I agree with you interpretation, but the possibility of secondary leukemia has to be excluded. How long ago was the last BMT? Current HB, WBC, Plt? I assume that the CD34+ cells were negative for B-cell markers and CD10?  
I would recommend in this case - if possible, to stop G-CSF therapy and perform control biopsy after approx. 2 -3 weeks to see if the blast infiltration would diminish. If possible send also bone marrow for cytogenetics at the same time.

I doubt that this is GCSF-effect. True, single and scattered CD34+ blasts can be seen after GCSF, particularly if the BM examination has been performed shortly after GCSF therapy,but they do not form aggregates as seen in this case. CD34+ cells forming aggregates are strongly suggestive of a leukemic process, supporting Anjas suggestion of a therapy related MDS/AL. Other changes in favour of this diagnosis are the fibrosis, and the marked megakaryocytic dysplasia and one typical mikcromegakaryocyte (see arrow on the drawing). A certain amount of megakaryocytic hyperlasia and dysplasia can be seen after bone marrow transplantation, but typical micromegakaryocytes are always suggestive of MDS. Can you get more clinical information? When were the transplants performed? Peripheral blood counts, findings on the aspirate?  

Dear dr.Anna and dr.Nina. Thank you for urgent replies. Some IH stains are in progress. I will apend it and clinical details in a short time. The main thing is, that despite primary LCH diagnosis (skin), all changes in the spleen and liver were not so typical for residual LCH and generalized macrophageal reactions Rosai Dorfman like were predominant. So nobody knows what really happened before transplantion...

Hello, I am currently the attending phisicion of the discussed patient. There are some more clinical facts about the patient. The second reduced intensity aloPBSCT was performed in March 01, 2006. The patient is no longer on G-CSF now. In his peripheral blood he currently has WBC about 3,0 - 4,0 10E9/l, MNC <500. He is PLT and RBC transfusion dependent.  
 

still an extremely difficult case, indeed! i agree with anna and nina, (sec) mds/aml has to be strictly excluded (and if diagnosed at all, rule out donor-type stem cell malignancy!?); important to analyse chimerism!  
still, reactive changes are possible considering previous bm infiltration bei lch and all the therapies applied....

Dear Goda, thank you for participating in the discussion about this very difficult case. As in many hematopathologic cases no evaluation can be made without the clinicans' input.I am very much concerned about the possibitity of a secondary MDS/AML in this young patient. Could you please provide us a summary of the clinical history, including laboratory findings. what where the findings on the aspirate? Were ringsideroblasts present?  
Is there any evidence of a hemopoietic disorder in addition to the LCH before the first BMT ?

I added a few references about LCH and associated hematologic malignancies

Thank you very much for your comments. The case of this patient is a little bit confusing to us. I give you a short history.  
Skin lesions were stated from 10 months of age, treated as atopic dermatitis. 2003 January at 2 years of age skin biopsy revealed LCH. Multiple skull bone defects, diffuse skin, lymph node, gum, spleen and liver involvement, mild anemia, elevated ESR. Treated according DAL-HX 93 protocol group C.  
Good response to treatment. Continuation therapy till 2004 November, because of risk organs involvement.  
2004 December diffuse bone pain, high fever. Peripheral blood analysis: panhypoplasia. New skull bone defects, enlarged liver and spleen, anemia, leucopenia, thrombocytopenia, elevated ESR, bone marrow smears: hypoplasia. Treated with cyclophosphamid, etoposid, methotraxate, doxorubicine, corticosteroids. 2005 April: skull bone defects declined. Fever remained with corticosteroids withdrawal. Liver and spleen remained enlarged. Bone marrow biopsy showed infiltration with CD1+cells and activated CD68+ cells, hemophagocytosis. The same infiltration in spleen (needle biopsy) and liver. Treatment with ATG, cyclosporine and corticosteroids. Bone marrow smears: panhypoplasia, no pathological infiltration. Fever remained. He is reliant on blood components transfusion. Very rapid enlargement of spleen (170 mm) and liver. 2005-09-21 he was splenectomised. Histopatologically CD1+ cells and signes of hematophagocytosis.  
2005-11-17 1-allele mismatch unrelated donor, major blood group mismatch, reduced intensity conditioning regimen FluCamp/Mel. No engraftment. Autologous recovery of granulocytes on day +54. He is reliant on blood components transfusion. The boy still had signes of active disease. In general his condition is not bad. According to LCH international society recomendations (was consulted bey prof. Gadner) the second graft was given on 01.03.2006 from the same donor with the same conditioning. No engraftment again, chimerism 98% on day +33, but ANC <500. At the moment chimerism 0, transfusion dependent. In peripheral blood there are plenty of normoblasts, blasts 5-7 percent.  
Karyotype was done twice from bone marrow, at the begining and the after 1 transplant - normal (46XY), unfortunately, FISH is unavailable in our country.  

P.S. ringed sideroblasts were never have been found. Blast count in bone marrow was never >5 percent before transplant.

Let me add, that recently we set up FISH for HER2 in our centre. So in the next future, maybe, the panel will be expanded to hematopathology area by request of clinicians. The PCR is still more actual and will be introduced at first step in the our hospital campus area.

I have an idea of false positivity of CD34, cause there is a prominent variation of size of these pseudoblasts... I have requested the lab, maybe they get a new CD34 Ab (not the QBend clone) or something is wrong.

CD34 positivity look quite specific to me. How did the smears look like? In this case another biopsy after relatively short time may be of value even if the G-CSF have been stopped a while ago - if it is a secondary leukemia it will probably show a progression.

Here is some update information about the course of the disease. As we had some increased count of blasts in peripheral blood for some period of time (we could not exclude G-CSF influence), now the patient is withdrowned from CSF and blast count in periphery is rapidly increasing for the last three days. Today we had 40 percent. It seems to be non lymphoblastic phenotype, immunophenotypic results will be available tomorrow.  
 
I have a very troublesome question, as far as histological findings were not so typical to LCH could they be compatible in some features with HLH?  
 
I am very much thankfull to all of you for your very interesting comments.

I agree with Anja, CD34 staining is specific, on image 3 blastic morphology of CD34+ cells is clearly recognized. In the area shown on image 3 the amout of blasts reaches at least 20%, Some of the MPO+ cells show blastic morphology as well. Based on this finding together with the recent information we received from Goda,that blasts in the peripheral blood are rapidly increasing I am afraid that we have to call this condition acute nonlymphoblastic leukemia.  
Could you please inform us about the results of flow cytometry?  
Goda, you wrote that the bone marrow smears showed hypoplasia of hemopoiesis, no malignant cells. This discrepancy between the findings on the aspirate and the biopsy, is typical in cases, like this one, where marked marrow fibrosis is associated with the infiltrate.

Let me inform you that I have repeated CD34 (laboratory mistake)- reaction is more restricted. I will provide a photo in a moment.

The new CD34 are placed in. CD34+ population (medium cells with irregular vesicular nuclei and nucleoli) subjectivelly about 5%? The massive iron deposition in BM.  
So: Architectural distortion with CD34+ elevation, megakaryocytic dysplasia, iron deposition, depression of the erythropoesis, reticulin II fibrosis and macrophageal reaction. Histological changes are consistant with posttherapy MDS.

If the blasts are 40% the criteria for acute leukemia are fullfilled. The lineage is still unclear. It seems that the blasts are now CD34-, CD117-, CD68- and MPO-. How about lymphatic markers and CD56? The only way to classify this leukemia is immunophenotyping.

Thanx. Rebiopsy was done once more because blastemia is gowing up. I will append add data in a short time.

anja is correct, of course; 40% blasts in pb is per definitionem acute leucemia (if blast count is correct, there is no differential dx. subclassification should be dobe by fcm of pb blasts; in addition, according to who it is a therapy-related al.  
indeed, this is a sad clinical course of this young child.

As Anja and Stephan stated there is no question that we are dealing with an AL, probabaly therapy related. Flow cytometry in needed for classification of this AL. It would be also of interest to repeat conventional karyotyping which apparently was normal before the 2.transplant. Complex cytogenetic abnormalities are usually demonstrated in therapy related AL. Unfortunately all our present considerations seem to be of purely scientific interest, which probably will not affect the bleak prognosis for this child.  
The repeated CD34 incubation clearly shows that CD34- cells have an unequivocal blastic morphology. I also cannot tell which of the two CD34 incubations is the reliable one, we still cannot exclude that the first incubation does reflect the real situation. Increase of iron deposits are a typical finding in posttransplant bone marrow biopsies.  
Dear Goda, pleaase excuse my ignorance, what is HLH? In general I would like to suggest be careful with abbreviations, not everybody might be familar with.

Thank you. The TECHINICAL mistake was: CD34 stain was programmed like CD43, so a lot of "CD34+" cells in the first IH line really represent myeloid cells CD43+.  
I'm appologize for this. The mistake was catched in case with CD34+ GIST with staining of lymphoid tissue. The primary CD34 photo is DELETED.  
 
I hope that blast quantitation will be performed by flow or in the next BM biopsy the quality would be better.  
 
Maybe it's true, that in this case only single blasts are CD34+.

I was reading this interesting case and I would like to give a short comment: the transformation from a HCL in an acute leukaemia is a rare event; it is surprisingly that in this patient it appeared after the second transplantation (he was conditioned twice with reduced intense protocol). The differential diagnosis of a therapy related leukaemia should also be hold since this patient received in 2003 and 2004 immunosuppressive therapy; he had at least more than two years of follow-up time in between. The Donor cell leukaemia is a rare complication after allogeneic hematopoietic stem cell transplantation. To evaluate the origin of Blast I would suggest performing a CD34+ Chimerism from bone marrow and check the health status of the donor as well.

The morphology of the marrow and the phenotype in my opinion strongly suggest a secondary neoplastic process, rather than reactive changes or LCH infiltration. I would strongly consider MDS/AML due to the dyspoiesis, CD34+ groups of cells and fibrosis. THe time course is clearly unusual, considering the recent allogeneic transplant.  
Considering evolution of the disease, one could also think about a stem cell disorder with initial LCH phenotype and presentation, progressing to a more immature disorder. Although this is very speculative, there is one report about a common clonal origin of LCH and ALL, which might add some thoughts (Feldman AL et al, Lancet Oncol). However, it does not seem to represent the usual association of AL and LCH, based on clinical history.