High PTX3 expression is associated with a poor prognosis in diffuse large B‐cell lymphoma

Abstract Tumor‐associated macrophages (TAMs) are associated with a poor prognosis of diffuse large B‐cell lymphoma (DLBCL). As macrophages are heterogeneous, the immune polarization and their pathological role warrant further study. We characterized the microenvironment of DLBCL by immunohistochemistry in a training set of 132 cases, which included 10 Epstein–Barr virus‐encoded small RNA (EBER)‐positive and five high‐grade B‐cell lymphomas, with gene expression profiling in a representative subset of 37 cases. Diffuse large B‐cell lymphoma had a differential infiltration of TAMs. The high infiltration of CD68 (pan‐macrophages), CD16 (M1‐like), CD163, pentraxin 3 (PTX3), and interleukin (IL)‐10‐positive macrophages (M2c‐like) and low infiltration of FOXP3‐positive regulatory T lymphocytes (Tregs) correlated with poor survival. Activated B cell‐like DLBCL was associated with high CD16, CD163, PTX3, and IL‐10, and EBER‐positive DLBCL with high CD163 and PTX3. Programmed cell death‐ligand 1 positively correlated with CD16, CD163, IL‐10, and RGS1. In a multivariate analysis of overall survival, PTX3 and International Prognostic Index were identified as the most relevant variables. The gene expression analysis showed upregulation of genes involved in innate and adaptive immune responses and macrophage and Toll‐like receptor pathways in high PTX3 cases. The prognostic relevance of PTX3 was confirmed in a validation set of 159 cases. Finally, in a series from Europe and North America (GSE10846, R‐CHOP‐like treatment, n = 233) high gene expression of PTX3 correlated with poor survival, and moderately with CSF1R, CD16, MITF, CD163, MYC, and RGS1. Therefore, the high infiltration of M2c‐like immune regulatory macrophages and low infiltration of FOXP3‐positive Tregs is associated with a poor prognosis in DLBCL, for which PTX3 is a new prognostic biomarker.


| INTRODUC TI ON
Diffuse large B-cell lymphoma accounts for 30%-40% of newly diagnosed non-Hodgkin lymphomas. 1 Gene expression profiling has classified DLBCL into the GCB subtype, which is associated with a better prognosis, and ABC/non-GCB subtype, which has a more aggressive clinical evolution. 1 The poorer prognosis of patients with the ABC/non-GCB type is partly associated with the high infiltration of macrophages and the constitutive activation of the NF-κB pathway. [2][3][4] The macrophage lineage is heterogeneous. 5 Classical or M1 macrophages are potent effector cells with pro-inflammatory and antitumoral functions. 6 However, M2 macrophages are immunosuppressive and protumoral and are stratified into the M2a, M2b, and M2c subtypes. 7 M2c macrophages have immune regulatory functions, express CD163, PTX3, and IL-10 markers, and induce Tregs. [5][6][7][8] Tumor-associated macrophages are also named M2d and are abundant in solid cancers, such as gastric, ovarian, breast, and lung adenocarcinomas. [5][6][7][8] Although M1-like TAMs enhance antitumoral host immune responses, M2-like TAMs have been implicated in tumor progression, metastasis, resistance to therapy, angiogenesis, and immune suppression. [5][6][7][8] Pentraxin 3, also known as TNF-inducible gene 14 protein (TSG- 14), is a protein that contributes to the regulation of innate resistance to pathogens, inflammatory reactions, and the clearance of self-components. Pentraxin 3 regulates the inflammatory activity of macrophages and it is expressed by macrophages with M2-like polarization, namely, the M2c-like subtype. [5][6][7][8] In the context of IL-10 stimulation, B lymphocytes acquire regulatory properties. 9,10 Pentraxin 3 makes a crucial contribution to tumor inflammation and is highly expressed in liposarcomas 11 and lung 12 and pancreatic carcinoma 13,14 ; however, its role in hematolymphoid neoplasia remains unclear.
Due to the importance of TAMs in the pathogenesis of solid and hematolymphoid neoplasia, the role of the IL-10 molecule in the regulation of host immune responses and immune checkpoints, the targeting of TAMs and the IL-10 regulatory pathway is an important strategy for DLBCL therapy. 15,16,17 In this study, we undertook IHC and a gene expression analysis of DLBCL samples collected from patients receiving R-CHOP therapy to investigate the role of macrophages and Tregs in the pathogenesis of DLBCL and elucidate their impact on clinical outcomes.
The results obtained revealed that PTX3 is a powerful marker that identifies patients with different survival outcomes.

| Patients and samples
The training set comprised 132 cases of DLBCL diagnosed according to the 2016 WHO criteria. 17  Detailed information is shown in Tables 1, 2, and S1. Staging maneuvers and the assessment of treatment responses were standard.
The median age of patients was 69 years (range, 14-97 years) and the male / female ratio was 1.54. The IPI was retrospectively assessed in 106 patients (80%): low risk (37,34 PTX3 correlated with poor survival, and moderately with CSF1R, CD16, MITF, CD163, MYC, and RGS1. Therefore, the high infiltration of M2c-like immune regulatory macrophages and low infiltration of FOXP3-positive Tregs is associated with a poor prognosis in DLBCL, for which PTX3 is a new prognostic biomarker.

K E Y W O R D S
diffuse large B-cell lymphoma, IL-10, PD-L1, CD163, PTX3 Validation of the PTX3 gene marker in an independent series of DLBCL from Europe and North America was undertaken in GSE10846, which is publicly available in the NCBI database.
Digital image quantification using Fiji software was undertaken to assess the total percentage of positive cells in the microenvironment as previously described. 19,20 In summary, a large representative area was digitalized and the number of DAB-positive pixels was identified in the blue stack. The percentage of positive cells was calculated as follows: percentage = ([positive pixels / all pixels] × 100).

| Gene expression analysis
Gene expression profiling of a representative set of 37 cases was undertaken using RNA extracted from FFPE samples. The nCounter Immuno-oncology and Lymph2Cx assay panels were used (NanoString Technology). Housekeeping gene normalization was calculated using the log 2 ((normalData[,i]/hkGeomMeans[i])) formula.

| Statistical analysis
All statistical analyses were undertaken using SPSS software (version 26; IBM). The χ 2 and/or Fisher's exact tests and the Mann-Whitney U test were used for group comparisons, and the Kaplan-Meier and logrank tests and Cox's regression analysis for survival analyses. Overall survival was defined from the date of diagnosis to the last contact date.
Progression-free survival was defined from the date of diagnosis to disease progression. Bivariate correlation was carried out using Pearson and Spearman's tests. The significance level was set at .05. 24-29

| Clinical and histological features of patients in the training set
Detailed information is shown in Tables 1, 2, and S1. The most relevant histological features of this series were as follows: DLBCL was positive for CD5 in 9.3% of cases, BCL2 in 69.0%, and EBER in 7.6%, and RGS1 was highly expressed in 59.3% of cases. The cell-of-origin analysis based on the Hans classifier showed non-GCB in 63.6% of cases.  (Table 1). Also, in the training set, we identified 10 cases that were positive for EBV by in situ hybridization. Those cases fell within the WHO category of EBV-positive DLBCL NOS.

| Distribution of markers in DLBCL and relationships with histopathological features in the training set
The frequencies of all markers are shown in Table 3, including means  (Table S3). Note: The immune microenvironment markers were initially evaluated as an ordinal variable a 0, 1+, 2+, and 3+ as <1%, 1%-5%, 5%-20%, and >20%, respectively, under the optical microscope. After digitalization the percentage of positive cells was quantified using Fiji software. Then, the best cut-off for the overall survival (OS) was found from the quantitative data (ie, the most significant P value). Figure 1 shows immunohistochemical images of the different immune markers with the evaluation reference under the microscope and a characteristic image of high values for each marker that were associated with poor OS.

TA B L E 3
Distribution of markers in the series of patients with diffuse large B-cell lymphoma (training set) The markers were also correlated with the double-expressor, MYD88 L265P mutation, and cell-of-origin in DLBCL NOS (Tables S3-S5).

| Relationships with OS and PFS in the training set
The best (ie, optimal) cut-off for each marker was defined based on the P value in the log-rank test for OS. On average, between all markers, 31.1% of cases were in the poor prognosis group (Table 3).
Detailed information is shown in Table 3. Overall survival figures are shown in Figure 2. In the univariate analysis, the markers associated with poor OS were macrophages with high expression of CD68, CD16, MITF, CD163, PTX3, and IL-10, and low infiltration of FOXP3+ Tregs. In the multivariate analysis of all of these markers and IPI, only PTX3 and IPI retained their significance (Table 5). Therefore, PTX3 was the most relevant marker in this model in addition to IPI.
Similar results were obtained for PFS (Table 5 and Figure S1). In the univariate analysis, the high expression of CD68, CD16, CD163, PTX3, and IL-10 correlated with unfavorable PFS.
The analysis was repeated for all markers stratifying for the DLBCL NOS, EBER+ DLBCL, and HGBCL subtypes ( Figure S1 and Table 6). The results showed that the previous findings in DLBCL were also confirmed in the DLBCL NOS group.

| Gene expression analysis in the training set
The results of the gene expression analysis are shown in Figure 3 and Table S7.  Abbreviations: GCB, germinal center B cell-like; IL-10, interleukin-10; MITF, microphthalmia transcription factor; PTX3, pentraxin 3; RGS1, regulator of G-protein signaling 1.  (Table S7). Genes were also annotated using the STRING database of known and predicted protein-protein interactions. Interactions included direct (physical) and indirect (functional) associations.
In Figure 3, the gene expression is shown using a relative  Abbreviations: -, no data; CI, confidence interval; HR, hazard ratio; OS, overall survival; PFS, progression-free survival.

| Evaluation of the PTX3 marker in the validation set
The correlation between PTX3 and the survival outcomes of patients was validated in an independent series from Tokai University Hospital ( Figure 4). The clinicopathologic characteristics of this validation series are shown in  Figure S2 (hazard risk = 1.8, P = .004; 95% CI, 1.2-2.8). In the Figure S2A-B, the analysis also included the stratification for EBER-and EBER+ DLBCL cases, and OS (A) and PFS (B).

| Validation of PTX3 marker in an independent series
The publicly available GSE10846 DLBCL gene expression dataset was used to validate the association of PTX3 and poor OS of pa-  (Table 7).

| D ISCUSS I ON
In this study, we showed that:  Tumor-associated macrophages are abundant in solid can- In comparisons with GCB, non-GCB PTX3 high was associated with a poorer prognosis, with a hazard risk of 3.9 (P =.007). Although a previous study reported that PTX3 was associated with a poor prognosis in pancreatic carcinoma, 13 this is the first study to show this relationship in hematological malignancies.
We identified the IL-10 marker. Interleukin-10 is a major immune regulatory cytokine that acts on many cells in  We have recently described the importance of PD-L1, CSF1R, TNFAIP8, and CASP8 in the pathogenesis and prognosis of DLBCL, and shown how systems biology can help improve the understanding of disease mechanisms. [59][60][61][62][63][64][65] In this research, we showed how PD-L1 correlated with CD16, RGS1, and IL-10. The relationship with PTX3 will be analyzed in future publications. Finally, we used the gene expression data of the GSE10846 dataset, which included 233 cases of DLBCL patients treated with a R-CHOP-like therapy. Using this independent series, we confirmed that high gene expression of PTX3 correlated with poor OS of patients.
In conclusion, we showed that the M2c immune regulatory pathway is associated with an unfavorable prognosis in de novo DLBCL.
These results provide novel insights into the pathogenesis of DLBCL with applicability to current and future clinical trials.

ACK N OWLED G M ENTS
This work was funded in part by grants KAKENI 24590430, 15K19061, and 18K15100 by the Japan Society for the Promotion of Science