Short- and long-term safety of the 2 x 106 CD34+ cells/kg threshold for hematopoietic reconstitution after high-dose chemotherapy and peripheral blood progenitor cell support

C. Corsini, F. Bertolini, P. Mancuso, S. Cinieri, F. Peccatori and G. Martinelli

Division of Hematology–Oncology, European Institute of Oncology, Milan, Italy (E-mail: francesco.bertolini@ieo.it)

High-dose chemotherapy (HDCT) followed by reinfusion of autologous peripheral blood progenitor cells (PBPC) is widely used in the treatment of cancer patients. However, controversy still remains around the issue of the adequate number of PBPC to be reinfused in order to obtain a prompt and stable hematopoietic rescue. According to recent studies and reviews [13], the recommended dose for autologous transplantation is 5–8 x 106 CD34+ cells/kg. This recommendation is mostly based upon faster platelet (rather than neutrophil) recovery with increasing CD34+ cell dose, and the fact that reconstitution of durable hematopoiesis capable of coping with further chemotherapy and/or radiation therapy, infections or surgery occurred in one single 1994 study on 34 cancer patients who received 8 x 106 CD34+ cells/kg after HDCT [4]. Here, we report our experiences of 433 patients who underwent a total of 1019 HDCT courses followed by reinfusion of PBPC collections containing a median of 2.4 x 106/CD34+ cells/kg. Median patient follow-up was 34 months.

Of the 433 patients evaluated, 281 had high-risk, stage II–III breast cancer (BC), 14 had ovarian cancer (OC), 22 had other solid tumors (ST), 93 had non-Hodgkin’s lymphoma (NHL), 18 had multiple myeloma (MM) and five had Hodgkin’s disease (HD). PBPC were mobilized by granulocyte colony-stimulating factor (G-CSF) alone in BC patients. ST, NHL, MM and HD patients were mobilized by cyclophosphamide 4–7 g/m2 plus G-CSF. OC patients were mobilized by epirubicin and docetaxel plus G-CSF. CD34+ cells were enumerated by flow cytometry using an FDA-approved, single plat-form kit (Procount; Becton Dickinson, Mountain View, CA). Results of intra- and interlaboratory controls (provided by Becton Dickinson and the European Working Group on Clinical Cell Analysis) [5] always fell within appropriate limits.

In all patients, the collection target was at least 2 x 106/CD34+ cells/kg for every planned HDCT course. BC patients underwent three courses of EC (cyclophosphamide and epirubicin; 49% of patients), T-EC (EC plus docetaxel; 21% of patients), ICE (ifosfamide, carboplatin and etoposide; 15% of patients) or T-ICE (ICE plus docetaxel; 15% of patients). OC and ST patients received three courses of ICE. NHL and HD patients received high-dose sequential therapy [4], and MM patients received a first course of melphalan and a second course of melphalan plus idarubicin. After HDCT, all patients received 5 µg/kg/day G-CSF until their neutrophil count exceeded 1000/µl.

Table 1 shows PBPC collection data and hematopoietic recoveries. A statistical comparison of G-CSF alone compared with chemotherapy plus G-CSF (Ch-G) data would be misleading for two reasons. First, CD34+ cell collection targets were different. Secondly, 39% of patients mobilized by G-CSF alone were previously treated with neo-adjuvant chemotherapy, whereas 87% of patients mobilized with Ch-G received multiple previous chemotherapy courses (P <0.01 by chi-square test). The 2 x 106/CD34+ cells/kg per planned HDCT course target was not reached in 52 out of 281 (18%) and 12 out of 204 (5%) patients mobilized by G-CSF alone and Ch-G, respectively (P <0.01). Among 204 patients mobilized by Ch-G, 52 were at the second mobilization attempt after mobilization with G-CSF alone. The frequency of patients mobilized by G-CSF alone who failed to collect 2 x 106/CD34+ cells/kg per planned HDCT course was not significantly increased among previously treated patients (19% compared with 17% in untreated patients). Following a crossover approach, patients who failed mobilization with G-CSF alone were mobilized by Ch-G, and 43 out of 52 (82%) collected >2 x 106/CD34+ cells/kg per planned HDCT course. On the other hand, 10 patients who failed Ch-G mobilization were mobilized by G-CSF alone, and 4 out of 10 collected >2 x 106/CD34+ cells/kg (P = 0.013).


View this table:
[in this window]
[in a new window]
 
Table 1. PBPC collections and haematopoietic recoveries
 
Overall, the median number of CD34+ cells reinfused per HDCT course was 2.4 x 106/kg (range 0.7–31.7), and prompt white cell and platelet engraftment was observed in the large majority of patients. The median number of red cell and platelet transfusions was 2 (range 0–6) and 1 (range 0–19), respectively, i.e. values similar to those reported previously in other programmes based upon single- or multiple-HDCT courses [14]. A delay in white cell engraftment (i.e. 17–30 days to reach 1000 neutrophils/µl) was observed in none of 281 recipients of PBPC mobilized by G-CSF alone and in 14 of 204 (6.8%) recipients of PBPC mobilized by Ch-G. Of these 14 patients, neutrophil engraftment took 17–20 days in 11 patients, 21–22 days in two patients (one with NHL and one with MM) and 30 days in the remaining patient with NHL. In this group of patients with delayed neutrophil engraftment, the median number of CD34+ cells per kilogram was 3.0 x 106 (range 1.3–23.2; P = 0.45, not significant compared with patients with faster engraftment). A delayed platelet engraftment (i.e. 17–40 days) was observed in one out of 281 recipients of PBPC mobilized by G-CSF alone (this BC patient engrafted in 21 days) and in 14 out of 204 (6.8%) recipients of PBPC mobilized by Ch-G. Of these 14 patients, platelet engraftment was 17–20 days in nine patients, 21–23 days in three patients (one with MM and two with NHL), and 24–40 days in the remaining two patients (both with NHL). In this group of patients with delayed platelet engraftment, the median number of CD34+ cells per kilogram was 3.3 x 106 (range 1.6–23.2; P = 0.56, not significant compared with patients with faster engraftment). Using multivariate Cox’s analysis, more than two previous chemotherapy courses (P = 0.01) but not CD34+ cells were associated with a delayed neutrophil or platelet engraftment. CD34+ cell collections below median values or of the lower quartiles were not associated with disease progression or reduced survival.

Regarding the durability of long-term hematopoiesis, after a median follow-up of 34 months (range 9–57), irreversible grade 2–4 cytopenia was not observed despite further radiotherapy administered to 67 and 21% of BC and NHL patients, respectively, and further chemotherapy administered to 33 and 48% of BC and NHL patients, respectively. No myelodysplastic syndromes (MDS) were observed. Two BC patients developed secondary acute leukemia (one AML FAB M4 and one pro-B ALL).

To the best of our knowledge, this is the largest series of data about autologous PBPC mobilization, single platform CD34+ cell enumeration, and HDCT and PBPC reinfusion reported from a single institution. Our data indicate that in a HDCT program, a PBPC collection policy based upon standardized, single platform CD34+ cell evaluation and a 2 x 106/CD34+ cells/kg threshold is associated with prompt hematopoietic recovery in >99% of patients mobilized with G-CSF. In more intensively pretreated patients mobilized with Ch-G, a prompt recovery was observed in 14 out of 179 patients (92%), whereas hematopoietic recovery was delayed for 1 or 2 and 3 or 4 weeks in the remaining 6 and 1%, respectively. Regarding PBPC mobilization failures, in the present study, 18% of patients mobilized by G-CSF alone failed to collect at least 6 x 106/CD34+ cells/kg in order to receive three HDCT courses. In 82% of these patients, a crossover Ch-G mobilization was effective. In previously treated patients who failed to mobilize after Ch-G, the crossover to mobilization with G-CSF alone was significantly less effective. Finally, it should be noted that all patients had a sustained long-term engraftment despite frequent boost radiotherapy and/or secondary chemotherapy, and the incidence of MDS and/or secondary hematopoietic malignancies was better or similar when compared with results reported in other studies where the PBPC collection threshold was two- to four-fold higher [14].

C. Corsini, F. Bertolini, P. Mancuso, S. Cinieri, F. Peccatori & G. Martinelli

Division of Hematology–Oncology, European Institute of Oncology, Milan, Italy (E-mail: francesco.bertolini@ieo.it)

References

1. Siena S, Schiavo R, Pedrazzoli P et al. Therapeutic relevance of CD34+ cell dose in blood cell transplantation for cancer therapy. J Clin Oncol 2000; 18: 1360–1377.[Abstract/Free Full Text]

2. Demirer T, Bensinger WI, Buckner CD. Peripheral blood stem cell mobilization for high-dose chemotherapy. J Hematother 1999; 8: 103–113.[ISI][Medline]

3. To LB, Haylock DN, Simmons PJ et al. The biology and clinical uses of blood stem cells. Blood 1997; 89: 2233–2258.

4. Siena S, Bregni M, Di Nicola M et al. Durability of hematopoiesis following autografting with peripheral blood hematopoietic progenitors. Ann Oncol 1994; 5: 935–941.[Abstract]

5. Barnett D, Granger V, Kraan J et al. Reduction of intra- and interlaboratory variation in CD34+ stem cell enumeration using stable test material, standard protocols and targeted training. CD34 Task Force of the European Working Group of Clinical Cell Analysis (EWGCCA). Br J Haematol 2000; 108: 784–792.[ISI][Medline]