Here we reported two cases of DI-MDS with monosomy 7. In our case 1, CDI was the initial manifestation of MDS, which might have led to misdiagnosis or delayed treatment. MDS associated with DI has rarely been reported. To our knowledge, only five MDS cases with CDI have been reported till now [4, 6, 7, 9, 10]. The information of reported DI-MDS cases was summarized in Table 2.
Table 2
Characteristics of five reported cases with MDS and CDI
Reference | Age (years) | MDS subtype | Partial/complete deletion of chromosome 7 | MRI abnormal | Treatment of MDS | Outcome of CDI | Time to AML (months) | OS (months) |
4 | 74 | RAEB 1 | Yes | No | Supportive care | Controlled by desmopressin | 2 | 2 |
6 | 6 | RAEB 1 | Yes | No | Allo-HCT | Controlled by desmopressin and cured after HCT | No | NA |
7 | 53 | RAEB 2 | No (Norma karyotype) | Nodular lesion on pituitary stalk & absent of posterior “bright spot” of neurohypophysis on T1-weighed MRI | Chemotherapy and Allo-HCT | Controlled by desmopressin and need for desmopressin persists after allo-HCT | No | 18+ |
9 | 60 | MDS-MLD | No (Norma karyotype) | Attenuation of “bright spot” | Chemotherapy | Recovered after chemotherapy | 1 | NA |
10 | 73 | NA | Yes | Absent of posterior “bright spot” & symmetrical enhancing lesions in the hypothalamus | NA | Temporary controlled by desmopressin | 3 | 3 |
MDS: myelodysplastic syndrome; CDI: central diabetes insipidus; MRI: magnetic resonance imaging; NA: not available; AML: acute myeloid leukemia; OS: overall survival; RAEB: refractory anemia with excess blasts; Allo-HCT: allogeneic hematopoietic cell transplant; MDS-MLD: MDS with multilineage dysplasia. |
Although the reason for why DI occurs in MDS is unclear, the co-occurrence of AML and DI has several possible explanations. For instance, leukemic infiltration of the pituitary gland or hypothalamus, leukostasis, thrombosis, hemorrhage and infection have been presumed as causes. In our case 1, the MRI revealed a slightly thickened pituitary stalk and a small nodule in his left pituitary, which may indicate a pituitary infiltration. It is worth noting that the WBCs of both patients were lower than normal, which makes leukostasis unlikely.
The chromosomal anomalies we identified were closely related to AML/MDS with DI. Partial or complete monosomy of chromosome 7, which is the most commonly reported chromosome abnormality in DI-AML, was detected in both cases by MC analysis or SNP-based microarray; this abnormality was also found in 3 out of 5 reported cases of MDS with DI [4, 6, 10]. One possible explanation for this correlation is that monosomy 7 may affect expression of the neutrophil migration gene located on the 7q22 gene region. This impairs the migratory and chemotactic functions of neutrophils and may be related to blast infiltration of the pituitary gland in these patients [11, 12].
A few studies have suggested the prognosis of DI-AML to be poor [1, 8, 13]. It is reasonable to suspect a poor prognosis for DI-MDS based on the similar cytogenetic abnormality in DI-AML. In all three reported DI-MDS who didn’t perform allogeneic HCT, progression to AML occurred within three months [4, 9, 10]. In our case 2, despite being treated with decitabine, rapid progression to AML occurred. These results suggest allogenic HCT may be the only effective therapy for DI-MDS and should be perform as soon as possible. In all reported cases and our cases, the symptoms of polydipsia and polyuria could be controlled by desmopressin [4, 6, 7, 9, 10]. Desmopressin was no longer needed after MDS were well controlled in our case 1 and two reported cases [6, 9]. However, need for desmopressin persisted even after allogenic HCT in one case [7]. Both of our cases showed fluctuation of the severity of DI with MDS status. Thus, it would be worthwhile to investigate how MDS status influences the incidence or severity of DI in the milieu of fewer blasts.
Cytogenetic aberrations have important diagnostic, prognostic and therapeutic roles in MDS. However, the “false normal karyotype” of MC analysis often occurs because of the lack of metaphase nuclei in MDS. FISH and SNP-A-based karyotyping don’t rely on metaphase nuclei, while FISH is limited to detection of the known lesions. SNP-A-based karyotyping can reveal unbalanced defects in as few as 10% of cells [14], thus to identify cryptic abnormalities that are below the resolution of MC analysis. Meanwhile, SNP-A-based karyotyping can identify segmental uniparental disomy (UPD) that is undetectable by MC or FISH. Tiu et al. reported that SNP-A-based karyotyping detected chromosomal defects in 54% of MDS with normal MC results, and the presence and number of new SNP-A detected lesions, including UPD, are independent predictors of overall and event-free survival [15]. Recently, Yang Ou et al. reported the presence of UPD was an independent prognostic factor in MDS patients with normal karyotype [16]. Makishima H. et al. revealed that combined SNP-A karyotyping with routine MC in MDS improved the cytogenetic detection of monosomy 7, del (7q), del (5q), del (20q) and trisomy 8 [17]. By combining SNP-A-based karyotyping with MC, we uncovered additional cytogenetic abnormalities in our case 1, especially those associated with monosomy 7 and AOH. It is difficult to judge the AOH from somatic or germline cells because of the absence of germline controls for our patients. Until now, we have been unable to associate AOH with DI-MDS. In our case 1, monosomy 7 was not found by MC but was detected by the SNP-A-based karyotyping. These results indicate that the significance of a “normal karyotype” in reported DI-MDS cases may need to be reconsidered. The SNP-A analysis achieved superb resolution and can detect UPD, which overcomes the technical limitations of MC or FISH, such as the need for metaphase nuclei, hypocellularity, incompetence in unknown lesions and relatively low resolution. Thus, application of this technique to assess MDS patients with DI will help us to further elucidate the pathogenesis of DI- MDS.