Accepted Manuscript Residual Cancer Lymphocytes in Patients With Chronic Lymphocytic Leukemia After Therapy Show Increased Expression of Surface Antigen Cd52 Detected By Quantitative Fluorescence Cytometry Michaela Pevna, Michael Doubek, Petr Coupek, Olga Stehlikova, Martin Klabusay PII:
S2152-2650(14)00148-7
DOI:
10.1016/j.clml.2014.06.006
Reference:
CLML 432
To appear in:
Clinical Lymphoma, Myeloma and Leukemia
Received Date: 17 October 2013 Revised Date:
9 May 2014
Accepted Date: 4 June 2014
Please cite this article as: Pevna M, Doubek M, Coupek P, Stehlikova O, Klabusay M, Residual Cancer Lymphocytes in Patients With Chronic Lymphocytic Leukemia After Therapy Show Increased Expression of Surface Antigen Cd52 Detected By Quantitative Fluorescence Cytometry, Clinical Lymphoma, Myeloma and Leukemia (2014), doi: 10.1016/j.clml.2014.06.006. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
RESIDUAL CANCER LYMPHOCYTES IN PATIENTS WITH CHRONIC LYMPHOCYTIC LEUKEMIA AFTER THERAPY SHOW INCREASED
QUANTITATIVE FLUORESCENCE CYTOMETRY
RI PT
EXPRESSION OF SURFACE ANTIGEN CD52 DETECTED BY
Running title: CD52 detection in chronic lymphocytic leukemia
SC
Michaela Pevna1, Michael Doubek2, Petr Coupek1, Olga Stehlikova2, and
1
M AN U
Martin Klabusay1
International Clinical Research Center– Integrated Center of Cellular Therapy
and Regenerative Medicine
St. Anne’s University Hospital Brno
Department of Internal Medicine – Hematooncology
EP
2
TE D
Pekarska 53, 656 91 Brno, Czech Republic
University Hospital Brno
AC C
Jihlavska 20, 625 00 Brno, Czech Republic
1
ACCEPTED MANUSCRIPT
Corresponding author: Martin Klabusay, M.D., Ph.D.
RI PT
Associate Professor Address:
SC
St. Anne’s University Hospital
M AN U
International Clinical Research Center
Integrated Center of Cellular Therapy and Regenerative Medicine Pekarska 53
Tel.:
EP
Czech Republic
TE D
56591 Brno
+420603167625
AC C
Email address:
[email protected]
Number of Words
abstract: 244 manuscript: 3086
2
ACCEPTED MANUSCRIPT
MICROABSTRACT: The quantitative determination of the expression of CD20 and CD52
RI PT
antigens in chronic lymphocytic leukemia (CLL) is important for treatment with monoclonal antibodies. Patients with CLL in complete or partial
remission have higher level of CD52 antigen expression compared to patients
SC
with CLL untreated, in progression, or diagnosed with small lymphocytic lymphoma. Our results support possible significance of alemtuzumab
M AN U
consolidation.
ABSTRACT:
TE D
Background: Rituximab and alemtuzumab, monoclonal antibodies (mAbs) used in recent years to treat chronic lymphocytic leukemia (CLL), are directed against antigens CD20 and CD52. CD20 is not highly expressed by
EP
CLL tumor cells, and rituximab does not have significant effectiveness in CLL unless combined with chemotherapy. Alemtuzumab targets CD52, which is
AC C
much more highly expressed, and is currently the most effective agent in monotherapy for CLL. Variability in expression of both antigens among these patients may be related to different individual therapeutic responses to mAb therapy. Patients and Methods: A total 95 patients diagnosed with CLL and/or small lymphocytic lymphoma (SLL) were divided into four groups: A)
3
ACCEPTED MANUSCRIPT
untreated, B) in complete or partial remission, C) disease in progression, and D) diagnosed with SLL. Flow cytometry of peripheral blood cells included gating of CD5+CD19+tumor population, within which mean fluorescence
RI PT
intensity of fluorescein isothiocyanate (FITC) conjugated with anti-CD20 or anti-CD52 antibody was measured. The resulting expression of the two
antigens was deduced from the calibration curve using QuantumTM FITC
SC
particles.
M AN U
Results: Expression of CD20 showed no significant differences among the four groups of patients. However, significantly higher expression of surface antigen CD52 was recorded in patient group B in complete or partial remission (p< 0.001).
Conclusion: The residual population of CLL cells after therapy is
TE D
characterized by increased surface detection of CD52. Although the exact cause of this phenomenon is unknown, our results provide a basis to consider
EP
the potential for CLL consolidation therapy using alemtuzumab.
AC C
Key words: small lymphocytic lymphoma, remission, surface CD52, CD20, quantitative fluorescence cytometry
4
ACCEPTED MANUSCRIPT
INTRODUCTION: Despite considerable progress in its therapy in recent years (1), chronic
RI PT
lymphocytic leukemia (CLL) remains an incurable disease, with the exception of when hematopoietic stem cell transplantation (HSCT) is pursued. CLL is a tumor disease originating from the B-cell lineage. Antigens CD20 and CD52
SC
expressed on the surface of CLL cells are the targets for the therapeutic
M AN U
activities of monoclonal antibodies. If rituximab (an anti-CD20 antibody) in combination with chemotherapy or alemtuzumab (an anti-CD52 antibody) in monotherapy is used in first-line CLL treatment, the therapeutic responses achieved are in the range of 83–95% (2–4).
The antigen CD20 is a classic marker of B-lymphocytes (5). CD20 is a
TE D
phosphoprotein with four transmembrane domains and both of its ends extend into the cytoplasm (6). Outside the plasma membrane there are two loops of
EP
protein, one large and one small. It is believed that CD20 acts as a storeoperated calcium channel, which is significant for the cell homeostasis of
AC C
calcium and for passage through the cell cycle (7, 8). In contrast to other Blymphoproliferative diseases, the expression of CD20 in CLL is low (9). CD52 is expressed by cells within the immune system (lymphocytes,
monocytes, macrophages and eosinophils) and by cells of the epididymal tissue in the male genital tract (10). CD52 is a glycopeptide whose peptide chain is composed of just 12 amino acids (11). The greater part of the molecule consists in the side hydrocarbon complex bound outside the plasma
5
ACCEPTED MANUSCRIPT
membrane. The C-terminus of the peptide chain is loosely bound to the outer layer of the cell membrane by a glycosylphosphatidylinositol lipid (GPI) anchor. The GPI anchor is a posttranslational modification common to a group
RI PT
of proteins typically constituting a component of lipid rafts (12). Lipid rafts are sphingolipids- and cholesterol-enriched membrane micro domains (13). Proteins with GPI anchor may spontaneously transfer from one cell and
SC
incorporate themselves into the cell membrane of a different cell (12). CD52’s
M AN U
function in the immune system is not known, and its expression by CLL cells is lower than that in, for example, either mantle cell lymphoma or normal Blymphocytes (14).
We assume that in addition to cell cycle arrest and direct apoptosis induction the effectiveness of the monoclonal antibodies on tumor cells occurs
TE D
through effector mechanisms of the immune system, the necrosis-causing complement-dependent cytotoxicity (CDC) and antibody-dependent cellular
EP
cytotoxicity (ADCC). Although we know neither the complex working of these mechanisms in vivo nor their role in individual compartments, studies
AC C
have been able to illustrate them separately in vitro. After binding to rituximab the CD20 molecule moves to lipid rafts, where it triggers calcium influx and crosslinking of surface-bound rituximab by a secondary antibody enhances a calcium influx and signal transduction, which leads to apoptosis (15). In contrast, the CD52 is preferentially localized in lipid rafts and crosslinking of the bound alemtuzumab on CLL cells leads to the formation of aggregates of lipid rafts and induction of cell death (16). Even if this constitutes merely a
6
ACCEPTED MANUSCRIPT
redistribution and increased local concentration of antigens within the cell membrane caused by rituximab or artificially by the secondary antibody, we can conclude that the more antigens that are available for the lipid rafts, the
RI PT
more efficient will be the effect indirectly inducing apoptosis.
Nevertheless, it has been demonstrated that only in the presence of a
complement does in vitro lysis of CLL cells by rituximab correlate with a
SC
quantitatively determined level of CD20 expression (17). The effect of the
M AN U
CDC is again related to the translocation of the CD20 and rituximab complex into lipid rafts (18). Rituximab binds to the large extracellular loop of CD20. The second-generation anti-CD20 antibody ofatumumab, which has higher CDC activity than does rituximab, binds to the small loop closer to the cell membrane (19). Alemtuzumab mediates CDC lysis of CLL cells much more
TE D
efficiently than does rituximab, probably due to the much greater expression of CD52(20).
EP
The main mechanism mediating ADCC is probably the dominant cytotoxic effect of NK cells. The ADCC effect of rituximab does not differ among cell
AC C
lines, despite differences in CD20’s expression (21). Neither the level of CD52 expression is proven to be related to cell lysis via the ADCC mechanism after binding of alemtuzumab (20).
PATIENTS AND METHODS:
7
ACCEPTED MANUSCRIPT
Patients The study group consists of 95 patients diagnosed with CLL and/or small lymphocytic lymphoma (SLL). Patients were divided into four groups: A)
RI PT
CLL patients newly diagnosed or so far untreated, B) CLL patients minimal
residual disease (MRD) -positive after treatment in complete (CR) or partial (PR) remission (22), C) pretreated patients with CLL in progression, and D)
SC
patients diagnosed with SLL. Samples of peripheral blood were collected
M AN U
from patients in each group. An overview of the patients is given in Table 1.
Sample preparation
The sample of fresh peripheral blood containing 2×105 leukocytes, whose volume in µl we obtained from the WBC count (x = 2 × 105/WBC count), was
TE D
filled to the volume of 100 µl using phosphate buffered saline. The total volume of 100 µl for the diluted sample was chosen (to include leukopenia
EP
patients after therapy) as a constant volume and number of cells is required in order to obtain the same default reaction conditions for all the samples. A
AC C
reduction in the number of cells to 2×105 leucocytes was determined on the basis of the titration curve such that the total volume of the 100 µl sample after adding 5 µl of fluorescent labeled monoclonal antibodies against the target antigen would bring us to a saturating concentration for all cells (Fig. 1A). The following antibodies were used to label the sample: Anti-CD19 antibody conjugated with R-phycoerythrin (R-PE) and anti-CD5 antibody conjugated with PE-Cy5 (Invitrogen, Carlsbad, CA, USA) was used to identify
8
ACCEPTED MANUSCRIPT
the CLL population. Anti-CD20 and anti-CD52 antibodies conjugated with fluorescein isothiocyanate (FITC, Invitrogen) were used to determine the level of expression of the target antigens CD20 and CD52. The samples were
RI PT
incubated for 20 min at room temperature, then lysed with formic acid and osmolarity was then adjusted using a standard protocol and ImmunoPrep
SC
Reagent System reagents (Beckman-Coulter, Fullerton, CA, USA).
Quantitative fluorescence cytometry
M AN U
The samples were analyzed using a CytomicsTM FC500 (Beckman-Coulter, Fullerton, CA, USA). The system linearity was tested (Fig. 1B) (23). The gating procedure in all four groups of patients involved gating of lymphocytes on an FS/SS dot plot, an exclusion of doublets and subsequently gating the
TE D
CD5+CD19+ population. Mean fluorescence intensity (MFI) of the target antigen CD20 or CD52 was determined from a histogram. A total of 10 000 cells were measured at the CD5+CD19+ gate with the flow rate of 300–600
EP
cellss−1 at the flow chamber. QuantumTM FITC calibrated microparticles with
AC C
predetermined values of fluorescence in molecules of equivalent soluble fluorochrome (MESF) units (Bangs Laboratories, Fishers, IN, USA) were analyzed using the same protocol and the same setting as those of the sample. The calibration curve was made using Quick-Cal® software (Bangs Laboratories) and the MFI values of each peak of the calibrated values. Finally, the level of expression of the target antigen in MESF units of fluorochrome FITC was deduced from the calibration curve.
9
ACCEPTED MANUSCRIPT
Statistical analyses The nonparametric Kruskal–Wallis test was used for the four groups of
RI PT
patients. Mann–Whitney U tests were subsequently made for every possible pair combination from the four groups.
M AN U
SC
RESULTS:
Antigen CD20
The nonparametric Kruskal–Wallis test did not demonstrate inhomogeneity in distribution of the values of the quantitatively established CD20 among the four groups of patients. Also, the Mann–Whitney U test showed no difference
TE D
in the distribution of quantitatively determined CD20 values between any possible pair from the four groups of patients. Most different from one another
EP
were the groups A (patients with newly diagnosed or untreated CLL) and C (pretreated patients with CLL in progression), but this difference was not
AC C
statistically significant (P< 0.076) (Fig. 2, Table 2). In the group B of patients in CR / PR at least one month elapsed since the end of therapy that included anti-CD20 monoclonal antibody (in 7 of them).
CD52 Antigen For the values of quantitatively determined CD52, the Kruskal–Wallis test did demonstrate inhomogeneity in distribution of the four monitored groups
10
ACCEPTED MANUSCRIPT
(P< 0.001). Subsequent analysis of the individual groups by Mann–Whitney U test showed there to be a significantly higher value of quantitatively determined CD52 antigen (compared to each of the remaining groups) in
RI PT
group B, which were CLL patients MRD-positive after treatment in complete
or partial remission (P< 0.001) (Fig. 3, Table 2). An overview of this group of
patients is shown in Table 3, and flow cytometric analyses of peripheral blood
SC
from one of the patients are shown in Fig. 4. There was no significant
M AN U
correlation between % residual CLL population and MESF values (Fig. 5). Four of 21 patients in this group received alemtuzumab during the last therapy, from which at least six months passed. In two of these patients, 6 and 12 months after the end of therapy, the measured values were 451 758 and 460 946 MESF-FITC for CD52 antigen. The remaining two patients, 17 and 33
TE D
months after the last treatment, had higher quantitatively determined values of CD52 antigen expression – 904 861 and 1 569 584 MESF-FITC. All the
EP
patients in this group exhibited increasing proportion of the CD5+CD19+ cell population in the gate of lymphocytes and showed progression of CLL in 3
AC C
years after collection of the peripheral blood sample for the quantitative fluorescence cytometry (data not shown). Mann-Whitney U test showed a difference between groups A (patients with
newly diagnosed or untreated CLL) and D (SLL) (P
