Abstract
Background/Aim: The combination of venetoclax (VEN) and azacitidine (AZA) (VEN+AZA) leads to higher complete remission rates and longer overall survival (OS) in patients with untreated acute myeloid leukemia (AML) who are ineligible for intensive combination chemotherapy. In practice, the doses of VEN and AZA are reduced at the attending physician’s discretion to avoid adverse events; however, the impact of dose and duration reductions has not been fully clarified. We analyzed whether the efficacy was maintained with reduced VEN+AZA compared to AZA monotherapy in the real world. Patients and Methods: A total of 33 patients were included; 17 (10 newly diagnosed, 7 primary refractory or relapsed) received VEN+AZA, and 16 (7 newly diagnosed, 9 primary refractory or relapsed) received AZA. We analyzed complete remission (CR) and CR with incomplete hematologic recovery (CRi) rates, OS, and the incidence of adverse events. Results: CR/CRi were achieved in 7/17 (41.2%) and 11/17 (64.7%) patients in the VEN+AZA group and 0/15 (0%) and 2/15 (6.7%) patients in the AZA group, respectively. The CR/CRi rate was higher in the VEN+AZA group than in the AZA group (p=0.001). OS was longer in the VEN+AZA group than in the AZA group (p=0.03), with a median of 506 days [95% confidence interval (CI)=234-585 days] and 208 days (95% CI=52-343 days), respectively. Conclusion: The doses of the VEN+AZA combination were reduced at the attending physician’s discretion, resulting in a higher CR/CRi rate and longer OS than AZA monotherapy and is considered useful for AML in the real world.
Acute myeloid leukemia (AML) requires intensive combination chemotherapy. Conversely, low-intensity chemotherapy, such as low-dose cytarabine or azacitidine (AZA) monotherapy, has been used in elderly patients because of intolerance to conventional intensive combination chemotherapy. However, AZA monotherapy is associated with poor prognosis (1). The VIALE-A trial showed that the combination of venetoclax (VEN) and AZA (VEN+AZA) led to higher complete remission (CR) rates and longer overall survival (OS) than the placebo and AZA in patients untreated with AML ineligible for intensive combination chemotherapy (2). A high incidence of myelosuppression and febrile neutropenia (FN) was reported with VEN+AZA in the trial. In practice, the doses of VEN and AZA are reduced at the attending physician’s discretion to avoid adverse events considering the patient’s age, comorbidities, and systemic status. The effect of dose and duration reduction of VEN+AZA on efficacy has not been fully clarified. In this study, we analyzed the reduction rate of VEN and AZA and whether efficacy was maintained with reduced VEN+AZA compared with AZA monotherapy in the real world.
Patients and Methods
Patient selection. This study retrospectively included patients with AML who received at least one cycle of VEN+AZA or AZA between April 2016 and May 2023 at Showa University Fujigaoka Hospital, Kanagawa, Japan. Patients who received VEN+AZA or AZA for bridging allogeneic hematopoietic stem cell transplantation (HSCT) were excluded; patients who relapsed after HSCT were also excluded. AML was diagnosed based on the World Health Organization Classification system.
This study was conducted in accordance with the principles of the Declaration of Helsinki and approved by the Showa University Research Ethics Review Board (2023-078-B). Informed consent was waived and opt-out agreement was applied by the Showa University Research Ethics Review Board.
VEN and AZA administration. VEN and AZA were administered as follows: VEN was administered orally at 100, 200, and 400 mg on days 1, 2, and 3, respectively, during cycle 1, 400 mg daily thereafter until day 28, and 400mg for 28 days from cycle 2. Following the criteria for dose modifications of the VIALE-A trial, VEN was decreased to 50, 100, and 200 mg if co-administered with a moderate cytochrome P450 inhibitor and to 10, 20, and 50 mg with a potent inhibitor. AZA was administered subcutaneously or intravenously at 75 mg per square meter of body-surface area on days 1-7, every 28 days. Dose adjustments and discontinuations were determined by the treating physicians. The reduction rate was calculated based on the daily dose and duration of administration by dividing the actual reducing dose by the basal dose. In the case of co-administration with a moderate or strong cytochrome P450 inhibitor, basal doses were defined as the doses specified in the criteria for dose modification. For example, if VEN was administered at 50 mg (day 1), 100 mg (day 2), and 200 mg (day 3-14) with a moderate cytochrome P450 inhibitor and AZA was administered at 75 mg per square meter of body-surface area (day 1-5), the reduction rate of VEN was 50.0%, and that of AZA was 28.6%. When the fractions were rounded off from the calculated doses of AZA, the reduction rate was considered to be 0%. The reduction rates were calculated in the first cycle for all patients and in all cycles for patients who received multiple cycles.
Assessment of outcomes. The treatment response was evaluated according to the International Working Group response criteria (3). CR was defined as an absolute neutrophil count of >1,000 cells per cubic millimeter, a platelet count of >100,000 per cubic millimeter, red-cell transfusion independence, and bone marrow with less than 5% blasts. Complete remission with incomplete hematologic recovery (CRi) was defined as complete remission, except for neutropenia (absolute neutrophil count, ≤1,000 per cubic millimeter) or thrombocytopenia (platelet count, ≤100,000 per cubic millimeter). Cases in which a bone marrow examination was not performed at the attending physician’s discretion based on peripheral blood findings were considered non-CR/CRi.
OS was defined as the number of days from the first day of VEN+AZA or AZA treatment to the date of death or last follow-up. The clinical data cut-off date was December 31, 2023. Adverse events were summarized using the National Cancer Institute Common Terminology Criteria for Adverse Events Version 5.0.
Statistical analysis. Statistical analyses were performed using JMP® 15 software (SAS Institute, Cary, NC, USA). OS was calculated using the Kaplan-Meier method. The log-rank test was used to compare survival curves. Fisher’s exact test was used to compare the ratios between groups, such as the CR/CRi rate and the ratio of the incidence of FN and cytopenia. We used the Mann-Whitney U-test to compare the number of total cycles between the VEN+AZA and AZA groups and the reduction rate between newly diagnosed and primary refractory or relapsed AML in the VEN+AZA group.
Results
Patient background. Table I shows the baseline characteristics of the patients who were administered VEN+AZA or AZA. A total of 33 patients were included: 17 received VEN+AZA (VEN+AZA group), and 16 received AZA (AZA group). The median patient ages were 78 (range=69-85) and 79.5 (range=58-88) years, respectively. Newly diagnosed AML (including pre-treatment for decreasing blasts) occurred in 10 and 7 patients, and primary refractory or relapsed AML occurred in 7 and 9 patients in the VEN+AZA and AZA groups, respectively. Two patients in each group had received prior AZA treatment for myelodysplastic syndrome (MDS) and progressed to AML.
Baseline characteristics of patients receiving VEN+AZA or AZA.
Dose, duration, and reduction rate. Patients received a median of 7 and 2.5 cycles (range=1-24 and 1-11) of VEN+AZA and AZA, respectively. The total number of cycles was higher in the VEN+AZA group than in the AZA group (p=0.03). One and six patients received only one cycle of VEN+AZA and AZA, respectively.
In the VEN+AZA group, the mean dose reduction and duration of VEN administration during the first cycle were 26.5% (0-75%) and 20.2 days (7-28 days), respectively. The mean reduction rates during the first and all cycles were 42.6% (0-93.8%) and 59.9% (0-87.5%), respectively. The mean dose reduction and duration of AZA treatment during the first cycle was 18.0% (0-35.3%). The duration of AZA treatment was seven days for all patients. The mean reduction rates during the first and all cycles were 18.0% (0-35.3%) and 24.2% (0-35.3%), respectively. Among the 17 patients, six received the full dose and duration of VEN, and four received the full dose and duration of both VEN and AZA during the first cycle. Among the 16 patients who received multiple cycles, only two received the full dose and duration of VEN, and no patient received either VEN or AZA during all cycles. There was no difference in the reduced rates of VEN and AZA between the newly diagnosed and primary refractory or relapsed AML subgroups (p=0.67 and 0.40, respectively).
In the AZA group, the mean dose reduction and duration of AZA treatment during the first course were 7.3% (0-42.9%) and 5.9 days (range=4-7). The mean reduction rates during the first and all cycles were 16.1% (0-59.2%) and 23.7% (0-59.2%), respectively.
Response and survival. CR and CRi were achieved in 7/17 (41.2%) and 11/17 (64.7%) patients in the VEN+AZA group and 0/16 (0%) and 2/16 (12.5%) in the AZA group, respectively. The CR/CRi rate was higher in the VEN+AZA group than in the AZA group (p=0.001). In the newly diagnosed and primary refractory or relapsed AML subgroups, CR/CRi was achieved in 6/10 (60.0%) and 5/7 (71.4%) patients in the VEN+AZA group, respectively, which was higher than in the AZA group (p=0.03 and 0.02). Figure 1 shows the Kaplan–Meier survival curves for OS. OS was longer in the VEN+AZA group than in the AZA group (p=0.03), with a median of 506 days [95% confidence interval (CI)=234-585 days] and 208 days (95% CI=52-345 days), respectively.
Kaplan–Meier survival curve depicting overall survival (OS) of the overall population. OS was longer in the venetoclax (VEN) + azacitidine (AZA) (VEN+AZA) group than in the AZA group (p=0.03), with a median of 506 days [95% confidence interval (CI)=234-585 days] and 208 days (95% CI=52-343 days), respectively.
Eleven and 14 patients died during the follow-up in the VEN+AZA and AZA groups, respectively (64.7% and 87.5%). Seven and eight patients died because of progressive disease (PD), and three and five because of infection (disease status: two and three had PD, one and two were without PD). One patient in the VEN+AZA group died of cerebral hemorrhage with PD. One patient in the AZA group died of hemophagocytic syndrome with PD.
Toxicity. Regarding toxicity, in the VEN+AZA group, 100% and 58.8% of the patients experienced grade 4 neutropenia and FN, respectively, whereas 60% and 47% were in the AZA group. The incidence of grade 4 neutropenia was higher in the VEN+AZA group than in the AZA group (p=0.003); however, there was no significant difference in the incidence of FN (p=0.50).
Discussion
This report demonstrates that VEN+AZA reduced at the attending physician’s discretion can achieve a higher CR/CRi rate and longer OS than AZA monotherapy in the real world. Although the sample size was small, it was meaningful to compare VEN+AZA with AZA to demonstrate the usefulness of VEN+AZA.
We demonstrated that VEN+AZA has a higher CR/CRi rate and longer OS than AZA monotherapy and is considered useful for AML. In the VEN+AZA group, our study’s CR/CRi rate was 64.7%, almost equal to that in the VIALE-A trial (66.4%) (2). Moreover, in the VEN+AZA group, the median OS is 506 days (16.9 months), which is slightly longer than the VIALE-A trial (14.7 months) (2). The patient characteristics differed from those in the VIALE-A trial because our cohort included patients with primary refractory or relapsed AML and prior AZA treatment for MDS. In the subgroups of newly diagnosed AML and primary refractory or relapsed AML, the CR/CRi rates were 60.0% and 71.4%, respectively. Compared with previous studies (4-8), especially those with high CR/CRi rates in patients with primary refractory or relapsed AML, this suggests that VEN+AZA reduction at the attending physician’s discretion is useful in the real world, even in refractory or relapsed cases.
Conversely, in analyzing a small number in the subgroup of newly diagnosed AML cases in our cohort, the CR/CRi rate was slightly lower than that in the VIALE-A trial. We focused on the possibility of excessive VEN reduction as the reason for this. It remains unclear whether a reduced dose of VEN+AZA results in a weakened effect, whereas the full dose and duration of VEN+AZA treatment has shown high efficacy. Shortening the course of VEN to 14 days, starting the first cycle in frail elderly patients, has been proposed (9). A previous retrospective study suggested that shortening the duration of VEN to 14 days (50% reduction rate) may reduce the risk of complications and be as effective as VEN for 28 days (10). The details of the reduction rate are not available. However, recent studies have reported that newly diagnosed patients with AML treated with VEN+AZA in a real-world setting had inferior outcomes compared with those in the VIALE-A trial (11-15). In practice, the dose and duration of VEN+AZA are reduced at the attending physician’s discretion without established criteria. In our cohort, only 23.5% of patients received the full dose and duration of both VEN and AZA during the first cycle, and none received either VEN or AZA during the entire cycle. The reduced rate of all cycles was > 50% in 75% (12/16) of the patients in the VEN+AZA group. Only one previous study reported that the response rate and OS were not different when newly diagnosed and primary refractory or relapsed AML were combined and compared based on whether the VEN reduction rate was 50% or more (16). However, HSCT cases were included, and newly diagnosed AML cases alone were not analyzed. Thus, excessive VEN+AZA reduction may not be advisable in patients newly diagnosed with AML. Furthermore, because there was no difference in the reduction rate between newly diagnosed and primary refractory or relapsed AML, reduction of VEN+AZA at the attending physician’s discretion may be acceptable in patients with primary refractory or relapsed AML rather than in newly diagnosed AML. Acceptance of the reduced dose and duration of VEN+AZA may differ between newly diagnosed and primary refractory or relapsed AML. Clarification of this reduction is desired.
Regarding safety, the incidence of grade 4 neutropenia was higher in the VEN+AZA group than in the AZA group; however, there was no difference in the incidence of FN. The incidence of FN in the VEN+AZA group (58.8%) was higher than that in the VIALE-A trial (30%), although our cohort showed a high reduction rate (2). Specifically in elderly patients, attending physicians often prefer reduced treatment to avoid severe neutropenia and subsequent FN. In our cohort, since the incidence of FN was not different between the VEN+AZA and AZA groups and was higher than that in the VIALE-A trial, the reduction of VEN+AZA alone may not have resulted in less FN. Further studies are needed to consider reduction and appropriate supportive care.
Study limitations. This study was limited by its small sample size. We did not investigate genetic abnormalities to classify the cytogenetic risk, despite the important prognostic role of cytogenetics and mutation testing (17, 18). As this was a retrospective study, clinical data were limited to information obtained from medical records. Future studies should involve larger sample sizes to validate our findings and confirm their clinical relevance.
Conclusion
VEN+AZA was reduced at the attending physician’s discretion, resulting in a higher CR/CRi rate and longer OS than AZA monotherapy and is considered useful for AML in the real world. Notably, the analysis of a small number of cases suggests that excessive reduction of VEN+AZA may have impaired efficacy in newly diagnosed AML. The acceptance of the reduced VEN+AZA may differ between newly diagnosed and primary refractory or relapsed AML. Further studies are needed to establish criteria for the reduction of VEN+AZA.
Acknowledgements
We thank the medical staff at Showa University Fujigaoka Hospital for providing patient care. Dr. Inoue provided statistical expertise. We thank Editage (www.editage.jp) for English language editing.
Footnotes
Authors’ Contributions
YB, NH, TS, and KY conceived the study. YB, MA, NK, HS, and TF performed patient care. YB developed the statistical analysis plan and performed the statistical analyses. NH, HS, and KY contributed to the interpretation of the results. YB drafted the manuscript. TF supervised the study. All Authors reviewed and approved the final version of the manuscript.
Conflicts of Interest
The Authors declare that they have no conflicts of interest.
- Received January 21, 2024.
- Revision received February 24, 2024.
- Accepted March 4, 2024.
- Copyright © 2024 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.