Primary sinonasal DLBCL: High prevalence of MCD subtype and CNS risk

Dear Editor,

Primary sinonasal diffuse large B-cell lymphoma (SN-DLBCL) is a rare extranodal (EN) lymphoma subtype. A nationwide retrospective cohort study from Denmark reported an incidence of 0.14 per 100,000 person-years, with 5-year overall survival (OS) and progression-free survival (PFS) rates of 56% and 50%, respectively.¹ However, robust nationwide, population-based incidence data for SN-DLBCL outside Denmark remain scarce. In Asian populations, sinonasal lymphomas are more commonly represented by extranodal natural killer/T-cell (EN NK/T-cell) lymphoma, and data specific to SN-DLBCL are limited.² Consequently, the clinical behaviour, molecular characteristics and risk of central nervous system (CNS) involvement in SN-DLBCL are not well-defined, particularly in Asian cohorts. This study reports a single-centre retrospective case series of SN-DLBCL in an Asian population.

In this retrospective study, 8 patients with primary SN-DLBCL were identified, all of whom were pathologically confirmed by biopsy. The baseline characteristics, treatment regimen and outcome of these patients are summarised in Table 1.

Table 1. Baseline characteristics, therapeutic regimen and outcome of primary SN-DLBCL patients.

The median age at diagnosis was 61.5 years (range, 56–74), and 5 patients were male. The clinical presentation was characterised by a high prevalence of nasal obstruction (62.5%), periorbital swelling with pain (50%) and diplopia (50%), collectively highlighting the locally aggressive nature of SN-DLBCL. In this cohort of 8 patients, next-generation sequencing (NGS) was performed in 7 patients, except 1 patient pathologically diagnosed at an outside hospital before referral. Genetic analysis demonstrated that all 7 patients were classified as the MYD88/CD79B-mutated (MCD) (85.7%, 6/7)/MCD-like (14.3%, 1/7) subtype DLBCL. In this study, MCD referred to cases with co-occurring MYD88 and CD79B mutations, whereas MCD-like referred to cases harboring MYD88 mutation without CD79B mutation. The most common additional mutations were in PIM1 (71.4%, 5/7) and KMT2D (42.9%, 3/7). Alterations in KMT2C, EP300, NOTCH1 and MPEG1 were each found in 28.6% of the cohort (2/7). This predominant molecular feature is consistent with previous studies³ and bears a strong resemblance to the profile of primary CNS DLBCL. CNS involvement is an uncommon complication, occurring in approximately 2–10% of patients with DLBCL, and carries a very poor prognosis.⁴ The CNS-International Prognostic Index (CNS-IPI) is the most commonly used tool for estimating risk of CNS recurrence, and CNS prophylaxis is recommended for high-risk patients.⁵ In the rituximab era, several additional high-risk features not captured by the CNS-IPI have been recognised, including dual expression of /B-cell lymphoma 2; the activated B-cell-like molecular subtype; certain EN sites involvement (testicular, uterine, breast); and intravascular large B-cell lymphoma.⁶ Furthermore, involvement of anatomical sites in the head and neck region (e.g. paranasal sinuses, hard palate) and areas close to the CNS (epidural space, dura, paravertebral region and orbit) has also been reported to confer an increased risk.⁷ A few studies have reported the association between SN-DLBCL and an increased risk of CNS involvement. Murawski et al.⁸ analysed 11 consecutive trials from the German High-Grade Non-Hodgkin Lymphoma Study Group. Ninety-three SN-DLBCL patients were identified out of 4155 patients (2.24%), and further analysis revealed that only paranasal sinus involvement was associated with a relative high risk of CNS relapse (2-year incidence 6.4%). Eriksen et al. analysed 116 SN-DLBCL patients registered in the Danish National Pathology Registry between 1980 and 2018. Among these patients, 30 of 116 (25.9%) developed recurrence and localised spread/progression, with 9 had CNS involvement.⁹ In this cohort, 2 of 8 patients (25%) presented with CNS involvement at diagnosis and 1 patient suffered from first CNS disease after complete remission (CR). At diagnosis, Cases 2 and 3 presented with CNS involvement, and successfully achieved CR induced by rituximab-cyclophosphamide, doxorubicin hydrochloride, vincristine sulphate and prednisolone-like immunochemotherapy combined with high-dose methotrexate (HD-MTX). Case 2 declined subsequent consolidation therapy with autologous hematopoietic stem cell transplantation (auto-HSCT) and maintenance therapy with Bruton’s tyrosine kinase (BTK) inhibitors, experienced isolated CNS relapse 6.4 months after completion of induction chemotherapy, and died 7 months after relapse. In contrast, Case 3 received BTK inhibitor therapy throughout the induction, consolidation and maintenance phases, which may have contributed to sustained remission and prolonged survival. Of the remaining 6 patients without CNS involvement at diagnosis, all have either completed or are currently scheduled to receive 2 cycles of HD-MTX as CNS prophylaxis. Long-term outcomes differed by treatment. Case 1 managed without auto-HSCT consolidation or BTK inhibitor maintenance, achieved initial CR but developed an isolated fatal CNS relapse 15.77 months after completion of induction chemotherapy. In contrast, Cases 4 and 5, who received auto-HSCT and/or BTK inhibitor therapy, have maintained sustained CR. With a median follow-up of 19.07 months (range, 4.33–25.80) for the entire cohort, the median CNS involvement-free survival was 20.37 months (95% confidence interval 0.00–49.75). Neither median PFS nor OS had been reached (Supplementary Fig. S1).

Findings show that SN-DLBCL patients predominantly exhibited an MCD/MCD-like molecular subtype and demonstrated a high frequency of CNS involvement. This propensity may be attributed to both the anatomical proximity of the sinonasal region to the CNS and the intrinsically aggressive molecular features. Therefore, the NGS testing appears essential for risk stratification. In resource-limited settings, a cost-effective alternative, such as digital polymerase chain reaction to detect key mutations (e.g. MYD88 and CD79B), should be considered to identify patients with the MCD molecular profile. These clinical observations further suggest that a prophylactic strategy consisting only 2 cycles of HD-MTX may be insufficient for patients with the MCD/MCD-like subtype of SN-DLBCL. Therefore, the authors recommend that CNS prophylaxis with HD-MTX be combined with auto-HSCT consolidation and/or BTK inhibitors maintenance, which has shown promising CNS penetrance and efficacy in genetically MCD subtype. However, these observations are tempered by limitations, including the single-centre, retrospective design, small sample size, limited follow-up duration and variations in treatment regimens, which may affect statistical power, the assessment of long-term relapse risk, and the generalisability of the findings, underscoring the need for further studies to validate these results and develop evidence-based management strategies.

Supplementary material

Fig. S1. Kaplan–Meier curves of the 8 primary SN-DLBCL patients for CNS involvement-free survival, progression-free survival and overall survival.

REFERENCES

1. Prg E, Eriksen PRG, Clasen-Linde E, et al. Sinonasal B-cell lymphomas: A nationwide cohort study, with an emphasis on the prognosis and the recurrence pattern of primary diffuse large B-cell lymphoma. Hematol Oncol 2022;40:160-71.
2. Kanumuri VV, Khan MN, Vazquez A, et al. Diffuse large B-cell lymphoma of the sinonasal tract: analysis of survival in 852 cases. Am J Otolaryngol 2014;35:154-8.
3. Peng F, Igawa T, Urata T, et al. High Prevalence of MYD88 and CD79B Mutations in Primary Sinonasal Diffuse Large B-Cell Lymphoma:Identification of an MCD-like Subtype. Am J Surg Pathol 2025;49:159-68.
4. Lin Z, Chen X, Liu L, et al. The role of central nervous system (CNS) prophylaxis in preventing DLBCL patients from CNS relapse: A network meta-analysis. Crit Rev Oncol Hematol 2022;176:103756.
5. 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. J Clin Oncol 2016;34:3150-6.
6. Chua BJG, Low CE, Yau CE, et al. Recent updates on central nervous system prophylaxis in patients with high-risk diffuse large B-cell lymphoma. Exp Hematol Oncol 2024;13:1.
7. Zahid MF, Khan N, Hashmi SK, et al. Central nervous system prophylaxis in diffuse large B-cell lymphoma. Eur J Haematol 2016;97:108-20.
8. Murawski N, Held G, Ziepert M, et al. The role of radiotherapy and intrathecal CNS prophylaxis in extralymphatic craniofacial aggressive B-cell lymphomas. Blood 2014;124:720-8.
9. Eriksen PRG, Clasen‐Linde E, de Nully Brown PD, et al. Sinonasal B‐cell lymphomas: A nationwide cohort study, with an emphasis on the prognosis and the recurrence pattern of primary diffuse large B‐cell lymphoma. Hematol Oncol 2022;40:160-71.