Abstract
Hodgkin lymphoma (HL) and idiopathic multicentric Castleman disease (iMCD) are markedly different conditions. However, in some cases, histological similarities caused by elevated cytokines, including interleukin-6, can lead to a misdiagnosis of HL as Castleman disease (CD). We herein report a patient with HL who had been diagnosed with CD by an expert panel and for whom an additional biopsy was useful for determining the correct diagnosis. Furthermore, we analyzed the positron emission tomography/computed tomography findings at the diagnosis and found that the maximum standardized uptake value was useful for distinguishing HL from iMCD.
Keywords: Hodgkin lymphoma, Castleman disease, PET/CT, SUVmax, interleukin-6
Introduction
Hodgkin lymphoma (HL) and idiopathic multicentric Castleman disease (iMCD) should not be mistaken for one another. HL is a hematologic malignancy characterized by large, typically multinucleated cells called Hodgkin and Reed-Sternberg (HRS) cells (1), while iMCD represents a group of unusual nonmalignant lymphoproliferative disorders (2). In iMCD, interleukin (IL)-6 hyperproduction plays a critical role in forming pathological features (2). However, elevated cytokines, including IL-6, produced by HL sometimes give secondary modifications to histological findings (3), and HL tumor cells are surrounded by extensive nonspecific infiltration of inflammatory cells, including T cells, eosinophils, and plasma cells, which also appear in plasma cell-type iMCD.
These histological similarities can lead to a misdiagnosis of HL as Castleman disease (CD). Thus, when physicians sense a discrepancy between the clinical features and pathological diagnosis, it is important that they do not hesitate to perform an additional biopsy from another region.
We herein report a patient with HL who had been diagnosed with CD by an expert panel and for whom an additional biopsy was useful for determining the correct diagnosis. In addition, we analyzed the positron emission tomography/computed tomography (PET/CT) findings at the diagnosis and found that the maximum standardized uptake value (SUVmax) was useful for distinguishing HL from iMCD.
Case Report
A 27-year-old woman was referred to our department with suspicion of iMCD. At the previous hospital, she had presented with enlargement of the right supraclavicular lymph node, a fever of 37.4°C, nocturnal sweating, and weight loss of 4 kg/year. She had a moderate performance status (Eastern Cooperative Oncology Group performance status 1) and no significant medical history, including immune deficiencies. She was a nonsmoker and consumed alcohol sparingly. She did not have any remarkable family history.
18F-fluorodeoxyglucose (FDG)-PET/CT revealed abnormal uptakes in the bilateral supraclavicular lymph nodes (right: 25 mm, left: 21 mm) and mediastinal lymph nodes (max 49 mm), with an SUVmax of 11.5 (Fig. 1). A lymph node biopsy from a left supraclavicular lymph node (SUVmax 5.8) revealed atrophic lymph follicles, widening of the paracortex region, and proliferation of small vessels with tufted endothelial cells. Plasma cells were enriched in the interfollicular areas. HL-associated large, atypical cells were absent on Hematoxylin and Eosin staining (Fig. 2). CD15- or CD30-positive large cells were not identified on immunostaining. She was diagnosed with plasma cell-type iMCD and referred to our hospital for treatment.
Figure 1.
FDG-PET/CT findings. The FDG-PET/CT image shows multiple lymph node enlargement. The FDG uptake was highest in the mediastinal lymph node (SUVmax 11.5).
Figure 2.
Histological appearance of lymph node biopsy from a left supraclavicular node (1st biopsy). Hematoxylin and Eosin staining, original magnification A: ×40, B: ×400. A: Atrophic lymph follicles. B: Plasma cells were enriched in the interfollicular areas. HL-associated large, atypical cells were absent.
We conducted further work-up upon her arrival. Laboratory findings showed increased leukocytosis, hypergammaglobulinemia, elevated erythrocyte sedimentation rate and C-reactive protein (CRP) values, and extremely high IgE values (Table). These findings, including normocytic normochromic anemia, elevated CRP levels, and hypergammaglobulinemia, as well as her clinical features, such as enlarged lymph nodes, splenomegaly, a high fever, and nocturnal sweating, were consistent with iMCD (2). Furthermore, an in-house pathological review by an expert panel also reached the same iMCD conclusion. We did not, however, have confidence in the diagnosis due to the enlarged lymph nodes and the abnormally high FDG uptakes in the mediastinal lymph nodes, which is atypical for iMCD (4).
Table.
Laboratory Findings
| Hematology | Serology | |||||||
| WBC (103/µL) | 14.3 | (3.3-8.6) | CRP† (mg/dL) | 5.81 | (<0.14) | |||
| Neutro (%) | 78.6 | (40.6-76.4) | β2-MG (mg/L) | 2.31 | (0.8-2) | |||
| Lymph (%) | 15.5 | (16.5-49.5) | ||||||
| Mono (%) | 3.1 | (2.0-10.0) | IgG (mg/dL) | 2,105 | (861-1,747) | |||
| Eosino (%) | 2.5 | (0.0-8.5) | IgA (mg/dL) | 460 | (93-393) | |||
| Baso (%) | 0.3 | (0.0-2.5) | IgM (mg/dL) | 124 | (50-269) | |||
| RBC (106/µL) | 4.34 | (3.86-4.92) | IgE (U/mL) | 11,221 | (<232) | |||
| Hb (g/dL) | 10.9 | (11.6-14.8) | ||||||
| MCV (fL) | 81.8 | (83.6-98.2) | sIL-2R (U/mL) | 1,535 | (122-496) | |||
| Plt (103/µL) | 352 | (158-348) | ||||||
| ESR* 1h (mm) | 91 | (3-15) | C3 (mg/dL) | 130 | (73-138) | |||
| C4 (mg/dL) | 26 | (11-31) | ||||||
| Biochemistry | CH50 (U/mL) | 80.6 | (31.6-57.6) | |||||
| TP (g/dL) | 8.1 | (6.6-8.1) | ||||||
| Alb (g/dL) | 3.6 | (4.1-5.1) | Infection | |||||
| CK (U/L) | 21 | (41-153) | HBs‡ antibody | Negative | ||||
| AST (U/L) | 10 | (13-30) | HCV§ antibody | Negative | ||||
| ALT (U/L) | 9 | (7-23) | HIV|| | Negative | ||||
| LDH (U/L) | 187 | (124-222) | TP¶ antibody | Negative | ||||
| ALP (U/L) | 102 | (38-113) | ||||||
| γGTP (U/L) | 35 | (9-32) | ||||||
| ChE (U/L) | 239 | (201-421) | ||||||
| UA (md/dL) | 5.8 | (2.6-5.5) | ||||||
| BUN (mg/dL) | 10 | (8-20) | ||||||
| Cr (mg/dL) | 0.54 | (0.46-0.79) | ||||||
| Na (mmol/L) | 139 | (138-145) | ||||||
| K (mmol/L) | 3.9 | (3.6-4.8) | ||||||
| Cl (mmol/L) | 103 | (101-108) | ||||||
| Ca (mmol/L) | 9.3 | (8.8-10.1) |
*ESR: erythrocyte sedimentation rate, †CRP: C-reactive protein, ‡HBs: hepatitis B surface, §HCV: hepatitis C virus, ||HIV: human immunodeficiency virus, ¶TP: Treponema pallidum
Therefore, we performed a thoracoscopic biopsy of the mediastinal lymph node that showed the highest FDG uptake. A microscopic examination revealed nodular sclerosis with extensive inflammatory cell infiltration, which included eosinophils and plasma cells. Hodgkin and Reed/Sternberg (HRS) cells and lacunar cells were also evident. On immunohistochemistry, HRS cells were positive for CD30 and negative for CD15. The MIB-1 index was over 90%. Epstein Barr Virus-encoded RNAs (EBERs) were detected in HRS cells by in situ hybridization (Fig. 3). Based on these observations, we diagnosed her with nodular sclerosis classic HL [advanced stage III; lesions found in bilateral supraclavicular lymph nodes, mediastinal lymph nodes and spleen, according to the Lugano classification (5) and her International Prognostic Score of 2 (6)].
Figure 3.
Histological appearance of a thoracoscopic biopsy from a mediastinal node (2nd biopsy). Hematoxylin and Eosin staining, original magnification A: ×100, B: ×400. Nodular sclerosis was observed (A) with extensive infiltration of inflammatory cells, including eosinophils and plasma cells. HRS cells and lacunar cells were also evident (indicated by yellow arrows) (B).
Six cycles of brentuximab vedotin (BV)+adriamycin/vinblastine/dacarbazine chemotherapy (BV: 1.2 mg/kg day 1, 15; adriamycin: 25 mg/m2 day 1, 15; vinblastine: 6 mg/m2 day 1, 15; and dacarbazine: 375 mg/m2 day 1, 15) were started. This led to partial remission. After the initiation of therapy, her clinical features, such as a high fever, fatigue, nocturnal sweating, and weight loss, had normalized.
Discussion
HL is characterized by the pathological feature of multinucleated Reed-Sternberg cells in the patient's lymph node tissue (1). To reach a diagnosis, the microscopic analysis of enlarged lymph node tissue is critically important, and nodular sclerosis-type HL is characterized by collagen fibrosis, with large tumor nodules showing scattered lacunar cells and classic HRS cells set against a background of eosinophils, plasma cells, and other cells to varying degrees (7). Immunohistochemical findings show CD30+, EBER+, and CD15- large, atypical cells.
iMCD is a group of unusual nonmalignant lymphoproliferative disorders. The median age at the diagnosis is 30-40 years old for unicentric CD and 50-60 years old for multicentric CD (8). Our 27-year-old patient was relatively young for the development of multicentric CD. Approximately 120 new cases are diagnosed as iMCD each year in Japan, and the annual incidence is estimated at 1 per 1 million people (9). iMCD is typically accompanied by slowly progressing clinical features, such as an enlargement of lymph nodes, a high fever, general malaise, and weight loss. Laboratory findings, such as elevated CRP levels, elevated IL-6 levels, microcytic anemia, hypoalbuminemia, low lactate dehydrogenase levels, hypergammaglobulinemia, elevated IgE values, and elevated vascular endothelial growth factor levels, are observed in many cases (2). According to the iMCD diagnostic criteria from the CD, TAFRO, and Related Disease Research Group (Japan) (10), having enlarged (>1 cm) lymph nodes with histological findings compatible with iMCD is essential for a definitive diagnosis. It is also necessary to exclude malignant neoplasms. Thus, a histological analysis of enlarged lymph node tissue is critically important.
CDs are histologically divided into four subgroups according to the Japanese guidelines: hyaline vascular type, plasma cell type, hypervascular type, and mixed type (9). The features of the plasma cell type are an increased number of follicles with large hyperplastic germinal centers and an increased number of plasma cells (11). The anti-IL-6 receptor therapeutic agent tocilizumab (12) is used for patients with moderate to severe iMCD and is effective in disease control. However, it is difficult to achieve a cure, and patients need to continue treatment for the rest of their lives.
As stated above, CD and HL present with similar symptoms, such as enlarged lymph nodes, splenomegaly, a high fever, and nocturnal sweating. Both plasma cell-type iMCD and nodular sclerosis-type HL in particular are noted for their nonspecific inflammatory cell infiltration. If there are no specific HL features, such as large tumor nodules or HRS cells, making an accurate diagnosis is difficult. In our patient, we speculated that IL-6 played an important role in the histological similarity between this case and iMCD. IL-6 promotes B-cell differentiation into plasma cells and is a growth factor for plasma cells (13). In many HL cases, IL-6 expression is activated by HLXB9 and regulated by PI3K signaling (3). A recent study revealed that IL-6 markedly influences the tumor microenvironment in classical HL (14). Elevated IL-6 may cause plasma cell enrichment in neighboring lymph nodes without any HL-associated large, atypical cell changes. As the treatment strategies differ markedly between the two diseases, thorough investigations when the patient arrives are critical in providing a reliable diagnosis.
In the current case, the PET/CT finding was key to raising doubts about the fitness of the diagnosis, although PET/CT imaging for CD is not covered by Japanese public insurance. Oksenhendler et al. recently reported the FDG uptake in CD (4). The SUVmax differed among disease types, with the median SUVmax being 5.1 (range: 2.2-10, n=21) in unicentric CD and 4.5 (range: 2.2-15, n=17) in iMCD; furthermore, CD regions had lower SUVmax values overall than HL regions (range 9.5-13.8, n=4). To determine the utility of PET/CT in distinguishing HL from iMCD, we retrospectively selected 64 patients with HL (HL group) and 65 patients with iMCD (iMCD group) from 2 institutes: The Jikei University Hospital (Tokyo, Japan) and Hôpital Saint-Louis (Paris, France). Each patient underwent FDG-PET/CT at the diagnosis. The median SUVmax differed between the groups, being 13.85 (range 3.3-50.5) in the HL group and 5.4 (range 2.2-16.8) in the iMCD group (Fig. 4). Furthermore, the FDG uptake in the iMCD group was significantly lower than that in the HL group (p<0.0001, Mann-Whitney U test). The receiver operating characteristic curve analysis showed that the SUVmax of 8.235 could predict the diagnosis of HL or iMCD with a sensitivity of 84.62% and a specificity of 85.94% using the Youden index (Fig. 5). This analysis revealed that the SUVmax at the diagnosis using FDG-PET/CT was useful for distinguishing HL from iMCD.
Figure 4.
The comparison of the FDG uptake between the HL and iMCD groups. The Mann-Whitney U test was used to assess differences among the three groups. The FDG uptake in the iMCD group was significantly lower than that in the HL group (p values are indicated by stars, ****p<0.0001).
Figure 5.
The ROC curve for predicting the diagnosis of HL versus CD based on the SUVmax.
Although additional invasive surgical procedures are burdensome for patients, the current case taught us that it is crucial to perform an additional biopsy when a physician feels there is a discrepancy between the clinical features and pathological diagnosis. A case of HL occurrence after an 11-year follow-up for iMCD has been reported (15). Thus, iMCD patients with a high SUVmax should be followed closely in the long term, keeping the possibility of HL occurrence in mind.
In addition to the SUVmax findings in PET/CT, the maximum lymph node size on CT can help distinguish HL from iMCD. The size of the swollen lymph node in iMCD is smaller than that in unicentric CD (mean value 5.7 cm) (16), while HL nodes vary in size. In cases of multiple lymph nodes associated with large lesions or bulky masses (>10 cm), it is necessary to consider HL rather than iMCD.
Whether or not HL overlaps with iMCD remains an open question. Indeed, there are several reports regarding the coexistence of HL and iMCD (17-19). In the current case, we believe our diagnosis was HL-compatible, as chemotherapy was effective for whole lesions and improved clinical symptoms, including B symptoms. Even if iMCD is not diagnosed, it is necessary to start combination chemotherapy for HL, considering the speed at which HL progresses. Thus, it is important to carefully monitor cases of HL to avoid overlooking the signs of relapse as well as any potential underlying iMCD.
iMCD is a rare disease, and it is critically important to rule out other malignancies. In HL, inflammatory modifications caused by tumors can display iMCD-like histological features. We should remain cautious of deciding to start treatment for iMCD even when patients have been diagnosed with it.
The authors state that they have no Conflict of Interest (COI).
Acknowledgement
We thank Drs. Atsuhito Nakayama, Atsuro Oishi, and Akihide Yoshimi (Division of Cancer RNA Research, National Cancer Center, Japan) for their valuable expert advice and the manuscript writing/editing process.
References
- 1.Hodgkin.. On some morbid appearances of the absorbent glands and spleen. Med Chir Trans 17: 68-114, 1832. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Dispenzieri A, Fajgenbaum DC. Overview of Castleman disease. Blood 135: 1353-1364, 2020. [DOI] [PubMed] [Google Scholar]
- 3.Nagel S, Scherr M, Quentmeier H, et al. HLXB9 activates IL6 in Hodgkin lymphoma cell lines and is regulated by PI3K signalling involving E2F3. Leukemia 19: 841-846, 2005. [DOI] [PubMed] [Google Scholar]
- 4.Oksenhendler E, Boutboul D, Fajgenbaum D, et al. The full spectrum of Castleman disease: 273 patients studied over 20 years. Br J Haematol 180: 206-216, 2018. [DOI] [PubMed] [Google Scholar]
- 5.Cheson BD, Fisher RI, Barrington SF, et al. Recommendations for initial evaluation, staging, and response assessment of Hodgkin and non-Hodgkin lymphoma: the Lugano classification. J Clin Oncol 32: 3059-3068, 2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Hasenclever D, Diehl V; International Prognostic Factors Project on Advanced Hodgkin's Disease. A prognostic score for advanced Hodgkin's disease. N Engl J Med 339: 1506-1514, 1998. [DOI] [PubMed] [Google Scholar]
- 7.WHO classification of tumours of haematopoietic and lymphoid tissues. 4th (revised) ed. Swerdlow SH, Campo E, Harris NLet al. , Eds. World Health Organisation, Lyon, 2017. [Google Scholar]
- 8.Ye B, Gao SG, Li W, et al. A retrospective study of unicentric and multicentric Castleman's disease: a report of 52 patients. Med Oncol 27: 1171-1178, 2010. [DOI] [PubMed] [Google Scholar]
- 9.Yoshizaki K. [A reference guide for management of Castleman disease]. Rinsho Ketsueki (Jpn J Clin Hematol) 58: 97-107, 2017. [DOI] [PubMed] [Google Scholar]
- 10.Fujimoto S, Koga T, Kawakami A, et al. Tentative diagnostic criteria and disease severity classification for Castleman disease: a report of the research group on Castleman disease in Japan. Mod Rheumatol 28: 161-167, 2018. [DOI] [PubMed] [Google Scholar]
- 11.Keller AR, Hochholzer L, Castleman B. Hyaline-vascular and plasma-cell types of giant lymph node hyperplasia of the mediastinum and other locations. Cancer 29: 670-683, 1972. [DOI] [PubMed] [Google Scholar]
- 12.Nishimoto N, Kanakura Y, Aozasa K, et al. Humanized anti-interleukin-6 receptor antibody treatment of multicentric Castleman disease. Blood 106: 2627-2632, 2005. [DOI] [PubMed] [Google Scholar]
- 13.Hirano T. Interleukin 6 (IL-6) and its receptor: their role in plasma cell neoplasias. Int J Cell Cloning 9: 166-184, 1991. [DOI] [PubMed] [Google Scholar]
- 14.Gholiha AR, Hollander P, Glimelius I, et al. Revisiting IL-6 expression in the tumor microenvironment of classical Hodgkin lymphoma. Blood Adv 5: 1671-1681, 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Park J, Lee JE, Kim M, et al. Discordant lymphocyte-depleted classical Hodgkin's and peripheral T-cell lymphoma arising in a patient 11 years after diagnosis of multicentric Castleman's disease. Int J Hematol 98: 114-121, 2013. [DOI] [PubMed] [Google Scholar]
- 16.Talat N, Belgaumkar AP, Schulte KM. Surgery in Castleman's disease: a systematic review of 404 published cases. Ann Surg 255: 677-684, 2012. [DOI] [PubMed] [Google Scholar]
- 17.Lyapichev KA, You MJ, Vega F, Solis LM, Medeiros LJ. Classic Hodgkin lymphoma and Castleman disease: an entity appears to be emerging. Virchows Arch 477: 437-444, 2020. [DOI] [PubMed] [Google Scholar]
- 18.Mohtaram A, Afif M, Sghiri T, et al. Coexistence of Hodgkin's lymphoma and Castleman's disease: a case report with successful response to chemotherapy and radiotherapy. Case Rep Oncol Med 2013: 487205, 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Filliatre-Clement L, Busby-Venner H, Moulin C, Roth-Guepin G, Perrot A. Hodgkin lymphoma and Castleman disease: when one blood disease can hide another. Case Rep Hematol 2017: 9423205, 2017. [DOI] [PMC free article] [PubMed] [Google Scholar]