Flow Cytometry of Blood and Bone Marrow Cells From ...€¦ · AIRY CELL LEUKEMIA' (HCL) is a rare B-cell dis- order characterized by cytopenia, splenomegaly and infections. Hairy

Flow Cytometry of Blood and Bone Marrow Cells From Patients With Hairy Cell Leukemia: Phenotype of Hairy Cells and Lymphocyte Subsets After

Treatment With 2-Chlorodeoxyadenosine

By Gunnar Juliusson, Rodica Lenkei, and Jan Liliemark

By flow cytometry and an extensive set of markers, we characterized leukemic cells from the blood and bone marrow of 68 symptomatic patients with hairy cell leukemia (HCL). Hairy cells identified in the large cell gate always expressed CDl9, CD20, HLA-DR. CD#RA, and B-ly 7. Other markers were occasionally expressed, such as CD38, CD45R0, CD23, CD15, CD4 CD5, and CD10 (expressed on more than 20% of the hairy cells in 44%. 25%. 2196, l8%, 12%. IO%, and 5% of evaluated cases, respectively). During treatment with 2- chlorodeoxyadenosine (CdA), the median lymphocyte counts decreased from 2,0001pL to 3OO/pL. Flow cytometry was repeated at the nadir (n = 24) of lymphocyte counts, at 3 months (n = 46). at 6 months (n = 50). at 1 year (n = 39). and at 2 years (n = 12) after treatment. The initial decrease of CD8' and CD20+ cells was greater than that of CD4+ and natural killer iNK) cells, leading to an increasing CD41CD8

AIRY CELL LEUKEMIA' (HCL) is a rare B-cell dis- order characterized by cytopenia, splenomegaly and

infections. Hairy cells express Igs and the B-cell markers CD19 and CD20.' A characteristic feature is the positive reaction with the interleukin-2 (IL-2) receptor, CD25.3 A variant form with a prolymphocyte-like cell morphology and poor response to interferon treatment is de~cribed.~ Expres- sion of the pan T-cell marker CD5, characteristically found in B-cell chronic lymphocytic leukemia, has also shown to associate with poor response to interferon.'

Deoxycoformycin (dCF)6 is a tight-binding inhibitor of adenosine deaminase (ADA), leading to a spectrum of therapeutic activity similar to that of the purine analogues fludarabine' and 2-chlorodeoxyadenosine (CdA),8,9 which are resistant to demination by ADA" because of the halogen substitution at the 2-position. It was shown in 1984 that dCF may produce complete remissions in HCL.",'' However, dCF induces a prolonged immunos~ppression, '~~~~ and patients with other diseases than HCL have developed late opportunistic infections." Treatment of HCL with CdA may

H

From the Division of Clinical Hematology and Oncology, Depart- ment of Medicine, Huddinge Hospital, and SBL Diagnostics/Calab Medical Laboratory, and Department of Clinical Pharmacology, Karolinska Hospital, Karolinska Institute, Stockholm; and the De- parment of Hematology, University Hospital, Linkdping, Sweden.

Submitted September 7, 1993; accepted February 18, 1994. Supported by the Swedish Cancer Society (project no. 2409-B91-

OIXAB), The Swedish Society for Medical Science, The Karolinska Institute's Research Funds, and The Medical Research Council.

Address reprint requests to Gunnar Juliusson, MD, PhD, Depart- ment of Hematology, University Hospital, S-58185 Linkdping, Swe- den.

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. section 1734 solely to indicate this fact.

0 1994 by The American Society of Hematology. 0006-497I/94/8312-0027$3.00/0

ratio. Median nadir values of CD4+, CD8+, CD20'. and NK cells were 128/l,cL, 78/pL, lO/l,cL, and 13/pL, respectively. The subsequent recovery was quicker for CD8+ and NK cells, leading to a normalization within 3 months, whereas CD20+ and CD4+ cells required 1 or 2 years to enter the normal range. The CD4ICD8 ratio thus decreased after the nadir and remained less than 1 . CD#RA+ CD4 cells and CD45RA+/ CD45RO' double-positive cells were less affected by CdA. Activated T cells, ie, HLA-DR+ cells, rarely decreased below the normal range and often recovered with an overshoot. CD10+ cells increased in the bone marrow posttreatment as an indication of normal B-cell regeneration in 16 of 36 (44%) patients. The quick regeneration of certain lymphoid subsets might explain the lack of late infections in CdA-treated HCL patients. 0 1994 by The American Society of Hematology.

result in an even higher complete remission rate8'9.'6"x from a single week of therapy, either as a continuous intravenous infusion' or as daily subcutaneous injections.'' Among patients with pretreatment anemia, treatment is frequently com- plicated by neutropenic fever,'8.20.Z1 either during the treatment course, which might be caused by rapid tumor cell lysis, or during the following weeks, which may be caused by opportunistic

To assess the phenotypic variability of hairy cells, and to evaluate the long-term effects of CdA on immune cells we used flow cytometry analyses with an extensive set of markers on blood and bone marrow cells in patients with active disease before and after CdA treatment.

PATIENTS AND METHODS

A total of 75 patients, 63 men and 12 women, with symptomatic HCL were studied, some of whom are previously reported.'s'20 Diag- nosis was made through morphology with cytochemistry on repeated bone marrow aspirations and biopsy specimens. Thirty-four patients were previously untreated, and 41 had received prior treatment, 1 of them (patient JJ in Juliusson and Liliemark") including deoxycoformycin. Their mean age was 53 years (range, 29 to 75), and sampling of blood and bone marrow was performed with a median of 10 months from diagnosis. All patients were subsequently treated with CdA for 7 days, either as a continuous infusion of 0.085 mg/ ,,/dla or as daily subcutaneous injections of 3.4 rnglm'.'' These treatment schedules give the same concentrations of CdA nucleotides in leukemic cells22~2' (and Liliemark and Juliusson, manuscript in preparation) and similar clinical results.'' Peripheral blood counts were in most cases evaluated once or twice weekly during the first months after CdA treatment, and such data were available from 70 patients. Blood and bone marrow sampling for flow cytometry was repeated 3 months (n = 46), 6 months (n = 50). 1 year (n = 39). and 2 years (n = 12) after C M treatment. Thus, flow cytometry was performed on at least 4 occasions in most patients. In addition, blood for flow cytometry was sampled from some patients at lymphocyte nadir 1 week (n = 24) and 1 month from the start of C M treatment (n = 7). Ten patients received a second course of C M because of residual disease. Approval was obtained from the Institu- tional Review Board. Informed consent was provided according to the Declaration of Helsinki.

3672 Blood, Vol83, No 12 (June 15), 1994: pp 3672-3681

For personal use only.on November 16, 2017. by guest www.bloodjournal.orgFrom

http://www.bloodjournal.org/

http://www.bloodjournal.org/site/subscriptions/ToS.xhtml

FLOW CMOMETRY IN CDA-TREATED HCL

Table 1. Monoclonal Antibodies

Antigen Antibody Name. Specificity

CD3 CD4 CD5

CD8 CD1 0

CD13

CD14

CD15 CD16

CD19

CD20 CD23

CD25

CD33

CD38

CD45RA

CD45RO

CD56 CD57 CD7 1 HLA-DR

TCR y6 K l h

Leu 4 Leu 3 Leu 1

Leu 2 J5

MY 7

Leu M3 MY 4

Leu M1 Leu 11

Leu 12

Leu 16 Leu 20

IL-2 R

W 9

Leu 17

Leu 18

UCHLl

Leu 19 Leu 7 OKT9 MHC II

B-ly 7

TCRG1

T-cell receptor T4. Helperiinducer cells Mature T-cells, thymocytes, B-cell subset,

B-cell chronic lymphocytic leukemia cells TB. Cytotoxic/suppressor cells cALLa. Early B and T precursors, common

acute lymphoblastic leukemia, stem cells, renal cells

Myelomonocyte lineages, intestinal cells, renal cells, central nervous system cells

Mature monocytes Myelomonocyte lineages, monoblasts, B-

Granulocytes, monocytes, embryonic cells Fcfllll. Natural killer cells, macrophages,

neutrophils, B cells, B-cell precursors, follicular

dendritic cells B cells, pre-B cells FwRII. Mature B cells, monocytes,

cell subset

macrophages, eosinophils, platelets, follicular dendritic cells

IL-2 receptor alpha chain. Activated B-cells, T-cells, monocytes, thymocytes

Granulocyte and macrophage precursors, monocytes

T10. Early or activated B and T cells, thymocytes, pre-B cells, germinal centre B cells, Ig-secreting plasma cells

macrophages

subsets, monocytes

Virgin T cells, most B cells, monocytes,

Activated and memory T cells, B-cell

N-CAM. NK cells NK cells, subsets of T cells, B cells T9. Transferrin receptor, all activated cells B cells, monocytes, macrophages, myeloid

and erythroid precursors, dendritic cells, some epithelial cells

intestinal mucosa T cellszs Minor peripheral T-cell subset lg light chains, B cells

Hairy cellsz‘ activated CD8+ ce11s,26

Data partly from Barcley et al.*’

Flow cytumetry. Blood was collected in vacuum tubes with EDTA as anticoagulant. Bone marrow cells were aspirated from the dorsal iliac crest and immediately put into heparinized tubes. Flow cytometry was performed with a FACStar (Becton Dickinson, Moun- tain View, CA) and a large panel of antibodies (Table l), including B-ly 7.24 Natural killer (NK) cells were defined as non-T cells (CD3-) expressing one or both of the markers CD16 and CD56.” All fluorescence-activated cell sorter (FACS) analyses were performed by a single investigator (R.L.). Double staining with antibodies conjugated to fluorescein isothiocyanate (FITC) or phycoerythrin (PE) was routinely performed on whole blood as previously described?’ In 84% of the pre-CdA-treatment samples, the analyses were separately performed in gated cells containing (1) the small lymphocytes and (2) the hairy cells mixed with some monocytes

3673

(Fig 1). Normal values for the different markers were assessed with an identical technique in blood from 75 controls.

Statistical analysis. Parametric and nonparametric description of data, Student’s t-test for paired or independent samples, the Wil- coxon matched pairs test, and the Kolmogornov Smirnov two-sample test were performed using the Statisticalh4ac software (StatSoft, Tulsa, OK). Patients receiving 2 courses were censored at the date of the second course.

RESULTS

Phenotype of hairy cells. The hairy cell phenotype could be assessed in 68 patients, either by analysis of the peripheral blood, bone marrow cells, or both (Table 2). The median percentage of hairy cells in the large cell gate (Fig l), as assessed by the CD20 positivity, was 90% (Fig 2), and contained less than 5% monocytes. In three quarters of the patients there were more than 80% hairy cells in the gate, enabling the identification of aberrant expression of cell markers without double fluorescent staining, which, however, confirmed expression of certain cellular markers in many patients.

Almost all hairy cells expressed CD19, CD20, CD25, HLA-DR, and CD45RA (Fig 2). A median of two-thirds of the cells expressed B-ly 7, and 8% of the patients had less than 20% B-ly 7-positive cells in the large cell gate. How- ever, unusual expression of one or another marker was found in a majority of the cases (Table 3). CD38, CD45R0, CD15, CD4, CD10, CD5 (Fig lb, plot C), andlor CD23 were occasionally expressed on a majority of the hairy cells (Fig 2). Also, CD25 and B-ly 7 were expressed only on a minority of the hairy cells in some patients.

Total lymphoid cell counts afrer C& treatment. Lym- phoid cell counts from the start of CdA treatment are shown in Fig 3. The cell populations included hairy cells in a majority of the patients at start of CdA treatment, whereas only 2 patients, both previously reported,” had residual circulating hairy cells more than 2 weeks from treatment. Patients receiving two CdA courses were censored at time of the second course. The mean blood lymphocyte count returned to 1 ,OW/ yL at 4 months from start of treatment.

Initial changes a f e r treatment in dejned lymphocyte subsets. The lymphocyte gate of all samples contained less than 3% monocytes. Through the flow cytometry data achieved at lymphocyte count nadir we were able to determine the early kinetics of the CdA effects on different cell populations. Figure 4 shows the median percentage of cell subsets for 24 patients with nadir data available. The varia- tion of the percentages in between patients was great, with less change in between measurements on different time points in individual patients (Fig 4); however, the changes were in the same direction in most patients. CD20+, CD8+ cells and NK cells were found in lower percentages at nadir than pretreatment, whereas the percentage of CD4+ cells were higher at nadir (differences significant when analyzed with either the t-test for paired samples or the Wilcoxon matched pairs test, Fig 4). This indicates a slower decrease of CD4+ cells after CdA treatment than of the other cell populations. Correspondingly, in these 24 cases, the CD4/ CD8 ratio was significantly higher at nadir as compared with




3674 JULIUSSON, LENKEI, AND LlLlEMARK

i Y

FE ->

c& ->

R1U.W 16 -> hlW 16 ->

RlUW 16 ->

IFSCl and rick ( S W angle light .cMter. The percentages of mils In the dlflerent quadrants of graphs B and C are indiuted. Lymphocytes are Fig 1. Twocolor flow Eytognmr on bone manow mlls from one patient la1 and b l d from mothet patient Ib). Graphs A show forward

found in regions R1 Ired1 of graphs A, and are plolted in gnphs B, showing the fluorescence intsnrlty for CD20 on the X-axis, and CDS on the Y-axls. Hairy celb are shown in regions R2 (green1 in graphs A, with plm in gnphs C. In Id. graph A contains 60% lymphocytes and 40% large wIIs and graph B shows predominantly T mlb, whereas in graph C there are 89.7% Cm- B dls. In Ib), graph A contains 88% lymphocytes and 12% large cells, and graph C shorn 90% B cells; 60% of these B all8 are COS', indiuting an abemnt expression of CD5 on the hairy dl..

the pretreatment value (median, 1.81 and 1.27, respectively; P = ,037. t-test for paired values). On the other hand, a significantly higher proportion of NK cells and C D 8 cells were found at subsequent time points as compared with nadir. Thus, the recovery from nadir was more rapid for NK cells and CD8' cells than for CD4+ cells and B cells, leading to a slowly but progressively decreasing CD4/CD8 ratio. The absolute cell counts for various lymphoid cell subsets

in relation to treatment are shown in Fig 5 and Table 4. The median CD4' cell count was close to the normal range 1 year after the s m of CdA treatment, whereas the CD8+ cells normalized within 3 months. Of interest was that several specific B- and T-cell subsets had a quicker recovery than the overall B- and T-cell populations (Fig 5 and Table 4). The results were not different whether the patients had

received CdA intravenously or subcutaneously (Kolmogor- nov Smimov nonparametric two-sample test. Table 5).

of T-cell receptor y6 lymphocytes decreased initially from T-cell receptor 76 positive lymphocytes. The percentage

a median of 2.2% to 1.4% (normal, 1% to 10%). At the 3- month evaluation, the median value was 6%. one-third had more than 10%. and a few patients had an overshoot to 40% or more. Subsequently. both relative and absolute values n o d z e d .

CDIO' cells in bone marrow afer CdA rreamenf. The mean percentage ? SD CDIO' cells from bone marrow in

The corresponding figures at 3 months, 6 months, 1 year, and the lymphocyte gate was 1.30% ? 5.92% before treatment.

2 years after CdA treatment were 3.53% 2 8.018, 5.43% ? 6.13, 3.56% ? 4.24%. and 1.33% ? 1.590, respectively.




FLOW CMOMETRY IN CDA-TREATED HCL 3675

Table 2. Percentage of Positive Cells in Hairy Cell Gate Table 3. Number of Patients With Aberrant Phenotype Expression

Marker No. of Patients Mean so With Aberrant

Single fluorescent staining Marker Limit (%)

ExpressionEvaluable (010)

Double Staining

CD45RA CD20 CD19 HLA-DR B-ly 7 CD38 CD45RO CD4 CD7 1

Percentage of positive cells among CD19' cells

CD25 CD23 CD1 5 CD5 CDlO

36 68 68 54 64 60 32 62 58

58 39 55 63 62

89.4 10.7 85.3 14.3 84.1 17.5 83.3 19.3 61.2 24.9 23.4 20.8 14.8 15.2 11.0 12.3 6.4 8.7

68.5 27.5 11.5 14.5 11.3 14.4 8.7 16.4 4.4 15.4

These values were significantly different from the baseline value by nonparametric statistics at 3, 6, and 12 months (Wilcoxon matched pairs test: P = .036, P = .0007, and P = .0017, respectively). Sixteen of 36 evaluable patients (44%) with less than 2% CDIO' cells of the pretreatment bone marrow lymphoid cells showed an increase to 5% or more at follow-up, with the kinetics shown in Fig 6.

DISCUSSION

By flow cytometry with an extensive set of antibodies and double fluorescent staining, we have analyzed the phenotype of hairy cells from 68 patients with symptomatic disease. Furthermore, by frequent samplings from 50 patients after CdA treatment, we were able to determine the selective tox- icity of this agent towards specific lymphoid cell subsets and their subsequent recovery.

CD38 CD1 5

CD23

CD5

CD4 CDlO

CD25 Bly ?/CD19

Double staining with CD19 in 4 cases, otherwise defined as more than 20% CD4* in the hairy cell gate and percentage of CD19 + percentage of CD4 greater than 110.

t Ratio between percentage of B-ly 7' and percentage of CD19 in different tubes.

Percentage Positive Cells In Hairy Cell Gate

>20 >5

>20 >5

>20 >5

>20 >20 >5

>20 <20 < 50

24/55 (44) 32/55 (58) 10155 (18) 20139 (51)

8/39 (21) 24/63 (38)

6/63 (1 0) 7/59 (12) 8/62 (13) 3/62 (5) 4/58 (7)

13/66 (20)

The phenotype of hairy cells is known to be CD19', CD20', CD22', CD25', B-ly 7', CD5-, and CD1 lc+.2.24.30 We also found that HLA-DR and CD45RA"' were consis- tantly expressed. In addition, hairy cells from a large proportion of the patients express other markers. CD1 I C is a monocyte marker that is typically found on hairy cells,"* but we also frequently found expression of CD15, another monocyte marker. Expression of CD4, the T-helpedinducer cell marker; CDIO, the common acute lymphoid leukemia antigen indicating early B-cell differentiation; and CD5, the T- cell marker typically found on B-cell chronic lymphocytic leukemia, were also verified by double fluorescent staining in significant numbers of patients. Thus, hairy cells com- monly have aberrant phenotypes, with simultaneous expression of divergent markers of differentiation and cell lineage.

20

IO

0

No CD19

CD19

CD20

CD19* CD19

CD19 Not

I ' 1 3rdquartile I 2ndquartile I lstquartile

CD45RA CD20 CD19 HLA-DR B-ly 7 CD38 CD45RO CD4 CD71

Percentage of CD19+ Cells

CD25 CD15 CD23 CD5 CDlO

Fig 2. The percentage of cells staining positive with specified markers in the hairy cell gate. Data on CD25, CD15, CD23, CD5, and CD25 indicate the percentage of positive cells among CD19' cells as assessed by double-fluorescence staining. Shaded areas represent the four quartiles.





CelIdpL Blood Lymphocyte Counts 100000

I I ._...- l

10000

1000

70 66 50 35 27 20 12

10 1 10 1 00 1000

Days from Start of CdA Treatment

Hairy cells from some patients also lacked expression of characteristic markers, such as CD2S3 and B-ly 7, at least in a proportion of the leukemic cells. The identification of minor subpopulations of hairy cells with aberrant phenotype mostly requires double fluorescent staining using the relevant combination of markers. Because far from all combinations were possible to determine, it is likely that aberrant expression is more common than we have been able to document. Cells from the disease entity called HCL variant4 lack CD25,

Fig 3. Total lymphocyte counts during and after CIA treatment. Day 1 is the baseline. Patients receiving two courses are censored at the time of the second course. F " ) Median values; (-1 quartiles, 5% and 95% valuos; and ( . . . l extreme values. The number of patients is indicated in the graph.

but were recently found to express IL-2 receptor p chains.33 Our phenotype data rather indicate a spectrum of HCL vari- ants than one typical variant entity.

We did not find any correlation between the phenotype and the response to treatment (data not shown). CdA gives a much higher complete remission rate in HCL than interferon,8 and thus divergent response to CdA treatment according to any marker is difficult to assess.

There is limited previous data available on the immune

Median Percentage

l

I \ 0.0004

Fig 4. The percentage of mUs staining positive CDU) with CD4, CD8, and CD20, and NK cells in the lym- l o phocyte gate of blood cells in relation to CdA treat- 0.003 ment. (-1 CD4 and NK cells; ( - - - I CD8 and CD20. P NK values achieved from the t-test for paired values indi- 0 J

cate differences in the percentages at indicated time Baseline 1 Week 1 Month 3 Months 6 Months points compared with the percentages at nadir, and are shown in italics for CD8 and NK cells. Time from Start of CdA treatment




FLOW CYTOMETRY IN CDA-TREATED HCL 3611

Cells/pl CD4/HLA c 1000 " .-...-- l

Cells/pl CD4

1 10 1 00 1000 Days

Cells/pl CD8 B IOOOO --.- 1

1000

100

1 -. .-. . . . . . . . ................

I . ..

1 00

10

............... I

Cellslpl CDS/HLA D I O O O ~ " -

100

10

.... ........" ......... .........

1 10 loo Days 1000 1 I . . . " . . . I . . . ".:.I . . . . . . . .

Cells/pl NK

100

10

Fig 5. Absolute number of cells per microliter in blood with the phenotypes (A) CD4'. (B) CD8'. (C) CD4'/HLA+, (D) CD8+/HLA+, (E) CD3-/ CD16' or CD56' (ie, NK cells), (F) CD20+, (G) CDZO'ICD5'. (H) CD20'/CD23+, according to days from start of CdA treatment. Thick solid lines indicate median values, thin solid lines quartiles, and dotted lines extreme values. Normal ranges are indicated by horizontal hatched lines.

suppression caused by CdA ~ e a t m e n t . " . ' ~ . ~ ' . ~ ~ ~ ~ ' An early decrease in both CD4' and CD8' cells after treatment is found, and a slow recovery during the subsequent year was indicated from sporadic observations. From the combined data collected at the Scripp's clinic and the MD Anderson Cancer Center, it was concluded that the mean CD4 count nadir was 272/yL and occurred 4 to 6 months from treatment." This discrepancy to our data probably reflects the lack of systematic phenotype analyses in the early posttreatment phase in the US studies.

In the present study, we have characterized the degree and

the kinetics of both the decrease and the recovery as regards several of the lymphocyte subsets. During CdA treatment, there is a rapid decrease of most of the studied lymphocyte subsets, but CDS, CD20, and NK-cell subsets decreased significantly more than CD4 cells. The median CD8 cell count at nadir was well below lOO/yL, whereas the corresponding CD4 count was 128/yL. The recovery from nadir was quicker for CD8 cells, NK cells, and T-cell receptor yS' cells, and the median blood counts of these subsets were all normal within 3 months, whereas CD4+ cells required 1 year or more for normalization, leading to a slowly but progres-





- Cells/pl CD20 F 1000

1 00

10

1

CD20/CD5

H '

100

10

Cells/pl CD20/CD23 000

100

10

1 1 10

Fig 5. (Cont'd).

sively decreasing CD4/CD8 ratio. Of specific interest was that the activated HLA-DR+ subset of CD4+ cells was fully normalized already at 3 months after CdA treatment, as were the NK cells, the CD5+ B cells, and the CD45RNCD45RO cells (Fig 5 and Table 4). Furthermore, the absolute numbers of HLA-DR+ subset of CD8+ cells decreased within the normal range, and often recovered with an overshoot. We have previously documented post-CdA-treatment overshoot of other leucocytes, such as monocytes and eosinophils." However, CD4'/CD45RA' cell counts had a late nadir, almost within the normal range.

CdA-treated HCL patients are at great risk for neutropenic fever if they have pretreatment pancytopenia, especially anemia.'' However, no infections present later than 1 month from CdA treatment, despite continuing low levels of several lymphocyte subsets during 6 to 12 months. The activated CD38' and HLA-DR' T-cell subsets, the CD5' B cells, and the NK cells that rapidly normalize after CdA-treatment might thus, in combination with the recovering neutrophils," be an adequate first-line defense.

As regards the B cells, the CD23+ B cells also recovered promptly from 3 to 6 months, whereas the total CD20' cell population required 1 year for normalization of the cell counts. Increased relative numbers of B-cell progenitors, ie, CDIO' cells, appeared in the bone marrow after treatment in 16 of 36 patients (Fig 6). We have seen the similar phenomenon during successful CdA treatment of chronic lymphocytic leukemia.36 It is unlikely that these CDlO+ B cells represented leukemia cells, because they were seen primarily in patients with leukemia cell clones laclung CDIO, and in those achieving complete remission. There was a normalization of the percentage of CD10' cells with longer follow-up without further treatment. This could not be caused by a selective depletion of bone marrow subsets other than CD10' cells, according to bone marrow cellularity, morphology, and flow cytometry data (not shown). It is more likely that we observed a regeneration of normal B cells, enabled by an effective and brief cytotoxic treatment. Such B-cell regeneration might be a common but hitherto un- known phenomenon after successful chemotherapy, because CDIO' cells were very rare in the peripheral blood, and systematic phenotype studies on bone marrow cells have not been performed previously. However, CDlO+ bone marrow cells are described to appear after autologous bone marrow tran~plantation.'~ B-cell regeneration is probably also mir- rored by the recent finding that CdA3x,39 and fl~darabine,''',~' in contrast to conventional chem~therapy,".~' may increase polyclonal gammaglobulin levels in patients with chronic lymphocytic leukemia. However, we cannot exclude the pos- sibility that CdA has a direct stimulatory effect on normal B-cell progenitors, which, if it would occur with a lower dose range than the cytotoxic effect, could be exploited in B-cell dysfunction states.

As yet, no studies have attempted to optimize CdA scheduling. Monitoring of the immune cells by flow cytometry and the time to recovery from cytopenias after various treatment schedules might facilitate the establishment of an optimal regime without the need for large-scale clinical trials. The immunosuppressive effects of CdA may also be possible to use in nonmalignant disorders, such as rheumatoid arthritis and multiple ~clerosis.~ How cytometry monitoring might then again prove useful when establishing adequate dosing and scheduling of the treatment.

Cytopenias and infections are the major side effects for most agents used for treatment of malignancy, and have to be considered when selecting treatment. DCF is highly effective therapy for hairy cell leukemia, but has to be administered in repeated doses. A long-standing suppression of CD4+ cells has, in disorders other than HCL, been associated





Table 4. Absolute Blood Lymphoid Cell Counts in Relation to CdA Treatment

Normal

Baseline 1 wk 3 mo 6 mo 12 mo 24 mo Mean ? SD Range

No. of patients 59 24 46 50 39 12 75

CD20 1,590 f 6,204 22 f 35 73 f 118 136 2 174 206 f 156 273 f 188 252 f 116 108-481

CD20iCD5 511 f 1,712 9 f 15 76 f 186 85 f 136 88 f 64 74 f 40 47 f 29 12-121

CD20lCD23 213 f 916 7 f 16 21 f 57 67 f 144 100 % 116 116 f 76 50 f 43 14-110

CD8 986 f 2,049 105 2 125 499 f 750 538 -C 438 572 f 407 851 f 789 568 2 255 256-1,050

CD8lHLA 246 f 979 2 0 2 2 8 233% 568 141 f 159 128 f 136 199 f 275 44 f 33

CD4 924 f 916 207 f 271 313 2 205 365 f 287 429 2 183 551 -C 295 858 t 330 429-1,492

CDUHLA 111 f 284 49 2 123 84 f 111 8 9 % 104 77 2 6 0 102 f 117 49 f 23 15-117

CD4iCD45RA 394 2 373 124 f 63 99 f 75 112 f 90 141 f 104 184 f 98 412 f 189 108-627

CD45RNCD45RO 157 f 200 38 f 39 169 f 348 165 f 159 217 2 296 254 f 294 321 f 205 81-431

CD38 646 f 573 179 -C 167 619 % 641 629 ?. 475 788 f 428 1,075 % 612 963 f 373 486-1,761

CD57 588 Ifr 887 102 f 92 446 f 555 410 f 302 451 f 393 861 f 810 292 f 198 81-698

NK 216 f 409 1 5 f 8 176 f 168 176 f 116 239 f 170 341 2 245 163 f 86 77-332

75

11-1 10

Values are in cells per microliter (mean f SD).

with severe opportunistic infections, such as candida pneu- monitis, central nervous system infections with Listeria and Aspergillus, and disseminated herpes zoster.I5 In Steis et al's

of dCF in HCL, the median nadir of CD4+ cell counts was 54/j~L,4~ and in other studies the median CD4 nadir was between 70 and 1 5 5 / j ~ L . ' ~ . ' ~ * ~ . ~ ~ Normalization of CD4 counts and CD8 counts may occur a median of 23 and 15 months after the start of dCF treatment, respectively." Be- cause the T-cell counts remain low during the full duration of dCF treatment,I3 the immune depression is likely to be shorter after treatment with CdA than with dCF. However, the immune suppression is likely to be dose-related for both

Table 5. Absolute Blood Lymphoid Cell Counts in Relation to CdA Treatment and Mode of Administration

Marker and Intravenous Subcutaneous Normal Time (n = 17) (n = 43) (5%-95%)

CD4 Baseline 720 (426-963) 660 (409-1.169) 429-1,492 1 wk 85 (56-354) 158 (99-202) 3 mo 300 (129-540) 249 (173-367) 6 m0 340 (195-517) 306 (213-398) 12 mo 415 (260-551) 399 (325-486)

Baseline 595 (429-764) 546 (300-875) 256-1,050 1 wk 47 (21-134) 80 (58-103) 3 mo 540 (175-660) 234 (138-539) 6 mo 549 (226-869) 440 (211-580) 12 mo 526 (297-918) 429 (301-661)

CD8

CD20 Baseline 184 (88-382) 165 (60-506) 108-481 1 wk 4 (3-26) 18 (4-24) 3 mo 83 (15-203) 53 ( 12-57) 6 m0 145 (41-234) 129 (38-115) 12 mo 275 (66-407) 180 (99-212)

There was no statistical difference between the modes of administration (Kolmogornov Smirnov nonparametric two-sample test). Val- ues are median cells per microliter with quartile values in parenthe- ses.

dCF and CdA, and the lowest effective dose for HCL has not been established for either of the drugs.

Fludarabine is not recommendable for HCL, but it is an important nucleoside analog for the treatment of other lymphoid neoplasms' and is structurally closely related to CdA.3S Fludarabine also produces a severe depletion of CD4+ cells when administered in repeated courses to patients with chronic lymphocytic leukemia and lymph0ma.4~.~~,~'

Interferon, the established first-line treatment of HCL;' does not induce T-cell depletion to a degree similar to that of the nucleoside analogues. However, HCL by itself causes immunologic impairment, the complete response rate to interferon is considerably lowe? than for CdA, and the im- provement is also slower." Thus, patients may die from opportunistic infections during induction treatment with in- terfer0n.4~~~~

In conclusion, CdA is highly cytotoxic towards all lymphoid subsets during treatment. These short-term effects cause a significant acute morbidity through neutropenic in-

Percent CD10-positive cells in bone marrow

30

25

0 3 6 9 Months 12

Fig 6. Percentage of CD10+ cells in the lymphocyte gate of bone marrow cells from l6 individual patienb according to time from start of CdA treatment.





fections.’8 Thus, the evaluation of other dose schedules, facil- itated by the use of subcutaneous administration’9.21 and flow cytometry monitoring, should be performed. Prolonged immunosuppression from CdA is not a significant clinical prob- lem in HCL. Partial regeneration of several normal subsets occurs rapidly, whereas the full recovery may be delayed, although this recovery is quicker than after dCF treatment. Drug-induced immune suppression will be a major concern when choosing between the available treatment options for lymphoproliferative disorders,” and this has to be considered in comparative trials in the near future.

REFERENCES I . Bouroncle BA, Wiseman BK, Doan C A Leukemic reticuloen-

dotheliosis. Blood 13:609, 1958 2. Jansen J, LeBien TW, Kersey JH: The phenotype of the neo-

plastic cells of hairy cell leukemia studied with monoclonal antibodies. Blood 59:609, 1982

3. Korsmeyer SJ, Greene WC, Cossman J, Hsu SM, Jensen JP, Neckers LM, Marshall SL, Bakhshi A, Depper JM, Leonard WJ, Jaffe ES, Waldmann TA: Rearrangement and expression of immuno- globulin genes and expression of Tac antigen in hairy cell leukemia. Proc Natl Acad Sci USA 80:4522, 1983

4. Catovsky D, O’Brien M, Melo JV, Wardle J, Brozovic M: Hairy cell leukemia (HCL) variant: an intermediate disease between HCL and B prolymphocytic leukemia. Semin Oncol 11:362, 1984

5. Lauria F, Raspadori D, Foa R, Zinzani PL, Buzzi M, Gugliotta L, Macchi S, Tura S: Reduced hematologic response to alpha-interferon therapy in patients with hairy cell leukemia showing a peculiar immunologic phenotype. Cancer 65:2233, 1990

6. Brogden RN, Sorkin EM: Pentostatin: A review of its pharma- codynamic and pharmacokinetic properties, and therapeutic potential in lymphcproliferative disorders. Drugs 46652, 1993

7. Ross SR, McTavish D, Faulds D: Fludarabine. A review of its pharmacological properties and therapeutic potential in malignancy. Drugs 45:737, 1993

8. Beutler E: Drug profiles. Cladrihine (2-chlorodeoxyadenosine). Lancet 340:952, 1992

9. Bryson HM, Sorkin EM: Cladribine: A review of its pharmaco- dynamic and pharmacokmetic properties and therapeutic potential in haematological malignancies. Drugs 46:872, 1993

IO. Simon LN, Bauer RJ, Tolman RL, Robins RK: Calf intestine adenosine deaminase. Substrate specificity. Biochemistry 9:573, 1970

1 I . Spiers ASD, Paresh SJ, Bishop MB: Hairy-cell leukemia: Induction of complete remission with pentostatin (2‘-deoxycoformycin). J Clin Oncol 2:1336, 1984

12. Kraut EH, Bouroncle BA, Grever MR: Pentostatin in the treatment of advanced hairy cell leukemia. J Clin Oncol 7:168, 1989

13. Urba WJ, Baseler MW, Kopp WC, Steis RG, Clark JW, Smith JW 11, Coggin DL, Longo DL: Deoxycoformycin-induced immunosuppression in patients with hairy cell leukemia. Blood 73:38, 1989

14. Kraut EH, Neff JC, Bouroncle BA, Gochnour D, Grever MR: Immunosuppressive effects of pentostatin. J Clin Oncol 8:848, 1990

15. O’Dwyer PJ, Spiers ASD, Marsoni S: Association of severe and fatal infections and treatment with pentostatin. Cancer Treatment Rep 70: 1 1 17, 1986

16. Piro LD, Carrera CJ, Carson DA, Beutler E: Lasting remission in hairy-cell leukemia induced by a single infusion of 2-chlorodeoxyadenosine. N Engl J Med 322: 1 1 17, I990

17. Estey EH, Kurzrock R, Kantarjian HM, O’Brien S, McCredie KB, Beran M, Koller C, Keating MJ, Hirsch-Ginsberg C, Huh YO,

Stass S, Freireich EJ: Treatment of hairy cell leukemia with 7” chlorodeoxyadenosine. Blood 79:882, 1992

18. Juliusson G , Liliemark J : Rapid recovery from cytopenia in hairy cell leukemia following treatment with 2-chloro-2’-deoxyadenosine (CdA): Relation to opportunistic infections. Blood 79:888. 1992

19. Juliusson G, Liliemark J, Hippe E, Blichfeldt P, Wallman K, Hagberg H, Stolt C-M, Bergheim J, Winquist I, Hedenus M, Carlsson M, Vilen L, Hellquist L, Brinck L, Kristofferson R, Ly BE, Wart J: Subcutaneous injections of 2-chloro-2’-deoxyadenosine (CdA) as treatment for symptomatic hairy cell leukemia (HCL). Blood 80359a, 1992 (abstr, suppl 1 )

20. Juliusson G, Liliemark J: Pretreatment anemia but not neutro- penia predicts neutropenic fever following treatment with 2-chloro- 2‘-deoxyadenosine (CdA) for symptomatic hairy cell leukemia (HCL). Fifth International Conference on Malignant Lymphoma, Lugano, Switzerland, June 9-12, 1993, p 150 (abstr)

21. Ortho Biotech Inc, Raritan, NJ: FDA drug registration file for cladribine (2-chlorodeoxyadenosine). Data presented in Nelson MC. Turner WJ: Leustatin (Cladribine) Injection. Guide to Continuouh Infusion. Drug Information Booklet. 1993

22. Liliemark J, Pettersson B, Juliusson G: The relationship between plasma 2-chloro-2‘-deoxyadenosine (CdA) and cellular CdA nucleotides (CdAN) after intermittent and continuous i.v. infusion in patients with chronic lymphocytic leukemia. Blood 78:33a, 1991 (abstr, suppl 1)

23. Liliemark J, Albertioni F, Hassan M, Juliusson G: On the bioavailability of oral and subcutaneous 2-chloro-2’-deoxyadenosine in humans: Alternative routes of administration. J Clin Oncol 10:1514, 1992

24. Visser L, Shaw A, Slupsky J , Vos H, Poppema S: Monoclonal antibodies reactive with hairy cell leukemia. Blood 74:320, 1989

2.5. Mulligan SP, Travade P, Matutes E, Dearden C, Visser L. Poppema S, Catovsky D: B-ly 7, a monoclonal antibody reactive with hairy cell leukemia, also defines an activation antigen on normal CDX+ T-cells. Blood 76:959, 1990

26. Micklem KJ, Dong Y, Willis A, Pulford KA, Visser L, Dur- kop H, Poppema S, Stein H, Mason DY: HML-I antigen on mucosa- associated T cells, activated cells, and hairy leukemic cells is a new integrin containing the p7 subunit. Am J Pathol 139:1297, 1991

27. Barclay AN, Birkland ML, Brown MH, Beyers AD, Davis SJ, Somoza C , Williams A F The Leucocyte Antigen Facts Book. London, UK, Academic, 1993

28. Lanier LL, Le AM. Civin CI, Loken MR, Philips JH: The relationship of CD16 (Leu 1 1 ) and Leu 19 (NKH-I) antigen expression on human peripheral blood NK cells and cytotoxic T lymphocytes. J lmmunol 135:4480, 1986

29. Lenkei R, Bjork 0, Ost A, Biberfeld P: Expression of CD15 as predictor of relapse in children with acute lymphoblastic leukaemia of the pre-B type. Leuk Res 15:189, 1991

30. Robbins BA, Ellison DJ, Spinosa JC, Carey CA, Lukes RJ, Poppema S, Saven A, Piro LD: Diagnostic application of two-color flow cytometry in 16 I cases of hairy cell leukemia. Blood 82: 1277, 1993

31. Zola H, Siderus N. Flego L, Sparrow R, van der Weyden MB, Nimmo J, Hart DNJ, Houghton S, Boyd AW: Expression of CD45 isoforms in chronic B-cell leukaemias. Leuk Res 17:209, 1993

32. Vardiman JW, Gilewski TA, Ratain MJ, Bitter MA, Bradlow BA, Golomb HM: Evaluation of Leu-MS (CD1 I C ) in hairy cell leukemia by the alkaline phosphatase anti-alkaline phosphatase technique. Am J Clin Pathol 90:250, 1988

33. de Totero D, Tazzari PL, Lauria F, Raspadori D, di Celle PF, Carbone A. Gobbi M, Foa R: Phenotypic analysis of hairy cell





leukemia: “Variant” cases express the interleukin-2 receptor p chain, but not the a chain (CD25). Blood 82528, 1993

34. Carrera CJ, Piro LD, Saven A, Cox K, Beutler E, Carson D, Kipps TJ: Restoration of lymphocyte subsets following 2-chlorodeoxyadenosine remission induction in hairy cell leukemia. Blood 76:260a, 1990 (abstr, suppl 1)

35. Lauria F, Raspadori D, Rondelli D, Benfenati D, Tura S: Lymphocyte modifications following 2-chlorodeoxyadenosine treatment in 22 patients with hairy cell leukemia. Blood 80:431a, 1992 (abstr, suppl 1)

36. Juliusson G , Liliemark J: Complete remission of B-cell chronic lymphocytic leukaemia after oral cladribine. Lancet 34154, 1993

37. Smedmyr B, Bengtsson M, Jakobsson A, Simonsson B, Oberg G , Totterman TH: Regeneration of CALLA (CD10+), TdT+ and double-positive cells in the bone marrow and blood after autologous bone marrow transplantation. Eur J Haematol 46:146, 1991

38. Juliusson G, Elmhorn-Rosenborg A, Liliemark J: Response to 2-chloro-2’-deoxyadenosine in patients with B-cell chronic lymphocytic leukemia resistant to fludarabine. N Engl J Med 327:1056, 1992

39. Juliusson G , Liliemark J: High complete remission rate from 2-chloro-2’-deoxyadenosine (CdA) in previously treated patients with B-cell chronic lymphocytic leukemia: Response predicted by rapid decrease of blood lymphocyte count. J Clin Oncol 11:679, 1993

40. Wijermans PW, Gemts WBJ, Haak HL: Severe immunodefi- ciency in patients treated with fludarabine monophosphate. Eur J Haematol 50:292, 1993

41. Schering AG, Berlin (data on file) 42. Rozman C, Montserrat E, Vinolas N: Serum immunoglobulins

in B-chronic lymphocytic leukemia. Natural history and prognostic significance. Cancer 61:279, 1988

43. Steis RG, Urba WJ, Kopp WC, Alvord WG, Smith JW 11, Longo D L Kinetics of recovery of CD4’ T cells in peripheral blood of deoxycoformycin-treated patients. J Natl Cancer Inst 83:1678, 1991

44. Thaler J, Griinewald K, Gattringer C, Ho AD, Weyrer K, Dietze 0, Stauder R, Fluckinger T, Lang A, Huber H: Long-term follow-up of patients with hairy cell leukaemia treated with pentostatin: Lymphocyte subpopulations and residual bone marrow infil- tration. Br J Haematol 84:75, 1993

45. Blick M, Lepe-Zuniga JL, Doig R, Quesada JR: Durable complete remission after 2‘-deoxycoformycin in patients with hairy cell leukemia resistant to interferon alpha. Am J Hematol 33:205, 1990

46. Bergmann L, Fenchel K, Jahn B, Mitrou PS, Hoelzer D: Immunosuppressive effects and clinical response of fludarabine in refractory chronic lymphocytic leukemia. Ann Oncol 4:371, 1993

47. Juliusson G: Immunological and genetic abnormalities in chronic lymphocytic leukaemia. Impact of the purine analogues. Drugs (in press)

48. Golomb HM, Fefer A, Golde DW, Ozer H, Portlock C, Silber R, Rappeport J, Ratain MJ, Thompson J, Bonnem E, Spiegel R, Tensen L, Burke JS, Vardiman J W : Survival experience of 195 patients with hairy cell leukemia treated in a multi-institutional study with interferon-alfa 2b. Leuk Lymph 4:99, 1991

49. Berman E, Heller G, Kempin S, Gee T, Tran L-L, Clarkson B: Incidence of response and long-term follow-up in patients with hairy cell leukemia treated with recombinant interferon alfa-2a. Blood 752339, 1990

50. Keating MJ: Immunosuppression with purine analogues- The flip side of the gold coin. Ann Oncol 4:347, 1993 (editorial)




1994 83: 3672-3681

G Juliusson, R Lenkei and J Liliemark after treatment with 2-chlorodeoxyadenosinehairy cell leukemia: phenotype of hairy cells and lymphocyte subsets Flow cytometry of blood and bone marrow cells from patients with

http://www.bloodjournal.org/content/83/12/3672.full.htmlUpdated information and services can be found at:

Articles on similar topics can be found in the following Blood collections

http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:

http://www.bloodjournal.org/site/misc/rights.xhtml#reprintsInformation about ordering reprints may be found online at:

http://www.bloodjournal.org/site/subscriptions/index.xhtmlInformation about subscriptions and ASH membership may be found online at:

Copyright 2011 by The American Society of Hematology; all rights reserved.Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American


http://www.bloodjournal.org/content/83/12/3672.full.html

http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests

http://www.bloodjournal.org/site/misc/rights.xhtml#reprints

http://www.bloodjournal.org/site/subscriptions/index.xhtml





Documents

Flow Cytometry of Blood and Bone Marrow Cells From ...€¦ · AIRY CELL LEUKEMIA' (HCL) is a rare B-cell dis- order characterized by cytopenia, splenomegaly and infections. Hairy