Are You Confident of the Diagnosis?
What to be alert for in the history
Extranodal NK/T cell lymphoma (ENKL) is the most common lymphoma of the nasal cavity. The most common presentation of ENKL is with upper aerodigestive tract (UAT) symptoms. Presenting symptoms include persistent nasal obstruction, rhinorrhea, swelling, and bleeding. In advanced disease, the involvement of adjunct bony tissue is evident with destruction, swelling, or necrosis of the involved areas. Invasion into the hard palate leads to characteristic midline perforation.
More than 80% of ENKL arise from UAT including nasal cavity and the paranasal areas such as nasopharynx, oral cavity, oropharynx, and hypopharynx. The other sites of involvement include lung, liver, spleen, the gastrointestinal tract, orbit, adrenals, testes, and the bone marrow. Central nervous system involvement is estimated at 5% to 10%.
ENKL are divided into two subtypes: UAT NK-cell lymphoma (UNKL) includes all lymphomas confined to the nasal cavity, nasopharynx, and the UAT; whereas extraupper aerodigestive tract NK-cell lymphoma (EUNKL) embraces lymphomas occurring at all other sites. Patients with a primary lesion within the nasal cavity and secondary spread to other organs are considered as UNKL. While UNKL and EUNKL share the same pathology, their clinical course differs significantly, with EUNKL likely to run a more aggressive course.
Presenting symptoms include persistent nasal obstruction, rhinorrhea, swelling, and bleeding. In advanced disease, the involvement of adjunct bony tissue is evident with destruction, swelling, or necrosis of the involved areas.
The presence of B symptoms such as night sweats and weight loss may give a clue toward an underlying malignancy. The presence of B symptoms such as low-grade fever, drenching night sweats or unintentional weight loss of 10% or more of body mass in the preceding 6 months were associated with a two-fold risk (P <.0003; relative risk, 2.202; 95% CI, 1.446 to 3.353) of death in one study.
Characteristic findings on physical examination
Careful physical examination of the UAT is essential to determine the extent of the disease. Thorough examination of the skin is important as skin is the most common extranasal site to be involved. Clinically, the patient presents with diffuse, nodular, plaque-like, lobular panniculitis-like, or lichenoid infiltration of dermis or subcutaneous tissue. ENKL skin lesions are typically described as smooth-surfaced, violaceous nodules and plaques that can ulcerate. The lesions have a predilection for the trunk but also can involve the extremities, head, and neck.
Other possible dermatologic presentations include patches, plaques, ulcer, necrosis, granulomatosis, and bullae. These dermatologic findings may be present with or without the primary nasal tumor, making diagnosis challenging.
As mentioned above, the skin is the most frequent extranasal site of involvement for ENKL. The fact that a patient may have only cutaneous involvement without any nasal involvement makes ENKL a challenging diagnosis in such cases.
Expected results of diagnostic studies
After careful history and physical examination, imaging of the nasal cavity and hard palate is usually the next step. Computerized tomography (CT) scan is a standard choice. However, magnetic resonance imaging (MRI) is gaining favor as a modality of choice owing to its ability to discriminate between soft tissue and bony masses.
Fluorine fluorodeoxyglucose positron emission tomography CT scan (FDG-PET CT) is a test with high sensitivity for nasal as well as extranasal involvement. Its use is limited when the involvement of skin and bone marrow are suspected. Panendoscopy is usually performed to determine the extent of disease involvement of UAT.
The diagnosis of ENKL is established with histopathologic examination of biopsy specimens. The characteristic feature of ENKL tumor cells is growth around vessels (angiocentricity, giving it the name of “angiocentric lymphoma”) and vascular destruction. Biopsy results show angiodestructive with admixed inflammatory infiltrate and necrosis. This is intermixed with areas of coagulative necrosis and the presence of apoptotic bodies.
The tumor cells are small to medium in size with occasional large and anaplastic cells. Tumor cells are characterized by moderately pale cytoplasm with azurophilic granules.
Another characteristic feature is the presence of cucumber-like tumor cells with elongated nuclei. There is diffuse lymphoid infiltration with the presence of inflammatory cells such as eosinophils, histiocytes, and other small lymphocytes including plasma cells that are believed to be reactive in nature. The World Health Organization (WHO) criteria require the presence of Epstein-Barr virus (EBV)-encoded small nuclear RNA-1 (EBER-1, vide infra) and the presence of cytotoxic granules for diagnosis.
IHC of the specimen shows the presence of cytoplasmic CD3 (epsilon) but the absence of surface CD3 or T-cell receptor. T/NK marker such as CD2 is usually positive along with the presence of NK-cell marker such as CD56. CD16 may or may not be present. T-cell specific markers such as CD5, CD7 and CD57 are absent.
The immunophenotype of the tumor cells is CD2+, CD56+, cytoplasmic CD3ε+, surface CD3–, CD5–, CD4–, CD8–, EBER+, clonal episomal EBV in tumor cells, and T-cell receptor genes in germline configuration. The phenotype suggests that these tumors arise in NK cells. Tumor cells also possess markers of cytotoxic cells such as granzyme B, T-cell restricted intracellular antigen (TIA)-1, and perforin. EBV is almost always detected by EBER1 in ENKL cells.
Detection of EBV in normal lymphoid cells of nasal mucosa is extremely uncommon. Hence the detection of EBV should lead to consideration of ENKL or other EBV-associated neoplasms.
There are no sine qua non cytogenetic changes associated with ENKL. More than two thirds of patients have abnormal karyotype including pseudodiploidy, hyperdiploidy, and hypodiploidy. Deletion of the long arm of chromosome 6, usually in the region of 6q21-6q25 is a common cytogenetic abnormality. Loss of heterozygosity at 13q is reported in one third of cases at diagnosis but is uniformly present at relapse.
The diagnosis and management of ENKL may be complicated by several factors:
– Biopsy specimens often show widespread necrosis. The pathologic findings may be subtle at times, leading to false-negative results.
– The tumor may be confused with a benign inflammatory reaction. However, extensive ulceration, involvement of adjacent tissue especially the bone, and the presence of EBV as outlined above may help establish the diagnosis.
– Variants such as EBV negative or CD56 negative ENKL have been described. In such cases, the diagnosis of peripheral T-cell lymphoma, not otherwise specified, should be considered as there is a definite phenotypic overlap between the two.
– CD56 may also be expressed in a subgroup of acute myelogenous leukemia (AML), presenting diagnostic challenge, especially if there is extramedullary or cutaneous involvement. However, CD56+ AML is invariably EBV negative, distinguishing it from ENKL.
Who is at Risk for Developing this Disease?
The classic form of ENKL manifests itself in Asian and Central and South American adults (median age of presentation in the 50s) as a tumor mass in the anterior nasal cavity. ENKL is more common in males than in females, with 4.5-fold higher prevalence in Asians and in South Americans of American Indian descent compared with Western countries.
The Korean Dermatopathology Research Group conducted a nationwide multicenter retrospective study that included 80 cases of cutaneous lymphomas. Twelve of 80 patients studied had NK / T cell lymphoma. In another retrospective study conducted at a single center in Japan, 133 new patients with cutaneous lymphomas were seen at the dermatology clinic. Of these, 3.8% (5 patients of 133) were diagnosed with ENKL by reviewing pathology and EBER-1.
The incidence in these studies differs significantly from the data derived from DACLG (Dutch and Austrian Cutaneous Lymphoma Group) and the National Cancer Institute Surveillance, Epidemiology, and End Results (SEER-16) studies in which 7 of 1905 and 12 of 3810 patients were diagnosed with ENKL respectively.
The exact mechanism for such striking geographic predilection is not yet known.
What is the Cause of the Disease?
The direct role of EBV in the pathogenesis of ENKL has been widely debated. This was further complicated by the presence of EBV in non-malignant cells as well as the association of EBV with other non-ENKL nasal tumors. However, there is a consensus now that EBV plays a direct role in NK-cell transformation in ENKL. The comparative study of EBV genome in normal and malignant NK cells supports the direct role in pathogenesis.
As mentioned above, the in situ hybridization (ISH) for EBER-1 within the transformed cells is critical to diagnosis. It is a test with such high sensitivity that negative ISH for EBER-1 should lead to re-evaluation and consideration of an alternative pathology. While B-cells might be positive for EBER-1 in chronic EBV infections unassociated with ENKL, the presence of EBER-1 in a single non-B cell should prompt the possibility of bone marrow involvement in ENKL.
The presence of mutated (30-base pair deletion) latent membrane protein (LMP-1) is a characteristic finding in transformed NK cells compared with wildtype LMP-1 that is found in normal NK cells. Mutation of LMP-1 is thought to play a crucial role in both cellular transformation as well as immune evasion.
The measurement of circulating fragmented EBV DNA (<500 base pairs in length) derived from dead tumor cells is important for diagnosis, monitoring, and prognosis of the disease.
Inhibition of EBV nuclear antigen (EBNA) inhibits the growth of ENKL cells in vitro, suggesting that EBNA may be a promising target for treatment of ENKL.
Systemic Implications and Complications
InstructionThe signs and symptoms of systemic involvement by the disease depend on the site of involvement. Sites of involvement other than the skin, such as the muscle, the adrenal gland, and the female genital tract (ovaries and uterus), lung, liver, spleen, the gastrointestinal tract, orbit, testes, and the bone marrow have been reported. The involvement of orbital cavity may result in impairment of extraocular muscles and/or proptosis. The involvement of bone marrow produces anemia, leukopenia, thrombocytopenia or combination thereof. Frequent infection secondary to leukopenia is a common cause of death.
Standard of care guidelines are not available for ENKL as large, prospective, randomized controlled trials have not been carried out. Such lack of evidence is due to several factors:
– rarity of the disease
– aggressive course
– heterogeneity in the diagnosis and staging
– geographic/ethnic restrictions
Hence, most treatment guidelines are consensus-based, largely derived from retrospective trials. Larger prospective randomized, multicenter trials are needed to establish evidence-based standard-of-care. Treatment modalities available for patients with ENKL are summarized in the Table I.
|Early-Stage Disease||Advanced- Stage Disease|
|Radiotherapy||Hematopoietic stem cell transplant (HSCT)|
|Chemoradiotherapy (concurrent or radiotherapy followed by chemotherapy)|
Optimal Therapeutic Approach for this Disease
The approach to treatment largely depends upon the extent of the disease.
Patients with localized disease usually respond well to radiotherapy. Front-line radiotherapy has resulted in superior survival compared with front-line chemotherapy.
Generally, high-dose radiotherapy (>50 Gy) is used as higher dose radiotherapy is shown to produce better 5-year overall survival (OS) than low-dose radiotherapy.
Non-anthracycline chemotherapy agents such as methotrexate (MTX), ifosfamide, and L-asparginase are preferred over anthracycline agents such as idarubicin, hydroxydaunorubicin, doxorubicin and are shown to produce higher complete remission (CR) and OS.
May be the optimal treatment for the early stages of ENKL; however, evidence from a large randomized prospective trial is still missing. Concurrent chemoradiotherapy or radiotherapy followed by chemotherapy may produce better CR and OS in patients with stage II disease. Our recommendation is to treat patients of stage I with radiotherapy alone. In case of relapse, the patients can be treated with radiotherapy alone again or with combination chemoradiotherapy. Patients with stage II can be treated with combination chemoradiotherapy as the first modality of treatment.
Overall, the prognosis for late-stage disease is grim. The addition of L-asparaginase has changed the treatment of advanced ENKL disease. The overall trend has been to move away from anthracycline-based regimens to non-anthracycline-based regimens.
Rationale for use of non-anthracycline based regimens
P-glycoprotein, a product of the multidrug resistance (MDR)-1 gene, is abundantly expressed on tumor cells of ENKL. P-glycoprotein is involved in active transport of various drugs (efflux) outside the tumor cells, leading to refractoriness to conventional anthracycline-containing chemotherapeutic regimens. This may explain, at least in part, the modest response to agents such as doxorubicin and vincristine, which are the major components of conventional chemotherapy regimens (eg, CHOP).
L-asparaginase and SMILE regimen
In vitro data showed that L-asparaginase induces apoptosis in NK-cell lymphoma/leukemia cell lines and patient tumor cells. L-asparaginase is an enzyme that breaks down serum L-asparagine, leading to depletion of L-asparagine in the cells. ENKL cells lack asparagine synthetase, making them more susceptible to the actions of L-asparaginase. L-asparaginase selectively acts on lymphoid cells, causing minimal myelosuppression.
Principal side effects associated with L-asparaginase are hemostatic complications, allergy, or pancreatitis. Patients with allergy to Escherichia coli L-asparaginase may receive Erwinia asparaginase (20,000U/m2) with the same schedule.
SMILE (steroid, MTX, ifosfamide, L-asparginase, etoposide) regimen consists of MTX 2g/m2 on day 1, ifosfamide 1500mg/m2, etoposide 100mg/m2 and dexamethasone 40mg from days 2 to 4, and E coli L-asparaginase 6000U/m2every other day from day 8 to day 20 (7 doses). Because of the profound myelotoxicity with the regimen, granulocyte colony-stimulating factor(G-CSF) should be used from day 6 until the neutrophil count recovers.
For patients who are 65 years or younger with good organ function with advanced ENKL disease, SMILE is a reasonable approach. Good functional status is defined as Eastern Cooperative Oncology Group (ECOG) performance status 0 to 2, WBC of 3000/mm3 or more, ANC 1200/mm3 or more, ALC 500/mm3 or more, platelet 50,000/mm3 or more, normal AST/ALT, total bilirubin 2.0 mg/dL or less, creatinine 1.5 mg/dL or less, left ventricular ejection fraction (LVEF) 50% or more, and O2 saturation 90% or more at room air.
In patients who are older than 65 years or have mild organ damage, dose-reduced SMILE (dose reduction by 33% to 75%) is a reasonable approach. In patients with severe organ dysfunction, a trial of steroid + L-asparaginase can be given.
Severe organ dysfunction is defined as ECOG performance status 4, WBC 1000/mm3 or less, ANC 500/mm3 or less, platelet 20,000/mm3 or less, AST/ALT within 10 times the normal limit, total bilirubin 5.0 mg/dL or less, or creatinine 3.0 mg/dL or less. If patient tolerates it well or if the functional status improves, a dose-reduced SMILE can be justified in such patients. Note, however, that prognosis in such patients remains poor.
A recent prospective phase II trial evaluated a related regimen, L-asparaginase, MTX, and dexamethasone (AspaMetDex) in patients with relapsed or refractory disease. Typically, these patients have extremely poor survival. The results of the trial showed objective responses in 14 of the 18 patients after three cycles. Eleven patients entered complete remission (61%), and only four of them relapsed. The median OS was 1 year, with a median response duration of 12 months.
Various other chemotherapeutic regimens such as DeVIC (dexamethasone, VP16, ifosfamide, carboplatin) have been tried with variable success.
Unusual Clinical Scenarios to Consider in Patient Management
As mentioned above, ENKL may present in various forms depending on the site of involvement. Recently, Summers et al described a case of overlap of ENKL and aggressive NK-cell leukemia (ANKL) in a 15-year-old Caucasian male. He presented with painful subcutaneous nodules and plaques over his anterior thighs, shins, and lower abdomen while receiving infliximab therapy for inflammatory bowel disease. He also was noted to have pulmonary nodules, liver nodules, and hepatosplenomegaly on CT scan, as well as thrombocytopenia and transaminitis.
Skin biopsy in this case revealed atypical small to intermediate EBV-positive lymphoid cells of NK-cell origin infiltrating the subcutaneous adipose tissue, mimicking subcutaneous T-cell lymphoma. Similar atypical EBV-positive lymphocytes were found in the bone marrow, liver, stomach, and colon.
What is the Evidence?
Lee , MW. “Characteristics of cutaneous lymphomas in Korea”. Clin Exper Dermatol . vol. 28. 2003. pp. 639-46. (This article reviews findings of the Korean Dermatopathology Research Group that conducted a multicenter clinicopathologic review of a nationwide collection of 80 cutaneous lymphomas over a recent 3-year period. The review discusses various dermatologic manifestations of the disease.)
(A recent paper describing results of a French multicenter prospective phase II trial of an L-asparaginase-containing regimen in 19 patients with relapsed or refractory disease.)
Lee , J, Suh , C, Park , YH, Ko , YH, Bang , SM, Lee , JH. “Extranodal natural killer T-cell lymphoma, nasal-type: A prognostic model from a retrospective multicenter study”. J Clin Oncol . vol. 24. 2006. pp. 612-8. (A multicenter retrospecitve study of 262 patients with ENKL. The study proposes a staging system specific for ENKL.)
Kim , TM, Heo , DS. “Extranodal NK/T-cell lymphoma, nasal type: New staging system and treatment strategies”. Cancer Sci . vol. 100. 2009. pp. 2242-8. (An excellent review that proposes new staging system based on involvement of UAT, correlating it with the prognosis and treatment strategies.)
Suzuki , R, Takeuchi , K, Ohshima , K, Nakamura , S. “Extranodal NK/T-cell lymphoma: diagnosis and treatment cues”. Hematolog Oncol . vol. 26. 2008. pp. 66-72. (An excellent review of ENKL disease with diagnostic and therapeutic strategies.)
Greer , J, Mosse , C. “Natural killer-cell neoplasms”. Curr Hematol Malignancy Rep . vol. 4. 2009. pp. 245-52. (A review describing biology of natural killer cells transitioning to the pathology of various NK cell neoplasms including ENKL.)
Hasserjian , RP, Harris , NL. “NK-cell lymphomas and leukemias”. Am J Clin Pathol . vol. 127. 2007. pp. 860-8. (This review describes the pathologic diagnosis of ENKL. It also addresses the diagnostic pitfalls and the spectrum of ENKL disease.)
Fujita , A, Hamada , T, Iwatsuki , K. “Retrospective analysis of 133 patients with cutaneous lymphomas from a single Japanese medical center between 1995 and 2008”. J Dermatol. vol. 37. 2010. pp. 1-7. (This Japanese retrospective analysis of cutaneous lymphomas describes epidemiology of various cutaneous lymphomas with emphasis on dermatologic manifestations.)
Greer , JP. “Therapy of peripheral T/NK neoplasms”. Hematology. 2006. pp. 331-7. (This is an excellent review giving a broad overview of NK neoplasms with diagnostic and therapeutic strategies.)
Summers , E, Samadashwily , G, Florell , SR. “A unique presentation of an Epstein-Barr virus-associated natural killer/T-cell lymphoproliferative disorder in a white male adolescent”. Arch Dermatol . vol. 147. 2011. pp. 216-20. (A recent case report describing unusual skin presentation in a young Caucasian male after receiving immunosuppressive therapy with infliximab.)
Jaccard , A, Gachard , N, Marin , B, Rogez , S, Audrain , M, Suarez , F. “Efficacy of L-asparaginase with methotrexate and dexamethasone (AspaMetDex regimen) in patients with refractory or relapsing extranodal NK/T-cell lymphoma, a phase 2 study”. Blood . vol. 117. 2011. pp. 1834-9.
Copyright © 2017, 2013 Decision Support in Medicine, LLC. All rights reserved.
No sponsor or advertiser has participated in, approved or paid for the content provided by Decision Support in Medicine LLC. The Licensed Content is the property of and copyrighted by DSM.