Double, double, TP53, and trouble in myelofibrosis

In this issue of Blood, Gagelmann and colleagues provide novel insights into the impact of TP53 mutational status on outcomes following allogeneic hematopoietic stem cell transplantation (allo-HSCT) for myelofibrosis (MF).¹ Their observations, which demonstrate that the multi-hit TP53 mutational state predicts adverse outcome following allo-HSCT, are of real practical relevance to transplant physicians counselling prospective recipients. One of their important findings is the identification of a subgroup of patients who have a particularly poor prognosis in spite of allo-HSCT, for whom the MF research community should consider ways to improve the efficacy of this procedure.

The adverse prognostic impact of TP53 mutational status in patients with diseases such as acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS) has been recognized increasingly in recent years.^2-4 Indeed, elegant work from Bernard et al highlights the important prognostic distinction between monoallelic TP53 abnormality and the multi-hit state (see figure) at diagnosis in patients with MDS.² In that study, patients with monoallelic TP53 mutated (TP53mut) MDS were observed to experience survival and treatment outcomes similar to those of patients with wild-type. This result was in stark contrast to the dismal prognosis of patients with multi-hit TP53 abnormalities.² 

In the context of MF, no reports have been made, until now, on the impact of TP53 mutational status at the time of allogeneic-HSCT on survival outcomes. In addition, current molecular prognostic scoring systems (such as the mutation-enhanced international prognostic score system for transplantation-age patients with primary myelofibrosis [MIPSS70]⁵ and the genetically inspired prognostic scoring system for primary myelofibrosis [GIPSS])⁶ consider the presence of high-risk molecular mutations (ASXL1, EZH2, SRSF2, IDH1/2, and U2AF1) and/or adverse karyotypic abnormalities (complex, with/without abnormalities of 17p) at the time of diagnosis but not specifically the presence of a monoallelic vs multi-hit TP53 state for prognostication.

The presence of TP53 abnormalities at diagnosis in patients with myeloproliferative neoplasia (MPN) has been shown previously to both increase the risk of leukemic transformation and shorten overall survival.⁷ The retrospective study presented here, the result of an international multicenter collaborative effort, demonstrates that patients with monoallelic TP53mut MF have an outcome following allo-HSCT comparable to that of patients with TP53 wild-type (TP53WT) MF (6-year survival: 56% vs 64%, respectively). Patients with TP53 multi-hit (TP53MH), however, fared particularly poorly following a transplant, with only 25% survival at 6 years. Adverse outcomes in patients with TP53MH were predominantly due to disease relapse posttransplant, with the cumulative incidence of relapse being 52% in such patients, compared to 17% for single-hit and 21% for patients with TP53WT, respectively. Leukemic transformation was a more frequent relapse presentation in patients with TP53MH, compared with patients with monoallelic TP53mut/TP53WT. As expected, TP53mut was frequently observed in association with a complex karyotype. Patients with TP53mut were less likely to have received ruxolitinib pre-HSCT, suggesting early recourse to allo-HSCT, presumably driven by physician recommendation, based on predicted adverse prognosis of TP53mut. Conditioning intensity did not have an impact on outcomes for patients, echoing previous findings in not only patients with MF⁸ but also those with other myeloid malignancies.⁴ 

Once again, the transplant community faces a formidable adversary in the form of TP53MH myeloid disease. The observation of an increased relapse risk (and higher likelihood of presentation with leukemic transformation at relapse) in patients with TP53MH MF highlights a need for an alternative approach to allo-HSCT in this subgroup, in contrast to the approach used for those with monoallelic TP53mut/WT disease.

The best way to manage such ultra-high-risk patients remains unclear, although allo-HSCT is still likely to offer the greatest curative potential. However, whether alternative pretransplant, conditioning, or posttransplant maintenance strategies (alone or in combination) could improve post allo-HSCT outcomes in these patients requires further research. TP53-directed agents, such as eprenetapopt, have shown promise in early-phase trials assessing the impact of posttransplant maintenance in combination with azacitidine in the context of MDS/AML.⁹ We do not know currently whether TP53-targeting agents similarly could have efficacy in the context of allo-HSCT for MF. Indeed, whether pretransplant therapy employing such agents has added benefit, vs moving immediately to transplant with early institution of posttransplant maintenance strategies, is unknown. Important to note is that patients with MF frequently experience prolonged poor graft function within the first year post-HSCT, often due to persistent bone marrow fibrosis. Consequently, delivery of posttransplant maintenance may be hampered by worsening cytopenias but warrants prospective exploration.

Alternative agents that have shown promising activity in the context of TP53mut AML/MDS include the macrophage checkpoint inhibitor magrolimab, which targets the anti-CD47/signal-regulatory protein α (SIRPα) axis¹⁰; whether its effects extend to other TP53mut myeloid diseases, including MF, is a question worthy of consideration. Current preemptive relapse strategies include the use of donor lymphocyte infusion (DLI), usually to correct increasing mixed-donor chimerism as a marker of incipient relapse. A notable point is that more than 70% of the patients with TP53mut in this cohort demonstrated co-mutations in either JAK2 or CALR. At present, monitoring of these markers and/or TP53 itself is insufficiently standardized to be adopted as a routine measurable residual disease (MRD) approach post-HSCT. The utility of MRD detection in this ultra-high-risk group, and whether a sufficient window of opportunity is present to allow intervention prior to full-blown relapse (for example, with drugs with or without DLI), are other potentially interesting avenues to explore.

Undoubtedly, the important findings presented in this issue of Blood highlight the pressing need for further research in this particularly difficult-to-treat patient subgroup and should galvanize the MPN and transplant communities to change the outlook for these patients by developing and exploring novel treatment approaches.

Conflict-of-interest disclosure: P.K. reports editorial collaboration with and serving as a speaker for Astellas; serving as an advisory board member for Jazz; serving as an advisory board member for and receiving conference support from Gilead; and serving as a speaker for Stemline Menarini and AbbVie.