Diffuse large B-cell lymphoma (DLBLC) is the most prevalent subtype of non-Hodgkin’s lymphoma (NHL), accounting for 30-50% of all adult NHL cases, with a median age of onset of more than 60 yrs. DLBCL has a wide variation in clinical presentation, response to treatment and prognosis due to its high heterogeneity. The 2008 World Health Organization (WHO) guidelines on the classification of lymphohematopoietic system tumors have distinguished more than 15 subtypes of DLBCL, and with the discovery of new molecular markers in recent years, the diagnostic and prognostic system of DLBCL has been continuously improved. At the same time, the International Prognostic Index (IPI) of DLBCL, which is widely used in clinical practice, is mainly derived from the chemotherapy alone era (CHOP), while the first-line treatment of DLBCL is mostly combined with immunochemotherapy. In this paper, we summarize and explain the new prognostic markers of DLBCL (DLBCL non-specific type, the same below) in recent years. Clinical prognostic indicators The most widely used prognostic system for DLBCL in clinical practice is still the IPI prognostic scoring system, which contains five clinical prognostic parameters, including age, lactate dehydrogenase (LDH), physical status (ECOG), disease stage and extra-nodal infiltration, and divides DLBCL patients into four different prognostic subgroups based on the scores, with 5-year survival ranging from 26-73%, while IPI, because IPI is widely performed in the clinic because it contains only 5 easy-to-use indicators, but since the summary of the prognostic value of the IPI system is based on DLBCL patients treated with chemotherapy alone, its prognostic value is reduced in the current era when rituximab is widely used as first-line therapy. The current change in the first-line treatment of DLBCL, with rituximab combined with CHOP (like) regimens becoming the preferred regimen for most DLBCL patients, has led more and more research teams to explore prognostic score systems that are more appropriate for the rituximab era. Among them, the modified IPI (R-IPI) prognostic scoring system, while keeping the 5 parameters of the original IPI unchanged, simplifies the original 4 risk stratification groups into 3, “very good prognosis” group (0 points), “good prognosis” group (1-2 points) and “poor prognosis” group (1-2 points). ) and the “poor prognosis” group (3-5 points), with 4-year overall survival of 94%, 79%, and 55%, respectively, which has served as a good differentiator for DLBCL in the rituximab era. By analyzing the NCCN database of 1650 DLBCL patients treated with combined immunochemotherapy from 2000-2010, the NCCN-IPI prognostic score system was proposed based on the original 5 prognostic parameters of IPI and further subdivided for age and LDH, which divided all patients into 4 risk stratification groups with 5-year OS of In addition to IPI and its modified prognostic score system, more studies have found that some clinical prognostic indicators applicable to other tumors are also clinically relevant in DLBCL patients, such as patient Glasgow score, absolute counts and ratios of peripheral blood lymphocytes and monocytes, serum free light chain and serum vitamin D levels, etc. In conclusion, clinical indicators are widely used as a prognostic score system for DLBCL because of their simplicity and ease of use, but they have limitations because they do not reflect the biological characteristics of the tumor itself. Cellular origin and genomic analysis DLBCL is a malignant tumor of B-cell origin, which is classified according to its origin as germinal center (GCB) and post-germinal center (non-GCB) or activated cell (ABC) origin. The cellular origin is considered to be closely related to disease pathogenesis and has prognostic value. The current gold standard for DLBCL cell origin detection is based on gene expression profiling (GEP), which more accurately classifies DLBCL into GCB, ABC and primary mediastinal large B-cell lymphoma, and further studies have found that GCB-DLBCL has a longer OS when treated with R-CHOP as well; a team of researchers from Canada used a digital gene expression analysis method based on 20 related A team from Canada used Lymph2Cx, a digital gene expression analysis method based on 20 related genes, to analyze the origin of DLBCL cells in paraffin-embedded tissue samples and found it to be highly accurate. Although the GEP method is more accurate, it is still not routinely performed even in developed countries such as Europe and the United States because of its high cost and poor clinical applicability due to the high specimen requirements. At present, the cell origin analysis based on immunohistochemistry technique is more widely used in clinical practice. most widely. In the pre-Rituximab era, cellular origin was considered one of the most important prognostic indicators for DLBCL patients, but with the advent of rituximab and the limitations of immunohistochemistry (IHC) itself, its prognostic value has become controversial, and several studies have even shown conflicting results. And we found by analyzing 244 DLBCL patients with different cellular origin typing systems that there were also large differences between Han, Choi, Tally and Visco-Young typing systems, and among 101 patients with primary DLBCL treated with R-CHOP (like) regimen based on Hans and Tally typing had Statistical differences in progression-free survival (PFS) were observed in 101 patients with primary DLBCL treated with the R-CHOP (like regimen) based on Hans and Tally typing, while no statistical differences were observed for any of the four cell origin typing systems on OS. The advent of high-throughput technologies, including whole-exome sequencing and deep sequencing, has also led to the identification of multiple molecular mutations, including MYD88, EZH2, CARD11, FOXO1, and single nucleotide polymorphisms, involving abnormalities in multiple signaling pathways, including BCR, NK-κB, NOTCH, Toll-like receptors, and PI3K, allowing for a better understanding of the pathogenesis of DLBCL Further studies have revealed that abnormalities in FOXO1, MYD88 and EZH2 may be associated with the prognosis of DLBCL, but further validation of large samples is needed. Novel molecular markers In recent years, with the rapid development of basic research and high-throughput technology, a variety of DLBCL-related molecular abnormalities have been gradually discovered. The discovery of these novel molecular abnormalities not only deepens the understanding of the pathogenesis of DLBCL, but also has a more important significance for the prognosis of DLBCL, and may even advance the improvement of a new typing system for DLBCL. MYC, as a proto-oncogene, plays an important role in the proliferation, apoptosis and other important pathophysiological processes of many tumors, and MYC gene rearrangement is a characteristic change in Burkitt’s lymphoma, but in recent years, studies have gradually found that the presence of MYC gene splitting can be detected in about 5-10% of patients with DLBCL using fluorescence in situ hybridization (FISH), and in-depth studies It was found that DLBCL patients with MYC gene splitting had significantly shorter survival than normal patients, with 5-year PFS and OS of 31% vs. 66% and 33% vs. 72%, respectively. BCL2 and BCL6 were considered to have some prognostic significance in the pre-Rituximab era, but their poor prognostic value was overcome with the introduction of rituximab. In recent years, with the widespread use of FISH technology, a new subtype of DLBLC with both MYC and BCL2/BCL6 gene abnormalities has been gradually discovered based on the above studies. For this DLBLC with specific molecular biological abnormalities, it is called double-hit lymphoma (DHL), and most of the clinical manifestations of DHL are highly aggressive with multiple poor prognostic factors, including Late disease stage, elevated LDH, high IPI scores, bone marrow and CNS involvement, and a median survival of less than 12 months for most patients, even after very aggressive immune-combination chemotherapy. Overexpression of MYC, BCL2 and BCL6 genes can also be detected in some DLBCL patients using the simpler and easier IHC, and the incidence is higher than that detected by genetic FISH techniques. Survival analysis also reveals that patients with MYC protein overexpression have a poorer prognosis, and if both MYC and BCL2/ BCL6 protein levels are overexpressed, then such patients become IHC-based DHL also has a poorer prognosis and shorter survival. We found that the abnormal rates of MYC, BCL2 and BCL6 based on FISH were 13.8%, 12.5% and 32.7%, respectively, while the abnormal rates based on IHC were significantly higher, 36.7%, 57.6% and 66.1%, respectively, in 250 cases of primary DLBCL, and both MYC abnormalities (FISH or/and IHC) and DHL (FISH or/and IHC) had significantly shorter survival and poorer prognosis, and further analysis revealed that abnormal amplification of gene copy number of MYC, BCL2 and BCL6 might also be one of the important factors in the prognosis of DLBCL. The expression of CD5 in DLBCL is about 5-10%, and a retrospective study found that most of the CD5+ DLBCL have more aggressive clinical manifestations and are more prone to extra-nodal and central nervous system infiltration, which is a poor prognostic factor for DLBCL. CD30 was thought to be more highly expressed in lymph nodes in mesenchymal large cell lymphoma and Hodgkin’s disease, but a recent study found that CD30 expression was present in about 25% of DLBCL patients, and these patients had a relatively better prognosis and longer survival compared to CD30-DLBCL, but further TP53 is an important pathogenic and prognostic indicator in a variety of tumors, and is present in approximately 20% of patients with DLBCL. TP53 mutations are present in about 20% of patients in DLBCL, which are potentially associated with the conversion of inert lymphomas to DLBCL and are poor prognostic factors in DLBCL, and the poor prognostic value of TP53 mutations persists even in the R-CHOP era. Other prognostic indicators EBV-associated DLBCL in the elderly would have been a DLBCL-specific type, but our study found that EBV infection in our DLBCL patients still exists in young adults and is a poor prognostic indicator by detecting EBV DNA in peripheral blood and EBER expression in lymph node specimens, while peripheral blood EBV DNA testing is a poor prognostic indicator because it can be dynamically The dynamic monitoring of peripheral blood EBV titers is also closely related to the response to treatment and prognosis of DLBCL. The prognostic value of PET/CT in DLBCL has been increasingly investigated. The prognostic value of PET/CT assessment in DLBCL is relatively clear at the initial diagnosis and after the end of treatment, while the interpretation of the results of mid-term PET/CT assessment is still controversial. Meanwhile, as an emerging technology, PET/CT still needs to be explored and improved in many aspects, including the advantages and disadvantages of different parameters such as SUVmax, ΔSUV, whole body tumor metabolic volume (WBMTV) and different interpretation methods such as visual method and Deville’s quintile method. As a highly heterogeneous lymphoid tumor, DLBCL has great differences in tumor biological characteristics, clinical manifestations, treatment response and prognosis. However, the integration of new indicators and classical IPI prognostic score system still confuses clinicians; meanwhile, the prognostic value of the original indicators will be changed due to the emergence of new drugs such as lenalidomide and Ibrutinib, etc. Therefore, the prognostic score system of DLBCL needs to be improved and perfected under the comprehensive consideration of the above factors.