- Top of page
- Patients and methods
- Clinical data
- Transplant patients
Summary. We report on nine children with Shwachman–Diamond syndrome (SDS), eight of whom had clonal abnormalities of chromosome 7. Seven children had an isochromosome 7 [i(7)(q10)] and one a derivative chromosome 7, all with an apparently identical (centromeric) breakpoint. Children with SDS are predisposed to myelodysplasia (MDS) and acute myeloid leukaemia (AML) often with chromosome 7 abnormalities. Allogeneic transplants have been used to treat these children, however, they are a high-risk transplant group and require careful evaluation. Three of the children were transplanted but only one survived, who to our knowledge remains the longest surviving SDS transplant patient (4·5 years +). The six non-transplanted children are well. In classic MDS, chromosome 7 abnormalities are associated with rapid progression to acute leukaemia; however, we present evidence to suggest that isochromosome 7q may represent a separate disease entity in SDS children. This is a particularly interesting finding given that the SDS gene has recently been mapped to the centromeric region of chromosome 7. Our studies indicate that i(7)(q10) is a relatively benign rearrangement and that it is not advisable to offer allogeneic transplants to SDS children with i(7)(q10) alone in the absence of other clinical signs of disease progression.
Shwachman–Diamond Syndrome (SDS) is a rare autosomal recessive disorder affecting multiple organs with a wide range of clinical severity (Shwachman et al, 1964). It is characterized by exocrine pancreatic insufficiency, growth retardation, skeletal abnormalities and constitutional bone marrow failure, resulting in peripheral blood cytopenias. As a result, SDS patients are at high risk of developing myelodysplastic syndrome (MDS) and acute myeloid leukaemia (AML). There also appears to be a strong association with chromosome 7 abnormalities, especially i(7)(q10) (Smith et al, 1996; Dror et al, 1998; Okcu et al, 1998). Monosomy 7 and deletions of 7q are common, and invariably poor prognostic markers in MDS/AML (Labal de Vinuesa et al, 1987, Shannon et al, 1989; Murray et al, 1996). However, i(7)(q10), representing a duplication of the q arm, has rarely been reported with the exception of SDS patients. Isochromosomes are formed by misdivision of the centromere, which is interesting, as Goobie et al (2001) have recently mapped the SDS gene to the peri-centromeric region of chromosome 7 [maximum multipoint LOD (log of the odds) score of 10]. The relevance of this discovery and whether or not the SDS gene is affected by i(7)(q10) formation is as yet unknown.
In childhood MDS, bone marrow transplantation (BMT) has become accepted as the treatment of choice with a disease-free survival of 50% (Guinan et al, 1989). However, SDS is a high-risk transplant group, of the 15 patients (including three of our patients) who have undergone BMT only six have survived (40%) and nine (60%) died from a variety of transplant-related causes (Tsai et al, 1990; Barrios et al, 1991; Smith et al, 1995; Arseniev et al, 1996; Bunin et al, 1996; Davies et al, 1997; Okcu et al, 1998; Faber et al, 1999). In addition, although BMT can replace faulty haematopoietic progenitors, the underlying stromal defect remains (Dror et al, 1998; Dror & Freedman, 1999). Clearly this is a very high-risk transplant group with a background of underlying organ failure, which may be exacerbated by current transplant conditioning regimes (Savilathi & Rapola, 1984; Okcu et al, 1998).
Here we report on nine SDS children, of whom seven acquired an i(7)(q10) and one had a derived chromosome 7 with a centromeric breakpoint near the proposed site of the SDS gene. We suggest that i(7)(q10) may be a relatively benign finding in SDS and should not be considered in the same poor prognostic group as other chromosome 7 abnormalities in MDS and AML, and, after reviewing the differing outcomes of BMT, question whether allogeneic BMT should be carried out if i(7)(q10) is the sole abnormality.
Independent prognostic scoring system (IPSS). IPSS is calculated from cytogenetics, percentage blasts, haemoglobin level and platelet count. IPSS survival and prognostic data is based on elderly patients with MDS (Greenberg et al, 1997; Soléet al, 2000), therefore, where possible we have also included a Paediatric scoring system (PSS) based on childhood MDS (Passmore et al, 1995). The score is calculated from fetal haemoglobin (HbF), platelet count, neutrophil count, percentage blasts and cytogenetics.
Probes. The following probes were used: Williams syndrome critical region (WSCR) probe (7q11.23) FITC (fluorescein isothocyanate) (green) with D7S427 control probe (7q36) FITC (green) (Appligene Oncor, QBiogene, Middlesex, UK); whole chromosome 5 paint Cy3 (Red) (Cambio, Cambridge, UK); whole chromosome 7 paint FITC (green) (Cambio); chromosome 1/5/19 alpha satellite (D1Z7/D5Z2/D19Z3), Texas red (Appligene Oncor); chromosome 7 alpha satellite (D7Z1), FITC (green) (Appligene Oncor).
- Top of page
- Patients and methods
- Clinical data
- Transplant patients
A summary of serial cytogenetics results on all patients is presented in Table II. The numbers in square brackets  after the karyotype indicate the number of cells analysed.
Table II. Summary of cytogenetic results.
Figure 1A shows the partial karyotype and ideogram of the der(5;7) (patient A1). Figure 1B shows the partial karyotype and ideogram of the i(7q) (patients A2, B1, B2, C1, D1, D2 and E).
Figure 1. FISH studies were carried out on patient A1 and her sister A2 to identify the precise breakpoint on the derived chromosome. (A) Partial karyotype and ideogram of (left to right) chromosome 5, chromosome 7 and the derived chromosome der(5;7) from patient A1. (B) Partial karyotype and ideogram of chromosome 7 (left) and the isochromosome 7q (right) seen in patients A2, B1, B2, C1, D1, D2 and E. (C) The der(5;7) (arrowed) with whole chromosome 7 paint (green) and chromosome 5 alpha-satellite (red) showing that 5 centromere is present on the rearranged chromosome of patient A1. The normal chromosome 7 (green) can be seen below. (The centromeres of chromosomes 1 and 19 also hybridized with this probe). (D) The der(5;7) (arrowed) painted with whole chromosome 5 paint (red), and the William syndrome critical region (WSCR) 7q11.2 and reference probe (green), revealed that the breakpoint in patient A1 was proximal to the WSCR at 7q11.23. The normal chromosome 5 (red) is on the right and the normal 7 (green) above. (E) The der(5;7) (arrowed) with alpha satellite probes for chromosome 5 (red) and 7 (green) showing centromeric fusion. (The centromeres of chromosomes 1 and 19 also hybridized with this probe). (F) The i(7)(q10) (arrowed) from patient A2 was confirmed with (WSCR) 7q11.2 and reference probe in green. The normal chromosome 7 (green) is on the right. FISH confirmed that the breakpoint on both sisters (patients A1 and A2) is at the centromere of chromosome 7. In C–F, an orange arrow indicates the abnormal chromosome.
Download figure to PowerPoint
- Top of page
- Patients and methods
- Clinical data
- Transplant patients
Cytogenetic analysis of bone marrow from nine SDS children confirmed a high incidence of i(7)(q10), when combined with published SDS patients, a striking 75% have chromosome 7 abnormalities, 46% having i(7)(q10) alone (Table III). Patient A1 had a centromeric breakpoint and a similar patient has been reported (Smith et al, 1995), suggesting that these may represent variants of i(7)(q10). Subtle centromeric heteromorphism of chromosome 7 has also been reported (Sokolic et al, 1999). Whole arm rearrangements are thought to be associated with dosage effects rather than locus-specific disruption (Labal de Vinuesa et al, 1987). However, the possibility of a qualitative disruption of a critical gene must be re-evaluated, following the mapping of the SDS gene to the peri-centromeric region of chromosome 7 (Goobie et al, 2001). This is the first report of the mapping of a cancer predisposition gene to the breakpoint of an acquired chromosomal abnormality. Furthermore, the possibility of a chance association must be negligible given the high LOD score and the very specific nature of the acquired abnormality. Mutations in the SDS gene may predispose to isochromosome formation or i(7)(q10) formation may disrupt the gene. A full understanding awaits identification of the SDS gene itself.
Table III. Clonal chromosome abnormalities in patients with Shwachman–Diamond syndrome.
|Cytogenetic abnormality||Diagnosis||Age (years)||Status||Reference|
|46XX,i(7)(q10)/46,XX||RA|| 1·5||Alive||Patient B2, present study|
|46,XY,i(7)(q10)/46,XY||NP, TCP, A, PCP||13||Alive||Patient C1, present study|
|46,XY,i(7)(q10)/46,XY||NP, TCP|| 7||Alive||Patient D1, present study|
|46XX,i(7)(q10)/46,XX||PCP|| 8||Alive||Patient D2, present study|
|46,XY,i(7)(q10)||NP, TCP|| 7||Alive||Patient E, present study|
|46,XY,i(7)(q10)/46,XY||RA|| 2||Alive||Smith et al (1996)|
|46,XY,i(7)(q11)||RCP|| 6||Alive||Dror et al (1998)|
|46,XY,i(7)(q10)/46,XY,del(20)(q11)/46,XY||RCP|| 5||Alive||Dror et al (1998)|
|46XX,i(7)(q10)||NP, TCP|| 6||Dead||Maserati et al (2000)|
|46XX,i(7)(q10)/46,XX||NP||16||Dead d 71 post MMD||Patient A2, present study|
|46XX,i(7)(q10)||RA|| 3·5||Dead d 31 post MMD||Patient B1, present study|
|46XX,i(7)(q10)/46,XX||Hypocellular|| 8||Dead 2 months post MMD||Davies et al (1997)|
|45,XY,−7/46,XY,i(7)(q10)||Hypoplasia, MF|| 9||Dead d 32 post BMT||Okcu et al (1998)|
|46,XY,der(7)t(4;7)(q31;q11)/46,XY||RA > AMLM5|| 5||Dead 1 years post BMT||Smith et al (1995, 1996)|
|45XX,der(5;7)(p10:q10)/46,XX||RA||12||Alive 3·5 years post MMD||Patient A1, present study|
|46,XY,del(7)(q22q34)/47,XY,del(7)(q22q34),+21/46,XY||NP > MDS (?)||13||Alive 12 months + post MUD||Kalra et al (1995), Davies et al (1997)|
|46XX,del(7)(q11.2q32)/46,XX||RA||11||Alive||Smith et al (1996)|
|46–47,XY,−2,−4,del(5)(q23q33),del(7)(q22),+2–3r,+2–4mar||RAEBT > AMLM6||42||Dead||Smith et al (1996)|
|45XX,t(6;13)(q21;q32),−7/46,XX||PCP > MDS -AML|| 7·5||Dead 2months post MUD||Kalra et al (1995), Davies et al (1997)|
|45,XY,−7,mar(18)||AML|| 9||Dead sepsis||Woods et al (1981)|
|45,XY,−7||RAEB|| 8||Dead d 93 post MUD||Okcu et al (1998)|
|47,XY,+1,del(9)(q22)||RAEBT > AMLM5|| 9||Dead||Smith et al (1996)|
|46,XY,inv(14)(q11q32)||MDS (?)|| 5||Alive 18 months + post MSD||Faber et al (1999)|
|46,XY,inv(9)||MDS > AMLM4||24||Dead 10 months post MSD||Arseniev et al (1996)|
|46,XY,add(11)(p?),−15,−22,+mar1,+mar2||RAEBT > AMLM2|| 8||Dead||Smith et al (1996)|
|45–50,XY,−18,t(21;?)(q22;?),dic(22;?)(p11;?)/46,XY||AML||38||5 years +||Seymour & Escudier (1993)|
|47,XY,+21,+4q,mar(1q)||PCP > MDS > AMLM4|| 2·5||Dead||Woods et al (1981)|
|53,XY,+G,+G||ALL L1|| 1·5||Alive 12 months +||Woods et al (1981)|
Although SDS patients are predisposed to myeloid disease, there is not the same clear relationship between chromosome 7 abnormalities and leukaemia, as seen in non-SDS patients. In many patients, the i(7)(q10) has been identified through routine monitoring and may previously have gone unnoticed in the absence of disease progression. There is compelling evidence to suggest that i(7)(q10), resulting in duplication of 7q, is cytogenetically and clinically distinct from other categories of chromosome 7 abnormality. MDS is commonly associated with monosomy 7, deletion of 7q, and complex abnormalities of 5q and 7q, where they define a poor prognostic group associated with rapid transformation to acute leukaemia (Johansson et al, 1991; Nowell, 1992). However, Anderson and Pederson-Bjergaard (2000) found no instances of i(7)(q10) in a total of 411 de novo and treatment-related MDS/AML patients, despite a high incidence of centromeric breaks.
SDS patients with i(7)(q10) do not show classic myelodysplastic features, do not appear to transform to AML and rarely develop secondary changes during disease progression.
Patients D1 and D2 had spontaneous reductions in clone size without treatment, and E had an improvement in bone marrow morphology with no change in level of the i(7)(q10) clone (Table II). Prior to transplant, patient B1 had i(7)(q10) for 5 years, and Maserati et al (2000) reported a similar patient with an i(7)(q10) clone for 9 years who eventually died of aplasia, not AML or MDS. This patient may provide some insight into the natural history of i(7)(q10) and would support our own patient data. Clearly these i(7)(q10) patients are inconsistent with the rapid expansion of an aggressive chromosome abnormality. No SDS children with i(7)(q10) have transformed to acute leukaemia and there is no evidence of additional abnormalities occurring within the i(7)(q10) clone, although separate minor del(20q) clones have been reported (patient A2 and Dror et al, 1998). Taken together, we believe there is evidence to suggest that isochromosome 7q is specific to SDS and is a different, much less aggressive, disease entity than other chromosome 7 abnormalities seen in MDS/AML. This of course has profound implications for the management of children with SDS.
As a result of the prior association of chromosome 7 abnormalities and poor prognosis, SDS children with i(7)(q10) may be offered transplants. However, the ethical decisions are difficult, SDS patients have a high incidence of transplant-associated problems; 46% survived for 9 months and 40% are still alive as far as we are aware (Savilathi & Rapola, 1984; Tsai et al, 1990; Okcu et al, 1998) (Table IV). If pulmonary function is compromised, and repeated infection and transfusion dependency are high, then transplant may be the only option; on the other hand, transplanting while a child is well and infection free may increase survival chances, but could risk transplant-related death. Faber et al (1999) suggested that early childhood transplants may be beneficial, as cardiotoxicity is reduced and treatment options are not yet limited by transformation to AML. While patient A1 and others in the literature show that transplants can be successful, none of these cases had an i(7)(q10) (Smith et al, 1995; Davies et al, 1997; Faber et al, 1999). Of the 13 i(7)(q10) SDS cases, four children (4/4) have died as a result of BMT (< 3 months post transplant), one of aplasia (Maserati et al, 2000), and eight non-transplanted cases are alive and have not transformed to acute leukaemia (Tables III and IV). Obviously a large collaborative study is required to determine the true prognostic significance of i(7)(q10) in SDS and how this may impact on patient management.
Table IV. Bone marrow transplants in SDS patients.
|Diagnosis||Type||Complications||Indications for transplant||Survival||Reference|
|NP||MMD||Adenovirus||CA||Dead 71 d||Patient A2, present study|
|MDS (RA)||MPD||Cardiac failure||CA, pulmonary/ renal impairment||Dead 1 month||Patient B1, present study|
|MDS (RA)||MMD||GVHD, PH||CA, myelofibrosis||Dead 32 d||Okcu et al (1998)|
|Hypocellular||MMD||Respiratory distress||CA, hypocellular||Dead 2 months||Davies et al (1997)|
|MDS (RA)||MMD||None||CA||Alive 52 months +||Patient A1, present study|
|AML M5a||MUD||Graft rejection||CA, AML M5||Dead 12 months||Smith et al (1995)|
|Hypocellular, dysplastic changes||MUD||None||CA, hypocellular||Alive 12 months +||Davies et al (1997)|
|MDS unclassified||MUD||Graft failed, Infection||CA, hypocellular||Dead 2 months||Davies et al (1997)|
|Aplasia||MSD||Cardiac failure||Repeated pyrogenic infection||Dead 23 d||Tsai et al (1990)|
|Aplasia||MSD||GVHD||Pneumonia, renal failure||Alive 9 months +||Barrios et al (1991)|
|Aplasia||MMD||Veno-occlusive disease||Transfusion dependent,||Alive 10 months +||Bunin et al (1996)|
|MDS (RAEB)||MUD||GVHD||CA||Dead 93 d||Okcu et al (1998)|
|MDS hypoplasia, dyserythropoesis||MUD||None||CA, Pyrogenic infection.||Alive 18 months +||Faber et al (1999)|
|MDS > AMLM4||MSD||Cytomegalovirus,||Severe transfusion dependent||Dead 10 months||Arseniev et al (1996) |
|AML||MUD||Hepatic/cardiac dysfunction||CA||Alive 97 d +||Seymour & Escudier (1993)|
Accumulated evidence strongly suggests that i(7)(q10) is specific to SDS, represents a different disease entity from other chromosome 7 abnormalities in myeloid leukaemia and should not be considered in the same poor risk category. Although SDS children are at high risk of developing MDS/AML, they are prone to lethal transplant complications and transplant decisions should not be based on the presence of i(7)(q10) alone in the absence of other signs of haematological disease progression.