Poor Stem Cell Mobilization: incidence and Risk Factors

High dose chemotherapy followed by autologous stem cell transplant (ASCT) is an effective treatment for many patients with haematological malignancies. Adequate stem cell mobilization (SCM) and collection are essential for a successful ASCT. Unfortunately, a small group of patients fail to mobilize sufficient stem cell for transplant. In order to explore factors influencing SCM we conducted a retrospective analysis on a large series of candidates for ASCT in a Portuguese transplantation center.

We analyzed 233 patients who underwent consecutively SCM from 2007 to 2012. Patients with CD34+ <10/μL in peripheral blood at maximum stimulation were considered non-mobilizers (NM) and patients that didn’t harvest the minimal count of 2×10^6 CD34+/kg at maximum number of aphaeresis cycles were considered poor mobilizers (PM). Patient diagnosis, sex, age, body weight, previous radiotherapy, number of cycles of previous chemotherapy, status disease at mobilization and premobilization platelet count, prior use of fludarabine or alkylating agents were analyzed for correlation with mobilization.

Two patients received granulocyte colony-stimulating factor (G-CSF), all others received G-CSF plus chemotherapy (G+C) for the initial mobilization. 68 patients had non-Hodgkin’s lymphoma (NHL), 23 Hodgkin’s lymphoma (HL), 94 multiple myeloma (MM), 46 acute myeloid leukemia (AML) and 2 non-hematological malignancies (NHM). The median age was 54 years (min-max: 18-69) and 108 were females. A successful SCM was achieved in 217 patients (93.1%) and the NM ratio differed with respect to diagnosis (NHL: 11.7%, AML: 9.5%, HL: 9%, MM: 2.1%, NHM: 0%). At univariate analysis NM was significantly associated with the number of cycles of previous chemotherapy (p=0.003), status disease at mobilization (p=0.000) and prior use of fludarabine (p=0.013). Multivariate analysis confirmed the higher number of cycles of previous chemotherapy (p=0.002) and prior use of fludarabine (p=0.000) as negative predictive factors for SCM. For the 217 patients who underwent stem cell collection, 504aphaeresis were performed. Median number of aphaeresis necessary to harvest 2×10^6 CD34+/kg was 2 (min-max:1-6). Median CD34+/kg harvested was 4.6x10^6 (min-max:1.82-52.7), with 43.3% patients collecting 2x10^6 CD34+/kg in one aphaeresis. MM patients had the highest total CD34 cell yield (median 6.9x10^6 CD34+/kg, min-max: 2-52.7) and required less aphaeresis to collect 2×10^6 CD34+/kg. At univariate analysis diagnosis of MM (p=0.000), status disease at mobilization (p=0.007) and the number of cycles of previous chemotherapy (p=0.002) were significantly associated with PM. Multivariate analysis confirmed the diagnosis of MM (p=0.000) and the lower number of cycles of previous chemotherapy (p=0.002) as favorable predictive factors for the total CD34+ cell yield. Grade III or IV hematopoietic toxicity of chemotherapy had no significant effect on the SCM. Incidence of febrile neutropenia was similar in all patients without morbidity. There was no mortality associated with SCM.

G+C SCM appears to be safe in all patients without major toxicity or morbidity. Our results enhances the need for effective first-line mobilization agents in patients with higher number cycles of chemotherapy and use of fludarabine prior SCM; these patients probably would benefit for a early SCM in order to achieve a higher CD34+ cell yield.

No relevant conflicts of interest to declare.