Hodgkin transformation of chronic lymphocytic leukemia

The M. D. Anderson Cancer Center experience

Authors


Abstract

BACKGROUND

Hodgkin transformation is a rare complication of chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL). In this study, the authors assessed the incidence, presenting characteristics, and outcomes of patients with CLL/SLL who developed Hodgkin lymphoma (HL).

METHODS

An electronic database search of patients with CLL/SLL who presented at The University of Texas M. D. Anderson Cancer Center Department of Leukemia between 1975 and 2005 was performed.

RESULTS

Among 4121 patients with CLL/SLL, 18 patients (0.4%) developed HL. Presenting features included B-symptoms (67%), lymph node enlargement (79%), splenomegaly (43%), hepatomegaly (29%), hypercalcemia (6%), infection (6%), and mental status changes (6%). The median age was 72 years (range, 49–81 years), and there was a male preponderance (78%). The median time from CLL to HL diagnosis was 4.6 years (range, 0–12.9 years). Fourteen patients (78%) had been previously treated for CLL/SLL. Ten patients (56%) had received >1 prior therapy. The median β2-microglobulin level was 4.5 mg/L, and the median lactate dehydrogenase level was 610 IU/L. Epstein–Barr virus (EBV) was positive by in situ hybridization for EBV-encoded RNA in 3 of 4 tested patients. Fourteen patients (78%) received chemotherapy. The overall response rate was 44% (complete response rate, 19%). The median overall survival duration was 0.8 years (range, 0.03 years–6.7+ years). The median failure-free survival (FFS) duration was 0.4 years.

CONCLUSIONS

The rates of response, survival, and FFS in patients with Hodgkin transformation of CLL/SLL were inferior to those reported in patients with de novo HL and were similar to those in patients with Richter syndrome. Cancer 2006. © 2006 American Cancer Society.

Chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL) rarely is complicated by the development of neoplasms that morphologically and immunophenotypically resemble Hodgkin lymphoma.1 Although the term “Hodgkin variant of Richter transformation” has been used, the term “Hodgkin transformation of CLL/SLL” describes this disease more accurately. The large cells of Hodgkin transformation of CLL/SLL are characterized by morphologic and immunophenotypic features of Hodgkin and Reed–Sternberg (H-RS) cells of typical Hodgkin lymphoma and express CD15 and CD30. In vitro studies have shown that these H-RS cells are derived from mature B-cells that undergo transformation, and this transformation is accompanied by a change in the cellular milieu, with numerous T-cells and histiocytes: a cellular background that is typical of Hodgkin lymphoma.2

Two types of Hodgkin transformation of CLL/SLL have been described. Type 1 transformation is characterized by H-RS cells scattered in a background of CLL cells.3–5 In type 2 transformation, H-RS cells present in a typical polymorphous, inflammatory background separate from the CLL cells.1, 6–10 Histologic and immunophenotypic findings suggest that H-RS cells in patients with type 1 transformation represent histologic progression of the underlying CLL cells, especially when the H-RS cells express B-cell markers. Although, in type 2 transformation, 2 different disease types are considered to be present, the 2 lesions may be related. It is unknown whether the 2 types of Hodgkin transformation of CLL/SLL are associated with distinct clinical and prognostic features. A clonal relationship between CLL and H-RS cells was demonstrated in 3 of 4 patients who had Hodgkin transformation of CLL/SLL by using single-cell polymerase chain reaction analysis and DNA sequencing.2

The outcome of patients with de novo Hodgkin lymphoma is generally favorable. With the use of combined-modality treatment and multiagent chemotherapy, ≥80% of patients with de novo Hodgkin lymphoma and >90% of patients with limited-stage Hodgkin lymphoma (usually defined as Stage IA or IIA disease with the greatest tumor dimension measuring <10 cm) can be cured.11 In contrast, only 34% to 47% of patients with Richter syndrome respond to multiagent chemotherapy, such as fractionated cyclophosphamide, vincristine, liposomal daunorubicin, and dexamethasone or cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) with or without rituximab. The median overall survival of patients with Richter syndrome was 8 months, and the median failure-free survival (FFS) was 7 months.12

In a meta-analysis, the clinical outcomes of patients who had Hodgkin transformation of CLL/SLL reportedly were inferior to the outcomes of patients who had de novo Hodgkin lymphoma and superior to the outcomes of patients who had Richter syndrome.1, 13 The objective of the current study was to summarize the experience at The University of Texas M. D. Anderson Cancer Center regarding the incidence, presenting characteristics, and rates of response, overall survival, and FFS in patients with Hodgkin transformation of CLL/SLL.

MATERIALS AND METHODS

An electronic database of patients with CLL/SLL who presented at the Department of Leukemia in The University of Texas M. D. Anderson Cancer Center between 1975 and 2005 was searched to identify patients who developed Hodgkin transformation of CLL/SLL. All records were reviewed to determine clinical, laboratory, and pathologic features at presentation as well as disease stage, treatment, and clinical outcome. All diagnoses of Hodgkin lymphoma were made by lymph node biopsy or core biopsy if a superficial lymph node was not easily accessible. Staging evaluations at the time of presentation included complete physical examinations, bone marrow aspirations and biopsies, and chest radiography studies or computed tomography scans of the chest, abdomen, and pelvis, when available. Staging and treatment were determined after a review of all clinical, laboratory, and pathologic data in a multidisciplinary conference. Standard or investigational treatment was administered either at our center or in the community by collaborating physicians. The evaluation of response for each patient who was included in the analysis was performed at M. D. Anderson Cancer Center. Signed informed consent forms were obtained before all procedures and before all therapy, as required by the Institutional Review Board.

Endpoints and Statistical Methods

Complete remission (CR) was defined as the complete disappearance of all detectable clinical and radiographic evidence of disease, the disappearance of all disease-related symptoms, and the normalization of biochemical abnormalities definitely assignable to the disease for at least 1 month. Partial remission (PR) was defined as a reduction ≥50% in the sum of the products of the greatest dimensions of bidimensionally measurable disease. Any lesser response was considered a failure.

Survival was measured from the start of treatment until death from any cause or until last follow-up. FFS was defined as the time from the start of treatment until disease progression, recurrence, or death.

The prognostic scoring system for advanced Hodgkin lymphoma was used to assess an individual pa tient's risk of disease progression.14 The Richter syndrome score, which was developed to predict the risk of death in patients with Richter syndrome, also was evaluated.12 This score takes into account 5 adverse prognostic factors (Zubrod performance status >1, LDH levels >1.5 times the upper limit of normal, platelet count <100 × 109/L, tumor size ≥5 cm, and >1 prior therapy) and predicts the median survival duration as follows: scores of 0 or 1, 1.12 years; a score of 2, 0.9 years; a score of 3, 0.3 years, and scores of 4 or 5, 0.1 years.12

RESULTS

Among 4121 patients with CLL/SLL, 18 patients (0.4%) developed Hodgkin transformation. The presentation shared some features with de novo Hodgkin lymphoma (Table 1). The median age was 72 years (range, 49–81 years). There were 14 men (78%) and 4 women (22%). The median time from CLL diagnosis to Hodgkin transformation of CLL/SLL was 4.6 years (range, 0–12.9 years) and included 3 patients who presented with concomitant CLL/SLL and Hodgkin lymphoma. Hodgkin lymphoma was confirmed by excisional biopsy of involved tissue in 15 patients and by core biopsy in 3 patients (2 lymph node biopsies and 1 abdominal mass biopsy). Disease involvement was documented in the following sites: lymph nodes, 9 patients; bone marrow, 6 patients; spleen, 1 patient; buccal mucosa, 1 patient; and abdominal mass, 1 patient. Patient characteristics are summarized in Table 2. The median β2-microglobulin level was 4.5 mg/L (normal range, 0.8–2.0 mg/L), and the median lactate dehydrogenase level was 610 IU/L (normal range, 313–618 IU/L). Fourteen patients (78%) had been treated previously for CLL/SLL. Ten patients (56%) had received >1 prior therapy for the treatment of CLL/SLL. Prior therapies included fludarabine with or without cyclophosphamide with or without rituximab (85% of patients), chlorambucil (31%), rituximab (23%), alemtuzumab (15%), platinum-based combination regimens (15%), and other therapies (38%) (Table 3).

Table 1. Clinical Presentation of 18 Patients with Hodgkin Transformation of Chronic Lymphocytic Leukemia/Small Lymphocytic Lymphoma
Clinical presentationPercentage of patients
  • *

    Records for organomegaly or lymphadenopathy were available in 14 patients.

B-symptoms67
Fever50
Lymph node enlargement*79
Fatigue28
Splenomegaly*43
Hepatomegaly*29
Night sweats28
Weight loss (>10% of body weight)17
Mental status changes6
Infection6
Hypercalcemia-related symptoms6
Table 2. Presenting Characteristics of 18 Patients with Hodgkin Transformation of Chronic Lymphocytic Leukemia/Small Lymphocytic Lymphoma
CharacteristicValue
  1. WBC indicates white blood cells; ALC, absolute lymphocyte count; ANC, absolute neutrophil count; Ig, immunoglobulin; ULN, upper limit of normal.

Median (range)
 Age, y72 (49–81)
 WBC, × 109/L7.2 (0.6–122.4)
 ALC, × 109/L1.7 (0.1–106.5)
 ANC, × 109/L2.7 (0–15.0)
 Platelet count, ×109/L164 (13–452)
 IgG, mg/dL565 (258–3080)
 IgA, mg/dL62 (11–377)
 IgM, mg/dL62 (21–7800)
No. of patients/evaluable patients (%)
 Age >60 y14/18 (78)
 Zubrod performance status >13/13 (23)
 Hemoglobin >11 g/dL14/18 (78)
 ALC >5 × 109/L5/17 (29)
 ANC <1.5 × 109/L4/17 (24)
 Platelet count <100 × 109/L6/18 (33)
 Albumin <3.4 g/dL10/17 (59)
 Lactate dehydrogenase >ULN8/17 (47)
 β2-microglobulin ≥4 mg/dL8/12 (67)
Table 3. Prior Therapies in 18 Patients with Hodgkin Transformation of Chronic Lymphocytic Leukemia/Small Lymphocytic Lymphoma
Prior therapyNo. of patients%
Fludarabine643
Fludarabine and cyclophosphamide321
Fludarabine and cyclophosphamide and rituximab429
Rituximab429
Rituximab and alemtuzumab17
Alemtuzumab17
Chlorambucil429
Other536

Bone marrow cytogenetics were examined in 12 patients, and the results were normal in 5 patients (42%); abnormalities included −Y in 4 patients (33%); 11q− in 1 patient (8%); and complex cytogenetic ab normalities in 2 patients (17%), including del(13)(q14q22) and trisomy 12 in 1 patient and t(9;15)(p22;q13),−11,−17 in another patient. Epstein–Barr virus (EBV) was detected by in situ hybridization for EBV-encoded RNA in 3 of 4 patients tested (75%).

Histopathologic Features

CLL/SLL and Hodgkin lymphoma were identified in a single biopsy specimen in 11 patients and in separate biopsy specimens in 7 patients. Five patients with Hodgkin lymphoma presented without morphologic or immunophenotypic evidence of CLL/SLL (Fig. 1A), and 1 patient had CLL/SLL in bone marrow and Hodgkin lymphoma in a cervical lymph node at the same time. One patient was diagnosed histologically with CLL/SLL and EBV-positive large B-cell lymphoproliferative disorder in December 2003: A follow-up biopsy in October 2004 showed the histologic and immunophenotypic features of classical Hodgkin lymphoma without morphologic evidence of CLL/SLL.

Figure 1.

(A) This photomicrograph shows bone marrow infiltration by Hodgkin lymphoma. Note the Hodgkin cells, histiocytes, lymphocytes, plasma cells, and fibrosis. (B) This photomicrograph shows bone marrow with chronic lymphocytic leukemia (CLL) composed of small lymphocytes (top) and Hodgkin cells (bottom, arrows). (C) This photomicrograph shows CD15 expression in Hodgkin cells. Note the membrane and dot-like enhancement of the Golgi region (immunoperoxidase stain; original magnification, × 500). (D) This photomicrograph shows CD30 expression in a Reed–Sternberg cell of Hodgkin lymphoma variant. Note dot-like enhancement of the Golgi region (immunoperoxidase stain; original magnification × 500).

In the 6 patients with Hodgkin lymphoma that involved the trephine bone marrow biopsy, classic Hodgkin lymphoma and CLL/SLL were present at the same time in 3 patients (Fig. 1B). In 3 patients who had a history of CLL/SLL, Hodgkin lymphoma occurred with no background of CLL/SLL.

The biopsy samples were infiltrated by a monotonous population of small lymphocytes with round nuclei consistent with CLL/SLL. On immunohistochemical analysis, the infiltrate consisted of CD20-positive cells that expressed CD5 and CD23. In the samples with composite CLL/SLL and Hodgkin lymphoma, there were areas of CLL/SLL, areas of CLL/SLL with H-RS cells, and areas of fibrosis and a more heterogeneous cell population, including lymphocytes, histiocytes, plasma cells, rare eosinophils, and a moderate number of large atypical cells, some of which were binucleate with large nucleoli, all suggesting concurrent Hodgkin lymphoma. The latter were CD15-positive (Fig. 1C), CD30-positive (Fig. 1D), CD20-negative, and CD45-negative.

In the patient who was diagnosed with CLL/SLL and EBV-positive large B-cell lymphoproliferative disorder in December 2003, the histologic sections from the cervical lymph node biopsy specimen showed large effacement of the lymph node architecture by a polymorphous population of cells, including small round lymphocytes, plasmacytoid lymphocytes, large lymphoid cells with prominent central nucleoli, and occasional Reed–Sternberg-like and giant cells. Extensive necrosis and mitotic figures were easy to identify. The large cells were positive for CD20, CD30, CD45, CD79a, and EBV latent membrane protein (EBV-LMP). The small cells were monotypic immunoglobulin light chain-positive, CD5-positive, CD19-positive, CD20-positive (dim), CD22-positive (dim), CD23-positive (dim), and CD38-positive. A follow-up biopsy of a right neck mass in October 2004 showed the morphologic and immunophenotypic features of classic Hodgkin lymphoma (CD15-positive, CD30-positive, EBV-LMP-1-positive, CD3-negative, CD4-negative, CD5-negative, CD7-negative, CD8-negative, CD10-negative, CD20-negative, BCL2-negative, keratin-negative, MART-1-negative, HMB-45-negative, and myeloperoxidase-negative) in addition to some features of non-Hodgkin lymphoma (CD45-positive). There was no morphologic evidence of CLL in the 2004 specimen.

Therapy and Response

Hodgkin transformation of CLL/SLL was treated with chemotherapy in 14 patients. Nine patients received Hodgkin lymphoma-type therapy, including 5 patients who received doxorubicin, bleomycin, vinblastine, and dacarbazine (ABVD); 3 patients who received cyclophosphamide, vinblastine, procarbazine, and prednisone (CVPP); and 1 patient who received CVPP/ABVD. One of the 3 patients who received CVPP also received consolidation therapy with involved-field radiation. Five patients received other cytotoxic regimens with or without rituximab (Table 4). One patient with EBV-related disease, as evidenced by in situ hybridization for EBV-encoded RNA in a lymph node biopsy, received the antiviral cidofovir, 1 patient received rituximab, and 2 patients received no therapy. The overall response rate was 44% (Table 5). Responses were noted in patients who received ABVD (1 CR and 2 PRs), CVPP (1 CR), CVPP/ABVD (1 PR), and CHOP (1 PR).

Table 4. Treatment for Hodgkin Transformation of Chronic Lymphocytic Leukemia/Small Lymphocytic Lymphoma
TherapyNo. of patients%
  • ABVD indicates doxorubicin, bleomycin, vinblastine, and dacarbazine; CVPP, cyclophosphamide, vinblastine, procarbazine, and prednisone; CHOP, cyclophosphamide, doxorubicin, vincristine, and prednisone.

  • *

    One patient also received involved-field radiation therapy after CVPP.

  • Each of these 5 patients received fludarabine, cytarabine, cyclophosphamide, and cisplatin; fludarabine, cyclophosphamide, and rituximab (FCR); mitoxantrone, vincristine, vinblastine, and prednisone; rituximab; and cidofovir.

  • One patient who received FCR also received salvage therapy with doxorubicin, bleomycin, and vinblastine followed by involved-field radiation therapy.

ABVD528
CVPP3*17
CVPP/ABVD15.5
CHOP15.5
CHOP and rituximab15.5
Other528
No therapy211
Table 5. Response to Therapy of 16 Patients with Hodgkin Transformation of Chronic Lymphocytic Leukemia/Small Lymphocytic Leukemia
TherapyNo. of patientsNo. of patients (%)
CRPROverall
  • CR indicates complete remission; PR, partial remission; ABVD, doxorubicin, bleomycin, vinblastine, and dacarbazine; CVPP, cyclophosphamide, vinblastine, procarbazine, and prednisone; CHOP, cyclophosphamide, doxorubicin, vincristine, and prednisone

  • *

    This patient also received involved-field radiation therapy after CVPP.

  • This group includes 1 patient who received fludarabine, cyclophosphamide, and rituximab and also received salvage therapy with doxorubicin, bleomycin, and vinblastine followed by involved-field radiation therapy.

ABVD5123 (60)
CVPP31*1 (33)
CVPP/ABVD111 (100)
CHOP111 (100)
CHOP and rituximab111 (100)
Other50 (0)
Total16347 (44)

Overall Survival and FFS

The median overall survival duration was 0.8 years (range, 0.03–6.7+ years), and the median FFS duration was 0.4 years (range, 0.03–6.7+ years) (Figs. 2, 3, respectively). All patients died from disease recurrence or progressive disease (without responding to treatment). Details of the clinical presentation and outcomes of the 18 patients are summarized in Table 6.

Figure 2.

Overall survival in 18 patients who had Hodgkin transformation of chronic lymphocytic leukemia/small lymphocytic lymphoma.

Figure 3.

Failure-free survival in 18 patients with Hodgkin transformation of chronic lymphocytic leukemia/small lymphocytic lymphoma.

Table 6. Characteristics of 18 Patients with Hodgkin Transformation of Chronic Lymphocytic Leukemia/Small Lymphocytic Lymphoma
PatientAge, YearsGenderHgb, g/dLWBC, × 109/LALC, × 109/LPLT, × 109/LLDH, IU/Lβ2-M, mg/LRai stageNo. of prior RxCLL→HL, YearsHL score*RS scoreTherapyResponseSurvival, yearsSurvival statusFFS, yearsFailure
  • Hgb indicates hemoglobin; WBC, white blood cells; ALC, absolute lymphocyte count; PLT, platelets, LDH, lactate dehydrogenase; β2-M, β2-microglobulin; Rx, treatment; CLL, chronic lymphocytic leukemia; HL, Hodgkin disease; RS, Richter syndrome; FFS, failure-free survival; ND, not done; ABVD, doxorubicin, bleomycin, vinblastine, and dacarbazine; PR, partial remission; CR; complete remission; ABV, doxorubicin, bleomycin, and vinblastine; CVPP; cyclophosphamide, vinblastine, procarbazine, and prednisone; IF R/T, involved-field radiation therapy; CHOP, cyclophosphamide, doxorubicin, vincristine, and prednisone; R, rituximab; NOVP, mitoxantrone, vincristine, vinblastine, and prednisone; FACP, fludarabine, cytarabine, cyclophosphamide, and cisplatin; FCR, fludarabine, cyclophosphamide, and rituximab; N/A, not available.

  • *

    The HL score was determined based on serum albumin level, Hgb level, gender, age, Ann Arbor stage, WBC count, and lymphocyte count (see Hasenclever and Diehl, 199814).

  • The RS score was determined based on Zubrod performance status, LDH levels, platelet counts, tumor size, and the number of prior therapies (see Tsimberidou et al., 200612).

172Male9.94.11.0222485ND313.850ABVDPR2.9Dead2.4Failed
272Male11.56.10.561536ND425.642ABVDCR2.6Dead2.2Failed
371Male10.56.711.1250752ND336.751ABVDPR1.9Dead1.1Failed
466Female9.33.40.14312763.8477.353ABVDFail0.8Dead0.2Failed
573Male10.68.13.523510196.0344.443ABVFail0.7Dead0.6Failed
678Male7.20.60.1296822.3413.561CVPPFail0.6Dead0.2Failed
778Female10.76.31.645218832.2326.633CVPP and  IF R/TCR6.7Alive6.7Nonfailed
875Female11.046421371257.7200.031CVPPFail2.2Dead0.4Failed
955Male9.22.70.4456105.0413.562CVPP/ABVDPR0.3Dead0.2Failed
1067Male12.011.11.71944615.9100.041CHOP and RCR2.2Dead1.7Failed
1172Male12.66.2ND225NDND126.931CHOPPR1.4Alive0.7Failed
1273Female9.37.72.51014425.5300.041NOVPFail0.9Dead0.1Failed
1349Male8.819.22.72638314.0331.861FACPFail0.2Dead0.2Failed
1469Male8.846.843.11902752.0304.960FCR, ABV,  IF R/TFail4.4Alive0.5Failed
1557Male6.98.00.393928ND431.165RFail0.03Dead0.03Failed
1657Male9.345.443.61013364.0335.062CidofovirFail0.3Dead0.3Failed
1777Male9.42.31.1208577ND324.951NoneN/A0.4Dead0.4Failed
1881Male9.41221061385011.54112.963NoneN/A0.2Dead0.2Failed

Prognostic Scores

A prognostic score for advanced Hodgkin lymphoma was assigned to each patient.14 The distribution of scores was as follows: 3 (3 patients), 4 (4 patients), 5 (4 patients), and 6 (7 patients) (Table 6). Therefore, all patients had a prognostic score ≥3, indicating that they all had features of high-risk Hodgkin lymphoma. The median progression-free survival duration for patients who had scores of 3 or 4 was 0.7 years compared with 0.2 years for those who had scores of 5 or 6 (P = .10; log-rank test) (Fig. 4).

Figure 4.

Use of the prognostic score to predict freedom from disease progression (Prog) in 18 patients with Hodgkin transformation of chronic lymphocytic leukemia/small lymphocytic lymphoma.14 HL indicates Hodgkin lymphoma.

The Richter syndrome scores12 were as follows: 0 (2 patients), 1 (8 patients), 2 (3 patients), 3 (4 patients), and 5 (1 patient). The median survival duration in patients who had Richter syndrome scores of 0 or 1 was 1.4 years compared with 0.3 years in patients who had scores of 2, 3, or 4 (P = .21 [log-rank test]; P = .18 [Breslow–Gehan–Wilcoxon test]) (Fig. 5).

Figure 5.

Survival in 18 patients with Hodgkin transformation of chronic lymphocytic leukemia/small lymphocytic lymphoma according to risk group defined by the Richter syndrome (RS) score.12

DISCUSSION

The current analysis describes the clinical characteristics and treatment outcomes of patients with Hodgkin transformation of CLL/SLL in a single institution. The outcomes of patients with Hodgkin transformation of CLL/SLL was similar to the outcomes of patients with Richter syndrome and was significantly inferior to the outcomes reported in patients with de novo Hodgkin lymphoma.

The incidence of this subtype of Richter syndrome was 0.4% among patients with CLL/SLL. Although Hodgkin lymphoma has been reported as among the most common secondary cancers in patients with CLL/SLL,15 it has been observed in our series and in other studies16 that other cancers, such as basal cell or squamous cell skin carcinoma, breast cancer, prostate cancer, and large cell lymphoma12 occur more frequently.

The clinical presentation and laboratory findings in the current series share some features with de novo Hodgkin lymphoma and are comparable to those from previously published series.1, 6, 7, 9, 17–25 However, the majority of those reports focused on the pathologic features of the disease. Although only 4 patients were tested for EBV in the current series, the incidence of EBV was consistent with previous reports,3, 18, 26–28 implying an important role for this virus in the pathogenesis of Hodgkin transformation of CLL/SLL. In a previous study, EBV was positive by in situ hybridization for EBV-encoded RNA in 12 of 13 patients with CLL/SLL and possible Hodgkin transformation of CLL/SLL.3 EBV DNA also was detected by polymerase chain reaction analysis of lymph nodes from 2 patients with the Hodgkin variant of CLL.26 In another study, EBV LMP-1 was identified in 4 patients in the absence of expression of the EBV lytic cycle switch protein Zta BZLF1 (ZEBRA), suggesting that the genome of EBV is integrated in the absence of active viral replication in these patients.18

Fourteen patients (78%) received chemotherapy. Regimens varied over the years, and assignment to different treatments was based on therapy availability, patient preference, and physician choice. The overall response rate was 44% (CR, 19%). Hodgkin lymphoma-type chemotherapy, such as ABVD and CVPP, induced responses in 5 of 9 treated patients (CR, 2 patients). However, the response duration was short, with a median FFS duration of 0.4 years and a median survival duration of 0.8 years. Although the number of patients was small, these results suggest that more effective treatments are needed for patients who have Hodgkin transformation of CLL/SLL. Newer, more intensified, combined-modality regimens, such as escalated bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, and prednisone (BEACOPP) and doxorubicin, vinblastine, mechlorethamine, vincristine, bleomycin, etoposide, and prednisone (Stanford V) may be more effective than ABVD or an equivalent in this disease and should be tested.29, 30 High-dose therapy followed by autologous stem cell transplantation has been used in the treatment of patients with advanced de novo Hodgkin lymphoma that is recurrent or refractory to multiagent therapy.11, 31, 32 Although the data regarding its effectiveness are inconclusive, this approach may be investigated in patients with recurrent or refractory Hodgkin transformation of CLL/SLL. However, aggressive approaches, such as escalated BEACOPP and stem cell transplantation, may not be appropriate for many frail, elderly patients. Other more novel approaches that include therapy with EBV-stimulated cytotoxic T-cells have shown activity in patients with EBV-positive Hodgkin lymphoma.26 In particular, this approach induced durable responses (up to 40 months) in 5 of 14 patients with recurrent, EBV-positive Hodgkin lymphoma.26 Because EBV may play a role in the induction and perhaps proliferation of Hodgkin transformation of CLL/SLL, we speculate that this therapeutic strategy may have activity in EBV-positive Hodgkin transformation of CLL/SLL.

All patients with Hodgkin transformation of CLL/SLL had high-risk Hodgkin lymphoma, as evidenced by their prognostic scores of ≥3.14 Although interpretation of these data is difficult because of the small number of patients and the retrospective nature of the current analysis, this system was predictive of the duration of progression-free survival (median survival: 0.7 years for patients with scores of 3 or 4, and 0.2 years for patients with scores of 5 or 6). Likewise, the Richter syndrome score also may be able to predict survival (median survival: 1.5 years for patients with scores of 0 or 1 and 0.3 years for patients who had scores of 2, 3, or 4). Neither of these differences reached statistical significance, however, probably because of the very small number of patients. It is noteworthy that these patients were not treated uniformly, and 2 patients never received therapy.

In conclusion, the treatment of patients who have Hodgkin transformation of CLL/SLL is challenging. The current results suggest that Hodgkin lymphoma-type multiagent chemotherapy, such as ABVD, is effective, but patients eventually develop recurrent disease after a short period of time. The newer, intensified, combined-modality regimens may be more effective. Stem cell transplantation may be considered for patients who respond to chemotherapy, because the remission duration is short. Future studies should focus on the pathogenetic mechanisms of Hodgkin transformation of CLL/SLL, and more specific treatments should be developed. For instance, the role of EBV in the pathogenesis of this disease needs to be investigated further, because patients with CLL/SLL and EBV infection may benefit from antiviral therapy, and such therapy may decrease the probability of Hodgkin transformation in this subset of patients.

Acknowledgements

We thank Mark Brandt for his assistance with the medical graphics

Ancillary