Abstract
MM patients are living longer with increasingly effective therapies, but long-term complications including second primary malignancies (SPMs) are becoming new challenges in designing optimal patient care. It has been demonstrated in large studies that amongst others, risk is particularly high for SPMs such as myelodysplastic syndromes (MDS) and acute myeloid leukemia (AML). Importantly, such increased risk of MDS/AML has also been observed in MGUS patients, suggesting that increased risk for MDS/AML may not only be treatment related but inheritably high in MGUS/MM. Thus, there is need to investigate for biomarkers that uncover cellular alterations predisposing for higher risk of MDS/AML in MM.
Here, we started by investigating in 312 newly diagnosed MM patients the presence of MDS-like phenotypic abnormalities (MDS-PA) in bone marrow (BM) neutrophil, monocytic, and erythroid lineages, using multidimensional flow cytometry 8 color combinations (CD138, CD27, CD38, CD56, CD45, CD19, CD117, CD81; and HLADR, CD45, CD36, CD13, CD34, CD117, CD11b, CD71). Up to 33/312 (11%) patients showed MDS-PA at diagnosis, which were more frequently observed in the neutrophil lineage (6%), followed by monocytic (4%) and erythroid (4%) lineages. Four cases had multilineage MDS-PA.
Afterwards, we investigated if the presence of MDS-PA was associated with underlying somatic mutations by performing targeted sequencing of 54 MDS/AML related genes (depth ≥500x) in 44 patients from the previous series (10 with MDS-PA and 34 without). Next generation sequencing was performed, at diagnosis and after HDT/ASCT in FACS sorted CD34+ hematopoietic stem cells (HSCs) and dysplastic cell lineages from patients with MDS-PA, as well as in HSC from cases without MDS-PA. CD138+ BM plasma cells (PCs) from both cohorts were also sequenced using the same panel. Six out of the 10 cases with MDS-PA showed somatic mutations. Namely, HSCs from one patient had two mutations in TET2 [allele fraction (AF): 18%, ≥ 26017x] one in CALR (AF: 14%, 1158x) and another in ASXL1 (AF: 7%, 1339x). None of these mutations were present in myeloid/erythroid cells. A second patient had NPM1 mutated in HSCs (AF: 7%, 12825x), which was absent in neutrophils. A third case had TET2 mutated in HSCs (AF: 16%, 1233x) as well as in dysplastic monocytes (AF: 27%, 16647x) and neutrophils (AF: 23%, 21719x). In the fourth case, a mutation in BCORL1 was noted in dysplastic erythroid cells (AF: 10%, 796x). The fifth patient had TET2 mutated in both HSCs and dysplastic monocytes (AF: 45%-63%; ≥21799x). The sixth case had PHF6 mutated in HSCs (AF: 8%; 800x). In none of the patients were the mutations found in HSCs and/or dysplastic lineages, present in PCs. Within the control cohort of the 34 patients without MDS-PA, only two of them displayed somatic mutations in HSCs; one case had DNMT3A mutated (AF: 26%, 1900x) and the other TET2 (AF: 13%, 3400x).
After demonstrating a correlation between MDS-PA and MDS/AML-related somatic mutations, we sought to analyze the prognostic significance of such alterations in MM. Since the follow-up of the present series of 312 cases is relatively short, we focused on a large series of 965 patients with longer follow up (median of 6.5 years) enrolled in GEM clinical trials, and for which the presence of CD56+ aberrant monocytes could be readily investigated. Noteworthy, this particular MDS-PA was again observed in a similar frequency as noted above (n=63; 6.5%) and as compared to the overall MM population, patients with MDS-PA showed significantly higher age, lower hemoglobin values and higher BMPC infiltration at diagnosis. Furthermore, they experienced more frequently hematological toxicity including anemia and neutropenia during treatment. Most interestingly, as compared to the overall MM population, patients with MDS-PA had significantly inferior progression-free (medians of 24 vs 37 months; P=.006) and overall survival (medians of 47 vs 73 months; P=.01).
In conclusion, we showed for the first time that a fraction of newly diagnosed MM patients harbors MDS/AML-related somatic mutations in HSCs and myeloid/erythroid lineages, and that such patients could be predicted through flow-based screening for MDS-PA. The presence of MDS-PA identifies a subset of patients that experience more frequently hematological toxicity and display inferior survival; accordingly, screening for MDS-PA could become an important biomarker to tailor treatment in MM.
Paiva:Celgene: Honoraria, Research Funding; Janssen: Honoraria; Takeda: Honoraria, Research Funding; Sanofi: Consultancy, Research Funding; EngMab: Research Funding; Amgen: Honoraria; Binding Site: Research Funding. Oriol:Amgen: Honoraria, Other: Expert board committee; Janssen: Honoraria, Other: Expert board committee. Mateos:Amgen: Honoraria; Takeda: Honoraria; Celgene: Honoraria; Janssen: Honoraria.
Author notes
Asterisk with author names denotes non-ASH members.