17 yrs old male with multiple THREPINE BIOPSIES (2003,2004,2005,2006) and liver/spleen/lymph node biopsies (2004).  
HISTOLOGY: BM B06-1140 (2006) erythroid depression and reactive central limfoid follicle in BM (on the slides).  
BM B04-27041 (2004): the same changes. The erythroid precursors are unvisible in HE. The reticulin fibrosis and Fe deposits are absent. Perivascular interstitial and cewntral lymphoid follicles/infiltrates are obvious (CD3+>CD20+). The slightly elevated amount of TdT+ > CD34+ precursors. The single CD57+>CD56/GranB+ small lymphocytes. CD8+>>CD4+.  
BM B05-22799 (2005): The same changes. Focal I reticulin fibrosis appears. The prominent central follicles, interstitial and peritrabecular? lymphoid infiltrates. The slight eosinophilia is present.  
SUMMARY of BM changes: Erythroid depression, lymhoid infiltrates, most probably reactive, slight architectural distortion and eosinophilia. Focal I reticulin fibrosis.  
LYMH NODE B04-3220: Reactive lymphadenopathy.  
LIVER B04-3220: There are mononuclear portal and intrasinusoidal infiltration with some nuclear irregularities in parenchyma. Intrasinusoidal atypical lympho's: LCA+, CD20-, CD3-, CD4/CD8-, GranzymB-: lymphoid (NK?)proliferation, uncertain behaviour?.  
SPLEEN B04-3220: Follicular hyperplasia and some small lymphocytic CD3+ > GranzymB+, CD8+ slightly > CD4+, CD56/CD57(-) intrasinusoidal and interstitial infiltration: tends to be REACTIVE CHANGES.  
CLINICAL INFO (dr.Laimonas Griškevicius). The patient was first evaluated for anemia by a child hematologist at the age of 15 in October 2003. HB count was 62 g/l (normocytic, low reticulocytes), PLT, WBC normal. On presentation spleen was 3 cm below costal margin. BM biopsy revealed absent erythropoiesis. PRCA was diagnosed, the patient was treated with prednisolone and CsA with a good response: rising Hb and reticulocytes. During the course of disease, liver, spleen and multiple bone marrow biopsies were obtained (see pathology report). On discontinuation of prednisolone the patient repeatedly developed transfusion dependent anemia and was treated with 2 courses of ATG and CsA. The Hb count was not sustainable and chronic prednisolone therapy had to be continued. The patient was refered to adult hematology at the age of 18 in June 2006. At presentation, Cushingoid features due to chronic prednisolone treatment were noted. There was no peripheral lympadenopathy, spleen was palpable at 5 cm BCM, liver was not enlarged. Chest X ray normal. Abdominal sonoscopy: spleen 157 mm. WBC: 5.26 x109/l, normal differential, normal small lymphocytes, Hb 67 g/l, retic. 0.1%, MCV 84, PLT 378 x 109/l. Peripheral FC (obtained on chronic prednislone treatment) (CD16 positive NK cells 41% - 390/mkl): CD45+, CD2+, CD7+, CD11c+, CD16+, HLA-DR+, CD38+, CD3-, CD4-, CD5-, CD8-, CD25-, CD56-, CD57-, TCRab-, TCRgd-, cCD3-. The same phenotype was observed in June 2005, NK count was 1835/mkl in bone marrow. Coombs direct and indirect negative. Since NK cells have TCR genes in nonarranged germline postition, TCR clonality studies were considered noninformative and not performed. LDH 221 U/l (normal < 190 U/l) on CsA treatment. Beta2 microglobulin 4,17 mg /l (normal < 3 mg/l), creatinine 110 mkmo/l (slightly increased due to CsA).  
Given the splenomegaly, continuosly increased NK numbers with aberrant immunophenotype (CD56-), absent erythropoiesis and good response to prednisolone, the proposed diagnosis is:  
CHRONIC NK CELL LEUKEMIA. PURE RED CELL APLASIA.  
QUESTION: DIFFERENTIAL WITH REACTIVE NK LYMPHOCYTOSIS.  

ugnius 2006-08-16 11:27	BM histological data are not so impressive. I will append some photos right now.
ugnius 2006-08-16 14:16	Histollogically the most suspitious are intrasinusoidal lymphoid infiltrates in the liver. But I cannot find any argues from histology side alone for definite diagnosis. And nowadays I cannot prove NK phenotype in histology except mentioned CD56 and oth T markers.
franco 2006-08-16 21:52	Your diagnosis of CHRONIC NK CELL LEUKEMIA associated with PURE RED CELL APLASIA is the best one. You can exclude reactive lymphocytosis. Diffuse intrasinusoidal neoplastic infiltration can be seen in BM, spleen and liver. PRCA can regress after proper chemotherapy and regression of neoplastic infiltration.   Differential diagnosis with LGL can be almost impossible due to overlapping cases and different immunophenotypic features.
sirje 2006-08-17 08:38	I agree basically with dr. Franko. Here the differential dgn should be made between NK LGL leukemia, Chronic NK lymphocytosis and reactive NK lymphocytosis. All these conditions may present with PRCA. As the WBC is normal with no lympocytosis (only small number of granular ly-tes), the reactive (viral induced?)condition should be excluded.
ugnius 2006-08-17 09:25	Thank you for comments. What we can do additionally with IH stains on the histological material and what is your practice with immunoverification of minimal T/NK infiltrates esp. in cases of T/NK leukemias in BM or everywhere?
cancerr 2006-08-17 09:41	Curiously, the patient had a EBV positivity by PCR 57000 copies/ml (May 2005) and received 1 course of Rituximab without any response in anemia. He was borderline positive for EBV virus (200 copies/ml) 1 week ago. Otherwise he is EBV IgG +, CMV IgG -.   Given the long standing NK expansion with aberrant immunophenotype (CD56-), splenomegaly, I would favor chronic NK-cell leukemia rather than lymphocytosis reactive or indolent.   I wonder whether there are any possibilities to prove (or disprove) the monoclonality of NK cells?
anpo 2006-08-17 11:15	One can get indications for clonal proliferation in NK cells by studying KIR (CD158 family) - see attached file. Also with extended antibody panel NK-cell leukemias have pathological profile as compared to reactive NK-cell increase. attachment: NK.pdf
hurwitz 2006-08-18 15:08	I share Vito's opinion that the most probable diagnosis is NK-GLL associated with PRCA. Thanks for the pb images which shows a few characteristic LGL.The most convincing finding are the 41% of CD16+ cells on FC. Regarding IHC on the BMB I would suggest to perform the following stains: CD 45; CD3,CD8; CD4; CD20, CD 56, CD57, TIA and Granzyme B, as well as EBV LMP and if possible ISH for EBER1. The beautifully demonstrated intrasinusoidal infiltration in the liver is an argument against a reactive process. Referring to Dr. Griskevicius comment on the demonstration of EBV sequences, this might be a coincidental finding since in contrast to aggressive NK-cell leukemia the indolent forms are not EBV associated.
ugnius 2006-08-18 15:49	Dear Prof.Nina- CD56, CD57, GranzymB and LCA were applied.   Obviously intrasinusoid lymhoid cells in liver are LCA+ CD3 and Co(-). In BM scant infiltrates (with some amount of CD57 cells) are doubtfull as malignant. Granzym B+ cells are more prominent in the spleen, but it seems to be noneoplastic. EBV LMP I will try, EBER ISH and TIA1 are out of my panel.
cancerr 2006-08-19 13:11	The chronic active EBV infection (CAEBV) may be another possibility with clonal expansion of NK cells. Another EBV PCR sample (the 3rd time tested during the desease course) turned out positive with the viral load of 3700 cop/ml. It seems that sporadic "European" CAEBV may have a variable clinical course as recently reviewd by Kimura (Rev. Med. Virol. 2006; 16: 251–261.)   It definitely would be nice to demonstrate the restriction in KIR (CD158 family) receptor expression as suggested by dr. Porwit-MacDonald and to see if the intrasinusoidal infiltration in the liver harbours EBV RNRs and LMP1 (thanks Dr. Hurwitz).   Unfortuantely, we do not have any monoclonals against CD158 at present. Could anyone suggest a lab where ISH for EBER1 can be performed?   The major therapeutic consideration for this young man is an allogeneic SCT.
hurwitz 2006-08-21 17:11	Dear Dr.Griškevicius, I will try to find out about KIR; EBER1 and LMP staings for this case and let you know where to ent the blocks. I will also ask a hematologist from the department of hematology, University of Basel to comment on treatment options.
cancerr 2006-08-21 18:43	Thank you very much, dr. Hurwitz
dirnhofer 2006-08-22 22:05	just got an e-mail from nina hurwitz and logged in. of course, i would be willing to do the ebv-analysis (ihc and ish). just send me representative blocks. kind regards
rovoa 2006-08-23 13:48	Within all NK entities the T-LGL Leukemia has been frequently associated with PRCA. In this case, the immunophenotyp ruled out this differential diagnosis. From all NK Diseases the patient’s immune profile is closed to an aggressive NK leukemia and is also very suggestive the data of positive EBV infection during the follow-up. However the typical clinical presentation of an aggressive NK leukemia is exactly as her title refers “aggressive” with persistent high fever, systemic symptoms, hepato-splenomegaly with deranged liver function and coagulation profile. Practically all patients with aggressive NK cell leukemia die from the disease within a few weeks or months of presentation. According to the described data this patient showed an indolent clinical follow-up. In summary making an integrative approach with the diagnostic data together with the clinical evolution we face a non typical presentation of a NK disorder not further classifiable. Considering the indolent course I would not suggest any treatment just watch and wait.
ugnius 2006-08-23 15:05	EBV LMP1 in BM, spleen and liver is negative.
ffrenchma 2006-08-30 16:59	This case of a young girl with an aplastic anemia associated with a NK proliferation and a chronic evolution does not appear to be far from some cases included in the series of this paper : "Rabbani GR, Phyliky RL, Tefferi A. A long-term study of patients with chronic natural killer cell lymphocytosis. Br J Haematol. 1999 Sep;106(4):960-6."
kurt 2006-08-31 09:11	here the full text of to the mentioned article {PMID: 10519998} attachment: j.1365-2141.1999.01624.x.pdf
ugnius 2006-08-31 11:33	[comment sent by email] This case is sent to dr.S.Dirnhofer/ dr.N.Hurwitz for review.
hurwitz 2006-10-17 19:00	I discussed the case with Stephan Dirnhofer. Stains for EBER1 were negative, so we could not find any evidence for EBV association.   Comparing the findings on the BMB we could observe an increase of the lymphoid infiltrate and the eosinophilia from the first biopsy in 2004 to the biopsies in 2005 and 2006. The amount of infiltrate beeing similar in 2005 and 2006. This is just an observation which might be due to sampling and cannot be used as argument for or against reactive versus neoplastic. Lets wait for Anjas results of her CD158 stain.
ugnius 2006-10-18 13:00	Thank you all participating in futher evaluation of the case. From my side the findings were not interpreted as positive fo any lymphoproliferation.
	» Add comment (Login)