ABSTRACT
The molecular subtype characterized by co‐occurring MYD88 and CD79B alterations (MCD) represents a biologically distinct subset of diffuse large B‐cell lymphoma (DLBCL) with chronic active B‐cell receptor signaling and a high risk of central nervous system (CNS) involvement. The clinical impact of Bruton tyrosine kinase inhibitors (BTKi) in this subtype remains unclear. We retrospectively analyzed 155 patients with newly diagnosed DLBCL harboring genetic features consistent with the MCD subtype. At a median follow‐up of 34.1 months, the estimated 3‐year progression‐free survival (PFS) rate was 76.1%. BTKi exposure (n = 56) was associated with significantly improved PFS compared with no BTKi exposure (3‐year PFS: 93.8% vs. 66.6%, p < 0.001) and remained independently associated with improved PFS after adjustment for IPI risk (HR 0.16, p < 0.001). Overall survival did not differ significantly between groups. Notably, all 15 CNS relapse events occurred in patients who did not receive BTKi, whereas no CNS relapse was observed in the BTKi‐treated group. BTKi exposure was independently associated with a markedly reduced risk of CNS relapse (HR 0.06, p = 0.002) after adjustment for CNS‐IPI risk and CNS prophylaxis. These findings suggest that BTK inhibition may improve outcomes and mitigate CNS relapse in MCD DLBCL.
Keywords: BCR signaling, BTK inhibitor introduction, CNS relapse, diffuse large B‐cell lymphoma, survival
1. Introduction
Diffuse large B‐cell lymphoma (DLBCL) is the most common subtype of non‐Hodgkin lymphoma and remains a biologically heterogeneous disease [1, 2, 3]. Although the addition of rituximab to CHOP chemotherapy has significantly improved survival over the past 2 decades, approximately one third of patients still experience relapse or refractory disease [4, 5, 6, 7]. Clinical risk models such as the International Prognostic Index (IPI) remain widely used for prognostic stratification [8]. However, these clinical parameters primarily reflect tumor burden and host factors and do not fully capture the biological diversity underlying DLBCL.
Recent genomic studies have significantly refined the molecular classification of DLBCL [9, 10, 11, 12]. Beyond the traditional cell‐of‐origin framework, integrative genomic analyses have identified distinct genetic subtypes characterized by recurrent somatic alterations and dysregulated oncogenic signaling pathways. The probabilistic classification framework proposed by Wright et al. identified several genetically defined subtypes, including MCD, BN2, N1, and A53 [11]. These molecular subtypes exhibit different biological characteristics and clinical outcomes when treated with standard R‐CHOP therapy. In particular, the MCD subtype, characterized by frequent MYD88 and CD79B mutations, chronic active B‐cell receptor (BCR) signaling and activation of downstream NF‐κB pathways [11].
The identification of BCR‐dependent molecular subtypes has raised interest in therapeutically targeting this signaling pathway. Bruton tyrosine kinase (BTK) plays a central role in BCR signaling, and BTK inhibitor (BTKi) have demonstrated activity across several B‐cell malignancies [13]. Evidence supporting this approach has emerged from retrospective analyses of major clinical trials. In the PHOENIX trial, genetic subtype analysis suggested that patients aged ≤ 60 years with MCD or N1 subtypes treated with R‐CHOP plus ibrutinib experienced improved outcomes compared with those receiving R‐CHOP alone, whereas patients with the BN2 subtype did not derive similar benefit [14]. Notably, the number of patients within these genetic subgroups was relatively limited (MCD, n = 31; N1, n = 13), highlighting the need for further studies evaluating targeted strategies in molecularly defined populations. More recently, exploratory molecular analyses from the POLARIX study further suggested that treatment benefit may vary according to genetic subtype. Among 56 patients with MCD DLBCL identified by the LymphGen classifier, Pola‐R‐CHP was associated with numerically improved 2‐year progression‐free survival (PFS) compared with R‐CHOP (83% vs. 75%; hazard ratios [HRs] 0.62, 95% confidence intervals [CIs] 0.19–2.01), without statistical significance. Collectively, these findings support the concept that therapeutic benefit in DLBCL may depend on the underlying molecular subtype and highlight the need for biologically informed treatment strategies [15, 16]. These findings collectively suggest that therapeutic benefit in DLBCL may depend on the underlying molecular subtype. In this context, we investigated the clinical impact of BTKi exposure in a cohort of patients with LymphGen‐defined MCD subtype DLBCL.
Another important clinical challenge in DLBCL is central nervous system (CNS) relapse. Although relatively uncommon, CNS relapse is associated with extremely poor prognosis and limited therapeutic options. Several studies have suggested that specific molecular features, including MYD88 mutations and activation of BCR/NF‐κB signaling pathways, may associated with CNS events in certain DLBCL subsets [17, 18]. However, the mechanisms underlying CNS involvement remain incompletely understood, and effective strategies to prevent CNS relapse are still lacking [19, 20]. Whether targeted inhibition of BCR signaling may influence the risk of CNS relapse has not been well investigated.
In the present study, we analyzed the 155 patients with MCD subtype DLBCL and characterized the mutational landscape of this population. We further evaluated the association between BTK inhibitor exposure and clinical outcomes, including PFS, overall survival (OS), and CNS relapse. In addition, exploratory analyses were conducted according to BCR genotype groups defined by MYD88 and CD79B alterations to better understand how these genomic features may influence treatment outcomes.
2. Methods
2.1. Study Population
Between March 2019 and June 2025, a total of 648 newly diagnosed DLBCL patients underwent targeted next‐generation sequencing (481‐gene panel) at Fudan University Shanghai Cancer Center. Among the 648 patients classified by the LymphGen algorithm, MCD was the most common subtype (155/648, 23.9%), followed by BN2 (114/648, 17.6%), A53 (59/648, 9.1%), EZB (56/648, 8.6%), ST2 (32/648, 4.9%), and N1 (10/648, 1.5%); 222 patients (34.3%) were classified as other subtypes. We retrospectively analyzed 155 patients with MCD subtype DLBCL treated at Fudan University Shanghai Cancer Center. Clinical characteristics, treatment information, molecular findings, and follow‐up data were obtained from medical records. Baseline variables included age, sex, ECOG performance status, serum lactate dehydrogenase (LDH), number of extranodal sites, and the CNS involvement. Cell‐of‐origin classification was performed using the Hans immunohistochemistry algorithm to distinguish GCB from non‐GCB subtypes. MYC rearrangement status was assessed by fluorescence in situ hybridization when available. International Prognostic Index (IPI) and CNS‐IPI scores were calculated according to standard criteria. Extranodal involvement at baseline was evaluated in detail.
2.2. Treatment
Initial therapy was selected at the discretion of the treating physicians. Most patients received an R‐CHOP–based regimen. In selected cases, lenalidomide or a BTKi was incorporated during treatment. CNS prophylaxis, either intrathecal methotrexate or high‐dose methotrexate, was administered according to clinical risk assessment. Patients were classified as receiving BTK inhibitor exposure if zanubrutinib was administered in combination with frontline therapy within the first six treatment cycles, with at least one cycle including zanubrutinib. In addition, only patients who subsequently achieved complete response (CR) and continued zanubrutinib as maintenance therapy were included in the BTK inhibitor group.
Patients who received zanubrutinib only after six cycles of induction therapy, including those treated in the setting of insufficient response, salvage therapy, or maintenance following non‐CR status, were not considered part of the BTK inhibitor exposure group.
2.3. Response and Outcomes
Response was evaluated using whole‐body PET/CT with or without confirmatory contrast‐enhanced CT/MRI per Lugano 2014 criteria [21]. According to established criteria. The overall response rate included patients achieving complete or partial response. PFS was measured from treatment initiation to progression, relapse, or death. OS was calculated from treatment initiation to death from any cause. CNS progression‐free survival referred to time to CNS relapse or progression. Patients without events were censored at the date of last follow‐up.
2.4. Genomic Profiling and Molecular Classification
DNA was extracted from FFPE tumor tissues using the QIAamp FFPE Tissue Kit (Qiagen) and quantified with the Qubit dsDNA HS Assay Kit. Libraries were prepared using the KAPA HyperPrep Kit following Covaris M220 fragmentation to 200–350 bp. Targeted sequencing was performed using the Hemasalus 481‐gene panel (Geneseeq), which covers genes involved in B‐cell receptor signaling as well as rearrangement‐prone loci including MYC, BCL2, and BCL6. Sequencing was conducted on the Illumina HiSeq 4000 platform with paired‐end reads, achieving a mean target depth of approximately 1000×.
Raw sequencing data were demultiplexed and subjected to quality control to remove low‐quality reads and reads containing ambiguous bases. Clean reads were aligned to the human reference genome (hg19) using the Burrows–Wheeler Aligner (BWA). Local realignment around indels and base‐quality score recalibration were performed using the Genome Analysis Toolkit (GATK), and PCR duplicates were removed using Picard. Single‐nucleotide variants and small insertions/deletions were identified using VarScan2 with a variant allele frequency threshold of 5% for tissue samples. Structural variants were detected using DELLY and visually inspected using the Integrative Genomics Viewer (IGV).
Copy number alterations were evaluated for 103 genes included in the panel using ADTEx. However, genome‐wide chromosomal arm‐level copy number events were not systematically assessed. Therefore, molecular subtype assignment was primarily based on mutation profiles with available gene‐level copy number information and should be interpreted as mutation‐based LymphGen‐like classification.
2.5. Statistical Analysis
Categorical variables are presented as counts and percentages. Baseline characteristics were compared between the chemoimmunotherapy (CIT) and BTKi‐CIT groups. Continuous variables were summarized as median (range) and compared using the Wilcoxon rank‐sum test. Categorical variables were presented as frequencies and percentages and compared using the χ 2 test or Fisher's exact test, as appropriate. Survival probabilities were estimated using the Kaplan–Meier method and compared with the log‐rank test. Cox proportional hazards models were used to evaluate associations between BTKi exposure and PFS or OS, adjusting for IPI or CNS‐IPI risk groups. HRs with 95% CIs are reported. Because no CNS relapse events occurred in the BTKi group, Firth‐penalized Cox regression was applied for CNS relapse analyses to reduce bias from complete separation. The multivariable CNS model included BTKi exposure and CNS‐IPI category (low vs. intermediate/high). All analyses were performed using R (version 4.4.1). A two‐sided p value < 0.05 was considered statistically significant.
3. Results
3.1. Patient Characteristics
A total of 155 patients were included (Table 1). The median age was 62 years (range, 33–83), and 82 patients (52.9%) were older than 60 years. Men accounted for 56.8% of patients (n = 88). Most patients had preserved functional status, with 146 (94.2%) presenting with an ECOG performance status of 0–1. Elevated LDH was observed in 66 patients (42.6%), and 55 (35.5%) had involvement of more than one extranodal site. Among the 155 patients, bone involvement was observed in 31 cases (20.0%), followed by testicular or seminal vesicle involvement in 29 patients (18.7%) and breast involvement in 24 patients (15.5%). Other extranodal sites included kidney (9 patients, 5.8%), bone marrow (9 patients, 5.8%), adrenal gland (7 patients, 4.5%), and ovary (4 patients, 2.6%). Baseline CNS involvement was rare, documented in 2 patients (1.3%). The majority of tumors were classified as non‐GCB (N = 97, 62.6%), while 57 (36.8%) were categorized as GCB according to the Hans immunohistochemistry algorithm; 1 cases (0.6%) were unclassified. Overall, 72 patients (46.5%) fell into the low IPI risk category, whereas 23 (14.8%) were classified as high risk. Based on the CNS‐IPI, 29 patients (18.7%) were considered high risk. Among the 56 patients who received BTKi‐CIT, 52 (94.6%) were treated with zanubrutinib at the standard dose of 160 mg twice daily, 3 were treated with zanubrutinib at a reduced dose of 80 mg twice daily and one was treated with 14.8 ibrutinib. The median duration of zanubrutinib treatment was 4.5 cycles (interquartile range [IQR], 3–5 cycles), with a range of 1–8 cycles. Seven patients in our cohort received BTKi as maintenance therapy beyond the completion of frontline chemoimmunotherapy, with a median duration of 24 months (range 3–48 months). Supporting Information S1: Table S1 presented the features of patients treated with CIT and BTKi‐CIT.
TABLE 1.
Baseline characteristics of the study population (N = 155).
| Characteristic | Total (N = 155) |
|---|---|
| Age, median (range), years | 62 (33–83) |
| Age > 60 years, n (%) | 82 (52.9%) |
| Sex, n (%) | |
| Female | 67 (43.2%) |
| Male | 88 (56.8%) |
| ECOG performance status, n (%) | |
| 0–1 | 146 (94.2%) |
| ≥ 2 | 9 (5.8%) |
| Elevated LDH, n (%) | 66 (42.6%) |
| Extranodal involvement > 1 site, n (%) | 55 (35.5%) |
| CNS involvement at diagnosis, n (%) | 2 (1.3%) |
| MYC rearranged, n (%) | 11 (7.1%) |
| Cell of origin, n (%) | |
| GCB | 57 (36.8%) |
| Non‐GCB | 97 (62.6%) |
| Unknown | 1 (0.6%) |
| IPI risk group, n (%) | |
| Low | 72 (46.5%) |
| Low‐intermediate | 28 (18.1%) |
| High‐intermediate | 32 (20.6%) |
| High | 23 (14.8%) |
| CNS‐IPI risk group, n (%) | |
| Low | 71 (45.8%) |
| Intermediate | 55 (35.5%) |
| High | 29 (18.7%) |
| Frontline treatment backbone, n (%) | |
| R‐CHOP | 144 (92.9%) |
| DA‐R‐EPOCH | 5 (3.2%) |
| Pola‐R‐CHP | 6 (3.9%) |
| Received BTKi, n (%) | 56 (36.1%) |
| Received CNS prophylaxis, n (%) | 65 (41.9%) |
| Intrathecal MTX | 59 (38.1%) |
| High‐dose MTX | 7 (4.5%) |
| Consolidative radiotherapy, n (%) | 25 (16.1%) |
Abbreviations: BTKi, Bruton tyrosine kinase inhibitor; CNS, central nervous system; CNS‐IPI, Central Nervous System International Prognostic Index; DA‐R‐EPOCH, dose‐adjusted rituximab, etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin; ECOG, Eastern Cooperative Oncology Group; GCB, germinal center B‐cell–like; HD‐MTX, high‐dose methotrexate; IPI, International Prognostic Index; LDH, lactate dehydrogenase; MTX, methotrexate; Pola‐R‐CHP, polatuzumab vedotin, rituximab, cyclophosphamide, doxorubicin, and prednisone; R‐CHOP, rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone.
3.2. Genomic Alterations
Targeted sequencing of 155 DLBCL cases identified recurrent alterations in MYD88 (78.7%), CDKN2A (53.5%), CD79B (48.4%), CDKN2B (36.1%), KMT2D (26.5%), ETV6 (21.9%) and TP53 (18.7%). Additional recurrent mutations included CD58 (14.2%), and BTG2 (9.7%). Genomic landscape was presented in Figure 1. In 122 patients harboring MYD88 mutations. The mutation spectrum was overwhelmingly dominated by the canonical p. L265P hotspot mutation, which accounted for the majority of MYD88 alterations.
FIGURE 1.
Genomic landscape of recurrent alterations in 155 patients with diffuse large B‐cell lymphoma. The oncoprint illustrates the distribution of genetic alterations identified by targeted next‐generation sequencing using a 481‐gene panel. Each column represents an individual patient, and each row corresponds to a gene. The bar plot above the heatmap indicates the total number of alterations detected per patient. Genes are displayed in descending order of mutation frequency. The bar plot on the right shows the proportion of patients harboring alterations in each gene, with percentages and absolute case numbers indicated. Genetic alterations are color‐coded by type, including missense mutations, insertions/deletions, truncating mutations, splice‐site variants, copy number loss, copy number gain, gene fusions, and other alterations. Samples are arranged according to overall mutation burden.
3.3. Patient Treatment
R‐CHOP remained the predominant frontline regimen and was administered in 144 patients (92.9%). Lenalidomide was incorporated into treatment in 11 patients (7.1%). BTK inhibitors were used in 56 patients (36.1%) during induction therapy. CNS prophylaxis was administered in 65 patients (41.9%), including 59 (38.1%) who received intrathecal methotrexate and 7 (4.7%) who received high‐dose methotrexate.
3.4. Response and Survival Analysis
Among the 155 patients included in the analysis, 127 (81.9%) achieved CR, and 13 (8.4%) achieved partial response (PR), resulting in an overall response rate (ORR) of 90.3%. Stable disease (SD) was observed in 1 patient (0.6%), while progressive disease (PD) occurred in 8 patients (5.2%). Six patients (3.9%) were not evaluable for response.
At a median follow‐up of 34.1 months (95% CI, 31.5–37.5), 36 PFS events were observed. The estimated 3‐year PFS rate for the entire cohort was 76.1% (95% CI, 69.2%–83.6%). Patients receiving BTKi had significantly improved PFS compared with those who did not receive BTKi (log‐rank p < 0.001, Figure 2A). The 3‐year PFS rate was 93.8% (95% CI, 77.2%–1.0%) in the BTKi group versus 66.6% (95% CI, 57.5%–87.2%) in the non‐BTKi group. After adjustment for IPI risk group, BTKi exposure remained independently associated with improved PFS (HR, 0.16; 95% CI, 0.05–0.52; p < 0.001). For overall survival (OS), 7 deaths were recorded. The estimated 3‐year OS rate was 91.5% (95% CI, 86.5%–96.8%). Although OS was numerically higher among patients receiving BTKi (Figure 2B), the difference did not reach statistical significance (log‐rank p = 0.096). In multivariable Cox analysis adjusting for CNS‐IPI, BTKi exposure was independently associated with OS (HR, 0.22; 95% CI, 0.03–1.74; p = 0.012).
FIGURE 2.
Kaplan–Meier estimates of progression‐free survival (A) and overall survival (B) according to BTK inhibitor exposure in the entire group. BTKi, Bruton tyrosine kinase inhibitor; OS, overall survival; PFS, progression‐free survival.
Sensitivity analysis was performed in the 144 patients treated with CIT (n = 88) and BTKI‐CIT (n = 56). Improved PFS was observed in BTKI‐CIT group compared with CIT group (log‐rank p = 0.0004, Supporting Information S1: Figure S1A), with a estimated 3‐year PFS rate of 93.8% (95% CI, 87.2%–100.0%) for BTKi‐CIT group and 65.9% (95% CI, 56.1%–77.3%) for CIT group.
The estimated 3‐year OS rate of 96.7% (95% CI, 90.5%–100.0%) for BTKi‐CIT group and 90.4% (95% CI, 83.7%–97.6%) for CIT group (Supporting Information S1: Figure S1B). After adjustment for IPI risk group, BTKi exposure remained independently associated with improved PFS (HR, 0.15; 95% CI, 0.05–0.50; p = 0.002).
3.5. CNS Event Analysis
A total of 49 patients were considered standard candidates for CNS prophylaxis based on a CNS‐IPI score ≥ 4 or testicular involvement. Among them, 33 patients received CNS prophylaxis, including 32 who underwent intrathecal methotrexate and 1 who received high‐dose methotrexate. During follow‐up, CNS relapse occurred in 5 (15.2%) of these 33 patients. Among the remaining 16 patients who did not receive CNS prophylaxis, 3 (18.8%) developed CNS relapse. Additionally, among patients who did not receive BTKi in the frontline setting, 15 of 99 patients (15.2%) developed CNS relapse. The median time from initiation of frontline treatment to CNS relapse was 14.9 months (95% CI, 9.2–21.5 months). Of the 15 CNS relapses, 11 (73.3%) were parenchymal, 3 (20.0%) were leptomeningeal, and 1 (6.7%) was isolated intraocular relapse.
Overall, all the 15 CNS relapses occurred in patients who did not receive BTKi as part of frontline therapy (15/99, 15.2%). No CNS relapse events were documented in the BTKi group (0/56). Kaplan–Meier analysis demonstrated a significantly lower cumulative incidence of CNS relapse among patients treated with BTKi (log‐rank p = 0.0054, Figure 3). In Firth‐penalized Cox regression to account for complete separation, BTKi exposure was significantly associated with a reduced risk of CNS relapse in univariable analysis (HR, 0.06; 95% CI, 0.0005–0.46; p = 0.002). After adjustment for CNS‐IPI risk category (low vs. intermediate/high) and CNS prophylaxis, BTKi exposure remained independently associated with lower CNS relapse risk (HR, 0.06; 95% CI, 0.0005–0.448; p = 0.002), whereas CNS‐IPI risk category (HR, 0.51; 95% CI, 0.17–1.40; p = 0.19) and CNS prophylaxis (HR, 0.39; 95% CI, 0.10–1.18; p = 0.099) was not significantly associated with CNS relapse in the adjusted model. Sensitivity analysis performed in the 144 patients treated with CIT and BTKI‐CIT demonstrated a significantly lower cumulative incidence of CNS relapse among patients in BTKi‐CIT group (log‐rank p = 0.0042, Supporting Information S1: Figure S2). Firth‐penalized Cox regression in this population confirmed BTKi exposure was associated with less CNS relapse (HR, 0.05; 95% CI, 0.0004–0.39; p = 0.008) after adjustment for CNS‐IPI risk category (HR, 0.26; 95% CI, 0.06–0.83; p = 0.02).
FIGURE 3.
Central nervous system progression–free survival according to BTK inhibitor exposure in the entire group. BTKi, Bruton tyrosine kinase inhibitor; CNS, central nervous system.
3.6. MCD Subtype Analysis
TP53 alteration was identified in 29 patients (18.7%), while 126 patients (81.3%) were TP53 wild‐type. In multivariable Cox regression including BTKi exposure, IPI risk group, and TP53 alteration status, BTKi exposure remained independently associated with improved PFS (HR, 0.16; 95% CI, 0.05–0.51; p = 0.002). Lower IPI risk categories were associated with favorable PFS. In contrast, TP53 alteration was not significantly associated with PFS in the adjusted model (HR, 0.63; 95% CI, 0.22–1.81; p = 0.39; Supporting Information S1: Figure S3).
Patients were further stratified according to BCR genotype (MYD88 + CD79B, MYD88‐only, CD79B‐only, and double negative). When stratified by BCR genotype, the estimated 3‐year PFS was 84.5% for MYD88‐only, 76.0% for double‐negative, 77.9% for CD79B‐only, and 66.6% for MYD88 + CD79B cases (Supporting Information S1: Figure S4). No statistically significant difference in PFS was observed among the four groups (Cox model p = 0.20). Due to the limited number of OS events (n = 11), formal multivariable survival analysis was not performed.
4. Discussion
DLBCL is increasingly recognized as a biologically heterogeneous disease in which genomic alterations play a major role in determining tumor behavior and treatment response. In this study, we integrated targeted genomic profiling with clinical outcome data in a cohort of 155 patients with MCD subtype DLBCL treated in routine clinical practice. Several observations from our analysis provide insight into the potential clinical relevance of BCR signaling and its therapeutic targeting.
Our findings support the biological rationale for targeting BCR signaling in genomically defined subsets of DLBCL. In our cohort, MYD88 and CD79B mutations were among the most frequent genomic alterations, consistent with previous genomic studies describing the MCD molecular subtype [11]. This subtype is characterized by chronic active BCR signaling and constitutive activation of downstream NF‐κB pathways, creating a dependency on BTK‐mediated signaling for lymphoma cell survival. In our analysis, exposure to BTK inhibitors was associated with significantly improved PFS compared with patients who did not receive BTKi, and this association remained significant after adjustment for established clinical risk factors. These findings are consistent with prior studies suggesting that BCR‐driven molecular subtypes may be particularly sensitive to BTK inhibition [14]. Taken together, these data support the concept that therapeutic strategies targeting BCR signaling may improve outcomes in biologically defined subgroups of DLBCL.
Additionally, our results provide insight into the potential relationship between targeting BCR signaling and CNS events. CNS relapse remains one of the most devastating complications of DLBCL and is associated with extremely poor prognosis once it occurs. Current strategies for identifying patients at risk largely rely on clinical models such as the CNS‐IPI, which incorporate factors including age, stage, lactate dehydrogenase level, and involvement of specific extranodal sites [22]. However, the ability of these clinical models to accurately predict CNS relapse remains limited. Prophylactic strategies such as intrathecal chemotherapy or systemic high‐dose methotrexate are frequently used in high‐risk patients, yet evidence supporting the effectiveness of these interventions remains inconsistent [22]. In our cohort, all CNS relapse events occurred in patients who did not receive BTK inhibitors as part of frontline therapy, whereas no CNS relapse events were observed among patients exposed to BTKi. Although the total number of events was limited, this observation raises the hypothesis that inhibition of BCR signaling may influence biological processes involved in CNS dissemination. BTK signaling has been implicated in multiple aspects of B‐cell migration and survival, and MYD88‐driven signaling pathways are known to play a central role in lymphomas arising in immune‐privileged sites such as the central nervous system and testis [23, 24]. It is therefore biologically plausible that inhibition of BTK signaling may interfere with mechanisms that facilitate lymphoma cell survival or trafficking within the CNS microenvironment. While these findings require confirmation in larger studies, they suggest that targeted inhibition of BCR signaling may represent a potential strategy for modifying CNS relapse risk in molecularly defined subsets of DLBCL.
Finally, our study also explored genomic heterogeneity within BCR‐driven disease. Although the MCD subtype is classically defined by co‐occurring MYD88 and CD79B mutations, clinical outcomes within this group may still vary. In exploratory analyses, stratification according to BCR genotype (MYD88 + CD79B, MYD88‐only, CD79B‐only, and double negative) did not reveal statistically significant differences in progression‐free survival, although numerical differences were observed. We also evaluated the potential prognostic impact of TP53 alterations, which have been associated with adverse outcomes in several DLBCL studies [25, 26]. In the multivariable analysis, TP53 alterations were not independently associated with progression‐free survival after adjustment for clinical risk factors and BTKi exposure. These findings suggest that the prognostic impact of specific genomic alterations may depend on the broader molecular and therapeutic context.
Several limitations of this study should be acknowledged. The retrospective design and the limited number of events, particularly for overall survival and CNS relapse, restrict the statistical power of some analyses. In addition, although targeted sequencing enabled comprehensive detection of recurrent mutations, genome‐wide copy number alterations were not fully captured, which may affect the precision of molecular subtype classification. Another limitation is the heterogeneity of induction regimens. Although only six patients received pola‐R‐CHP, precluding a meaningful comparison between pola‐R‐CHP and R‐CHOP, sensitivity analyses excluding all patients who received polatuzumab vedotin or lenalidomide yielded results consistent with the primary analyses (Supporting Information S1: Figure S5). These findings suggest that the observed association between BTKi‐containing therapy and improved survival outcomes was not driven by the small number of patients treated with alternative induction regimens. Finally, treatment exposure was heterogeneous, reflecting real‐world clinical practice rather than a controlled clinical trial environment.
Despite these limitations, our study provides several clinically relevant insights. By integrating genomic profiling with clinical outcome data, we demonstrate that BTK inhibitor exposure may be associated with improved outcomes in patients with BCR‐driven DLBCL and may potentially influence the risk of CNS relapse. These findings highlight the importance of incorporating molecular information into risk stratification and therapeutic decision‐making in DLBCL. Prospective studies are warranted to further clarify the role of BTK inhibition in molecularly defined subgroups and to determine whether targeted inhibition of BCR signaling may represent an effective strategy for CNS relapse prevention.
Author Contributions
Q.Z., S.J., and Y.L. drafted the manuscript. Q.Z., S.J., and X.Z. conceptualized and designed the study. Q.Z., S.J., Y.L., and W.Z. provided statistical input into the implementation and statistical analysis plan, under the supervision of J.C. Q.Z., S.J., Y.L., W.Z., Y.L., J.J., F.L., C.L., X.L., H.S., J.W., R.T., X.Z., and J.C. contributed to the participant enrollment and data acquisition. R.W. was responsible for quality control of NGS data, bioinformatics analysis, and interpretation of results. L.B. was responsible for NGS experimental operations. Q.Z., and X.Z. co‐conceived, implemented and supervised the study. All authors read, commented and approved the final manuscript.
Ethics Statement
The present study was designed and conducted in accordance with Good Clinical Practice and the Declaration of Helsinki.
Consent
All patients provided written informed consent before enrollment. All authors approved the final manuscript and the submission to this journal.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Supporting Information S1
Acknowledgments
We acknowledged receiving funding from National Natural Science Foundation of China (NSFC) (No. 82470189, 82400234 and 81870155), Clinical Research Special Project of Shanghai Municipal Health Commission (No.20244Y0073), Chinese Society of Clinical Oncology (CSCO) Chaoyang Oncology Research Fund Project (No. Y‐Young2024‐0156), Innovation Program of Shanghai Science and Technology Committee (20Z11900300) and Clinical Research Plan of Shanghai Hospital Development Center (SHDC2020CR3046B). The funder had no role in data collection, analysis, or interpretation, writing of the manuscript. We would like to thank all the participants in the study.
Contributor Information
Xiaoyan Zhou, Email: xyzhou100@163.com.
Qunling Zhang, Email: zhangqunling@fudan.edu.cn.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. Teras L. R., DeSantis C. E., Cerhan J. R., Morton L. M., Jemal A., and Flowers C. R., “2016 US Lymphoid Malignancy Statistics by World Health Organization Subtypes,” CA: A Cancer Journal for Clinicians 66, no. 6 (2016): 443–459, 10.3322/caac.21357. [DOI] [PubMed] [Google Scholar]
- 2. Liu W., Liu J., Song Y., et al., “Burden of Lymphoma in China, 1990‐2019: An Analysis of the Global Burden of Diseases, Injuries, and Risk Factors Study 2019,” Aging 14, no. 7 (2022): 3175–3190, 10.18632/aging.204006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Li X., Li G., Gao Z., Zhou X., and Zhu X., eds., “The Relative Frequencies of Lymphoma Subtypes in China: A Nationwide Study of 10002 Cases by the Chinese Lymphoma Study Group,” in Annals of Oncology (Oxford Univ Press Great Clarendon St, Oxford, 2011). [Google Scholar]
- 4. Coiffier B., Thieblemont C., Van Den Neste E., et al., “Long‐term Outcome of Patients in the LNH‐98.5 Trial, the First Randomized Study Comparing Rituximab‐CHOP to Standard CHOP Chemotherapy in DLBCL Patients: A Study by the Groupe d'Etudes Des Lymphomes de l'Adulte,” Blood 116, no. 12 (2010): 2040–2045, 10.1182/blood-2010-03-276246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Pfreundschuh M., Kuhnt E., Trumper L., et al., “CHOP‐like Chemotherapy With or Without Rituximab in Young Patients With Good‐Prognosis Diffuse Large‐B‐Cell Lymphoma: 6‐Year Results of an Open‐Label Randomised Study of the MabThera International Trial (MInT) Group,” Lancet Oncology 12, no. 11 (2011): 1013–1022, 10.1016/s1470-2045(11)70235-2. [DOI] [PubMed] [Google Scholar]
- 6. Coiffier B. and Sarkozy C., “Diffuse Large B‐cell Lymphoma: R‐CHOP Failure‐What to Do?,” Hematology American Society of Hematology. Education Program 2016, no. 1 (2016): 366–378, 10.1182/asheducation-2016.1.366. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Coiffier B., Lepage E., Briere J., et al., “CHOP Chemotherapy Plus Rituximab Compared With CHOP Alone in Elderly Patients With Diffuse Large‐B‐Cell Lymphoma,” New England Journal of Medicine 346, no. 4 (2002): 235–242, 10.1056/nejmoa011795. [DOI] [PubMed] [Google Scholar]
- 8. Ruppert A. S., Dixon J. G., Salles G., et al., “International Prognostic Indices in Diffuse Large B‐Cell Lymphoma: A Comparison of IPI, R‐IPI, and NCCN‐IPI,” Blood 135, no. 23 (2020): 2041–2048, 10.1182/blood.2019002729. [DOI] [PubMed] [Google Scholar]
- 9. Schmitz R., Wright G. W., Huang D. W., et al., “Genetics and Pathogenesis of Diffuse Large B‐Cell Lymphoma,” New England Journal of Medicine 378, no. 15 (2018): 1396–1407, 10.1056/nejmoa1801445. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Chapuy B., Stewart C., Dunford A. J., et al., “Molecular Subtypes of Diffuse Large B Cell Lymphoma Are Associated With Distinct Pathogenic Mechanisms and Outcomes,” Nature Medicine 24, no. 5 (2018): 679–690, 10.1038/s41591-018-0016-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Wright G. W., Huang D. W., Phelan J. D., et al., “A Probabilistic Classification Tool for Genetic Subtypes of Diffuse Large B Cell Lymphoma With Therapeutic Implications,” Cancer Cell 37, no. 4 (2020): 551–568.e14, 10.1016/j.ccell.2020.03.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Lacy S. E., Barrans S. L., Beer P. A., et al., “Targeted Sequencing in DLBCL, Molecular Subtypes, and Outcomes: A Haematological Malignancy Research Network Report,” Blood 135, no. 20 (2020): 1759–1771, 10.1182/blood.2019003535. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Stanchina M. D., Montoya S., Danilov A. V., et al., “Navigating the Changing Landscape of BTK‐Targeted Therapies for B Cell Lymphomas and Chronic Lymphocytic Leukaemia,” Nature Reviews Clinical Oncology 21, no. 12 (2024): 867–887, 10.1038/s41571-024-00956-1. [DOI] [PubMed] [Google Scholar]
- 14. Wilson W. H., Wright G. W., Huang D. W., et al., “Effect of Ibrutinib With R‐CHOP Chemotherapy in Genetic Subtypes of DLBCL,” Cancer Cell 39, no. 12 (2021): 1643–1653.e3, 10.1016/j.ccell.2021.10.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Morschhauser F., Leung W., Raghavan V., et al., “Deciphering the Clinical Benefit of Pola‐R‐CHP Versus R‐CHOP in Different Genetic Subtypes Beyond Cell of Origin in the POLARIX Study,” supplement, Blood 142, no. S1 (2023): 3000, 10.1182/blood-2023-178623. [DOI] [Google Scholar]
- 16. Morschhauser F., Leung W., Raghavan V., et al., “ABCL‐183 Deciphering the Clinical Benefit of Pola‐R‐CHP Versus R‐CHOP in Different Genetic Subtypes Beyond Cell of Origin (COO) in the POLARIX Study,” Clinical Lymphoma, Myeloma and Leukemia 24 (2024): S460, 10.1016/s2152-2650(24)01493-9. [DOI] [Google Scholar]
- 17. Jiang S., Zhang Q., Jin J., and Zhang W., “Genomic Risk Factors for Central Nervous System Relapse in Patients With Diffuse Large B‐cell Lymphoma,” Blood Research 60, no. 1 (2025): 36, 10.1007/s44313-025-00087-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Lionakis M. S., Dunleavy K., Roschewski M., et al., “Inhibition of B Cell Receptor Signaling by Ibrutinib in Primary CNS Lymphoma,” Cancer Cell 31, no. 6 (2017): 833–843.e5, 10.1016/j.ccell.2017.04.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Bobillo S., Joffe E., Sermer D., et al., “Prophylaxis With Intrathecal or High‐Dose Methotrexate in Diffuse Large B‐Cell Lymphoma and High Risk of CNS Relapse,” Blood Cancer Journal 11, no. 6 (2021): 113, 10.1038/s41408-021-00506-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Lin Z., Chen X., Liu L., Zeng H., Li Z., and Xu B., “The Role of Central Nervous System (CNS) Prophylaxis in Preventing DLBCL Patients From CNS Relapse: A Network Meta‐Analysis,” Critical Reviews in Oncology 176 (2022): 103756, 10.1016/j.critrevonc.2022.103756. [DOI] [PubMed] [Google Scholar]
- 21. Cheson B. D., Fisher R. I., Barrington S. F., et al., “Recommendations for Initial Evaluation, Staging, and Response Assessment of Hodgkin and Non‐Hodgkin Lymphoma: The Lugano Classification,” Journal of Clinical Oncology 32, no. 27 (2014): 3059–3068, 10.1200/jco.2013.54.8800. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Schmitz N., Zeynalova S., Nickelsen M., et al., “CNS International Prognostic Index: A Risk Model for CNS Relapse in Patients With Diffuse Large B‐Cell Lymphoma Treated With R‐CHOP,” Journal of Clinical Oncology 34, no. 26 (2016): 3150–3156, 10.1200/jco.2015.65.6520. [DOI] [PubMed] [Google Scholar]
- 23. Grommes C., Pastore A., Palaskas N., et al., “Ibrutinib Unmasks Critical Role of Bruton Tyrosine Kinase in Primary CNS Lymphoma,” Cancer Discovery 7, no. 9 (2017): 1018–1029, 10.1158/2159-8290.cd-17-0613. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Tan S., He H., Ni J., et al., “Meta‐Analysis Evaluating the Efficacy and Safety of Various Bruton Tyrosine Kinase (BTK) Inhibitors for Central Nervous System Lymphoma: Novel Covalent BTK Inhibitors, Except for Ibrutinib, Also Demonstrate Good Efficacy in the Treatment of Primary Central Nervous System Lymphoma,” Cancer 131, no. 18 (2025): e70083, 10.1002/cncr.70083. [DOI] [PubMed] [Google Scholar]
- 25. Xu‐Monette Z. Y., Wu L., Visco C., et al., “Mutational Profile and Prognostic Significance of TP53 in Diffuse Large B‐Cell Lymphoma Patients Treated With R‐CHOP: Report From an International DLBCL Rituximab‐CHOP Consortium Program Study,” Blood 120, no. 19 (2012): 3986–3996, 10.1182/blood-2012-05-433334. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Chiappella A., Diop F., Agostinelli C., et al., “Prognostic Impact of TP53 Mutation in Newly Diagnosed Diffuse Large B‐Cell Lymphoma Patients Treated in the FIL‐DLCL04 Trial,” British Journal of Haematology 196, no. 5 (2022): 1184–1193, 10.1111/bjh.17971. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting Information S1
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.