Venetoclax and azacitidine therapy in acute myeloid leukemia patients with severe renal impairment.

To the Editor: Acute myeloid leukemia (AML) is typically diagnosed in patients with a mean age of 65–72 years. AML in elderly patients has a dismal prognosis, as intensive chemotherapy and allogeneic hematopoietic stem cell transplantation (allo-HSCT) are not feasible. Recently venetoclax (Ven), combined with hypomethylating agents such as azacitidine (Aza) and decitabine, has proven effective in AML in both treatment-naïve and relapsed or refractory settings.1 This therapy is now widely used as a standard first-line therapy for transplant-ineligible elderly AML patients. Elderly patients often have multiple comorbidities including chronic kidney disease at the time of AML diagnosis. When used as second-line therapy, the renal function may deteriorate due to preceding infection, antibiotics, or chemotherapeutic agents. However, the efficacy and safety of Ven/Aza therapy in patients with severe renal impairment have not been established, as patients with poor renal function (creatinine clearance is less than 30 mL/min) were excluded from a pivotal study.1 Dose adjustment according to renal function is not recommended for Ven, as it is primarily metabolized by the liver. The pharmacokinetics of Ven in patients with severe renal impairment is comparable to those in patients with normal kidney function.2 Conversely, the metabolism of Aza has not been fully characterized, with limited data on patients with renal impairment, and the dose and duration are often arbitrarily modified by treating physicians. Besides the optimal dose of each agent, the risk of developing tumor lysis syndrome (TLS) has not yet been addressed. Herein, we report a series of 15 AML patients with severe renal impairment treated with Ven/Aza. To our knowledge, this study is the first to report on the clinical outcomes of this population. We retrospectively reviewed the clinical records of 176 AML patients who had been treated with at least one cycle of Ven/Aza at six tertiary medical centers in Japan, from March 2021 to December 2022. We identified 15 patients (8.5%) with severe renal impairment, defined as less than 35 mL/min creatinine clearance calculated using the Cockcroft–Gault equation. Bone marrow examination was performed to evaluate treatment response during or after the first cycle using the modified International Working Group criteria for AML. Adverse events were graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events Version 4.03. The TLS was assessed based on the Cairo–Bishop definition. The institutional review board of Kobe City Medical Center General Hospital approved this study (approval number: zn230217). This study was conducted in accordance with the principles of the Declaration of Helsinki. The baseline patient characteristics are listed in Table 1. All patients were Japanese with a median age of 83 years. Twelve of the 15 patients had de novo AML or MDS/AML. Four of the 15 patients had an adverse karyotype, and high-risk mutations were confirmed in six out of nine patients whose high-throughput target sequencing data were analyzed as previously reported.3 The Eastern Cooperative Oncology Group performance status was 0–1, except for one case (#7) with 2. The known causes of renal insufficiency included hypertension (n = 10), diabetes mellitus (n = 2), thromboangiitis obliterans (n = 1), and drug-induced nephrotoxicity (n = 2). One patient (#14) was undergoing hemodialysis before Ven/Aza therapy. The left ventricular ejection fraction was greater than 60%, except in three cases (#9, #13, and #14). All patients had normal bilirubin and aminotransferase levels, with the exception of one patient (#10). He was administered Ven/Aza because his profound hyperbilirubinemia was attributed to leukemic infiltration. The median follow-up time was 180 days (range, 149 days to not estimable). Ven/Aza therapy was initiated at a median of 5 days (range, 1–26) after diagnosis or confirmed relapse. Hydroxyurea was used in three cases to reduce peripheral white blood cell counts to less than 30 000/μL. Ven was uptitrated to the maximum dose predefined by antifungal prophylaxis (400 mg without prophylaxis, 200 mg with fluconazole, and 50 mg with posaconazole and itraconazole), except in two patients (#8 and #11), who were treated with the half-dose at most. Ven was continued at least for 14 days in the first cycle except in patient #10 with clinical TLS. The Aza dose was reduced by 20%–25% in eight patients. FLT3 inhibitors were not used in combination with Ven/Aza in the cases with mutated FLT3. Clinical response after one cycle of Ven/Aza was observed in 12 out of 15 patients (80%), with 9 achieving complete remission (CR) or CR with incomplete count recovery, and 3 achieving partial remission (Figure 1A). For those who had an initial response, the second cycle was initiated after a median of 35.5 days (range, 28–41). The second cycle deepened the response to CR in patient #4, who then underwent allo-HSCT, without significant morbidity or evidence of relapse. As a whole, Ven/Aza was continued for a median of two cycles (range, 1–14), and the median overall survival was 150 days (range, 113 days to not estimable). The causes of interruption included progressive disease (n = 4), adverse effects (n = 5), patient preference (n = 2), and bridging to allo-HSCT (n = 1). Three patients continued to undergo treatment. Two patients (#3 and #5) exhibited months of treatment-free remission after cessation of Ven/Aza. The most common adverse effect observed was cytopenia. All but one patient (#5) had grade ≥3 thrombocytopenia. Grade ≥3 anemia was observed in all patients. All but one (#12) had grade ≥3 neutropenia, and febrile neutropenia was reported in 12 patients. Patients #9 and #14 died of infectious pneumonia after four and one cycles of Ven/Aza, respectively, without evidence of progressive AML. Blood cell counts required careful monitoring as severe myelosuppression was observed in all patients. The downregulation of CYP enzymes observed in patients with renal failure might have contributed to the increased hematotoxicity. However, we could not fully determine the exact effect of renal function on the increased hematotoxicity, as our patient population predominantly comprised very elderly individuals of Japanese ethnicity.4 Acute kidney injury (AKI) during chemotherapy is associated with an increased mortality rate in AML patients, for which multiple risk factors have been identified.5 The pre-existing renal dysfunction undoubtedly confers a risk of developing AKI, and AKI often results from clinical TLS. The frequency of TLS in AML patients receiving Ven/Aza was estimated to be 8.8% in a previous study.6 To prevent TLS, 11 patients received febuxostat, and 9 received massive (≥1L) intravenous hydration. Although laboratory TLS was reported in three patients in the present study, only one patient (#10) developed clinical TLS, which was likely associated with sustained hyperbilirubinemia. After he recovered from continuous hemodiafiltration and mechanical ventilation, Ven was resumed on Day 35. Another patient (#8), who started undergoing hemodialysis after having received Ven/Aza, had been considered for it 2 years before the initiation of Ven/Aza. Serum creatinine levels were consistently decreased in our patients, except for a few patients in whom a transient increase occurred (Figure 1B), implying that disease control of AML may be associated with improvement in renal function. In summary, we demonstrated a consistent response rate of Ven/Aza in AML patients with renal insufficiency. Clinical TLS occurred in only one case, and the risk of developing AKI could be managed with appropriate measures. As grade ≥3 cytopenia was observed in all patients, myelosuppression should be carefully monitored in such cases. The main limitations of this study were the small sample size and the lack of a control group. Further studies with more patients are warranted to fully evaluate the impact of renal impairment on the clinical outcomes of Ven/Aza. Ryosuke Naka, Tadakazu Kondo, and Takayuki Ishikawa designed the study, analyzed the data, and wrote the manuscript. Masashi Nishikubo, Hiroyuki Muranushi, Yasunori Ueda, Tomomi Oka, Fumiya Wada, Junya Kanda, Shohei Yamamoto, Mitsumasa Watanabe, Shinri Okada, Kazunori Imada, Yukiharu Nakabo, Yu Mizutani, Yasuhito Nannya, Seishi Ogawa collected and provided important clinical data. All authors contributed to the writing and approved the final version of the manuscript. The authors would like to thank the medical, nursing, and laboratory staff of the participating departments for their contribution to this study. The authors would like to thank Honyaku Center Inc. for English language editing. KI received honoraria from Nippon Shinyaku and AbbVie GK. JK received honoraria from Nippon Shinyaku and AbbVie GK, and research funding from AbbVie GK. The other authors declare no conflict of interest. The data that support the findings of this study are available on request from the corresponding author.