Objectives: In acute myeloid leukemia (AML), leukemia cells invade the bone marrow, disrupting hematopoiesis and reducing blood cell counts. Recombinant human thrombopoietin (rhTPO) is widely used in China to enhance platelet recovery post-chemotherapy. This study aims to evaluate the therapeutic advantages and safety of rhTPO in treating post-chemotherapy thrombocytopenia in AML using real-life clinical data. Additionally, a nomogram model was constructed to predict platelet recovery in AML patients following chemotherapy, based on basic clinical data.
Methods:(1) To investigate the clinical benefits and safety of rhTPO, a retrospective analysis was conducted on 347 AML patients (excluding M3) who received their first induction chemotherapy at our hospital from 2016 to May 2022. They were divided into an experimental group and a control group based on whether they received rhTPO treatment. The platelet counts' recovery level, recovery rate, and the amount of platelet transfusions after induction chemotherapy were compared and analyzed between the two groups. To assess the impact of rhTPO on chemotherapy and survival, the remission rate after induction chemotherapy and patient survival status, including overall survival (OS) and progression-free survival (PFS), were recorded. The incidence rate of adverse reactions was collected to evaluate the safety profile of rhTPO.
(2) In order to establish a nomogram model for predicting the platelet recovery rate after chemotherapy in AML patients, clinical data such as gender, age, disease subtype, ECOG score, KPS score, treatment regimen, initial platelet level before chemotherapy, and other routine blood test indicators were collected for statistical analysis. Platelet recovery rate was used as the outcome variable, and patient clinical information as independent variables. A univariate Logistics regression was initially employed for variable selection, with variables having a P-value <0.1 progressing to multivariate analysis; these selected variables were then included in a multivariate Logistics regression model. Subsequently, the model's discriminative ability, calibration, and clinical utility were tested.
Results:(1) Compared to the control group, the experimental group treated with rhTPO exhibited higher platelet recovery levels (99.0×10^9/L vs. 50.0×10^9/L), and this difference was statistically significant (P<0.001). With the exception of the duration of BPC<10×10^9/L, which did not differ significantly between the experimental and control groups, the experimental group showed a shorter duration for BPC<20×10^9/L and time required for platelet count recovery to BPC>30, 50, 80, and 100×10^9/L thresholds, with statistically significant differences. There were no significant differences in platelet transfusion amounts, chemotherapy remission rates, overall survival, progression-free survival, and adverse reaction rates between the experimental group and the control group.
(2) Multivariate Logistics regression analysis indicated that the use of rhTPO, Hb, MPV, initial platelet level, and ASXL1 gene mutation are independent risk factors affecting the rate of platelet recovery. The Area Under Curve (AUC) for the model was 0.789 (95%CI: 0.723-0.846) in the training set and 0.759 (95%CI: 0.67-0.848) in the validation set; Hosmer-Lemeshow test showed a good fit with P=0.978 and P=0.144 respectively.
Conclusion:(1) rhTPO can significantly enhance the level of platelet recovery, making it appropriate for clinical applications to elevate patients' platelet levels in preparation for the next stage of chemotherapy. rhTPO is able to shorten the recovery time for platelet counts to reach 20, 30, 50, 80, 100×10^9/L thresholds, but it remains uncertain as to its impact on platelet transfusion amounts. Its safety profile is positive, with no severe adverse reactions or detrimental effects on chemotherapy and survival observed.
(2) Multivariate Logistics regression has identified the use of rhTPO, Hb, MPV, and baseline platelet levels as favorable factors for platelet recovery, while ASXL1 gene mutation is a negative factor.
(3) This study is the first to construct a predictive model on the rate of platelet recovery, demonstrating good discriminative ability and calibration. Decision curve analysis suggests patients can benefit from this model.
No relevant conflicts of interest to declare.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal