Prognosis
Risk stratification vs actual outcome, 35 months post-HSCT
IPSS-M Score Very High
3.24
Very High (threshold >1.5); range 3.21 to 4.36 over missing genes. Official IPSS-M calculator, Bernard et al. NEJM Evid 2022
ELN 2022
Adverse
Monosomy 7 overrides IDH2
Median OS
9.7 months
ELN Adverse, newly diagnosed
Actual OS Alive
37+ months
Since diagnosis (Sept 2023). MRD negative, full chimerism
Drivers Unprecedented
5
One per oncogenic axis, plus monosomy 7
Expected vs Actual
Source: Median OS values from published
cohort studies. Actual survival counted from diagnosis, as of November 2026.
Risk Model Comparison
| Model | Category | Median OS | Source |
|---|---|---|---|
| IPSS-M | Very High (3.24) | 1.0 year | Bernard et al. NEJM Evid 2022 |
| ELN 2022 | Adverse | 9.7 months | Lachowiez et al. Blood Adv 2023 |
| Chr7-aberrant + PTPN11 | HR 2.24 | N/A | Halik et al. J Hematol Oncol 2024 |
| Chr7-aberrant + IDH2 | HR 0.51 (protective) | N/A | Halik et al. 2024 |
| VenAza (RAS pathway) | Lower-benefit | 5.5 months | Döhner et al. Blood 2024 |
| R/R AML on HMA/VEN, VEN-PRS adverse group (EMD, prior HMA, NF1, PTPN11, FLT3, TP53) | Adverse (n=53) | 4.6 months | Shahswar et al. Leukemia 2025 |
Note: HR = hazard ratio. IPSS-M
13
[13] E 2022
Molecular International Prognostic Scoring System for Myelodysplastic Syndromes. NEJM Evid (2022)
trained on 2,957 MDS patients. ELN
2022
14[14] H 2022
Diagnosis and management of AML in adults: 2022 recommendations from an international expert panel on behalf of the ELN. Blood (2022)
risk stratification. AML genomic
classification per Papaemmanuil et al.
12[12] E 2016
Genomic Classification and Prognosis in Acute Myeloid Leukemia. N Engl J Med (2016)
.
Outside the Model
Every scoring system places this profile in its worst category. But the
scoring systems are built on population data. This mutation profile does
not exist in the population. The patient scores not just worst:
the patient is outside the system's ability to score.
IPSS-M 13
And the patient is alive. MRD-negative. Full donor chimerism. 35 months post-HSCT. Acute GvHD grade 2 at day +52 and chronic GvHD from day +83, treated; moderate at 12 months, mild at 2 years, now topical steroids only. That is also outside the model.
This is a data point worth studying: the most adverse profile documented in 8,655 screened patients (what that counts) has survived 35 months post-HSCT.
With EZH2 V662A as a fifth variant, Likely Pathogenic in this project's research classification (ClinVar: VUS; on in-silico evidence; V662A itself has not been tested in the lab, and Chase et al. 10
EZH2-mutant MDS with chromosome 7 abnormalities carries the worst survival among EZH2-mutant subgroups 11
An in-house reimplementation of the PyClone-VI model 16
Type 1 progression mutations 15
And the patient is alive. MRD-negative. 35 months post-HSCT. That is the other side of being outside the model.
IPSS-M 13
[13] E 2022
Molecular International Prognostic Scoring System for Myelodysplastic Syndromes. NEJM Evid (2022)
was trained on 2,957 patients. None
had this quadruple combination. The model extrapolates from patients who
are less complex and assumes the pattern continues linearly. It likely
does not.
And the patient is alive. MRD-negative. Full donor chimerism. 35 months post-HSCT. Acute GvHD grade 2 at day +52 and chronic GvHD from day +83, treated; moderate at 12 months, mild at 2 years, now topical steroids only. That is also outside the model.
This is a data point worth studying: the most adverse profile documented in 8,655 screened patients (what that counts) has survived 35 months post-HSCT.
With EZH2 V662A as a fifth variant, Likely Pathogenic in this project's research classification (ClinVar: VUS; on in-silico evidence; V662A itself has not been tested in the lab, and Chase et al. 10
[10] A 2020
Mutational mechanisms of EZH2 inactivation in myeloid neoplasms. Leukemia (2020)
cite SET-domain missense loss of function
from earlier in vitro assays), a novel variant never
described in literature, carrying 0 carriers among the
20,739 GENIE myeloid samples whose panel covers EZH2. The
profile moves from unprecedented to genuinely unknown. The exact five-variant combination has no estimable population frequency. Zero in 8,655 patients bounds it below 1 in 2,889 (95%, exact binomial).
No prognostic model was designed to handle this.
EZH2-mutant MDS with chromosome 7 abnormalities carries the worst survival among EZH2-mutant subgroups 11
[11] S 2023
Clinical characteristics and outcomes of EZH2-mutant myelodysplastic syndrome: a large single institution analysis of 1774 patients. Leuk Res (2023)
. SETBP1 + monosomy 7 co-occurrence was
first characterized by Makishima et al.
2[2] H 2013
Somatic SETBP1 mutations in myeloid malignancies. Nat Genet (2013)
as a hallmark of aggressive
myeloid transformation.
An in-house reimplementation of the PyClone-VI model 16
[16] S 2020
PyClone-VI: scalable inference of clonal population structures using whole genome data. BMC Bioinformatics (2020)
orders the clones by cancer cell
fraction: DNMT3A (CCF 1.00), EZH2 (0.92), SETBP1 (0.87), PTPN11 (0.74),
IDH2 (0.05). DNMT3A and EZH2 cannot be ordered (overlapping intervals),
so the founding clone carries DNMT3A and/or EZH2, two epigenetic
regulators. IDH2 (metabolic effect on TET2) is a late minor subclone.
No existing treatment protocol addresses this combination.
Type 1 progression mutations 15
[15] H 2017
Dynamics of clonal evolution in myelodysplastic syndromes. Nat Genet (2017)
. The Type 1 hits in this profile
(PTPN11, signaling; IDH2, metabolic) acquired during transformation define the clonal
trajectory from MDS to AML.
And the patient is alive. MRD-negative. 35 months post-HSCT. That is the other side of being outside the model.
OncoKB Actionability
| Variant | OncoKB Level | Actionable Drug | Status |
|---|---|---|---|
| IDH2 R140Q | Level 1 | Enasidenib | FDA-approved 2017 |
| PTPN11 E76Q | No level (OncoKB: Likely Neutral) | SHP2 inhibitors (TNO155; SHP2i+venetoclax synergy) | Clinical trials |
| DNMT3A R882H | Oncogenic | Azacitidine (indirect) | Standard of care |
| SETBP1 G870S | Oncogenic | None (PP2A activators experimental) | Preclinical |
| EZH2 V662A | Unknown (not in OncoKB) | Tazemetostat CONTRAINDICATED (LoF) | N/A |
Only 2 of 5 drivers have FDA-approved targeted therapy (IDH2:
enasidenib
21
[21] EM 2017
Enasidenib in mutant IDH2 relapsed or refractory acute myeloid leukemia. Blood (2017)
) or active clinical trials (PTPN11:
SHP2 inhibitors
23[23] MJ 2020
Identification of TNO155, an Allosteric SHP2 Inhibitor for the Treatment of Cancer. J Med Chem (2020)
). The remaining 3 have no direct
inhibitors. SHP2 inhibitor (RMC-4550) + venetoclax synergy
22[22] B 2023
Allosteric SHP2 inhibition increases apoptotic dependency on BCL2 and synergizes with venetoclax in FLT3- and KIT-mutant AML. Cell Rep Med (2023)
has been shown preclinically in
FLT3- and KIT-mutant AML, not in PTPN11-mutant AML. This is a targetability gap that underscores the complexity of
this profile.
Source: OncoKB levels and oncogenicity from
the project's OncoKB API query (AML, March 2026); OncoKB per Chakravarty et
al. JCO Precis Oncol 2017
35
[35] D 2017
OncoKB: A Precision Oncology Knowledge Base. JCO Precis Oncol (2017)
. EZH2 V662A classified by this project as
loss-of-function; tazemetostat is indicated for gain-of-function EZH2
mutations only.
References
- Bernard E, Tuechler H, Greenberg PL, et al. Molecular International Prognostic Scoring System for Myelodysplastic Syndromes. NEJM Evid (2022). DOI
- Döhner H, Wei AH, Appelbaum FR, et al. Diagnosis and management of AML in adults: 2022 ELN recommendations. Blood (2022). PubMed
- Papaemmanuil E, Gerstung M, Bullinger L, et al. Genomic Classification and Prognosis in Acute Myeloid Leukemia. N Engl J Med (2016). PubMed
- Chase A, Score J, Lin F, et al. Mutational mechanisms of EZH2 inactivation in myeloid neoplasms. Leukemia (2020). PubMed
- Ball S, Aguirre LE, Jain AG, et al. Clinical characteristics and outcomes of EZH2-mutant myelodysplastic syndrome. Leuk Res (2023). PubMed
- Makishima H, Yoshida K, Nguyen N, et al. Somatic SETBP1 mutations in myeloid malignancies. Nat Genet (2013). PubMed
- Gillis S, Roth A. PyClone-VI: scalable inference of clonal population structures using whole genome data. BMC Bioinformatics (2020). DOI
- Makishima H, Yoshizato T, Yoshida K, et al. Dynamics of clonal evolution in myelodysplastic syndromes. Nat Genet (2017). PubMed
- Stein EM, DiNardo CD, Pollyea DA, et al. Enasidenib in mutant IDH2 relapsed or refractory acute myeloid leukemia. Blood (2017). PubMed
- LaMarche MJ, Acker M, Argintaru A, et al. Identification of TNO155, an Allosteric SHP2 Inhibitor for the Treatment of Cancer. J Med Chem (2020). PubMed
- Popescu B, Stahlhut C, Tarver TC, et al. Allosteric SHP2 inhibition increases apoptotic dependency on BCL2 and synergizes with venetoclax in FLT3- and KIT-mutant AML. Cell Rep Med (2023). PubMed
- Chakravarty D, Gao J, Phillips SM, et al. OncoKB: A Precision Oncology Knowledge Base. JCO Precis Oncol (2017). PubMed