If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Université Paris Diderot, Sorbonne Paris Cité, UMR-S 1165, Paris, FranceINSERM, U1165-Paris, Paris, FranceLaboratoire de pathologie, Hôpital Saint-Louis, AP-HP, Paris, France
Université Paris Diderot, Sorbonne Paris Cité, UMR-S 1165, Paris, FranceINSERM, U1165-Paris, Paris, FranceLaboratoire de pathologie, Hôpital Saint-Louis, AP-HP, Paris, France
Université Paris Diderot, Sorbonne Paris Cité, UMR-S 1165, Paris, FranceINSERM, U1165-Paris, Paris, FranceLaboratoire de pathologie, Hôpital Saint-Louis, AP-HP, Paris, France
Université Paris Diderot, Sorbonne Paris Cité, UMR-S 1165, Paris, FranceINSERM, U1165-Paris, Paris, FranceLaboratoire de pathologie, Hôpital Saint-Louis, AP-HP, Paris, France
). The prognostic value of skin lesions infiltrated only by non-blastic MDS tumor cells has not yet been characterized.
We studied 24 MDS patients with non-blastic skin infiltrate and compared them with 20 leukemia cutis patients. This study adheres to the declaration of Helsinki principles, and patient consent for experimentals was not required because the French laws consider human tissue left over from surgery as discarded material. Detailed patient data are given in Supplementary Table S1 online.
Between 1995 and 2012, 800 patients were diagnosed with MDS in Hôpital-Saint-Louis, Paris. One hundred and fifty patients underwent skin biopsy, and we identified 24 skin involvements by non-blastic tumor cells, defined as medium-sized immature myeloid cells (Figure 1) with (i) abundant eosinophilic cytoplasm and (ii) twisted nuclei or pseudo-Pelger-Huet anomaly, a specific myelodysplasia marker on blood smears (
). The tumor cells had a combined myeloid and monocytic phenotype, expressing both myeloperoxydase (100% of cases) and CD163 (100%) or CD68 (96%). They did not express CD34, CD117, or CD56. The proliferative index with Mib-1 was low (<10% positive cells) in 56% of cases, or intermediate (10 to 66% positive cells) in 44% of cases, but never high (≥66% positive cells). Mature neutrophils and normal CD3+ lymphocytes were numerous (46% and 100% of cases, respectively) and edema was frequent (67%).
Figure 1Myelodysplastic syndrome patients with non-blastic tumor cell infiltrate (“myelodysplasia cutis”) and leukemia cutis patients had different clinical, histological, and immuno-phenotypical features. In myelodysplasia cutis patients (a) erythematous plaques, frequently annular, were predominant, corresponding to (b) a skin infiltrate composed of myeloid cells with the pseudo-Pelger-Huet anomaly (nucleus with “pince-nez” appearance at high magnification, arrows), mixed with lymphocytes. (c) The myeloid cells expressed myelomonocytic markers CD68 and myeloperoxydase but not blastic markers CD34 and CD56. In contrast, in leukemia cutis patients, (a) nodules were predominant, corresponding to (b) a skin infiltrate composed of monomorphous blastic cells (with large round nuclei at high magnification). (c) These cells expressed both myelomonocytic markers CD68 and myeloperoxydase and blastic markers CD34 and CD56. Scale bars=b 100 μm, high magnifications 25 μm; c 50 μm.
Nineteen of the 24 MDS cases were tested for bone marrow cytogenetics. Abnormalities were found in 6 of them (32%). To determine whether the same genetic abnormalities were also found in the skin tumor cells, fluorescent in situ hybridization (FISH) analyses were performed on 4 μm-thick paraffin-embedded skin sections using relevant probes and the Histology-FISH-Accessory-Kit (Dako, Denmark). Scoring of the hybridization signals, performed on 200 consecutive morphologically intact nuclei with a 10% normal cutoff value (
Detection of three common translocation breakpoints in non-hodgkin’s lymphomas by fluorescence in situ hybridization on routine paraffin-embedded tissue section.
), showed common cytogenetic abnormalities in the bone marrow and the non-blastic myeloid cells in 4/6 patients (Supplementary Figure S2 online). This demonstrated that the immature tumor cells in the skin were clonally related to the myeloid malignancy. A clonal relationship of this sort has so far only been demonstrated in neutrophilic dermatoses associated with AML (
In all the 24 patients, the median overall survival (OS) from skin diagnosis was 62 months, longer compared with the 52 months expected OS calculated with the International Prognostic Scoring System (IPSS;
Diagnosis and classification of myelodysplastic syndrome: International Working Group on Morphology of myelodysplastic syndrome (IWGM-MDS) consensus proposals for the definition and enumeration of myeloblasts and ring sideroblasts.
In 16 cases, the patients received hydroxychloroquine, dapsone, colchicine, or thalidomide, with no significant benefit. High-dose oral prednisone provided complete response in 18 patients, most often with steroid dependence.
We thus characterized a skin condition in the course of MDS with non-blastic tumor cell skin infiltrate and a lower risk of progression to AML than for leukemia cutis.
We further compared these 24 patients with “myelodysplasia cutis” with 20 true leukemia cutis, in the course of AML (7 with previous MDS). Histologically, leukemia cutis was defined by a skin infiltration by blast cells with (i) medium to large monomorphous cytoplasm, (ii) large round nuclei, and (iii) expression of the myeloid or monocytic markers, myeloperoxydase (80% of cases), CD163 (85%), and CD68 (95%).
Discriminant features, using Fisher’s t-test (Figure 2) were, for histopathology, the positivity of CD34, CD56, or CD117 for leukemia cutis (P<0.05) and the presence of CD3+ lymphocytes (P<0.001), edema (P<0.01), and a lower Mib-1 proliferative index (P<0.05) for myelodysplasia cutis.
Figure 2Statistical analyses comparing 24 myelodysplasia cutis patients and 20 leukemia cutis patients. (a) Using Fisher’s t-test with R-software (R 3.1.2 The R Foundation for Statistical Computing R, Vienna, Austria) and after correction for multiple comparison, eight clinical and histopathological discriminant features were identified. We performed a quantitative study of cells labeled with CD163, CD68, myeloperoxydase, CD117, CD56, CD34, and CD3 in the skin sections. We analyzed a minimum of three different fields at an original magnification 250 × and counted the percentage of stained cells in one hundred tumor cells for CD163, CD68, myeloperoxydase, CD117, CD56, CD34, and lymphocytes for CD3. The cutoff value was 80% positive cells for CD163, CD68, and myeloperoxydase, 50% for CD117, CD56, and CD34, and 20% for CD3. (b) Plotting Kaplan–Meier curves for the overall survival (OS) in myelodysplasia cutis patients versus leukemia cutis patients, and comparing them with the log-rank test, showed that myelodysplasia cutis patients had a significantly longer OS compared with patients with leukemia cutis (62 vs. 5 months, P<0.001).
Clinical discriminant features were the presence of nodules for leukemia cutis (P<0.01) and the presence of erythematous plaques (P<0.001), annular pattern (P<0.05), fever, or arthralgia (P<0.01) for myelodysplasia cutis.
Regarding evolution, persistent nodules were the hallmark of leukemia cutis, whereas flares and relapses characterized evolution in myelodysplasia cutis. In most myelodysplasia cutis patients (14/24), skin lesions occurred before the bone marrow MDS diagnosis, with a mean time-lapse of 41 months (range 2–108). In the other 10/24 patients, skin lesions occurred during MDS evolution with a shorter time-lapse of 15 months (range 6-19). Skin lesions in all 20 leukemia cutis patients occurred after the bone marrow diagnosis of the hemopathy.
Regarding survival, OS from the skin diagnosis was considerably longer in myelodysplasia cutis patients than in leukemia cutis patients (62 vs. 5 months, P<0.001). In the seven leukemia cutis patients with previous MDS, the skin lesions occurred when MDS evolved to AML, with a 2-month OS after the occurrence of the skin lesions, similar to the time-lapse reported by
Using the Cox multivariate hazard model, the blastic versus non-blastic nature of the skin infiltrate in MDS patients was associated with survival, independently from IPSS. In Cox univariate analysis, the blastic versus non-blastic nature of the skin infiltrate was significantly associated with survival (P<0.005), whereas IPSS was not (P=0.17).
Altogether, we identified clinical and pathological features discriminating myelodysplasia cutis from leukemia cutis, with a significantly longer survival for myelodysplasia cutis.
Retrospective analysis of the relevant literature enabled us to find reports of “histiocytoid Sweet’s syndrome” (
): patients had plaques with fever and arthralgia and a myeloid histiocytoid non-blastic skin infiltrate with edema, mixed neutrophils, and CD3+ lymphocytes––i.e., five of the eight discriminant features we identified (the three others being not specified). In addition, in two MDS cases without AML, the same cytogenetic abnormality was found in the skin and the bone marrow, as in four of our myelodysplasia cutis. Therefore, we think that cases reported as “histiocytoid Sweet’s syndrome” in the course of MDS may be better classified as myelodysplasia cutis.
One of the original results of our study is to highlight the fact that skin lesions in myelodysplasia cutis can precede by months or years MDS diagnosis in the bone marrow. This underlines the value of a long follow-up, particularly in elderly patients with a normal initial bone marrow examination. More studies will be necessary to identify peculiar markers involved in non-blastic MDS cells skin homing, as it has been described for blast cells (
The discriminant features we identified between myelodysplasia cutis and leukemia cutis have translational value, given the far better prognosis of myelodysplasia cutis.
ACKNOWLEDGMENTS
We thank Dr C. Juillard, Dr F. Cordoliani, and Dr M. Rybojad who provided patient data and Angela Swaine Verdier who edited the English language.
Detection of three common translocation breakpoints in non-hodgkin’s lymphomas by fluorescence in situ hybridization on routine paraffin-embedded tissue section.
Diagnosis and classification of myelodysplastic syndrome: International Working Group on Morphology of myelodysplastic syndrome (IWGM-MDS) consensus proposals for the definition and enumeration of myeloblasts and ring sideroblasts.
38998-3&id=gr1.jpg){kind=link}
38998-3&id=gr2.jpg){kind=link}