Limited Survival Benefit and High Healthcare Utilization With AML‐Directed Therapy in Older Patients With TP53 ‐Mutated Acute Myeloid Leukemia

To the Editor,

1.

Optimal management of TP53‐mutated acute myeloid leukemia (AML) and myelodysplastic neoplasms (MDS) remains an unmet clinical need. These malignancies predominantly affect older adults and are frequently associated with complex karyotypes, adverse‐risk cytogenetic features, and modest response to treatment leading to poor survival [1, 2, 3, 4]. Older patients are often suboptimal candidates for intensive chemotherapy and allogeneic stem cell transplantation (ASCT), both of which can be curative in particular disease subtypes, as well as clinical trial enrollment due to co‐morbidities, frailty, and limited performance status typically associated with age.

Hypomethylating agents (HMA) and novel therapies have expanded the treatment landscape for older patients and those unfit for high‐intensity chemotherapy. However, recent clinical trials studying the addition of venetoclax and other targeted therapeutics including eprenetapopt, sabatolimab, and magrolimab have not definitively shown survival benefit compared to HMA monotherapy in patients with TP53‐mutated AML or high‐risk MDS [5, 6, 7, 8]. In retrospective analyses, more intensive treatment approaches did not lead to improved survival in TP53‐mutated myeloid neoplasms [9, 10]. The American Society of Hematology (ASH) Clinical Practice Guidelines provide limited consensus recommendations for the management of TP53‐mutated disease in older adults [11]. We recently conducted an analysis of the National Cancer Database and demonstrated that a substantial proportion of older patients with AML did not receive any leukemia‐directed therapy [12]. In the absence of effective disease‐modifying therapies in TP53‐mutated AML and MDS, treatment may provide limited clinical benefit in older patients who are at higher risk for treatment‐related toxicity. This study seeks to assess whether disease‐directed therapy improves overall survival (OS) in patients 75 years and older with TP53‐mutated AML or MDS who were ineligible for curative‐intent ASCT and the effect of treatment on healthcare resource utilization including number of transfusions, inpatient days, and days on hospice.

We analyzed patients aged 75 years and older with TP53‐mutated AML and MDS diagnosed at our institution, an academic safety net hospital, between January 2014 and June 2025 using the UMass Leukemia Registry. OS was measured as days from initial bone marrow biopsy to death and compared using Kaplan–Meier analyses and log‐rank tests. Where the upper 95% confidence intervals (CIs) could not be estimated because they did not exceed a threshold of 0.5 during follow‐up, these values are reported as “not estimable” (NE). Mean number of transfusions, inpatient days (excluding initial diagnosis admission), and hospice days were compared between treated and untreated patients who were not candidates for ASCT. Trends of first‐line treatments were evaluated. Multihit state was determined using the International Consensus Classification definition.

Among 151 patients with TP53‐mutated AML or MDS, 50 (33.1%) were 75 years or older at diagnosis. Twenty‐two (44%) had AML and 28 (56%) had MDS. One patient was excluded from the analysis due to lack of survival data. Four patients (8.2%) underwent ASCT, and they were excluded from the analysis. Seven patients did not receive any cancer‐directed therapy and were managed with hospice or best supportive care (BSC), including five patients with AML and two patients with MDS (Table 1).

TABLE 1.

mOS for patients with TP53‐mutated AML and MDS who did not undergo ASCT by multihit state, treatment response, and initial treatment.

			N	mOS (months)	95% CI (months)	p
All patients			45	5.3	2.7–8.1	N/A
Sex	Male		22	2.9	1.5–15.3	0.2
	Female		23	7.2	3.1–19.3
Diagnosis	MDS		26	7.2	4.6–25.8	< 0.05
	AML		19	2.7	1.4–5.4
AML	Treated patients		14	3.0	1.4–NE	1.0
	Untreated patients		5	1.5	0.5–NE
MDS	Treated patients		24	9.1	4.6–25.8	< 0.05
	Untreated patients		2	0.3	0.1–NE
TP53 allelic state	Multihit		33	4.6	1.5–7.2	0.2
	Not multihit		11	9.1	2.5‐NE
Best response (AML only)	Complete remission a		6	5.4	3.0–NE	0.09
	Did not achieve complete remission		13	1.5	0.9–NE
Initial treatment	AML	Hospice	3	3.1	1.3–NE	< 0.05
		Best supportive care	2	1.4	0.5–NE
		HMA	3	1.1	0.9–NE
		HMA combination	8	3.4	1.3–NE
		Induction chemotherapy	3	8.1	1.5–NE
	MDS	Hospice	1	0.5	NE–NE	< 0.05
		Best supportive care	1	0.1	NE–NE
		ESA	2	N/A	NE–NE
		HMA	19	9.1	4.6‐NE
		HMA combination	2	8.2	7.2‐NE
		Induction chemotherapy	1	2.5	NE–NE

Abbreviations: AML, acute myeloid leukemia; ESA, erythropoiesis‐stimulating agent; HMA, hypomethylating agent; MDS, myelodysplastic syndrome; NE, not estimable.

^a Multihit state is unknown for one patient.

Patients with AML who received AML‐directed treatment did not have significantly longer median OS (mOS) compared to untreated patients (3.0 months, 95% CI: 1.4‐NE vs. 1.5 months, 95% CI: 0.5‐NE, p = 1.0) (Table 1 and Figure 1A). For patients with AML, 35.7% (5 of 14) died prior to a second bone marrow biopsy to assess for treatment response. Three out of five patients (60%) discontinued treatment after one cycle for reasons that included neutropenic fever, failure to thrive, and hospital delirium. Patients with MDS who received treatment had longer mOS (9.1 months, 95% CI: 4.6–25.8) compared to patients who did not receive treatment (0.3 months, 95% CI: 0.1‐NE, p < 0.05) (Table 1 and Figure 1A). Patients whose disease was multihit TP53 had shorter mOS compared to patients whose disease was not multihit; however, this difference was not statistically significant (Table 1).

FIGURE 1.

(A) Kaplan–Meier curve for probability of survival by diagnosis and whether patients underwent treatment. Frequency of first line treatments in 2014–2017 and 2018–2025 for patients with TP53‐mutated AML (B) and MDS (C).

Patients with AML who received AML‐directed treatment had higher healthcare utilization including more transfusions following diagnosis (24.5 vs. 2.4 units, p < 0.05) and more days in the hospital (20.6 vs. 1.6 days, p < 0.05) compared to patients who did not receive treatment (Table 2). Patients who were treated with HMA alone spent a mean of 12.4 days (95% CI: 7.5–17.4) in the hospital compared to 25.7 days (95% CI: 9.8–41.6) for patients treated with HMA combination therapy (p = 0.10) (Table 2). Hospice days averaged 4.2 days for treated patients with AML compared to 41.0 days for patients who did not receive leukemia‐directed treatment, although this difference was not statistically significant (p = 0.18) (Table 2). Patients who were treated with HMA and venetoclax spent 26.6 days in the hospital following their index hospitalization compared to 12.4 days for patients treated with single agent HMA (p = 0.15) (Table 2). Patients treated with HMA and venetoclax also had a higher number of total transfusions (30.6 vs. 22.9, p = 0.41) (Table 2). However, difference in healthcare utilization between patients treated with HMA and venetoclax versus HMA monotherapy was not statistically significant.

TABLE 2.

Mean number of transfusions, days in the hospital, and time on hospice for patients with TP53‐mutated AML and MDS who were not candidates for ASCT.

	All patients	Treated	Untreated	p
AML
Number of transfusions	19.7	24.5	2.4	< 0.05
Number of days in the hospital	15.6	20.6	1.6	< 0.05
Number of days on hospice	12.7	4.2	41.0	0.18
MDS
Number of transfusions	21.3	23.0	1.5	< 0.05
Number of days in the hospital	12.2	13.1	1.5	< 0.05
Number of days on hospice	15.3	15.4	14	N/A

Abbreviations: AML, acute myeloid leukemia; MDS, myelodysplastic syndrome.

Prior to 2018, four of six (66.7%) patients who were treated received induction chemotherapy (Figure 1B,C). Beginning in 2018, HMA‐based therapies predominated (34 of 37 patients, 91.9%) (Figure 1B,C). Most patients receiving HMA‐based therapy for AML also received venetoclax (71.4%) (Figure 1C). HMA/venetoclax was used 5 of 10 (50.0%) patients treated in 2024–2025. In AML, mOS was 8.1 months (95% CI: 1.5‐NE) with induction chemotherapy, 1.1 months (95% CI: 0.9‐NE) with HMA alone, 3.4 months (95% CI: 1.3‐NE) with HMA combination therapy, and 1.4 months (95% CI: 0.5‐NE) with BSC. In MDS, mOS was 9.1 months (95% CI: 4.6‐NE) with HMA alone, and 0.1 months (95% CI: 0.1‐NE) with BSC. Two patients with MDS were treated with erythropoiesis stimulating agents (ESAs) due to low‐risk disease by Molecular International Prognostic Scoring System (IPSS‐M) despite TP53 mutation. These patients were stable on treatment for over 4 years at the time of data cutoff.

In our study, AML‐directed therapies did not prolong survival for older patients with TP53‐mutated AML who did not undergo ASCT and were linked to increased healthcare burden compared to patients who did not receive treatment. The benefit of venetoclax in TP53‐mutated myeloid neoplasms has been questioned [5, 13, 14, 15, 16]. Furthermore, the addition of venetoclax increases the risk of adverse effects, particularly cytopenias and their clinical sequelae [5, 15]. This may have contributed to more days in the hospital in patients treated with HMA and venetoclax compared to HMA monotherapy in our study. In the absence of demonstrated efficacy in TP53‐mutated AML, it is imperative to consider the addition of venetoclax in older patients with caution. Venetoclax may help with reduction of blast burden, though the value of this is mostly restricted to ASCT‐eligible patients who are being considered for expedited ASCT [15].

Patients who were treated for TP53‐mutated MDS showed survival benefits compared to untreated patients. MDS is a heterogeneous disease, and patients with low‐risk disease, even in the presence of high‐risk molecular features, may respond favorably to ESA therapy. Prior studies have shown that the prognoses for TP53‐mutated AML and MDS with excess blasts are comparable [2, 17]. Our study included patients with TP53‐mutated MDS without excess blasts, and this likely explains the difference in survival outcomes between patients with TP53‐mutated myeloid neoplasms.

The high proportion of patients in this study who did not have subsequent disease assessment with bone marrow biopsy highlights the frailty of this patient population and underscores the importance of nuanced assessments of overall health and functional status in treatment decisions. While HMA/venetoclax is considered a lower intensity intervention, it may carry significant toxicity in real world practice [18]. For patients 75 years and older with TP53‐mutated AML who are not candidates for ASCT, BSC may be appropriate. In our cohort, hospice was underutilized. Our data attests to the value of early integration of Palliative Care services, as this may afford high‐value care on a patient‐by‐patient basis while minimizing global healthcare costs for patients with TP53‐mutant MDS or AML [4, 19, 20].

This study is limited by its single institution setting and small sample size. More detailed analyses, such as survival analysis by multihit state for MDS and AML, could not be individually conducted. We acknowledge the heterogeneity of patients with MDS, and effect of blast enumeration on outcomes was not evaluated. We did not have the ability to perform single cell sequencing and could not definitively qualify allelic state by its strict definition. Although some findings did not reach statistical significance, the results support the hypothesis that older patients with TP53‐mutated AML who are ineligible for ASCT may be more appropriate for hospice rather than AML‐directed therapy if no appropriate clinical trial is available. Additional research is needed to evaluate the benefits of disease‐directed treatment for older patients with intermediate‐risk AML and to identify which older patients are most likely to benefit from therapy. Our results are hypothesis‐generating for future multicenter studies on the topic of healthcare utilization in TP53‐mutant myeloid neoplasms, and we hope to inspire collaborative science that will help lead to improved outcomes for this highest risk subset [15].

Author Contributions

Tina Y. Zhang: conception/design, acquisition of data, data analysis and interpretation, manuscript writing, final approval of manuscript. Alex Y. Ge: data analysis, manuscript writing, final approval of manuscript. Lloyd Hutchinson: acquisition of data, final approval of manuscript. Anne W. Higgins: acquisition of data, final approval of manuscript. Patricia M. Miron: acquisition of data, final approval of manuscript. Rory M. Shallis: critical review and interpretation. Shyam A. Patel: conception/design, manuscript writing, final approval of manuscript.

Funding

The authors have nothing to report.

Disclosure

This study was presented as a poster at the 67th American Society of Hematology Annual Meeting on December 6, 2025 in Orlando, FL.

Ethics Statement

This study was approved by the University of Massachusetts Institutional Review Board under general study protocol (IRB#: H00019393) for the retrospective analysis of patients with hematologic malignancies.

Consent

The authors have nothing to report.

Conflicts of Interest

S.A.P. served on Advisory Boards for Bristol Myers Squibb, Syndax, Taiho, and Geron. S.A.P. received travel support for the 2024 AAMDSIF Conference and funding from the Department of Medicine for the UMass CHIP Clinic. R.M.S. has served in a consulting/advisory role and/or received honoraria from Bristol Myers Squibb, Cardinal Health, Geron, Gilead Sciences Inc, Kura Oncology, Rigel, Servier, Syndax Pharmaceuticals, and TScan Therapeutics and has participated in a Steering Committee for Servier. Other authors declare no competing financial interests.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.