Page 193 - 2021_06-Haematologica-web

MDS progression involves cohesin and RAS mutations
Discussion
Our study characterizes the landscape of mutations during progression toward sAML and how mutations, alone or in combination, contribute to leukemic transfor- mation, based on the analysis of a large number of serial samples from patients who progressed to sAML. Moreover, the comparison of a control and a validation cohort supports the identification of mechanisms mostly related to evolution to sAML.
Previous studies have documented the clonal evolution and greater clonal heterogeneity during cancer develop- ment and, specifically, as MDS evolves to advanced stages and transforms into sAML,8,9,14-16,20 and according to this fact, the study demonstrates that MDS patients, irrespective of their diagnostic subtype, gain more muta- tions and a higher VAF during disease progression. Therefore, these results are evidence that progression toward sAML is associated with pronounced genomic instability and a heavy mutational burden.
The mechanisms of disease evolution showed a great heterogeneity.14-17 However, this study identified four dis- tinct types of mutational dynamics and their roles during sAML progression. It is of particular note that this study shows a higher incidence of mutations in the cohesin complex and Ras signaling genes than in previously pub- lished MDS series,8,9,21,22 probably because this cohort consisted only of patients who progressed to sAML. Moreover, increasing mutations (type 1) were mainly found in genes of this pathway, such as STAG2, most of which were loss of function mutations. Previous studies of cohesin mutations had already shown that these muta- tions could be related to sAML progression,15,21,22 but the dynamics observed in this study confirm that mutations of cohesin complex genes could be an early event in sAML progression, and the loss of function of these genes could play an important role in the leukemic transforma- tion. Similarly, the dynamics of Ras signaling genes observed here, mainly newly acquired mutations (type 3), highlight the importance of this pathway in sAML pro- gression. Earlier studies have described how alterations in Ras pathway genes, such as NRAS and FLT3, could drive the progression to sAML23-25 but these results confirm that they are late events that may drive leukemic transfor- mation. Furthermore, this study reveals a significant co- occurrence of mutations in these two pathways and, also, in the main genes of these pathways (STAG2 and NRAS), excluding the FLT3-ITD mutations which could have a different behavior than other mutations in Ras pathway and they could be involved in an independent mechanism of sAML progression.26,27 Although Walter et al.28 described that NRAS and cohesin mutations tend to be mutually exclusive, this work included a low number of patients and, conversely, this co-occurrence was briefly described in another study with a higher number of patients.21 Thus, this study, due to detecting in a cohort of sAML-progressing patients, demonstrates that this pro- gressive combination of the cohesin complex, mainly STAG2, and Ras signaling mutations, mostly NRAS, could play an important role in the progression of MDS to sAML. In addition, the results from the validation cohort confirmed the impact of this co-occurrence on sAML pro- gression not only in the discovery cohort, but also in the validation cohort. Therefore, these findings support a hypothesis of genetic “predisposition”, that early muta-
tions shape the future trajectories of clonal evolution from MDS to sAML. Therefore, a new model of genetic evolution could be suggested consisting of cohesin muta- tions as an early event in the evolution of the disease that trigger to acquire new mutations, mainly Ras signaling mutations. Consequently, this clone expands, driving the disease evolution (Figure 3, model of clonal evolution using Fishplot R package).29 Recent studies have described that cohesins are involved in DNA damage repair, chro- matin accessibility and transcription factor activity30-32 and our results show than cohesin-mutated patients dis- played a higher number of mutations, thereby cohesins could cause an instability where new mutations are gen- erated, mainly Ras mutations, leading the disease pro- gression. These are not absolute rules and, unfortunately, this novel model does not fully explain the progression to sAML, but it could explain the evolution in 15-20% of all sAML transformations. Moreover, cohesin mutations potentiate the subsequent acquisition of Ras mutations, so these mutations could be used to identify patients whose disease is progressing before symptoms associated with progression to sAML are manifested.
On the other hand, the mutations that were stable dur- ing the evolution (type 4) were found in splicing and DNA methylation genes. This is in line with the finding of some recent reports showing that variants affect these pathways in this steady-state pattern.16 Moreover, muta- tions in DNA methylation and RNA splicing pathways are well known to have a heavier mutational burden than those in other genes, suggesting an early event in MDS development.8,9 Considering these findings together, these results showed that mutations in these pathways, which have high VAF and are stable during the disease evolution, could be directly involved in MDS pathogene- sis (driver role) but not in sAML progression (passenger role). The mutations whose VAF decreased during pro- gression (type 2) were distributed randomly throughout all the genes without showing a particular pattern. This could be the result of clone sweeping, a previously described event,15 that is specific to each patient rather than to a specific pattern of each gene or pathway.
Furthermore, a mechanism that could be linked to the evolution of patients who receive disease-modifying treatment before progression to sAML was detected. Several studies have described the mutational dynamics in treated MDS patients and have demonstrated that ther- apy alters clonal distribution, but the predictive impact of the dynamics is still unclear.33-37 A significantly higher proportion of newly acquired or increasing mutations in chromatin modifiers at the time of sAML was identified in treated patients. Thus, these results suggest that muta- tions in chromatin-modifier genes could be related to the evolution of treated patients. However, more studies with larger numbers of patients are required to validate this result.
In summary, MDS progression to sAML is character- ized by greater genomic instability, irrespective of the MDS subtypes at diagnosis, and there are four types of mutational dynamics during the disease evolution, increasing and newly acquired mutations (type 1 and type 3, respectively) being of particular importance. Moreover, a co-occurrence of cohesin complex and Ras signaling mutations could play an important role in the 15-20% of MDS patients who evolved to sAML. With regard to treatment, we found that mutations in chro-
haematologica | 2021; 106(8)
2221