A multicentre analytical comparison study of inter-reader and inter-assay agreement of four programmed death-ligand 1 immunohistochemistry assays for scoring in triple-negative breast cancer.

AIMS: Studies in various cancer types have demonstrated discordance between results from different programmed death-ligand 1 (PD-L1) assays. Here, we compare the reproducibility and analytical concordance of four clinically developed assays for assessing PD-L1-positivity in tumour-infiltrating immune cells in the tumour area (PD-L1-IC-positivity) in triple-negative breast cancer (TNBC). METHODS AND RESULTS: Primary TNBC resection specimens (n = 30) were selected based on their PD-L1-IC-positivity per VENTANA SP142 (<1%: 15 cases; 1-5%: seven cases; >5%: eight cases). Serial histological sections were stained for PD-L1 using VENTANA SP142, VENTANA SP263, DAKO 22C3 and DAKO 28-8. PD-L1-IC-positivity and tumour cell expression (≥1 versus <1%) were scored by trained readers from seven sites using online virtual microscopy. The adjusted mean of PD-L1-IC-positivity for SP263 (7.8%) was significantly higher than those for the other three assays (3.7-4.9%). Differences in adjusted means were statistically significant between SP263 and the other three assays (P < 0.0001) but not between the three remaining assays when excluding SP263 (P = 0.0961-0.6522). Intra-class correlation coefficients revealed moderate-to-strong inter-reader agreement for each assay (0.460-0.805) and poor-to-strong inter-assay agreement for each reader (0.298-0.678) on PD-L1-IC-positivity. CONCLUSIONS: In this first multicentre study of different PD-L1 assays in TNBC, we show that PD-L1-IC-positivity for SP142, 22C3 and 28-8 was reproducible and analytically concordant, indicating that these three assays may be analytically interchangeable. The relevance of the higher PD-L1-IC-positivity for SP263 should be further investigated.

cbpManager: a web application to streamline the integration of clinical and genomic data in cBioPortal to support the Molecular Tumor Board.

BACKGROUND: Extensive sequencing of tumor tissues has greatly improved our understanding of cancer biology over the past years. The integration of genomic and clinical data is increasingly used to select personalized therapies in dedicated tumor boards (Molecular Tumor Boards) or to identify patients for basket studies. Genomic alterations and clinical information can be stored, integrated and visualized in the open-access resource cBioPortal for Cancer Genomics. cBioPortal can be run as a local instance enabling storage and analysis of patient data in single institutions, in the respect of data privacy. However, uploading clinical input data and genetic aberrations requires the elaboration of multiple data files and specific data formats, which makes it difficult to integrate this system into clinical practice. To solve this problem, we developed cbpManager. RESULTS: cbpManager is an R package providing a web-based interactive graphical user interface intended to facilitate the maintenance of mutations data and clinical data, including patient and sample information, as well as timeline data. cbpManager enables a large spectrum of researchers and physicians, regardless of their informatics skills to intuitively create data files ready for upload in cBioPortal for Cancer Genomics on a daily basis or in batch. Due to its modular structure based on R Shiny, further data formats such as copy number and fusion data can be covered in future versions. Further, we provide cbpManager as a containerized solution, enabling a straightforward large-scale deployment in clinical systems and secure access in combination with ShinyProxy. cbpManager is freely available via the Bioconductor project at https://bioconductor.org/packages/cbpManager/ under the AGPL-3 license. It is already used at six University Hospitals in Germany (Mainz, Gießen, Lübeck, Halle, Freiburg, and Marburg). CONCLUSION: In summary, our package cbpManager is currently a unique software solution in the workflow with cBioPortal for Cancer Genomics, to assist the user in the interactive generation and management of study files suited for the later upload in cBioPortal.

Using immuno-PET imaging to monitor kinetics of T cell-mediated inflammation and treatment efficiency in a humanized mouse model for GvHD.

PURPOSE: Hematopoietic stem cell transplantation is the only curative treatment for several hematological malignancies and immune deficiency syndromes. Nevertheless, the development of graft-versus-host disease (GvHD) after transplantation is a severe complication with high morbidity and mortality. The aim of this study was to image human T cells during GvHD development and their migration into GvHD-related organs. By using a radiolabeled anti-human CD3 monoclonal antibody (mAb), we were able to visualize GvHD progression in a humanized mouse model. METHODS: Human peripheral blood mononuclear cells (PBMC) were transferred into immunodeficient mice (initially n = 11 mice/group) to induce GvHD. One group additionally received regulatory T cells (Treg) for prevention of GvHD. T cell migration was visualized by sequential small animal PET/MRI using 89Zr-labeled anti-human CD3 mAb. Flow cytometry and immunohistochemistry were used to measure T cell frequencies in relevant organs at different time points after engraftment. RESULTS: Using radiolabeled anti-CD3 mAb, we successfully visualized human T cells in inflamed organs of mice by 89Zr-anti-CD3-PET/MRI. Upon GvHD progression, we observed increased numbers of CD3+ T cells in the liver (22.9% on day 3; 94.2% on day 10) and the spleen (4.4% on day 3; 58.8% on day 10) which correlated with clinical symptoms. The liver showed distinct spot-like lesions representing a strong focal accumulation of T cells. Administration of Treg prior GvHD induction reduced T cell accumulation in the liver from 857 ± 177 CD3+ cells/mm2 to 261 ± 82 CD3+ cells/mm2 and thus prevented GvHD. CONCLUSION: 89Zr-labeled anti-human CD3 mAb can be used as a proof of concept to detect the exact spatio-temporal distribution of GvHD-mediating T cells. In the future, radiolabeled T cell-specific mAb could be employed as a predictive early biomarker during the course of GvHD maybe even before clinical signs of the disease become evident. Furthermore, monitoring T cell migration and proliferation might improve tailored GvHD therapy.

Concise Review: Increasing the Validity of Cerebrovascular Disease Models and Experimental Methods for Translational Stem Cell Research.

Interspecies differences, anatomical and physiological aspects, as wells as simplified study designs contribute to an overestimation of treatment effects and limit the transferability of experimental results into clinical applications. Confounders of cell therapies for cerebrovascular disorders (CVD) include common CVD comorbidities, frequent medications potentially affecting endogenous and transplanted stem cells, as well as age- and immune-system-related effects. All those can contribute to a substantial modeling bias, ultimately limiting the prospective quality of preclinical research programs regarding the clinical value of a particular cell therapy. In this review, we discuss the nature and impact of most relevant confounders. We provide suggestions on how they can be considered to enhance the validity of CVD models in stem cell research. Acknowledging substantial and sometimes surprising effects of housing conditions, chronobiology, and intersex differences will further augment the translational value of animal models. We finally discuss options for the implementation of high-quality functional and imaging readout protocols. Altogether, this might help to gain a more holistic picture about the therapeutic impact of a particular cell therapy for CVD, but also on potential side and off-site effects of the intervention. Stem Cells 2017;35:1141-1153.

Astrocytic expression of the CXCL12 receptor, CXCR7/ACKR3 is a hallmark of the diseased, but not developing CNS.

Based on our previous demonstration of CXCR7 as the major mediator of CXCL12 signaling in cultured astrocytes, we have now compared astrocytic expression of the CXCL12 receptors, CXCR7 and CXCR4, during CNS development and disease. In addition, we asked whether disease-associated conditions/factors affect expression of CXCL12 receptors in astrocytes. In the late embryonic rat brain, CXCR7+/GFAP+ cells were restricted to the ventricular/subventricular zone while CXCR4 was widely absent from GFAP-positive cells. In the early postnatal and adult brain, CXCR7 and CXCR4 were almost exclusively expressed by GFAP-immunoreactive astrocytes forming the superficial glia limitans. Contrasting the situation in the intact CNS, a striking increase in astrocytic CXCR7 expression was detectable in the cortex of rats with experimental brain infarcts, in the spinal cord of rats with experimental autoimmune encephalomyelitis (EAE) and after mechanical compression, as well as in the in infarcted human cerebral cortex and in the hippocampus of Alzheimer's disease patients. None of these pathologies was associated with substantial increases in astrocytic CXCR4 expression. Screening of various disease-associated factors/conditions further revealed that CXCR7 expression of cultured cortical astrocytes increases with IFNγ as well as under hypoxic conditions whereas CXCR7 expression is attenuated following treatment with IFNß. Again, none of the treatments affected CXCR4 expression in cultured astrocytes. Together, these findings support the hypothesis of a crucial role of astrocytic CXCR7 in the progression of various CNS pathologies.

High-dosage granulocyte colony stimulating factor treatment alters monocyte trafficking to the brain after experimental stroke.

Ischemic stroke elicits a prompt inflammatory response that is characterized by a well-timed recruitment of peripheral immune cells to the brain. Among these, monocytes play a particularly important, but multifaceted role and have been increasingly recognized to affect stroke outcome. Granulocyte colony stimulating factor (GCSF) is known for its immunosuppressive actions on mononuclear cells, but previous studies in the stroke field were mainly confined to its neuroprotective actions. Herein, we investigated whether GCSF affects post-stroke inflammation in a mouse model of focal brain ischemia by modulating monocyte responses. Treatment with GCSF was controlled by vehicle injection, sham surgery and naive animals. Despite a significant monocytosis, high-dosage GCSF reduced the number of brain-infiltrating monocytes/macrophages four days after stroke. Lower numbers of mononuclear phagocytes in the brain were associated with smaller cerebral edema and improved motor outcome after stroke. GCSF treatment over 72h, but not 24h diminished integrin expression on circulating Ly6C+ inflammatory monocytes. In vitro experiments further revealed that GCSF strongly promotes interleukin (IL)-10 secretion by activated mononuclear cells. Blockade of the IL-10 receptor partly reversed GCSF-induced downregulation of integrin surface expression. Overall, our results suggest that high-dosage GCSF mitigates monocyte infiltration after stroke, likely by attenuating integrin-mediated adhesion to the brain endothelium in an IL-10-dependent manner. Lower amounts of mononuclear cells in the brain translate to less severe brain edema and functional impairment and thus support a harmful role of Ly6C+ inflammatory monocytes in the acute stage of stroke.

Bone marrow cell transplantation time-dependently abolishes efficacy of granulocyte colony-stimulating factor after stroke in hypertensive rats.

BACKGROUND AND PURPOSE: We aimed to determine a possible synergistic effect of granulocyte colony-stimulating factor (G-CSF) and bone marrow-derived mononuclear cells (BM MNC) after stroke in spontaneously hypertensive rats. METHODS: Male spontaneously hypertensive rats were subjected to middle cerebral artery occlusion and randomly assigned to daily injection of 50 µg/kg G-CSF for 5 days starting 1 hour after stroke (groups 1, 2, and 3) with additional intravenous transplantation of 1.5×10E7 BM MNC per kilogram at 6 hours (group 2) or 48 hours (group 3) after stroke, or control treatment (group 4). Circulating leukocyte counts and functional deficits, infarct volume, and brain edema were repeatedly assessed in the first week and first month. RESULTS: G-CSF treatment led to a significant neutrophilia, to a reversal of postischemic depression of circulating leukocytes, and to a significantly improved functional recovery without affecting the infarct volume or brain edema. BM MNC cotransplantation was neutral after 6 hours, but reversed the functional effect of G-CSF after 48 hours. Short-term investigation of combined G-CSF and BM MNC treatment at 48 hours indicated splenic accumulation of granulocytes and transplanted cells, accompanied by a significant rise of granulocytes in the circulation and the ischemic brain. CONCLUSIONS: G-CSF improved functional recovery in spontaneously hypertensive rats, but this effect was abolished by cotransplantation of BM MNC after 48 hours. In the spleen, transplanted cells may hinder the clearance of granulocytes that were massively increased by G-CSF. Increased circulation and infiltration of granulocytes into the ischemic brain may be detrimental for stroke outcome.

Allometric dose retranslation unveiled substantial immunological side effects of granulocyte colony-stimulating factor after stroke.

BACKGROUND AND PURPOSE: Granulocyte colony-stimulating factor (GCSF) showed robust neuroprotective and neuroregenerative properties after stroke in rodents but failed to meet study end points in patients. Because immunologic side effects of GCSF may have escaped preclinical testing because of nonallometric dose translation, we hypothesized those as possible reasons. METHODS: Stroke was induced in C57BL/6 mice by 45-minute filament middle cerebral artery occlusion. GCSF was administered at 50 and 832.5 µg/kg body weight. Treatment was controlled by vehicle injection, sham surgery, and naive animals. Immune cell counts were assessed in blood, spleen, and brain by multidimensional flow cytometry 1 day after stroke. RESULTS: High-dose GCSF significantly altered myeloid and T-cell subpopulations in blood and spleen and caused a tremendous increase of monocytes/macrophages infiltrating the ischemic brain. CONCLUSIONS: Dose-dependent immunomodulation superimposes central nervous system-specific effects of GCSF after stroke. Adaption of dose or treatment time may overcome this drawback.

PD-L1 Immunohistochemistry in Gastric Cancer: Comparison of Combined Positive Score and Tumor Area Positivity Across 28-8, 22C3, and SP263 Assays.

PURPOSE: The clinical application of PD-L1 immunohistochemistry (IHC) testing is complicated by the availability of multiple IHC assays, scoring algorithms, and cutoffs. This study assessed the analytical comparability of three commercially available PD-L1 assays and two scoring algorithms used to assess PD-L1 status in gastric cancer (GC) samples. METHODS: Serial sections of 100 resected GC samples, with PD-L1 expression levels across the dynamic range, were stained with three in vitro diagnostic-grade PD-L1 assays (28-8, 22C3, and SP263). Three trained pathologists blindly and independently scored slides using combined positive score (CPS) and tumor area positivity (TAP) algorithms. Comprehensive statistical analyses were performed to evaluate analytical concordance. Digital image analysis (DIA) was used to objectively compare the technical performance of each assay by simulating CPS and TAP. RESULTS: Comparable staining patterns were observed with these three PD-L1 assays. Despite discernible variation in staining intensity, reproducible evaluations of PD-L1 positivity were observed. Inter- and intra-assay assessments of all three assays, using either CPS or TAP and the same PD-L1 cutoffs, demonstrated moderate to almost-perfect (interassay Cohen's kappa [κ] range, 0.47-0.83) and substantial to almost-perfect (intra-assay κ range, 0.77-1.00) agreement. Interpathologist assessment exhibited a significant level of concordance (intraclass correlation coefficient ≥0.92). No difference in technical performance was observed using DIA. CONCLUSION: This study highlights analytical concordance in PD-L1 testing between three major PD-L1 assays when TAP and CPS are applied. Comparability of the technical assay performance was further supported by independent DIA. These observations support cross-application flexibility of the different PD-L1 assays and scoring algorithms to characterize PD-L1 expression in GC.

Multistain deep learning for prediction of prognosis and therapy response in colorectal cancer.

Although it has long been known that the immune cell composition has a strong prognostic and predictive value in colorectal cancer (CRC), scoring systems such as the immunoscore (IS) or quantification of intraepithelial lymphocytes are only slowly being adopted into clinical routine use and have their limitations. To address this we established and evaluated a multistain deep learning model (MSDLM) utilizing artificial intelligence (AI) to determine the AImmunoscore (AIS) in more than 1,000 patients with CRC. Our model had high prognostic capabilities and outperformed other clinical, molecular and immune cell-based parameters. It could also be used to predict the response to neoadjuvant therapy in patients with rectal cancer. Using an explainable AI approach, we confirmed that the MSDLM's decisions were based on established cellular patterns of anti-tumor immunity. Hence, the AIS could provide clinicians with a valuable decision-making tool based on the tumor immune microenvironment.

Discordance of HER2-Low between Primary Tumors and Matched Distant Metastases in Breast Cancer.

We examined differences in HER2 expression between primary tumors and distant metastases, particularly within the HER2-negative primary breast cancer cohort (HER2-low and HER2-zero). The retrospective study included 191 consecutive paired samples of primary breast cancer and distant metastases diagnosed between 1995 and 2019. HER2-negative samples were divided into HER2-zero (immunohistochemistry [IHC] score 0) and HER2-low (IHC score 1+ or 2+/in situ hybridization [ISH]-negative). The main objective was to analyze the discordance rate between matched primary and metastatic samples, focusing on the site of distant metastasis, molecular subtype, and de novo metastatic breast cancer. The relationship was determined by cross-tabulation and calculation of Cohen's Kappa coefficient. The final study cohort included 148 paired samples. The largest proportion in the HER2-negative cohort was HER2-low [primary tumor 61.4% (n = 78), metastatic samples 73.5% (n = 86)]. The discordance rate between the HER2 status of primary tumors and corresponding distant metastases was 49.6% (n = 63) (Kappa -0.003, 95%CI -0.15-0.15). Development of a HER2-low phenotype occurred most frequently (n = 52, 40.9%), mostly with a switch from HER2-zero to HER2-low (n = 34, 26.8%). Relevant HER2 discordance rates were observed between different metastatic sites and molecular subtypes. Primary metastatic breast cancer had a significantly lower HER2 discordance rate than secondary metastatic breast cancer [30.2% (Kappa 0.48, 95%CI 0.27-0.69) versus 50.5% (Kappa 0.14, 95% CI -0.03-0.32)]. This highlights the importance of evaluating potentially therapy-relevant discordance rates between a primary tumor and corresponding distant metastases.

Changes in T2 relaxation time after stroke reflect clearing processes.

BACKGROUND AND PURPOSE: CT and MR imaging techniques are frequently used for the diagnosis and progress monitoring of ischemic stroke in clinical practice and research. After stroke, both methods are characterized by a transient pseudo-normalized imaging signal, the so-called fogging phenomenon. This study evaluates potential pathophysiological changes associated with fogging, as well as its influence on the correct determination of the ischemic lesion in a rat stroke model. METHODS: Male spontaneously hypertensive rats were subjected to permanent middle cerebral artery occlusion. Ischemic lesion volume, brain edema and gray scale value spread within the ischemic lesion were determined on T2-weighted MR sequences at days 1, 4, 8, 11 and 29 after stroke onset, and compared with immunohistochemistry for astrogliosis, microglia/macrophage infiltration and angiogenesis. RESULTS: All animals showed MR fogging at days 4, 8 and 11 after stroke. The transient normalization of T2 signals occurred independently from the development of infarct volumes, but coincided well with the spatio-temporal occurrence of necrosis, angiogenesis and microglia/macrophage infiltration. CONCLUSIONS: Our results suggest that the fogging effect reflects the clearance of necrotic tissue within the ischemic lesion and is thus not relevant for the determination of the lesion volume.

CD15 immunostaining improves placental diagnosis of fetal hypoxia.

INTRODUCTION: Fetal hypoxic events with unclear predictive value are a common indication for placenta examination. We evaluated whether the use of CD15 immunostaining can improve the assessment of severity and duration of fetal hypoxia. METHODS: We compared placentas (37-42 gestational weeks) from stillborns/newborns with birth asphyxia (BA) and non-hypoxic newborns. Placental findings were studied in following groups: (1) acute BA (n = 11) due to placental abruption, (2) non-acute BA (n = 121) due to non-acute conditions, (3) non-BA (n = 46) in pregnancies with preeclampsia and gestational diabetes, and (4) controls (n = 30). RESULTS: A high expression of CD15 in feto-placental resistance vessels (FRVs) was present in non-acute BA (95.9%), but absent in acute BA, non-BA and controls (p < 0.0001). Furthermore, we found no causal relationship of high expression of CD15 in FRVs to coexisting placental conditions, including severity and mechanisms/patterns of placental injury, fetal erythroblastosis, and maternal conditions. According to a multivariate analysis, only a high expression of CD15 in FRVs was independently associated with severe non-acute fetal hypoxia ([OR] = 15.52; 95% [CI] = 5.92-40.67). DISCUSSION: We have defined a characteristic pattern of CD15 expression in FRVs that allows to interpret various clinical/placental conditions with respect to fetal hypoxia, with an improved predictability compared to conventional analyses. This approach represents a novel diagnostic strategy for placenta examination, which could indirectly assess severity and duration of intrauterine hypoxia in a heterogeneous population of newborns.

Expression of programmed death ligand-1 (PD-L1) in metastatic and postchemotherapy viable testicular germ cell tumors.

INTRODUCTION: Chemotherapy for testicular germ cell tumors (GCT) is highly effective, with few patients who do not respond. Clinical studies to evaluated novel treatments are challenging given the rarity of these patients. Therefore, we sought to evaluate PD-L1 staining on metastatic and postchemotherapy viable testicular GCTs as a surrogate for potential benefit for immunotherapy targeting the PD-1/PD-L1 axis. METHODS: Ethics research committee approval for this retrospective study was obtained by four participating institutions (CHU de Québec, St. Joseph's Health Care, Halifax Health Science Centre, Johannes Gutenberg University). Patients with viable metastatic testicular GCTs pathology samples were included. Patients with pure teratoma were excluded. PD-L1 staining with the 22C3 clone was evaluated on samples with >100 viable tumor cells using the combined positive score (CPS). RESULTS: From 51 patients identified at participating institutions, 24 postchemotherapy and 18 chemotherapy-naive metastatic samples were available for PD-L1 staining, with 9 matched prechemotherapy samples and 7 matched orchiectomy pathology samples, respectively. The median CPS score was 55.6 (IQR 16-100) for all metastatic samples, 44.9 (IQR 13-100) for postchemotherapy metastatic samples, and 68.8 (IQR 38-100) for chemotherapy-naïve metastatic samples, with the median number of viable tumor cells at 545, 500, and 550, respectively. Differences were not significant between chemotherapy-naïve and postchemotherapy samples (P = 0.07), though among non-seminoma GCT metastatic samples, CPS scores were significantly lower postchemotherapy (P = 0.02). Significant differences among postchemotherapy metastatic tumors were also seen according to predominant subtype, with lower CPS scores for predominant yolk sac and higher values for predominant seminoma and choriocarcinoma. In 7 patients with matched specimens pre- and postchemotherapy, a significant increase in CPS was observed for seminoma (26.7 vs. 81.7, P = 0.045), but not nonseminoma GCTs. Comparing all chemotherapy naïve-samples, PD-L1 expression was higher in metastatic samples versus testicular samples (mean CPS 68.8 vs. 39.8, P = 0.02). This was also seen in matched chemotherapy-naïve samples (mean CPS 77.9 vs. 33.1, P = 0.01). CONCLUSION: Our results suggest that most patients with refractory GCTs postchemotherapy will not benefit from PD-1/PD-L1 immunotherapy. However, the high PD-L1 expression in patients with predominant or pure seminoma post-chemotherapy suggests this may represent a subgroup for whom further trials may be considered.

Interassay and interobserver comparability study of four programmed death-ligand 1 (PD-L1) immunohistochemistry assays in triple-negative breast cancer.

Different immunohistochemical programmed death-ligand 1 (PD-L1) assays and scorings have been reported to yield variable results in triple-negative breast cancer (TNBC). We compared the analytical concordance and reproducibility of four clinically relevant PD-L1 assays assessing immune cell (IC) score, tumor proportion score (TPS), and combined positive score (CPS) in TNBC. Primary TNBC resection specimens (n = 104) were stained for PD-L1 using VENTANA SP142, VENTANA SP263, DAKO 22C3, and DAKO 28-8. PD-L1 expression was scored according to guidelines on virtual whole slide images by four trained readers. The mean PD-L1 positivity at IC-score ≥1% and CPS ≥1 ranged between 53% and 75% with the highest positivity for SP263 and comparable levels for 22C3, 28-8, and SP142. Inter-assay agreement was good between 28-8 and 22C3 across all scores and cut-offs (kappa 0.68-0.74) and for both assays with SP142 at IC-score ≥1% and CPS ≥1 (kappa 0.61-0.67). The agreement between SP263 and all other assays was substantially lower for all scores. Inter-reader agreement for each assay was good to excellent for IC-score ≥1% (kappa 0.73-0.78) and CPS ≥1 (kappa 0.68-0.74), fair to good for CPS ≥10 (kappa 0.52-0.67) and TPS ≥1% (kappa 0.53-0.72). The percentage of overlapping cases in the positive/negative category was >90% between IC-score ≥1% and CPS ≥1 but below when comparing IC-score ≥1% with CPS ≥10. We demonstrate an overall good inter-reader agreement for all PD-L1 assays in TNBC along with assay specific differences in positivity and concordances, which may aid to select the right test strategy in routine diagnostics.

Multimodal Deep Learning for Prognosis Prediction in Renal Cancer.

BACKGROUND: Clear-cell renal cell carcinoma (ccRCC) is common and associated with substantial mortality. TNM stage and histopathological grading have been the sole determinants of a patient's prognosis for decades and there are few prognostic biomarkers used in clinical routine. Management of ccRCC involves multiple disciplines such as urology, radiology, oncology, and pathology and each of these specialties generates highly complex medical data. Here, artificial intelligence (AI) could prove extremely powerful to extract meaningful information to benefit patients. OBJECTIVE: In the study, we developed and evaluated a multimodal deep learning model (MMDLM) for prognosis prediction in ccRCC. DESIGN SETTING AND PARTICIPANTS: Two mixed cohorts of non-metastatic and metastatic ccRCC patients were used: (1) The Cancer Genome Atlas cohort including 230 patients and (2) the Mainz cohort including 18 patients with ccRCC. For each of these patients, we trained the MMDLM on multiscale histopathological images, CT/MRI scans, and genomic data from whole exome sequencing. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Outcome measurements included Harrell's concordance index (C-index) and also various performance parameters for predicting the 5-year survival status (5YSS). Different visualization techniques were used to make our model more transparent. RESULTS: The MMDLM showed great performance in predicting the prognosis of ccRCC patients with a mean C-index of 0.7791 and a mean accuracy of 83.43%. Training on a combination of data from different sources yielded significantly better results compared to when only one source was used. Furthermore, the MMDLM's prediction was an independent prognostic factor outperforming other clinical parameters. INTERPRETATION: Multimodal deep learning can contribute to prognosis prediction in ccRCC and potentially help to improve the clinical management of this disease. PATIENT SUMMARY: An AI-based computer program can analyze various medical data (microscopic images, CT/MRI scans, and genomic data) simultaneously and thereby predict the survival time of patients with renal cancer.

Nivolumab Reduces PD1 Expression and Alters Density and Proliferation of Tumor Infiltrating Immune Cells in a Tissue Slice Culture Model of Renal Cell Carcinoma.

BACKGROUND: In the treatment of clear cell renal cell carcinoma (ccRCC), nivolumab is an established component of the first-line therapy with a favorable impact on progression free survival and overall survival. However, treatment-related adverse effects occur and, to date, there is no approved predictive biomarker for patient stratification. Thus, the aim of this study was to establish an ex vivo tissue slice culture model of ccRCC and to elucidate the impact of nivolumab on tumor infiltrating immune cells. METHODS: Fresh tumor tissue of ccRCC was treated with the immune checkpoint inhibitor nivolumab using ex vivo tissue slice culture (TSC). After cultivation, tissue slices were formalin-fixed, immunohistochemically stained and analyzed via digital image analysis. RESULTS: The TSC model was shown to be suitable for ex vivo pharmacological experiments on intratumoral immune cells in ccRCC. PD1 expression on tumor infiltrating immune cells was dose-dependently reduced after nivolumab treatment (p < 0.01), whereas density and proliferation of tumor infiltrating T-cells and cytotoxic T-cells were inter-individually altered with a remarkable variability. Tumor cell proliferation was not affected by nivolumab. CONCLUSIONS: This study could demonstrate nivolumab-dependent effects on PD1 expression and tumor infiltrating T-cells in TSC of ccRCC. This is in line with results from other scientific studies about changes in immune cell proliferation in peripheral blood in response to nivolumab. Thus, TSC of ccRCC could be a further step to personalized medicine in terms of testing the response of individual patients to nivolumab.

Exploiting Gangliosides for the Therapy of Ewing's Sarcoma and H3K27M-Mutant Diffuse Midline Glioma.

The ganglioside GD2 is an important target in childhood cancer. Nevertheless, the only therapy targeting GD2 that is approved to date is the monoclonal antibody dinutuximab, which is used in the therapy of neuroblastoma. The relevance of GD2 as a target in other tumor entities remains to be elucidated. Here, we analyzed the expression of GD2 in different pediatric tumor entities by flow cytometry and tested two approaches for targeting GD2. H3K27M-mutant diffuse midline glioma (H3K27M-mutant DMG) samples showed the highest expression of GD2 with all cells strongly positive for the antigen. Ewing's sarcoma (ES) samples also showed high expression, but displayed intra- and intertumor heterogeneity. Osteosarcoma had low to intermediate expression with a high percentage of GD2-negative cells. Dinutuximab beta in combination with irinotecan and temozolomide was used to treat a five-year-old girl with refractory ES. Disease control lasted over 12 months until a single partially GD2-negative intracranial metastasis was detected. In order to target GD2 in H3K27M-mutant DMG, we blocked ganglioside synthesis via eliglustat, since dinutuximab cannot cross the blood-brain barrier. Eliglustat is an inhibitor of glucosylceramide synthase, and it is used for treating children with Gaucher's disease. Eliglustat completely inhibited the proliferation of primary H3K27M-mutant DMG cells in vitro. In summary, our data provide evidence that dinutuximab might be effective in tumors with high GD2 expression. Moreover, disrupting the ganglioside metabolism in H3K27M-mutant DMG could open up a new therapeutic option for this highly fatal cancer.

Safety and Activity of the Combination of Ceritinib and Dasatinib in Osteosarcoma.

Osteosarcoma (OS) is the second most common cause of cancer-related death in pediatric patients. The insulin-like growth factor (IGF) pathway plays a relevant role in the biology of OS but no IGF targeted therapies have been successful as monotherapy so far. Here, we tested the effect of three IGF specific inhibitors and tested ceritinib as an off-target inhibitor, alone or in combination with dasatinib, on the proliferation of seven primary OS cells. Picropodophyllin, particularly in combination with dasatinib and the combination ceritinib/dasatinib were effective in abrogating the proliferation. The ceritinib/dasatinib combination was applied to the primary cells of a 16-year-old girl with a long history of lung metastases, and was more effective than cabozantinib and olaparib. Therefore, the combination was used to treat the patient. The treatment was well tolerated, with toxicity limited to skin rush and diarrhea. A histopathological evaluation of the tumor after three months of therapy indicated regions of high necrosis and extensive infiltration of macrophages. The extension of the necrosis was proportional to the concentration of dasatinib and ceritinib in the area, as analysed by an ultra performance liquid chromatography-tandem mass spectrometer (UPLC-MS/MS). After the cessation of the therapy, radiological analysis indicated a massive growth of the patient's liver metastases. In conclusion, these data indicate that the combination of ceritinib/dasatinib is safe and may be used to develop new therapy protocols.