Introduction

TAFRO syndrome, first described in 2010, includes thrombocytopenia, anasarca (edema, pleural effusion and ascites), fever, reticulin myelofibrosis and organomegaly (hepatosplenomegaly and lymphadenopathy) [1]. Similar cases have since been reported [2–10]. Although some of the histopathological features of TAFRO syndrome are similar to those of mixed type of multicentric Castleman disease (MCD) [1], other clinical characteristics can distinguish between the two conditions. For example, polyclonal hypergammaglobulinemia, multiple lymphadenopathy, thrombocytosis, and a chronic clinical course are characteristics of MCD; whereas normal immunoglobulin level, thrombocytopenia, relatively small lymphadenopathy, marked pleural effusion, ascites, and edema, and acute or subacute onset and clinical course are characteristics of TAFRO. The differential diagnosis of TAFRO syndrome also includes disorders with similar symptoms, such as lymphoma, autoimmune disease, and acid-fast bacillus infection.

Some patients with TAFRO syndrome die of disease [1, 4, 6], whereas others may be saved by early aggressive treatment. Thus, determining the diagnostic criteria and treatment strategies for TAFRO syndrome may benefit these patients. Several meetings have evaluated the diagnosis and treatment of TAFRO syndrome [3]. However, its epidemiology, etiology, diagnosis and treatment have not been defined. Multicenter retrospective study in Japan, based on an on-line patient registry (UMIN000011809, https://www.facebook.com/CastlemanTAFRO), has therefore been initiated.

In addition the Ministry of Health, Labour and Welfare of Japan has organized and funded, through Health and Labour Sciences Research Grants for Research on Rare and Intractable Diseases, a nation-wide research team on TAFRO syndrome (H27-Nanchi, etc. (Nan)-General-008).

To promote the research on TAFRO syndrome, the research team on TAFRO syndrome first defined its preliminary diagnostic criteria and disease severity classification. Then we verified these criteria and classification using clinicopathologic data on 28 patients, who were reported in research meetings, with this condition and similar symptoms (e.g. MCD with serositis and thrombocytopenia), retrospectively.

Methods

Twenty-seven members of the TAFRO research group, including hematologists, rheumatologists, pulmonologists, pathologists, radiologist and basic researchers, attended a meeting on October 31,2015, to discuss the diagnostic criteria, disease severity classification and treatment strategies for TAFRO syndrome based on data from patients they have treated, those reported [1–12] to have this disease, and patients registered in the retrospective study. Subsequent on-line meetings have led to serious, active and enthusiastic discussion.

To verify these criteria and classification, 28 cases with TAFRO syndrome and similar conditions reported in previous meetings were analyzed. Thrombocytopenia was defined as <100,000 platelets/mm3; anasarca as either pleural effusion or ascites on CT-scan; fever/inflammation as body (axillar) temperature >37.5 °C and/or elevated CRP level >2 mg/dl; reticulin myelofibrosis as a pathological diagnosis using gitter staining; organomegaly as hepatomegaly, splenomegaly or lymphadenopathy on CT-scans; anemia as hemoglobin <10.0 g/dl; and renal insufficiency as serum creatinine >1.2 mg/dl in males or >1.0 mg/dl in females.

The study protocol was approved by the review board of Kanazawa Medical University and each collaborating institute.

Results

Preliminary diagnostic criteria and disease severity classification

The intensive discussions of a meeting held on Oct. 31, 2015 and several online meetings resulted in the 2015 version of the diagnostic criteria (Table 1) and disease severity classification (Table 2). Furthermore, we made recommendation of treatment strategy based on our clinical experiences (Table 3).

Table 1 The 2015 diagnostic criteria for TAFRO syndrome, as determined by All Japan TAFRO Syndrome Research Group in the Research Program for Intractable Disease by Ministry of Health, Labor and Welfare (MHLW) Japan
Full size table
Table 2 The 2015 disease severity classification for TAFRO syndrome
Full size table
Table 3 The 2015 treatment strategy for TAFRO syndrome
Full size table

Clinical features of reported cases and application of the diagnostic criteria

The 28 patients (12 male, 16 female) with TAFRO syndrome and similar conditions ranged in age from 21 to 78 with a median of 52 and a mean of 52.5 years old. The incidence of each of the parameters of TAFRO findings are shown in Table 4.

Table 4 Clinical and laboratory findings in all 28 patients investigated in this study, including the 18 patients with TAFRO syndrome
Full size table

According to the diagnostic criteria for TAFRO syndrome (Table 1), 18 of 28 patients (8 male, 10 female, median age 54, mean of 54.7 years old) in this series were diagnosed as TAFRO syndrome. By disease severity classification (Table 2) patients one (5.5 %), 11 (61.1 %), 4 (22.2 %) and 2 (11.1 %) were categorized into grades 1, 2, 3 and 4, respectively.

All patients in this study were negative for HHV8 (DNA and/or antibody), HIV and Epstein Barr virus (EBV). Some patients presented with anasarca (pleural effusion and ascites), thrombocytopenia and normo- to microcytic anemia on hemograms, renal dysfunction, ALP elevation, low level of LDH, relatively mild polyclonal hypergammopathy, some immunological abnormalities (but few specific autoantibodies), or elevation of IL-6 and/or vascular endothelial growth factor (VEGF). Other clinical findings included myelofibrosis, increased levels of megakaryocytes in bone marrow, and small or unclear lymphadenopathy (<1.5 cm in diameter) with weak accumulation of 18F-fluorodeoxy glucose (18FDG) on 18FDG-positron emission tomography (PET). Autoantibodies were sometimes detected, but patients did not fulfill the diagnostic criteria for autoimmune diseases, such as systemic lupus erythematosus (SLE). Lymph node histology in most patients was of mixed type, with few patients having hyaline-vascular type MCD histology. Most patients demonstrated dry tap on bone marrow aspiration, and bone marrow biopsy confirmed reticulin myelofibrosis and increased megakaryocytes in these patients.

Although glucocorticoids, immunosuppressive therapy, and tocilizumab were effective in some patients with this condition, others showed clinical deterioration despite treatment (2 of the 18 patients died in this series).

Discussion

Relations between TAFRO syndrome and MCD

MCD in western countries has been thought to be associated with HIV and/or HHV-8 infection. HHV-8 infection has been found to induce the expression of viral IL-6 (vIL-6), resulting in the development of hyper IL-6 syndrome [13, 14]. This results in the production of VEGF, causing angiogenesis; B cell differentiation, causing plasma cell expansion and polyclonal hypergammaglobulinemia; megakaryocyte differentiation in bone marrow causing thrombocytosis; and the production by the liver of acute inflammatory proteins, such as C-reactive protein (CRP), fibrinogen, and serum amyloid A (SAA). Hyper IL-6-syndrome can also induce hepcidin production in the liver, inhibiting iron absorption in the gastrointestinal tract and reducing iron recycling in the reticuloendothelial systems, thus causing microcytic anemia. Most Japanese patients with MCD are negative for HIV and HHV-8 infection [15] and display a chronic disease course. Moreover, MCD in Japan does not appear to progress to Kaposi’s sarcoma or B-cell lymphoma [16]. Recently, Fajgenbaum et al. summarized MCD cases in western countries, where more than half of them were also HHV-8 negative [17]. They classified such HHV-8 negative MCD cases as idiopathic MCD (iMCD). Therefore, most Japanese MCD patients are classified into iMCD.

Some of the characteristic findings in TAFRO syndrome, such as severe inflammation with hyper IL-6 and histopathological findings of lymph nodes, are similar to those in patients with iMCD. However, other characteristic findings such as thrombocytopenia, anasarca, renal insufficiency and reticulin myelofibrosis are not considered typical of iMCD. Few diseases are characterized by severe inflammation with thrombocytopenia and anasarca. In our cohort, most of patients had thrombocytopenia, and some of them had been diagnosed as having immune-thrombocytopenia (ITP), thrombotic thrombocytopenic purpura (TTP), anti-phospholipid antibody syndrome (APS), or disseminated intravascular coagulation (DIC).

Several questions remain regarding the mechanism and etiology of TAFRO syndrome. It is not clear whether TAFRO syndrome is a disease entity distinct from iMCD, a subset of iMCD, or a part of it overlaps with iMCD. Moreover, it is not clear whether this syndrome is neoplastic, autoimmune, infectious, or some other type of disease entity.

Future prospects

The multicenter retrospective clinical study of patients with TAFRO syndrome is now recruiting participants. This study register patients with MCD, as well as with TAFRO syndrome. The clinical characteristics of these two groups will be compared. The study will include TAFRO syndrome patients successfully treated with glucocorticoid treatment alone [1, 4], or in combination with other immunosuppresants [1, 4], tocilizumab [3, 5, 10, 11] and rituximab [12], as well as patients who died of disease, without improvement, despite treatment. As TAFRO syndrome is currently diagnosed by combinations of clinical conditions, this syndrome may be heterogeneous. To assess heterogeneity and establish the disease concept, it is necessary to compare patients with relatively mild TAFRO syndrome and those with severe, even fatal, disease. These diagnostic criteria should be used to design prospective clinical studies. Collected patients data and clinical materials may help in determining the pathophysiology and etiology of TAFRO syndrome. We also made the treatment strategy based on our clinical experiences (Table 3), and will strengthen evidence from now on.

Results of this study will also be coordinated with the research group on Castleman disease, represented by Dr. Kazuyuki Yoshizaki. This group is also funded by Health and Labour Sciences Research Grants for Research on Rare and Intractable Diseases from the Ministry of Health, Labour and Welfare of Japan (H27-Nanchi, etc. (Nan)-General-002). Differences between TAFRO syndrome and MCD will be discussed.

As the formulation of these clinical guidelines are based on update data, these guidelines will be revised, if necessary, based on further clinical and pathophysiological evidence.