Nodular lymphocyte-predominant Hodgkin lymphoma (NLPHL), a rare subtype of Hodgkin lymphoma, is an indolent tumor with frequent instances of disease recurrence but a favorable prognosis. To the best of the authors' knowledge, there are only limited descriptions of NLPHL in the cytology literature because it was only formally recognized as a distinct entity in 1994.
In the current study, all cases of NLPHL diagnosed on excisional biopsy (n = 6 cases) at the study institution between 2000 and 2011 that had undergone previous fine-needle aspiration (FNA) were reviewed, with a focus on cytomorphologic features.
Four of 6 cases were termed benign on FNA; however, there was retrospective recognition of characteristic LP cells in all cases. Unlike classical Hodgkin lymphoma, the tumor cells of NLPHL were often found to be mononucleate and presented in a background of small lymphocytes. Other features identified included epithelioid histiocytes and numerous bare atypical nuclei.
Nodular lymphocyte-predominant Hodgkin lymphoma (NLPHL), a rare subtype of Hodgkin lymphoma, was only formally recognized by the Revised European-American Lymphoma (REAL) classification in 1994.1 This indolent B-cell neoplasm is characterized by frequent instances of disease recurrence but has an overall favorable prognosis. It has a marked male predominance, and usually presents at an early stage. NLPHL shares more clinical and biological features with non-Hodgkin lymphomas (NHLs) derived from germinal center B cells rather than classical Hodgkin lymphoma (CHL).2 The management of NLPHL is controversial.2
The neoplastic cells of NLPHL are usually large with minimal cytoplasm, and demonstrate multilobated or folded nuclei resembling “popcorn.” Nucleoli in the malignant cells (lymphocyte-predominant [LP] cells) are usually smaller than those noted in CHL. However, a subset of LP cells are binucleate with large prominent nucleoli, and are morphologically indistinguishable from the Reed-Sternberg (RS) cells of CHL.3 Using immunohistochemistry, LP cells are positive for CD45 and CD20 and negative for CD30 and CD15, which is the opposite expression pattern from true RS cells. A subset of LP cells (approximately 40%) expresses epithelial membrane antigen (EMA), whereas true RS cells are negative. Unlike CHL, NLPHL cases never demonstrate positivity by in situ hybridization for Epstein-Barr virus (EBV)-encoded RNA. On histologic sections, LP cells often are encircled by a ring of T cells; otherwise, the background nodules in NLPHL are B cell-rich. There tends to be an increase in epithelioid histiocytes and plasma cells, but no prominence of eosinophils, in contrast to CHL. Flow cytometry is usually not sensitive enough to detect a B-cell clone.
Fine-needle aspiration (FNA) has been increasingly used to aid in the diagnosis of NHLs, particularly the more common subtypes such as chronic lymphocytic leukemia/small lymphocytic lymphoma and diffuse large B-cell lymphoma (DLBCL).4 An abundance of neoplastic cells on aspirate smears and the ability to confirm clonality by flow cytometry permit cytopathologists to make such diagnoses on limited FNA samples. However, the recognition of Hodgkin lymphoma on FNA can be difficult due to the scarcity of diagnostic tumor cells, especially in the nodular sclerosis type of CHL,5 and some authors even suggest that tissue biopsy is needed in the majority of cases.6 Rarer subtypes of NHL, such as T-cell/histiocyte-rich large B-cell lymphoma (TCHRLBCL) or primary mediastinal B-cell lymphoma also may require excisional biopsy for accurate classification.4
The description of NLPHL is limited in the cytology literature,4, 7 with, to our knowledge, no studies specifically addressing its cytomorphologic features on FNA. The current study examines cases of established histologic diagnoses of NLPHL with previous FNA of the same mass, and correlates the cytomorphology and clinicopathologic characteristics of the 2 specimens.
MATERIALS AND METHODS
We reviewed the cytopathology of all cases of NLPHL diagnosed on subsequent excisional biopsy at the study institution between 2000 and 2011. Demographics, clinical history, and follow-up information were obtained. FNA was performed with ultrasound guidance by a radiologist using a 22-gauge needle. Air-dried, Diff-Quik–stained smears and alcohol-fixed, Papanicolaou-stained smears were prepared. Needle rinses were performed in balanced salt solution and submitted in entirety for flow cytometry in all 6 cases; no cell block preparations were made. A core needle biopsy specimen was obtained in 3 cases. Cytologic material was reviewed to assess for the presence of LP cells, and to attempt to identify other morphologic features suggestive of NLPHL.
There were 6 cases, with 5 male patients and 1 female patient; the average age was 49 years (range, 32 years-63 years). All patients presented with a mass, ranging in size from 3.0 cm to 4.8 cm. The mass locations were the groin in 3 cases, the axilla in 1 case, the parotid gland in 1 case, and the submandibular gland in 1 case. Cytologic diagnoses were benign/reactive in 4 cases and “atypical” in 2 cases. Flow cytometry demonstrated no phenotypic abnormalities in 5 cases and was cancelled in 1 case because of low cellularity. Surgical follow-up on all 6 cases resulted in a diagnosis of NLPHL. The extent of disease at the time of definitive diagnosis was stage I in 1 case, stage II in 2 cases, and stage IIIA in 3 cases (Ann Arbor Staging System). Two patients had a remote history of CHL, but this material was unavailable for review at the time of our study and thus it is unclear whether this represented true CHL or NLPHL. One patient had a remote history of DLBCL. The patients were managed in a variety of ways including watchful waiting, chemotherapy, and radiotherapy. Follow-up ranged from 1 month to 6.6 years with an average of 2 years. Two cases were diagnosed quite recently at our institution and therefore only minimal follow-up was available. One patient had refractory disease and underwent bone marrow transplant. All patients were alive at the time of last follow-up. Clinicopathologic features are summarized in Table 1.
Table 1. Clinicopathologic Features of 6 Cases of Nodular Lymphocyte-Predominant Hodgkin Lymphoma
Mass Size, cm
Ann Arbor Stage
BMT indicates bone marrow transplant; chemo, chemotherapy; CHL, classical Hodgkin lymphoma; DLBCL, diffuse large B-cell lymphoma; rad, radiotherapy.
2 y, chemo/rad treatment
6.6 y, no treatment
1.7 y, refractory to treatment, followed by BMT
1.6 y, no treatment
2 mo, no treatment
1 mo, no treatment
The aspirate smears demonstrated abundant cellularity in 5 of the 6 cases. The low-power impression in all cases was similar to that of a reactive lymph node, showing numerous small lymphocytes admixed with many epithelioid histiocytes (Fig. 1) present both singly and in small clusters, with some forming small granulomata. Only on higher magnification was the presence of large, atypical lymphoid cells evident. Cytomorphologic features of these large atypical cells included: 1) large mononuclear cells with multilobated or convoluted nuclei (LP cells); 2) very large bare nuclei; and 3) variable presence of prominent nucleoli. These features are summarized in Table 2.
Table 2. Cytomorphologic Features of 6 Cases of Nodular Lymphocyte-Predominant Hodgkin Lymphoma
Convoluted, “popcorn”-type nuclei were more prominent on Papanicolaou-stained material (Fig. 2), while markedly enlarged nuclei and prominent nucleoli were better appreciated on slides stained using Diff-Quik (Fig. 3). The “popcorn”-type LP cells were also present on a cytospin preparation in 1 case (Fig. 4). Bare nuclei were readily identified in smears using both staining methods (Fig. 5). Many of the LP cells had a suggestion of “rimming” by small lymphocytes, similar to features noted in histologic sections (Fig. 6), although this may not actually correlate to the histologic feature, as the origin of these lymphocytes (B cell or T cell) was not established on the direct smears. In all cases, the background lymphocytes were small and monotonous. Scattered plasma cells were present in several cases, but eosinophils were not appreciated in any of the cases. In 3 cases, a core needle biopsy was available and 2 of these demonstrated LP cells within the tissue noted on retrospective review; however, immunostains (such as CD20) were not performed in these 2 cases. In retrospect, all 6 cases had some combination of these characteristic NLPHL findings, although they were often subtle.
FNA is increasingly being used in the assessment of lymphadenopathy to spare patients from a potentially unnecessary, invasive excisional biopsy. Its usefulness in the diagnosis of infectious processes, sarcoidosis, and common types of NHL have been well documented.8-11 In several rarer types of lymphoma, however, cytomorphologic features are not well described.
Such is the case with NLPHL, a rare entity that may be mistaken for a reactive process or CHL on cytology material. Immunohistochemical staining can suggest the diagnosis if characteristic LP cells from a cell block or core needle biopsy express CD20 and CD45 (with or without EMA), and are negative for CD30 and CD15. However, in practice, the interpretation of stains can be limited by the scarcity of diagnostic tumor cells. Additional features that distinguish NLPHL from CHL are the absence or paucity of eosinophils, neutrophils, and plasma cells as well as a lack of necrosis and metachromatic stroma, which is suggestive of fibrosis.
Although reactive lymphadenopathy can also be included in the differential diagnosis for NLPHL, smears of the former entity should demonstrate lymphocytes in various stages of transformation, with a range of cell sizes. There should be scattered immunoblasts, tingible body macrophages, and a polymorphous population of follicular center cells. In contrast, the cases of NLPHL in the current series were dominated by a monotonous background of small mature lymphocytes, and lacked the degree of polymorphism that would be expected in a large reactive lymph node.
Other entities that can be associated with RS-like cells include NHLs (eg, T-cell lymphoma, anaplastic large cell lymphoma, and chronic lymphocytic leukemia/small lymphocytic lymphoma), posttransplant lymphoproliferative disorders (PTLDs), and EBV infections. In a separate study from our institution, Iacobuzio-Donahue et al described 9 such cases with RS-like cells. Seven of these 9 cases were explained by evolving lymphoma, PTLD, or EBV infection.12 The authors emphasized that these pitfalls should be considered, and further evaluation advised, when RS-like cells are encountered in a lymph node without other diagnostic features of Hodgkin lymphoma. Similar cells rarely have been described in patients with systemic lupus erythematosus as well.13 Immunohistochemical studies and flow cytometry can be of use in distinguishing NLPHL from these entities if adequate material is available. NLPHL has a characteristic immunoprofile, typically does not demonstrate a clonal population by flow cytometry, and is negative for EBV-encoded RNA by in situ hybridization.
A final diagnostic consideration would be TCHRLBCL, a very rare, aggressive lymphoma that usually presents in middle-aged men with “B” symptoms and an advanced stage of disease. To the best of our knowledge, there is a very limited description of this entity in the cytology literature.4, 14 There is considerable morphologic overlap with NLPHL, including a predominance of reactive lymphocytes and histiocytes with rare neoplastic cells and a lack of background eosinophils. Furthermore, the 2 entities share a similar immunoprofile and both often show no evidence of a clone by flow cytometry. Although the malignant cells of TCHRLBCL may mimic LP cells, they usually demonstrate more size variation with morphology ranging from centroblastic to pleomorphic, more suggestive of RS cells.3 TCHRLBCL typically lacks both nodular architecture and an intact dendritic cell network by CD21 and/or CD23 immunostaining, whereas the presence of these features is important in the diagnosis of NLPHL. The distinction is not always straightforward on tissue sections either, because NLPHL can demonstrate areas of diffuse growth resembling TCHRLBCL. Fan et al even describe a “T cell-rich B-cell lymphoma-like” pattern of NLPHL that can only be classified separately from TCHRLBCL after the demonstration of a nodular component in association with the diffuse area in the excisional biopsy.15
Cases of NLPHL can also progress, becoming histologically indistinguishable from de novo TCHRLBCL3; thus, the relation between these 2 entities remains controversial.3 The distinction should be made by histologic examination in correlation with clinical presentation, and is of utmost importance because of the marked difference in treatment and prognosis. Although NLPHL is a low-grade, indolent lymphoma that some clinicians defer to treat, TCHRLBCL is an aggressive disease that usually has a poor outcome.
In our series of NLPHL cases, characteristic LP cells were retrospectively identified in all cases, including the 4 that were originally termed benign. In the cases that were initially misdiagnosed on cytology, factors contributing to the incorrect impression of benignity include 1) a paucity of characteristic LP cells; 2) the misinterpretation of LP cells as follicular center cells or immunoblasts in a background of reactive lymphocytes; 3) an abundance of epithelioid histiocytes, giving the impression of a reactive lymph node; 4) a lack of the typical milieu seen in CHL, leading to a low level of suspicion for Hodgkin lymphoma; 5) flow cytometry results demonstrating no phenotypic abnormality, making NHL unlikely, and 6) nonspecific clinical findings/lack of “B” symptoms. Fortunately, all patients underwent further evaluation with excisional biopsy that allowed the histologic diagnosis to be made in a timely manner.
In conclusion, although making a primary definitive diagnosis of NLPHL on FNA would be challenging, characteristics suggestive of this entity on aspirate smears include enlarged multilobated nuclei, large bare nuclei, prominent nucleoli, and scattered epithelioid histiocytes. We do not suggest that an outright diagnosis of NLPHL should be made on FNA, but rather that cases with such morphologic features should be flagged as “atypical” to prompt an excisional biopsy for more precise classification. NLPHL was commonly misdiagnosed as benign reactive lymphoid tissue in our series, and therefore a careful search on high magnification for LP cells is recommended in the evaluation of lymph node FNAs, especially in adult males with large (> 3 cm) lymph nodes to facilitate timely diagnosis and treatment in these patients.