Comparable Leukemic Cells Reduction but Improved Normal Hematopoietic Regeneration with G-CSF Priming Plus Low Dose Chemotherapy Regimen Relative to Standard Chemotherapy in AML Mice Model
Ya Zhang,Xiaomin Feng,Aili Chen,Shaoyan Hu,Qianfei Wang,Gang Huang
DOI: https://doi.org/10.1182/blood-2019-129465
IF: 20.3
2019-01-01
Blood
Abstract:Pediatric acute myeloid leukemia (pAML) is a heterogeneous disease. The clinical outcome of pAML has improved significantly over the past four decades with current long-term survival rates of ~70%. This improvement is due to intensification of chemotherapeutic regimens, better risk-group stratifications, better salvage at relapse, and improved supportive care. Nevertheless, novel therapies are still needed as the cure rates for some subtypes of pAML remain unacceptably low. In addition, the profound marrow suppression caused by near-myeloablative induction therapy may lead to life-threatening complications, such as severe infections, which not only prolong the hospitalization but also cause significant morbidity and mortality during or after treatment. Moreover, intensive chemotherapy usually affects the patient's quality of life and they can experience low physical, emotional, or social function. To determine the impact of reducing the intensity of induction chemotherapy on the outcome of pAML, we recently performed a study using G-CSF priming plus low-dose chemotherapy (GLDC) to treat pAML, compared with standard chemotherapy (SDC) during induction. Our study showed that complete remission (CR), 4-year event free survival (EFS), overall survival (OS), and cumulative incidence of relapse (CIR) rates of 46 pAML patients, initially given a GLDC for the first induction remission, were comparable to those of 94 children admitted to the same hospital and given SDC. However, GLDC showed a much lower toxicity than SDC. After the first course of induction, the recovery of white blood cell (WBC) and platelet counts were significantly faster in patients receiving GLDC than SDC. Grade III or IV infectious complications occurred in 4 (8.7%) patients receiving GLDC and 23 (24.4%) patients receiving SDC, and 2 patients receiving SDC died of complications (Hu Y. et al, 2019). Even though the clinical efficacy of GLDC has been seen in multiple clinical trials, the detailed mechanisms remain unclear. Some clinical and pre-clinical studies showed that G-CSF could push AML cells into cell-cycle, accelerate WBC and platelet recovery even when used after induction chemotherapy, and reduce the viability of AML cells when co-cultured with BM stroma, but not cultured alone. However, all of these studies were not established with LDC. To determine the potential mechanism(s) of GLDC in pAML, we established Mll-Af9 transplanted chimeric mice to develop AML, and we assessed the clearance of leukemic cells and the impact on normal cells by GLDC or SDC. Based on previous studies, we used combined cytarabine (100mg/kg) and doxorubicin (3mg/kg) (DA) in mice as SDC, half dose of SDC as LDC, and G-CSF (300ug/kg) plus half dose of SDC as GLDC in our study design. The G-CSF priming times were inconsistent in previous studies. To establish the best priming effect, especially for the stem cells, we first analyzed the cell cycle stimulation by G-CSF to the SLAM hematopoietic stem cells (SLAM-HSCs) and the more quiescent stem-like HSC population (endothelial protein C receptor expressing, EPCR+ HSCs) in WT mice. We found that G-CSF can stimulate SLAM-HSCs and EPCR+ HSCs into cell-cycle as early as 12hrs and 24hrs post G-CSF injection, so we used G-CSF one day priming before the chemotherapy estimating that even the deeper quiescent stem cells have been stimulated into the cell-cycle at that time point. We then arranged the Mll-Af9 transplanted mice with similar disease burden into 5 groups to receive different treatments for 6 days: LDC, GLDC, SDC, and also the G-CSF or the PBS groups as control (G-CSF was given on D1-D6, DA was given on D2-D6). We found that GLDC and SDC had similar clearance to the leukemic cells (CD45.2+), which were better than LDC, as observed 2 days after the treatment. However, the recovery of normal cells (CD45.1+) was significantly faster in the GLDC group compared with LDC and SDC groups. We also found that G-CSF could simulate the differentiation of both leukemic cells and normal cells, shown by the increase of the CD11b+Ly6G+ population of both CD45.2+ and CD45.1+ cells in the G-CSF group compared to the PBS group. Studies to further elucidate the additional mechanisms of GLDC are still ongoing. In conclusion, our clinical and mouse modeling studies consistently suggest that GLDC promotes leukemia cells clearance and normal hematopoietic regeneration simultaneously, which may be a better regimen for pAML. Disclosures No relevant conflicts of interest to declare.