Dear Editor,

TP53, also known as the guardian of the genome, is a tumor suppressor gene whose pathogenic mutations are associated with poor outcomes across all tumor types. Its incidence in de novo acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS) ranges from 5 to 18% and increases to as high as 30–35% in therapy-related myeloid neoplasms [1, 2]. Despite advancements in the treatment of high-risk myeloid malignancies, TP53-mutated myeloid neoplasms (TP53mt-MN) remain a therapeutic challenge due to the “undruggable” nature of the gene. Outcomes remain dismal due to poor responsiveness to therapy, a high risk of relapse and short survival [3, 4]. Due to their unique biology and poor outcomes, TP53mt-MN are increasingly being recognized as distinct clinical entities [5, 6].

The initial response rate and median survival for TP53mt AML is approximately 20–50% and 6 months, respectively [3, 7, 8]. Allogeneic hematopoietic stem cell transplantation (allo-HSCT) can potentially cure 20% of these patients depending on the disease and transplant characteristics [9]. However, due to the high risk of relapse, its role as a consolidative option in TP53mt disease remains controversial.

Due to the poor outcomes, there has been a recent focus on questioning the benefit of disease-directed treatment in older patients with TP53mt-MN. Many physicians may consider that the toxicity of disease-directed therapy may outweigh potential benefit in older patients, particularly since curative intent allo-HSCT is not routinely offered. Within this context, we utilized a multi-institutional real-world database to evaluate the treatment outcomes of older patients with TP53mt-MN (“Methods” in Supplementary).

We identified 515 patients, ≥60 years old, with TP53mt-MN (Supplementary Fig. 1). In total, 451 patients received disease-directed therapy and were included in the final analysis (Table 1). Most patients had de novo AML (n = 202) followed by clinically defined secondary AML (sAML, n = 171), myeloproliferative neoplasm blast phase (MPN-BP, n = 50), and MDS excess blasts 2 (MDS-EB2, n = 28). The median age was 70 years (range 60–90 years), and 47% were ≥70 years. The proportion of patients with multi-hit TP53mt and complex cytogenetics (CG) was 86% and 83%, respectively (Supplementary Fig. 2).

Table 1 Baseline characteristics and response to therapy of the entire cohort.
Full size table

Treatment consisted of hypomethylating agent + venetoclax (HMA + VEN), intensive chemotherapy, HMA-based therapy (excluding VEN), and other low-intensity therapy in 43%, 31%, 20%, and 6% of the patients, respectively. Fifty-three (12%) patients proceeded to allo-HSCT with 83% having <5% blasts at the time of transplant. The most common reasons for not pursuing allo-HSCT were inadequate disease response to therapy (62%), poor performance status/comorbidities (41%), advanced age (24%), and socioeconomic barriers (19%).

The composite complete remission [cCR = CR + CR with incomplete count recovery (CRi)] rate amongst evaluable patients (n = 356) was 34%. The cCR by treatment was 40%, 25%, 22%, and 12.5% for HMA + VEN, intensive chemotherapy, HMA-based therapy (excluding VEN), and other low-intensity therapy, respectively (p = 0.01). Ninety-five patients were unevaluable for response assessment due to death during induction or referral to hospice. The median duration of response was 6.7 months (range, 5.9–7.5 months).

The median EFS (mEFS) and OS (mOS) for the entire cohort were 2.9 months (range, 2.2–3.5 months) and 6.9 months (range, 6.1–7.6 months), respectively (Fig. 1A, B). In transplanted patients, the mOS from the time of transplant was 21.9 months with 3-year OS of 61%. (Fig. 1C). The mOS by disease group was 16.4 months, 6.6 months, 6.7 months and 5.2 months in MDS-EB2, de novo AML, sAML, and MPN-BP, respectively (Supplementary Fig. 3A, p = 0.0004). The mOS by age was 7.6 months, 6.9 months, and 6.3 months in 60–69 years, 70–79 years and ≥80 years, respectively (Supplementary Fig. 3B, p = 0.006). The mOS based on response was 13.3 months, 14.6 months, and 5.2 months for CR, CRi, and no CR/CRi, respectively (Supplementary Fig. 3C, p < 0.0001). Predictors of outcome are shown in Supplementary Table 1.

Fig. 1: Survival in older patients with TP53-mutated AML.
figure 1

A Kaplan–Meier curve showing event-free survival of all patients, B overall survival of all patients, and C overall survival of the subgroup who underwent allogeneic hematopoietic cell transplantation.

Full size image

MDS-EB2 vs. other myeloid neoplasms (HR: 0.56, 95% CI: 0.34–0.94, p = 0.02), cCR (HR: 0.31, 95% CI: 0.23–0.41, p = <0.001), and allo-HSCT (HR: 0.32, 95% CI: 0.21–0.49, p = <0.001) were associated with improved EFS. Multi-hit TP53mt (HR: 1.62, 95% CI: 1.11–2.39, p = 0.01) was associated with worse EFS in MVA. MDS-EB2 vs. other myeloid neoplasms (HR: 0.45, 95% CI: 0.24–0.84, p = 0.01), cCR (HR: 0.51, 95% CI: 0.38–0.68, p = <0.001), and allo-HSCT (HR: 0.31, 95% CI: 0.19–0.49, p = <0.001) were associated with improved OS. Age ≥70 years (HR: 1.34, 95% CI: 1.02–1.76, p = 0.03) and multi-hit TP53mt (HR: 1.96, 95% CI: 1.26–3.06, p = 0.003) were associated with worse OS in MVA.

In our study of older patients with high-risk TP53mt-MN, we report that disease-directed therapy results in modest responses and survival outcomes. We find that responses to treatment and allo-HSCT are independent predictors of outcomes. Lastly, a small proportion of patients were able to proceed to allo-HSCT and that allo-HSCT substantially improved outcomes for this high-risk population.

Irrespective of the type of induction chemotherapy (intensive vs. non-intensive), response rates in TP53mt-MN are low and often short-lived with a mOS of approximately 6 months [7, 8, 10]. Recent data have demonstrated that intensive induction with the traditional “7 + 3” leads to a low response and that alternative options based on the biology of TP53, instead of the fitness of the patient, should be explored [3, 8]. In the pivotal VIALE-A trial, the combination of HMA + VEN, when compared to HMA alone, resulted in a higher response rate, but that did not translate into a survival benefit in TP53mt AML [7]. Since then, several studies have demonstrated that in the absence of allo-HSCT, there is no difference in outcomes for TP53mt AML based on the type of induction used. Many centers are now administering HMA alone for transplant-ineligible patients with TP53mt AML [8].

The treatment of TP53-mutated disease has been immensely challenging. Investigational agents targeting this subset have largely been unsuccessful in randomized trials despite early promise [11, 12]. In MDS, the p53 reactivator, eprenetapopt, combined with HMA did not improve outcomes when compared to HMA alone [13]. The ENHANCE-2 trial investigating the combination of the anti-CD47 monoclonal antibody, magrolimab, with HMA did not improve the survival of TP53mt AML [1]. These studies did validate that the outcomes of TP53-mutated disease continue to remain poor regardless of the type of disease, age or intensity of therapy.

In our cohort of older patients, 12% proceeded to allo-HSCT. Several studies have shown that the rate of transplant in TP53mt-MN is low (3-25%) regardless of age and an added reason for this may include the increased risk of post-transplant relapse. The rate of transplant for TP53mt-MN has ranged from 3 to 25% based on results from both clinical trials and large registries focusing on younger and older patients [1, 8, 14]. Despite this concern, several studies have demonstrated that allo-HSCT is the only treatment modality that can prolong survival for this disease. In the BMTCTN trial for patients with MDS, those with a TP53 mutation undergoing allo-HSCT had improved survival compared with non-transplant therapy (3 years OS: 23% vs. 11%; P = 0.04) [14]. Our post-transplant outcomes of median survival close to 2 years are in line or somewhat better compared with previously published data. Baranwal et al. reported a median OS of 1.03 years in a multicenter study with 64.2% of the recipients having any disease present at the time of transplant and 8.2% having >5% blasts, which was an independent predictor of poor survival [9]. Lontos et al. reported a 2-year OS of 29% in a single-center analysis where only 39% of the patients were in remission at the time of transplant and identified VAF and CGs to be independent predictors of outcomes [15]. The reason for our outcomes being different from these two reports is not entirely clear but one explanation could be that the majority of patients in our analysis were in remission at the time of transplant and we did identify cCR as an independent predictor of outcomes. Further studies to validate the difference in these outcomes are warranted. Nonetheless, these findings reinforce the need for a transplant consultation and careful weighing of risks and benefits and the goals for individual patients, including older patients with TP53mt-MN.

We acknowledge the limitations of our study with its inherent selection bias and inter-institutional variability. Analysis for LOH was not available for patients with a single TP53 mutation and a VAF > 10% but <50% so a CK or a karyotypically apparent 17p deletion was used to define MH-TP53 per ICC 2022. An ideal comparison would have also involved a cohort of patients treated with best supportive care. However, outside the setting of a clinical trial, such a comparison would be significantly biased.

TP53mt-MN constitutes a distinct entity with an aggressive biology associated with poor outcomes. These poor outcomes are applicable to all patients regardless of age. Our large, real-world, multi-center analysis, focusing on only older patients with TP53mt-MN, replicates the modest outcomes seen with this disease across the age-spectrum and reiterates the potential benefit of allo-HSCT in a small subset of patients. The decision to treat older patients with TP53mt-MN should be based on individual patient characteristics and desires. Lastly, like several other studies, we continue to highlight the dire need for novel therapeutics for this disease.