High expression of PD-L1 marks the poor prognosis of pancreatic ductal adenocarcinomas (PDAC). However, the regulatory mechanism of PD-L1 remains elusive. We recently reported that cancer Forkhead box protein 3 (Cancer-FOXP3 or C-FOXP3) promoted immune evasion of PDAC by recruiting Treg cells into PDAC via upregulation of CCL5. In this study, we confirmed that PD-L1 was overexpressed in PDAC samples from two independent cohorts of patients with radical resection. Moreover, C-FOXP3 was colocalized and correlated with the expression of PD-L1 in tumor cells at the mRNA and protein levels, and this finding was confirmed by the The Cancer Genome Atlas (TCGA) database. Chromatin immunoprecipitation (ChIP) revealed that C-FOXP3 directly bound to the promoter region of PD-L1 in pancreatic cancer cells. Furthermore, overexpression of C-FOXP3 activated the luciferase reporter gene under the control of the PD-L1 promoter. However, mutation of the binding motif-a completely reversed the luciferase activity. In addition, C-FOXP3-induced upregulation of PD-L1 effectively inhibited the activity of CD8 + T cells. Based on our recent finding that the CCL-5 antibody achieved a better response to PDAC models with high C-FOXP3 levels, we further demonstrated that the PD-L1 antibody strengthened the antitumor effect of CCL-5 blockade in xenograft and orthotopic mouse models with high C-FOXP3 levels. In conclusion, C-FOXP3 directly activates PD-L1 and represents a core transcription factor that mediates the immune escape of PDAC. Combined blockade of PD-L1 and CCL-5 may provide an effective therapy for patients with PDAC that have high C-FOXP3 levels.

Background Identifying gene mutation signatures will enable a better understanding for the occurrence and development of colorectal cancer (CRC), and provide some potential biomarkers for clinical practice. Currently, however, there is still few effective biomarkers for early diagnosis and prognostic judgment in CRC patients. The purpose was to identify novel mutation signatures for the diagnosis and prognosis of CRC. Methods Clinical information of 531 CRC patients and their sequencing data were downloaded from TCGA database (training group), and 53 clinical patients were collected and sequenced with targeted next generation sequencing (NGS) technology (validation group). The relationship between the mutation genes and the diagnosis, pathological type, stage and prognosis of CRC were compared to construct signatures for CRC, and then analyzed their relationship with RNA expression, immunocyte infiltration and tumor microenvironment (TME). Results Mutations of TP53, APC, KRAS, BRAF and ATM covered 97.55% of TCGA population and 83.02% validation patients. Moreover, 57.14% validation samples and 22.06% TCGA samples indicated that patients with mucinous adenocarcinoma tended to have BRAF mutation, but no TP53 mutation. Mutations of TP53, PIK3CA, FAT4, FMN2 and TRRAP had a remarkable difference between I-II and III-IV stage patients (P < 0.0001). Besides, the combination of PIK3CA, LRP1B, FAT4 and ROS1 formed signatures for the prognosis and survival of CRC patients. The mutations of TP53, APC, KRAS, BRAF, ATM, PIK3CA, FAT4, FMN2, TRRAP, LRP1B, and ROS1 formed the signatures for predicting diagnosis and prognosis of CRC. Among them, mutation of TP53, APC, KRAS, BRAF, ATM, PIK3CA, FAT4 and TRRAP significantly reduced their RNA expression level. Stromal score, immune score and ESTIMATE score were lower in patients with TP53, APC, KRAS, PIK3CA mutation compared non-mutation patients. All the 11 gene mutations affected the distributions of immune cells. Conclusion This study constructed gene mutation signatures for the diagnosis, treatment and prognosis in CRC, and proved that their mutations affected RNA expression levels, TME and immunocyte infiltration. Our results put forward further insights into the genotype of CRC.

No abstract

Background The purpose of this study was to construct a new typing model for diffuse large B-cell lymphoma (DLBCL) patients based on the B-cell receptor (BCR) and explore its potential molecular mechanism. Methods BCR repertoire sequencing and whole-exome sequencing were performed on formalin-fixed paraffin-embedded samples from 12 DLBCL patients. Subsequently, a typing model was built with cluster analysis, and prognostic indicators between the two groups were compared to verify the typing model. Then, mutation and bioinformatics analyses were conducted to investigate the potential biomarkers of prognostic differences between the two groups. Results Based on BCR sequencing data, we divided patients into two clusters (cluster 1 and cluster 2); this classification differed from the traditional typing method (GCB and non-GCB), in which cluster 1 included some non-GCB patients. The progression-free survival (PFS), overall survival (OS), metastasis and Shannon diversity index of IGH V-J and survival after chemotherapy were significantly different (P < 0.05) between the two clusters, but no statistical significance was found between the GCB and non-GCB groups. The mutation status of 248 genes was significantly different between cluster 1 and cluster 2. Among them, FTSJ3, MAGED2, and ODF3L2 were the specific mutated genes in all patients in cluster 2, and these genes could be considered critical to the different prognoses of the two clusters of DLBCL patients. Conclusion We constructed a new typing model of DLBCL based on BCR repertoire sequencing that can better predict the survival time after chemotherapy. FTSJ3, MAGED2, and ODF3L2 may represent key genes for the difference in prognosis between the two clusters.

Liver cancer is a common malignant tumor worldwide, which is a serious threat to the health of people. We try to investigate some mutations and clinical indicators as candidate markers for the development of liver cancer through targeted region capture technology combined with next-generation sequencing. We collected peripheral blood and liver cancer tissue samples from 32 liver patients concurrently. The SeqCap EZ Prime Choice Probe was used to perform the targeted enrichment; this probe captures 1,000 known cancer-associated genes. We calculated the tumor mutation burden (TMB) for each patient. The high-frequency mutations and these relative genes were identified. Eventually, survival analysis was performed based on the mutations and clinical indicators. In 32 liver patients, a total of 29 high-frequency mutations were investigated. They were located in 25 genes, which were enriched in 9 cellular components (CCs), 6 molecular functions (MFs), and 21 biological processes (BPs). Among them, EZH2 c.1544A&gt;G and CCND1 c.839A&gt;T had the highest mutation frequency (5/32). In the protein–protein interaction (PPI) network, EZH2-DNMT3A, NOTCH1-CCND1, and ABL1-CCND1 were the top three pairs. The survival analysis showed that there were significant differences in progression-free survival (PFS) and overall survival (OS) between the Karnofsky performance score (KPS) groups. The PFS and OS in the TMB high group were higher than those in the TMB low group. OS and tumor stage had a remarkable relationship. In conclusion, EZH2 c.1544A&gt;G and CCND1 c.839A&gt;T might be potential biomarkers of liver cancer. TMB might be used as a prognosis and survival indicator of liver cancer.