Successful mobilization of hematopoietic peripheral blood progenitor cells with paclitaxel-based chemotherapy as initial or salvage regimen in patients with hematologic malignancies

Autologous hematopoietic progenitor cell transplantation is a standard of care in several hematologic diseases, but many patients are unable to mobilize a sufficient number of cells for transplantation. Paclitaxel is a plant alkaloid effective against ovarian and breast cancers, and has also been proven active in multiple myeloma and non-Hodgkin’s lymphoma, among other human neoplasms.1,2 We and others have described the efficacy of paclitaxel-based chemotherapy in mobilizing large amounts of hematopoietic progenitors in patients with ovarian or breast cancer.3–5 However, data on the use of paclitaxel and rhG-CSF for hematopoietic cell mobilization in patients with hematologic malignancies is scarce; only recently McKibbin et al. have described this schedule in 26 patients after failure of a prior mobilization regimen.6 To further determine the potential clinical utility of paclitaxel with rhG-CSF for hematopoietic progenitor mobilization in patients with non-solid tumors, we investigated: (i) the mobilizing ability and toxicity of this schedule as initial treatment, or as salvage therapy in patients who failed a mobilization attempt with rhG-CSF, and (ii) the efficacy and tolerability of cyclophosphamide (Cy), given in combination with paclitaxel and rhG-CSF after mobilization failure with filgrastim alone.

Table 1.Patients’ characteristics.

Between January 1999 and January 2008, 75 patients with a primary diagnosis of a hematologic malignancy who were scheduled for autologous transplant received paclitaxel in the mobilization schedule (Table 1). The time elapsed from the last treatment was at least three weeks. All patients gave informed consent.

Group A included 19 patients with risk factors for failure to achieve successful mobilization, representing 12% out of a total 156 first-line mobilizations with rhG-CSF during the same study period. Most patients displayed coexistence of various factors associated with poor mobilization success (Table 1). Patients received paclitaxel 170 mg/m i.v. by continuous infusion for 24 hours (day 1) followed by 8 μg/kg s.c rhG-CSF (P-G) daily until the last apheresis.4 Thirty-three patients received the same schedule as above as a second mobilization attempt (group B). Group C included 23 patients who were treated with a protocol containing paclitaxel as above, followed 24 hrs. later by Cy 4 g/m i.v. as a one hour infusion. On the third day, rhG-CSF was started s.c. at a dose of 8 μg/kg each day until the completion of leukapheresis (P-Cy-G).4 For patients in both B and C groups, paclitaxel containing schedules were administered after a first failed mobilization attempt with s.c rhG-CSF (10 μg/Kg), that had induced maximal CD34 cells in peripheral blood <7/μL (P-Cy-G) or 7–14/μL (P-G).

Table 2.Parameters of peripheral blood progenitor cell mobilization in 60 patients (acute leukemias excluded) treated with paclitaxel-rhG-CSF (P-G) either as first line therapy –group A-, or salvage therapy –group B-, or with paclitaxel-cyclophosphamide-rhG-CSF (P-Cy-G).

Peripheral blood counts, and CD34 cell concentrations were assessed on days 5, 7, and daily afterwards, and in each apheresis product. CD34 cell evaluations were performed as previously described.4 Leukapheresis procedures were initiated when total white blood cell count exceeded 5–10 × 10/L, or when peripheral blood CD34 cells were >15/μL. Daily leukaphereses were executed using a COBE Spectra (COBE BCT, Lakewood, CO, USA) blood cell separator, by processing 3 total blood volumes daily, until a target number of 2 × 10/CD34/kg recipient body weight was achieved.

The analysis of both paclitaxel containing schedules (P-G and P-Cy-G) showed that patients in the P-G group met the criteria to start leukapheresis earlier (median day 9, range 6–23) than those patients receiving P-Cy-G (median day 14, range 9–19; p=0.0001), resulting in higher rhG-CSF costs in the latter group (€1086 vs. €1520, p=0.0006).

When excluding the 15 patients diagnosed with acute leukemia, as it has been previously established that these patients mobilize poorly,7 and to avoid skewed data between groups, the ability to mobilize was similar in patients receiving P-G and P-Cy-G in terms of peak peripheral blood CD34 cells (50.0 vs. 49.5/μL respectively; p=0.4452), and yields of CD34 cells in the first apheresis (2.0 vs. 2.6 × 10/Kg respectively; p=0.2070). The use of paclitaxel resulted in 83% of patients achieving the minimum threshold number of CD34 >2 × 10/Kg (73% in group A, 91% in group B, and 81% in group C, p=0.7550). Thus, the use of paclitaxel-based regimens allowed successful mobilization in 86% of patients that had failed previous mobilization with rhG-CSF (Table 2).

The use of paclitaxel containing schemes was less efficient in the setting of acute leukemia, with an overall success rate of 44%. Age, disease status, time elapsed after last chemotherapy, and previous radiotherapy did not correlate with the mobilization outcome (p>0.05). The mobilization therapy was generally well tolerated, with grades III and IV neutropenia or thrombopenia significantly lower in P-G than in P-Cy-G patients (p=0.0002, and p=0.0121 respectively). Thus, the median duration of neutrophils <1 × 10/L was 1 day (range 0–3), and 6 days (range 4–12) respectively; p=0.0003, while that of platelet counts below 50 × 10/L was 0 days (range 0–3) and 4 days (range 0–16) respectively; p=0.0215. Only 12% of patients had clinical infection (8% vs. 22% in P-G vs. P-Cy-G respectively; p=0.0864), and no differences were observed between regimens in terms of fever (6% vs. 17%; p=0.1130), nor in number of patients requiring RBC, or platelet transfusion (p>0.05). No procedure-related deaths occurred.

A significant proportion of patients receiving standard mobilization for the purpose of autologous transplantation fail to mobilize bone marrow cells into the periphery. Currently, AMD3100 is the most promising mobilizing agent under investigation, but this CXCR4 antagonist is restricted to clinical studies. Additionally, pre-clinical and clinical studies show that acute myeloid leukemia, and chronic lymphocytic leukemia cells may be mobilized by AMD3100 via CXCR4 inhibition,8–10 thus limiting the use of this agent in patients with certain hematologic malignancies. Our findings show that paclitaxel is effective in mobilization of PBSC in patients with hematologic malignancies, not only as salvage therapy,6 but also in patients with adverse prognostic factors for mobilization. The overall mobilization success rate was 75%, being lower in acute leukemia (44%) than in other hematologic malignancies (83%). The addition of Cy to this regimen did not increase the collection yield, whereas it aggravated the injurious effect of chemotherapy on bone marrow. Since rhG-CSF in the first days of mobilization is likely to be not relevant, a delayed rhG-CSF administration might be considered as a strategy to reduce mobilization costs in P-Cy-G patients.11

In conclusion, our data show that mobilization of CD34 cells using paclitaxel is an effective, safe, and predictable strategy that allows efficient mobilization in patients with hematologic malignancies as first or second line priming schedule in poor mobilizers.