Mixed hematopoietic chimerism after allogeneic bone marrow ... - Nature

Bone Marrow Transplantation, (1997) 19, 697–702  1997 Stockton Press All rights reserved 0268–3369/97 $12.00

Mixed hematopoietic chimerism after allogeneic bone marrow transplantation: the impact of quantitative PCR analysis for prediction of relapse and graft rejection in children P Bader1, W Ho¨lle1 , T Klingebiel1, R Handgretinger1, N Benda2, PG Schlegel 1, D Niethammer1 and J Beck1 University Children’s Hospital, 1Department of Pediatric Hematology and Oncology, and 2 Department of Biomedical Statistics, Tu¨bingen, Germany

Summary: It still remains unclear whether patients with mixed hematopoietic chimerism (MC) after allogeneic bone marrow transplantation (allo-BMT) have an increased risk of developing relapse or graft failure. To address this question, we monitored the individual dynamics of chimerism after allo-BMT in pediatric patients within a prospective case control study. The individual ratio of donor to recipient peripheral white cells was determined by quantification of genomic variable number of tandem repeats (VNTRs) with a polymerase chain reaction (PCR) approach. Within the study period from 1 January 1994 until 1 July 1996 we investigated 50 sequences of 46 pediatric patients after allo-BMT (32 with malignant, 18 with nonmalignant diseases). We found complete chimerism (CC) in 34/50 cases, MC in 12/50 follow-ups and 4/50 patients revealed autologous recovery (AC). Eight of 12 patients with MC showed increasing autologous patterns and subsequently relapsed or rejected their graft, 3/12 showed decreasing amounts of recipient DNA and turned to CC upon further followup. One patient of 12 who had severe combined immunodeficiency (SCID), attained engraftment with a stable MC pattern. Three patients of 34 with CC relapsed lacking a transitional MC interval. However, the time span between last CC confirmation and relapse in each of these three patients was 6 months or longer. We suggest that these patients also developed a stage of transitional MC but that the critical timepoint of molecular confirmation by PCR was missed as time intervals in the individual follow-up of these three patients were too long (> >6 months). In summary, the results demonstrate that the individual risk of developing relapse or graft failure is significantly enhanced in the MC situation (P , 0.0005). Therefore the quantitative analysis of MC at short time intervals might be of great value to identify high risk patients which will have a significantly enhanced risk for relapse or graft rejection. Keywords: mixed hematopoietic chimerism; quantitative

PCR; minisatellite regions; allogeneic bone marrow transplantation

At present allogeneic BMT (allo-BMT) treatment regimens are considered to be of advantage in particular malignant and non-malignant diseases.1–3 The success of this treatment modality is mainly affected by relapse or graft rejection. Factors responsible for relapse and/or graft rejection may be insufficient conditioning regimens or a deficient graft-versus-leukemia (GVL) effect due to decreased amounts of donor effector cells or to their functional ineffectiveness. In order to evaluate the individual risk for relapse or graft rejection in correlation to the chimerism status after allo-BMT, various studies on this subject have been performed previously.4–7 Several studies on patients suffering from severe aplastic anemia (SAA) or chronic myelogenous leukemia (CML) gave evidence that the risk of relapse or graft failure was increased when MC was detectable after allo-BMT.8,9 However, the prognostic value to predict relapse after MC detection for patients with acute leukemias remains controversial.4,9,10 The conflicting results in the literature might be explained on the basis of the experience that relapses from acute leukemias generally develop within a shorter time period in comparison to cases with chronic leukemias. Therefore, studies with relatively long intervals in patient follow-up may have missed the critical and transient stage of MC. The hypothesis of a critical and transient stage of mixed chimerism was tested within a prospective study of patients after allo-BMT from a single institution. Sequential analysis of the individual chimerism status was performed within sufficiently shorttime intervals. The results of this study demonstrate that the individual risk of relapse or graft failure is significantly enhanced in the MC situation.

Patients and methods Patients

Correspondence: Dr P Bader, University Children’s Hospital, Department of Pediatric Hematology and Oncology, Ru¨melinstraβe 23, D-72070 Tu¨bingen, Germany Received and accepted 5 December 1996

Patients shown in this study received allo-BMT exclusively at the University Children’s Hospital, Tu¨bingen between 1 January 1994 and 1 July 1996. The clinical data are summa-

Prognostic value of mixed hematopoietic chimerism P Bader et al

698

rized in Table 1 (malignant diseases) and Table 2 (nonmalignant diseases). Each patient received unmanipulated bone marrow either from HLA-identical (30/50) or from HLA-identical unrelated donors (13/50) or from partially mismatched family donors (7/50). Clinical grading of aGVHD was performed according to Ref. 11. Methods In order to evaluate the possible dynamics of chimerism after allo-BMT we established a quantitative PCR approach for amplification of the following minisatellite regions: F13 A1, 12 TOPIP2,13 1p32,14 D11S533, 15 IL2Rb,16 D1S8017 and D17S30 18 as described recently. 19 Briefly, pretransplant recipient and donor DNA samples were amplified with each of the seven primer pairs to obtain at least one informative locus. Standardized mixed chimeric samples were generated by mixing pretransplant recipient and donor DNA in a range of percentages for each individual case. PCR analysis for the informative locus was carried out on sequential patient samples using the informative primer pair. PCR products were separated by polyacrylamide gel electrophoresis. After staining with ethidium bromide, signals were analyzed densitometrically and results were taken on the basis of individual standard curves. The sensitivity level to detect recipient between host DNA was generally in the range of 0.8%. Post-transplant DNA samples were investigated in sequences and the signal intensities were compared to the standard curves. Statistics Statistical analysis was performed using Fisher’s exact test. A P value of P , 0.05 was considered significant. Evaluation of the patients Following engraftment peripheral blood samples were investigated weekly when the leukocyte count exceeded 500/ml. After discharge of the patients, the majority of the samples were collected and analyzed monthly within an observation period of 18 months after allo-BMT, with the exception of a very few patients who could not be investigated within the short intervals defined by our study design (see below). The follow-up of the patients from 1 January 1994 until 1 July 1996 ranged from 2 to 32 months (median 14.4 months). If an MC was detectable, we intensified the monitoring regimen by analyzing patient samples twice a week during the first 100 days post-transplant.

Results We investigated the individual dynamics of chimerism in 50 follow-ups after allo-BMT from 46 pediatric patients with malignant (n = 32) or nonmalignant (n = 18) diseases. Four patients received allo-BMT more than once. We found CC in 34/50 cases, 12/50 patients with initial CC developed MC, and 4/50 patients showed AC within the observation period.

Patients with malignant diseases The CC status was detected initially in 22 cases within the group of 32 patients with malignant diseases after alloBMT. Eight patients developed MC during the observation period. Two of 32 patients (UPN 09 and UPN 26) initially showed autologous recovery with subsequent relapse. Three of 22 patients with CC (UPN 03, UPN 12 and UPN 21, see Figure 1) relapsed after transplantation without prior detection of MC. However, the time interval between the last follow-up and diagnosis of hematological relapse was greater than 6 months for all three patients. We suppose that we failed to detect MC because of this gap in the observation period. The individual results of VNTR-quantification within the follow-up of patients who developed MC post-transplant are shown in Figure 2. Concerning the individual dynamics of MC the patients given in Figure 2 showed particular characteristics. Patients who initially exhibited MC and subsequently showed increasing amounts of autologous DNA generally relapsed or rejected the graft. In addition, two patients (UPN 22 with MDS and UPN 20 with AML) who developed MC beyond day +100 for the first time relapsed as well. Just two patients were exempt from this general trend, (1) patient UPN 18 who suffered from AML rejected his graft 11 months post-transplant and continued to remain in complete remission for 2 years after BMT at the end of this study; (2) patient UPN 23 with MDS showed increasing amounts of autologous marrow repopulation after the first transplant and subsequently relapsed on day +80 post-transplant. He received a second bone marrow transplantation 1 year later. Again, he developed MC with increasing autologous patterns posttransplant. On day +40 when dysplastic cells were observed in the peripheral blood, we discontinued the cyclosporin A prophylaxis and administered donor lymphocytes on day +45 post-transplant. The patient returned to CC on day +60 and dysplastic cells were no longer detectable in bone marrow or peripheral blood. The patient remained in complete continuous remission 6 months after the second transplant. In addition, one of eight patients with MC turned subsequently to CC and remained in complete continuous remission (UPN 08 with ALL) 24 months from BMT. We therefore suppose that patients with MC and decreasing autologous patterns might represent a group with a favorable prognosis. Although the numbers of patients with malignant diseases and MC in this study are still relatively small (n = 8), statistical analysis showed a significant correlation (P = 0.0079) between the finding of increasing autologous signals and hematological relapse. Patients with nonmalignant disease We investigated follow-ups of 18 patients with nonmalignant diseases after allo-BMT. The CC status was initially detectable in 12/18 cases, 4/18 patients showed MC, and two patients recovered with their autologous marrow. The clinical course of the patients with nonmalignant diseases and MC is shown in Figure 3. Two of four patients developed increasing autologous patterns within the observation period (1) patient UPN 30 with severe aplastic ane-

ALL, ALL, ALL, ALL, ALL, ALL, ALL, ALL, ALL, ALL, ALL, ALL,

AML, AML, AML, AML, AML, AML, AML, AML, AML,

MDS MDS MDS MDS MDS MDS MDS MDS MDS CML, CP CML, CP

01 02 03 04 05 06 07 08 09 09a 10 11

12 13 14 15 16 17 18 19 20

21 21a 22 23 24 25 26 27 27a 28 29

1 1 1 1 1 1 1 2 1

2 1 1 1 1 1 1 2 1 1 2 2

7 8 14 1 8 8 4 2 3 17 7

5 4 14 1 6 10 3 1 7

9 4 10 9 13 12 17 12 4 4 13 14

Age (years)

TBI, CYC, ETO BU, CYC, ATG, MEL BU, CYC, ATG BU, CYC, ATG, MEL TBI, CYC, ATG TBI, CYC, ATG, THI TBI, CYC, ATG, ETO BU, CYC, ETO TBI, CYC, ATG, THI TBI, CYC, ETO, ATG TBI, CYC, ETO, ATG

BU, CYC BU, CYC BU, CYC BU, CYC BU, CYC BU, CYC BU, CYC BU, CYC TBI, CYC, ETO, ATG

TBI, ETO TBI, ETO TBI, ETO TBI, ETO TBI, ETO TBI, ETO TBI, ETO TBI, ETO TBI, THI, CYC, ATG HDC, OKT3 TBI, ETO, CYC, ATG TBI, ETO, CYC, ATG

Conditioning regimen

HLA-ids HLA-ids HLA-ids HLA-ids MUD MUD MUD oth-re oth-re MUD MUD

HLA-ids HLA-ids HLA-ids HLA-ids HLA-ids HLA-ids HLA-ids HLA-ids MUD

HLA-ids HLA-ids HLA-ids HLA-ids HLA-ids HLA-ids HLA-ids oth-re oth-re oth-re MUD MUD

Donor

CsA, CsA, CsA CsA CsA, CsA, CsA, CsA CsA CsA, CsA, MTX MTX

MTX MTX MTX

MTX MTX

CsA CsA CsA CsA CsA CsA CsA CsA CsA, MTX

CsA CsA CsA CsA CsA CsA CsA CsA CsA CsA, OKT3 CsA, MTX CsA, MTX

GVHD prophylaxis

no no aGVHD no no aGVHD no aGVHD aGVHD aGVHD no

aGVHD aGVHD aGVHD aGVHD no aGVHD aGVHD aGVHD aGVHD

aGVHD aGVHD aGVHD aGVHD aGVHD aGVHD no no no no aGVHD aGVHD

I I II

III

I

II I I I

I II I II

II II

II I I II III II

GVHD grade 589 713 230 261 155 188 125 474 0 225 481 103 395 649 555 284 173 452 752 450 185 320 42 164 46 557 347 0 58 138 81 45

CCb CC CC CC CC CC MC MC MC CCb CC MC CC CC CC AR MC MCc CC CC

DFS (days)

CC CC CCb CC CC CC CC MC AR MC CC CC

Chimerism

relapse died, complication died, relapse alive, well alive, well alive, well died, relapse alive, relapse alive, well died, CMV, cGVHD alive, well

died, relapse alive, well alive, well alive, well alive, well alive, well alive, well alive, relapse alive, relaspe

alive, well alive, well died, relapse alive, well died, complication alive, well alive, well alive, well graft rejection alive, relapse alive, well alive, well

State of disease

AML = acute myelogenous leukemia; ALL = acute lymphoblastic leukemia; AR = autologous recovery; ATG = antithymocyte globulin; BU = busulfan; CC = complete chimerism; CML = chronic myelogenous leukemia; CP = chronic phase; CR = complete remission; CsA = cyclosporin; CYC = cyclophosphamide; DFS = disease-free survival; ETO = etoposide; MC = mixed chimerism; MDS = myelodysplastic syndrome; MTX = methotrexate; THI = thiotepa; TBI = total body irradiation; UPN = unique patient number. a Patients who received two bone marrow transplants. b Time interval between last investigation of chimerism and relapse was more than 6 months. c Patient received donor lymphocytes.

CR CR CR CR CR CR CR CR CR

CR CR CR CR CR CR CR CR CR CR CR CR

Disease

Patient characteristics – malignant diseases

UPN

Table 1


699


700

Table 2

Patient characteristics – nonmalignant diseases

UPN

Disease

Age Conditioning regimen (years)

30 30a 31 32 33 34 35 36 37

SAA SAA SAA SAA SAA SAA SAA SAA SAA

11 11 11 5 6 3 5 7 2

38 39 40 41 42 43 44 45 46

SCD THA THA ADR ADR HIS HIS WAS SCID

13 4 15 8 9 1 2 1 1

Donor

GVHD prophylaxis

GVHD grade

Chimerism

DFS (days)

TNI, CYC TNI, CYC, ATG BU, CYC, ATG BU, CYC, ATG CYC, ATG CYC, ATG TNI, CYC, ATG, OKT3 BU, CYC, ATG BU, CYC, ATG

HLA-ids HLA-ids HLA-ids HLA-ids HLA-ids HLA-ids MUD MUD HLA-ids

CsA CsA CsA, MTX CsA CsA CsA CsA, MTX CsA CsA

no no no aGVHD aGVHD aGVHD no aGVHD no

MC CC CC CC CC CC CC CC CC

120 388 923 89 103 104 842 332 716

graft rejection alive, well alive, well alive, well alive, well alive, well alive, well alive, well alive, well

BU, BU, BU, BU, BU, BU, BU, BU, no

HLA-ids HLA-ids HLA-ids MUD MUD oth-re HLA-ids MUD oth-re

CsA, CsA, CsA CsA, CsA CsA CsA, CsA no

aGVHD aGVHD aGVHD no no no no aGVHD no

MC MC CC CC CC AR AR CC MC

180 360 210 0 0 0 0 253 219

alive, graft rejection alive, well died, GVHD died, progression died, progression died, progression died, progression alive, well alive, well

CYC, ATG CYC, ATG CYC, ATG CYC, ATG CYC, ATG ETO, CYC, ATG ETO, CYC CYC, ATG

MTX MTX MTX MTX

I I I II I II III

II

State of disease

ADR = adrenoleukodystrophy; AR = autologous recovery; ATG = antithymocyte globulin; BU = busulfan; CML = chronic myelogenous leukemia; CC complete chimerism; CsA = cyclosporin A; CYC = cyclophosphamide; DFS = disease-free survival; ETO =etoposide; HIS =histiocytosis; MC = mixed chimerism; MTX =methotrexate; SAA = severe aplastic anemia; SCD = sickle cell disease; SCID = severe combined immunodeficiency; THA = thalassemia major; TNI total node irradiation; UPN = unique patient number; WAS = Wiskott–Aldrich syndrome. a Patient received two bone marrow transplants.

UPN 03 12 21

0

ALL 99

0

0 0

UPN

99

0

0

0

AML

70

MDS

20 15 5

5

5 5 0

0

0 0

0 0 0 0

0 0 0

0

0 0 5 10204080 99

99

5

20

25

50

55

80

15

5