Abstract Lenalidomide (LEN) can induce RBC transfusion independence (RBC-TI) in 60–70% of del(5q) myelodysplastic syndrome (MDS) patients. Current recommendation is to continue LEN in responding patients until failure or progression, with likelihood of toxicity and a high cost for healthcare systems. This HARMONY Alliance study investigated the outcome of MDS del(5q) patients who discontinued LEN in RBC-TI. We enrolled 118 patients with an IPSS-R low-intermediate risk. Seventy patients (59%) discontinued LEN for intolerance, 38 (32%) per their physician decision, nine (8%) per their own decision and one (1%) for unknown reasons. After a median follow-up of 49 months from discontinuation, 50/118 patients lost RBC-TI and 22/30 who underwent cytogenetic re-evaluation lost complete cytogenetic response. The median RBC-TI duration was 56 months. In multivariate analysis RBC-TI duration after LEN discontinuation correlated with low transfusion burden before LEN therapy, treatment ≥ 12 LEN cycles, younger age and higher Hb level at LEN withdrawal. Forty-eight patients were re-treated with LEN for loss of response and 28 achieved RBC-TI. These data show that stopping LEN therapy in MDS del(5q) patients in RBC-TI allows prolonged maintenance of TI in a large subset of patients.
Deletion of chromosome 5q occurs in 15%–20% of MDS patients and is associated with favorable prognosis if present as a single aberration or with only one additional cytogenetic aberration. The TP53 mutations, reported in 5%–10% of MDS, are enriched in del(5q) MDS (~20%), therapy-related MDS and MDS with complex karyotype and are associated with high-risk disease, AML transformation, treatment resistance and poor outcome. (1, 2) Recently, Bernard et al. showed that the number of TP53 aberrations is prognostic for death and leukemic transformation. (3) The PPM1D mutations are found in clonal hematopoiesis of indeterminate potential (CHIP) and appear more frequent in therapy-related MDS compared to de novo MDS (15% vs 3%). (4, 5) Activating PPM1D mutations are considered to act similarly to TP53 loss-of-function mutations. Loss of the C-terminal localization domain of PPM1D activates PPM1D and inhibits p53 activation. (6) However, the prevalence of PPM1D mutations, their impact and role in lenalidomide (LEN) resistance and disease progression in MDS with del(5q) still remains unknown. We performed a retrospective analysis of 234 patients ≥18 years old, with WHO 2016 defined del(5q) MDS (n = 175, 74.8%) or other MDS with del(5q) (n = 38, 16.2%) or sAML with del(5q) (n = 21, 9%) (supplementary methods). Patients with del(5q) alone or with one additional chromosomal abnormality except monosomy 7 or del(7q) and a blast count of <5% in bone marrow (BM) and < 1% in peripheral blood (PB) comprised the group of WHO 2016 defined del(5q) MDS. All other cases of MDS with del(5q) and complex karyotype, chromosome 7 abnormalities or blasts >5% in the BM or > 1% in PB were included in the group of other MDS with del(5q). Overall survival information was available for 216 of the 234 patients, specifically for 164 patients with WHO 2016 defined del(5q) MDS, 31 patients with other MDS with del(5q) and 21 patients with sAML with del5(q). There were 65 patients with WHO 2016 defined del(5q) MDS were treated with LEN and had information about treatment response available (Figure S1(A)). Del(5q) was the sole cytogenetic abnormality in 202 patients (86.3%). Ten patients (4.3%) harbored del(5q) and at least one additional chromosomal abnormality and 22 (9.4%) had a complex karyotype. The median age was 72.2 years (range 35–93). As expected, there was a female predominance (72.2%). Forty-four percent of the patients were transfusion-dependent at the time of diagnosis; 68 patients (29%) progressed to AML, and 19 (8.1%) underwent allogeneic hematopoietic cell transplantation (HCT) (Table S1). At time of diagnosis PPM1D mutations were detected in 13 of 234 (5.6%) MDS del(5q) patients, 11 of which had mutations in the hotspot region between amino acids 427 and 542 (Figure S1(B); supplementary methods and Table S2 for sequencing details). The mutation frequency was 6.3% (11 of 175) in patients with WHO 2016 defined del(5q) MDS, 5.3% (2 of 38) in patients with other MDS with del(5q) and 0% (0 of 21) in sAML from MDS with del(5q). One of the 13 PPM1D-mutated patients harbored a trisomy 8 in addition to del(5q), and two had a complex karyotype. Three PPM1D-mutated patients had a TP53 co-mutation (23%), including the two patients with complex karyotype and one with WHO 2016 defined del(5q) MDS. The PPM1D mutations co-occurred with CSNK1A1, SF3B1, ETV6, KIT, ASXL1, TET2 and DNMT3A mutations. Three of the 13 (23%) PPM1D-mutated patients had no additional mutations. Also, TP53 mutations were found in 35 of 234 (15%) patients. Twelve of the 35 (34%) TP53-mutated patients had a complex karyotype. We next investigated the prognostic impact of PPM1D mutations in 164 WHO 2016 defined del(5q) MDS patients (Tables S1 and S3). This cohort included 11 PPM1D-mutated patients, 16 PPM1D-wildtype/TP53-mutated patients and 137 PPM1D-/TP53-wildtype patients. All TP53 mutations were monoallelic in this group (supplementary methods). The PPM1D mutated patients were numerically older compared to PPM1D/TP53 wildtype patients (78.3 vs 71 years, p = .31). After a median follow up of 2.6 years, two of 11 (18.2%) PPM1D-mutated patients transformed to AML. The AML transformation rate was 6.3% for TP53-mutated/PPM1D-wildtype patients and 20.4% for PPM1D-/TP53-wildtype patients (Table S3). None of the 11 PPM1D-mutated patients and one of the 16 PPM1D-wildtype/TP53-mutated patients underwent HCT. The 2-year OS was 100% for PPM1D-mutated patients (n = 11) and PPM1Dwt/TP53mut patients (n = 16), and 85% for PPM1D-/TP53-wildtype patients (n = 137) with WHO 2016 defined del(5q) (Figure 1(A)). For multivariate analysis four variables were considered based on univariate analysis (age, sex, IPSS risk group, PPM1D mutation status). Only age and IPSS risk group were independent predictors of OS (Table S4). We then investigated the prognostic effect of PPM1D mutations in 52 patients with other MDS with del (5q) (n = 31) and sAML (n = 21) with del(5q) (Table S1). Note, PPM1D was mutated in two of 52 patients, both showing a concurrent TP53 mutation and complex karyotype. Thus, we could not evaluate the prognostic effect of PPM1D independently of a complex karyotype and a TP53 mutation. Nine patients had a monoallelic and six patients a biallelic TP53 aberration. Overall survival was shorter for the TP53mut monoallelic ± PPM1Dmut patients (n = 9) and significantly shorter for the TP53mut biallelic ± PPM1Dmut patients (n = 6) compared to TP53-wildtype and PPM1D-wildtype patients (n = 37; 2-y-OS 11% vs 0% vs 53%, respectively, Figure 1(B)). To evaluate the hematologic response to LEN in WHO 2016 defined del(5q) MDS we analyzed 65 LEN treated patients (Tables S1 and S5). Nine of 65 (13.9%) patients were TP53 (n = 5, 7.7%) or PPM1D-mutated (n = 4, 6.2%). Of 65 patients with WHO 2016 defined del(5q) MDS who were treated with LEN, 54 achieved hematologic response (83.1%) and 11 (16.9%) did not. Treatment response was independent of PPM1D (p = .35) (Figure 1(C)) or TP53 (p = .15) mutation status (Figure 1(D)). After a median follow up of 3.1 years, 40 of the 65 (61.5%) LEN treated patients became refractory or progressed to AML (Figure S2(A)). The median time to AML progression was 2.6 years. The rate of LEN resistance or disease progression was independent of the PPM1D (p = .62, Figure S2(B)) or TP53 (p = .38) mutation status (Figure S2(C)). Lastly, we investigated clonal evolution under LEN treatment. Follow-up samples were available after LEN treatment for 22 patients with MDS with del(5q) (19 of 22 with WHO 2016 defined del(5q) MDS) (Table S1), who either achieved a complete remission (n = 5) or developed resistance to LEN, which was followed by MDS progression (n = 7) or AML transformation (n = 10). All samples were screened at diagnosis, time of LEN resistance and/or time of AML transformation by NGS (supplementary methods and Table S6). Of the five patients achieving complete hematological remission four patients displayed no mutations, while one patient was PPM1D-mutated and ASXL1-mutated prior LEN. After 76 months on LEN, the VAF decreased from 27.6% to 4.8% for PPM1D and from 12.1% to 1.1% for ASXL1 in this patient. Of the 17 patients with LEN resistance or MDS/AML progression, two patients (11.8%) carried mutations in PPM1D and three patients (17.6%) in TP53 prior to LEN treatment (p = .64). At the time of LEN resistance or MDS/AML progression, we observed three (17.6%) PPM1D-mutated and eight (47.1%) TP53-mutated patients (p = .03) (Figure 1(E),(F)). The one novel PPM1D and the five novel TP53 mutations were not detected in the diagnostic sample at a median sequencing depth of 2528 reads (range 1393–12583 reads) and a median limit of detection of 0.72% (range 0.56%–1.77%). Two of eight TP53-mutated patients co-expressed PPM1D mutations. The prevalence of PPM1D-mutated and/or TP53-mutated patients increased from 29.4% prior LEN treatment to 52.9% (p = .09) at the time of LEN resistance/progression (Figure 1(G)). At the time of LEN resistance or AML progression, the VAF of PPM1D mutations increased from 10.2% to 23.3% and of TP53 mutations from 5.9% to 23.2% (Figure S2(D),(E)). This corresponds to a 2.5% and 3% increase of the VAF per year in PPM1D-mutated and TP53-mutated patients, respectively. Novel ETV6, RUNX1, WT1, U2AF1, SF3B1 and SRSF2 mutations were observed in patients with LEN resistance or MDS/AML progression (Figure S3(A)–(I)). In summary, we found a 5.6% and 15% prevalence of PPM1D and TP53 mutations prior to LEN treatment, respectively in 234 MDS/sAML patients with del(5q). All patients with WHO 2016 defined del(5q) MDS harbored a TP53 monallelic state. PPM1D and monoallelic TP53 mutations had no prognostic impact in MDS patients with WHO 2016 defined del(5q), while TP53 mutations, especially when biallelic, predicted poor OS in patients with sAML and other MDS with del(5q). Furthermore, neither the hematologic response to LEN nor MDS and AML progression risk was affected by PPM1D and TP53 mutation status in patients with WHO 2016 defined del(5q) MDS, although this analysis is preliminary due to the limited number of patients bearing these mutations. Lastly, we found that LEN resistance and disease progression were associated with the acquisition of novel TP53 and PPM1D mutations and a VAF increase suggesting that hematopoietic clones with these mutations are less inhibited by the selective pressure of LEN than PPM1D and TP53 wildtype clones and therefore expand over time. Future studies need to investigate whether sequential genetic analysis for the detection of clonal evolution is useful to identify patients at risk of adverse outcomes and to choose an appropriate treatment to prevent transformation to AML. We would like to thank all participating patients, contributing doctors and our technicians Blerina Neziri and Martin Wichmann for their excellent support. This work was supported by an ERC grant under the European Union's Horizon 2020 research and innovation program (No. 638035), by grant 70 112 697 from Deutsche Krebshilfe, DFG grants HE 5240/6-1 and HE 5240/6-2 and DJCLS 06 R/2017 from Deutsche José Carreras Stiftung. P.V. was supported by the Austrian Science Fund (FWF) SFB project F4704-B20. Open Access funding enabled and organized by Projekt DEAL. The authors declare no conflict of interest. Written informed consent from patients was obtained according to the Declaration of Helsinki. V.P and M.H designed the research; V.P, M.M., A.K.,R.G., R.S, C.K, P.K, J.S, S.K, M.H. performed the research; M.M, J.K.,A.M, G.G, C.F, C.G, K.S., A.G. C.T., U.G., T.S., G.K, C.K., B.S., N.K, D.H., K.D., W.S., P.V., A.G, F.T., T.H., U.P. contributed patient samples and clinical data; V.P, M.M., R.G, M.H. analyzed the data; V.P and M.H wrote the manuscript. All authors read and agreed to the final version of the manuscript. The study was approved by the review board of Hannover Medical School (ethical vote 5558/2010). All data are available from the corresponding author and in the supplementary data file. Appendix S1: Supporting Information Figure S1 Flow Diagram of analyzed patients and location of PPM1D mutations. (A) Molecular analysis of 234 patients with MDS or sAML and del(5q). Overall survival (OS) information was available for 216 of the 234 patients, i.e. for 164 MDS patients with WHO 2016 defined del(5q), 31 patients with other MDS with del(5q) and 21 patients with sAML with del5(q). Sixty-five patients with WHO 2016 defined del(5q) MDS were treated with lenalidomide (LEN) and had available information about treatment response (LEN cohort). For 22 patients with WHO 2016 defined del(5q) MDS or other MDS with del(5q), follow up (FU) samples after LEN treatment were available and they were included in the follow up cohort. (B) Schematic PPM1D gene structure with localization and frequency of mutations in 234 MDS/sAML patients with del(5q) based on reference PPM1D sequence ENST00000305921.7. Figure S2 Impact of PPM1D and TP53 mutations on the development of lenalidomide resistance or disease progression of patients with WHO 2016 defined del(5q) MDS. (A) Development of LEN resistance or disease progression of 65 patients with WHO 2016 defined del(5q) MDS treated with LEN. (B) Frequency of PPM1D mutated patients who developed LEN resistance and/or progressed under LEN treatment in comparison to PPM1D wildtype patients. (C) Frequency of TP53 mutated patients who developed LEN resistance and/or progressed under LEN treatment in comparison to TP53 wildtype patients. (D,E) Variant allele frequency of PPM1D (D) and TP53 (E) mutations prior LEN treatment and at the time of LEN resistance or AML progression. Figure S3 Clonal evolution under lenalidomide treatment. (A-J) Graphical illustration of the clonal evolution in PPM1D and/or TP53 mutated patients (n = 9) at the time of LEN resistance or disease progression under LEN treatment. Shown are time of diagnosis, time of LEN resistance or AML transformation and the variant allele frequency at each timepoint. Table S1 Comparison of clinical characteristics of MDS/sAML patients with del(5q). Table S2 Cycling conditions and primers for PPM1D sanger sequencing. Table S3 Comparison of clinical and molecular characteristics of 164 MDS patients with WHO 2016 defined del(5q)* and available survival information. Table S4 Univariate and multivariate analysis for OS in 164 MDS patients with WHO 2016 defined del(5q)* and available survival information. Table S5 Comparison of clinical and molecular characteristics of 65 MDS patients with WHO 2016 defined del(5q)* treated with lenalidomide and available response and survival information. Table S6 Genes covered by our custom myeloid panel for NGS analysis. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
Half of the myelodysplastic syndromes (MDS) have normal karyotype by conventional banding analysis. The percentage of true normal karyotype cases can be reduced by 20-30% with the complementary application of genomic microarrays. We here present a multicenter collaborative study of 163 MDS cases with a normal karyotype (≥10 metaphases) at diagnosis. All cases were analyzed with the ThermoFisher® microarray (either SNP 6.0 or CytoScan HD) for the identification of both copy number alteration(CNA) and regions of homozygosity (ROH). Our series supports that 25 Mb cut-off as having the most prognostic impact, even after adjustment by IPSS-R. This study highlights the importance of microarrays in MDS patients, to detect CNAs and especially to detect acquired ROH which has demonstrated a high prognostic impact.
Background: Iron accumulation occurs in MDS patients by upregulated iron absorption due to ineffective erythropoiesis with decreased hepcidin levels ( Santini et al, 2011 ) and by chronic blood transfusion therapy. Elevated liver iron concentration (LIC) is regularly found in transfusion dependent patients. Abnormal cardiac iron (50–1) was observed in few studies and severe cardiac iron overload was almost not present although cardiac death appeared in 8% of deceased patients ( Nachtkamp et al, 2016 ). Little is known about corresponding pancreatic iron and iron accumulation in the bone marrow. The release of NTBI in the bone marrow may directly affect hematopoiesis by oxidative stress ( Porter et al, 2016 ). Aims: To compare the relationship of iron concentration and fat content in different organs in low ‐ risk MDS patients. Methods: We studied the iron accumulation in bone marrow, liver, spleen, pancreas and septal heart muscle in 14 mainly low‐risk MDS patients (age 39–87 y), partly transfused and chelated. Tissue iron concentration and fat content were assessed by magnetic resonance imaging (MRI ‐ R2 ∗ ) with 3D data acquisition at 3 Tesla using chemical shift relaxometry ( Pfeifer et al, 2015 ). Results: About 50% of patients showed iron overload in the spleen, pancreas and vertebral bone marrow (VBM). Mean LIC (±SD) was found as 1.709 ± 0.914 mg/g liver (10.3 ± 5.5 mg/g dry weight ). Patients with iron overload in the pancreas showed as well abnormal pancreatic fat contents (11–32%) and fasting glucose values (101–186 mg/dL). As expected, septal cardiac iron was normal for all patients. Bone marrow (VBM) iron was elevated in 8/14 patients (R2 ∗ = 149–330 s −1 ) with the highest R2 ∗ in a non‐chelated patient. Summary/Conclusion: The feasibility of multi‐organ iron and fat measurements by MRI in heart, liver, spleen, pancreas and bone marrow was shown in a small group of elderly MDS patients within a reasonable scan time (30 min). The iron concentration and fat content in different organs will contribute further information about disease process in low ‐ risk MDS patients, as well as success of iron chelation therapy.
Abstract In the current World Health Organization (WHO)-classification, therapy-related myelodysplastic syndromes (t-MDS) are categorized together with therapy-related acute myeloid leukemia (AML) and t-myelodysplastic/myeloproliferative neoplasms into one subgroup independent of morphologic or prognostic features. Analyzing data of 2087 t-MDS patients from different international MDS groups to evaluate classification and prognostication tools we found that applying the WHO classification for p-MDS successfully predicts time to transformation and survival (both p < 0.001). The results regarding carefully reviewed cytogenetic data, classifications, and prognostic scores confirmed that t-MDS are similarly heterogeneous as p-MDS and therefore deserve the same careful differentiation regarding risk. As reference, these results were compared with 4593 primary MDS (p-MDS) patients represented in the International Working Group for Prognosis in MDS database (IWG-PM). Although a less favorable clinical outcome occurred in each t-MDS subset compared with p-MDS subgroups, FAB and WHO-classification, IPSS-R, and WPSS-R separated t-MDS patients into differing risk groups effectively, indicating that all established risk factors for p-MDS maintained relevance in t-MDS, with cytogenetic features having enhanced predictive power. These data strongly argue to classify t-MDS as a separate entity distinct from other WHO-classified t-myeloid neoplasms, which would enhance treatment decisions and facilitate the inclusion of t-MDS patients into clinical studies.
