Latest insights on TP53 mutations in AML and MDS from ASH 2022

TP53 mutations (TP53^mut) are associated with an adverse prognosis and treatment resistance in patients with acute myeloid leukemia (AML) and myelodysplastic syndromes (MDS).^1,2 There is currently a lack of treatments that specifically target TP53^mut and the most effective treatment strategy for patients with TP53^mut remains unclear.^1,3 The allelic state of TP53^mut may also have prognostic implications in patients with AML and MDS.^4,5

Several presentations at the 64th American Society of Hematology (ASH) Annual Meeting and Exposition discussed the latest treatment options, key clinical and genetic factors of patients with TP53^mut, and the impact of these factors on survival outcomes. Below, we summarize presentations by Daver,¹ Garcia-Manero,² Badar,³ Bahaj,⁴ and Meggendorfer⁵ on TP53^mut in AML and MDS.

Treatment options

Clinical outcomes of current standard of care¹

Study design and patient characteristics

This study included adult patients from the Flatiron Health database with newly diagnosed AML between 2014 and 2021 and with confirmed TP53^mut, chromosome 17p deletions (17p del), or both. The objective of the study was to assess the effectiveness of first-line intensive chemotherapy (IC), hypomethylating agents (HMA), and venetoclax with HMA in patients with AML with TP53^mut/17p del. In total, 370 patients were included, of which 94 received venetoclax and HMA, 135 received IC, and 141 received HMA alone. The majority of the patient population presented an adverse cytogenetic profile and there were low rates (8%) of patients receiving allogeneic hematopoietic stem cell transplantation (allo-HSCT).

Key findings

Bone marrow blast clearance after first-line treatment was reported in 54% of patients overall, 67% for patients who received venetoclax and HMA, 62% for patients who received IC, and 19% for patients who received HMA only. Overall survival (OS) outcomes for the three treatment cohorts are displayed in Table 1.

Table 1. OS by treatment type*

CI, confidence interval; HMA, hypomethylating agents; IC, intensive chemotherapy; OS, overall survival; Q1, first quartile; Q3, third quartile. *Adapted from Daver.1
Treatment received	Median OS, months (95% CI)	Median follow-up, months (Q1–Q3)
Venetoclax + HMA	7.4 (6.0–8.8)	6 (3–9)
IC	9.4 (7.2–10.4)	7 (4–13)
HMA	5.9 (4.3–7.5)	5 (2–11)

Following multivariate analysis, OS was associated with:

• age (adjusted hazard ratio [aHR], 1.02; 95% CI, 1.01–1.04);
• time from diagnosis to first-line therapy initiation (aHR, 0.98; 95% CI, 0.96–0.99);
• number of comorbidities: 0 vs ≥2 (aHR, 0.63; 95% CI, 0.47–0.86);
• Eastern Cooperative Oncology Group (ECOG): 0–1 vs ≥2 (aHR, 0.59; 95% CI, 0.40–0.86); and
• treatment with allo-HSCT (aHR, 0.35; 95% CI, 0.19–0.63).

When comparing the therapies received:

• venetoclax and HMA did not significantly improve OS when compared with HMA alone (aHR, 0.93; 95% CI, 0.67–1.29) or IC (aHR, 1.00; 95% CI, 0.67–1.50); and
• IC did not improve OS compared with HMA alone (aHR, 1.08; 95% CI, 0.76–1.55).

This large cohort study confirmed that frontline standard of care options for patients with TP53^mut/17p del AML are currently correlated with overall limited survival outcomes.

Oral decitabine and cedazuridine²

Study design and patient characteristics

The phase III ASCERTAIN trial included 133 patients with intermediate- or high-risk MDS/chronic myelomonocytic leukemia. The objective of the study was to determine OS in this patient population with regards to TP53 mutational status assessed by next-generation sequencing (NGS). Patients were randomized 1:1 to receive either:

• Cycle 1 of oral decitabine and cedazuridine followed by Cycle 2 of intravenous decitabine; or

• Cycle 1 of intravenous decitabine followed by Cycle 2 of oral decitabine and cedazuridine.

Both cohorts then received oral decitabine and cedazuridine for Cycle 3 onwards. Of the 133 patients included in the study, 125 were evaluable for genetic analysis. Patient samples were analyzed using NGS, with 44 patients carrying TP53^mut.

Key findings

• As shown in Table 2, patients with TP53^mut had worse OS (HR, 1.7; 95% CI, 1.00–2.87), as well as leukemia-free survival (HR, 1.63; 95% CI; 0.98–2.72), when compared with patients with wild-type TP53 (TP53^wt).
• Patients with TP53^mut were further analyzed by allelic state, with 30 patients classified as monoallelic and 14 classified as biallelic (defined as more than one TP53 copy or 17p deletion with at least one TP53 mutation).

Table 2. Overall survival by TP53 mutational status*

CI, confidence interval; NE, not estimable; TP53mut, mutated TP53; TP53wt, wild-type TP53. *Data from Garcia-Manero.2
TP53 mutational status	Median overall survival, months (95% CI)
TP53wt	33.7 (29.0–NE)
TP53mut	25.5 (14.4–NE)
TP53mut monoallelic	29.2 (19.7–NE)
TP53mut biallelic	13.0 (5.3–29.0)

In patients with MDS/chronic myelomonocytic leukemia treated with oral decitabine and cedazuridine, this preliminary analysis emphasized worse survival outcomes with high-risk TP53 mutational status, particularly related to TP53^mut biallelic state. Patients with biallelic TP53^mut profile treated with oral decitabine/cedazuridine showed favorable survival outcomes versus historical results.

Allo-HSCT³

Study design and patient characteristics

This retrospective, multicenter study was carried out through the Consortium on Myeloid Malignancies and Neoplastic Diseases (COMMAND) in collaboration with acute leukemia experts from ten academic centers in the US. The study aimed to assess long-term survival predictors in a large cohort of patients with TP53^mut AML who had received allo-HSCT. This study included 370 patients with TP53^mut AML diagnosed from 2012 to 2021, of which 13% received allo-HSCT after first-line therapy and 5% received allo-HSCT after second-line therapy.

Key findings

In patients who received allo-HSCT after first-line therapy:

• the rate of complete remission (CR) with negative measurable residual disease (MRD) was 37% prior to allo-HSCT;
• the rate of CR/CR with incomplete hematological recovery (CRi) with positive MRD was 39% prior to allo-HSCT; and
• the CR/CRi rate 100 days post-allo-HSCT was 88%.

In patients who received allo-HSCT after second-line therapy:

• the rate of CR with negative MRD was 35% prior to allo-HSCT;
• the rate of CR/CRi with positive MRD Was 45% prior to allo-HSCT; and
• the CR/CRi rate 100 days post-allo-HSCT was 50%.

In patients who received allo-HSCT after first-line therapy, the median OS was higher compared with patients who received allo-HSCT after second-line therapy (30.5 months vs 20.2 months; p = 0.01). Univariate analysis of event-free survival (EFS) revealed that factors associated with EFS included:

• CR/CRi at Day 100 post-allo-HSCT versus progressive disease (17.53 months vs 5.55 months; p < 0.001); and
• chronic graft versus host disease (cGvHD) versus no cGvHD (23.67 months vs 7.0 months; p = 0.0001).

Univariate analysis showed that the factors associated with OS also included:

• CR/CRi at Day 100 post-allo-HSCT versus progressive disease (25.73 months vs 12.87 months; p = 0.003), and
• cGvHD versus no cGvHD (46.1 months vs 14.37 months; p = 0.0001).

Multivariate analysis for EFS and OS demonstrated that CR/CRi at Day 100 post-allo-HSCT and cGvHD retained significance in terms of association with EFS and OS:

• CR/CRi 100 days post-allo-HSCT EFS (HR, 0.24; 95% CI, 0.10–0.57; p = 0.001)
• CR/CRi 100 days post-allo-HSCT OS (HR, 0.22; 95% CI, 0.10–0.50; p < 0.001)
• cGvHD EFS (HR, 0.21; 95% CI, 0.09–0.46; p < 0.001)
• cGvHD OS (HR, 0.34; 95% CI, 0.15–0.75; p = 0.007)

Overall, this multicenter study suggested that allo-HSCT is associated with improved survival in patients with TP53^mut AML; CR/CRi at Day 100 post-allo-HSCT and cGvHD were identified as long-term predictors for significantly better survival outcomes in this patient cohort.

Clinical and genetic factors

Variant allele frequency⁴

Study design and patient characteristics

This study aimed to assess TP53 allelic configuration and prognosis in patients with myeloid neoplasia. It included 7,400 patients with MDS and AML that were analyzed by NGS revealing that 1,010 patients had TP53^mut. Patients were classified as either TP53 single-hit (one TP53 mutation, isolated 17p deletion) or TP53 multi-hit (a TP53 mutation and another TP53 mutation, or 17p deletion, or TP53 locus uniparental disomy). Random-forest regression was then performed to separate monoallelic and biallelic mutations using survival as a benchmark. A new predictive algorithm was applied to better resolve allelic configurations and help improve prognosis predictions.

Key findings

Regardless of configuration, patients with TP53^mut had worse OS when compared with patients with TP53^wt (p < 0.001). Patients with TP53 multi-hit had worse OS than patients with TP53 single-hit in patients with AML (p = 0.035) and MDS (p < 0.0001), while patients with TP53 single-hit had worse OS than patients with TP53^wt in patients with AML (p < 0.0001) and MDS (p = 0.0069).

Random-forest regression survival analysis demonstrated that a variant allele frequency (VAF) cut-off of 23% could separate patients into probable monoallelic (VAF <23%) and probable biallelic (VAF ≥23%) and obligatory biallelic (VAF >50%). When applying this new algorithm to the TP53 single-hit and TP53 multi-hit models, there was a statistical survival difference in all groups, demonstrating the impact of a VAF cut-off of 23% on prognosis.

In summary, the study suggested a novel approach to resolve TP53^mut allelic configuration and improve prognosis beyond traditionally used single and double TP53 hit groups in clinical practice.

Allelic configuration, blast count, and karyotype⁵

Study design and patient characteristics

This study included 773 patients with AML and 747 patients with MDS, with the aim of analyzing the frequency of TP53 single-hit and multi-hit events in patients with AML and MDS and determining the impact of TP53 allelic state on OS. All patients were analyzed using whole-genome sequencing.

Key findings

TP53 alterations occurred at a frequency of 13% in patients with MDS and 11% in patients with AML (Table 3). When analyzing patients with MDS by blast percentage, patients with MDS with <5% blasts were more likely to have TP53 single hit (83%) than TP53 multi-hit (17%). However, patients with MDS with ≥5–10% blasts and with ≥10% to <20% blasts were more likely to have TP53 multi-hit (67% and 61%, respectively) than TP53 single-hit (33% and 39%, respectively).

Table 3. Frequency of TP53 alterations*

AML, acute myeloid leukemia; CN-LOH, copy neutral loss of heterozygosity; MDS, myelodysplastic syndromes. *Data from Meggendorfer.5
TP53 alterations, %	MDS (n = 96)	AML (n = 84)
Single-hit
TP53 mutation + wild type	50	24
17q deletion + wild type	7	14
17p CN-LOH + wild type	1	2
Multi-hit
TP53 mutation + TP53 mutation	13	16
17q deletion + TP53 mutation	20	38
17p CN-LOH + TP53 mutation	9	6

In terms of OS, TP53 single-hit and TP53 multi-hit were associated with an inferior OS when compared with TP53^wt in patients with AML and patients with MDS across all blast groups. TP53 multi-hit was also associated with an inferior OS when compared with TP53 single hit in patients with AML (p < 0.001) and patients with MDS with ≥5–10% blasts (p = 0.029; Table 4).

Table 4. OS in TP53 single-hit versus multi-hit MDS subgroups and AML*

AML, acute myeloid leukemia; MDS, myelodysplastic syndromes; OS, overall survival; TP53wt, TP53 wild type. *Adapted from Meggendorfer.5
Median OS, months	Total cohort	TP53wt	TP53 alterations	TP53 single-hit	TP53 multi-hit
MDS <5% blasts	75	80	56	57	17
MDS ≥5–10% blasts	54	60	11	22	10
MDS ≥10% to <20% blasts	27	34	10	13	8
AML	18	21	3	8	1

Univariate and multivariate analysis revealed that TP53 single-hit, TP53 multi-hit, blast count (<5% to ≥5%), and complex karyotype had an independent adverse impact on OS, with TP53 multi-hit being the strongest prognostic factor. TP53 single-hit and TP53 multi-hit remained significant independent factors in all MDS blast subgroups.

Conclusion

The above studies highlight that TP53^mut remains an adverse prognostic factor in patients with AML and MDS. Current treatment options are limited, although allo-HSCT has been shown to be associated with improved long-term outcomes.^1,3 New emerging treatments such as oral decitabine and cedazuridine have also demonstrated encouraging preliminary results in this patient population, though further studies are necessary.²

There have also been advancements in our understanding of TP53^mut, with the prognostic significance of allelic configuration becoming more apparent. Multi-hit TP53^mut has demonstrated an association with dismal outcomes when compared with both single-hit TP53^mut and TP53^wt.^2,4,5 Further prospective studies are warranted to elucidate the impact of allelic configuration on survival outcomes in patients with AML and MDS.