Prognostic impact of biomarkers in PMBCL: rationale for early integration of immune checkpoint inhibitors

Abstract

Aim: This research aims to guide future strategies for personalized treatment of primary mediastinal large B-cell lymphoma (PMBCL), particularly to identify high-risk patients who may benefit from incorporating immune checkpoint inhibitors (ICIs) in the first-line setting.

Methods: A retrospective, single-center study included 254 newly diagnosed PMBCL patients treated with rituximab, dose-adjusted etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin (R-DA-EPOCH), rituximab, modified protocol NHL-BFM-90 (RmNHL-BFM-90), or R-DA-EPOCH combined with nivolumab. Clinical parameters, immunohistochemical markers [programmed death ligand-1 (PD-L1), programmed death-1 (PD-1), cytotoxic T-lymphocyte-associated protein 4 (CTLA-4), human leucocyte antigen (HLA)-DR, Ki-67, multiple myeloma oncogene 1 (MUM1)], molecular markers (mutations in tumor protein p53 (TP53), CD58, beta-2-microglobulin (B2M), and exportin 1 (XPO1) genes; short tandem repeats at 6p21.3 [major histocompatibility complex (MHC) class I/II], 9p24.1 (PD-L1/PD-L2), 16p13.13 [class II, MHC, transactivator gene (CIITA)]), and cytogenetic profiles [myelocytomatosis oncogene (MYC)/8q24, B-cell lymphoma 2 (BCL2)/18q21, BCL6/3q27, del17p13, and karyotype abnormalities] were analyzed.

Results: The addition of nivolumab to R-DA-EPOCH as a first-line regimen significantly improved event-free survival (EFS; P = 0.018). This study identified that adverse prognostic factors for PMBCL include allelic imbalance at specific loci 6p21.3 (MHC class I/II), 9p24.1 (PD-L1/PD-L2), and 16p13.13 (CIITA). Incorporating nivolumab into the R-DA-EPOCH regimen as a first-line therapy has shown potential in reducing adverse prognostic factors.

Conclusions: These findings suggest that high-risk patients may benefit significantly from the early incorporation of ICIs into their treatment plans.

Keywords: PMBCL; genomic instability; immune checkpoint inhibitors; loss of heterozygosity.

Figure 1

Schematic overview of the study design. PMBCL: primary mediastinal large B-cell lymphoma; R-DA-EPOCH: rituximab, dose-adjusted etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin; RmNHL-BFM-90: rituximab, modified protocol NHL-BFM-90; PD-L1: programmed death ligand-1; PD-1: programmed death-1; CTLA-4: cytotoxic T-lymphocyte-associated protein 4; HLA: human leucocyte antigen; MUM1: multiple myeloma oncogene 1; MYC: myelocytomatosis oncogene; BCL2: B-cell lymphoma 2; CCA: conventional cytogenetic analysis; CMA: chromosome microarray analysis; TP53: tumor protein p53; B2M: beta-2-microglobulin; XPO1: exportin 1; STR: short tandem repeat; MHC: major histocompatibility complex; CIITA: class II, MHC, transactivator gene

Figure 2

Kaplan-Meier survival analysis of PMBCL patients treated with R-DA-EPOCH (n = 162), RmNHL-BFM-90 (n = 69), and R-DA-EPOCH with nivolumab (n = 23). (A) The analysis showed no statistically significant differences in OS between the groups (P = 0.41); (B) PFS analysis reveals a similar trend favoring R-DA-EPOCH with nivolumab, with no statistically significant difference (P = 0.23); (C) the RFS outcomes are comparable across the three treatment groups, with no significant differences observed (P = 0.47); (D) the EFS demonstrates a statistically significant improvement in the R-DA-EPOCH with nivolumab group (P = 0.018), suggesting that the addition of nivolumab reduces adverse events such as treatment failure, relapse, and progression requiring second-line therapy or auto-HSCT. PMBCL: primary mediastinal large B-cell lymphoma; R-DA-EPOCH: rituximab, dose-adjusted etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin; RmNHL-BFM-90: rituximab, modified protocol NHL-BFM-90; OS: overall survival; PFS: progression-free survival; RFS: relapse-free survival; EFS: event-free survival; auto-HSCT: autologous hematopoietic stem cell transplantation

Figure 3

Frequency of immunohistochemical, molecular, and cytogenetic markers in patients with PMBCL. This figure illustrates the frequency distribution of negative (green, absence) and positive (yellow, presence) status of various markers in patients with PMBCL. Markers are sorted by their positive frequency, and the total number of patients analyzed for each marker is displayed on the right. PMBCL: primary mediastinal large B-cell lymphoma; BCL2: B-cell lymphoma 2; MYC: myelocytomatosis oncogene; TP53: tumor protein p53; XPO1: exportin 1; PD-L1: programmed death ligand-1; IHC: immunohistochemical; B2M: beta-2-microglobulin; HLA: human leucocyte antigen; PD-1: programmed death-1; MUM1: multiple myeloma oncogene 1; CTLA-4: cytotoxic T-lymphocyte-associated protein 4

Figure 4

Genome-wide distribution of copy number alterations and cnLOH in PMBCL samples (n = 15). This figure illustrates the chromosomal localization and frequency of genomic aberrations detected by CMA. Each vertical bar represents a distinct event across the cohort. Blue bars indicate copy number gains (amplifications or duplications), red bars denote deletions, and purple bars represent cnLOH. cnLOH: copy-neutral loss of heterozygosity; PMBCL: primary mediastinal large B-cell lymphoma; CMA: chromosome microarray analysis

Figure 5

Association of clinical, immunohistochemical, and molecular markers with poor outcomes (R/R or PR). ^* P < 0.05. R/R: relapsed or refractory; PR: partial remission; OR: odds ratios; CI: confidence intervals; TP53: tumor protein p53; IHC: immunohistochemical; HLA: human leucocyte antigen; LOH: loss of heterozygosity; XPO1: exportin 1; PD-1: programmed death-1; PD-L1: programmed death ligand-1; B2M: beta-2-microglobulin; MUM1: multiple myeloma oncogene 1; LDH: lactate dehydrogenase

Figure 6

Event-free survival based on STR profiles 9p24.1 (PD-L1/PD-L2), 16p13.13 (CIITA), 6p21.3 (HLA) and first-line therapy. This figure illustrates EFS in patients with stable STR profile versus those with AI involving 9p24.1 (PD-L1/PD-L2), 16p13.13 (CIITA), and 6p21.3 (HLA). (A) EFS of patients treated with chemotherapy (R-DA-EPOCH or RmNHL-BFM-90) stratified by STR profile stability. Patients with AI at 9p24.1 (PD-L1/PD-L2) and/or 16p13.13 (CIITA) and/or 6p21.3 (HLA) loci had significantly lower EFS compared to those with stable STR profiles (P = 0.028); (B) EFS of patients stratified by therapy type and STR profiles. Patients treated with nivolumab combined with chemotherapy (nivolumab and R-DA-EPOCH) exhibited no adverse events, achieving 100% EFS despite the presence of AI at the loci analyzed. In contrast, patients receiving chemotherapy alone demonstrated significantly reduced EFS when AI was present (P = 0.004). STR: short tandem repeat; PD-L1: programmed death ligand-1; CIITA: class II, major histocompatibility complex, transactivator gene; HLA: human leucocyte antigen; EFS: event-free survival; AI: allelic imbalance; R-DA-EPOCH: rituximab, dose-adjusted etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin; RmNHL-BFM-90: rituximab, modified protocol NHL-BFM-90

Figure 7

Overall survival in R/R PMBCL treated with nivolumab plus chemotherapy vs. chemotherapy alone. The Kaplan-Meier survival curve illustrates the OS in patients with R/R PMBCL treated with nivolumab in combination with chemotherapy (blue curve) compared to those receiving chemotherapy alone (green curve). R/R: relapsed or refractory; PMBCL: primary mediastinal large B-cell lymphoma; OS: overall survival