Impact of frontline fludarabine and cyclophosphamide combined treatment on peripheral blood stem cell mobilization in B-cell chronic lymphocytic leukemia

Olivier Tournilhac, Bruno Cazin, Stéphane Leprètre, Marine Diviné, Karim Maloum, Alain Delmer, Bernard Grosbois, Pierre Feugier, Frederic Maloisel, Florence Villard, Bruno Villemagne, Dominique Bastit, Karim Belhadj, Nabi Azar, Mauricette Michallet, Geraud Manhès and Philippe Travade


Ongoing studies in B-cell chronic lymphocytic leukemia are evaluating autologous peripheral blood stem cell (PBSC) transplantation in first remission following fludarabine therapy. However, fludarabine could impair PBSC harvest. In 38 patients after frontline oral fludarabine and cyclophosphamide (FDR-CY) therapy, we prospectively evaluated steady state filgrastim- or lenograstim-primed PBSC mobilization to collect 2.0 × 106/kg or more CD34 cells. The first mobilization, performed a median of 178 days (range, 69-377 days) from the last FDR-CY course, was unsuccessful in 32 patients. This result was significantly associated with a low platelet count before mobilization but not with age, interval from last FDR-CY course, initial stage, remission status, or other blood parameters. Finally, after 1, 2, and 3 mobilizations in 27, 10, and 1 patients, 2.0 × 106/kg or more CD34 cells were collected in only 12. Explorations of the mechanism of poor mobilization and adaptation of PBSC harvest policies after fludarabine treatment are therefore warranted.


In B-cell chronic lymphocytic leukemia (B-CLL), treatment must be instituted for advanced (ie, B and C) stage patients.1 Because the duration of response is correlated to the response rate,2,3 autologous stem cell transplantation has been tested in the disease2,4-6 with encouraging results. Therefore, ongoing cooperative studies are evaluating autologous peripheral blood stem cell (PBSC) transplantation at the time of first remission. To obtain the best response before PBSC harvest, fludarabine (FDR) is considered the most efficient drug in the treatment of B-CLL and is increasingly used as frontline therapy.7,8 An oral formulation of this drug is available.9 The adjunction of cyclophosphamide10,11 (CY) or rituximab and CY12 with FDR appears still more efficient and could lead to molecular remissions.

However, it has been suggested that occasionally in B-CLL13-15 and in other disorders,16-19 FDR could adversely affect PBSC mobilization. Conversely, high CD34 cell yields can be obtained in B-CLL patients treated with non-FDR–containing regimens.20 This controversial issue has never been prospectively evaluated in a large number of B-CLL patients in first remission. The aim of this prospective study was to evaluate steady state filgrastim- or lenograstim-primed PBSC mobilization in 38 B-CLL patients in first remission after oral frontline treatment with FDR-CY.

Study design


The study involved 38 patients with B-CLL (median age, 53.5 years [range, 38-66 years]; 33 men, 5 women) in complete remission (CR) (n = 26) or partial remission (PR) (n = 12), according to the National Cancer Institute criteria,21 after frontline treatment with FDR-CY. All gave informed consent. They had been previously enrolled in a clinical trial (ME 98123 Schering S.A.) that evaluated the efficacy and safety of 6 monthly courses of FDR (30 mg/m2 per day) and CY (200 mg/m2 per day) both given orally for 5 consecutive days.11 All these patients (34 stage B and 4 stage C) had received 6 FDR-CY courses, except 4 who had received either 5 courses (n = 2), 4 courses (n = 1), or 2 courses plus 3 courses of a CHOP-like regimen (n = 1) because of toxicity. After evaluation performed 2 months after the last FDR-CY course, responding patients were considered for PBPC collection at the time of first CR or PR. According to the aforementioned trial, high-dose therapy should not be performed before relapse.

Mobilization and apheresis

Mobilization was initiated at least 2 months after the last FDR-CY course, according to published results.13 To collect a number of CD34 cells equaling at least 2.0 × 106/kg body weight, all patients underwent a first steady state PBSC mobilization using either filgrastim (10 μg/kg per day) or lenograstim (7 μg/kg per day) given once a day for 4 to 6 consecutive days until adequate blood CD34 circulation was achieved. Apheresis was initiated when circulating CD34 cell levels reached 10 × 106/L. Two to 3 blood mass volume was processed during each apheresis performed with continuous-flow blood cell separators. If the first cell collection failed, further steady state mobilizations using the same procedure could be performed.

Statistical analysis

To avoid redundant patient-dependent variables caused by repeated mobilization courses in the same patient, we focused statistical analysis on the results of their first mobilization course only. Univariate analysis was performed using the χ2 test to compare categorical variables and by analysis of variance to compare means. Multivariate logistic regression analysis was performed taking the final result of the first mobilization (ie, number of collected CD34 cells) as dependent variables, and the variables found significant in the univariate analysis or described in the literature as independent variables. Data analysis and statistics analysis were performed using SAS software (SAS Institute). P < .05 was considered statistically significant.

Results and discussion

The 38 patients underwent a total of 50 mobilizations: 1, 2, or 3 mobilizations in 27, 10, and 1 patients, respectively (Table 1). The first mobilization was performed after a median of 178 days (range, 69-377 days) from the last FDR-CY course. Apheresis was initiated in 17 (45%) patients, giving a median number of 1.48 × 106/kg (range, 0.47-3.45) CD34 cells. Only 6 (16%) patients achieved a CD34 cell number that equaled or exceeded 2.0 × 106/kg. In 21 (55%) patients, apheresis was not performed because of a poor blood CD34 level (less than 10 × 106/L).

Table 1.

Steady-state filgrastim- or lenograstim-primed PBSC mobilization to collect 2.0 × 106/kg or more CD34 cells

Univariate analysis showed a significant association between number of CD34 cells collected and platelet count immediately before mobilization (P = .009), with a trend in correlation with platelet count at evaluation 2 months after the last FDR-CY (P = .051) but no correlation with other studied parameters (Table 2). Therefore, all patients with platelet counts lower than 150.0 × 109/L, either at evaluation or before mobilization, failed to mobilize adequately. In addition, achieving at least 10 × 106/L circulating CD34 cells was associated with platelet counts either before mobilization (P = .009) or at evaluation (P = .010) (data not shown). Multivariate analysis using age, interval time from last FDR-CY course to mobilization, initial Binet stage, disease status at evaluation, hemoglobin level, and platelet count before mobilization persistently found a significant association between number of collected CD34 cells and platelet count (P = .04).

Table 2.

Univariate analysis to predict number of collected CD34 cells equaling or exceeding 2 × 106/kg after first mobilization

A second mobilization was made in 11 patients after a median of 288 days (range, 243-517 days) from the last FDR-CY course and of 140 days (range, 68-285 days) from the first mobilization. Apheresis was initiated in 8 of these patients, giving a median number of CD34 cells of 1.44 × 106/kg (range, 0.55-2.56). Only 5 patients achieved a CD34 cell number that equaled or was greater than 2 × 106/kg—2 with this second mobilization only, and 3 achieved it by cumulative results from the first and second mobilizations. In addition, though they already had 2.0 × 106/kg or more CD34 cells stored after the first procedure, 2 other patients had second and successful mobilizations. A third mobilization was performed in 1 patient after 573 days from the last FDR-CY course and 300 days from the second mobilization, leading to a total of 2.25 × 106/kg CD34 cells harvested.

The combined results of all 3 mobilizations are detailed in Table 3. These poor results were probably not caused by bone marrow involvement16 by tumor cells at the time of mobilization because most of these patients were in CR (even when assessed by bone marrow biopsy), nor were they caused by multiple previous treatments given that FDR-CY was their first therapeutic line. As previously reported,13-19 a negative impact of FDR on further PBSC mobilization is therefore clearly suspected, but the precise mechanism of such toxicity remains to be clarified. Jeopardized stem cell mobilization in B-CLL could depend on alterations of bone marrow stroma. The chemokine receptor CXCR4 for the stromal factor SDF-1, which is highly and functionally expressed by B-CLL cells and acts in bone marrow homing and cell survival,22 could be one of the involved mechanisms.

Table 3.

Combined results of first, second, and third mobilizations

Prospective studies in B-CLL are currently evaluating the impact of transplantation after frontline therapy using FDR-containing regimens. Based on this small but homogeneous series of treated B-CLL patients, these results could lead to the adaptation of PBSC harvest policies. Delaying PBSC mobilization until after FDR treatment is complete does not seem to improve the results. Limiting the number of FDR courses to the minimum needed to achieve the maximal response before mobilization and using combined mobilization with chemotherapy and G-CSF could be alternatives.


We thank Marisha Burden (University of Oklahoma College of Medicine) for technical help in manuscript preparation and Dr Odile Guibon (Laboratoire Schering, Lys-lez-Lannoy).


  • Reprints:
    Philippe Travade, Service d'Hématologie Clinique, Hôtel Dieu, Centre Hospitalier Universitaire - 11, bd Léon Malfreyt, 63003 Clermont-Ferrand Cedex, BP 69, France; e-mail: ptravade{at}
  • Prepublished online as Blood First Edition Paper, September 11, 2003; DOI 10.1182/blood-2003-05-1449.

  • The publication costs of this article were defrayed in part by page charge payment. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 U.S.C. section 1734.

  • Submitted May 9, 2003.
  • Accepted August 29, 2003.


View Abstract