Dendritic cells Dendritic cell recovery after autologous stem ... - Nature

Bone Marrow Transplantation (2002) 30, 261–266  2002 Nature Publishing Group All rights reserved 0268–3369/02 $25.00 www.nature.com/bmt

Dendritic cells Dendritic cell recovery after autologous stem cell transplantation D Damiani1, R Stocchi1, P Masolini1, A Michelutti1, A Sperotto1, A Geromin1, C Skert1, M Cerno1, M Michieli3, M Baccarani2 and R Fanin1 1 Chair and Division of Haematology, Bone Marrow Transplant Unit, Department of Medical and Morphological Research, University Hospital, Udine, Italy; 2Institute of Haematology and Medical Oncology ‘L and A Sera`gnoli’, University Hospital, Bologna, Italy; and 3Centro di Riferimento Oncologico (CRO), Aviano, Italy

Summary: There is persistent immunosuppression not only in allogeneic but also in autologous stem cell transplantation because humoral and cellular immunity may take a year or more to return to normal, with increased risk of infectious complications. This immune defect may also involve antigen presentation, in particular dendritic cell (DC) function. We evaluated DC subset reconstitution in 58 patients who underwent bone marrow (BM) or peripheral blood (PB) autologous haematopoietic stem cell transplantation (HSCT). In all patients DC type 1 (DC1) and DC type 2 (DC2) were already significantly lower than in normal individuals before conditioning therapy (DC1/␮l 3.1 ⴞ 1.0, DC2/␮l 3.0 ⴞ 1.1). On day 0 and day ⴙ7 the mean DC1 and DC2 numbers were very low in both groups. Patients who received unmanipulated marrow or peripheral blood stem cells reached pre-conditioning levels of DC1 and DC2 cells on day ⴙ20. In patients receiving selected CD34 cells, DC increased slowly and pre-transplant counts were observed only on day ⴙ60. Nearly ‘normal’ levels of DC1 and DC2 could be observed in the first group from day ⴙ180, and were maintained thereafter; in CD34⫹ selected patients DC1 and DC2 counts remained lower than normal. Our data emphasise that circulating antigen presenting cells (APC) recover quickly. It remains to be determined if DC frequency in PB reflects their tissue function. The relatively low incidence of infections in patients undergoing autologous transplantation, despite defective lymphocyte reconstitution, could be related to functionally efficient DC. Bone Marrow Transplantation (2002) 30, 261–266. doi:10.1038/sj.bmt.1703637 Keywords: dendritic cell; autologous stem cell transplantation; immunological recovery

High-dose chemotherapy followed by haemopoietic stem cell transplantation (HSCT) is increasingly used for various

haematologic malignancies and solid tumours. It is now well-recognised that there is persistent immunosuppression not only in allogeneic but also in autologous stem cell transplantation because humoral and cellular immunity may take a year or more to return to normal, with increased risk of infectious complications.1 Functional reconstitution of the lymphoid compartment after autologous bone marrow/ peripheral blood HSCT depends either on the presence of adequate numbers of antigen-specific T and B cells in the transplant or the maturation of lymphoid progenitors and perhaps multipotent stem cells into antigen-specific T and B cells within thymic and bone marrow microenvironments. The early post-transplant period is characterised by a reduction of CD4+ T cells (due to a persistently low level of naive CD4+/CD45RA+ T cells) and by elevated numbers of CD8+ T cells. The number of B cells is also reduced.2–7 This immune defect may also involve antigen presentation. On this subject, it has been suggested that the patients’ antigen presenting cells (APC) are abnormal either numerically or functionally after transplant.8 Dendritic cells (DC) are the most powerful APC in inducing T cell activation. They originate from pluripotent stem cells in the bone marrow; then, they migrate from peripheral blood to tissues, where they achieve the capacity to capture and process antigens. In lymphoid organs DC present these processed antigens to and activate T cells.9 Recently, two subsets of DC have been identified in peripheral blood. DC type 1 (lineage negative, HLA-DR positive and CD11c positive) are myeloid and activate (through interaction between CD80 and CD28, in the presence of IL-12) T helper type 1 (Th1), which produce IL-1 and IFN␥ (pro-inflammatory cytokines); DC type 2 (lineage negative, HLA-DR positive and CD123 positive) may be lymphoid and activate (through interaction between CD86 and CD28, in the presence of IL-10) T helper type 2 (Th2), which produce IL-4 and IL-10 (anti-inflammatory cytokines).8,10–12 In this paper we evaluated the kinetics of lymphocyte and dendritic cell subset reconstitution in 58 patients undergoing bone marrow or peripheral blood autologous HSCT and its relationship with infectious complications. Patients and methods

Correspondence: Dr D Damiani, Division of Haematology, University Hospital, P le S Maria della Misericordia, 33100 Udine, Italy Received 26 September 2001; accepted 1 May 2002

The kinetics of lymphocyte and dendritic cell reconstitution were studied in 58 consecutive patients affected by NHL

Dendritic cell recovery D Damiani et al

262

(n ⫽ 28), HL (n ⫽ 6), multiple myeloma (n ⫽ 15), acute leukaemias (n ⫽ 4) and breast cancer (n ⫽ 5) who received intensification therapy and autologous stem cell rescue (HSCT) from January 1997 to December 2000. Patients with NHL received the BAVC (cytosine-arabinoside, etoposide, cyclophosphamide and carmustine) conditioning regimen, patients with HL received the BEAM (cytosinearabinoside, etoposide, melphalan and carmustine) conditioning regimen, myeloma patients received busulphan and melphalan, patients with acute leukemia received busulphan and cyclophosphamide and breast cancer patients received etoposide, carboplatinum and iphosphamide. A median of 2.6 ⫻ 106/kg (range 0.77–4) CD34⫹ cells were infused in peripheral blood HSCT (n ⫽ 46) and 1.07 ⫻ 106/kg (range 0.4–1.87) in bone marrow HSCT (n ⫽ 12). Twelve of 15 myeloma patients received highly purified CD34⫹ stem cells (median 2.8 ⫻ 106/kg, range 2–5), obtained after immumomagnetic separation by CliniMacs (Miltenyi Biotec, Bergisch Gladbach, Germany). Antibiotic prophylaxis included ciprofloxacin and itraconazole. GCSF (filgrastim, Neupogen, Amgen, Thousand Oaks, CA, USA) at a dose of 5 ␮g/kg was administered from day ⫹4, until the leukocyte count exceeded 2 ⫻ 109/l for 3 consecutive days. Patients’ characteristics are summarised in Table 1. Dendritic cell reconstitution Dendritic cell recovery was evaluated before starting the conditioning regimen, at day 0 before stem cell infusion, at day ⫹7 and at the same points as lymphocytes. DC1 (CD11c⫹, myeloid origin) and DC2 (CD123⫹, lymphoid origin) subsets were identified by using a three-colour flow cytometric assay on lysed whole blood to minimise selective loss. For each test 100 ␮l of blood were incubated with 10 ␮l of the HLA-DR-PerCP (BD), with 20 ␮l of a mixture of lineage-related antibodies FITC (lineage cocktail 1 FITC, BD), including the following monoclonal antibodies: CD3, CD14, CD16, CD19, CD20 and CD56, and with 10 ␮l of the CD11c-PE or the CD123-PE antibodies. At the end of a 15 min incubation, red cells were lysed as described above, samples were washed twice and immediately analysed. A minimum of 50 000 events was acquired for each experiment. Dendritic cells express high levels of HLA-DR and lack lineage-related antigens. After gating lineage-negative events, the two DC1 and DC2 subsets were identified in the high HLA-DR expressing population, on the basis of their high CD11c or CD123 intensity. Nega-

tive controls, with irrelevant isotypic antibodies were prepared in each experiment, as appropriate. The absolute number of dendritic cells was calculated from the WBC count multiplied by the proportion of each subpopulation among the WBC, as determined by flow cytometric analysis. Data were expressed as DC1 and DC2 mean number ⫾ 2 standard deviation per microliter of PB. Immune reconstitution Peripheral blood lymphocyte subsets were analysed by flow cytometry before starting the conditioning regimen, at time 0, before stem cell infusion, at time ⫹30, ⫹60, ⫹90, ⫹240, ⫹360, ⫹3 years. One hundred ␮l of peripheral blood anticoagulated with EDTA were incubated at laboratory temperature for 20 min with the following monoclonal antibodies: CD3FITC, CD19PE, CD4PE, CD8FITC, CD16PE, CD56PE, CD45ROFITC, CD45RAFITC. At the end of incubation red cells were lysed by Facs Lysis solution (BD, Bruxelles, Belgium), washed twice and analysed within 1 h. Acquisition and analysis were performed with a FacsCalibur (BD) flow cytometer with Lysis II software. WBC cell counts were determined using an automated cell counter (CELL DYN 3200). Total lymphocyte count was determined by flow cytometry by incubating a sample of whole blood with the CD14PE/CD45FITC (BD) antibodies. The absolute number of lymphocytes was calculated by multiplying the percentage of CD14−/CD45⫹⫹ by the total WBC. The absolute number of cells in any given lymphocyte population was calculated by multiplying the percentage of positive cells for each lymphocyte marker by the absolute lymphocyte number. Data were expressed as mean number ⫾ 2 standard deviations per microliter of PB. Normal controls Circulating dendritic cell and lymphocyte subsets of 15 healthy donors were evaluated as described above, to obtain normal reference values. Data analysis Mean differences at different times between patients and normal controls were assessed by the Mann–Whitney U test. Results

Table 1

Patient characteristics

Patients Sex (M/F) Median age Disease

Bone Marrow Transplantation

Not selected

Selected

46 28/18 40 (16–58) 28/46 NHL 6/46 HL 3/46 MM 4/46 AL 5/46 BC

12 7/5 55 (41–63) 12/12 MM

Transplant-related data are summarised in Tables 2 and 3. Haematological recovery occurred promptly in all patients, without any differences between patients receiving bone marrow or peripheral blood stem cells. The median time to reach a neutrophil count >1 ⫻ 109/l was day ⫹10 (9–14) in PB patients, day ⫹13 (11–15) in PB selected patients and day ⫹12 (10–30) in BM patients. The median time to reach a platelet count >20 ⫻ 109/l was day ⫹11 (9–30), day ⫹14 (10–21) and day ⫹12 (10–30), respectively. Not considering CD34⫹ selected patients 23/46 (50%) had a febrile episode; only in 8/46 (17.3%) was a gram ⫹ bacter-


Table 2

PB not selected (n ⫽ 48)

PB selected (n ⫽ 12)

2.6 (0.77–4) ⫹10 (9–14) ⫹11 (9–30) 7 (6–9) 19 (16–20) 3 (2–5) 4 (3–7)

2.8 (2–5) ⫹13 (11–15) ⫹14 (10–21) 10 (8–12) 21 (18–24) 4 (2–6) 5 (4–9)

CD34⫹ ⫻ 106/kg Neutrophils >1 ⫻ 109/l Platelets >20 ⫻ 109/l G-CSF (days) In patients (days) Red cells unit Platelet apheresis

Table 3

263

Graft composition and transplant-related data according to stem cell source BM (n ⫽ 12)

1.07 (0.4–1.87) ⫹12 (10–30) ⫹12 (10–30) 8 (7–13) 23 (19–26) 4 (2–7) 6 (5–8)

10

Infectious complications after transplant

9

Selected (%)

Not selected (%)

8 DC2 normal value

7

5/12 5/12 2/12 4/12

(41.6) (41.6) (16.6) (33.3)

23/46 (50) 15/46 (32.6) 8/46 (17.3) 16/46 (34)

DC1 normal value

6 DC/ml

None FUO Bacteraemia Stomatitis (III–IV WHO)

5

DC1

4

DC2

3

aemia detected. A stomatitis grade III–IV, was the most frequent toxic side-effect (16/46 patients, 34.7%). In CD34⫹ selected patients 5/12 (41.6%) had a febrile episode; in 2/12 (16.6 %) a gram ⫹ bacteraemia was detected. A stomatitis grade III–IV, was the most frequent toxic sideeffect (4/12 patients, 33.3%). No deaths occurred during the early post-transplant period (follow-up of 12 months). We did not test T helper 1 and T helper 2 cell functions on the basis of DC phenotype; we tested DC1 and DC2 in patients who had infection and we found, as expected, low levels of both.

2 1 0

BL

0

7

30 Time (days)

60

90

180

Figure 1 DC1 and DC2 recovery in selected patients. Each time point is expressed as mean value ⫾ 2 standard deviations. BL indicates day before starting conditioning therapy; day 0 indicates day of stem cell infusion (before infusion).

10 9

Dendritic cell reconstitution

8 DC1

DC2 normal value

7

DC1 normal value

6 DC/m l

In all patients DC1 and DC2 cells were already significantly lower than in normal individuals before conditioning therapy (DC1 mean/␮l 3.1 ⫾ 1.0 vs mean 6.0 ⫾ 3.3 P ⫽ 0.025, DC2 mean 3.0 ⫾ 1.1 vs 6.7 ⫾ 3.5 P ⫽ 0.025). On day 0 before stem cell infusion, and on day ⫹7 the mean DC1 and DC2 number was very low (0.03 ⫾ 0.01 and 0.25 ⫾ 0.01 on day 0; 0.06 ⫾ 0.01 and 0.04 ⫾ 0.01/␮l on day ⫹7) in both groups (selected and not selected), but the recovery kinetics were quite different. As shown in Figures 1 and 2, patients who received unmanipulated marrow or peripheral stem cells reached pre-conditioning levels of DC1 and DC2 cells at day ⫹30 (3.9 ⫾ 1.2 and 2.8 ⫾ 1.3, respectively). On the contrary, in patients receiving selected CD34⫹ cells, dendritic cells increased slowly and pre-transplant counts were observed only on day ⫹60. No difference was observed in the kinetics of DC1 and DC2 reconstitution. Nearly ‘normal’ levels of DC1 and DC2 could be observed in the first group of patients from day ⫹180 (6.3 ⫾ 2.3 and 5.8 ⫾ 2.0), and were maintained thereafter (6.3 ⫾ 3.0 and 6.8 ⫾ 3.0 at 1 year; 7.3 ⫾ 3.3 and 6.6 ⫾ 3.0 at 3 years post transplant). In CD34⫹ selected patients DC1 and DC2 counts were lower than normal up to 1 year post transplant (4.3 ⫾ 0.2 and 4.3 ⫾ 0.1, respectively).

DC2

5 4 3 2 1 0

BL

0

7

30 Time (days)

60

90

180

Figure 2 DC1 and DC2 recovery in unselected patients. Each time point is expressed as mean value ⫾ 2 standard deviations. BL indicates day before starting conditioning therapy; day 0 indicates day of stem cell infusion (before infusion).

Immune reconstitution The B cell count (defined as the number of cells expressing the pan-B CD19 antigen) and the NK cell count (defined by the CD56 expression and the lack of CD3 antigen) rose to normal levels after 3 months post transplant in all Bone Marrow Transplantation


264

patients. CD3+ cells rose to normal level after 3 years post transplant. CD8+ cells rose over the normal level reaching a peak at day ⫹60 (811 ⫾ 150/␮l vs 557 ⫾ 78/␮l) and fell to normal levels within 6 months post transplant in the unselected group, and within 1 year in the selected group. CD4+ lymphocytes remained less than 50% normal, so that all patients had a low CD4+/CD8+ ratio (