopsy in FGFR2 fusion-positive cholangiocarcinoma patients during targeted therapy.

PURPOSE: FGFR2 fusions occur in 10%-15% of intrahepatic cholangiocarcinoma (iCCA) patients, potentially benefiting from FGFR inhibitors (FGFRi). We aimed to assess the feasibility of detecting FGFR2 fusions in plasma and explore plasma biomarkers for managing FGFRi treatment. EXPERIMENTAL DESIGN: We conducted a retrospective study on 18 patients with iCCA and known FGFR2 fusions previously identified in tissue samples from prior FGFRi treatment. Both tissue and synchronous plasma samples were analyzed using a custom hybrid capture gene panel with next-generation sequencing (VHIO-iCCA panel) and validated against commercial vendor results. Longitudinal plasma analysis during FGFRi was performed. Subsequently, we explored the correlation between plasma biomarkers, liver enzymes, tumor volume, and clinical outcomes. RESULTS: Sixteen patients (88.9%) were positive for FGFR2 fusion events in plasma. Remarkably, the analysis of plasma suggests that lower levels of circulating tumor DNA (ctDNA) are linked to clinical benefits from targeted therapy and result in improved progression-free survival and (PFS) overall survival (OS). Higher concentrations of cell-free DNA (cfDNA) before FGFRi treatment were linked to worse OS, correlating with impaired liver function, and indicating compromised cfDNA removal by the liver. Additionally, increased ctDNA or the emergence of resistance mutations allowed earlier detection of disease progression compared to standard radiological imaging methods. CONCLUSIONS: VHIO-iCCA demonstrated accurate detection of FGFR2 fusions in plasma. The integration of information from various plasma biomarkers holds the potential to predict clinical outcomes and identify treatment failure prior to radiological progression, offering valuable guidance for the clinical management of patients with iCCA.

Weakly Supervised Deep Learning Predicts Immunotherapy Response in Solid Tumors Based on PD-L1 Expression.

Programmed death-ligand 1 (PD-L1) IHC is the most commonly used biomarker for immunotherapy response. However, quantification of PD-L1 status in pathology slides is challenging. Neither manual quantification nor a computer-based mimicking of manual readouts is perfectly reproducible, and the predictive performance of both approaches regarding immunotherapy response is limited. In this study, we developed a deep learning (DL) method to predict PD-L1 status directly from raw IHC image data, without explicit intermediary steps such as cell detection or pigment quantification. We trained the weakly supervised model on PD-L1-stained slides from the non-small cell lung cancer (NSCLC)-Memorial Sloan Kettering (MSK) cohort (N = 233) and validated it on the pan-cancer-Vall d'Hebron Institute of Oncology (VHIO) cohort (N = 108). We also investigated the performance of the model to predict response to immune checkpoint inhibitors (ICI) in terms of progression-free survival. In the pan-cancer-VHIO cohort, the performance was compared with tumor proportion score (TPS) and combined positive score (CPS). The DL model showed good performance in predicting PD-L1 expression (TPS ≥ 1%) in both NSCLC-MSK and pan-cancer-VHIO cohort (AUC 0.88 ± 0.06 and 0.80 ± 0.03, respectively). The predicted PD-L1 status showed an improved association with response to ICIs [HR: 1.5 (95% confidence interval: 1-2.3), P = 0.049] compared with TPS [HR: 1.4 (0.96-2.2), P = 0.082] and CPS [HR: 1.2 (0.79-1.9), P = 0.386]. Notably, our explainability analysis showed that the model does not just look at the amount of brown pigment in the IHC slides, but also considers morphologic factors such as lymphocyte conglomerates. Overall, end-to-end weakly supervised DL shows potential for improving patient stratification for cancer immunotherapy by analyzing PD-L1 IHC, holistically integrating morphology and PD-L1 staining intensity. SIGNIFICANCE: The weakly supervised DL model to predict PD-L1 status from raw IHC data, integrating tumor staining intensity and morphology, enables enhanced patient stratification in cancer immunotherapy compared with traditional pathologist assessment.

Identification of Precise 3D CT Radiomics for Habitat Computation by Machine Learning in Cancer.

Purpose To identify precise three-dimensional radiomics features in CT images that enable computation of stable and biologically meaningful habitats with machine learning for cancer heterogeneity assessment. Materials and Methods This retrospective study included 2436 liver or lung lesions from 605 CT scans (November 2010-December 2021) in 331 patients with cancer (mean age, 64.5 years ± 10.1 [SD]; 185 male patients). Three-dimensional radiomics were computed from original and perturbed (simulated retest) images with different combinations of feature computation kernel radius and bin size. The lower 95% confidence limit (LCL) of the intraclass correlation coefficient (ICC) was used to measure repeatability and reproducibility. Precise features were identified by combining repeatability and reproducibility results (LCL of ICC ≥ 0.50). Habitats were obtained with Gaussian mixture models in original and perturbed data using precise radiomics features and compared with habitats obtained using all features. The Dice similarity coefficient (DSC) was used to assess habitat stability. Biologic correlates of CT habitats were explored in a case study, with a cohort of 13 patients with CT, multiparametric MRI, and tumor biopsies. Results Three-dimensional radiomics showed poor repeatability (LCL of ICC: median [IQR], 0.442 [0.312-0.516]) and poor reproducibility against kernel radius (LCL of ICC: median [IQR], 0.440 [0.33-0.526]) but excellent reproducibility against bin size (LCL of ICC: median [IQR], 0.929 [0.853-0.988]). Twenty-six radiomics features were precise, differing in lung and liver lesions. Habitats obtained with precise features (DSC: median [IQR], 0.601 [0.494-0.712] and 0.651 [0.52-0.784] for lung and liver lesions, respectively) were more stable than those obtained with all features (DSC: median [IQR], 0.532 [0.424-0.637] and 0.587 [0.465-0.703] for lung and liver lesions, respectively; P < .001). In the case study, CT habitats correlated quantitatively and qualitatively with heterogeneity observed in multiparametric MRI habitats and histology. Conclusion Precise three-dimensional radiomics features were identified on CT images that enabled tumor heterogeneity assessment through stable tumor habitat computation. Keywords: CT, Diffusion-weighted Imaging, Dynamic Contrast-enhanced MRI, MRI, Radiomics, Unsupervised Learning, Oncology, Liver, Lung Supplemental material is available for this article. © RSNA, 2024 See also the commentary by Sagreiya in this issue.

Analysis of cell free DNA to predict outcome to bevacizumab therapy in colorectal cancer patients.

To predict outcome to combination bevacizumab (BVZ) therapy, we employed cell-free DNA (cfDNA) to determine chromosomal instability (CIN), nucleosome footprints (NF) and methylation profiles in metastatic colorectal cancer (mCRC) patients. Low-coverage whole-genome sequencing (LC-WGS) was performed on matched tumor and plasma samples, collected from 74 mCRC patients from the AC-ANGIOPREDICT Phase II trial (NCT01822444), and analysed for CIN and NFs. A validation cohort of plasma samples from the University Medical Center Mannheim (UMM) was similarly profiled. 61 AC-ANGIOPREDICT plasma samples collected before and following BVZ treatment were selected for targeted methylation sequencing. Using cfDNA CIN profiles, AC-ANGIOPREDICT samples were subtyped with 92.3% accuracy into low and high CIN clusters, with good concordance observed between matched plasma and tumor. Improved survival was observed in CIN-high patients. Plasma-based CIN clustering was validated in the UMM cohort. Methylation profiling identified differences in CIN-low vs. CIN high (AUC = 0.87). Moreover, significant methylation score decreases following BVZ was associated with improved outcome (p = 0.013). Analysis of CIN, NFs and methylation profiles from cfDNA in plasma samples facilitates stratification into CIN clusters which inform patient response to treatment.

LOXL2-mediated chromatin compaction is required to maintain the oncogenic properties of triple-negative breast cancer cells.

Oxidation of histone H3 at lysine 4 (H3K4ox) is catalyzed by lysyl oxidase homolog 2 (LOXL2). This histone modification is enriched in heterochromatin in triple-negative breast cancer (TNBC) cells and has been linked to the maintenance of compacted chromatin. However, the molecular mechanism underlying this maintenance is still unknown. Here, we show that LOXL2 interacts with RuvB-Like 1 (RUVBL1), RuvB-Like 2 (RUVBL2), Actin-like protein 6A (ACTL6A), and DNA methyltransferase 1associated protein 1 (DMAP1), a complex involved in the incorporation of the histone variant H2A.Z. Our experiments indicate that this interaction and the active form of RUVBL2 are required to maintain LOXL2-dependent chromatin compaction. Genome-wide experiments showed that H2A.Z, RUVBL2, and H3K4ox colocalize in heterochromatin regions. In the absence of LOXL2 or RUVBL2, global levels of the heterochromatin histone mark H3K9me3 were strongly reduced, and the ATAC-seq signal in the H3K9me3 regions was increased. Finally, we observed that the interplay between these series of events is required to maintain H3K4ox-enriched heterochromatin regions, which in turn is key for maintaining the oncogenic properties of the TNBC cell line tested (MDA-MB-231).

Cancer Core Europe: Leveraging Institutional Synergies to Advance Oncology Research and Care Globally.

Cancer Core Europe brings together the expertise, resources, and interests of seven leading cancer institutes committed to leveraging collective innovation and collaboration in precision oncology. Through targeted efforts addressing key medical challenges in cancer and partnerships with multiple stakeholders, the consortium seeks to advance cancer research and enhance equitable patient care.