Page 183 - 2021_04-Haematologica-web

LH suppression and hematopoietic recovery
Methods
Patient selection
We screened adult female patients younger than 55 years at The University of Texas MD Anderson Cancer Center (Houston, TX) with newly diagnosed acute myeloid leukemia (AML) or acute lymphoblastic leukemia (ALL), treated with intensive chemother- apy between January 2000 and December 2017. We identified those who received at least one leuprolide injection for prevention or treatment of abnormal uterine bleeding with intensive chemotherapy (leuprolide received between day -7 and day 90 of induction chemotherapy start). The control group consisted of patients who had never received leuprolide (Figure 1). Baseline variables including age, types of treatment, laboratory parameters as well as clinical outcomes were collected and analyzed. All peripheral blood counts, performed between the start of treatment with induction chemotherapy until the date of last follow-up, were extracted from the electronic medical records and analyzed. We compared short and long-term count recovery, transfusion requirements and survival between the leuprolide and control groups.
This study was performed in accordance with the Declaration of Helsinki and was approved by the MD Anderson Institutional Review Board.
Statistical analysis
Patient characteristics were summarized using median (range) for continuous variables and frequency (percentage) for categorical variables. Fisher’s exact test and Wilcoxon rank-sum test were used to assess differences in categorical and continuous variables.
Given that leuprolide was more commonly given in younger patients, propensity score matching was used to adjust for covari- ate imbalances including the age difference between the respec- tive case and control groups (Figure 1). Using the nearest-neighbor algorithm, patients from the leuprolide groups were matched to control at a 1:2 ratio.15 A logistic regression model was used to esti- mate propensity scores. All subsequent analyses including count recovery, transfusion requirements and survival were performed on matched cohorts. Absolute neutrophil count (ANC) and platelet recovery were defined as achievement of an ANC ≥1x109/L, and a platelet count ≥100x109/L after first induction chemotherapy. Scatterplots of all peripheral blood cell counts for each patient, collected between induction chemotherapy (day 0) and last follow-up date, were extracted from health records, plot- ted and compared between leuprolide and control matched groups. Lowess smooth curves were used for indicating longitudi- nal trajectories of counts and differences were assessed using the generalized estimation equation model.16,17 The probabilities of recovery were estimated using the Kaplan-Meier method. Overall survival (OS) was calculated as the time interval from treatment start date to the date of death, and was censored at the last follow- up date for patients who were alive. Event free survival (EFS) was defined as the time interval between the date of response and the date of relapse or death, whichever was first and was censored at last follow-up in patients alive and in remission. The Kaplan- Meier method was used to estimate the probability of OS and EFS, and log-rank test was used to compare survival between matched cohorts. Univariate and multivariate analyses were per- formed to determine the differential effect of leuprolide on count recovery and the interaction with baseline characteristics, treat- ment type, and relapse status. All computations were done in R version 3.4.4.
Results
Patient population
We identified 454 pre-menopausal women with AML and 257 with ALL, newly diagnosed and treated with intensive chemotherapy. Among those patients, 66 with AML and 65 with ALL had received leuprolide (Figure 1). Those who never received leuprolide were used as control cohorts and included 388 patients with AML and 192 patients with ALL (Figure 1; Table 1). Among patients with AML who received leuprolide, 33 (52%) received it between day -7 and day 15 of induction chemotherapy start, compared to 22 patients (45%) with ALL who received leuprolide during this time interval (Online Supplementary Table S2). Leuprolide was given in various dosage forms either through subcutaneous or intramuscu- lar injections depending on the platelet count at time of administration. The median cumulative dose of leuprolide per patient, given throughout cycles of chemotherapy was 22.5 mg (range, 3–78.75 mg) for those with AML and 22.5 mg (range, 11.25–135 mg) for those with ALL (Online Supplementary Table S2). Patients who received leuprolide were significantly younger than the control cohorts (AML: 33 years vs. 47 years, P<0.001; ALL: 31 years vs. 41 years, P<0.001). Baseline characteristics were well balanced after propensity score matching (Table 1). For patients with AML, the most commonly given treatment regimens con- sisted of triplets including the backbone of idarubicin and cytarabine in addition to a nucleoside analog (cladribine, clofarabine or fludarabine).18-20 The rest of the patients received a combination of idarubicin and cytarabine.
Figure 1. Patient selection. Control cohorts consisted of patients who had never received leuprolide. Propensity matching was done to adjust for covari- ate imbalances including the age difference between the respective case and control groups. All subsequent analyses were done comparing matched cohorts. AML: acute myeloid leukemia; ALL: acute lymphoblastic leukemia.
haematologica | 2021; 106(4)
1099