Outcomes in patients with newly diagnosed TP53-mutated acute myeloid leukemia with or without venetoclax-based therapy
Abstract
Background
Venetoclax (VEN) in combination with a hypomethylating agent (HMA) has become the standard of care for patients aged >75 years and for those not eligible for intensive chemotherapy who have newly diagnosed acute myeloid leukemia (AML). The benefit of VEN-based therapy in patients who have newly diagnosed AML with mutations in the TP53 gene (TP53mut) over standard therapy is undefined.
Methods
In this single-institutional, retrospective analysis, the authors assessed the clinical outcomes of 238 patients with newly diagnosed TP53mut AML and compared the clinical characteristics, response to different therapies, and outcomes of those who received VEN-based (n = 58) and non–VEN-based (n = 180) regimens.
Results
Patients who received VEN-based regimens were older (aged >65 years: 81% vs 65%; P = .02) and had higher response rates (complete remission, 43% vs 32%; P = .06) than those who received non–VEN-based regimens. Compared with patients who received non–VEN-based regimens, no difference in overall survival (median, 6.6 vs 5.7 months; P = .4) or relapse-free survival (median, 4.7 vs 3.5 months; P = .43) was observed in those who received VEN-based regimens, regardless of age or intensity of treatment.
Conclusions
The addition of VEN to standard treatment regimens did not improve outcomes in younger or older patients who had TP53mut AML. These data highlight the need for novel therapies beyond VEN to improve the outcome of patients with TP53mut AML.
Introduction
Recent advances and wide access to next-generation DNA sequencing has further revealed the genetic diversity and heterogeneity of acute myeloid leukemia (AML).1 Originally classified by morphology and, later, in groups defined by recurrent chromosomal abnormalities, the advent of higher resolution DNA sequencing has uncovered recurrent somatic mutations that have important implications in overall prognosis, response to therapy, and pathobiology.2 Current prognostic and treatment paradigms incorporate karyotype and mutational analysis to allow more accurate prognostication and long-term treatment planning in AML.
Mutations in the TP53 gene (TP53mut) are among the most common abnormality across malignancies3 and have been identified in 15% to 20% of patients with newly diagnosed AML.4-7 TP53mut AML is frequently associated with therapy-related AML, complex karyotype, and often with deletion of the short arm of chromosome 17 (17p−), portending an adverse prognosis.8-10 TP53mut AML has also been associated with lower rates of remission with standard therapy, high rates of relapse, and poor overall survival (OS).9 Furthermore, studies have shown that patients with TP53mut AML may not derive additional benefit from intensive chemotherapy and may actually have similar outcomes and less toxicity with lower intensity approaches.10, 11 An important exception are those patients who have TP53 mutations with a low variant allelic frequency (VAF), in which the mutated clone may not represent the dominant driver.12 Despite the recent approval of several new agents for the treatment of AML, satisfactory treatment of TP53mut AML remains elusive and is the focus of ongoing drug development.
Recently, the BCL-2 inhibitor venetoclax (VEN) has been incorporated into several treatment regimens for selected patients with newly diagnosed AML. On the basis of the modest activity of VEN monotherapy in relapsed/refractory AML,13 VEN was studied in combination with lower intensity chemotherapy regimens for patients with newly diagnosed AML who were older or were considered ineligible for intensive chemotherapy. After early phase studies demonstrating the activity of VEN in combination with hypomethylating agents (HMAs)14 or low-dose cytarabine (LDAC),15 randomized phase 3 studies evaluating the respective combinations confirmed the benefit of adding VEN to lower intensity therapy in this population. In the phase 3 placebo-controlled trial, 431 patients with a median age of 76 years were randomized to receive 5-azacitidine either with or without VEN. The composite complete remission (CR) rate (CRc) (CR or CR with incomplete hematologic recovery [CRi]) was 66.4% for patients who received the combination, with a median remission duration (CRd) of 17.5 months and a median OS of 14.7 months.11 In the phase 3 placebo-controlled trial of LDAC with or without VEN in 210 patients with a median age of 76 years, the CRc rate was 48% in the combination arm, but the combination did not meet the survival end point.16 The improvement in response rates appeared to be maintained across most subgroups of AML, with somewhat attenuated responses among the high-risk subgroups.14 Compared with other genomically defined subgroups within the trials, patients who had TP53mut AML had lower response rates and inferior OS with VEN-based combination therapy.11, 14
Given the diminished activity of VEN in TP53mut AML relative to other genomically defined subtypes, we reviewed our own institutional experience with VEN-based therapy to determine whether the addition of VEN improved outcomes compared with treatment without VEN (non–VEN)-based therapy in patients with newly diagnosed TP53mut AML.
Materials and Methods
Patients
We conducted a retrospective review of patients who had newly diagnosed AML with a TP53 mutation at the time of diagnosis who were treated at our institution between 2014 and 2019 and had baseline molecular testing available. Baseline patient and disease characteristics of TP53mut AML were analyzed and compared among subsets of AML with regard to response to therapy, response duration, and survival outcomes. Patients were treated on protocols approved by the Institutional Review Board, with written informed consent obtained before enrollment, in accordance with the Declaration of Helsinki.
Treatments and Response
Patients received various therapies as initial treatment for newly diagnosed AML (see Supporting Table 1), These were categorized as 1) low-intensity therapy, which included HMA-based (5-azacitidine or decitabine) regimens and LDAC-based combinations, or 2) high-intensity therapy, which included regimens using combinations with cytarabine at a minimum dose of 1000 mg/m2 daily (range, 1000-2000 mg/m2 daily). A group that received what was denoted HMA-based therapy was a subset of the low-intensity group that had received only HMA-based treatment. For the purpose of analysis, each of these groups was then divided by those who did or did not receive combination with VEN. Patients who had received HMA for myelodysplastic syndromes (MDS) before their diagnosis of AML were excluded from the analysis.
Responses were assessed using standard criteria and were based on morphologic evaluation.17 CR was defined as <5% blasts in the bone marrow, neutrophil count ≥1.0 × 109/L, platelet count ≥100 × 109/L, and no evidence of extramedullary leukemia. CRi met all criteria for CR except for either residual neutropenia (absolute neutrophil count <1.0 × 109/L) or thrombocytopenia (platelet count <100 × 109/L). Minimal residual disease (MRD) was evaluated using 8-color multiparameter flow cytometry with a sensitivity of 0.1%. A distinct cluster showing ≥20 cells with aberrant expression of ≥2 antigens was regarded as positive for MRD.18
Cytogenetic and Molecular Assessment
Conventional G-band karyotyping was performed on the bone marrow at diagnosis, before the initiation of therapy as described previously.19 Three cytogenetic risk groups were defined: diploid (patients with normal karyotype), intermediate (patients with nondiploid intermediate-risk karyotype defined according to the European LeukemiaNet [ELN] 2017 recommendations), and adverse (patients with adverse-risk karyotype according to ELN). There were no patients in this cohort with a favorable karyotype. Molecular testing to detect known somatic mutations was performed by the institutional Clinical Laboratory Improvement Amendments-certified Molecular Diagnostics Laboratory. Amplicon-based, targeted next-generation sequencing-based mutation profiling was performed using DNA extracted from the bone marrow aspirate material on the Illumina MiSeq sequencer (Illumina), as described previously.20 The exons (codons) covered on the TP53 gene were exon 2 (codons 1-25), exon 4 (codons 33-45), exon 4 (codons 41-80), exon 4 (codons 72-112), exon 4 through 6 (codons 107-214), exon 6 (codons 210-224), and exons 7 through 10 (codons 234-367). Adequate coverage was defined as a total coverage depth of at least 250X.10 The limit of detection for variant calling was 2%.
Statistical Analysis
Relapse-free survival (RFS) was calculated from the time of CR/CRi until relapse or death resulting from any cause and was censored if the patient was alive at last follow-up. OS was measured from the start of treatment until death or was censored at last follow-up; Survival end points were not censored at allogeneic stem cell transplantation (ASCT).2 CRd was defined as the time from achieving a CR or until relapse from any cause. CRd was censored at death or HSCT. The Fisher exact test and the Mann-Whitney U test were used for categorical and continuous variables, respectively. Survival distribution was estimated using Kaplan-Meier curves, and survival differences were evaluated using the log-rank test. All differences with P < .05 were considered statistically significant (2-tailed test). Multivariate analysis was done using a Cox proportional hazard model. Statistica (version 13.5; TIBCO software) was used for statistical analysis.
Results
Patient Characteristics
We reviewed 238 patients with newly diagnosed TP53mut AML who were treated at our institution between 2014 and 2019. Their median age was 69 years (range, 20-90 years), and 53% were men). The baseline characteristics overall and within each group are summarized in Table 1. These patients exhibited characteristic hematologic parameters that have been previously described in newly diagnosed TP53mut AML, with lower white blood cell and platelet counts and a low percentage of bone marrow blasts.10 Most patients within the cohort were categorized as having adverse cytogenetic risk, predominantly exhibiting complex karyotype (82%), with 69% (n = 42) harboring abnormalities of chromosomes 5 and/or 7. Fifty-eight patients (24%; median age, 73 years) were treated with VEN therapy, and 180 patients (76%; median age, 70 years) were treated with non-VEN therapy. Compared with patients who received non-VEN therapy, those who received VEN were older, and a larger percentage were aged >65 years (81% vs 65%; P = .02). However, among patients who received lower intensity therapy with or without VEN, there was no significant difference in age (median, 71 years; range, 37-90 years). Overall, there were no other significant differences in baseline characteristics among patients treated with or without VEN-based therapies.
| Characteristic | Median [Range] or No. (%) | P | ||
|---|---|---|---|---|
| Overall, N = 238 | Venetoclax-Based Therapy, N = 58 | Nonvenetoclax-Based Therapy, N = 180 | ||
| Age, y | 69 [20-90] | 73 [25-86] | 70 [20-90] | .02 |
| >65 y | 163 (68) | 47 (81) | 116 (65) | |
| Men | 123 (68) | 28 (48) | 95 (53) | .51 |
| WBC, 103/μL | 3 [0.4-77.3] | 3.3 [0.7-72.6] | 2.9 [0.4-77.3] | .82 |
| Platelets, 103/μL | 31 [2-321] | 34 [5-136] | 30 [2-321] | .61 |
| PB blasts, % | 9 [0-97] | 10 [0-91] | 9 [0-97] | .93 |
| Bone marrow blasts, % | 32 [3-97] | 33 [8-88] | 32 [3-97] | .51 |
| Cytogenetic risk category | ||||
| Adverse | 200 (84) | 52 (90) | 148 (82) | .17 |
| Diploid | 15 (6) | 2 (3) | 13 (7) | .37 |
| Intermediate | 14 (6) | 2 (3) | 12 (6) | .53 |
| TP53 VAF | 42.8 [1-100] | 42.7 [1-100] | 43.3 [1-95] | |
- Abbreviations: PB, peripheral blood; VAF, variant allelic frequency; WBC, white blood cells.
Treatment Response and Early Mortality
Response rates in patients who had TP53mut AML treated with or without VEN-based treatment by age and intensity of therapy are summarized in Table 2. The choice of initial therapy by the treating physician was based on age, performance status, comorbidities, and pretreatment disease characteristics. Among the patients who received lower intensity-based therapy, overall, the addition of VEN was associated with significantly higher rates of CRc (57% vs 39%; P = .02). Patients aged ≥65 years had a significantly higher CRc rate with VEN-based therapy compared with those who received non–VEN-based therapy (57% vs 39%; P = .03). Within the HMA group, patients who received VEN had significantly higher CR rates compared with those who received non–VEN-based therapies (46% vs 28%; P = 0 01). The CRc rates among patients who received treatment with HMA-VEN were comparable to the rates among those who received intensive chemotherapy (56% vs 52%; P = .68). When comparing the entire cohort across therapies, there was no significant difference in the CRc rate between patients receiving VEN-based and non–VEN-based treatment regimens (P = .18).
| Subgroup | No. of patients with response or early mortality/Total No. at Risk (%) | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CR | CRi | CRc | 4-Week Mortality | 8-Week Mortality | |||||||||||
| VEN-Based | Non–VEN-Based | P | VEN-Based | Non–VEN-Based | P | VEN-Based | Non–VEN-Based | P | VEN-Based | Non–VEN-Based | P | VEN-Based | Non–VEN-Based | P | |
| Regimen | |||||||||||||||
| Low intensity | 25/53 (47) | 35/126 (28) | .01 | 5/53 (9) | 14/126 (11) | .6 | 30/53 (57) | 49/126 (39) | .02 | 4/53 (8) | 11/126 (9) | .8 | 12/53 (23) | 23/126 (18) | .4 |
| HMA-based | 23/50 (46) | 26/93 (28) | .01 | 5/50 (10) | 14/93 (15) | .4 | 28/50 (56) | 40/93 (43) | .1 | 4/50 (8) | 11/93 (12) | .4 | 12/50 (24) | 21/93 (23) | .9 |
| High intensity | 0/5 (0) | 23/54 (43) | .06 | 1/5 (20) | 5/54 (9) | .4 | 1/5 (20) | 28/54 (52) | .1 | 0/5 (0) | 7/54 (13) | .3 | 1/5 (20) | 8/54(15) | .7 |
| Age, y | |||||||||||||||
| <65 | 2/11 (18) | 24/64 (38) | .2 | 2/11 (18) | 8/64 (13) | .6 | 4/11 (36) | 32/64 (50) | .4 | 0/11 (0) | 3/64 (5) | .5 | 1/11 (9) | 5/64 (8) | .9 |
| ≥65 | 23/47 (49) | 34/116 (29) | .02 | 4/47 (9) | 11/116 (9) | 1.0 | 27/47 (57) | 45/116 (39) | .03 | 4/47 (9) | 15/116 (13) | .47 | 12/47 (26) | 26/116 (22) | .6 |
- Abbreviations: CR, complete remission; CRc, composite complete remission rate; CRi, complete remission with incomplete hematologic recovery; HMA, hypomethylating agent; OR, odds ratio; RFS, relapse free survival; VEN, venetoclax.
The rates of early mortality by regimen and age are summarized in Table 2. There was no significant difference between 4-week and 8-week mortality rates among patient treated with or without VEN (4-week mortality: 7% [VEN] vs 10% [non-VEN]; P = .5; 8-week mortality: 22% [VEN] vs 17% [non-VEN]; P = .4).
Survival Outcomes
The OS and RFS outcomes by subgroup are summarized in Table 3. Overall, there was no difference in survival between patients treated with or without VEN-based regimens, with a median OS of 5.7 versus 6.6 months, respectively (P = .4) (Fig. 1A). This lack of difference in OS was preserved when patients were stratified by age (Fig. 1B,C) and when patients received lower intensity therapy with or without VEN (Fig. 1D). Similarly, there was no difference in OS between patients with TP53mut AML who were treated specifically with HMA therapy alone with or without VEN (median OS, 6.37 vs 7.20 months; P = .87) (see Supporting Fig. 1C) or those who were treated with a 10-day course of decitabine (median OS, 7.26 vs 6.6 months; P = .25) (see Supporting Fig. 1D). The OS rates were comparable between those who received HMA-VEN and those who received high-intensity chemotherapy without VEN (median OS, 6.37 vs 7.03 months; P = .07) when censored for ASCT (see Supporting Fig. 1E). Because there were only 5 patients in the cohort that received treatment with VEN and higher intensity therapy, a meaningful comparison could not be conducted.
| Variable | Overall | Low Intensity | HMA-Based | Aged <65 Years | Aged ≥65 Years | |||||
|---|---|---|---|---|---|---|---|---|---|---|
| VEN-Based, N = 58 | Non–VEN-Based, N = 180 | VEN-Based, N = 53 | Non–VEN-Based, N = 126 | VEN-Based, N = 50 | Non–VEN-Based, N = 93 | VEN-Based, N = 11 | Non–VEN-Based, N = 64 | VEN-Based, N = 47 | Non–VEN-Based, N = 116 | |
| OS | ||||||||||
| Median, mo | 5.7 | 6.6 | 6.4 | 6.5 | 5.7 | 6.6 | 4.1 | 8.5 | 5.7 | 5.1 |
| 1-y, % | 22.0 | 28.0 | 23.0 | 26.0 | 20.0 | 26.0 | 27.0 | 32.0 | 21.0 | 21.0 |
| 2-y, % | 12.0 | 10.0 | 12.0 | 7.0 | 10.0 | 4.0 | — | 15.0 | 11.0 | 11.0 |
| RFS | ||||||||||
| Median, mo | 3.5 | 4.7 | — | — | 3.4 | 4.41 | 7.9 | 3.1 | 3.4 | 5.2 |
| 1-y, % | 13.0 | 24.0 | — | — | 9.0 | 6.0 | 30.0 | 22.0 | 10.0 | 25.0 |
| 2-y, % | 13.0 | 11.0 | — | — | 9.0 | — | — | 19.0 | 10.0 | 5.0 |
- Abbreviations: CR, complete remission; CRc, composite complete remission rate; CRi, complete remission with incomplete hematologic recovery; HMA, hypomethylating agent; OR, odds ratio; RFS, relapse free survival; VEN, venetoclax.
Similarly, there was no difference in RFS between patients treated with or without VEN-based therapy (median RFS, 3.5 vs 4.7 months, respectively; P = .43) (Fig. 2A). This was also maintained when subgroups were examined by age (Fig. 2B,C) and among patients who received HMA-based therapy with or without VEN (Fig. 2D). Of the responding patients, 24 (22%) were able to proceed to ASCT in the entire cohort, including 3 (10%) in the VEN group and 21 (27%) in the non-VEN group (P = .07). The median age of patients who achieved a remission and did not proceed to ASCT was 73 years (range, 35-90 years) compared with 59 years (range, 20-69 years; P < .001) for those who were able to receive ASCT in first remission (CR1). Among responding patients aged ≥65 years, the rates of ASCT among patients treated with or without VEN were 2% and 4%, respectively (P = .4). There was no significant difference in OS among patients who underwent ASCT in CR1 compared with those who did not (median OS, 13.6 vs 12.5 months, respectively; P = .093) (Fig. 3).
The VAF of TP53 mutations has also been shown to predict for outcomes in AML.12 We had the baseline VAFs available for 202 patients (85%) in our study. The median VAF overall was 42.8%, with no significant difference between patients treated with or without VEN (Table 1). When analyzing patients who had TP53 VAFs <40% and ≥40%, there was no difference in survival between those who received therapy with or without VEN (see Supporting Fig. 1A,B). Among patients with a VAF <40%, the median OS was 3.9 versus 7.3 months (P = .16) for in those who received VEN-based or non–VEN-based therapy, respectively. Among those with a VAF ≥40%, the median OS was 5.7 versus 5.1 months, respectively (P = .77).
Multivariate Analysis
We performed a multivariate analysis (MVA) to investigate whether the addition of VEN influences the outcomes of patients with TP53mut AML. Several known risk factors were included in the model (age, white blood cell count, platelet count, bone marrow blasts, cytogenetic risk category). On univariate analysis, age (<65 vs >65 years; P = .02), cytogenetic risk category (adverse/intermediate vs diploid; P < .001), and platelet count (<20,000 vs >20,000/µL; P = .044) were significantly associated with inferior OS, and receipt of ASCT (P < .001) was associated with significantly improved OS. By MVA, only cytogenetic risk category, platelet count, and receipt of ASCT maintained significance (P < .001), regardless of age or platelet count as categorical or continuous variables. Neither the intensity of the therapy (intensive, lower intensity, or HMA) nor the use of VEN-based therapy influenced OS in patients with TP53mut AML (Table 4). Similarly, by MVA, the only factors independently associated with a response (CR/CRi) were cytogenetic risk category and platelet count. The addition of VEN-based therapy to the model did not influence response rates (Table 5).
| Variable | Overall Survival, N = 238 | P | |
|---|---|---|---|
| HR | 95% CI | ||
| Age: ≥65 vs <65 y | 1.11 | 0.81-1.52 | .52 |
| Cytogenetic category: Adverse/intermediate vs diploid | 2.92 | 1.59-5.37 | <.001 |
| Platelet count: ≥20,000 vs <20,000 × 103/μL | 1.79 | 1.32-2.42 | <.001 |
| Stem cell transplantation: Yes vs no | 0.4 | 0.24-0.67 | <.001 |
| VEN-based therapy: Yes vs no | 0.99 | 0.70-1.39 | .94 |
- Abbreviations: HR, hazard ratio; VEN, venetoclax.
| Variable | Multivariate Analysis: CR and CRi, N = 238 | P | |
|---|---|---|---|
| OR | 95% CI | ||
| Cytogenetic category: Adverse vs diploid/intermediate | 0.96 | 0.15-1.77 | .02 |
| Platelet count: ≥20 vs <20 × 103/μL | 0.57 | −0.004, 1.14 | .052 |
| VEN-based therapy: Yes vs no | 0.42 | −0.18, 1.03 | .17 |
- Abbreviations: CR, complete remission; CRi, complete remission with incomplete hematologic recovery; OR, odds ratio; VEN, venetoclax.
Discussion
Several groups have reported on the adverse outcomes of patients with newly diagnosed TP53mut AML.4, 9, 10 There has been optimism that the development of more effective therapies like the BCL2 inhibitor VEN may provide the opportunity for better outcomes in this difficult subset. Indeed, the addition of VEN to standard low-intensity backbones such as HMAs or LDAC have been associated with higher response rates across the cytogenetic and genomic risk groups of AML, including those with adverse risk.11, 14 In a phase 1 and 2 study of newly diagnosed older and/or unfit patients who received VEN combined with HMA, the CRc rate among those with adverse-risk cytogenetics was 60%, with a median CRd of 6.7 months and a median OS of 9.6 months; among those who had TP53mut AML, the CRc was 47%, with a median CRd of 5.6 months and a median OS of 7.2 months.14 Similarly, compared with non–VEN-based therapy, our data also demonstrated higher rates of CRc with the addition of VEN in patients with TP53mut AML. However, this did not translate into an improvement in RFS or OS compared with standard therapy without VEN. When stratified by TP53 VAF (<40% or >40%),VEN-based therapy did not demonstrate a difference in survival outcomes compared with non–VEN-based therapy, which is in line with the previously published report of the prognostic and therapeutic impact of TP53 VAF in patients with TP53mut AML.12 This lack of benefit with VEN-based therapy was observed across the subgroups regardless of age or type of therapy and regardless of the duration of HMA (ie, 5-day or 7-day course vs 10-day course) with VEN.
In a retrospective, single-center study reporting the outcomes of a VEN and HMA combination in 31 patients with TP53mut AML across both frontline (N = 13) and relapsed/refractory (N = 18) settings, the OS was 10.8 months, and the median leukemia-free survival was 7.6 months among responders (52%), of whom 31% underwent ASCT in CR. The investigators concluded that there was a trend toward better response rates with a VEN and HMA combination in the frontline setting.21 By comparison, we reported on 238 newly diagnosed patients with TP53mut AML and compared the response rates and long-term outcomes of those who received VEN-based and non–VEN-based regimens. Although we similarly observed an improvement in response rates with VEN, the OS was 5.7 months, RFS was 3.4 months, and only 10% of patients underwent ASCT after CR in the frontline setting. Our study represents the largest report comparing the outcomes of patients with newly diagnosed TP53mut AML treated with or without VEN-based treatment and also stratified by age.
Although response rates may have numerically improved with the addition of VEN to standard therapy, the high risk of relapse in patients with TP53mut AML remains an important challenge.22 As we observed from our study, VEN-based therapy did not improve RFS compared with non–VEN-based therapies in older or younger patients. In the MVA, adverse cytogenetic risk category and low platelet count (<20,000/µL) were strong, independent, negative prognostic factors for OS. This is in line with the previously published reports of poor outcomes in patients with TP53mut AML.23, 24 The addition of VEN was not associated with improved rates of CR/CRi by MVA. Only ASCT was associated with significantly improved OS (odds ratio, 0.4; 95% CI, (0.24-0.67; P < .001) and may provide an opportunity for better outcomes in a subset of patients with TP53mut AML.
In a retrospective study of 83 patients with a median age of 60 years who had TP53mut AML/MDS and underwent first ASCT, the 1-year OS rate was 35% in the entire cohort. Patients with fewer comorbidities, good performance status, and those in CR1 or CR2 benefited the most, with a 1-year OS rate of 67%.25 With higher response rates attributed to VEN-based treatment, we would anticipate the prospect of higher rates of ASCT in CR1. However, in our entire cohort of 238 patients with a median age of 69 years, only 24 (22%) underwent SCT, among whom a greater percentage of patients in the non–VEN-based cohort (27%) were able to proceed with ASCT in CR1 compared with those in the VEN-based cohort (9%), suggesting that the higher response rates did not directly translate into higher rates of ASCT. This difference is likely a function of younger age and a potentially more fit, ASCT-eligible population in the non–VEN-based cohort. Older age, short-lived responses, and ineligibility to receive ASCT may have contributed to the lower ASCT rates in patients treated with VEN-based therapy. Among responding patients who did receive ASCT, there was an improvement, although it was not statistically significant, in OS compared with those who did not proceed with ASCT.
In addition to relapse, early mortality, often defined at 4 and 8 weeks, is a major source of treatment failure in newly diagnosed AML. Although it is sometimes attributed only to treatment-related toxicity in a susceptible population, early mortality is as much a function of the effectiveness of the therapy at producing remissions as it is of death related to drug toxicity. With this in mind, there was no difference in 4-week or 8-week mortality with or without VEN-based therapy and when stratified by older and younger patients—suggesting the inadequacy of either approach to produce meaningful responses in this difficult population and underscoring the need for novel targeted therapies exploiting a p53-independent mechanism in TP53mut AML. Several novel agents targeted at improving the outcomes of TP53mut AML are currently in clinical development.
Eprenetapopt (formerly APR246), a PRIMA-1 (p53 reactivation and induction of massive apoptosis) analogue, restores the loss of function of mutant TP53 to induce a massive apoptotic response, and preclinical studies have shown the synergistic efficacy of eprenetapopt and 5-azacitidine combination in TP53mut AML/MDS.26 Although phase 2 study results of eprenetapopt and 5-azacitidine in combination for patients with TP53mut AML/MDS, (n = 55; MDS, n = 40;
AML, n = 11; MDS/myeloproliferative neoplasms, n = 4) reported an encouraging overall response rate of 71%, with a 44% CR rate and an improved OS in responding patients (14.6 vs 7.5 months; P = .0005)27; recent results of a randomized phase 3 study reported a CR rate of 33.3% in the eprenetapopt and 5-azacitidine combination arm compared with 22.4 % in the 5-azacitidine monotherapy arm (P = .13) among patients with TP53mut MDS (n = 154), and the study did not meet its primary end point of CR.28 Magrolimab, an anti-CD47 monoclonal antibody, engages the innate immune response to overcome the don't eat me signal of CD47-expressing leukemic cells and aids in leukemic cell death.29-31 Magrolimab in combination with 5-azacitidine has shown preliminary efficacy in patients with TP53mut AML32 (ClinicalTrials.gov identifier NCT03248479). Flotetuzumab, an investigational CD123 × CD3ε bispecific, dual-affinity retargeting (DART) antibody construct, exploits adaptive immunity to target and kill CD123-expressing leukemic cells,33, 34 and preliminary data suggest activity in patients with relapsed/refractory TP53mut AML35 (ClinicalTrials.gov identifier NCT02152956).
Biologic mechanisms of VEN resistance in TP53mut AML are currently under investigation. Several groups have reported that a functional TP53 gene, in its function as a transcriptional regulator of proapoptotic proteins, is essential for the activity of the BCL2 inhibitor VEN.36, 37 Apoptosis is a carefully orchestrated programmed cell death network that depends on the interactions between the pro-survival BCL-2 subfamily proteins (BCL-2, BCL-XL, and MCL1) and BH3-only proapoptotic proteins (PUMA, BID, and BIM).38 Antiapoptotic protein expression in AML is heterogeneous, and BCL2-XL, MCL1-XL, and BCL-XL are variably upregulated. Although the efficacy of VEN is reliant on BCL-2 expression, MCL-1 and BCL-XL upregulation is an important mechanism of VEN resistance. Preclinical studies have demonstrated the rationale of combining VEN—a BCL-2 inhibitor—with MCL-1 inhibition, either directly39 or indirectly through MEK inhibition, to overcome VEN resistance.40 Genome-wide CRISPR/Cas9 screening studies investigating TP53-knockout and the proapoptotic effector BAX-knockout AML cell lines confirm the importance of functional TP53 and BAX protein to produce an apoptotic response in AML cells exposed to VEN.36, 37 TP53 acts as a sensor of cellular stress and a transcriptional regulator of several proapoptotic genes.36, 37 Mutant or dysfunctional TP53 leads to altered mitochondrial homeostasis, protection from mitochondrial stress, and consequent diminished sensitivity to BCL2 inhibition.37 Restoring or replacing the transcriptional activity of mutant TP53 or upregulating proapoptic effectors may help restore sensitivity of the leukemic blasts to BCL2 inhibition. Therefore, whereas the addition of VEN to standard therapeutic backbones did not improve the outcomes in TP53mut AML, rational combinations of VEN with agents that modulate alternative apoptotic, transcriptional, or survival pathways may be a promising approach.
There are several limitations to our study, including a retrospective study design and heterogeneous treatments received based on age, comorbidities, and pretreatment karyotype. However, the treatment regimens were appropriately grouped, and this heterogeneity provided an opportunity to analyze the outcomes of VEN-based and non–VEN-based therapies according to the age and the type of therapy. Although it was not the aim of our study, we note that the response rates in patients with TP53mut AML receiving HMA-VEN were comparable to the rates in those receiving intensive chemotherapy, with similar OS. This observation generates the important question of whether intensive chemotherapy or HMA-VEN is the better choice in a younger, fit population. Although the current study was not designed to address this question, it provides some perspective for such an analysis. Furthermore, not all TP53 mutations are functionally equal (eg, gain vs loss of function),41 but a clear and comprehensive analysis on the impact of each type of mutation in AML therapy remains to be done. In its absence, we have reported the results of a highly annotated group of prospectively treated patients, organized by age and therapy, who had TP53 mutational testing with VAF performed at baseline.
In conclusion, we demonstrate that the incorporation of VEN into standard treatment regimens did not improve outcomes in patients with TP53mut AML relative to non–VEN-based therapies. Compared with those who have wild-type TP53, these patients exhibit lower response rates, and shorter CR durations regardless of age or type of chemotherapy. This study highlights the need for a better understanding of the mechanism of VEN resistance in TP53mut AML coupled with the urgent need for novel therapies to improve CR rates and survival in patients with TP53mut AML.
Funding Support
This work was supported in part by The University of Texas MD Anderson Cancer Center Support Grant (CA016672) from the National Cancer Institute.
Conflict of Interest Disclosures
Marina Konopleva reports grants and other support from AbbVie, F. Hoffman La-Roche, Stemline Therapeutics, Forty-Seven, Eli Lilly, Cellectis, Calithera, Ablynx, Agios, Ascentage, AstraZeneca, Reata Pharmaceutical, Rafael Pharmaceutical, Sanofi, Janssen, and Genentech outside the submitted work; has a patent (US 7,795,305 B2 CDDO-Compounds and Combination Therapy) with royalties paid to Reata Pharmaceutical, a patent (Combination Therapy With a Mutant IDH1 Inhibitor and a BCL-2) licensed to Eli Lilly, and a patent (62/993,166 Combination of a MCL-1 Inhibitor and Midostaurin, Uses and Pharmaceutical Compositions Thereof) pending to Novartis. Courtney D. Dinardo reports institutional research support from AbbVie, Agios, Calithera, Cleave, Bristol-Myers Squibb/Celgene, Daiichi-Sankyo, Forma, ImmuneOnc, and Loxo; and personal fees from AbbVie, Agios, Aprea, Celgene/Bristol-Myers Squibb, ImmuneOnc, Novartis, Takeda, and Notable Labs outside the submitted work. Michael Andreeff reports grants and research support from Daiichi-Sankyo, the Breast Cancer Research Foundation, United Therapeutics, ONO Pharmaceuticals, Karyopharm, The Cancer Prevention and Research Institute of Texas, the National Institutes of Health/National Cancer Institute, Amgen, and AstraZeneca; personal fees from Daiichi-Sankyo and Aptose; membership on advisory committees for the Centre for Drug Research & Development and Cancer UK; membership on review panels and board membership at the National Cancer Institute-Cancer Therapy Evaluation Program, the German Research Council, the Leukemia Lymphoma Foundation, the National Cancer Institute-Rare Diseases Clinical Research Network; and the Chronic Lymphocytic Leukemia Foundation; and has an ownership interest in Reata, Aptose, Eutropics, SentiBio, Oncoceutics, and Oncolyze all outside the submitted work. Naval Daver reports research funding from Daiichi-Sankyo, Bristol-Myers Squibb, Pfizer, Gilead, Sevier, Genentech, Astellas, Daiichi-Sankyo, AbbVie, Hanmi, Trovagene, FATE, Amgen, Novimmune, Glycomimetics, and ImmunoGen; and personal fees from Daiichi-Sankyo, Bristol-Myers Squibb, Pfizer, Novartis, Celgene, AbbVie, Astellas, Genentech, ImmunoGen, Servier, Syndax, Trillium, Gilead, Amgen, and Agios outside the submitted work. Naveen Pemmaraju reports research support from Novartis, Stemline Therapeutics, Samus Therapeutics, AbbVie, Cellectis, Affymetrix, Daiichi-Sankyo, and Plexxikon; personal fees from Pacylex Pharmaceuticals, ImmunoGen, Bristol-Myers Squibb, and Blueprint Medicines; grants from Affymetrix and the Sager Strong Foundation; travel support from Stemline Therapeutics, Celgene, MustangBio, DAVA Oncology, and AbbVie; and honoraria from Incyte, Novartis, LFB Biotechnologies, Stemline Therapeutics, Celgene, AbbVie, MustangBio, Roche Diagnostics, Blueprint Medicines, DAVA Oncology, and Springer Science + Business Media, LLC, all outside the submitted work. Koji Sasaki reports research funding and personal fees from Novartis, honoraria from Otsuka, and personal fees from Pfizer Japan outside the submitted work. Elizabeth J. Shpall reports personal fees from Bayer HealthCare Pharmaceuticals, Novartis, Magenta, Adaptimmune, Partner Therapeutics, Mesoblast, and Axio; license agreements or patents with Takeda (listed as a coinventor on a provisional patent application on Takeda owned by MD Anderson and licensed to Takeda); and honoraria from Magenta, Novartis, Partner Therapeutics, and Bayer HealthCare Pharmaceuticals all outside the submitted work. Guillermo Garcia-Manero reports grants or research support from Amphivena, Helsinn, Novartis, AbbVie, Bristol-Myers Squibb, Astex, Onconova, H3 Biomedicine, Merck, Curis, Janssen, Genentech, Forty Seven, and Aprea; and personal fees from Bristol-Myers Squibb, Astex, Helsinn, and Genentech outside the submitted work. Hagop M. Kantarjian reports research grants from AbbVie, Amgen, Ascentage, Bristol-Myers Squibb, Daiichi-Sankyo, Immunogen, Jazz, Pfizer, and Sanofi; and honoraria from AbbVie, Actinium, Adaptive Biotechnologies, Amgen, Aptitude Health, BioAscend, Daiichi-Sankyo, Delta Fly, Janssen Global, Novartis, Oxford Biomedical, Pfizer, and Takeda Oncology all outside the submitted work. Tapan M. Kadia reports research funding from AbbVie, Amgen, Bristol-Myers Squibb, Genentech, Jazz, Pfizer, Pulmotech, Cellenkos, Ascentage, Genfleet, Astellas, AstraZeneca, and Janssen; and personal fees or honoraria from AbbVie, Agios, Daiichi Sankyo, Genentech, Jazz, PinotBio, Novartis, Pfizer, Sanofi-Aventis, and Genzyme all outside the submitted work. The remaining authors made no disclosures.
Author Contributions
All persons who meet authorship criteria are listed as authors, and all authors certify that they have participated sufficiently in the work to take public responsibility for the content, including participation in the concept, design, analysis, writing, or revision of the article. Sangeetha Venugopal, Mahran Shoukier, and Tapan Kadia: Conception and design of study, acquisition of data, analysis and/or interpretation of data, drafting the article. Marina Konopleva, Courtney Dinardo, Farhad Ravandi, Nicholas J. Short, Michael Andreeff, Gautam Borthakur, Naval Daver, Naveen Pemmaraju, Koji Sasaki, Guillermo Montalban-Bravo, Guillermo Garcia-Manero, and Hagop Kantarjian: Acquisition of data, revising the article critically for important intellectual content, and approval of the final version for publication. Kayleigh Marx, Sherry Pierce, Uday R. Popat, Elizabeth J. Shpall, and Rashmi Kanagal-Shamanna: Acquisition of data, analysis and/or interpretation of data, revising the article critically for important intellectual content, and approval of the final version for publication.
