A cure for the kiss of death? Open Access

In this issue of Blood Advances, Lontos¹ et al provide a comprehensive report of the survival outcomes among 240 patients who underwent a transplant at the MD Anderson Cancer Center (MDACC) for acute myeloid leukemia (AML) or myelodysplastic syndrome (MDS) with TP53 mutations (TP53^mut) over a 10-year period. Regardless of the intervention, TP53^mut myeloid malignancies have such poor outcomes that some have called this mutation a kiss of death. By discovering clinical and disease characteristics that distinguish patients with satisfactory transplant outcomes from those who suffer dismal results, the authors increased our confidence in selecting the patients with TP53^mut MDS or AML who are most likely to benefit from allogeneic hematopoietic cell transplant (allo-HCT).

The rarity of long-term survival has prompted some authors to suggest that HCT for TP53^mut AML or MDS may be futile. In contrast, HCT is the most potent available antimyeloid therapy, and even rare success may justify risky endeavors. Nonetheless, for patients who suffer from MDS or AML with a TP53^mut, the utility of HCT may not be universal. A previous European Bone Marrow Transplant (EBMT) registry report of 179 patients with TP53^mut disease found that the 2-year overall survival (OS) after HCT was 65% among those without complex karyotype (CK) or loss of 17p; however, it decreased to 24% among those with either of these abnormalities, suggesting that allo-HCT can be a curative modality.² In patients with MDS with TP53 genetic defects, the reported survival 2 years after HCT is 19%.³ 

An optimal approach would reserve the resource-intensive and potential toxic HCT procedure for patients who are moderately likely to be cured.³ Lontos et al described the entire set of patients who received an allograft for AML or MDS with TP53^mut between 2012 and 2022 at the MDACC; although stringent selection criteria were undoubtedly in place to ensure transplant eligibility, this cohort was large and comprehensive. The adverse risk for these patients was clear; 71% showed CK and 30% had multiple TP53 mutations detected. In the entire cohort, the 1-year OS was 41%. The similarity in the 2-year OS that was observed in the MDACC and EBMT cohorts (24% and 29%, respectively) is reassuring.

Further evaluation suggested that baseline factors, which are often viewed as additional challenges to posttransplant success, may not have adversely impacted the outcomes. Specifically, the most common disease status (53% of patients) at the time of transplant was primary induction failure; furthermore, pretransplant TP53^mut persistence was common. Surprisingly, progression free survival (PFS) was not impacted by age, diagnosis (AML vs MDS), detectable TP53^mut before transplant, the number of TP53 mutations, nor the HCT comorbidity index.

The authors next attempted to identify subgroups with favorable and adverse post-HCT results. The authors selected the classification and regression tree (CART) analysis,⁴ a method well suited to identify predictors of PFS outcomes even when the baseline covariates are highly correlated.⁵ CART identified a favorable subgroup (TP53 variant allele frequency [VAF] <50% and absence of CK, del5q, or del7q; n = 39 patients) with a remarkable 2-year PFS of 60%. In the CART-defined intermediate risk group (TP53 VAF <50% with any of CK, del5q, or del 7q), ∼15% of patients had sustained PFS, revealed by the unusually long follow-up in the present study. Future efforts could focus on augmenting the efficacy of HCT for these patients with post-HCT maintenance therapy or other strategies. The final group, patients with a TP53 VAF >50%, had dismal outcomes with 2-year PFS <5%. Overall, these data refine the conclusion that HCT may be an opportunity for longer survival of patients with AML/MDS with TP53^mut at lower VAFs who are transplant candidates.

Comparing transplants against nontransplant alternatives for TP53^mut disease is extremely challenging, and strong provider and patient opinions may presently preclude a randomized trial.⁶ In the Bone Marrow Transplant Clinical Trials Network 1102 study,⁷ a donor with no-donor comparison found that patients with TP53^mut who were not transplanted had reduced OS (hazard ratio, 3.89; confidence interval, 1.87-8.12), and a multivariable analysis suggested that the relative benefit of transplant was even greater for patients with multi-hit TP53^mut (defined as VAF >50%, the presence of multiple TP53 mutations, 17p deletion, or CK). A recent retrospective report of patients with TP53^mut MDS/AML who underwent transplant and those who did not and who were treated at the MDACC⁸ suggests that HCT is associated with improved survival, an observation potentially confounded by patient selection. For these highest risk patients, substantial uncertainty remains.

The MDACC HCT experience for patients with TP53^mut MDS or AML is consistent with 2 potential conclusions. First, there is no direct evidence that the activity of HCT is improved by forcing patients into a deep remission. Elimination of measurable residual disease (MRD) by pretransplant therapies may only identify cases of treatable AML or MDS.⁹ The many studies in which improved outcomes were observed among patients who were MRD negative may simply reflect this fact.⁹ A recent decision analysis suggested that, for patients with TP53^mut MDS, an immediate transplant optimized surival.¹⁰ The lack of demonstrated additional benefit of pretransplant therapies over transplant alone should be considered by patients who are planning HCT before they accept the delays and risk of toxicity that follow exposure to pretransplant chemotherapy.

A second opinion is that TP53^mut AML and MDS is a heterogenous entity. Although most TP53^mut AML and MDS are undeniably bad actors, patients with a TP53 VAF <50% and no CK may be cured with HCT (even if MRD positive). Unfortunately, our current clinical assessments are limited and homogenous assessment of TP53 VAF is at best an imperfect surrogate for relapse potential.

What may be needed is an assay that detects the emergence of clonal states that have acquired the quartet of genomic instability, proliferative capacity, chemotherapy resistance, and evasion of immune surveillance. Leukemic (or preleukemic) clones in TP53^mut disease may transition to a state that is licensed to acquire additional cytogenetic abnormalities, which then coincides with the 4 traits of lethality listed earlier. Preventing this state or understanding its vulnerabilities may direct future therapies. The data from the study by Lontos et al are a step forward in our selection of those who might benefit most from HCT. For most patients with TP53^mut AML or MDS, HCT remains the only hope for cure from what historically has been the kiss of death.

Conflict-of-interest disclosure: The authors declare no competing financial interests.