Fax: (507) 266-4972.
Pentostatin, cyclophosphamide, and rituximab regimen in older patients with chronic lymphocytic leukemia
Article first published online: 18 MAY 2007
Copyright © 2007 American Cancer Society
Volume 109, Issue 11, pages 2291–2298, 1 June 2007
How to Cite
Shanafelt, T. D., Lin, T., Geyer, S. M., Zent, C. S., Leung, N., Kabat, B., Bowen, D., Grever, M. R., Byrd, J. C. and Kay, N. E. (2007), Pentostatin, cyclophosphamide, and rituximab regimen in older patients with chronic lymphocytic leukemia. Cancer, 109: 2291–2298. doi: 10.1002/cncr.22662
- Issue published online: 18 MAY 2007
- Article first published online: 18 MAY 2007
- Manuscript Accepted: 22 JAN 2007
- Manuscript Revised: 9 JAN 2007
- Manuscript Received: 29 NOV 2006
- National Institutes of Health National Cancer Institute (NCI). Grant Numbers: CA 95241, NCI K12 CA90628, CA113408
- chronic lymphocytic leukemia;
- elderly patients;
- renal failure;
- β-2 microglobulin;
The prevalence of chronic lymphocytic leukemia (CLL) increases with age. Although chemoimmunotherapy (CIT) has dramatically improved response rates in patients with CLL, some CIT regimens are not well tolerated by many patients ≥70 years of age.
Sixty-four previously untreated patients with CLL and serum creatinine <1.5 times the upper limit of normal who met National Cancer Institute (NCI) 96-WG criteria for treatment received pentostatin (2 mg/m2), cyclophosphamide (600 mg/m2), and rituximab (375 mg/m2). The authors measured performance status at study entry and used age, weight, and baseline creatinine to calculate creatinine clearance (CrCl).
Eighteen of 64 (28%) patients were ages ≥70 years. Although individuals ages ≥70 years were more likely to have delayed treatment cycles (28% vs 7%; P = .03), there were no significant differences in the number of cycles administered, need for dose reductions, or grade 3–4 hematologic, infectious, or other toxicities. No significant differences in overall response rate, complete response rate, or progression-free survival were observed by age. Twenty-five (39%) patients had a CrCl < 70 mL/min (range, 34–67). Although individuals with CrCl < 70 were more likely to require dose reduction (24% vs 5%; P = .05), there were no significant differences in the number of cycles administered or grade 3–4 hematologic, infectious, or other toxicities. No significant difference in overall response rate, complete response rate, or progression-free survival were observed between patients with CrCl ≥ 70 mL/min and those with CrCl < 70 mL/min.
In this clinical trial, the PCR regimen was well tolerated by older patients and individuals with CrCl ≤ 70. The efficacy of PCR was not significantly affected by age or renal function. These findings suggest PCR may be a good therapeutic option for older patients and those with modestly decreased renal function. Cancer 2007; 109:2291–8. © 2007 American Cancer Society.
The prevalence of chronic lymphocytic leukemia increases with age.1–4 The median age at the time of diagnosis is 65 years to 70 years, and because most patients are observed for several years before treatment, more than half of patients who require therapy are older than 70 years of age.1–5 Accordingly, identification of effective treatment regimens that can be safely administered to older patients is a critical need in chronic lymphocytic leukemia therapy. Despite these facts, older patients are typically under-represented in most clinical trials. The median age of patients with chronic lymphocytic leukemia accrued to first-line therapy trials is often ≤60 years of age,6–9 and many trials exclude patients aged >65 years10, 11 or 70 years9 because therapies under development are thought to be too toxic for older patients. Clinical trials also typically have strict eligibility criteria, which mandate that only patients with well preserved organ function can be enrolled, further limiting application of these therapies to older individuals. Because of these limitations, it is often unclear whether treatment regimens shown to be highly efficacious in phase 2 clinical trials can be effectively and safely administered to unselected patients with chronic lymphocytic leukemia who are often older and/or infirm.
The major advance in the treatment of chronic lymphocytic leukemia over the last decade has been the development of chemoimmunotherapy regimens that combine purine nucleoside analogs and monoclonal antibodies. The fludarabine and rituximab combination (FR)7, 12 and the fludarabine, cyclophosphamide, rituximab combination (FCR)6 have been the most widely tested and used chemoimmunotherapy regimens. In phase 2 trials, these approaches achieved a response in >90% of patients.6, 12 Notably, the FCR regimen induced a complete response in 70% of patients and led to molecular complete responses in some individuals.6 Perhaps most importantly, it appears these chemoimmunotherapy approaches may prolong survival.13
Unfortunately, the toxicity of some chemoimmunotherapy regimens precludes their administration to many older patients. The M.D. Anderson Cancer Center (Houston, Tex) group has reported that the FCR regimen is not well tolerated in individuals ages ≥70 years of age14 or those with β-2-microglobulin (B2M) >4.0 mg/dL.6 Specifically, when administered as a first-line therapy to patients ≥70 years of age, 75% of patients experienced grade 3–4 myelosuppression, 11% experienced grade 3–4 infection, and fewer than half were able to complete the intended 6 cycles of induction.14 The investigators conclude that, although efficacious, a high incidence of early discontinuations of treatment due to hematological and nonhematological toxicity limited clinical benefits of FCR for patients ages ≥70 years.14
Along with fludarabine, 2 other purine nucleoside analogs are widely used to treat lymphoid malignancies: 2-chlorodeoxyadenosine (2-CDA) and pentostatin. Whereas all 3 purine nucleoside analogs have demonstrated single-agent efficacy in chronic lymphocytic leukemia,15–22 and both fludarabine and 2-CDA have demonstrated higher response rates than alkylating agents in randomized trials,23–26 no randomized controlled trials comparing these agents have been reported to date. Based on indirect evidence, some have suggested that pentostatin is better tolerated and less myelosuppressive than fludarabine when used in combination with cyclophosphamide in chronic lymphocytic leukemia patients.27, 28
We recently reported the results of a phase 2, multicenter study of pentostatin, cyclophosphamide, and rituximab (PCR) as first-line treatment for patients with chronic lymphocytic leukemia, which demonstrated PCR is an efficacious therapy in this disease.29 Because of the important issues detailed above concerning whether chemoimmunotherapy can be safely administered to older chronic lymphocytic leukemia patients, we examined how age, creatinine clearance (CrCl), β-2 microglobulin (B2M), and performance status (PS) relate to the efficacy and tolerability of this pentostatin-based chemoimmunotherapy regimen.
MATERIALS AND METHODS
As recently reported, we treated 64 previously untreated chronic lymphocytic leukemia patients meeting NCI 96 chronic lymphocytic leukemia Working Group criteria29 for treatment with pentostatin (2 mg/m2), cyclophosphamide (600 mg/m2) and rituximab (375 mg/m2) for a planned duration of 6 cycles.30 All patients received hydration during and after treatment with each dose of pentostatin. In accord with NCI working group recommendations, eligibility required an Eastern Cooperative Oncology Group performance status (ECOG PS)of 0–3 and serum creatinine of ≤1.5 times the upper limit of normal (≤1.8 mg/dL for women and ≤2.1 mg/dL for men).29 Patients were also required to have a total bilirubin and aspartate aminotransferase (AST) ≤3 times the upper limit of normal at the treating institution. There was no age restriction on eligibility. We measured ECOG PS at study entry and used age, weight, and baseline creatinine to calculate the CrCl of all patients at the time of study enrollment by using the Cockroft-Gault equation.31 Toxicity was graded by using the NCI Common Toxicity Criteria (v.2)32 supplemented by the NCI chronic lymphocytic leukemia Working Group guidelines for hematologic toxicity.29 Response to treatment was classified according to NCI Criteria.29 The 3 patients who had no evidence of disease at completion of therapy but who had residual cytopenia were considered “CR with residual cytopenia” and were grouped with complete response (CR) patients for this analysis.
Statistical Design and Analysis
Descriptive statistics and graphical evaluation were used to assess the distribution of age, CrCl, and performance status in this cohort as well as differences in the various endpoints of interest between dichotomized groups. On the basis of cutpoints used in previous studies,14, 33 patients who were aged 70 years or older were classified as “older” chronic lymphocytic leukemia patients. In addition, a cutpoint of 70 mL/min was used for CrCl based on the use of this threshold as an eligibility criterion for treatment in some chronic lymphocytic leukemia chemoimmunotherapy trials.14 Patients were also dichotomized on the basis of having a B2M level ≤4.0 mg/dL or >4.0 mg/dL based on previous reports that this cutpoint predicts early discontinuation of FCR therapy.6 Quantitative comparisons in the various endpoints of interest were evaluated by using Fisher exact test for categorical endpoints and Wilcoxon rank sum tests for continuous endpoints. Because of the small number of patients in the performance status 2 or 3 groups, these patients were combined with the performance status 1 patients for analysis. Overall, comparisons were limited by the available sample sizes, and sufficient power was available for detecting only relatively large differences between groups. For example, at least 80% power was available to detect a 30% to 40% or greater difference (eg, 30% vs 70%; 10% vs. 40%) in response or toxicity incidence rates between patients ages <70 (n = 46) vs ≥70 years (n = 18). To detect a 20% difference in rates (eg, 5% vs 25% or 75% vs 95%), we would have only 58% power. Similarly, at least 80% power was available to detect a >25% to 35% difference (eg, 30% vs 65%; 5% vs 30%) in response or toxicity between patients with CrCl ≥ 70 (n = 38) vs CrCl <70 (n = 24) as well as between patients with performance status (PS) PS = 0 (n = 34) and those with PS ≥ 1 (n = 30). To detect a 20% difference in rates (eg, 5% vs 25% or 75% vs 95%), we would have only 63% power. Finally, we explored differences in progression-free survival (PFS) between groups. Based on the sample size, we had at least 80% to detect a hazard ratio of ≥3. Statistical significant for all analyses was declared at P < .05.
The cumulative patient characteristics, overall response rate (ORR), and response according to prognostic parameters (IgVH gene mutation status, CD38, ZAP-70, cytogenetic abnormalities by fluorescent in situ hybridization [FISH]) for the 64 patients treated with PCR in this trial have been previously reported.30 In the current evaluation, we first assessed how age influenced the efficacy and tolerability of PCR therapy. Eighteen of 64 (28%) treated patients were ages ≥70 years (range, 70–80 years). Fourteen (78%) of the patients ages ≥70 completed the intended 6 cycles of induction. Although individuals ages ≥70 years were more likely to have a dose delay of >1 week at some point during the trial (28% vs 7%; P = .03), there was no difference in the median number of cycles administered, patients requiring dose reductions, or patients with grade 3–4 hematologic toxicity, infectious complications, or other nonhematologic toxicity (Table 1). The rates of any grade 3–4 hematologic toxicity was similar among individuals ages ≥70 years and those aged <70 years (61% vs 48%; P = .41). Although grade 3–4 neutropenia was more common among individuals ages ≥70 years (56% vs 35%; P = .16), this difference was not statistically significant, and no increase in grade 3–4 infectious complications was observed (6% vs 11%; P = .67). No significant differences in overall response (OR) (83% vs 93%; P = .34) or complete response rate (39% vs 41%; P = .86) was observed between patients ages ≥70 years compared with those aged <70 years.
|<70 y (n=46)||≥70 y (n=18)||P|
|No. (%)||No. (%)|
|Mean/median no. cycles received||5.3/6||5.6/6||.36|
|No. patients who required dose reduction||6 (13)||2 (11)||1.00|
|No. of patients who required dose delay >1 wk||3 (7)||5 (28)||.03|
|No. of patients with grade 3–4 infectious complication||5 (11)||1 (6)||.67|
|No. of patients with grade 3–4 nonhematologic toxicity other than infection||13 (28)||4 (22)||.76|
|No. of patients with any grade 3–4 hematologic toxicity||22 (48)||11 (61)||.41|
|Grade 3–4 neutropenia||16 (35)||10 (56)||.16|
|Grade 3–4 thrombocytopenia||10 (22)||3 (17)||.74|
|Grade 3–4 anemia||1 (2)||0 (0)||1.00|
|Overall response rate||43 (93)||15 (83)||.34|
|Complete response rate||19 (41)||7 (39)||.86|
We next evaluated whether CrCl related to efficacy and tolerability of PCR. Twenty-five (39%) patients had a CrCl < 70 mL/min (range, 34–67), including 14 (22%) with a CrCl ≤ 60 mL/min and 4 (6%) with CrCl ≤ 50 mL/min. Although individuals with CrCl < 70 mL/min were more likely to require a dose reduction (24% vs 5%; P = .05) at some point during treatment, there was no difference in the median number of cycles administered, patients who required a dose delay, or patients with nonhematologic toxicity based on renal function (Table 2). There was more grade 3–4 hematologic toxicity in individuals with CrCl < 70 mL/min, although this difference was not statistically significant (64% vs 42%; P = .12). Additionally, a trend toward increased grade 3–4 neutropenia among patients with CrCl < 70 (56% vs 32%; P = .07) was observed, although no increase in grade 3–4 infectious complications was apparent (8% vs 11%; P = 1.0). Decreased renal function had no effect on OR (92% vs 89%; P = 1.0) or complete response rates (36% vs 45%; P = .60).
|Creatinine clearance ≥70 mL/min* (n = 38)||Creatinine clearance <70 mL/min (n = 25)||P|
|No. (%)||No. (%)|
|Mean/median no, cycles received||5.4/6||5.7/6||.34|
|No. patients who required dose reduction||2 (5)||6 (24)||.05|
|No. of patients who required dose delay >1 wk||5 (13)||3 (12)||1|
|No. of patients with grade 3–4 infectious complication||4 (11)||2 (8)||1.00|
|No. of patients with grade 3–4 nonhematologic toxicity other than infection||8 (21)||8 (32)||.33|
|No. of patients with any grade 3–4 hematologic toxicity||16 (42)||16 (64)||.12|
|Grade 3–4 neutropenia||12 (32)||14 (56)||.07|
|Grade 3–4 thrombocytopenia||6 (16)||6 (24)||.52|
|Grade 3–4 anemia||1 (3)||0 (0)||1.0|
|Overall response rate||34 (89)||23 (92)||1.00|
|Complete response rate||17 (45)||9 (36)||.60|
We next evaluated the effect of performance status (PS) on efficacy and tolerability of PCR. Thirty-four (53%) patients were PS = 0, 24 (38%) were PS = 1, and 6 (9%) were PS = 2 or 3. Because of the low number of patients with PS = 2–3 (n = 6), we compared the tolerability of PCR between those with PS = 0 and those with PS ≥ 1 (Table 3). Individuals with PS ≥1 were more likely to require a dose reduction (23% vs 3%; P = .02) at some point during treatment. Although more patients with PS ≥ 1 required a dose delay of > 1 week at some point during treatment (20% vs 6%; P = .13), this difference was not statistically significant. Patients with PS ≥ 1 also experienced more grade 3–4 nonhematologic toxicity other than infection (40% vs 15%; P = .03) and had more grade 3–4 hematologic toxicity (67% vs 38%; P = .03) because of a significantly greater incidence of grade 3–4 neutropenia (60% vs 24%; P = .005). Despite these trends, no significant differences in the median number of cycles administered or the number of patients with infectious complications were observed among those with PS = 0 compared with those with PS ≥ 1. Performance status did not effect the OR (94% vs 87%; P = .41) or complete response rates (47% vs 33%; P = .31).
|PS = 0 (n = 34)||PS ≥ 1* (n = 30)||P|
|No. (%)||No. (%)|
|Mean/median no. cycles received||5.6/6||5.3/6||.34|
|No. patients who required dose reduction||1 (3)||7 (23)||.02|
|No. of patients who required dose delay >1wk||2 (6)||6 (20)||.13|
|No. of patients with grade 3–4 infectious complication||3 (9)||3 (10)||1.00|
|No. of patients with grade 3–4 nonhematologic toxicity other than infection||5 (15)||12 (40)||.03|
|No. of patients with any grade 3–4 hematologic toxicity||13 (38)||20 (67)||.03|
|Grade 3–4 neutropenia||8 (24)||18 (60)||.005|
|Grade 3–4 thrombocytopenia||6 (18)||7 (23)||.76|
|Grade 3–4 anemia||1 (3)||0 (0)||1.00|
|Overall response rate||32 (94)||26 (87)||.41|
|Complete response rate||16 (47)||10 (33)||.31|
On the basis of reports that B2M influences the tolerability of FCR,6 we next assessed how B2M influenced the efficacy and tolerability of PCR therapy. B2M at the time of study entry was available for 61 of 64 (95%) patients. Thirty-one of these 61 (51%) patients had a B2M ≤ 4.0 mg/dL and 30 (49%) had a B2M > 4.0 mg/dL. Although individuals with a B2M > 4.0 mg/dL were more likely to have a dose delay of >1 week at some point during the trial (28% vs 7%; P = .03), there was no difference in the median number of cycles administered, patients who required dose reductions or dose delays, or patients with grade 3–4 nonhematologic toxicity (Table 4). The rates of any grade 3–4 hematologic toxicity appeared to be higher among patients with B2M > 4.0 (63% vs 39%; P = .07) primarily because of an increased frequency of grade 3–4 neutropenia (53% vs 26%; P = .04). Despite this difference in grade 3–4 neutropenia, no increase in grade 3–4 infectious complications was observed among patients with B2M ≥ 4.0 (10% vs 6%; P = .67). No significant differences in overall response (83% vs 97%; P = 0.1) or CR rate (37% vs 45%; P = .61) were observed between patients B2M > 4.0 compared with those B2M ≤ 4.0.
|B2M ≤ 4.0 mg/dL (n = 31)||B2M > 4.0 mg/dL (n = 30)||P|
|Mean/median # cycles received||5.5/6||5.5/6||.96|
|No. patients who required dose reduction||4 (13)||4 (13)||1.00|
|No. of patients who required dose delay >1 wk||2 (6)||4 (13)||.42|
|No. of patients with grade 3–4 infectious complication||2 (6)||3 (10)||.67|
|No. of patients with grade 3–4 non-hematologic toxicity other than infection||7 (23)||9 (30)||.57|
|No. of patients with any grade 3–4 hematologic toxicity||12 (39)||19 (63)||.07|
|Grade 3–4 neutropenia||8 (26)||16 (53)||.04|
|Grade 3–4 thrombocytopenia||6 (19)||7 (23)||.76|
|Grade 3–4 anemia||1 (3)||0 (0)||1.00|
|Overall response rate||30 (97)||25 (83)||.10|
|Complete response rate||14 (45)||11 (37)||.61|
Finally, we evaluated how age, CrCl, performance status, and B2M related to PFS after treatment with PCR (median follow-up for living patients = 33 months). No differences in PFS were noted for any of these analyses (Figure 1A–D).
We have examined efficacy and tolerability of the PCR regimen among patients ages ≥70 years and those with modestly decreased CrCl. We found that older patients tolerated PCR as well as younger patients. Older patients were as likely as younger patients to complete the intended 6 cycles of therapy and to achieve complete or partial remission without excess grade 3–4 toxicity. This appears to contrast with the tolerability of the FCR regimen in older patients.6, 14 Additionally, within the restrictions of the eligibility criterion (serum creatinine ≤1.5 times the upper limit of normal range), CrCl did not dramatically influence the efficacy or tolerability of the PCR regimen. This is notable given that nearly 40% of patients enrolled on this study had a CrCl < 70 mL/min, a characteristic that would exclude them from participation in some ongoing phase 3 chemoimmunotherapy trials.33 Consistent with the concept that physiologic age is more important than chronologic age, poorer performance status at the time of enrollment did decrease tolerability; however, it did not significantly affect the number of treatment cycles given or treatment efficacy.
Previously performed phase 2 studies suggest that chemoimmunotherapy may improve both response rates and survival for patients with chronic lymphocytic leukemia13 and phase 3 trials to validate this approach are ongoing.33 Unfortunately, many patients with chronic lymphocytic leukemia who require therapy are ages ≥70 years1–4 and appear to tolerate some chemoimmunotherapy regimens less well than younger patients.6, 14 Older patients are also more likely to have decreased renal function and other comorbidities that can influence the ease of administration, tolerability, toxicity, and efficacy of therapy. Additional barriers, such as more limited mobility and/or dependence on others for transportation, also make compliance with regimens that require frequent office visits more difficult for such patients. Despite these characteristics, many such patients will have a prolonged life expectancy if successfully treated for their chronic lymphocytic leukemia.
After the observation that patients ages ≥70 years tolerate some chemoimmunotherapy regimens poorly, some clinicians have limited the use of chemoimmunotherapy to younger patients and have employed “less-toxic therapies” such as chlorambucil, rituximab monotherapy, steroid-based treatments, or single-agent purine nucleoside analogs for older patients.34–37 This approach sacrifices efficacy for tolerability. Importantly, a lack of durable control of the leukemia has potential to result in both diminished quality of life and length of life for many older patients. A second approach is to allow older patients to receive aggressive chemoimmunotherapy with regimens such as FCR, provided they have well preserved renal function and good performance status.33 Although this strategy expands the number of patients who can receive chemoimmunotherapy, it still precludes treatment of many older individuals. A third approach is to attempt to identify chemoimmunotherapy regimens that achieve both a high degree of efficacy and that can be well tolerated by older patients. The ideal treatment regimen for older chronic lymphocytic leukemia patients would be one that is highly efficacious, easy to administer, requires few office visits, and has modest toxicity.
Our analysis suggests that the PCR regimen may fulfill many of these criteria. The regimen appears to be both quite effective and well tolerated by chronic lymphocytic leukemia patients older than the age of 70 years as well as those with modestly reduced CrCl or elevated B2M. Importantly, patients aged ≥70 years, as well as those with modestly reduced renal function, increased B2M, or PS ≥ 1 were as likely to complete the intended 6 cycles of induction and achieve a response as those without these characteristics. Furthermore, no differences in progression-free survival were observed based on any of these characteristics, suggesting that clinical benefit of PCR for these patients was also independent of these features. Because the PCR regimen is administered on Day 1 of every 3-week cycle, this regimen also requires fewer office visits than other chemoimmunotherapy regimens (8 visits over 6-cycle period for PCR compared with 19 visits for FCR6 and 30–34 visits for FR6, 12), which may improve the ease of administration to older patients.
There are several reasons why older individuals may tolerate the PCR regimen better than the FCR regimen. First, pentostatin may be less myelosuppressive than fludarabine, resulting in a lower frequency of grade 3–4 neutropenia and better tolerability.27, 28 Second, we used a 2 mg/m2 dose of pentostatin rather than the 4 mg/m2 dose used in some other studies.27 The optimal dose of pentostatin for treating chronic lymphocytic leukemia is unknown, and it is uncertain whether increasing the dose from 2 mg/m2 to 4 mg/m2 enhances efficacy or simply increases toxicity. Third, the dose of pentostatin needed to achieve an equivalent effect to the fludarabine dose in FCR is unknown. It is possible 1 dose of pentostatin at 2 mg/m2 every 3 weeks is not equivalent to 3 doses of 25 mg/m2 of fludarabine every 4 weeks and that the enhanced tolerability is due to a lower total purine nucleoside analog dose. Finally, growth factor support was used routinely for all patients treated in our trial of PCR,30 whereas it was used at the treating physician's discretion in the published trial of FCR.6
Whereas the PCR regimen appears to be a significant step forward for older patients with chronic lymphocytic leukemia, there are some limitations to our analyses. First, this was a phase 2 study with a relatively small number of patients. The study was not statistically powered to detect small differences in efficacy and toxicity between the subgroups studied, and these could be more apparent in future, larger studies. Second, we used the calculated CrCl to estimate renal function. Although this is a standardized, widely used approach and is the approach being used to determine eligibility in ongoing phase 3 trials of chemoimmunotherapy for chronic lymphocytic leukemia, it is not as accurate as the measured glomerular filtration rate as assessed by inulin or iothalamate clearance.31 Third, our study does not provide data on the use of PCR in patients with more severe renal failure (creatinine > 1.5 × upper limit normal) in whom dose reductions for purine nucleoside analogs are required for safe treatment. Fourth, although trial eligibility criteria permitted enrollment of patients with an ECOG PS 0–3, few study participants had a PS ≥ 2, which limits our ability to analyze the tolerability of PCR in that patient subgroup. Finally, although this trial was conducted at 2 medical centers that serve large local and regional patient populations, they are also tertiary referral centers which could influence the applicability of these results to community practice.
Nevertheless, the PCR regimen seems to be very well tolerated among patients ages ≥70 years and those with decreased renal function. There remains a significant need for development of efficacious treatment regimens that can be tolerated by older patients with chronic lymphocytic leukemia. Clinical trials evaluating tolerability and the optimal duration of therapy in addition to efficacy are needed for patients with chronic lymphocytic leukemia, and such trials should actively seek to enroll older patients so that their results are applicable to the majority of chronic lymphocytic leukemia patients in need of therapy worldwide.
- 5The American College of Surgeons Commission on Cancer and the American Cancer Society. The National Cancer Data Base report on age, gender, treatment, and outcomes of patients with chronic lymphocytic leukemia. Cancer. 1999; 86: 2684–2692., , .
- 12Randomized phase 2 study of fludarabine with concurrent versus sequential treatment with rituximab in symptomatic, untreated patients with B-cell chronic lymphocytic leukemia: results from Cancer and Leukemia Group B 9712 (CALGB 9712). Blood. 2003; 101: 6–14., , , et al.
- 13The addition of rituximab to fludarabine may prolong progression-free survival and overall survival in patients with previously untreated chronic lymphocytic leukemia: an updated retrospective comparative analysis of CALGB 9712 and CALGB 9011. Blood. 2005; 105: 49–53., , , et al.
- 14Treatment of patients with CLL 70 years old and older: a single center experience of 142 patients. Leuk Lymphoma. 2005; 46: S86., , , et al.
- 26Cladribine alone and in combination with cyclophosphamide or cyclophosphamide plus mitoxantrone in the treatment of progressive chronic lymphocytic leukemia: report of a prospective, multicenter, randomized trial of the Polish Adult Leukemia Group (PALG CLL2). Blood. 2006; 108: 473–479., , , et al.
- 31National Kidney Foundation. K/DOQI clinical practice guidelines for chronic kidney disease: evaluation, classification, and stratification. Amer J Kidney Diseases. 2002; 39(2 Suppl 1 ): S1–266.
- 32National Cancer Institute: common toxicity criteria. Bethesda, Md: National Cancer Institute; 1988, 1988.