Indeterminate and oncogenic potential: CHIP vs CHOP mutations in AML with NPM1 alteration

Abstract

In AML patients, recurrent mutations were shown to persist in remission, however, only some have a prognostic value and persistent mutations might therefore reflect a re-established premalignant state or truly active disease causing relapse. We aimed to dissect the nature of co-mutations in NPM1 mutated AML where the detection of NPM1 transcripts allows highly specific and sensitive detection of complete molecular remission (CMR). We analysed 150 consecutive patients who achieved CMR following intensive treatment by next generation sequencing on paired samples at diagnosis, CMR and relapse (38/150 patients). Patients with persistence or the acquisition of non-DTA (DNMT3A, TET2, ASXL1) mutations at CMR (23/150 patients, 15%) have a significantly worse prognosis (EFS HR = 2.7, p = 0.003; OS HR = 3.6, p = 0.012). Based on clonal evolution analysis of diagnostic, CMR and relapse samples, we redefine pre-malignant mutations and include IDH1, IDH2 and SRSF2 with the DTA genes in this newly defined group. Only the persistence or acquisition of CHOP-like (clonal hematopoiesis of oncogenic potential) mutations was significantly associated with an inferior outcome (EFS HR = 4.5, p = 0.0002; OS HR = 5.5, p = 0.002). Moreover, the detection of CHOP-like mutations at relapse was detrimental (HR = 4.5, p = 0.01). We confirmed these findings in a second independent whole genome sequencing cohort.

Fig. 1. Persisting/acquired non-DTA mutations at CMR confer inferior survival in AML with mutated NPM1.

A Mutation frequencies of AML associated genes in diagnostic and complete molecular remission (CMR) samples. The percentage of each gene alteration among all the mutations per timepoint is depicted. *FLT3-ITD mutations were detected by gene scan. B Survival analysis of patients with NPM1^mut AML stratified by persistence or acquisition of DTA vs non-DTA mutations at CMR. Kaplan–Meier plots depicting event-free survival (EFS left panel) and overall survival (OS right panel) of NPM1^mut AML patients based on the combination of persistency and acquisition of non-DTA mutations at CMR. Patients showing non-DTA hits at CMR have a worse prognosis than those who do not. P values were calculated with the log-rank test and p values for pairwise comparisons are given.

Fig. 2. Clonal evolution defines persistent CHIP and CHOP-like mutations in NPM1^mut AML.

A Donut plot depicting the mutational status of patients at complete molecular remission (CMR). 81/150 patients (54%) had no mutation, whereas 69/150 (46%) had persistency/acquisition of single or combined mutations, for a total of 24 different groups (donut slices). B Clonal evolution analysis of NPM1^mut AML from diagnosis to CMR (n = 150 patients). Lost mutations are depicted in blue, persistent mutations in gray and acquired mutations in orange. Three main patterns emerged: mutations that were mostly or completely lost at CMR: NRAS, FLT3-TKD, STAG2, WT1, GATA2, and KRAS (all 100%), PTPN11 (95%), CEBPA (92%), IDH1 (81%), EZH2 (80%), IDH2 (71%); mutations that were mostly or exclusively acquired at CMR (TP53, CSNK1A1 and SETBP1, all 100%), and mutations with a more heterogeneous behavior: SRSF2 (mutation lost in 45% persistent in 38% and acquired in 17% of cases), TET2 (mutation lost in 52%, persistent in 31% and acquired in 17% of cases) and DNMT3A (mutation lost in 29%, persistent in 56% and acquired in 15% of cases). C Mutation frequencies of AML associated genes in relapse samples of 38/52 patients with clinical relapse. The percentage of each gene alteration among all the mutations per timepoint is depicted. *FLT3-ITD mutations were detected by gene scan. D Clonal evolution analysis of NPM1^mut AML from diagnosis to CMR and relapse (R) (n = 38 patients) allows for higher temporal resolution and identifies three main patterns: mutations which could either persist at CMR and be lost at R or completely absent at both CMR and R (BCOR, NRAS, FLT3-TKD); CHIP-like mutations present at diagnosis, CMR and R (TET2, IDH1, IDH2, DNMT3A, SRSF2); mutations with oncogenic potential: gained at CMR and persistent at R or acquired de novo at relapse (CEBPA, PTPN11, WT1, GATA2, RUNX1). E DNMT3A, TET2, IDH1, IDH2, and SRSF2 often act as foundation mutations onto which other potentially oncogenic (CHOP) hits arise as later events in AML pathogenesis. Venn diagram showing the novel proposed classification of CHIP-like mutations including: DNMT3A, TET2, IDH1, IDH2, and SRSF2, versus mutations with oncogenic potential (CHOP) in the context of NPM1^mut AML.

Fig. 3. Persistence or acquisition of novel defined premalignant CHIP like mutations vs CHOP like mutations are prognostic in AML with mutated NPM1.

A, B Survival analysis of patients with NPM1^mut AML stratified by clonal evolution patterns of novel defined CHIP-like mutations vs oncogenic mutations. Based on the clonal evolution analysis on diagnosis, remission and relapse samples we redefined CHIP-like mutations (DNMT3A, TET2, SRSF2, IDH2, and IDH1) versus all other mutations (CHOP-like). A Kaplan–Meier plots depicting event-free survival (EFS, left panel and OS, right panel) of NPM1^mut AML patients based on the persistency/acquisition of CHIP vs CHOP like mutations at CMR. P values were calculated with the log-rank test and p values for pairwise comparisons are given. B Cox proportional hazards multivariate model incorporating clonal evolution patterns by the presence/absence of CHOP-like mutations at CMR, clinical/molecular risk factors and allogeneic hematopoietic stem-cell transplantation. Overall survival (OS) hazard ratio (HR) at 95% confidence interval and p values for each variable are given.

Fig. 4. CHOP-like persistent or acquired mutations at relapse confer inferior outcome.

Survival analysis of patients with NPM1^mut AML experiencing clinical relapse stratified by clonal evolution patterns. Kaplan–Meier plots depicting overall survival (OS) of NPM1^mut AML patients following relapse all analysed by panel sequencing (n = 38). We stratified patients that acquired oncogenic mutations at relapse vs patients with no novel or only novel CHIP-like mutations at relapse. OS after relapse was calculated from the date of relapse until the date of death or censoring. P values were calculated with the log-rank test.

Fig. 5. Mutational analysis of independent WGS cohort recapitulates the panel-seq findings and confirms predictive power of CHOP-like mutations.

A Mutation frequencies of AML associated genes in diagnostic (D), complete molecular remission (CMR) and relapse samples (R). Mutations detected in at least two patients are depicted. B Clonal evolution analysis of NPM1^mut AML from diagnosis to CMR (n = 36 patients) of genes mutated in at least two patients. Lost mutations are depicted in blue, persistent mutations in gray and acquired mutations in orange. C Clonal evolution analysis of NPM1^mut AML from D to CMR and R (n = 8 patients) allows for higher temporal resolution and identifies three main patterns: mutations which could either persist at CMR and be lost at R or completely absent at both CMR and R (ANAPC5, NRAS, FLT3-TKD); CHIP-like mutations present at D, CMR and R (TET2, IDH1, SRSF2); mutations with oncogenic potential: gained at CMR and persistent at R or acquired de novo at R (RAD21, TBC1D9, EZH2, EMC7, NACAD, RUNX1); D Survival analysis of 36 patients from the WGS cohort stratified by clonal evolution patterns of novel defined CHIP-like mutations (DNMT3A, TET2, ASXL1, SRSF2, IDH1, IDH2): Kaplan–Meier plots depicting event-free survival (EFS, left panel) and overall survival (OS, right panel) of NPM1^mut AML patients based on the persistency/acquisition of CHIP vs CHOP like mutations at CMR. P values were calculated with the log-rank test.