CD30 expression in extranodal natural killer/T-cell lymphoma, nasal type among 622 cases of mature T-cell and natural killer-cell lymphoma at a single institution in South China

Background Mature T-cell and natural killer (NK)-cell lymphomas compose a heterogeneous group of non-Hodgkin lymphomas, and extranodal NK/T-cell lymphoma, nasal type (ENKTL) is an aggressive subtype with sporadic CD30 expression. However, the significance of CD30 expression in ENKTL is controversial. We aimed to classify a large cohort of patients with mature T-cell and NK-cell lymphomas according to the 2016 World Health Organization (WHO) classification guidelines and to study the association between CD30 expression and prognosis of patients with ENKTL. Methods We selected consecutive patients with mature T-cell and NK-cell lymphomas who attended our institution between September 1, 2009 and August 31, 2013. We classified the lymphomas according to the 2016 revision of the WHO classification of lymphoid neoplasms, analyzed the associations between CD30 expression and clinicopathologic features of ENKTL patients, and evaluated the prognostic implications of CD30 expression. Results We identified 622 consecutive patients with mature T-cell and NK-cell lymphomas, including 317 (51.0%) patients with ENKTL. In addition, CD30 expression was detected in 43 (47.3%) of a subset of 91 patients with ENKTL. No clinicopathologic features were associated with CD30 expression, and CD30 positivity showed no prognostic significance in patients with ENKTL. Conclusions ENKTL is the most common type of mature T-cell and NK-cell lymphoma diagnosed at our institution. CD30 is frequently expressed in ENKTL and represents a therapeutic target; however, it may not be a prognostic marker.

In general, mature T-cell and NK-cell lymphomas are aggressive tumors with a poor prognosis, and the use of systemic chemotherapy, including anthracycline, achieves complete remission in only a fraction of all cases. The 3-year overall survival (OS) and failure-free survival (FFS) rates of patients with mature T-cell and NK-cell lymphoma are approximately 52% and 32%, respectively [4]. The prognosis of patients with ENKTL is also unsatisfactory, even for those who present with early-stage disease [5]. The recent availability of brentuximab vedotin, an anti-CD30 monoclonal antibody, has led to improved OS and FFS rates among patients with refractory Hodgkin lymphoma, which is characterized by strong and uniform expression of CD30 [6]. A similar benefit was observed for patients with ALCL, a T-cell lymphoma also characterized by the strong and uniform expression of CD30 in neoplastic cells [7]. We previously reported that CD30 is expressed in ENKTL [8]; however, the significance of CD30 expression in ENKTL remains unclear.
Thus, we reviewed clinical records of a large cohort of ENKTL patients and analyzed the prognostic implications of CD30 expression. In addition, we also investigated the subtype distribution and incidence of mature T-cell and NK-cell lymphomas according to the 2016 WHO classification system in a large cancer center in South China.

Patient selection
We searched for all cases of mature T-cell and NK-cell lymphomas diagnosed in the Department of Pathology at the Sun Yat-sen University Cancer Center (Guangzhou, China) between September 1, 2009 and August 31, 2013. Diagnoses were based on the criteria of 2016 revision of WHO Classification of Lymphoid Neoplasms [3]. A series of ENKTL patients who were treated in the Department of Medicine at the Sun Yat-sen University Cancer Center during the same period were retrospectively selected to further investigate their clinical features and CD30 expression levels. The selection criteria were as follows: (1) confirmed diagnosis by pathologists, and (2) availability of complete follow-up records and pathologic materials. Patients with concomitant malignant neoplasms were excluded. Patient records were reviewed, and clinical data regarding age, gender, symptoms at presentation, site of involvement, plasma levels of Epstein-Barr virus (EBV)-DNA determined using quantitative polymerase chain reaction (Q-PCR), clinical stage, response to therapy, and status at last followup were extracted. The Institutional Review Board at the Sun Yat-sen University Cancer Center approved this study.

Patient follow-up
All patients were followed up every 6 months until May 2015. Disease progression and recurrence were diagnosed based on clinical examination, imaging assessments, and pathologic examination. OS was measured from the initiation of treatment to either the last followup or death from any cause. Progression-free survival (PFS) was measured from the initiation of treatment to the first indications of disease progression or relapse, death from any cause, or the last follow-up.

Histopathologic and immunohistochemical analysis
All the diagnoses were confirmed by seven senior pathologists from the Department of Pathology at the Sun Yatsen University Cancer Center after reviewing hematoxylin and eosin-stained tissue sections, immunohistochemistry (IHC) results, and other ancillary materials (including clinical records). A diagnostic consensus was made when the preliminary review by individual pathologists reached inconsistent diagnoses. IHC was performed on sections of 4-μm thickness. To determine the immunophenotype, IHC examination and in situ hybridization for Epstein-Barr virus-encoded small RNAs (EBERs) were performed. The main primary antibodies and antigen retrieval methods for each antibody are shown in Table 1. Appropriate positive and negative controls were included for each antibody.
The EBV Probe In Situ Hybridization Kit (DIG-AP, A300K.9901, PanPath Company, Amsterdam, the Netherlands) was used to detect EBERs according to the following procedure: (1) deparaffinization and rehydration of the paraffin sections using xylene and a series of graded ethanol solutions, respectively; (2) pretreatment of samples with 0.4% pepsin for 10 min; (3) hybridization with digoxigenin-conjugated EBV probes at 37 °C for 3 h; (4) signal detection using peroxidase-conjugated antidigoxigenin antibody and 3,3′-diaminobenzidine (DAB); and (5) counterstaining of sections with a hematoxylin solution. The positive signals were brownish-yellow in color and localized within the nuclei.
CD30 was detected using a ready-to-use antibody supplied by Zymed Laboratories (Zymed-08-155, Thermo Fisher Scientific Company, South San Francisco, CA, USA). IHC was performed on sections of 4-μm thickness. Heat-induced antigen retrieval was performed using a citrate acid buffer (pH 6.0) in a stainless steel high-pressure cooker for 10 min, with the pressure set to approximately 103 kPa. The antibody was detected using DAB on a DAKO EnVision system (Dako, Carpinteria, CA, USA) with hematoxylin as a counterstain. CD30 expression was semi-quantitated in positive tumor cells. A total of 100 tumor cells in two randomly selected high-power fields were counted, and the average percentage of positive cells was calculated as the rate of CD30 expression. CD30 positivity was defined as strong cell membrane and Golgi zone reactivity in ≥20% of tumor cells [9].

Statistical analysis
Categorical variables were compared using either the Chi square or Fisher's exact tests. OS and PFS were estimated using the Kaplan-Meier method. Differences were considered significant when the two-sided P value was less than 0.05. All analyses were performed with SPSS 19.0 software (SPSS, Inc. Chicago, IL, USA).
In total, 91 patients with ENKTL met the selection criteria mentioned above, and their data were collected along with available paraffin blocks and unstained slides to study the association between clinicopathologic features and CD30 expression.

Clinical features of ENKTL
The clinical features of 91 patients with ENKTL are shown in Table 4. The upper aerodigestive tract, in Table 1  particular the nasal cavity, was the most common site of presentation, followed by the skin, orbit, and gastrointestinal tract. The distribution of clinical stages was as follows: stage I disease in 37 (40.7%) patients, stage II in 34 (37.4%) patients, stage III in 5 (5.5%) patients, and stage IV in 15 (16.4%) patients. Forty-one (45.1%) patients had regional lymph node involvement. Eighty-one patients were tested for plasma EBV DNA using Q-PCR before treatment: 39 (48.1%) were positive, with EBV DNA loads ranging from 1.07 × 10 3 to 2.49 × 10 6 copies/mL; 42 (51.9%) were negative for viral loads.
Fifty patients underwent bone marrow examination, and one had bone marrow involvement.
There was also no association between CD30 positivity and age, sex, B symptoms, Ki-67 index, clinical stage, Ann-Arbor stage, lactate dehydrogenase level, primary site, therapeutic modality, plasma EBV-DNA status, or relapse (Table 5). However, patients who tested positive for plasma EBV-DNA had longer OS (P = 0.008) and PFS (P = 0.012) than patients negative for plasma EBV DNA (Fig. 4).

Discussion
Here we report a large cohort of 622 patients with mature T-cell and NK-cell lymphomas from a single institution in South China and outlined a basic subtype distribution according to the revised 2016 WHO classification system. We identified ENKTL as the most common mature T-cell and NK-cell lymphoma subtype (51.0%) and confirmed a high frequency of CD30 expression (47.3%) in ENKTL cases.
According to the 2008 WHO classification of tumors of hematopoietic and lymphoid tissues, mature T-cell and NK-cell lymphomas compose a group of heterogeneous lymphomas, including ENKTL, AITL, ALCL, PTCL-NOS, and other rare subtypes [1]. Recently, a revision of the WHO classification of lymphoid neoplasms and the accompanying monograph has been published [3]. This revised edition included refinements of rare types of lymphoma and introduced molecular changes in the definition of certain subtypes. Herein we classified 622 cases of mature T-cell and NK-cell lymphoma according to this revised WHO classification, which yielded new incidence data based on a large institutional study.
Within this cohort, ENKTL was the most common subtype, accounting for 51.0% of cases over a 4-year period. Similar frequencies have been reported in North China [10] and Japan [11], but these frequencies are much higher than those reported in the United States and in European countries [2]. Recently, Li et al. [12] identified the genetic risk of ENKTL based on a genome-wide association study of cases and controls from Guangdong province in South China, which indicated a genetic predisposition to ENKTL among this population.
CD30 is a member of tumor necrosis factor (TNF) receptor super-family and exerts numerous biological functions, including cell cycle arrest, apoptosis, and activation of the pro-survival transcription factor nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB). CD30 is preferentially expressed in activated B cells but is also expressed in T cells and NK cells [13]. Among malignant lymphomas, CD30 is expressed in Hodgkin and Reed-Sternberg cells of Hodgkin lymphoma and in almost all neoplastic cells of ALCL. Several studies with small case series have reported CD30 expression in mature T-cell and NK-cell lymphoma, with the reported percentage of ENKTL patients that were CD30-positive ranging from 36.4% to 41% [10,[14][15][16]. In the current study, CD30 was expressed in 47.3% of cases, which is much higher than the rate reported in another study of mature T-cell and NK-cell lymphomas (with the exception of ALCL) [17]. CD30 has been reported to be more frequently expressed in ENKTL than in other subtypes of mature T-cell and NK-cell lymphomas, and this may be due to the concomitant presence of EBV in ENKTL patients [1,18]. A type II latency pattern of EBV infection has been observed in ENKTL patients, whereby tumor cells express EBERs, EBV nuclear antigen-1 (EBNA-1), latent membrane protein (LMP)-1, and LMP-2. LMP-1 is the most critical EBV element involved in cell proliferation. It induces the expression of B-cell lymphoma-2 (Bcl-2) and TNF-α-induced protein-3 to inhibit apoptosis. LMP-1 also up-regulates c-Myc, interleukin (IL)-6, and IL-10 to accelerate cell proliferation. LMP-1 has also been shown to up-regulate CD30 [18]; thus, LMP-1 expression may be associated with the high rate of CD30 expression among patients with ENKTL. The prognostic significance of CD30 expression in lymphomas is not well reported. Hu et al. [9] reported that CD30 expression was associated with a favorable prognosis among patients with EBV-positive diffuse large B-cell lymphoma. Bisig et al. [19] reported that patients with CD30-positive PTCL-NOS tended to have a better clinical outcome than those with CD30-negative PTCL-NOS. However, the prognostic value of CD30 expression among patients with ENKTL is controversial. In one study comprising 36 patients with ENKTL, CD30 expression appeared to be associated with a favorable outcome [10]. In another study of 22 patients with ENKTL, Hong et al. [16] reported that CD30 expression was associated with a poor prognosis. Kim et al. [14] (n = 72) and Kuo et al. [15] (n = 22) revealed that there was no difference in OS among CD30-positive patients versus CD30negative patients. In our study, CD30 expression did not exhibit a prognostic significance. Since we included more cases than most (if not all) studies in the current literature, we speculate that CD30 may not be a prognostic marker for ENKTL. It is likely that a selection bias exists in our cohort because patients in our study tended to present at earlier disease stages than patients in other studies. Studies with larger referral bases are needed to determine the significance of CD30 on patient prognosis.
Although it does not appear that CD30 expression has a prognostic role in ENKTL, CD30 is a molecular target for brentuximab vedotin, an anti-CD30 monoclonal antibody, which has led to improved outcomes in patients with refractory classic Hodgkin lymphoma and ALCL [6,20,21]. Horwitz et al. [22] observed objective responses to brentuximab vedotin in patients with relapsed T-cell lymphomas, and these responses were seen among patients regardless of their CD30 expression status in the tumor samples. More recent researches also showed that treatment with brentuximab vedotin led to complete remission in patients with CD30-positive refractory extranodal NK/T-cell lymphoma [23] and cutaneous T-cell lymphoma [24]. Therefore, targeted therapy with an anti-CD30 antibody represents a promising management of patients with refractory ENKTL that expresses CD30. A phase 2 clinical trial is currently in processing for patients with T-cell lymphomas (including ENKTL) to determine the effect of brentuximab vedotin on CD30-positive cases [25].
Other potential therapeutic modalities include targeting EBV antigens or using adoptive immune therapy in patients with circulating EBV DNA [26]. In our study cohort, plasma EBV DNA was detected in 48.1% of the tested cases prior to treatment initiation; the presence of EBV DNA was associated with a poorer prognosis. This result is similar to that of a previous report [27]. Plasma EBV DNA is derived from apoptotic tumor cells. It is an accurate surrogate biomarker of lymphoma load and a prognostic factor for patients treated with conventional chemotherapy [28]. However, we did not observe an association between plasma EBV DNA load and CD30 expression.
In conclusion, ENKTL is the most common subtype of mature T-cell and NK-cell lymphoma treated at our institution, a reference cancer center in South China. We identified that CD30 was expressed in a high percentage of ENKTL cases and might be a potential therapeutic molecular target for this disease; however, CD30 was insufficient as a prognostic marker. Clinical trials are needed to determine the effect of anti-CD30 therapy in patients with ENKTL that express CD30.