FINAL DIAGNOSIS: RICHTER SYNDROME
Richter transformation was first described by Maurice Richter in 1928, in a young male patient with a past medical history of chronic lymphocytic leukemia (CLL) who died soon after developing sudden deterioration and diffuse lymphadenopathy. Richter described the histomorphology on the lymph nodes of the patient as "leukemic and tumor cells". The leukemic cells were enlarged with abundant cytoplasm, large nuclei and prominent nucleoli, and the tumor cells were smaller lymphoid cells admixed within these large cells. The latter appeared to correspond to CLL cells. In 1964, Lorthoraly et al. introduced the term "Richter syndrome" (RS) to define the progression of CLL to a "malignant reticulopathy". RS was defined as a secondary high-grade lymphoma, mostly a diffuse large B-cell lymphoma (DLBCL), arising in patients with B-CLL and was further expanded to include patients with transformation of CLL to prolymphocytic leukemia, Hodgkin lymphoma (so called Hodgkin variant of RS) and hairy cell leukemia, among other diagnoses. However, many of the cases were described between the 1960's and 1970's and lacked the cytogenetic and molecular tools to evaluate whether these neoplasms were arising from the same clone of CLL cells or if they were unrelated "collision lesions". Currently, RS is considered to arise largely from the same clone of CLL cells if the same rearrangement band or same sequence as B-CLL is proven by Southern blotting, or if the nucleotide sequence analyses of immunoglobulin heavy or light chain genes suggests the transformation.
The incidence of Richter transformation in CLL is reported to be in the range of 2.2 - 8%. In a study of 1011 patients with B-CLL (Mauro et al.), 18 (1.8%) cases were of DLBCL and 4 (0.4%) were of the Hodgkin variant of RS. Geographically, B-CLL is the most frequent disorder among hematologic malignancies in Western countries. There are no established risk factors for RS; however, younger age, the presence of peripheral blood involvement, diffuse involvement of the bone marrow, initial hemoglobin less than 12 g/dl, advanced Rai stage, increasing LDH levels and high beta-2-microglobulin levels are considered to be possible risk factors. Among these, it is important to mention that younger age (less than 55 years old) has been shown as a possible risk factor in some studies; this might be because older patients with advanced disease may not undergo a bone marrow or lymph node biopsy. Other studies indicate that previous fludarabine treatment, as in our patient, could also be a risk factor in the development of RS.(1,7)
The patients usually present with sudden deterioration, characterized by systemic symptoms as fever, weight loss and night-sweats ("B-symptoms") and rapid increase in the size a lymphoid mass or hepatosplenomegaly. The most common sites where the transformation occurs are the bone marrow or lymph nodes. Extranodal sites include the gastrointestinal tract (5.1%), lung (2.5%), pleura (10.2%), oropharynx (7.7%), bone (2.5%), marrow (5.1%), skin (5.1%), and also the central nervous system (12.8%). Bone marrow involvement, as seen in our patient, is not frequent. In rare cases, CNS symptoms may be the only initial manifestation of RS. One study (1,5) showed that 13% of RS patients had CNS involvement at the time of presentation. In our case, this was suspected as the patient presented with seizures, but the lesion was histologically consistent with a meningioma.
There are no laboratory findings that are specific for RS. There may be elevation of LDH levels, which can be a marker of tumor growth. 82% of patients present with LDH levels twice as high as the upper limit of normal, as opposed to CLL in which only 8% of cases demonstrate elevated LDH levels (1, 15). 44% of patients in one study presented with paraproteinemia (1). Lytic bone lesions or hypercalcemia may be present, probably due to increased bone resorption. In some patients, the presence of circulating immunoblasts in the peripheral blood is the first indication of RS.
The histologic diagnosis of classic RS is made in the presence of large, round or slightly irregular cells with moderate amounts of cytoplasm, vesicular nuclei and prominent nucleoli that are distinct from those of CLL/SLL cells. The specimen's architecture can be completely effaced by the DLBCL or the latter can coexist with CLL cells. The immunophenotype of the cells could be identical in RS and the CLL of origin, but could also be different. A variant of RS has been described when patients with CLL develop Reed-Sternberg like cells, and this is termed Hodgkin variant of RS. Dominic et al describe the immunophenotype of these Reed-Sternberg like cells as CD30 positive, EBER positive in 3 of 4 cases; CD20, CD15 and LMP-1 variably positive, and can be accompanied by CLL cells (8) . In our case, EBER was negative.
The etiology of Richter's transformation is poorly understood. There are two theories regarding Richter's transformation. One proposes that the disease arises from the same clone. The second one proposes that it is a separate or independent neoplasm, as demonstrated by characterization of immunoglobulin heavy chain rearrangement and light chain isotype analysis. Southern hybridization with immunoglobulin heavy chain (IgH) gene and light chain (IgL) can be used as a tool for distinguishing clonal evolution versus a new malignant clone. A rearrangement of a same-sized band for both B-CLL and secondary DLBCL supports an identical origin, and a rearrangement of a different-sized band suggests a new malignant event. However, cases with lambda light-chain class restricted (LCCR) B-CLL and cases with kappa LCCR RS were demonstrated to have identical rearrangements of the IgH chain gene, indicating that different immunophenotypes of CLL and RS do not necessarily indicate biclonality. Tools helpful in evaluating for the presence of biclonality include nucleotide sequencing of the IgH and/or IgL genes that can demonstrate the clonal relation between RS and a pre-existing CLL. Therefore, if identical rearrangement bands by Southern blotting are found in both RS and B-CLL, one could conclude that clonal evolution is present. But if one of those bands is different, this conclusion cannot be made and nucleotide sequencing of the CDR3 region including the VDJ region of the IgH gene is necessary.
Clonal evolution can be triggered by viral infections, but many cases can develop after acquisition of new karyotypic abnormalities. Among the cytogenetic abnormalities described in RS, trisomy 12 is described with a high frequency. This is associated with atypical lymphocyte morphology and also with immunophenotypic deviations; most cases of typical B-CLL are positive for CD19, CD20, surface immunoglobulin (weak) as well as CD5 and negative for FMC7. When trisomy 12 is present in RS, a different immunophenotype may be present such as CD5 negativity, FMC7 positivity and strong Ig staining, features not seen in typical CLL. Trisomy 12 is also associated with high rates of proliferation and disease progression, and studies have shown that it may be one of the primary changes in CLL undergoing transformation to RS (1-3, 9, 11).
Deletion of 11q may be seen in "typical" CLL. Disease progression has been described in patients with CLL/SLL who present with deletion 11q23. Another aberrancy of chromosome 11 is translocation t (11; 14)(q13; 32) that is found in "atypical" CLL. Patients with CLL and this translocation develop RS more frequently, and while this translocation is typically associated with mantle cell lymphoma, it can be present in RS, and a relationship between mantle cell lymphoma and RS has not been fully evaluated. Other chromosomal abnormalities include 14q+, del 13q and del 17p. When occurring in CLL, atypical lymphocyte morphology and immunophenotype have been reported.
In our patient, the cytogenetic study of the bone marrow showed a abnormal near-tetraploid female chromosome analysis with several consistent numerical and structural abnormalities, including t(14;19)(q32;q13) [see figure 'cytogenetic study']. This rare but recurrent translocation is found in patients with B-cell malignancies, mainly in chronic B-cell lymphoproliferative disorders, and is often associated with rapidly progressive disease, and overall poor prognosis as compared to the expected survival in chronic lymphocytic leukemia and low-grade B-cell lymphoma. The t(14;19) involves the BCL3 gene, which is located at the breakpoint on chromosome 19 and is juxtaposed to the immunoglobulin heavy chain gene locus on chromosome 14 (often in the switch alpha region) in a "head-to-head" configuration. The translocation does not interrupt the transcriptional integrity of BCL3, but is associated with overexpression of this gene, which encodes an I kappa B-like protein and modulates the activity of the NF-kappa B transcription factors (18-19).
In addition to chromosomal abnormalities, microsatellite instability (MSI) has also been described in RS. Studies have shown that there is hypermethylation of hMLH1 mismatch repair gene promoter (16) .
Interestingly, high levels of MSI have been observed in patients with Richter transformation but not in patients with non-transformed CLL. Other chromosomal abnormalities may be involved in the transformation of small cells to large ones, as loss of heterozygosity (LOH) on chromosomes 11, 17 and 20 has been reported in the large cells (Richter cells) of these tumors in contrast to small CLL cells.
There are multiple alterations in tumor suppressor genes related to RS. There is a high frequency of alterations of p53 in RS and this appears to be related to the presence of a new malignant clone rather than progression from an indolent CLL clone. Another strong inhibitor of the cell cycle progression, p16INK4A, has been shown to be inactivated in 33% of patients with transformed CLL, follicular lymphomas and blastic mantle cell lymphomas (1,2) . Of note, INK4A knockout mice can spontaneously develop B-cell lymphomas with aggressive morphology. Other tumor suppressor genes shown to be altered in RS are p21, in which immunohistochemical studies of RS have shown a higher rate of p53+/p21+ than in CLL (1-3) . While the majority of B-CLL cases strongly express p27, which is involved in G1 arrest, RS cases are p27 negative. This has been related to disease progression and poor prognosis.
The role of EBV in RS is not yet defined. Epstein-Barr virus (EBV), a lymphotropic human herpesvirus, has a prevalence of more than 90% in the population worldwide. Thanks to the action of natural killer cells, T-suppressor cells and T-cytotoxic cells, which restrict the proliferation of infected cells, the majority of carriers remain asymptomatic. However, immunosuppressed patients receiving treatment against these and other hematopoietic disorders may be at risk for developing EBV infection. It is known that when EBV infects a B-cell, its genome stays as a plasmid in the nuclei and then the virus maintains its genome (latent infection), preventing the cell from being destroyed, and activates cellular growth pathways, immortalizing the infected B-cells. In latent infection, the cells proliferate indefinitely, but the virus does not replicate and expresses minimal viral genes, avoiding its destruction. Latent membrane protein 1 (LMP-1) and EBV nuclear antigen 2 (EBNA2) are well known transforming agents for B-cells. LMP-1 is crucial in the oncogenic process, mimicking members of the family of tumor necrosis factor (TNF) receptors. LMP-1 has a cytoplasmic tail, very similar to the one of CD40, that binds to TNF-receptor associated factors (known as TRAF molecules), that when activated, cause the cell to proliferate. A retrospective study by Ansell et al found that 4 of 25 patients with RS (16%) were EBV positive (1) . The median survival rate of these patients was less than the ones without EBV infection (3 months vs. 9 months). Interestingly, EBV positive neoplastic B-cells are more resistant to fludarabine treatment than their negative counterparts. EBV appears to induce bcl-2 antiapoptotic protein, but the mechanisms are not fully understood.
The treatment for RS ranges from chemotherapy to transplantation. Currently, RS is considered to be resistant to conventional combined chemotherapeutic agents. Giles et al. evaluated a combined chemotherapy with cis-platinum, fludarabine and cytosine arabinoside, and another using cyclophosphamide instead of cis-platinum. Of 11 patients, 2 achieved complete remission and 3 achieved partial remissions. The response ranged from 5 to 43%, and the medial survival rate from 5 to 8 months. Many chemotherapeutic agents have been tried, but the results are limited. In particular, hyper-CVXD (fractionated cyclophosphamide, vincristine, liposomal daunorubicin, and dexamethasone) induced a 41% response of patients with RS and the median survival rate was 10 months. Allogeneic stem cell transplantation post-chemotherapy in a study demonstrated that 3 of 8 patients achieved durable remissions and were free of disease at 14, 47 and 67 months (17) . The management of CNS RS included systemic and intrathecal chemotherapy, and radiotherapy and two patients with CNS involvement remained alive after 13 and 17 months. The role of radiotherapy is limited to control pain and symptoms associate with bulky lymphadenopathy or extranodal disease. Patients with Hodgkin variant of RS are more resistant to therapy than those with classic RS.
In summary, the majority of patients with RS do not respond to available therapies, and a greater understanding of the mechanisms involved in the development of Richter syndrome along with the development of new therapeutic regimens are needed to improve prognosis and survival in these patients. Currently, our patient is receiving therapy with R-CHOP (Rituximab, Vincristine, Doxorubicin, Cyclophosphamide, Prednisolone) and has demonstrated a response.
Contributed by Rosemary Recavarren A, MD, Jie Hu, PhD and Lydia Contis, MD