An Autophagy-Related Gene Signature can Better Predict Prognosis and Resistance in Diffuse Large B-Cell Lymphoma

Prediction significance of autophagy-related genes in survival probability and drug resistance in diffuse large B-cell lymphoma

BACKGROUND/AIMS

Diffuse large B-cell lymphoma (DLBCL), the most common subtype of non-Hodgkin lymphoma, has significant prognostic heterogeneity. This study aimed to generate a prognostic prediction model based on autophagy-related genes for DLBCL patients.

METHODS

Utilizing bioinformatics techniques, we analyzed the clinical information and transcriptome data of DLBCL patients from the Gene Expression Omnibus (GEO) database. Through unsupervised clustering, we identified new autophagy-related molecular subtypes and pinpointed differentially expressed genes (DEGs) between these subtypes. Based on these DEGs, a prognostic model was constructed using Cox and Lasso regression. The effectiveness, accuracy, and clinical utility of this prognostic model were assessed using numerous independent validation cohorts, survival analyses, receiver operating characteristic (ROC) curves, multivariate Cox regression analysis, nomograms, and calibration curves. Moreover, functional analysis, immune cell infiltration, and drug sensitivity analysis were performed.

RESULTS

DLBCL patients with different clinical characterizations (age, molecular subtypes, ECOG scores, and stages) showed different expression features of autophagy-related genes. The prediction model was constructed based on the eight autophagy-related genes (ADD3, IGFBP3, TPM1, LYZ, AFDN, DNAJC10, GLIS3, and CCDC102A). The prognostic nomogram for overall survival of DLBCL patients incorporated risk level, stage, ECOG scores, and molecular subtypes, showing excellent agreement between observed and predicted outcomes. Differences were noted in the proportions of immune cells (native B cells, Treg cells, CD8+ T cell, CD4+ memory activated T cells, gamma delta T cells, macrophages M1, and resting mast cells) between high-risk and low-risk groups. LYZ and ADD3 exhibited correlations with drug resistance to most chemotherapeutic drugs.

CONCLUSIONS

This study established a novel prognostic assessment model based on the expression profile of autophagy-related genes and clinical characteristics of DLBCL patients, explored immune infiltration and predicted drug resistance, which may guide precise and individualized immunochemotherapy regimens.

Autophagy-Related Gene Signature Highlights Metabolic and Immunogenic Status of Malignant Cells in Non-Small Cell Lung Cancer Adenocarcinoma

Simple Summary

The role of autophagy in lung cancers is still controversial, mainly because the visualization of autophagy levels in patients remains challenging. One interesting approach consists of studying autophagy at the transcriptomic level. In this line, many transcriptomics analyses performed on autophagy genes focused on the discovery of new biomarkers to predict the efficiency of antitumor therapies. However, the majority of these studies were based on global transcriptomic analysis of the whole tumor microenvironment, and few investigations have been performed on malignant cells themselves. The goal of this study was not to determine another new predictive signature based on autophagy-related genes. Instead, we investigated the expression of autophagy genes to understand the involvement of this process in lung cancer homeostasis. Specifically, we discovered a new autophagy signature that correlates with the metabolic and immunogenic status of malignant cells, supporting the relationship between autophagy and tumor growth in lung cancer patients.

Abstract

Autophagy is a self-degradative mechanism involved in many biological processes, including cell death, survival, proliferation or migration. In tumors, autophagy plays an important role in tumorigenesis as well as cancer progression and resistance to therapies. Usually, a high level of autophagy in malignant cells has been associated with tumor progression and poor prognostic for patients. However, the investigation of autophagy levels in patients remains difficult, especially because quantification of autophagy proteins is challenging in the tumor microenvironment. In this study, we analyzed the expression of autophagy genes in non-small cell lung (NSCLC) cancer patients using public datasets and revealed an autophagy gene signature for proliferative and immune-checkpoint-expressed malignant cells in lung adenocarcinoma (LUAD). Analysis of autophagy-related gene expression profiles in tumor and adjacent tissues revealed differential signatures, namely signature A (23 genes) and signature B (12 genes). Signature B correlated with a bad prognosis and poor overall and disease-specific survival. Univariate and multivariate analyses revealed that this signature was an independent factor for prognosis. Moreover, patients with high expression of signature B exhibited more genes related to proliferation and fewer genes related to immune cells or immune response. The analysis of datasets from sorted fresh tumor cells or single cells revealed that signature B is predominantly represented in malignant cells, with poor expression in pan-immune population or in fibroblast or endothelial cells. Interestingly, autophagy was increased in malignant cells exhibiting high levels of signature B, which correlated with an elevated expression of genes involved in cell proliferation and immune checkpoint signaling. Taken together, our analysis reveals a novel autophagy-based signature to define the metabolic and immunogenic status of malignant cells in LUAD.