A Rare Case of Aggressive Disseminated Nasal-Type Epstein-Barr Virus-Positive Extranodal Natural Killer/T-Cell Lymphoma with Bone Marrow Involvement

Background

Nasal-type extranodal natural killer/T-cell lymphoma (ENKL), a rare subtype of non-Hodgkin lymphoma, is seen in 2–8% of lymphoma patients. It is most commonly seen in Asian and Central/South American populations and rarely in the United States, where incidence is highest amongst Hispanic and Asian Americans and lowest amongst Black and non-Hispanic White Americans [1–3]. The onset of disease typically occurs at age 50 years in Western countries, but at a mean age of 40 years, with male predominance at a ratio of 2: 3, in Asian populations [4]. Pathogenesis is poorly understood, but typically involves mutation and overexpression of the p53 tumor suppressor protein, the cyclin-dependent kinase inhibitor p21, and genes encoding components of the JAK/STAT pathway, among others, and includes a known association with Epstein-Barr virus (EBV) [5–7]. The majority of patients present with localized disease as a direct result of mass effect including vision impairment, diplopia, eye protrusion, ptosis, and nasal obstruction and its sequelae including facial pain, refractory sinusitis, and velopharyngeal insufficiency [1]. Additionally, tumor cell morphology is extremely variable. Extra attention should be given to the possibility of this diagnosis, given the aggressive nature of the tumor disease course.

Case Report

A 38-year-old woman with a past medical history of hypertension presented after 1 month of right-sided facial pain, dyspnea, non-productive cough, nausea, vomiting, green nasal discharge, and dark-brown eye discharge, despite a 7-day course of outpatient Augmentin therapy. On admission, the patient was septic and hypoxic and direct visualization of her nasal cavity revealed a right nasal mass. Tests for SARS-CoV-2, adenovirus, rhinovirus/enterovirus, influenza, parainfluenza, respiratory syncytial virus, Bordetella sp., Chlamydia pneumoniae, and Mycoplasma pneumoniae were all negative. Maxillofacial computerized tomography (CT) and magnetic resonance imaging (MRI) demonstrated local invasion from the right ethmoid cell wall through the cribriform plate and right frontal lobe vasogenic edema suggestive of metastatic disease. The patient was initiated on piperacillin-tazobactam, vancomycin, and amphotericin B. Subsequent antrostomy and ethmoidectomy were performed, and histopathological diagnosis revealed strong evidence of an aggressive lesion with characteristics including: mucosal ulceration; bony destruction with diffuse infiltration of lymphoid cells consisting of a broad cytological spectrum ranging from small to large neoplastic lymphoid cells; and large areas of extensive coagulative tumor necrosis with associated frequent mitosis. Immunohistochemical staining of the nasal tissue was positive for CD3, CD15, Ki-67 (~80%), CD4, CD43, CD56, and PD-L1 (Figure 1). Further testing of bone marrow biopsy tissue showed subtle involvement of the same lymphoid neoplasm as was found in the nasal tissue and with atypical lymphocytes of varying sizes admixed with small interstitial lymphoid aggregates and overall increased marrow hypercellularity. Immunohistochemical staining was positive for CD3, CD56, CD20, and PAX5. In-situ hybridization for EBV-encoded small RNA (EBER) was performed on the core biopsy and was positive for interstitial bone marrow involvement (Figure 2). Given the imaging findings and pathology features as above, the patient was diagnosed with stage IV nasal-type EBV-positive ENKL. The patient was then initiated on systemic chemotherapy with dexamethasone, methotrexate, ifosfamide, pegaspargase, and etoposide. Her hospital course was complicated by hypoxic respiratory failure, ifosfamide neurotoxicity, acute renal failure, and septic shock. Despite aggressive intervention, the patient died.

Discussion

Nasal-type ENKL is a rare type of malignant non-Hodgkin lymphoma. The clinical presentation is often aggressive and the disease course is commonly discovered in later stages; therefore, early diagnosis and prompt treatment of patients is pivotal [8]. Diagnosis is routinely established upon evaluation of a biopsy from the site of involvement followed by corresponding immunophenotype studies demonstrating key features of natural killer/T-cell markers and EBV. Simultaneous CT or MRI imaging is also routinely performed to determine localized, regional, or distant involvement of surrounding structures to aid in staging in conjunction with the Ann Arbor staging system. Additionally, bone marrow involvement in ENKL is noted to be uncommon, but core biopsy can be recommended for further staging of the disease, and management. It is postulated that EBV is a driver of this malignancy, demonstrated by a variety of EBV-oncogenic proteins which cause tumor proliferation. These include the latent membrane proteins (LMP) and more specifically, LMP-1, which inhibits apoptosis, stimulates cell-cycle progression and proliferation, and upregulates gene expression and downstream signaling pathways [7,9–11]. One of the genes that are promoted through LMP-1 stimulation is programmed death ligand 1 (PD-L1), the expression of which allows for escape from detection by activated T cells of the immune system. Additionally, LMP-1 regulates multiple signaling pathways such as JAK/STAT, MAPK, PI3K/Akt, and NFkB, which promote oncogenesis and interfere with apoptosis. However, because these proteins and pathways play a significant role in cell proliferation, they provide likely drug targets for oncogenic disruption [7,10,12]. Tumor cells will also commonly express biomarkers CD56, CD3, CD43, and CD45RO on an immunohistochemistry profile. These markers are consistent with T-lymphocyte proliferation, while EBV involvement is often demonstrated through molecular testing via in-situ hybridization or real-time reverse transcription polymerase chain reaction (RT-PCR) [9,13]. EBER transcripts are known to be abundantly present in infected tissue, with one study indicating levels that often exceeded 1 million copies/cell [13]. In-situ hybridization utilizes oligonucleotide probes that will hybridize to the corresponding chromosomal base pairs and allow for detection of EBV involvement through quantitative analysis of viral copies. Alternatively, RT-PCR allows for direct quantitation of EBV DNA load from peripheral blood. These molecular techniques can be useful to monitor disease course and perhaps be utilized as a predictor for prognosis given its surrogacy as a marker of tumor burden [9,13]. Another potential prognostic factor is Ki-67, which is a nuclear protein that is synthesized at the start of cell proliferation and is present in all active phases (G1, S, G2, and M) of the cell cycle with the exception of the resting G0 phase. Because of its involvement in proliferative activity, it has been used as a marker for the level of tumor aggression, represented as an index and calculated as a percentage from the number of positive cells that express nuclear Ki-67 divided by the total number of tumor cells per high-power field [12,14–16].

However, its value for prognostication is controversial, as outcomes among subtypes of lymphoma have been contradictory and at times inconclusive, with some studies demonstrating an inverse relationship, with poorer survival rates, while others have shown a lack of association or even opposite findings. Still, a meta-analysis of 4112 patients from 28 studies worldwide between 1990 and 2013 revealed Ki-67 as a good prognostic indicator for non-Hodgkin lymphoma subtypes including natural killer/T-cell lymphoma, in which high expression is correlated with a poor survival outcome [12,14].

Prognostication aside, management of the disease consists of radiation and chemotherapy. For localized disease in stable patients, the primary strategy to treat this malignancy appears to be chemotherapy plus radiation, or combined modality therapy (CMT), rather than single modality alone. Radiation doses of 50 Gy or greater are recommended, and can be done as: (1) sequential therapy with chemotherapy administered first, followed by consolidative therapy (sandwich model), (2) radiotherapy conducted concurrently with chemotherapy, or (3) sequential therapy, with radiation therapy administered first. Frontline chemotherapies such as dexamethasone, methotrexate, ifosfamide, L-asparaginase, and etoposide (SMILE) demonstrate some promise in relapsed and advanced-stage ENKL. In addition, these therapies exhibit a higher 5-year overall survival (OS) when combined with radiation, vs conventional regimens that integrate anthracycline regimens such as cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP), which show lower and suboptimal 5-year OS [2,3,5–7,17–21]. Retrospective studies appear to demonstrate superior outcomes from CMT compared with radiation or chemotherapy alone, and to our knowledge, there have been no prospective, controlled studies to date that draw direct comparisons. In addition, most studies have utilized suboptimal chemotherapy with anthracycline-containing regimens and radiation protocols that are no longer considered optimal [17]. Consequently, more research is needed to fully establish optimal CMT versus individual treatment with radiation therapy or chemotherapy for nasal-type ENKL. In addition, following completion of induction, response to therapy must be assessed to determine the need for post-induction management. Close surveillance is favored in patients with complete response; however, hematopoietic stem cell transplantation, either autologous (auto-HSCT) or allogeneic (allo-HSCT), may be warranted as post-induction consolidation therapy. So far, through retrospective analysis, auto-HSCT has not been shown to provide results significantly different from chemotherapy or radiotherapy alone, and is therefore not generally recommended for treatment of ENKL. Allo-HSCT, on the other hand, offers promising potential as a curative treatment due to its graft-versus-lymphoma effect in eradicating residual lymphoma cells. In a retrospective study, patients with advanced-stage disease and in complete remission at presentation were treated with SMILE followed by allografting. These patients showed an estimated 5-year progression-free survival and OS of 51% and 57%, respectively [22–24]. These results demonstrated a drastic improvement compared with 2-year survival rates (30–40%) in patients who had received non-L-asparaginase-containing regimens, as reported in other retrospective studies [5,6,22–24]. However, because of high treatment-related mortality, allo-HSCT is typically reserved for stable patients in relapse or for advanced-stage or refractory disease. Of note, there have been no randomized trials to date utilizing HSCT, further limiting its application as a post-induction treatment to ENKL [5,6]. Alternatively, in unstable patients in whom HSCT may not be ideal, immunotherapy utilizing agents that block the expression of PD-L1, such as pembrolizumab and nivolumab, have shown promise in several small studies [3,5,6,19]. An additional agent in this class is bortezomib, which is under phase II evaluation but has already shown a positive overall response rate [7].

Nevertheless and unfortunately, a large portion of patients with ENKL present in advanced stages, in which case prognosis is poor with an overall median survival of 4–7 months [17]. While CMT has shown promise as a primary strategy with notable success, optimization of the chemotherapy regimen and radiotherapy protocol is needed to continue improving treatment of nasal-type ENKL malignancy. More research and data are needed for the use of PD-L1 inhibitors in refractory and advanced cases. Furthermore, given similar immunophenotypic profiles observed among nasal-type ENKL, advancements in genetic analysis may also lead to novel targets for therapy in checkpoint and signaling pathway blockade.

Conclusions

Nasal-type ENKL is a rare variation of non-Hodgkin lymphoma that is associated with high morbidity and mortality. It is, furthermore, infrequently associated with bone marrow involvement. The malignancy is typically diagnosed by biopsy evaluation, with positive immunohistochemistry results revealing common biomarkers including CD56, CD3, CD43, and CD45RO. Additionally, positive findings from in-situ hybridization with an EBER probe further strengthen the case for EBV involvement. Treatment is primarily through combined radiation and chemotherapy. Prior research studies have utilized chemokine-modulating drugs and have demonstrated promise in their effectiveness; however, more research must be performed to fully understand their use in treating this aggressive malignancy