Cost–effectiveness of repeated aphereses in poor mobilizers undergoing high-dose chemotherapy and autologous hematopoietic cell transplantation


We retrospectively analyzed a cohort of 435 patients who – from 1992 to 1998 – received high-dose chemotherapy and autologous blood cell transplantation for a variety of nonleukemic malignancies at a single institution. Autologous cells and progenitors were mobilized in the blood, with daily administration of rhG-CSF (Filgrastim, Amgen, Thousand Oaks, CA, USA): 600 μg per day SQ, or after administration of various chemotherapy regimens, followed by the daily administration of rhG-CSF, 300 μg per day SQ, starting on day 6, as per institutional protocols. Patients were collected by apheresis: two blood masses were treated, using a Cobe Spectra (Gambro BCT) automated processor; autologous grafts were characterized for their content in CD34+ cells, using murine monoclonal antibodies directed to the CD34 and CD45 antigens, and flow cytometry, according to institutional protocols.

In this cohort of patients, 130 individuals were identified as ‘poor mobilizers’ because less than 3 × 106 CD34+ cells/kg were collected during the first attempt at mobilization and collection (three aphereses). As expected, patients were older in the ‘poor-mobilizer’ group than in the ‘good-mobilizer’ group: 47.5 years [15–67] vs 44 years [15–65], P=0.007; more patients had multiple myeloma or Hodgkin's disease in the ‘poor-mobilizer’ group (12 and 35 % vs 9 and 20%, respectively, P<0.001); more patients had bone marrow involvement in the ‘poor-mobilizer’ group (49 vs 34%, P=0.003). On average, patients in the ‘poor-mobilizer’ group also had received higher numbers of chemotherapy treatments before undergoing mobilization and collection: 7 [3–22] vs 6 [2–30], P=0.01. The proportion of patients mobilized with rhG-CSF alone was higher in the ‘poor-mobilizer’ group: 83 vs 49%, P<0.001.

Eventually, a proportion of ‘poor mobilizers’ underwent successive attempts at collection: 83 out of 130 individuals were submitted to a second, and potentially a third attempt at mobilization and aphereses, based on the decision of the referring physician. Out of these 83 patients, 59 (71.1%) eventually reached the threshold value of 3 × 106 CD34+ cells/kg for the cumulative number of collected blood progenitors. Analysis of patient characteristics shows that these 59 subjects differ from the 47 patients for whom a second attempt at mobilization/collection was not performed: they are older (52 years vs 43 years, P=0.003), include more men (66 vs 38%, P=0.004), have a different distribution of diagnoses, including more multiple myeloma (51 vs 13%, P<0.001), and include more patients with progressive diseases 24 vs 11%, P=0.04); all these ‘unfavorable’ factors may explain the willingness of transplant physicians to proceed with additional cycles of mobilization/aphereses, in order to obtain higher numbers of autologous progenitors, and perform autologous transplantation in safer conditions. A total of 24 patients experienced successive failures (ie the cumulative number of collected blood CD34+ cells did not reach 3 × 106 CD34+ cells/kg), and nevertheless underwent high-dose chemotherapy and auto-logous transplantation: the economics for this subset are analyzed separately, as these individuals cumulated high costs, both for the collection phase and for the transplantation phase of the treatment.

Clinical efficacy, as well as cost–effectiveness were evaluated for patients who were eventually reinfused with less than 3 × 106 CD34+ cells/kg obtained in a single attempt at collection (n=47), and for patients who were eventually re-infused with more than 3 × 106 CD34+ cells/kg, following more than one attempt at collection (n=59). Engraftment (and effectiveness) was evaluated as the first of two consecutive days during which the ANC rose above 0.5 × 109/l for neutrophils, and above 25 × 109/l for platelets. Direct medical costs were evaluated for each step of the procedure: mobilization, collection, cryopreservation, reinfusion, and immediate follow-up, from admission to discharge. Costs were measured in physical quantities, obtained from detailed observation of each patient medical records: number of mobilizations, number of leukaphereses, quantity of administered rhG-CSF, length of hospital stay, length of IV antibiotic administration, number of transfusions, conditioning regimen, and daily laboratory tests. Monetary values were attributed to each of these quantities on the basis of specific unit costs. A 1999 per diem room cost was computed and used for calculation, using a previously described methodology.1 rhG-CSF and antibiotics unit costs were the purchasing prices for the hospital. Transfusion unit costs were the official 1999 French prices (established each year through direct government regulations). Laboratory tests were evaluated using the prices of medical technical acts from the national nomenclature that is established and revised by the French public health-care provider (‘Sécurité Sociale’). Collection and cryopreservation costs were assessed in 1999 French francs on the basis of a previous study.2

There is a clinical benefit with a reduced time to platelet recovery, when one achieves collection of more than 3 × 106 CDC34+ cells during repeated attempts at mobilization/collection. Analysis of engraftment demonstrated no dose effect for neutrophil recovery, but showed the existence of a dose effect for platelet recovery: the ‘poor mobilizers’ who ultimately received more than 3 × 106 CD34+ cells/kg did recover platelets significantly earlier than patients who received less than 3 × 106 CD34+ cells/kg, with a difference of 4 days (14 days [13–15] vs 18 days [12–24], P<0.001), while the difference for neutrophils was only 1 day (12 days [12–14] vs 13 days [12–14], P>0.05); these observations cannot be explained by the use of post-transplant rhG-CSF, as a similar proportion of approximately 40% of the patients were administered this hematopoietic growth factor in vivo in both groups.

Computation of costs for these patients demonstrated that additional costs associated with repeated attempts at mobilization/collection were compensated by a decreased use of health-care resources during the high-dose chemotherapy with autologous blood cell transplantation phase of the procedure (Table 1); this is mostly because of a reduced number of platelet and RBC transfusions. Overall, the cost of the entire procedure appears to be similar for both groups. Taking into account the accelerated platelet recovery, it thus appears that this strategy is cost efficient.

Table 1 Costs for mobilization/collection and for high-dose chemotherapy/autologous transplantation, for patients obtaining <3 × 106/kg CD34+ in one attempt and more or less than 3 × 106/kg in more than one attempt.

We conclude that repeating attempts for mobilization and collection is legitimate in ‘poor-mobilizer’ patients who are candidates for high-dose chemotherapy and autologous transplantation. Support of high-dose chemotherapy with the reinfusion of autologous blood cells and progenitors is now routinely used for patients with advanced or poor-prognosis malignancies, including lymphoid malignancies and various types of carcinomas.3 The relation between numbers of collected (infused) cell and progenitor doses, usually assessed as the number of CD34+ cells/kg of body weight, and hematopoietic recovery has been established by many investigators,4 and bears several important implications, both medical and economical.1,5 Different strategies have been designed to obtain high numbers of cells and progenitors, including careful selection of candidate patients, optimization of mobilization regimens, optimization of collection procedures (extended or repeated aphereses), or attempts at expanding part or all of the autologous graft during culture with cytokine combinations.6 The existence of a ‘recommended minimal number’ of CD34+ cells/kg of recipient weight – usually evaluated between 2 and 5 × 106 CD34+ cells/kg – is now widely accepted, and used to make daily decisions for patients who undergo CD34+ cell monitoring in their peripheral blood and aphereses. While clinical and economical benefits associated with the collection of higher doses of CD34+ cells have been evaluated in a number of studies,4,7,8 none of these focused attention on ‘poor mobilizers’, and searched for guide-lines on how to manage this situation. In addition, none of these studies includes a complete economical assessment of the entire procedure (mobilization, collection, and transplantation), and provides economic criteria to optimize the collection strategy. Our economic evaluation measures the total cost of the mobilization, collection, and transplant procedure, on a group of individuals who failed to mobilize adequately progenitors in the peripheral blood. Previously published studies offset the cost of the mobilization procedure,8 or inversely focused only on this specific cost. All costs have to be computed together, as we show that the increase in collection costs may be partially or totally compensated by a decrease in post-transplant costs. This is especially important in view of the development of outpatient post-transplant management programs that lead to a significant decrease in post-transplant costs, increasing the relative weight of the mobilization and collection steps in the entire procedure. Obviously, a retrospective analysis of a cohort of patients who were not included in a unique clinical trial introduces bias in the analysis. However, it more realistically reflects the daily clinical practice of high-dose chemotherapy with stem cell support, including a realistic evaluation of economic parameters. A limitation of our analysis is the choice of the hematological recovery as the effectiveness criteria. More studies are needed to assess the impact of CD34+ cell dose on nonhematopoietic endpoints, including survival and quality of life. Finally, while we provide evidence that ‘poor mobilizers’ may benefit from repeated attempts at mobilization – collection, the small subset of really ‘hard to mobilize’ patients (those not achieving 3 × 106/kg CD34+ cells, despite several attempts at mobilization) deserves further investigation. For these individuals, aphereses and marrow collections can be considered as alternative sources of progenitors for transplantation. They also represent an incentive to develop and evaluate alternative technologies to produce ex vivo large quantities of cells and progenitors for clinical use.6


  1. 1

    Hartmann O, Le Coroller AG, Blaise D, Michon J, Philip I, Norol F et al. Peripheral blood stem cells and bone marrow transplantation for solid tumors and lymphomas : hematologic recovery and costs. Ann Intern Med 1996; 126: 600–607.

    Article  Google Scholar 

  2. 2

    Le Corroller AG, Moatti JP, Chabannon C, Faucher C, Fortanier C, Ladaique P et al. Optimization of peripheral blood stem cell collection by leukapheresis: a case of interaction between economic and clinical assessment of an innovation. Int J Technol Assessment Health Care 1999; 15: 161–172.

    CAS  Article  Google Scholar 

  3. 3

    Gratwohl A, Passweg J, Baldomero H, Urbano-Ispizua A . Hematopoietic stem cell transplantation activity in Europe 1999. Bone Marrow Transplant 2001; 27: 899–916.

    CAS  Article  Google Scholar 

  4. 4

    Weaver C, Hazelton B, Birch R, Palmer P, Allen C, Schwartzberg L et al. An analysis of engraftment kinetics as a function of the CD34 content of peripheral blood progenitor cell collections in 692 patients after the administration of myeloablative chemotherapy. Blood 1995; 86: 3961–3969.

    CAS  PubMed  Google Scholar 

  5. 5

    Faucher C, Fortanier C, Viens P, Le Coroller AG, Chabannon C, Camerlo J et al. Clinical and economical comparison of lenograstim-primed blood cells (BC) and bone marrow (BM) allogeneic transplantation. Bone Marrow Transplant 1998; 21: S92-S98.

    Google Scholar 

  6. 6

    Pecora AL, Stiff P, LeMaistre CF, Bayer R, Bachier C, Goldberg SL et al. A phase II trial evaluating the safety and effectiveness of the AastromReplicell system for augmentation of low-dose blood stem cell transplantation. Bone Marrow Transplant 2001; 28: 295–303.

    CAS  Article  Google Scholar 

  7. 7

    Ketterer N, Salles G, Raba M, Espinousse D, Sonnet A, Tremisi P et al. High CD34+ cell counts decrease hematologic toxicity of autologous peripheral blood cell transplantation. Blood 1998; 91: 3148–3155.

    CAS  PubMed  Google Scholar 

  8. 8

    Schulman KA, Birch R, Zhen B, Pania Z, Weaver CH . Effect of CD34+ cell dose on ressource utilization in patients after high-dose chemotherapy with peripheral-blood stem-cell support. J Clin Oncol 1999; 17: 1227–1233.

    CAS  Article  Google Scholar 

Download references


This work was supported in part by Institut Paoli-Calmettes. The authors thank Mr Christian Cailliot at Amgen France for his support during the conduct of this study.

Author information



Corresponding author

Correspondence to C Chabannon.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Chabannon, C., Le Corroller, AG., Viret, F. et al. Cost–effectiveness of repeated aphereses in poor mobilizers undergoing high-dose chemotherapy and autologous hematopoietic cell transplantation. Leukemia 17, 811–813 (2003).

Download citation

Further reading


Quick links