AUTHORS
Benjamin Frost, MD, Julia Keating, M.S.2, Dagam Jeong, PharmD3, Sean Lohle, BS4, Maysam Taktak, MS5, Evan C. Chen, MD6, Virginia O Volpe, MD7, Jacqueline S. Garcia, MD8, Eric Stephen Winer, MD9, Martha Wadleigh, MD10, Frederick D Tsai, MD, PhD11, Benson M. George, MD, PhD12, Rahul Vedula, MD13, Richard Stone, MD14, Daniel J DeAngelo, MD, PhD15, R. Coleman Lindsley, MD, PhD16, Marlise Rachael Luskin, MD, MSCE17, Shai Shimony, MD, MPH18
AFFILIATIONS
1 Brigham and Women's Hospital, Boston, MA, United States
2 Dana-Farber Cancer Institute, Boston, MA, United States
3 Dana-Farber Cancer Institute, Boston, MA, United States
4 Dana-Farber Cancer Institute, Boston, MA, United States
5 Dana-Farber Cancer Institute, Boston, MA, United States
6 Dana-Farber Cancer Institute, Boston, MA, United States
7 Dana-Farber Cancer Institute, Boston, MA, United States
8 Dana-Farber Cancer Institute, Boston, MA, United States
9 Dana-Farber Cancer Institute, Boston, MA, United States
10 Dana-Farber Cancer Institute, Boston, MA, United States
11 Dana-Farber Cancer Institute, Boston, MA, United States
12 Dana-Farber Cancer Institute, Boston, MA, United States
13 Dana-Farber Cancer Institute, Boston, MA, United States
14 Dana-Farber Cancer Institute, Boston, MA, United States
15 Dana-Farber Cancer Institute, Boston, MA, United States
16 Dana-Farber Cancer Institute, Boston, MA, United States
17 Dana-Farber Cancer Institute, Boston, MA, United States
18 Dana-Farber Cancer Institute, Boston, MA, United States
ABSTRACT
Introduction: Venetoclax (VEN) plus a hypomethylating agent (HMA) is standard therapy for patients (pts) with acute myeloid leukemia (AML) ineligible for intensive chemotherapy. As VEN/HMA indications expand and FDA approval of an all-oral regimen simplifies outpatient use, identifying pts at risk for tumor lysis syndrome (TLS) with VEN/HMA, termed treatment-emergent TLS (teTLS), is increasingly important. Whether teTLS is determined by disease genotype has not been established. IDH2-mutant AML is sensitized to VEN, and we hypothesized that IDH2 mutations would confer VEN-dependent teTLS risk. We assessed genotype-specific teTLS risk during cycle 1 of VEN/HMA and compared rates by genotype to HMA alone in newly diagnosed (ND) AML.
Methods: We included pts ≥18 years with ND AML treated with VEN/HMA or HMA alone at our center between Sept. 2014 and Apr. 2025. Pathogenic mutations at diagnosis were detected by targeted DNA sequencing. Blast percentage (%) was the highest at diagnosis from the bone marrow aspirate or core biopsy, or, if no diagnostic marrow, the peripheral blood. TeTLS was defined as Cairo-Bishop laboratory TLS (Cairo, BJHaem 2004) from C1D1-D7. Among VEN/HMA pts, genes mutated in ≥10 pts that met a univariable threshold of p<0.10 were included in a multivariable logistic regression model for teTLS, along with pre-specified variables: baseline WBC, LDH, elevated Cr (>ULN), and blast %. We tested VEN-by-mutation interactions using Firth’s logistic regression; IDH2 was pre-specified, and other mutations were exploratory.
Results: 350 pts were included, of whom 287 (82%) received VEN/HMA and 63 (18%) received HMA alone. TeTLS occurred in 85 (24.3%) pts, with similar rates with VEN/HMA (24.0%) and HMA alone (25.4%, p=0.82) despite more frequent prophylactic allopurinol (p<0.001) and IV fluids (p=0.0012) with VEN/HMA. Clinical TLS occurred in 5.6% (16/287) of VEN/HMA-treated pts. Pts with teTLS had more VEN interruptions than those without (8.7% vs 2.3%; p=0.026).
Among VEN/HMA-treated pts (n=287), pts with teTLS had higher baseline blast % (median 79% vs 50%, p<0.001), WBC (7.7 vs 2.9 ×10⁹/L, p<0.001), LDH (371 vs 316 U/L, p=0.015), and rate of elevated Cr (20.3% vs 11.0%, p=0.047) than those without. Age (median 74 vs 73), sex, race, ELN 2024 risk, extramedullary disease, prophylactic allopurinol or IV fluids, VEN ramp-up, CYP3A inhibitor use, and HMA type were similar in pts with and without teTLS (all p>0.05). Pts with teTLS had lower potassium (median 3.9 vs 4.0, p=0.0013) and calcium (8.5 vs 8.7, p=0.043) than those without, but similar uric acid and phosphorus levels.
IDH2, CEBPA, NPM1, SRSF2, SF3B1, DNMT3A, and TP53 mutations met the univariable screening threshold; IDH1 mutations did not (OR 1.67; p=0.24). In the multivariable setting, IDH2 (OR 3.00, 95% CI 1.07-8.46; p=0.036) and CEBPA (OR 3.72, 95% CI 1.00-13.8; p=0.045) mutations were associated with higher teTLS risk, while DNMT3A mutations were associated with lower risk (OR 0.36, 95% CI 0.15-0.82; p=0.020). Blast % was the strongest covariate (OR 1.02 per 1% increase, 95% CI 1.01-1.03; p<0.001), and a data-derived cut-point of 70% blasts best separated pts with and without teTLS (AUC 0.66; sensitivity 0.64, specificity 0.67). No other univariable associations, including NPM1 (OR 2.23, 95% CI 1.10-4.43; p=0.023), persisted in the multivariable model. NPM1 mutations were associated with higher blast % (p<0.001).
IDH2 mutations remained associated with teTLS among DNMT3A-mutated pts (OR 5.16, 95% CI 1.22-20.9; p=0.027). Three of 12 CEBPA-mutated pts had in-frame bZIP mutations; teTLS occurred in 1/3 (33.3%) pts with bZIP mutations and 5/9 (55.6%) without. One pt had co-mutated CEBPA and DNMT3A.
The IDH2-associated teTLS risk was VEN-dependent (interaction p=0.0015). Among IDH2-mutated pts, teTLS occurred in 43.5% (10/23) with VEN/HMA and 0% (0/11) with HMA alone despite similar blast % (70% vs 81%, p=0.15). In contrast, CEBPA-associated teTLS risk was not VEN-dependent (interaction p=0.79).
Conclusion: As VEN/HMA use in AML expands, identifying pts at risk for teTLS is essential. Blast % at diagnosis was the strongest determinant of teTLS. The association between IDH2 mutations and higher teTLS risk was VEN-dependent but independent of AML burden, marking IDH2 as a unique, VEN-specific risk factor for teTLS. CEBPA mutations also conferred higher teTLS risk. Pts with IDH2 or CEBPA mutations may warrant closer monitoring at VEN/HMA initiation.
DISCLOSURES
Benjamin Frost, MD: No relevant conflicts of interest to declare.