Defining Leukaemia Associated Phenotypes in newly diagnosed Acute Myeloid Leukaemia on a Beckman Dickinson flow cytometer

Transcription

1 Defining Leukaemia Associated Phenotypes in newly diagnosed Acute Myeloid Leukaemia on a Beckman Dickinson flow cytometer Purpose This procedure gives instruction to define leukaemia associated phenotypes (LAPs) in newly diagnosed acute myeloid leukaemia (AML). Background About half of the patients suffering from AML relapses within 12 months after achieving complete remission due to the presence and outgrowth of residual AML cells in bone marrow (BM), so called minimal residual disease (MRD). Several groups have shown that the frequency of residual cells detected by immunophenotypic methods by LAP technology is highly predictive for the occurrence of relapse. LAPs are established in newly diagnosed AML using a two steps immunophenotypic labelling procedure: in a first run the expression of several markers is measured using a screenings panel of monoclonal antibodies (moabs). LAPs are defined based on these marker expressions and are immediately verified in a second run. Policy All specimens for immunophenotyping should be processed preferentially within 24 hours after collection. Specimen anticoagulated (preferably with heparin) BM or peripheral blood (PB) of AML patients (>20% blasts in BM as established with morphologic techniques) Equipment and materials - Flow cytometer with at least 6 parameters: 2 light scatter and at least 4 fluorescence parameters. - Vortex mixer - Adjustable pipettes (e.g. 10 μl, 20 μl 100 μl, 1 ml, 2 ml) - Disposable pipette tips - Disposable polypropylene tubes - A large scala of different moabs which are conjugated in several fluorescent labels February 17, 2006 page 1 of 18

2 Abbreviations BM BSA LAP Moab MRD PB PBS PFA RT WBC bone marrow bovine serum albumin leukaemia associated phenotype monoclonal antibody minimal residual disease peripheral blood phosphate-buffered saline paraformaldehyde room temperature white blood cell Reagents 1. Ammonium Chloride Lysing Solution (10x): 82.6 g (1.5 M) Ammonium Chloride (NH4CL); Sigma A g (100 mm) Potassium Bicarbonate (KHCO3); Sigma P mg (0.96 mm) Tetra Sodium EDTA (Na4EDTA); Sigma ED4SS 1000ml Distilled H2O Dissolve these ingredients in 900 ml of distilled water. Adjust the ph by addition of either 1N HCL or 1N NaOH to a range of Add distilled water to a volume of 1000 ml. Store 6 months at 2-8 ºC. To prepare the working solution, dilute the concentrate 1:10 in distilled water. 2. Phosphate-buffered Saline (PBS). 3. PBS-0.1% BSA; PBS containing 0.1% Bovine Serum Albumin (BSA) or Fetal Bovine Serum or Human Serum Albumin. 4. PBS-1% PFA; PBS containing 1% Paraformaldehyde (PFA). 5. Moabs in working dilution, defined by titration. Procedure 1 Bulk lysis of cells 1.1 Resuspend the anticoagulated BM (or PB) sample to homogeneity 1.2 Optionally, homogenise BM by passing the sample twice through a 25 µm gauge syringe 1.3 Perform a white blood cell (WBC) count 1.4 Pipet the needed amount of cells in a 15 or 50 ml tube* 1.5 Lyse red blood cells by adding ammonium chloride working solution to the cells. Volume of ammonium chloride working solution has to be 15 times higher at minimum than the volume of the cell suspension. 1.6 Mix gently 1.7 Incubate for 15 min (or 10 min when shaking) at room temperature (RT) protected from light 1.8 Centrifuge for 10 min 800g at RT February 17, 2006 page 2 of 18

3 1.9 Discard the supernatant (e.g with a vacuum pump or Pasteur pipette) 1.10 Resuspend the cell pellet and add 10 ml PBS-0.1% BSA 1.11 Centrifuge for 5-10 min 500g at RT 1.12 Discard the supernatant (e.g. with a vacuum pump or Pasteur pipette) 1.13 Resuspend the cell pellet and add PBS-0.1% BSA (maximal cell concentration: 20x10 6 /ml). * when starting at this point with 30x10 6 cells and provided that you resuspend the cells in 1.5 ml PBS-0.1% BSA at point 1.13 you will end up with a cell suspension of about 10x10 6 /ml (1.5 ml cell suspension is sufficient for 15 staining procedures). This type of handling is necessary because of half of the cells will by lost by a lysing procedure. 2 Staining and analysing marker expression in a first run 2.1 Add 10 µl moabs conjugated with FITC, PE, PERCP and APC, or PBS in FACS tubes 1-6 according to the scheme in appendix Add 100 µl of the lysed cell suspension with a concentration of maximal 20x10 6 /ml to the tubes 2.3 Mix gently. 2.4 Incubate for 15 min (or 10 min when shaking) at RT protected from light. 2.5 Resuspend the cell pellet and add PBS-0.1% BSA. 2.6 Mix gently. 2.7 Centrifuge for 5 min 500g at RT. 2.8 Discard the supernatant (e.g. with a vacuum pump or Pasteur pipette). 2.9 Resuspend the cell pellet and add PBS-0.1% BSA Acquire for each tube 20,000-50,000 WBC events (characterised by CD45 expression and FSC, see Figure A in appendix 2) on flow cytometer Analyse the marker expressions on the leukaemic blast population(s) (See point 4. Data analysis) When flow cytrometric acquisition can not be performed on the same day, PBS-1% PFA should be added to the cell pellet at point 2.9. Samples can be stored at 2-8 C for 24 hours at maximum. February 17, 2006 page 3 of 18

4 3 Defining LAPs and verifying these in a second run 3.1 Define LAPs based on the marker expressions as measured and analysed in the first run according to the instructions in appendix 3. At least 2 LAPs with different aberrant marker expressions should be defined. If only one aberrant marker is present, make a second LAP with the same aberrant marker but with a different myeloid or primitive marker. 3.2 Verify the putative LAPs in a second run by pipetting the moabs which make up the LAP in a FACS tube. This incubation can be done with 4x10µl moabs, also other volumes can be used (minimum volume: 5 µl) with adjusted sample volumes. Use the moabs distributed by the MRD project group and/or own moabs, 3.3 Add the lysed cells 3.4 Mix gently 3.5 Incubate for 15 min at RT (or 10 min when shaking) protected from light. 3.6 Resuspend the cell pellet and add PBS-0.1% BSA 3.7 Mix gently 3.8 Centrifuge for 5 min 500g at RT 3.9 Discard the supernatant (e.g. with a vacuum pump or Pasteur pipette) 3.10 Resuspend the cell pellet and add PBS-0.1% BSA 3.11 Acquire 20,000-50,000 WBC events (characterised by CD45 expression and FSC, see Figure A in appendix 3) on flow cytometer When flow cytrometric acquisition can not be performed on the same day, then add PBS-1% PFA on the cell pellet at point Samples can be stored at 2-8 C for a 24 hours at maximum. 4 Data analysis Data analysis can be performed with Cellquest, Paint-a gate, FlowJo or other software. Cellquest software has been used to explain the gating strategy both for the percentage of different cell populations (immature and mature blasts, blasts with aberrant CD45 expression, lymphocytes, monocytes and granulocytes) and for marker expression in AML blasts. 4.1 Analyses of different cell populations The gating steps as described below are illustrated in appendix Load the cells in density plot CD45-SSC Gate the WBC compartment based on CD45 expression and FSC (region R1 in Figure A) Load the WBC (R1) in density plot SSC-FSC. If necessary dead cells and cell fragments can be further excluded by an extra gate (R2 in Figure B) in this plot Load the WBC in a CD45-SSC density plot (Figure C and F) Gate in density plot CD45-SSC if present the: 1 Immature leukaemic blasts (characterized by CD45dim-SSClow), 2 Mature leukaemic cells (characterized by CD45dim-SSCintermediate), 3 CD45 aberrant leukaemic cells (characterized by CD45aberrant-SSClow), 4 Lymphocytes (characterized by CD45high-SSClow), February 17, 2006 page 4 of 18

5 5 Monocytes (characterized by CD45high-SSCintermediate), 6 Granulocytes (characterized by CD45low-SSChigh), These different cell populations are illustrated in appendix 2. ad 1: ad 2: ad 5: Fine tune the region on immature blast population by back gating those blasts showing expression of a primitive marker (CD34, CD117, CD133). (See Figure E and H in which the CD34+ and CD133+ cells in R7 are back gated in Figure C and F). Mature blasts can be distinguished from monocytes by the presence of an aberrant phenotype or lack of CD14 expression (See Figure C -D and Figure F-G). Fine tune the gate on the monocytes by back gating those cells showing CD14 expression (see Figure F-G). 4.2 After 1 st run: Analyses of marker expressions on the leukaemia blast population Establish in all defined leukaemia blast gates (immature, mature and aberrant, depending of their presence) the marker expressions of the labelled cells. The intensity of marker expression (no, dim or positive expression) has to be determined by comparing the expression of these markers with the autofluorescence of the unlabelled cells (tube 1) Only markers expressed for >10% on the total leukaemia blast population are used in a LAP Check how the leukaemia cell population showing LAP expression is located in the FSC-SSC and CD45-SSC by backgating. NB: These back gating steps are necessary: 1 To recognize apoptotic cells (these cells are located low in the FSC scale, see appendix 4). These cells have to be ignored. 2 To establish the maturity and/or type of these cells by i. assessing the location of the cells in the CD45-SSC. When cells are located outside the area of the blasts region these cells belong to another cell population (see appendix 5), ii. assessing the expression of other immature/mature markers on cells which were simultaneously labelled in the same tube. With this approach it can be checked whether the marker expression of cells of interest originated from another (leukaemia) cell population. 4.3 After a second run: Establishment of a LAP The gating steps as described below are illustrated in appendix 6. February 17, 2006 page 5 of 18

6 4.3.1 Analyse whether the combination of marker expressions on the leukaemia blast population forms a reliable LAP Check in the FSC-SSC and CD45-SSC by back gating the localization of leukaemia cell population that express the LAP Only LAPs with minimal expression >10% on the leukaemia blast population are used for MRD monitoring. When the LAP includes the presence of an aberrant marker combined with a primitive marker (CD34, CD117, CD133), the percentage of the aberrant marker on blasts showing the primitive marker has to exceed by 20% expression (normal blasts may also show 5%-10% aberrant marker expression). 5 Data reporting and interpretation 5.1 Note on the LAP REPORT form (see appendix 7) if present: - The percentage of immature leukaemia blasts, mature leukaemia blasts, CD45 aberrant leukaemic blasts, lymphocyte, monocytes and granulocytes - The aberrant marker and its percentage of the blasts compartment which determines the LAP, the LAP and its percentage of the blast compartment - The sensitivity of the LAPs. (The specificity and stability of the LAPs can be established only by the centre with an extensive experience in MRD measurements). The total of these parameters determines the quality of the LAP for the detection of MRD in follow-up samples. On the worksheet Commentary (see appendix 7), present at the LAP REPORT form, a guide for these quality parameters is given. February 17, 2006 page 6 of 18

7 Appendix 1: Screenings panel Tube FITC conjugated Moab PE conjugated Moab PercP conjugated Moab APC conjugated Moab 1 PBS PBS CD45 PBS 2 CD34 CD22 CD45 CD117 3 CD15 CD13 CD45 CD14 4 HLADR CD33 CD45 CD11b 5 CD2 CD56 CD45 CD7 6 CD36 CD133 CD45 CD19 Clones of these Moabs can be found on the MRD website at clones of monoclonal antibodies. Before use Moabs have to be titrated and diluted in the working solution. February 17, 2006 page 7 of 18

8 Appendix 2: Analysis of different cell populations A No gate R1 B R2 In R1 + D In R1 and R2 and R5 C In R1 and R2 R8 R3: immature blasts R4: lymphocytes R5: mature blasts (CD14-) R6: granulocytes R4 R5 R6 R5 R3 R3 E In R1 and R2 and R3 R7 In R1 and R2 G In R1 and R2 and R5 R3: immature blasts R4: lymphocytes R5: monocytes (CD14+) R6: granulocytes F R4 R5 R6 R5 H R8 In R1 and R2 and R3 R3 R3 R7 February 17, 2006 page 8 of 18

9 Examples of the analysis of different cell populations of two different AML samples (sample 1: A-E, sample 2: F-H). Cell populations are analysed within the white blood cell (WBC) compartment, characterized by CD34/FSC (gate R1 in Fig A). Dead cells and cell fragments can be further excluded in a SSC/FSC plot (gate R2 in Fig B). Different cell types can be identified in a CD45/SSC plot (Fig C and F). R3: Immature leukaemic blasts (characterized by CD45dim-SSClow). The immature leukaemic blasts gate can be fine tuned by back gating the cells showing a primitive marker (Fig E and H). R4: Lymphocytes (characterized by CD45high-SSClow). R5: Mature leukaemic blasts or monocytes (characterized by CD45dim-SSCintermediate) - In Fig C cells are identified as mature leukaemic blasts based on the lack of CD14 expression (Fig D), - In Fig F cells are identified as monocytes based on the expression of CD14 The monocytes gate can be fine tuned by back gating the cells with CD14 expression (Fig G). R6: Granulocytes (characterized by CD45low-SSChigh). February 17, 2006 page 9 of 18

10 Appendix 3: Guidelines for defining LAPs Theoretical considerations The quality of a LAP for MRD detection depends on its 1) Specificity. The specificity depends on the percentage of LAP expression on normal cells. A high specificity can be achieved by including primitive markers (CD34, CD133 and CD117) if present on AML. The specificity of LAPs further depends on many other factors. In most cases LAP expression on normal cells is <0.1%. Below, a guideline for the design of LAP with an optimal specificity is given. 2) Sensitivity. The sensitivity of MRD detection depends among others on the percentage of LAP expression on the leukaemic blast population at diagnosis and the number of cells analyzed. For this reason only LAPs which are expressed >10% on the leukaemic blast population are considered. 3) Stability LAPs may undergo phenotypic shifts. During the disease, marker expression on AML may disappear, resulting in false-negativity. Especially dim expression of markers is susceptible for changes. Furthermore, CD19 expression has been shown to disappear on several occasions in the course of disease. Taking into account the quantity of BM material commonly available, immunophenotypic MRD detection methods enable detection of 1 MRD cell in 10,000 normal cells when using LAPs with high specificity, high sensitivity and high stability. I LAPs with primitive markers LAPs that include an aberrant marker expression, combined with a primitive marker (PM) and a myeloid marker (MM) usually have a high specificity. PM are CD34, CD133 and CD117. MM are CD13, CD33, CD117 and CD133. Markers that define a LAP when combined with a PM and MM are: CD2 CD11b CD14 CD15 CD19 CD22 CD36 February 17, 2006 page 10 of 18

11 CD56 CD13- CD33- HLADR- (only when combined with HLADR+ or CD33+) (only when combined with HLADR+ or CD13+) (only when combined with CD13+ or CD33+, [in this case CD117 as PM is less specific compared to CD34 or CD133; use the latter ones]) Which primitive marker should be used in a LAP? Only a PM of which the fluorescence of the conjugate exceeds the 2 nd log quadrant is appropriate. The higher the fluorescence the better the discrimination between normal and leukaemic cells. The fluorescence of the conjugated PM is highest when using PE label. In cases with equal intensities of expressions, use the PM with the highest percentage of expression. In cases with similar intensities and percentages of marker expressions, CD34 is the most reliable primitive marker: CD133 expression is often dim and CD117 may be down-regulated in peripheral blood stem cell products. Which myeloid marker should be used in a LAP? Preferably, use a MM that is most discriminative from normal cells. Be alert on - Differences in percentages between normal and leukaemic cells: e.g. normal CD34+ cells show about 50% CD133 expression, while normal CD34+ cells show >90% CD13 and CD33 expression: CD133 is thus most discriminative when AML blasts show 100% expression of these myeloid markers. - Differences in intensities of MM expression between normal and leukaemic cells. However, when the LAP includes a lack of a marker expression (CD13, CD33, HLADR) then use a MM marker which is expressed in high percentages on normal cells. - Markers expressed in high percentages on normal CD34 or CD133 positive cells are CD13, CD33 and HLADR. On normal CD117 positive cells these markers are CD13 and CD33 (normal CD117 positive cells are partly HLADR negative: promyelocytes). - The marker that is not expressed should be measured preferably in FL4: In this channel the autofluorescence of cells is very low. - The MM that is expressed should have a high fluorescence signal (beyond the 2 nd log quadrant). Use PE as conjugate for the MM antibody if necessary. Exclusion marker instead of myeloid marker? When a PM has a staining intensity that does not exceed the 2 nd log quadrant, even when used with the strongest fluorescence label, it sometimes makes sense to replace the MM by an exclusion marker to exclude monocytes and/or granulocytes. Of course, the exclusion marker has to be absent (or in low percentages) on the leukaemic cells. Exclusion markers are: CD36 (exclusion of monocytes), CD14 (exclusion of monocytes), CD15 (exclusion of granulocytes). February 17, 2006 page 11 of 18

12 II LAPs without primitive markers The specificity of these LAPs is less compared to LAPs in which a PM is included. In most cases LAPs can be defined by the absence of a MM. Most frequently defined LAPs without PM are: CD15+ CD13- CD45 CD14+ CD15- CD33+ CD45 CD14+ CD33+ CD56+ CD45 CD14- DR- CD11b CD45 CD14+ DR- CD56+ CD45 CD33+ February 17, 2006 page 12 of 18

13 Appendix 4: Apoptotic / viable status of cells A CD45ECD B C D CD45ECD Viable cells CD117CY5 E F G CD45ECD Apoptotic cells AML blasts are identified by CD45/SSC (Fig A). The CD34+CD117+ AML blasts (Fig B) are located in the centre of the CD45/SSC gate (Fig C) and have a normal FSC/SSC pattern (Fig D). However, the CD56+ AML blasts (Fig E), which are also located in the centre of the CD45/FSC gate, have a low FSC (Fig G). These CD56+ cells thus are apoptotic cells and CD56 can not be used as an aberrant marker in a LAP. February 17, 2006 page 13 of 18

14 Appendix 5: Differentiation status of cells A B C D E F G AML blasts are identified by CD45/SSC (Fig A). The AML blasts show CD34 expression (Fig B). The CD34+ cells are located in the centre of the CD45 gate (Fig C) and have a normal FSC/SSC pattern (Fig D). However, the AML blasts with CD14 expression (Fig E) are located in the outer range of the CD45 gate (Fig F). These cells are located somewhat lower in the FSC (Fig G). In conclusion, these CD14+ cells are originated form another cell population (monocytes). February 17, 2006 page 14 of 18

15 Appendix 6: Establishment of a LAP Defining of AML blasts in the WBC compartment A B C R2 R1 R3 No gate In R1 In R1+R2 LAP expression on AML blasts D E F R4 R5 R6 In R1+R2+R3 In R1+R2+R3 In R1+R2+R3+R4+R5 Backgating steps of cells showing CD34+CD33+CD45dimCD2+ expression G H I In R1+R2+R3+R4+R5+R6 In R1+R2+R3+R4+R5+R6 In R1+R2+R3+R4+R5+R6 February 17, 2006 page 15 of 18

16 Fig A, B, C: Defining of AML blasts in the WBC compartment A: WBC compartment is gated based on CD45 expression and FSC (R1). B: Cell fragments and/or dead cells can be further excluded in SSC/FSC plot (R2). C: AML blast compartment can be defined in a CD45/SSC plot (R3). Fig D, E, F: LAP expression on AML blasts D: Part of the AML blasts show CD34 expression (R4). E: These CD34+ cells show a dim CD45 expression. F: The CD34+ cells which show CD45dim expression show CD33 and CD2 expression, which defines the LAP Fig G, H, I: Back gating steps of AML blasts with LAP expression These steps are necessary to fine tune the characteristics of the LAP positive cells. G: Back gating of LAP+ cells in FSC/SSC: few cells with FSC are excluded. H: Back gating of LAP+ cells in CD45/SSC. I: Back gating of LAP+ cells in CD34/SSC. February 17, 2006 page 16 of 18

17 Appendix 7: LAP report form LAP REPORT form number registration number measured by centrum judged by centrum AML Populations (see commentory 1) % remarks Immature leukaemic blasts Mature leumaemic blasts Leukaemic blasts (CD45aberrant) Lymfocytes Monocytes Granulocytes LAPS (see commentory 2) Judgement of the quality of LAPs LAPs aberrancy % aberrancy LAP % LAP specificity sensitivity stability quality number of aberrancies Remarks February 17, 2006 page 17 of 18

18 Commentary 1 Populations Percentages of blasts, lymfocytes, monocytes and granulocytes of the white blood cell compartment are established by using CD45/FSC and CD45/SSC characteristics Discrimination between monocytes and mature leukaemic blasts on the basis of CD14 expression 2 LAPS aberrancy: Note the aberrant marker or combination of markers % aberrancy Note the % of the aberrant marker expression (or combinations) LAP Note only the intensity of marker expression when the marker showed no (-), dim (dim) or over (++) expression Note LAPs in order of colour (FITC, PE, PercP, APC) Define only LAPs if the combination of aberrant markers is expressed on >10% of the blasts % LAP Note the LAP as percentage of the blasts. specificity sensitivity stability quality is dependent on the LAP expression on normal cells 1 high specificity (<0.1% normal expression) 2 low specificity ( >0.1% normal expression) 0 unknown Specificity is estimated on the basis of the own experience with the LAP is dependent of the size of the population with aberrant marker expression (as % of the blasts) 1 high sensitivitity (LAP: >50%) 2 intermediate sensitivity (LAP 20-50%) 3 low sensitivity ( LAP: 10-20%) offers a guide about the sensitivity of a LAP for a phenotypic shift 1 stable LAP 2 unstable LAP 0 unknown offers the reliability of a LAP for MRD monitoring quality specificity sensitivity stability 1 excellent good moderate sufficient poor unknown when specificity or stability is unknown number of aberrancies: note the number of independent aberrancies February 17, 2006 page 18 of 18