Fax: (713) 794-4297
Fractionated cyclophosphamide, vincristine, liposomal daunorubicin, and dexamethasone plus rituximab and granulocyte-macrophage–colony stimulating factor (GM-CSF) alternating with methotrexate and cytarabine plus rituximab and GM-CSF in patients with Richter syndrome or fludarabine-refractory chronic lymphocytic leukemia
Article first published online: 18 MAR 2003
Copyright © 2003 American Cancer Society
Volume 97, Issue 7, pages 1711–1720, 1 April 2003
How to Cite
Tsimberidou, A. M., Kantarjian, H. M., Cortes, J., Thomas, D. A., Faderl, S., Garcia-Manero, G., Verstovsek, S., Ferrajoli, A., Wierda, W., Alvarado, Y., O'Brien, S. M., Albitar, M., Keating, M. J. and Giles, F. J. (2003), Fractionated cyclophosphamide, vincristine, liposomal daunorubicin, and dexamethasone plus rituximab and granulocyte-macrophage–colony stimulating factor (GM-CSF) alternating with methotrexate and cytarabine plus rituximab and GM-CSF in patients with Richter syndrome or fludarabine-refractory chronic lymphocytic leukemia. Cancer, 97: 1711–1720. doi: 10.1002/cncr.11238
- Issue published online: 18 MAR 2003
- Article first published online: 18 MAR 2003
- Manuscript Accepted: 30 NOV 2002
- Manuscript Revised: 26 NOV 2002
- Manuscript Received: 23 AUG 2002
- Richter syndrome;
- chronic lymphocytic leukemia;
Therapy for patients with Richter syndrome (RS) or fludarabine-refractory chronic lymphocytic leukemia (CLL) is unsatisfactory. A Phase II study was conducted to evaluate an alternating combination cytotoxic regimen given with rituximab and granulocyte-macrophage–colony stimulating factor (GM-CSF) in these patients.
Fludarabine-refractory CLL was defined as failure to respond to most recent prior fludarabine-containing regimen. Patients received up to six cycles of fractionated cyclophosphamide, vincristine, liposomal daunorubicin, and dexamethasone (hyper-CVXD) plus rituximab and GM-CSF alternating with methotrexate and cytarabine plus rituximab and GM-CSF. Response, toxicity, and survival data were compared with data from prior therapy with hyper-CVXD alone in this patient group.
Forty-nine patients with RS (n = 30 patients) or refractory CLL (n = 19 patients) were treated on study. Nine patients (18%) achieved a complete remission, and 11 patients achieved a partial remission (22%), for an overall objective response (OR) rate of 41%. With a median follow-up of 7.5 months and a maximum follow-up of 15.2 months, the 12-month failure free survival (FFS) rate was 27%, and the overall survival (OS) rate was 39%. Nine patients (18%) died during the first cycle of therapy, and two patients (4%) died during the second cycle. There were no significant differences between the rates of OR, OS, and FFS in the current study and those obtained with hyper-CVXD alone on a prior study.
The study regimen had activity and significant toxicity in patients with RS or fludarabine-refractory CLL. It was not clearly better compared with hyper-CVXD alone in this patient population. Cancer 2003;97:1711–20. © 2003 American Cancer Society.
The management of patients with Richter syndrome (RS) or refractory chronic lymphocytic leukemia (CLL) is complex, and their clinical outcome remains very poor, even with intensive chemotherapy regimens.1, 2 The introduction of fludarabine-based treatment strategies in the management of patients with CLL has increased response rates; however, the incidence of RS has remained constant.1, 3 The literature contains few reports of regimens specifically studied in patients with RS.1, 2 Most patients with RS receive therapy with variations on regimens that were developed for patients with non-Hodgkin lymphoma (NHL).1, 2, 4–7
We previously investigated combination cytotoxic regimens in patients with RS or fludarabine-refractory CLL, including cisplatinum, fludarabine, and cytosine arabinoside (ara-C) (PFA);1, 8 cyclophosphamide, fludarabine, and ara-C (CFA);1 and fludarabine, ara-C, cyclophosphamide, and prednisone (FACP).9 None of those regimens clearly improved patient outcome. The fractioned cyclophosphamide, vincristine, doxorubicin, and dexamethasone (hyper-CVAD) regimen, alternating with methotrexate and ara-C, is a very effective induction regimen for patients with previously untreated or recurrent acute lymphoblastic leukemia (ALL).10–14 In a recent comparison with the vincristine, doxorubicin, and dexamethasone (VAD) regimen, complete remission (CR) rates, percentages of patients in CR after 1 course of treatment, percentages of patients with first-cycle Day 14 blasts > 5%, and survival rates all improved significantly with the hyper-CVAD regimen.11 This regimen also is active in adult patients with other lymphoproliferative disorders, including mantle cell lymphoma (MCL), Burkitt-type ALL (both human immunodeficiency virus-associated and nonassociated), and multiple myeloma.15–19 Based on these data, we studied a modified hyper-CVAD regimen in patients with RS.20 This altered regimen (hyper-CVXD) substituted liposomal daunorubicin (DaunoXome) for doxorubicin and did not include methotrexate or ara-C. Liposomal daunorubicin has significant single-agent activity against refractory lymphomas.21 Combination regimens that include liposomal daunorubicin have significant activity in patients with refractory lymphoproliferative disorders, including patients with refractory NHL.22, 23 Investigators have suggested that liposomal encapsulation of cytotoxic agents may improve drug efficacy and reduce toxicity.24–26 In a cohort of 29 patients with RS who were treated with the hyper-CVXD regimen, 11 patients (38%) achieved CR, and 1 patient had a partial remission (PR), for an OR rate of 41%.20 The maturing positive results in patients with Burkitt-type ALL led us to reconsider the omission of alternating methotrexate and ara-C from the hyper-CVXD regimen that was studied previously in patients with RS.19
Reports also were emerging concerning the combination of rituximab with chemotherapy regimens for both patients with indolent NHL27 and patients with aggressive NHL.28–31 Rituximab is a chimeric murine/human monoclonal antibody with significant single-agent activity in patients with follicular NHL, CLL, and MCL.32–34 The achievement of molecular CR, as assessed by polymerase chain reaction (PCR)-based assay of t(14;18), in seven of eight patients with follicular NHL who were treated with six cycles of cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) combined with rituximab was very encouraging.27 OR and CR rates of 95% and 55%, respectively, were achieved in a cohort of 40 patients who were treated with this regimen.27 CHOP alone is associated with rare, if any, molecular CR in this clinical situation. In vitro studies also have demonstrated that rituximab sensitizes both drug-resistant B-cell NHL cell lines and malignant lymphocytes from patients to cell killing by cytotoxic drugs.35, 36 Therefore, we added rituximab to each course of chemotherapy on the protocol regimen.
Recombinant granulocyte-macrophage–colony stimulating factor (GM-CSF) also was added to the protocol regimen because of its potential to avoid or reduce associated myelosuppression. GM-CSF, rather than granulocyte-colony stimulating factor, was chosen because of the potential additional benefits of monocyte stimulation.8 Aside from potential activity against relevant opportunistic infections in these immunosuppressed patients, it has been postulated that GM-CSF may enhance the efficacy of rituximab by increasing cellular expression of CD20 and/or its antibody dependent cell toxicity.37–39 Thus, in an attempt to improve on the efficacy of the hyper-CVXD regimen, we conducted a Phase II study of hyper-CVXD alternating with methotrexate and ara-C, with both rituximab and GM-CSF, in patients with RS or fludarabine-refractory CLL.
MATERIALS AND METHODS
Patients with RS or fludarabine-refractory CLL were entered on the study between November 1999 and February 2001 after written informed consent was obtained according to institutional guidelines and after approval of the protocol by the Institutional Review Board. Eligibility criteria were 1) age > 15 years; 2) Zubrod performance score ≤ 3; and 3) normal cardiac, hepatic (serum bilirubin ≤ 2 mg/dL), and renal function (serum creatinine ≤ 2mg%), unless attributed to RS or CLL. RS was documented based on histologic assessment of fine-needle aspiration (FNA) and/or biopsy of involved tissue. A diagnosis of RS was not made based on FNA results alone. The pathologic diagnosis was confirmed independently in each patient (M.A.). Pretreatment evaluation included history and physical examination; complete blood count, differential, and platelets counts; serum chemistries (sequential multiple analysis 12/60), including liver and renal function studies; bone marrow aspiration for morphologic analysis and staining; bone marrow biopsy; cytogenetic analysis; flow cytometry, immunophenotyping; electrocardiogram, cardiac scan, and/or echocardiogram; and documentation of extent of disease. Patients with insufficient metaphases had ≤ 10 analyzable metaphases. Fludarabine-refractory CLL was defined as failure to respond to the most recent prior fludarabine-containing regimen—patients with recurrent disease after a response were not eligible for study therapy.
The study regimen consisted of six cycles of an augmented hyper-CVXD regimen, as detailed in Figure 1. Empiric antibiotic prophylaxis was given to all patients while they were on study as follows: fluconazole, 200 mg orally daily; acyclovir, 200 mg orally daily or valacyclovir, 500 mg orally daily; and trimethoprim/sulfomethoxazole DS, 1 tablet orally daily. Subsequent cycles of therapy were administered as soon as the neutrophil count was > 3 × 109/L or on Day 21, whichever occurred first (provided the neutrophil count was > 109/L on Day 21). The minimum time allowed between cycles was 14 days. The following dose adjustments were made if indicated: The vincristine dose was reduced to 1 mg if the serum bilirubin level was > 2 mg/dL; the liposomal daunorubicin dose was reduced by 25% if the serum bilirubin level was 2–3 mg/dL, by 50% if the serum bilirubin level was 3–4 mg/dL, and by 75% if the serum bilirubin level was > 4mg/dL; the cyclophosphamide dose was reduced by 50% for Grade 2 hemorrhagic cystitis, was reduced by 75% for Grade 3 hemorrhagic cystitis, and was stopped if Grade 4 hemorrhagic cystitis occurred. Patients commenced allopurinol therapy prior to the first cycle of therapy and were monitored closely for tumor lysis syndrome while they were on study.
Endpoints and Statistical Methods
A CR was defined as the complete disappearance of all detectable clinical and radiographic evidence of disease with the disappearance of all disease-related symptoms and normalization of biochemical abnormalities definitely assignable to the tumor. If bone marrow involvement was documented, then immature cells (blasts, lymphoma cells, or prolymphocytes) had to be reduced to a total of < 5% on bone marrow aspirate and/or biopsy. A PR was defined as a reduction ≥ 50% of all measurable disease with no appearance of new lesions. Any other responses were considered protocol failures. OS was measured from the time of entry onto this trial until death from any cause or last follow-up. Failure free survival (FFS) was defined from the time of entry onto this trial until disease progression, recurrence, or toxic death. Survival curves were calculated according to the method of Kaplan and Meier and were compared using the log-rank test. Toxicity was graded on a scale of 0 to 5 using the National Cancer Institute Common Toxicity Criteria (version 2.0). All patients who received any therapy on study were considered evaluable for toxicity. Univariate and multivariate regression analyses of prognostic factors for response and survival were performed. To compare the effects of the study augmented hyper-CVXD regimen with the effects of the hyper-CVXD-alone regimen, results of the study were compared with those obtained with the latter regimen in a similar patient population.20
The clinical characteristics of 49 patients who were treated on the study are summarized in Table 1. Thirty patients had RS, 19 patients had fludarabine-refractory CLL, and 42 patients on the study (86%) had received prior fludarabine therapy. Forty-eight patients (98%) had received prior chemotherapy; and the median number of previous regimens was 4; 42 patients (86%) had received ≥ 2 regimens. The median interval to a diagnosis of RS from a diagnosis of CLL was 52 months (range, 0–112 months).
|Characteristic||No. of patients||%|
|Performance status > 1||10||20|
|Elevated LDH (> 618 IU/L)||36||73|
|Elevated β2M (> 3 mg/L)||39/41||95|
|Elevated β2M (> 4 mg/L)||35/41||85|
|Bulky lymphadenopathy (> 5 cm)||37||56|
|Bone marrow involvement||46||94|
|del 13q14.3, del 11q22-q23, +12, del 17p13.3||4||9|
|del 13q14.3, del 11q22-q23, +12, or del 17p13 and additional chromosome abnormalities||16||37|
|Random structural abnormalities||3||7|
|Hemoglobin < 10 g/dL||20||41|
|Neutrophil count < 1 × 109L||45||92|
|Platelet count < 100 × 109L||28||57|
|Albumin < 3.5 g/dL||28||57|
|CD20 ≥ 20%||34/45||76|
|No. of prior treatment regimens|
|Prior fludarabine therapy||42||86|
|Prior rituximab therapy||25||51|
|Prior campath therapy||11||22|
Nine of 49 patients (19%) achieved a CR, and 11 patients (22%) achieved a PR, for an OR rate of 41%. The median time to documentation of CR was 2.4 months (mean, 2.7 months; range, 0.5–5.7 months). With a median follow-up of 8.2 months (range, 2–15 + months), 24 patients remained alive, with 15 patients in remission. Of the 9 patients who achieved a CR, the median CR duration was 10 months: Three patients remained in CR at 12 + months and 6 + months, including 1 patient at 10 + months who underwent allogeneic stem cell transplantation as consolidation after achievement of CR; 5 patients developed recurrent disease 2 months, 2 months, 3 months, 5 months, and 11 months after they achieved a CR, respectively; and 1 patient was lost to follow-up.
Responses by pretreatment characteristics are summarized in Table 2. Patients age ≥ 60 years did not differ in terms of the CR rate, the OR rate, or the death rate during first two study cycles compared with patients age < 60 years. Factors that were associated with a higher CR rate included baseline platelet counts ≥ 100 × 109/L, < 3 prior regimens, less bulky lymphadenopathy (< 5 cm), and no prior treatment with rituximab. Elevations in lactate dehydrogenase (LDH) levels or β2 microglobulin (β2M) levels; the presence of constitutional symptoms or splenomegaly; histology of RS or CLL; the severity of baseline anemia and baseline absolute neutrophils count (ANC); the presence of CD20 expression on malignant lymphocytes; the presence of low albumin or chromosomal abnormalities; and prior treatment with fludarabine did not have a significant impact on response to study therapy.
|Characteristic||CR (%)||P value||RR (CR + PR) (%)||P value||12-month survival rate (%)||P value|
|< 60||7/26 (27)||0.10||13/26 (50)||0.16||47||0.07|
|≥ 60||2/23 (9)||—||7/23 (30)||—||15||—|
|0–1||8/39 (20)||0.44||18/39 (46)||0.13||48||0.001|
|> 1||1/10 (10)||—||2/10 (20)||—||10||—|
|Absence||8/34 (24)||0.16||15/34 (44)||0.50||51||0.15|
|Presence||1/15 (7)||—||5/15 (33)||—||23||—|
|≤ 618 IU/L||2/13 (15)||0.74||4/13 (31)||0.38||38||0.19|
|> 618 IU/L||7/36 (19)||—||16/36 (44)||—||41||—|
|< 4 mg/L||1/5 (20)||0.85||4/7 (57)||0.35||33||0.51|
|≥ 4 mg/L||6/36 (17)||—||13/34 (38)||—||40||—|
|< 3 g/dL||5/21 (24)||0.39||4/14 (29)||0.27||21||0.04|
|≥ 3 g/dL||4/28 (14)||—||16/35 (45)||—||46||—|
|< 5 cm||5/14 (36)||0.04||8/14 (57)||0.14||63||0.30|
|> 5 cm||4/35 (11)||—||12/35 (34)||—||31||—|
|Richter syndrome||8/30 (27)||0.06||13/30 (43)||0.65||28||0.41|
|Fludarabine-refractory CLL||1/19 (5)||—||7/19 (37)||—||54||—|
|Yes||5/26 (19)||0.87||13/26 (50)||0.16||24||0.55|
|No||4/23 (17)||—||7/23 (30)||—||50||—|
|Yes||2/17 (12)||0.38||3/17 (18)||0.01||9||0.01|
|No||7/32 (22)||—||17/32 (53)||—||56||—|
|No. of prior regimens|
|0–2||7/15 (47)||< 0.0001||10/15 (67)||0.01||67||0.04|
|3–15||2/34 (6)||—||10/34 (29)||—||29||—|
|≥ 10 g/dL||7/29 (24)||0.20||14/29 (48)||0.20||48||0.057|
|< 10 g/dL||2/20 (10)||—||6/20 (30)||—||27||—|
|< 30 × 109L||7/30 (23)||0.25||16/30 (53)||0.02||48||0.07|
|≥ 30 × 109L||2/19 (10)||—||4/19 (21)||—||23||—|
|Absolute neutrophil count|
|≥ 1 × 109L||1/4 (25)||0.71||2/4 (50)||0.69||50||0.97|
|< 1 × 109L||8/45 (18)||—||18/45 (40)||—||38||—|
|≥ 100 × 109L||8/21 (38)||0.002||14/21 (67)||0.001||71||0.0002|
|< 100 × 109L||1/28 (4)||—||6/28 (21)||—||14||—|
|< 20%||2/11 (18)||0.86||5/11 (45)||0.80||45||0.75|
|≥ 20%||7/34 (21)||—||14/34 (41)||—||35||—|
|Yes||6/42 (14)||0.07||15/42 (36)||0.07||37||0.48|
|No||3/7 (43)||—||5/7 (71)||—||68 (8 months)||—|
|Yes||2/25 (8)||0.05||8/25 (32)||0.20||34||0.09|
|No||7/24 (29)||—||12/24 (50)||—||43||—|
The median survival of patients on the study was 8.5 months, with a 12-month survival rate of 39% (Fig. 2). Survival rates by pretreatment characteristics are summarized in Table 2. There was a significant trend toward shorter survival for patients age ≥ 60 years compared with patients age < 60 years (12-month survival rate, 15% vs. 47%, respectively). Factors that were associated with a longer 12-month survival rate included a performance score of 0–1, platelet count ≥ 100 × 109/L, 0–2 prior regimens, absence of hepatomegaly, albumin > 3 g/dL, and baseline hemoglobin ≥ 10 g/dL. There was no significant association between OS and the presence of B-symptoms, elevated LDH levels, elevated serum β2M levels, bulky disease, histology of RS versus refractory CLL, presence of splenomegaly, degree of neutropenia, expression of CD20, prior treatment with fludarabine, or prior rituximab treatment.
Table 3 shows the parameters that retained significance when a multivariate regression analysis for OS was performed. At 12 months, FFS was 27% (Fig. 3). None of these parameters were significant for FFS when a regression analysis was performed. Although OS curves were similar (Fig. 4), patients with fludarabine-refractory CLL had a significantly higher survival rate compared with patients who had RS when a 2-month landmark analysis was performed (Fig. 5) to determine whether any significant differences between the two patient cohorts had been masked by the high early death rate. The prognostic significance of baseline leukocyte count and platelet counts for OS was evident equally in patients with CLL and in patients with RS. The performance score is a previously well-established prognostic factor for patients with aggressive NHL.40 However, no other components of the International Prognostic Index (IPI)40 or the overall IPI score were significantly predictive of OS or FFS for patients with either CLL or RS.
|Performance status 0–1||0.005|
|Fludarabine-refractory CLL vs. Richter syndrome||0.01|
|Leukocyte count < 30 × 109L||0.03|
|Platelet count ≥ 100 × 109L||0.05|
Twenty-four patients have died: 11 patients during the first or second cycles of therapy; 9 patients from progressive disease; and 4 patients from a combination of myelosuppression, sepsis, and progressive disease, including 1 patient with severe, uncontrolled graft-versus-host disease after undergoing a sibling-donor stem cell transplantation prior to study entry. Nine patients with RS who developed recurrent disease were alive at the time of this analysis, six patients after undergoing further salvage chemotherapy regimens. One patient has been lost to follow-up.
The total numbers of cycles administered on study was 151 (median per patient, 2 cycles; range, 1–6 cycles). Nine patients (18%) died during the first cycle of therapy, and two patients (4%) died during the second cycle. The median time to the start of the second cycle in the 40 patients (82%) who did so was 29 days (range, 18–57 days). The primary causes of death during the first course of induction were systemic fungal infections in four patients (Aspergillus terreus, Candida albicans, Candida krusei, and Candida guillermondii in one patient each); systemic bacterial infections (Escherichia coli and Mycobacterium avium complex, Stomatococcus, and both coagulase-negative Staphylococcus and Stomatococcus, respectively); cardiopulmonary arrest in one patient; and rapidly progressive disease in one patient. One death during the first cycle of therapy was attributed to rituximab in a patient who developed shortness of breath and subsequent cardiopulmonary arrest that was associated with extensive bilateral pulmonary infiltrates after the administration of rituximab alone. This male patient age 72 years had heavily pretreated (including prior rituximab), refractory CLL with a pretreatment leukocyte count count of 188 × 109/L (71% lymphocytes, 28% prolymphocytes, and LDH of 2658 IU/L). This patient had cigarette smoking-induced, chronic obstructive pulmonary disease and prior biopsy-proven pulmonary infiltrate with CLL. The causes of death in two patients during the second cycle of therapy were systemic bacterial infection (Pseudomonas aeruginosa) and rapidly progressive disease, respectively.
The most common Grade 4 toxicity was neutropenia, which occurred in all cycles of therapy for a median duration of 7 days (range, 1–33 days), with Grade 4 thrombocytopenia occurring in 40% of cycles. Thirty-nine percent of cycles were complicated with Grade 3 or 4 sepsis, with fever of unknown origin occurring in a further 20% of cycles. Hemorrhagic episodes occurred in 5% of cycles, including one patient who developed hemorrhagic cystitis. Three patients experienced thrombotic episodes: One patient had cavernous sinus thrombosis with concomitant documentation of Aspergillus niger and Staphylococcus aureus sinus and systemic infection, one patient had retinal artery thrombosis, and one patient had deep venous thrombosis. Twelve percent of cycles were associated with Grade 3 neurotoxicity (mainly steroid-related mood lability or depression; 8%) or peripheral neuropathy associated with vincristine (4%). Ten percent of cycles were associated with Grade 3 or 4 oral mucositis and/or esophagitis. Three percent of cycles were associated with Grade 3 diarrhea. Grade 3 renal failure as part of a septic episode occurred in two patients. Twenty percent of initial rituximab infusions were associated with Grade 1 or 2 fever, chills, or pruritus.
Compared with a prior hyper-CVXD regimen, which did not include methotrexate, ara-C, rituximab, or GM-CSF, that was given to 29 patients with RS, the rates of OR, OS, and FFS were similar (Figs. 2, 3).20 The rates of early death (on the first two cycles of study therapy) also were similar (18% vs. 14%, respectively; P = 0.8). Forty-one percent of patients achieved a CR or PR with hyper-CVXD alone in the previous study, and 41% of patients also achieved a CR or PR on the current study. Survival also was similar for both regimens (Fig. 2), with 12-month survival rates of 39% and 36% (P = 0.9). The 12-month FFS rates (Fig. 3) also were comparable (27% and 19%, respectively; P = 0.8). The incidence, severity, and duration of neutropenia and thrombocytopenia in the first cycle and in all cycles on this study were directly comparable with what was seen on the prior study with hyper-CVXD alone.
The study augmented hyper-CVXD regimen was associated with an OR rate of 41% and an early mortality rate of 18%. At 12 months, the OS rate was 39%, and the FFS rate was 27%. A Zubrod performance score of 0–1, a diagnosis of fludarabine-refractory CLL (vs. RS), a baseline leukocyte count < 30 × 109/L, and a baseline platelet count ≥ 100 × 109/L were favorable prognostic factors for OS on this study. Compared with a prior hyper-CVXD regimen that did not include methotrexate, ara-C, rituximab, or GM-CSF, the rates of early death, OR, OS, and FFS were similar.20
Thus, the more complex and expensive, augmented hyper-CVXD regimen that was used in this study did not offer an evident significant benefit. Overall, it is difficult to distinguish any major differences between the effectiveness of the PFA,1 CFA,1 or FACP9 regimens or either the original or augmented hyper-CVXD regimen in our sequential studies of patients with RS. It seems reasonable to pursue different approaches in patients with refractory CLL and RS in future studies.
It was particularly disappointing that the addition of rituximab to the standard hyper-CVXD regimen did not enhance the regimen efficacy for patients with RS. For both patients with low-grade B-cell NHL and MCL, it has been reported that rituximab adds therapeutic benefit to CHOP without significant added toxicity.27, 41 However, the Dana-Farber group recently documented the failure of molecular CR associated with CHOP-rituximab therapy to predict for improved progression free survival in patients with previously untreated MCL.41 Thus, the effect of adding rituximab to combination cytotoxic therapy in the treatment of patients with lymphoproliferative disorders may differ among the various malignancies and requires further study.29 Rituximab therapy has been associated with sudden death, which was seen in one patient on this study. Sudden deaths have been reported as very rare occurrences associated with the administration of rituximab to patients with lymphoproliferative disorders42, 43—the great majority of patients, including those on high-dose therapy and those experiencing significant infusion-related side effects or tumor lysis syndrome, have transient symptoms and are able to receive further rituximab therapy.44, 45
A significant difference in terms of the types of systemic fungal infections seen in these immunosuppressed patients was seen on this study relative to our previous experience, with the emergence of Candida krusei and guillermondii as lethal pathogens. The shift from established candidal species, i.e., Candida albicans, as pathogens in patients with hematologic malignancies has been noted recently in the context of increasing fluconazole use as a prophylactic agent.46, 47 Fluconazole was used as prophylaxis on this study, and all but the single previously untreated patient with RS received prior fluconazole prophylaxis while receiving other cytotoxic and/or monoclonal antibody therapies. Opportunistic infections remain a major problem and a significant contributor to the high early death rate seen on current anti-NHL type regimens for patients with RS: The early integration of novel antifungal agents into future regimens seems warranted.48
It also was disappointing in this context that the inclusion of GM-CSF in the augmented hyper-CVXD regimen did not overtly reduce the incidence or lethality of systemic fungal infections. A large body of data on the ability of GM-CSF to reduce death from fungal sepsis in various animal models has been generated.49 Investigators have suggested that GM-CSF also may reduce the mortality associated with chemotherapy-induced myelosuppression in patients with hematologic malignancies.49, 50 GM-CSF has been approved in the U.S. as an adjunct therapy to improve the survival of elderly patients undergoing induction therapy for acute myeloid leukemia: The supportive data for this indication were based on an Eastern Cooperative Oncology Group randomized, double-blind, placebo-controlled study in which the apparent main benefit in patients who received GM-CSF was a reduction in overall and lethal, systemic fungal infections.51 This survival benefit has not been confirmed in other studies of this issue.52, 53 In this study, Grade 4 neutropenia occurred in all cycles of therapy, and only eight patients (16%) received the full six cycles of therapy intended on protocol. The incidence, severity, and duration of neutropenia and thrombocytopenia were comparable directly with the hyper-CVXD and augmented hyper-CVXD regimens.
The majority of patients on the current study (43 patients; 88%) had adequate metaphases to allow karyotypic analysis: Of these, 28 patients (65%) had clonal cytogenetic abnormalities, a finding consistent with our previous cohort, in which 50% of patients had similar cytogenetic abnormalities.2 However, none of the subgroups of cytogenetic abnormalities (Table 1) were predictive independently for the achievement of OR or the duration of OS or FFS. Only 4 patients (9%) had cytogenetic abnormalities that were considered characteristic for CLL (del 13q, del 11q, trisomy 12, or del 17p), whereas 16 patients (37%) had additional cytogenetic abnormalities. Because most of these patients had RS, these data suggest that clonal evolution may be involved in the pathogenesis of this malignancy, a suggestion consistent with the observation that patients with CLL and multiple cytogenetic abnormalities are more likely to develop RS.54 The establishment of a relation between these progressively acquired genetic lesions and specific therapeutic targets would be a major step forward in the treatment of patients with this intensively refractory, lymphoproliferative malignancy.
The currently reported data, along with our overall institutional experience in patients with RS,1, 2, 9, 20 indicate that these patients reasonably may be offered investigational therapies at the time of diagnosis. Of the available standard regimens, the hyper-CVXD protocol is active, is relatively straightforward to administer, and is at least as effective as the alternative regimes that we studied. The addition of methotrexate, ara-C, rituximab, and GM-CSF did not improve its efficacy in patients with RS. Our data to date indicate that patients with RS legitimately may be offered investigational therapies at the time of diagnosis or as consolidation approaches.
- 18Hyperfractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone and highly active antiretroviral therapy for patients with acquired immunodeficiency syndrome-related Burkitt lymphoma/leukemia. Cancer. 2002; 94: 1492–1499., , , et al.
- 47Nosocomial candidaemias due to species other than Candida albicans in cancer patients. Aetiology, risk factors, and outcome of 45 episodes within 10 years in a single cancer institution. Support Care Cancer. 1999; 7: 428–431., , , et al.
- 51A randomized placebo-controlled Phase III study of granulocyte-macrophage colony-stimulating factor in adult patients (> 55 to 70 years of age) with acute myelogenous leukemia: a study of the Eastern Cooperative Oncology Group (E1490). Blood. 1995; 86: 457–462., , , et al.
- 52Results of a randomized double-blind placebo-controlled trial evaluating sequential high-dose cytosine arabinoside/mitoxantrone chemotherapy with or without granulocyte/macrophage-colony-stimulating factor in high-risk myelodysplastic syndromes. J Cancer Res Clin Oncol. 1999; 125: 369–374., , , et al.
- 53Use of recombinant GM-CSF during and after remission induction chemotherapy in patients aged 61 years and older with acute myeloid leukemia: final report of AML-11, a Phase III randomized study of the Leukemia Cooperative Group of European Organisation for the Research and Treatment of Cancer and the Dutch Belgian Hemato-Oncology Cooperative Group. Blood. 1997; 90: 2952–2961., , , et al.