Correction: Standardisation and consensus guidelines for minimal residual disease assessment in Philadelphia-positive acute lymphoblastic leukemia (Ph+ALL) by real-time quantitative reverse transcriptase PCR of e1a2 BCR-ABL1.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Standardisation and consensus guidelines for minimal residual disease assessment in Philadelphia-positive acute lymphoblastic leukemia (Ph + ALL) by real-time quantitative reverse transcriptase PCR of e1a2 BCR-ABL1.

Minimal residual disease (MRD) is a powerful prognostic factor in acute lymphoblastic leukemia (ALL) and is used for patient stratification and treatment decisions, but its precise role in Philadelphia chromosome positive ALL is less clear. This uncertainty results largely from methodological differences relating to the use of real-time quantitative PCR (qRT-PCR) to measure BCR-ABL1 transcript levels for MRD analysis. We here describe the first results by the EURO-MRD consortium on standardization of qRT-PCR for the e1a2 BCR-ABL1 transcript in Ph + ALL, designed to overcome the lack of standardisation of laboratory procedures and data interpretation. Standardised use of EAC primer/probe sets and of centrally prepared plasmid standards had the greatest impact on reducing interlaboratory variability. In QC1 the proportion of analyses with BCR-ABL1/ABL1 ratios within half a log difference were 40/67 (60%) and 52/67 (78%) at 10-3 and 36/67 (53%) and 53/67 (79%) at 10-4BCR-ABL1/ABL1. Standardized RNA extraction, cDNA synthesis and cycler platforms did not improve results further, whereas stringent application of technical criteria for assay quality and uniform criteria for data interpretation and reporting were essential. We provide detailed laboratory recommendations for the standardized MRD analysis in routine diagnostic settings and in multicenter clinical trials for Ph + ALL.

Subtype-specific patterns of molecular mutations in acute myeloid leukemia.

Acute myeloid leukemia (AML) can be grouped into morphologically or genetically defined subtypes. Today, the AML phenotype-genotype associations, that is, FAB/WHO (French-American-British/World Health Organization) definitions and recurrent molecular mutations, are not fully understood. Therefore, we evaluated the impact of molecular mutations on the AML differentiation stage by molecular profiling of 4373 adult de novo AML patients in 7 cytomorphological subtypes. We investigated mutations in 20 genes, including myeloid transcription factors (CEBPA, RUNX1), tumor suppressors (TP53, WT1), DNA modifiers (DNMT3A, IDH1/2, TET2), chromatin modifiers (ASXL1, MLL), signal transduction genes (FLT3, KRAS, NRAS) and NPM1. The most frequently mutated genes per cytomorphological subtype were RUNX1 in M0 (43%), NPM1 in M1 (42%), DNMT3A in M2 (26%), NPM1 in M4 (57%), M5a (49%) and M5b (70%) and TP53 in M6 (36%). Although some gene mutations were frequent in several cytomorphological subtypes, a series of associations of co-occurring mutations with distinct phenotypes were identified for molecularly defined subcohorts. FLT3, NPM1 and WT1 mutations were associated with an immature phenotype in myeloblastic AML, whereas other combinations involving ASXL1, RUNX1, MLL-PTD, CEBPA or KRAS were more frequent in myeloblastic AML with maturation. Within the NPM1 mutated subcohort, ASXL1 mutations were significantly associated with a monoblastic differentiation and DNMT3A mutations with a monocytic phenotype.

Increasing intensity of therapies assigned at diagnosis does not improve survival of adults with acute myeloid leukemia.

We randomized 3375 adults with newly diagnosed acute myeloid leukemia (AML) or high-risk myelodysplastic syndrome to test whether increasingly intensive chemotherapies assigned at study-entry and analyzed on an intent-to-treat basis improved outcomes. In total, 1529 subjects <60 years were randomized to receive: (1) a first course of induction therapy with high-dose cytarabine and mitoxantrone (HAM) or with standard-dose cytarabine, daunorubicin and 6-thioguanine (TAD) followed by a second course of HAM; (2) granulocyte-colony stimulating factor (G-CSF) or no G-CSF before induction and consolidation courses; and (3) high-dose therapy and an autotransplant or maintenance chemotherapy. In total, 1846 subjects ⩾60 years were randomized to receive: (1) a first induction course of HAM or TAD and second induction course of HAM (if they had bone marrow blasts ⩾5% after the first course); and (2) G-CSF or no G-CSF as above. Median follow-up was 7.4 years (range, 1 day to 14.7 years). Five-year event-free survivals (EFSs) for subjects receiving a first induction course of HAM vs TAD were 17% (95% confidence interval, 15, 18%) vs 16% (95% confidence interval 14, 18%; P=0.719). Five-year EFSs for subjects randomized to receive or not receive G-CSF were 19% (95% confidence interval 16, 21%) vs 16% (95% confidence interval 14, 19%; P=0.266). Five-year relapse-free survivals (RFSs) for subjects <60 years receiving an autotransplant vs maintenance therapy were 43% (95% confidence interval 40, 47%) vs 40 (95% confidence interval 35, 44%; P=0.535). Many subjects never achieved pre-specified landmarks and consequently did not receive their assigned therapies. These data indicate the limited impact of more intensive therapies on outcomes of adults with AML. Moreover, none of the more intensive therapies we tested improved 5-year EFS, RFS or any other outcomes.

Array-based comparative genomic hybridization detects copy number variations with prognostic relevance in 80% of ALL with normal karyotype or failed chromosome analysis.

Pretreatment cytogenetics is an important parameter for risk stratification and therapy approach in acute lymphoblastic leukemia (ALL). However, in up to 30% of cases, chromosome banding analysis (CBA) fails or reveals a normal karyotype. To characterize the subset of ALL with normal karyotype or failed CBA, we performed fluorescence in situ hybridization (FISH) or PCR for BCR-ABL1 and MLL rearrangements as well as array comparative genomic hybridization (aCGH) in 186 adult patients. We further carried out FISH for MYC in cases with Burkitt leukemia phenotype. FISH or PCR revealed one of the respective rearrangements in 22% of patients. In 80% of cases, copy number variations (CNV) were identified by aCGH. In 22% of cases, all CNV were below the resolution of CBA. On the basis of results of FISH, RT-PCR and aCGH, patients were categorized into three groups. The novel subset of patients with submicroscopic CNV only showed an overall survival at 3 years of 84% compared with 64% for patients classified as adverse abnormalities and 77% for cases with other aberrations (P=0.046). Thus, ALL with non-informative CBA can be further classified by FISH and aCGH providing prognostic information, which may be useful for a more individualized therapy.

Additional mutations in SRSF2, ASXL1 and/or RUNX1 identify a high-risk group of patients with KIT D816V(+) advanced systemic mastocytosis.

Most patients with KIT D816V(+) advanced systemic mastocytosis (SM) are characterized by somatic mutations in additional genes. We sought to clarify the prognostic impact of such mutations. Genotype and clinical characteristics of 70 multi-mutated KIT D816V(+) advanced SM patients were included in univariate and multivariate analyses. The most frequently identified mutated genes were TET2 (n=33 of 70 patients), SRSF2 (n=30), ASXL1 (n=20), RUNX1 (n=16) and JAK2 (n=11). In univariate analysis, overall survival (OS) was adversely influenced by mutations in SRSF2 (P<0.0001), ASXL1 (P=0.002) and RUNX1 (P=0.03), but was not influenced by mutations in TET2 or JAK2. In multivariate analysis, SRSF2 and ASXL1 remained the most predictive adverse indicators concerning OS. Furthermore, we found that inferior OS and adverse clinical characteristics were significantly influenced by the number of mutated genes in the SRSF2/ASXL1/RUNX1 (S/A/R) panel (P<0.0001). In conclusion, the presence and number of mutated genes within the S/A/R panel are adversely associated with advanced disease and poor survival in KIT D816V(+) SM. On the basis of these findings, inclusion of molecular markers should be considered in upcoming prognostic scoring systems for patients with SM.

Molecular profiling of myeloid progenitor cells in multi-mutated advanced systemic mastocytosis identifies KIT D816V as a distinct and late event.

To explore the molecular profile and its prognostic implication in systemic mastocytosis (SM), we analyzed the mutation status of granulocyte-macrophage colony-forming progenitor cells (CFU-GM) in patients with KIT D816V(+) indolent SM (ISM, n=4), smoldering SM (SSM, n=2), aggressive SM (ASM, n=1), SM with associated clonal hematologic non-mast cell lineage disorder (SM-AHNMD, n=5) and ASM-AHNMD (n=7). All patients with (A)SM-AHNMD (n=12) carried 1-4 (median 3) additional mutations in 11 genes tested, most frequently TET2, SRSF2, ASXL1, CBL and EZH2. In multi-mutated (A)SM-AHNMD, KIT D816V(+) single-cell-derived CFU-GM colonies were identified in 8/12 patients (median 60%, range 0-95). Additional mutations were identified in CFU-GM colonies in all patients, and logical hierarchy analysis indicated that mutations in TET2, SRSF2 and ASXL1 preceded KIT D816V. In ISM/SSM, no additional mutations were detected and CFU-GM colonies were exclusively KIT D816V(-). These data indicate that (a) (A)SM-AHNMD is a multi-mutated neoplasm, (b) mutations in TET2, SRSF2 or ASXL1 precede KIT D816V in ASM-AHNMD,

WT1 mutations are secondary events in AML, show varying frequencies and impact on prognosis between genetic subgroups.

To investigate frequency and prognostic impact of Wilms tumor 1 (WT1) mutations (mut), we analyzed 3157 unselected acute myeloid leukemia patients for WT1mut in exons 7 and 9. In total, 188 WT1 mutations were detected (exon 7: n=150, exon 9: n=38); 141 were frameshift, 24 missense, 14 non-sense, 7 splice site and 2 indel mutations. In 175/3157 (5.5%) patients, a WT1mut was found. Higher frequencies were detected in patients with biallelic CEBPAmut (13.6%; P=0.001), followed by t(15;17)/PML-RARA (11.0%, P=0.004), and FLT3-ITD (8.5%, P<0.001). WT1mut were rare in DNMT3Amut (4.4%, P=0.014), ASXL1mut (1.7%, P<0.001), IDH2R140 (1.7%, P=0.001) and IDH1R132 (0.9%, P<0.001), and not detected in complex karyotypes (P=0.047). They were more frequent in females than in males (6.6 vs 4.7%; P=0.014) and in patients <60 years (P<0.001). Analysis of paired samples of 35 patients revealed a relatively unstable character of WT1mut (65.7% retained, 34.3% lost WT1mut at relapse). In the total cohort and subgroups with high WT1mut incidences (biallelic CEBPAmut, PML-RARA), WT1mut had no impact on prognosis. In normal karyotype AML, WT1mut patients had shorter event-free survival (EFS) (10.8 vs 17.9 m, P=0.008). In multivariate analysis, WT1mut had an independent adverse impact on EFS (P=0.002, hazard ratio (HR): 1.64) besides FLT3-ITD status (P<0.001, HR: 1.71) and age (P<0.001, HR: 1.28).

Molecular-defined clonal evolution in patients with chronic myeloid leukemia independent of the BCR-ABL status.

To study clonal evolution in chronic myeloid leukemia (CML), we searched for BCR-ABL-independent gene mutations in both Philadelphia chromosome (Ph)-negative and Ph-positive clones in 29 chronic-phase CML patients by targeted deep sequencing of 25 genes frequently mutated in myeloid disorders. Ph-negative clones were analyzed in 14 patients who developed clonal cytogenetic abnormalities in Ph-negative cells during treatment with tyrosine kinase inhibitors (TKI). Mutations were detected in 6/14 patients (43%) affecting the genes DNMT3A, EZH2, RUNX1, TET2, TP53, U2AF1 and ZRSR2. In two patients, the mutations were also found in corresponding Ph-positive diagnostic samples. To further investigate Ph-positive clones, 15 randomly selected CML patients at diagnosis were analyzed. Somatic mutations additional to BCR-ABL were found in 5/15 patients (33%) affecting ASXL1, DNMT3A, RUNX1 and TET2. Analysis of individual hematopoietic colonies at diagnosis revealed that most mutations were part of the Ph-positive clone. In contrast, deep sequencing of subsequent samples during TKI treatment revealed one DNMT3A mutation in Ph-negative cells that was also present in Ph-positive cells at diagnosis, implying that the mutation preceded the BCR-ABL rearrangement. In summary, BCR-ABL-independent gene mutations were frequently found in Ph-negative and Ph-positive clones of CML patients and may be considered as important cofactors in the clonal evolution of CML.

Velocity of early BCR-ABL transcript elimination as an optimized predictor of outcome in chronic myeloid leukemia (CML) patients in chronic phase on treatment with imatinib.

UNLABELLED: Early assessment of response at 3 months of tyrosine kinase inhibitor treatment has become an important tool to predict favorable outcome. We sought to investigate the impact of relative changes of BCR-ABL transcript levels within the initial 3 months of therapy. In order to achieve accurate data for high BCR-ABL levels at diagnosis, beta glucuronidase (GUS) was used as a reference gene. Within the German CML-Study IV, samples of 408 imatinib-treated patients were available in a single laboratory for both times, diagnosis and 3 months on treatment. In total, 301 of these were treatment-naïve at sample collection. RESULTS: (i) with regard to absolute transcript levels at diagnosis, no predictive cutoff could be identified; (ii) at 3 months, an individual reduction of BCR-ABL transcripts to the 0.35-fold of baseline level (0.46-log reduction, that is, roughly half-log) separated best (high risk: 16% of patients, 5-year overall survival (OS) 83% vs 98%, hazard ratio (HR) 6.3, P=0.001); (iii) at 3 months, a 6% BCR-ABL(IS) cutoff derived from BCR-ABL/GUS yielded a good and sensitive discrimination (high risk: 22% of patients, 5-year OS 85% vs 98%, HR 6.1, P=0.002). Patients at risk of disease progression can be identified precisely by the lack of a half-log reduction of BCR-ABL transcripts at 3 months.

BAALC expression: a suitable marker for prognostic risk stratification and detection of residual disease in cytogenetically normal acute myeloid leukemia.

High brain and acute leukemia, cytoplasmic (BAALC) expression defines an important risk factor in cytogenetically normal acute myeloid leukemia (CN-AML). The prognostic value of BAALC expression in relation to other molecular prognosticators was analyzed in 326 CN-AML patients (<65 years). At diagnosis, high BAALC expression was associated with prognostically adverse mutations: FLT3 internal tandem duplication (FLT3-ITD) with an FLT3-ITD/FLT3 wild-type (wt) ratio of 0.5 (P=0.001), partial tandem duplications within the MLL gene (MLL-PTD) (P=0.002), RUNX1 mutations (mut) (P<0.001) and WT1mut (P=0.001), while it was negatively associated with NPM1mut (P<0.001). However, high BAALC expression was also associated with prognostically favorable biallelic CEBPA (P=0.001). Survival analysis revealed an independent adverse prognostic impact of high BAALC expression on overall survival (OS) and event-free survival (EFS), and also on OS when eliminating the effect of allogeneic stem cell transplantation (SCT) (OS(TXcens)). Furthermore, we analyzed BAALC expression in 416 diagnostic and follow-up samples of 66 patients. During follow-up, BAALC expression correlated with mutational load or expression levels, respectively, of other minimal residual disease markers: FLT3-ITD (r=0.650, P<0.001), MLL-PTD (r=0.728, P<0.001), NPM1mut (r=0.599, P<0.001) and RUNX1mut (r=0.889, P<0.001). Moreover, a reduction in BAALC expression after the second cycle of induction chemotherapy was associated with improved EFS. Thus, our data underline the utility of BAALC expression as a marker for prognostic risk stratification and detection of residual disease in CN-AML.

High number of additional genetic lesions in acute myeloid leukemia with t(8;21)/RUNX1-RUNX1T1: frequency and impact on clinical outcome.

t(8;21)/RUNX1-RUNX1T1-positive acute myeloid leukemia (AML) is prognostically favorable; however, outcome is heterogeneous. We analyzed 139 patients with t(8;21)/RUNX1-RUNX1T1-positive AML (de novo: n=117; therapy-related: n=22) to determine frequency and prognostic impact of additional genetic abnormalities. All patients were investigated for mutations (mut) in ASXL1, FLT3, KIT, NPM1, MLL, IDH1, IDH2, KRAS, NRAS, CBL and JAK2. Sixty-nine of 139 cases (49.6%) had 1 mutation in addition to RUNX1-RUNX1T1, and 23/139 (16.5%) had ⩾2 additional mutations. Most common were KITmut (23/139; 16.5%), NRASmut (18/139; 12.9%) and ASXL1mut (16/139; 11.5%). FLT3-ITD, FLT3-TKDmut, CBLmut, KRASmut, IDH2mut and JAK2mut were found in 2.9-5.0%. Additional chromosomal abnormalities (ACAs) were found in 97/139 (69.8%). Two-year overall survival (OS) was 73.4% in 111 intensively treated patients. KITD816mut negatively impacted on OS in de novo AML (2-year OS: 59.1% vs 82.0%, P=0.03), ASXL1mut on EFS (de novo AML: 20% vs 59.1%, P=0.011; total cohort: 28.6% vs 56.7%, P=0.021). Sex chromosome loss was favorable (2-year EFS: 66.9% vs 43.0%, P=0.031), whereas +8 was adverse on EFS (2-year EFS: 26.7% vs 55.9%, P=0.02). In conclusion, t(8;21)/RUNX1-RUNX1T1-positive AML shows a high frequency of additional genetic alterations. Investigation for KITD816 and ASXL1mut combined with investigation of ACAs is recommended in t(8;21)/RUNX1-RUNX1T1-positive AML because of the prognostic significance of these parameters.

Landscape of genetic lesions in 944 patients with myelodysplastic syndromes.

High-throughput DNA sequencing significantly contributed to diagnosis and prognostication in patients with myelodysplastic syndromes (MDS). We determined the biological and prognostic significance of genetic aberrations in MDS. In total, 944 patients with various MDS subtypes were screened for known/putative mutations/deletions in 104 genes using targeted deep sequencing and array-based genomic hybridization. In total, 845/944 patients (89.5%) harbored at least one mutation (median, 3 per patient; range, 0-12). Forty-seven genes were significantly mutated with TET2, SF3B1, ASXL1, SRSF2, DNMT3A, and RUNX1 mutated in >10% of cases. Many mutations were associated with higher risk groups and/or blast elevation. Survival was investigated in 875 patients. By univariate analysis, 25/48 genes (resulting from 47 genes tested significantly plus PRPF8) affected survival (P<0.05). The status of 14 genes combined with conventional factors revealed a novel prognostic model ('Model-1') separating patients into four risk groups ('low', 'intermediate', 'high', 'very high risk') with 3-year survival of 95.2, 69.3, 32.8, and 5.3% (P<0.001). Subsequently, a 'gene-only model' ('Model-2') was constructed based on 14 genes also yielding four significant risk groups (P<0.001). Both models were reproducible in the validation cohort (n=175 patients; P<0.001 each). Thus, large-scale genetic and molecular profiling of multiple target genes is invaluable for subclassification and prognostication in MDS patients.

SF3B1 mutations correlated to cytogenetics and mutations in NOTCH1, FBXW7, MYD88, XPO1 and TP53 in 1160 untreated CLL patients.

We analyzed a large cohort of 1160 untreated CLL patients for novel genetic markers (SF3B1, NOTCH1, FBXW7, MYD88, XPO1) in the context of molecular, immunophenotypic and cytogenetic data. NOTCH1 mutations (mut) (12.3%), SF3B1mut (9.0%) and TP53mut (7.1%) were more frequent than XPO1mut (3.4%), FBXW7mut (2.5%) and MYD88mut (1.5%). SF3B1mut, NOTCH1mut, TP53mut and XPO1mut were highly correlated to unmutated, whereas MYD88mut were associated with mutated IGHV status. Associations of diverse cytogenetic aberrations and mutations emerged: (1) SF3B1mut with del(11q), (2) NOTCH1mut and FBXW7mut with trisomy 12 and nearly exclusiveness of SF3B1mut, (3) MYD88mut with del(13q) sole and low frequencies of SF3B1mut, NOTCH1mut and FBXW7mut. In patients with normal karyotype only SF3B1mut were frequent, whereas NOTCH1mut rarely occurred. An adverse prognostic impact on time to treatment (TTT) and overall survival (OS) was observed for SF3B1mut, NOTCH1mut and TP53 disruption. In multivariate analyses SF3B1mut, IGHV mutational status and del(11q) were the only independent genetic markers for TTT, whereas for OS SF3B1mut, IGHV mutational status and TP53 disruption presented with significant impact. Finally, our data suggest that analysis of gene mutations refines the risk stratification of cytogenetic prognostic subgroups and confirms data of a recently proposed model integrating molecular and cytogenetic data.

Monitoring of residual disease by next-generation deep-sequencing of RUNX1 mutations can identify acute myeloid leukemia patients with resistant disease.

We studied the utility and clinical relevance of RUNX1 (runt-related transcription factor 1) mutations and their application as residual disease detection markers using next-generation deep-sequencing. Mutation screening was prospectively performed in 814 acute myeloid leukemia patients. At diagnosis, 211/814 (25.9%) patients harbored mutations with a median clone size of 39% (range: 2-96%). Furthermore, in 57 patients paired samples from diagnosis and relapse were analyzed. In 47/57 (82.5%) cases the same alterations detected at diagnosis were present at relapse, whereas in 1/57 (1.8%) cases the mutation from the diagnostic sample was no longer detectable. Discrepancies were observed in 9/57 (15.8%) cases, also including the occurrence of novel RUNX1 mutations not restricted to those regions affected at diagnosis. Moreover, in 103 patients the prognostic impact of residual levels of RUNX1 mutations during complete remission was studied. Separation of patients according to median residual mutation burden into 'good responders' and 'poor responders' (median: 3.61%; range: 0.03-48.0%) resulted in significant differences of both event-free (median 21.0 vs. 5.7 months, P<0.001) and overall survival (OS; median 56.9 vs. 32.0 months, P=0.002). In conclusion, deep-sequencing revealed that RUNX1 mutations qualify as patient-specific markers for individualized disease monitoring. The measurement of mutation load may refine the assignment into distinct risk categories and treatment strategies.