Study of the Correlation between Phenotype - Including Acute Leukemia/MDS- and Genotype in Shwachman Diamond Syndrome in 60 Patients from the French SCN Registry.

Shwachman Diamond syndrome (SDS) is a rare multi organ genetic disease bearing a very high risk of haematological complications i.e. MDS/leukaemia and Bone Marrow Failure. The aim of this study is to explore genotype predisposition of the major complications observed in SDS’s patients and to explore prognosis factors of MDS/leukaemia.

Methods: Among 90 SDS patients included in the French Severe Chronic Neutropenia Registry, SBDS gene was screened in 63 patients and mutations have been found in 60 patients. Cut-off date was july 30th, 2007. Differences between groups of patients were analysed as survival data, by log rank test. The medical events analysed were: death (n=6), myelodysplasia or acute leukemia (n=6), bone marrow failure (n=6), all hematological events combined (n=12), the use of G-CSF as infectious prophylaxis (n= 11), the necessity of an orthopedic surgery (n=4) and the necessity of nutritional medical support (parenteral or enteral feeding, by mean of gastrostomy (n= 5), intrauterine growth retardation (n=19) and finally, major development retardation if it leads to a specialized school (n=10).

Results: Mutations were found in 60 patients (35 males, 25 females) belonging to 54 distinct families (in 6 families, two siblings were genotyped). The median age at last analysis was 10.3 years (0.5yr-38.6 yr). The great majority of patients present the recurrent genotype K62X/C84fs (n=38, 68%) while 19 other mutations were founded, which could be classified in truncating mutations leading to premature stop codons (nonsense, frameshift or splicing defect; n=8) or missense mutations (n=11). We compared patients with truncating mutations on both alleles to compound heterozygous patients carrying at least one missense mutation. Even if differences were observed for the distribution of events between genotype subgroups of patients, none of them raised statistically significance. However, to date, all leukemia has been observed in the group of patients with “truncating” mutations. The genotype of patients with leukemia was [K62X]+[C84fs] in 5 and [C84fs]+[V93fs] in one; while the genotype of patients with BM failure was [K62X]+[C84fs] (n=2), [C84fs]+[624+1G>C], [C84fs]+[C119R], K62X/undetermined, and [C84fs]+[E99fs], [C84fs]+[E44fs]. Among the 6 pairs of siblings tested, four had a similar outcome and two pairs were discordant for the haematological events (leukaemia in one family, Bone marrow failure in the second family). Further, we have analysed genotype, gender, G-CSF therapy and initial Neutrophils and monocytes count, Hemoglobin level, Platelet level as risk factors of Leukemia/MDS. In a multivariate model, none of these features predicts Leukemia/MDS in SDS patients.

Conclusion: The genotype of SDS did not appear to be correlated with clinical presentation or outcome. It remains possible than patients without truncating mutations (about 18%) may have a low rate of leukaemia but our survey lack of statistical powerful to demonstrate this hypothesis. We also failed to determine prognostic factors of Leukemia/MDS in SDS patients.