Chronic Relapsing Remitting Sweet’s Syndrome- a Harbinger Of Myelodysplastic Syndrome (MDS), Single Centre Analysis Of 31 Patients

Sweet’s syndrome (SS) is an acute febrile neutrophilic dermatosis. It has been associated with malignant disease, especially acute myeloid leukaemia (AML) and drugs, particularly granulocyte colony stimulating factor (GCSF). No cause is found in the rest and is labeled idiopathic. We describe 31 patients with a readily diagnosed form of SS, which we believe represent an autoimmune phenomenon secondary to the myelodysplastic syndrome (MDS).

A retrospective study was conducted to identify patients with SS with underlying diagnosed or occult haematological disorders over a 7 year period. The skin histology was reviewed independently by histopathologist and additionally frequency of CD4+ and CD8+ T cells, B cells and NK cells were investigated in 6 patients with chronic relapsing SS in comparison with 4 healthy age matched donors.

We identified 31 patients with SS in a cohort of 744 patients with MDS and 215 with AML seen between 2004-2011. The median age was 58 yrs (37-82 yrs), with male female ratio of 1.2;1 (male 17, female 14). Of these, 74% (N=23) were associated with myelodysplastic syndrome, 13 %( N=4) with AML, 6% (n=2) with chronic myeloid leukemia, 3% (n=1) with acute lymphoblastic leukaemia and 3% (n=1) with polycythaemia rubra vera.

We grouped the patients into those with a chronic relapsing/remitting type of skin eruption (n=15) and the second group consisting of patients with a single episode of classical SS (n=16).

Patients presenting with this chronic relapsing remitting form of SS (n=15) were not generally known to have MDS at the time of their initial skin eruption. The median time from diagnosis of SS to diagnosis of MDS was 17 months (1.4 years) (range 0-157 months). The WHO subtypes of MDS were RCMD (N= 13), RAEB-1 (N=1) and MDS/MPN-U (N=2).All except two patients (trisomy 8 and del 11q) had normal bone marrow karyotype. The IPSS risk groups were; Low (N=10), Int-1 (n=5).Transfusion dependency was subsequently seen in 6 of 15 patients. Progression to high-risk occurred in two patients (RAEB 1), whilst none had leukaemic transformation.

The clinical features of the chronic form were identical to those described by Sweet ; raised, tender plaques which were red and urticated. Some of the lesions had mamillated (“nipple like”) elevation on the surface of these plaques. They were found to be scattered on the torso and limbs, neck and face. Larger more nodular plum coloured lesions may also be found.

All 15 patients had constitutional symptoms including fever and sweats at the time of skin eruptions. Arthralgia was seen in a majority of patients (n=12).Additionally, other associated autoimmune conditions or dermatological conditions seen included seronegative rheumatoid arthritis (n=1), relapsing polychondritis (n=1), pyoderma gangrenosum (n=1) and Behcets disease (n-1).

Compared to MDS without SS (n=711), patients with SS and MDS were on an average 8 years younger, low/int-1 risk, less likely to receive MDS therapy and had lower propensity to leukaemic transformation (all p<0.001).

The frequency of γδ Tcells (3.6% ± 1.66% v 2.2% ± 0.6%, p=0.51), effector CD4+ T cells (12.6% ± 5.7 v 4.5% ± 1.8%, p=0.2) and resting Tregs (11.1 ±2.5 v 5.5% ± 0.6%, p=0.4) in patients were higher compared to healthy aged match donors.

The chronic relapsing remitting of SS was recalcitrant to treatment. Most patients had to be maintained on a higher doses of prednisolone (>15-20mg) to prevent recurrent episodes. The response to immunosuppressive therapy (IST) was variable with median of 4 ISTs (range 0-12) per patient. The treatment associated with complete resolution of the skin eruptions with no relapses were 5-azacitidine in four patients, infliximab in one patient and one with methotrexate but other agents were disappointing. Corticosteroids were effective in all patients; however doses of prednisolone below 15mgs resulted invariably in relapse of SS.

The cause in 16 patients could be attributed either to administration of GCSF or after chemotherapy. The eruption was brief and disappeared spontaneously or following withdrawal of GCSF

We describe a chronic debilitating episodic clinically distinctive skin eruption with features of Sweet’s syndrome but not always definitive histopathology often associated with immunological abnormalities affecting other systems (especially rheumatological) universally related to underlying occult ‘lower risk’ MDS.

No relevant conflicts of interest to declare.