CD20highCD138low tumor-infiltrating lymphocytes predominantly related to cytokine‒cytokine receptor interactions are associated with favorable outcomes in neuroblastoma patients.

Recent advances have revealed that the role of the immune system is prominent in the antitumor response. In the present study, it is aimed to provide an expression profile of tumor-infiltrating lymphocytes (TILs), including mature B cells, plasma cells, and their clinical relevance in neuroblastoma. The expression of CD20 and CD138 was analyzed in the Cangelosi786 dataset (n = 769) as a training dataset and in our cohort (n = 120) as a validation cohort. CD20 high expression was positively associated with favorable overall survival (OS) and event-free survival (EFS) (OS: P < 0.001; EFS: P < 0.001) in the training dataset, whereas CD138 high expression was associated with poor OS and EFS (OS: P < 0.001; EFS: P < 0.001) in both the training and validation datasets. Accordingly, a combined pattern of CD20 and CD138 expression was developed, whereby neuroblastoma patients with CD20highCD138low expression had a consistently favorable OS and EFS compared with those with CD20lowCD138high expression in both the training and validation cohorts (P < 0.0001 and P < 0.01, respectively). Examination of potential molecular functions revealed that signaling pathways, including cytokine‒cytokine receptor interactions, chemokine, and the NF-kappa B signaling pathways, were involved. Differentially expressed genes, such as BMP7, IL7R, BIRC3, CCR7, CXCR5, CCL21, and CCL19, predominantly play important roles in predicting the survival of neuroblastoma patients. Our study proposes that a new combination of CD20 and CD138 signatures is associated with neuroblastoma patient survival. The related signaling pathways reflect the close associations among the number of TILs, cytokine abundance and patient outcomes and provide therapeutic insights into neuroblastoma.

Cross-cohort analysis identified an immune checkpoint-based signature to predict the clinical outcomes of neuroblastoma.

BACKGROUND: Neuroblastoma (NB) places a substantial health burden on families worldwide. This study aimed to develop an immune checkpoint-based signature (ICS) based on the expression of immune checkpoints to better assess patient survival risk and potentially guide patient selection for immunotherapy of NB. METHODS: Immunohistochemistry integrated with digital pathology was used to determine the expression levels of 9 immune checkpoints in 212 tumor tissues used as the discovery set. The GSE85047 dataset (n=272) was used as a validation set in this study. In the discovery set, the ICS was constructed using a random forest algorithm and confirmed in the validation set to predict overall survival (OS) and event-free survival (EFS). Kaplan-Meier curves with a log-rank test were drawn to compare the survival differences. A receiver operating characteristic (ROC) curve was applied to calculate the area under the curve (AUC). RESULTS: Seven immune checkpoints, including PD-L1, B7-H3, IDO1, VISTA, T-cell immunoglobulin and mucin domain containing-3 (TIM-3), inducible costimulatory molecule (ICOS) and costimulatory molecule 40 (OX40), were identified as abnormally expressed in NB in the discovery set. OX40, B7-H3, ICOS and TIM-3 were eventually selected for the ICS model in the discovery set, and 89 patients with high risk had an inferior OS (HR 15.91, 95% CI 8.87 to 28.55, p<0.001) and EFS (HR 4.30, 95% CI 2.80 to 6.62, p<0.001). Furthermore, the prognostic value of the ICS was confirmed in the validation set (p<0.001). Multivariate Cox regression analysis demonstrated that age and the ICS were independent risk factors for OS in the discovery set (HR 6.17, 95% CI 1.78 to 21.29 and HR 1.18, 95% CI 1.12 to 1.25, respectively). Furthermore, nomogram A combining the ICS and age demonstrated significantly better prognostic value than age alone in predicting the patients' 1-year, 3-year and 5-year OS in the discovery set (1 year: AUC, 0.891 (95% CI 0.797 to 0.985) vs 0.675 (95% CI 0.592 to 0.758); 3 years: 0.875 (95% CI 0.817 to 0.933) vs 0.701 (95% CI 0.645 to 0.758); 5 years: 0.898 (95% CI 0.851 to 0.940) vs 0.724 (95% CI 0.673 to 0.775), respectively), which was confirmed in the validation set. CONCLUSIONS: We propose an ICS that significantly differentiates between low-risk and high-risk patients, which might add prognostic value to age and provide clues for immunotherapy in NB.

Clinical actionability of triaging DNA mismatch repair deficient colorectal cancer from biopsy samples using deep learning.

BACKGROUND: We aimed to develop a deep learning (DL) model to predict DNA mismatch repair (MMR) status in colorectal cancers (CRC) based on hematoxylin and eosin-stained whole-slide images (WSIs) and assess its clinical applicability. METHODS: The DL model was developed and validated through three-fold cross validation using 441 WSIs from the Cancer Genome Atlas (TCGA) and externally validated using 78 WSIs from the Pathology AI Platform (PAIP), and 355 WSIs from surgical specimens and 341 WSIs from biopsy specimens of the Sun Yet-sun University Cancer Center (SYSUCC). Domain adaption and multiple instance learning (MIL) techniques were adopted for model development. The performance of the models was evaluated using the area under the receiver operating characteristic curve (AUROC). A dual-threshold strategy was also built from the surgical cohorts and validated in the biopsy cohort. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), F1-score, and the percentage of patients avoiding IHC testing were evaluated. FINDINGS: The MIL model achieved an AUROC of 0·8888±0·0357 in the TCGA-validation cohort, 0·8806±0·0232 in the PAIP cohort, 0·8457±0·0233 in the SYSUCC-surgical cohort, and 0·7679±0·0342 in the SYSUCC-biopsy cohort. A dual-threshold triage strategy was used to rule-in and rule-out dMMR patients with remaining uncertain patients recommended for further IHC testing, which kept sensitivity higher than 90% and specificity higher than 95% on deficient MMR patient triage from both the surgical and biopsy specimens, result in more than half of patients avoiding IHC based MMR testing. INTERPRETATION: A DL-based method that could directly predict CRC MMR status from WSIs was successfully developed, and a dual-threshold triage strategy was established to minimize the number of patients for further IHC testing. FUNDING: The study was funded by the National Natural Science Foundation of China (82073159, 81871971 and 81700576), the Natural Science Foundation of Guangdong Province (No. 2021A1515011792 and No.2022A1515012403) and Medical Scientific Research Foundation of Guangdong Province of China (No. A2020392).

Clinical Significance of a CD3/CD8-Based Immunoscore in Neuroblastoma Patients Using Digital Pathology.

Background: Infiltrating immune cells have been reported as prognostic markers in many cancer types. We aimed to evaluate the prognostic role of tumor-infiltrating lymphocytes, namely CD3+ T cells, CD8+ cytotoxic T cells and memory T cells (CD45RO+), in neuroblastoma. Patients and Methods: Immunohistochemistry was used to determine the expression of CD3, CD8 and CD45RO in the tumor samples of 244 neuroblastoma patients. We then used digital pathology to calculate the densities of these markers and derived an immunoscore based on such densities. Results: Densities of CD3+ and CD8+ T cells in tumor were positively associated with the overall survival (OS) and event-free survival (EFS), whereas density of CD45RO+ T cells in tumor was negatively associated with OS but not EFS. An immunoscore with low density of CD3 and CD8 (CD3-CD8-) was indictive of a greater risk of death (hazard ratio 6.39, 95% confidence interval 3.09-13.20) and any event (i.e., relapse at any site, progressive disease, second malignancy, or death) (hazard ratio 4.65, 95% confidence interval 2.73-7.93). Multivariable analysis revealed that the CD3-CD8- immunoscore was an independent prognostic indicator for OS, even after adjusting for other known prognostic indicators. Conclusions: The new immunoscore based on digital pathology evaluated densities of tumor-infiltrating CD3+ and CD8+ T cells contributes to the prediction of prognosis in neuroblastoma patients.

Spatial heterogeneity of immune infiltration predicts the prognosis of nasopharyngeal carcinoma patients.

Spatial information on the tumor immune microenvironment is of clinical relevance. Here, we aimed to quantify the spatial heterogeneity of lymphocytes and cancer cells and evaluated its prognostic value in patients with nasopharyngeal carcinoma (NPC). The scanned immunohistochemistry images of 336 NPC patients from two different hospitals were used to generate cell density maps for tumor and immune cells. Then, Getis-Ord hotspot analysis, a spatial statistic method used to describe species biodiversity in ecological habitats, was applied to identify cancer, immune, and immune-cancer hotspots. The results showed that cancer hotspots were not associated with any of the studied clinical outcomes, while immune-cancer hotspots predicted worse overall survival (OS) in the training cohort. In contrast, a high immune hotspot score was significantly associated with better OS (HR 0.41, 95% CI 0.22-0.77, P = .006), disease-free survival (DFS) (HR 0.43, 95% CI 0.24-0.75, P = .003) and distant metastasis-free survival (DMFS) (HR 0.40, 95% CI 0.20-0.81, P = .011) in NPC patients in the training cohort, and similar associations were also evident in the validation cohort. Importantly, multivariate analysis revealed that the immune hotspot score remained an independent prognostic indicator for OS, DFS, and DMFS in both cohorts. We explored the spatial heterogeneity of cancer cells and lymphocytes in the tumor microenvironment of NPC patients using digital pathology and ecological analysis methods and further constructed three spatial scores. Our study demonstrates that spatial variation may aid in the identification of the clinical prognosis of NPC patients, but further investigation is needed.

Prognostic value of immune score in nasopharyngeal carcinoma using digital pathology.

BACKGROUND: Tumor-infiltrating lymphocytes have been reported as prognostic markers in tumors. We aimed to assess the prognostic value of total T cell (CD3+) density, cytotoxic T cell (CD8+) density and memory T cell (CD45RO+) density in patients with nasopharyngeal carcinoma (NPC). METHODS: The expression of CD3, CD8 and CD45RO was detected by immunohistochemistry in the training (n=221) and validation cohorts (n=115). The densities of these three markers were quantified by digital pathology both in the tumor and stroma. Then, we developed the immune score based on the density of these three markers and further analyzed its prognostic value. RESULTS: The high density of CD3+, CD8+ and CD45RO+ T cells both in the tumor and/or stroma were significantly associated with the decrease in mortality in the training cohort, respectively. High immune score predicted a prolonged overall survival (OS) (HR 0.34, 95% CI 0.18 to 0.64, p=0.001, disease-free survival (DFS) (HR 0.44, 95% CI 0.25 to 0.78, p=0.005) and distant metastasis-free survival (DMFS) (HR 0.43, 95% CI 0.21 to 0.87, p=0.018) in NPC patients. The findings were confirmed in the validation cohort. Multivariate analysis revealed that immune score remained an independent prognostic indicator for OS, DFS and DMFS. In addition, we established a nomogram with the integration of all independent variables to predict individual risk of death. CONCLUSIONS: We established an immune score model, which provides a reliable estimate of the risk of death, disease progress and distant metastasis in NPC patients.

Development and validation of an immune checkpoint-based signature to predict prognosis in nasopharyngeal carcinoma using computational pathology analysis.

BACKGROUND: Immunotherapy, especially immune checkpoint inhibition, has provided powerful tools against cancer. We aimed to detect the expression of common immune checkpoints and evaluate their prognostic values in nasopharyngeal carcinoma (NPC). METHODS: The expression of 9 immune checkpoints consistent with 13 features was detected in the training cohort (n = 208) by immunohistochemistry and quantified by computational pathology. Then, the LASSO cox regression model was used to construct an immune checkpoint-based signature (ICS), which was validated in a validation cohort containing 125 patients. RESULTS: High positive expression of PD-L1 and B7-H4 was observed in tumour cells (TCs), whereas PD-L1, B7-H3, B7-H4, IDO-1, VISTA, ICOS and OX40 were highly expressed in tumour-associated immune cells (TAICs). Eight of the 13 immune features were associated with patient overall survival, and an ICS classifier consisting of 5 features (B7-H3TAIC, IDO-1TAIC, VISTATAIC, ICOSTAIC, and LAG3TAIC) was established. Patients with high-risk scores in the training cohort had shorter overall (P < 0.001), disease-free (P = 0.002), and distant metastasis-free survival (P = 0.004), which were confirmed in the validation cohort. Multivariate analysis revealed that the ICS classifier was an independent prognostic factor. A combination of the ICS classifier and TNM stage had better prognostic value than the TNM stage alone. In addition, the ICS classifier was significantly associated with survivals in patients with high EBV-DNA load. CONCLUSIONS: We determined the expression status of nine immune checkpoints consistent with 13 features in NPC and further constructed an ICS prognostic model, which might add prognostic value to the TNM staging system.