Effective T Cell Regeneration Following High-Dose Chemotherapy Rescued with CD34؉ Cell Enriched Peripheral Blood Progenitor Cells in Children

Home , T helper cell

Bone Marrow Transplantation, (1999) 23, 347–353  1999 Stockton Press All rights reserved 0268–3369/99 $12.00 http://www.stockton-press.co.uk/bmt Effective T cell regeneration following high-dose chemotherapy rescued with CD34؉ cell enriched peripheral blood progenitor cells in children

A Heitger1, H Kern1, D Mayerl1, K Maurer1, D Nachbaur2, M Fru¨hwirth1, F-M Fink1 and D Niederwieser3

1University Children’s Hospital Innsbruck; 2Department of Internal Medicine, Division of Clinical Immunobiology and Bone Marrow Transplantation, University Hospital Innsbruck, Austria; and 3Division Hematology and Oncology, University of Leipzig, Germany

Summary: shown to be an effective treatment for high risk malignancies.1,2 Peripheral blood mononuclear cells (PBMC) that The ex vivo enrichment for the CD34؉ cell fraction of have been stimulated with G-CSF contain a small fraction PBPC, while it retains the capacity to restore haematopo- of haematopoietic CD34+ stem cells3,4 which are able suc- iesis and potentially reduces a contamination by tumour cessfully to restore haematopoiesis after myeloablative HD- cells, implements a depletion of T cells. To test whether CTX.5,6 The success of such treatment, however, may be such a setting adversely affects T cell reconstitution, we hampered by a contamination of the leukapheresis product monitored T cells in four paediatric patients after CD34؉ with residual circulating tumour cells.7–9 Therefore, the selected PBPC transplantation. The dose of CD34؉ cells, enrichment of G-CSF stimulated PBMC for CD34+ cells which were enriched to 74%, median, was 7.1 ؋ 106/kg, might be of benefit by reducing the risk of reinfusing vital median, that of T cells was 0.071 ؋ 106/kg, median. The tumour cells together with the leukapheresis product.10 This patients were homogenous with respect to features with a procedure, however, results in an effective depletion of T potential to effect T cell reconstitution (low median age, cells. Since previous evidence suggests that peripheral T (3.5 years); stage IV malignant tumours in first CR, cells reinfused together with bone marrow or PBPC sig- uncomplicated post-treatment course). The results of nificantly contribute to the reconstitution of T cells after sequential FACS analyses showed that by 9 months after myeloablative treatment,11–13 T cell depletion may treatment all four patients had recovered (1) a normal T adversely affect T cell regeneration.14 Recent reports of the cell count (CD3؉ cells 1434/␮l, median); (2) a normal kinetics of lymphoid reconstitution after CD34+ selected CD4؉ cell count (816/␮l, median), while CD8+ cells were autologous and allogeneic PBPC transplantation in children recovered (Ͼ330/␮l) already by 3 months; (3) a normal and adults15,16 showed that T cell reconstitution was CD4/CD8 ratio (1.8, median), as a result of an augmented delayed as compared to a rescue with unmanipulated grafts. growth of CD4؉ cells between 3 and 6 months (increase of We hypothesised that high residual thymic activity, as -CD4؉ cells 4.9-fold, median, CD8؉ cells 1.1-fold, median). present in early childhood, might allow for a rapid regener Expansion of cells with a CD45RA؉ phenotype (thymus- ation of T cells in the absence of suppressive factors, such derived T cells) predominated; from 3 to 6 months the as GVHD, and studied T cell regeneration in four paediatric -increase of CD4؉/CD45RA؉ T cells was 130-fold, that of patients after autologous CD34+ selected PBPC transplan CD4؉/CD45RO؉ cells was 1.7-fold; CD8؉/CD45RA؉ cells tation. These patients were most homogenous with respect -increased 9-fold, CD8؉/CD45RO؉ cells increased 2.1-fold, to factors with a potential relevance to the T cell regenerat indicating effective thymopoiesis. The findings demon- ive capacity (see below). The results show an effective and strate that in paediatric patients the setting of HD-CTX rapid reconstitution of T cells and T cell subsets based on rescued with autologous CD34؉ selected PBPC per se is a preferential use of a thymopoietic pathway, indicating that not predictive of an impaired T cell recovery. High thymic T cell depletion of PBPC by positive selection for CD34+ activity may be a key factor for the rapid restoration of cells does not per se adversely affect T cell reconstitution T cells. in the paediatric age group. Keywords: CD34+ enrichment; peripheral stem cell rescue; T cell reconstitution; children; solid tumours Patients and methods

High-dose chemotherapy (HD-CTX) with autologous per- Patients ipheral blood progenitor cell (PBPC) rescue has been The clinical characteristics of the patients are summarised in Table 1. Care was taken to avoid a variety of combi- Correspondence: Dr A Heitger, University Children’s Hospital, Anich- nations of factors with a potential relevance to T cell regen- stra␤e 35, A-6020 Innsbruck, Austria erative capacity. The four consecutively treated patients had Received 3 August 1998; accepted 21 September 1998 similar clinical features. The median age was low (3.5 CD34؉ enriched PBPC rescue and T cell recovery in children A Heitger et al 348 Table 1 Patient characteristics

Patient sex, Diagnosis Preparative CD34+ cells T cells Engraftment Outcome age (years) regimen reinfused reinfused (WBC (time after (×106/kg) (×106/kg) Ͼ1.0 ϫ 109/l) reinfusion of stem cells)

1 M, 3 neuroblastoma IV MEC 7.6 0.16 +11 alive, well (4 yrs 4 mo) 2 M, 2 neuroblastoma IV MEC 9.5 0.057 +12 alive, well (1 yr 11 mo) 3 M, 4 rhabdomyosarcoma IV MEC 6.6 ND +12 alive, well (1 yr 10 mo) 4 F, 8 neuroblastoma IV MEC 1.9 0.071 +13 dead of relapse (1 yr 6 mo)

m = male; f = female; yrs = years; mo = months; MEC = melphalan 160 mg/m2 in patients 1, 2; 140 mg/m2 in patients 3, 4; etoposide 60 mg/kg; carboplati- num 1500 mg/m2;ND= not determined.

years, range 2–8 years). All had advanced stage malignant number of reinfused T cells ranged from 0.057 to tumours, all had been treated with four to six courses of 0.16 × 106/kg (Table 1). conventional chemotherapy and were in first CR prior to receiving autologous PBPC rescue, and received the same Analysis of T cell regeneration preparative regimen. The engraftment was supported by rh- G-CSF (5–10 ␮g/kg/day s.c.) and was rapid and stable in T cells and T cell subsets were quantified by FACS. The all four cases. With respect to the post-transplant period following fluorochrome-labeled monoclonal antibodies none of the patients received extended field radiation ther- (MoAb), all purchased from Becton Dickinson (BD; Moun- apy and all patients had an uncomplicated post-transplant tain View, CA, USA), were used: negative controls, Simult- course, ie no severe infection, no evidence of a tumour est control (fluorescein (FITC), phycoerythrin (PE)), IgG1 relapse, while being investigated for T cell reconstitution. (peridinin chlorophyll protein (PerCP)); anti-CD3 (PerCP); anti-CD4 (PerCP or FITC); anti-CD8 (PerCP or PE); anti- ␣␤ Composition of CD34+ selected PBPC CD45RA (FITC); anti-CD45RO (PE); anti-TCR (FITC); anti-TCR␥␦ (FITC). To quantify a contamination with red The methods of collection and selection of CD34+ cells blood cells one portion of each sample was also stained have recently been described.17 Briefly, PBMC were with anti-glycophorin A (PE). Peripheral heparinized blood collected in the phase of haematopoietic recovery after was collected on the occasion of regular follow-up visits cytotoxic treatment and stimulation with rh-G-CSF with informed consent of the parents from 3 to 12 months (5 ␮g/kg/day). PBMC were separated from whole blood after treatment (missing examinations: patient No. 1 at 9 using a CS 3000 Plus cell separator (Baxter, Healthcare months for technical reasons, patient No. 4 at 12 months Fenwal Division, Deerfield, IL, USA). Three leukaphereses because of relapse with haematopoietic failure due to bone were performed on each patient. The PBMC of the first two marrow infiltration by tumour cells). PBMC were isolated leukapheresis products were pooled and CD34+ cells were from whole blood by Lymphoprep (Nycomed, Oslo, positively selected on a Ceprate SC Stem Cell Concen- Norway) density centrifugation and washed twice. The tration System loaded with a CD34-specific biotinylated isolated PBMC population usually contained less than 2% monoclonal antibody (CellPro, Bothell, WA, USA). This glycophorin A+ cells. Aliquots of 300 000 cells were first resulted in an enrichment for CD34+ cells of 74%, median incubated with MOPC 21 (mouse IgG1 ␬; Sigma, St Louis, (range 51–88%). Before use the cells were cryopreserved MO, USA) on ice for 20 min to block non-specific binding. in 10% DMSO by controlled rate freezing and stored in Then the cells were stained with the appropriate MoAb by liquid nitrogen. The third apheresis product was cryopre- incubation on ice for 30 min, washed twice and immedi- served as a back-up. ately analysed on a FACScan flow cytometer (BD). Lym- With respect to T cells, conventional chemotherapy for phocytes were identified by forward scatter (FSC) vs side neuroblastoma and rhabdomyosarcoma, as given to these scatter (SSC) and gated electronically to allow for a calcu- patients, invariably induces a severe depletion of T cells lation of absolute cell numbers (see below). (Ref. 18 and manuscript in preparation). For example, The analyses of two- and three-colour studies were per- patient No. 1 had only 68/␮l CD3+ cells 3 weeks prior to formed on a FACScan research software 2.1 (BD). Three- HD-CTX with CD4+/45RA+ cells being most severely colour analysis was performed to examine T cell subsets depleted. These abnormalities of T cells translated into the within a preselected T cell population, as described.20 To apheresis product. As examined in patient Nos 1, 2 and quantify the proportion of CD4+ and CD8+ cells, CD3+ cells 4, the content of total T cells was 2.4%, 4.1% and 7.4%, were identified by quadrant analysis of FL3 (PerCP) vs respectively. Sixty-eight per cent, median, of these were SSC, and gated. These CD3+ cells were then separated into CD4+ T cells of which the proportion of CD4+/CD45RA+ fractions of FL1 (anti-CD4, FITC) and FL2 (anti-CD8, PE) cells was severely reduced (13%, 26% and 8% of CD4+ and quantified by quadrant analysis. To quantify CD45RA+ cells in patients 1, 2 and 4, respectively; normal value 66– and CD45RO+ cells within the T helper and T 77%).19 The T cell content of the apheresis product after cytotoxic/suppressor subset, CD4+ or CD8+ cells were selection for CD34+ cells was 1.1%, median (range 0.54– identified by quadrant analysis of FL3 (PerCP) vs SSC, 2.9%) with a similar distribution of T cell subsets. The gated and then examined by quadrant graphs of FL1 (anti- CD34؉ enriched PBPC rescue and T cell recovery in children A Heitger et al 349 CD45RA, FITC) vs FL2 (anti-CD45RO, PE). The distri- cells (CD8+) over T helper cells (CD4+). This was the result bution of the different forms of the TCR was examined of an increased growth rate of CD8+ cells, most of them by a simultaneous expression of CD3 (PerCP) and TCR␣␤ displaying a suppressor phenotype, and a late recovery of (FITC) or TCR␥␦ (FITC), respectively (FL3 vs FL1). The CD4+ cells.13,23,26 The findings of T cell subset analysis in absolute number (cells/␮l) of each T cell subpopulation was this group of paediatric patients were different. Only at the calculated according to the formula: initial analysis 3 months after treatment was the number of CD3+/CD8+ cells higher than that of CD3+/CD4+ cells lymphocytes/␮l + cells/␮l = % positive cells × (Figure 2). At this time point CD8 cells had returned to 100 baseline levels in three of the four patients (349/␮l, median, (range 124–580/␮l); normal range 330–1400/␮l), while The absolute lymphocyte count was derived from the dif- + ferential count of the white blood count. The age-adapted CD4 cells were below the threshold of normal (141/␮l, median, (range 24–165/␮l); normal range 560–2700/␮l). normal values of T cells and T cell subsets were taken from + recent standard literature.21 Between 3 and 6 months the growth rate of CD4 cells exceeded that of CD8+ cells; CD4+ cells increased 4.9-fold while CD8+ cells increased only 1.1-fold (median). This Results resulted in a normal T helper cell count in three of four patients (Figure 2a). In all four patients, even in patient 3 Recovery of total T cells with a low T cell count, the CD4/CD8 ratio became normal by 6 months and remained so throughout the observation The median period of time required to recover total T cells + period (CD4/CD8 ratio 1.7, median, range 1.2–2.5). (CD3 ) into the age-adapted normal range was 6 months Double positive T cells (CD3+/CD4+CD8+) previously (Figure 1). By 9 months the T cell count of all four patients described as occurring in an elevated frequency in the early was well above the lower limit of 1.0 ϫ 109/l. The pro- phase of lymphocyte reconstitution after BMT27 were less portion of T cells contained within total lymphocytes at the than 2%, median, throughout the observation time. Further- initial evaluations was relatively low (48%, median, at 3 more, only one of the four patients (patient 3) showed a months, 52%, median, at 6 months; normal range 52–85%). transient abundance of double negative (CD3+/CD4−/CD8−) Total lymphocytes returned to a normal level by 6 months cells (24% at 3 months, 14% at 6 months). after treatment (2148/␮l, median). This reflects the long period of time for T cells to regenerate as compared to other lymphocytes, eg B cells22 or NK cells.23 The distribution of T cells expressing the TCR␣␤ and TCR␥␦, which may be a altered after autologous BMT24,25 remained normal 1400 + ␣␤+ Ϯ Ϯ (CD3 /TCR cells 92 5% (mean s.e.m.), 1200 CD3+/TCR␥␦+ cells 3 Ϯ 1%) throughout the observation period with a single exception (patient 3, 14% 1000 CD3+/TCR␥␦+ at 6 months after treatment). µ 800 600 Cells/ l Recovery of T cell subsets 400 Previous studies of adult autologous and allogeneic BMT 200 and high-dose chemotherapy have consistently shown a 0 long-lasting imbalance of the two major T cell subsets, 3 6 9 12 characterised by a predominance of T cytotoxic/suppressor Months after stx b 3000 1400

2500 1200

2000 1000 µ µ 800 1500

Cells/ l 600 Cells/ l 1000 400

500 200

0 0 3 6 9 12 3 6 9 12 Months after stx Months after stx

Figure 1 Reconstitution of total T cells (CD3+). Each symbol represents Figure 2 Reconstitution of T cell subsets. (a) T helper cells (CD4+); (b) an individual patient: patient 1 (̆); patient 2 (᭜); patient 3 (᭿); patient T cytotoxic/suppressor cells (CD8+). The symbols represent the cell counts 4(̄). The line indicates the median value of CD3+ cells. The shaded area of individual patients as in Figure 1. The line indicates the median value. indicates the normal range of total T cells according to age; stx, stem The shaded area indicates the normal range according to age; stx, stem cell rescue. cell rescue. CD34؉ enriched PBPC rescue and T cell recovery in children A Heitger et al 350 Distribution of CD45RA+ vs CD45RO+ T cell subsets between 3 and 6 months increased 130-fold (from 3/␮lto 392/␮l), while that of CD45RO+ T helper cells increased The expression of the high molecular weight isoform of the only 1.7-fold (from 130/␮l to 222/␮l). The subsequent CD45 antigen (CD45RA) designates T cells as being + + 28 growth rates of CD45RA cells and CD45RO cells were recently thymus derived. In contrast, T cells with a similar (1.2-fold vs 0.8-fold and 1.7-fold vs 2.2-fold at 9 CD45RO phenotype are considered to represent peripheral and at 12 months, respectively). Unexpectedly, in the T memory cells. The distinction of these two CD45 isoforms cytotoxic/suppressor subset the expansion of the two CD45 has allowed discrimination between a thymus-dependent isoform expressing T cell subgroups followed a similar pat- and a thymus-independent pathway of regeneration of the + + tern as in CD4 cells (Figure 3b). At 3 months after treat- T helper subset. Cells with a CD45RA phenotype were + + + ment, the median CD45RA /CD45RO ratio of CD8 cells found to regenerate only in the presence of residual thy- 20,29 was 0.07. Between 3 and 6 months after treatment, mus. This pattern was not as stringent in the T + + CD45RA cells increased 9-fold (from 21/␮l to 190/␮l, cytotoxic/suppressor subset, in which CD45RA cells were + 13,20,29 median) while CD45RO cells decreased by 50% (from regenerated even in the absence of residual thymus. 295/␮l to 143/␮l, median), such that the ratio had turned To determine the relative contribution of a thymus- to normal (median 1.5, range 1.0–1.9).30 From 6 months dependent and thymus-independent pathway of T cell + + after treatment, the growth rate of CD8 /CD45RA and that reconstitution, we sequentially examined the fraction of T of CD8+/CD45RO+ cells was similar (2.2-fold vs 1.2-fold, cells expressing the CD45RA antigen vs that expressing the + + + + + and 1.6-fold vs 1.4-fold, CD8 /CD45RA vs CD8 / CD45RO antigen within the CD4 and the CD8 subsets. + CD45RO at 9 and at 12 months, respectively). In the T helper subset, as expected, CD45RO+ cells rep- resented by far the predominant population at 3 months after treatment, the CD45RA+/CD45RO+ ratio being 0.02 Discussion (Figure 3a). At this time point the absolute count of T helper cells in all patients was extremely low (median + The positive selection of CD34 cells in the apheresis pro- 141/␮l, range 24–165/␮l). The subsequent analysis at 6 + duct used in this study implies that the amount of peripheral months after treatment – a time period at which total CD4 T cells available to contribute to T cell regeneration is cells were increasing rapidly – showed that predominantly reduced as compared to unmanipulated grafts. This, hypo- CD45RA+ cells expanded. Their absolute count (median) thetically, shifts T cell regeneration towards a pathway that employs haematopoietic stem cells and thymopoiesis as the a source of reconstitution. 1400 The ﬁndings of this study of paediatric patients undergo- + 1200 ing HD-CTX with CD34 selected PBPC transplantation can be summarised as follows: (1) A period of 6–9 months 1000 after treatment was required to quantitatively restore total µ 800 T cells; (2) A normal distribution of the major T cell sub- 600 sets was reconstituted rapidly as a combined result of an Cells/ l absence of an overshoot of CD8+ cells and a high growth 400 rate of CD4+ cells; (3) In both major T cell subsets cells 200 with a CD45RA+ phenotype predominantly expanded. 0 These ﬁndings indicate an effective regeneration of T cells, 0 1 2 3 4 5 6 7 8 9 10 11 12 by an effective use of a thymopoietic pathway. Months after stx In general, PBPC transplantation might improve T cell b regeneration as compared to BMT, as several studies in adults, including patients with leukaemia, NHL, MM and 1400 solid tumours, have suggested.31–34 These studies also 1200 showed a more rapid regeneration of a normal CD4/CD8 31 1000 ratio. In the autologous setting Talmadge et al observed

µ a normalised CD4/CD8 ratio and an improved cytotoxic 800 and proliferative response even 3 months after PBPC treat- 600 Cells/ l ment. Also, in HLA-matched allogeneic transplantation, 400 O¨ ttinger et al34 found recipients of PBPC to have a more 200 rapid reconstitution of T cells, which included the regeneration of CD4+/CD45RA+ naive T helper cells, and to better 0 recover an in vitro response to non-specific and specific 0 1 2 3 4 5 6 7 8 9 10 11 12 + Months after stx stimulatory antigens. The experience with CD34 selected PBPC transplantation in adults is more limited. Most Figure 3 Composition of the two major subsets by CD45RA+ and recently, Bomberger et al16 investigating patients with NHL + CD45RO cells. (a) T helper subset; (b) T cytotoxic/suppressor subset. showed a delay in recipients of autologous CD34+ selected The bars indicate the median value at each time point: (black), CD45RA+ cells; (hatched), CD45RA+/RO+ (double positive) cells; (grey), CD45RO+ PBPC as compared to a control population who received cells. The shaded area indicates the normal range according to age; stx, unsorted PBPC. Yet, in their study even the control popu- stem cell rescue. lation failed to recover a normal T cell count by 12 months CD34؉ enriched PBPC rescue and T cell recovery in children A Heitger et al 351 after treatment (mean T cell count Ͻ700/␮l). This finding Fourth, the age of the patients (2–8 years) was low. It has was in line with multiple previous studies, as recently been shown that age as it reflects residual thymic activity extensively reviewed by Archimbaud.35 Thus, there is evi- is an important factor for an effective regeneration of T dence that in adult PBPC transplantation a depletion of T cells, especially of T helper cells.44 In fact, the rapid T cells by positive selection of CD34+ cells adversely affects helper cell reconstitution in all patients in this study T cell regeneration. In the paediatric age group the knowl- occurred by a predominant expansion of cells with a naive edge is scarce. Takaue et al36 investigated 41 children with (CD45RA+) phenotype, which indicates the effective use of leukaemia and NHL after unsorted autologous PBPC trans- a thymopoietic regenerative pathway.20,29 It is intriguing to plantation and showed abnormally low CD4+ counts and speculate that in addition to a high number of haematopo- an inverse CD4/CD8 ratio for Ͼ12 months after treatment. ietic progenitor cells, young age, as it favours the regener- Matsuda et al15 reported five children having undergone ation via thymus-derived T cells, is a factor that accelerates CD34+ selected PBPC transplantation from allogeneic mis- T helper cell reconstitution. matched donors. The CD3+ cell counts remained low One interesting finding was that even within the T through the observation time of a maximum of 210 days cytotoxic/suppressor subset CD45RA+ cells were the pre- after transplantion with CD4+ cells being more severely dominantly expanding population. Since CD8+/CD45RA+ depressed than CD8+ cells. However, the interpretation of cells are capable of regenerating even in the absence of this observation is difficult, since four of the five patients thymus20,29 and their regeneration is not age-dependent,13 developed acute or chronic GVHD, a factor known to pre- no conclusion can be drawn as to whether this also indicates dictably affect T cell regeneration. No recent experience a predominant regeneration via thymus-derived CD8+ cells. exists with autologous CD34+ selected PBPC transplan- Irrespective of the hypothetical origin of CD45RA+ cells, tation in children. Thus, the findings of this study, a time all patients were able to rapidly, ie by 6 months after treat- period of 6 to 9 months after treatment required to recover ment, restore a normal distribution of CD45RA+ and normal T cells including normal T cell subsets (a normal CD45RO+ cells in both major T cell subsets. A further fac- CD4/CD8 ratio and a normal CD45RA/CD45RO ratio tor with a potential, but not clarified influence on T cell within the T helper and T cytotoxic/suppressor subset), are regeneration was the use of G-CSF in the immediate post- promising. They fit well into the range that has been transplant period, which was used to accelerate granulo- described for paediatric BMT. Foot et al37 reported a simi- cytic recovery. G-CSF has multiple effects on other haema- lar time period in paediatric allogeneic HLA-matched topoietic cell lineages as well, which include an increment BMT, which included T cell-replete and T cell-depleted in T cell numbers.45,46 grafts. Kook et al38 found that in paediatric allogeneic BMT Notably, within the observation period of 12 months after using T cell-depleted matched unrelated or partially treatment, none of the patients experienced infectious prob- matched donors a period of Ͼ12 months was required to lems beyond the expected range. This observation is not normalise the peripheral T cell number. trivial since recent evidence suggests that T cell function Notably, the present study examined a clinically very remains impaired even beyond the quantitative normaliz- homogenous group of patients in whom features with a ation of T cells and T cell subsets.47–50 One further problem potential adverse effect on T cell regeneration, except the with respect to T cell function is the restriction of the T selection process of CD34+ cells which included a T cell cell repertoire after conventional BMT and CD34+ selected depletion, were absent. First, none of the patients PBSC treatment.16,51,52 Yet, a rapid increase of naive T underwent long-lasting extensive conventional chemo- cells, as in these young patients, has been shown to set the therapy prior to and/or irradiation prior to or after HD-CTX scene for the restoration of an unrestricted and diverse T with PBPC rescue. Our recent experience with patients who cell repertoire.53 This needs to be confirmed. received tandem39 or triple HD-CTX with unmanipulated Together, the findings show that the setting of HD-CTX PBPC rescue, in line with recent reports,40,41 suggests that and autologous CD34+ selected PBPC transplantation does these factors may adversely affect the bone marrow regen- allow for a an effective and rapid T cell recovery in paedi- erative capacity including a poor regeneration of T cells atric patients. Such an approach might be desired in circum- (not shown). Second, no severe infection, especially no stances in which tumour cells potentially contaminate the infection with CMV, occurred during the post-transplant leukapheresis product and can be removed by a selection period. CMV infection is known to abrogate a successful T for CD34+ cells, such as in stage IV neuroblastoma.10 This cell regeneration, as exemplified in one patient after CD34+ concept warrants further testing. The extent as to which selected PBSC transplantation.15 Third, the dose of CD34+ factors with a potential adverse effect on engraftment (eg cells was high (median 7.1 × 106CD34+ cells/kg). It has a low number of CD34+ cells, partial bone marrow damage been shown that the period of time required for leukocytic by irradiation or prolonged pre-transplant chemotherapy) and thrombocytic recovery after PBPC rescue is linked to have a disadvantageous effect also on T cell regeneration the number of CD34+ cells used as a rescue.17,42 A high remains to be determined. stem cell support may also be critical to a rapid reconstitution of T cells.31,32 One recent study by Carabasi et al43 + investigating 10 adult patients after allogeneic CD34 selec- Acknowledgements ted PBPC transplantation using Ͼ9 × 106/kg CD34+ cells, median, showed a rapid normalisation of the CD4/CD8 This work was supported in part by the Children’s Cancer + + ratio even by 2 months and a regrowth of CD3 /CD45RA Research Institute (CCRI), St Anna Children’s Hospital, cells, indicating a regeneration via newly produced T cells. Vienna, Austria. CD34؉ enriched PBPC rescue and T cell recovery in children A Heitger et al 352 References tumors. A single center experience. Bone Marrow Transplant 1998; 20: 827–834. 1 Kessinger A, Armitage JO, Smith DN et al. High dose therapy 18 Mackall CL, Fleisher TA, Brown MR et al. Lymphocyte and autologous peripheral blood stem cell transplantation for depletion during treatment with intensive chemotherapy for patients with lymphoma. Blood 1989; 74: 1260–1265. cancer. Blood 1994; 84: 2221–2228. 2 Garaventa A, Hartmann O, Bernard JL et al. Autologous bone 19 Hannet I, Erkeller-Yuksel F, Lydyard P et al. Developmental marrow transplantation for pediatric Wilms’ tumor: the experi- and maturational changes in human blood lymphocyte subpop- ence of the European Bone Marrow Transplantation Solid ulations. Immunol Today 1992; 13: 215–218. Tumor Registry. Med Ped Oncol 1994; 22: 11–14. 20 Heitger A, Neu N, Kern H et al. Essential role of the thymus 3 Kanold J, Rapatel C, Berger M et al. Use of G-CSF alone to to reconstitute naive (CD45RA+) T-helper cells after human mobilize peripheral blood stem cells for collection from chil- allogeneic bone marrow transplantation. Blood 1997; 90: dren. Br J Haematol 1994; 88: 633–635. 850–857. 4 Klingebiel T, Handgretinger R, Herter M et al. Autologous 21 Brugnara C. Reference values in infancy and childhood. In: transplantation with peripheral blood stem cells in children Nathan DG, Orkin SH (eds). Hematology in Infancy and and young adults after myeloablative treatment, nonran- Childhood, 5th edn. WB Saunders Co: Philadelphia, 1998, domized comparison between GM-CSF and G-CSF for mobil- Appendix 29. ization. J Hematother 1995; 4: 307–314. 22 Locatelli F, Maccario R, Comoli et al. Hematopoietic and 5 Andrews RG, Bryant EM, Bartelmez SH et al. CD34+ marrow immune recovery after transplantation of cord blood progeni- cells devoid of T and B lymphocytes, reconstitute stable lym- tor cells in children. Bone Marrow Transplant 1996; 18: phopoiesis and myelopoiesis in lethally irradiated allogeneic 1095–1101. baboons. Blood 1992; 80: 1693–1701. 23 Lum LG. The kinetics of immune reconstitution after human 6 Link H, Arseniev L, Ba¨hre O et al. Transplantation of allo- marrow transplantation. Blood 1987; 69: 369–380. geneic CD34+ blood cells. Blood 1996; 78: 4903–4909. 24 Cela ME, Holladay MS, Rooney CM et al. ␥␦ T lymphocyte 7 Nagasu M, Aizawa S, Hojo H et al. Detecting of the minimal regeneration after T lymphocyte-depleted bone marrow trans- residual disease contaminated in peripheral blood stem cell plantation from mismatched family members or matched unre- transplantation in the B-cell malignant lymphoma patients. Am lated donors. Bone Marrow Transplant 1996; 17: 243–247. J Hematol 1992; 41: 107–112. 25 San Miguel JF, Hernandez MD, Gonzales M et al. A ran- 8 Moss TJ, Cairo M, Santana VM et al. Clonogenicity of circul- domized study comparing the effect of GM-CSF and G-CSF ating neuroblastoma cells: implications regarding peripheral on immune reconstitution after autologous bone marrow trans- blood stem cell transplantation. Blood 1994; 83: 3085–3089. plantation. Br J Haematol 1996; 94: 140–147. 9 Miyajima Y, Horibe K, Fukuda M et al. Sequential detection 26 Favrot M, Janossy G, Tidman N et al. T cell regeneration after of tumor cells in the peripheral blood and bone marrow of allogeneic bone marrow transplantation. Clin Exp Immunol patients with stage IV neuroblastoma with reverse transcrip- 1983; 54: 59–72. tion-polymerase chain reaction for tyrosine hydroxylase 27 Storek J, Ferrara S, Rodriguez C, Saxon A. Recovery of mRNA. Cancer 1996; 77: 1214–1219. mononuclear cell subsets after bone marrow transplantation: 10 Tchirkov A, Kanold J, Giollant M et al. Molecular monitoring overabundance of CD4+CD8+ dual-positive T cells remi- of tumor cell contamination in leukapheresis products from niscent of ontogeny. J Hematother 1992; 1: 303–316. stage IV neuroblastoma patients before and after CD34 selec- 28 Spits H, Lanier LL, Philips JH. Development of human T and tion. Med Ped Oncol 1998, 30: 228–232. natural killer cells. Blood 1995; 85: 2654–2670. 11 De Gast GC, Verdonck LF, Middeldorp JM et al. Recovery 29 Mackall CL, Granger L, Sheard MA et al. T-cell regeneration of T cell subsets after autologous bone marrow transplantation after bone marrow transplantation: differential CD45 isoform is mainly due to proliferation of mature T cells in the graft. expression on thymic-derived versus thymic-independent pro- Blood 1985; 66: 428–431. geny. Blood 1993; 82: 2585–2594. 12 Roux E, Helg C, Dumont-Griard F et al. Analysis of T-cell 30 De Paoli P, Battisitin S, Santini GF. Age-related changes in repopulation after allogeneic bone marrow transplantation: human lymphocyte subsets: progressive reduction of the CD4 significant differences between recipients of T-cell depleted CD45R (suppressor inducer) population. Clin Immunol Immu- and unmanipulated grafts. Blood 1996; 87: 3984–3992. nopathol 1988; 48: 290–296. 13 Mackall CL, Fleisher TA, Brown MR et al. Distinctions 31 Talmadge JE, Reed K, Ino K et al. Rapid immunologic recon- between CD8+ and CD4+ T-cell regenerative pathways result stitution following transplantation with mobilized peripheral in prolonged T-cell subset imbalance after intensive chemo- blood stem cells as compared to bone marrow. Bone Marrow therapy. Blood 1997; 89: 3700–3707. Transplant 1997; 19: 161–172. 14 Lowdell MW, Craston R, Ray N et al. The effect of T cell 32 Roberts MM, To LB, Gillis D et al. Immune reconstitution depletion with Campath-1M on immune reconstitution after following peripheral blood stem cells transplantation, autolog- chemotherapy and allogeneic bone marrow transplantation as ous bone marrow transplantation and allogeneic bone marrow treatment for leukemia. Bone Marrow Transplant 1998; 21: transplantation. Bone Marrow Transplant 1993; 12: 469–475. 679–686. 33 Henon PR, Liang H, Beck-Wirth G et al. Comparison of hem- 15 Matsuda Y, Hara J, Osugi Y et al. Allogeneic peripheral stem atopoietic and immune recovery after autologous bone marrow cell transplantation using positively selected CD34+ cells from or blood stem cell transplantation. Bone Marrow Transplant HLA-mismatched donors. Bone Marrow Transplant 1998; 21: 1992; 9: 285–291. 355–360. 34 O¨ ttinger HD, Beelen DW, Scheulen B et al. Improved immune 16 Bomberger C, Singh-Jairam M, Rodey G et al. Lymphoid reconstitution after allotransplantation of peripheral blood reconstitution after autologous PBSC transplantation with stem cells instead of bone marrow. Blood 1996; 88: 2775– FACS-sorted CD34+ hematopoietic progenitors. Blood 1998; 2779. 91: 2588–2600. 35 Archimbaud E. Immune reconstitution following transplan- 17 Nachbaur D, Fink FM, Nussbaumer W et al. CD34+ selected tation of selected hematopoietic stem cells. Hematol Cell Ther autologous peripheral stem cell transplantation (PBSCT) in 1997; 39: 264–268. patients with poor risk hematological malignancies and solid 36 Takaue Y, Okamoto Y, Kawano Y et al. Regeneration of CD34؉ enriched PBPC rescue and T cell recovery in children A Heitger et al 353 immunity and varicella-zoster infection after high-dose colony-stimulating factor (G-CSF) on peripheral leukocyte chemotherapy and peripheral blood stem cell autografts in and lymphocyte subsets in patients with chemotherapy- children. Bone Marrow Transplant 1994; 14: 219–223. induced leukopenia. Geburtshilfe und Frauenheilkunde 1994; 37 Foot ABM, Potter N, Donaldson C et al. Immune reconsti- 54: 670–674. tution after BMT in children. Bone Marrow Transplant 1993; 46 Dreger P, Haferlach T, Eckstein V et al. G-CSF-mobilized 11: 7–13. peripheral blood progenitor cells for allogeneic transplan- 38 Kook H, Goldman F, Padley D et al. Reconstruction of the tation: safety, kinetics of mobilization and composition of the immune system after unrelated or partially matched T-cell- graft. Br J Haematol 1994; 87: 609–613. depleted bone marrow transplantation in children: immuno- 47 Rozans MK, Smith BR, Burakoff SJ, Miller RA. Long-lasting phenotypic analysis and factors affecting the speed of recov- deficit of functional T cell precursors in human bone marrow ery. Blood 1996; 88: 1089–1097. transplant recipients revealed by limiting dilution methods. J 39 Maurer K, Heitger A, Schwaighofer H et al. Double high-dose Immunol 1986; 136: 4040–4048. chemotherapy with autologous peripheral stem cell rescue in 48 Armitage RJ, Goldstone AH, Richards JDM, Cawley JC. relapsed Wilms’ tumor. Bone Marrow Transplant 1997; 20: Lymphocyte function after autologous bone marrow transplan- 1111–1113. tation (BMT): a comparison with patients treated with allo- 40 Brice P, Marolleau JP, Pautier P et al. Hematological recovery geneic BMT and with chemotherapy alone. Br J Haematol and survival of lymphoma patients after autologous stem-cell 1986; 63: 637–647. transplantation: comparison of bone marrow and blood pro- 49 Anderson KC, Soiffer R, DeLage R et al. T-cell-depleted auto- genitor cells. Leuk Lymphoma 1996, 22: 449–456. logous bone marrow transplantation therapy: analysis of 41 Morton J, Morton A, Bird R et al. Predictors for optimal mobi- immune deficiency and late complications. Blood 1990; 76: lization and subsequent engraftment of peripheral blood pro- 235–244. genitor cells following intermediate dose cyclophosphamide 50 Miller RA, Daley J, Ghalie R, Kaizer H. Clonal analysis of and G-CSF. Leukemia Res 1997; 21: 21–27. T-cell deficiencies in autotransplant recipients. Blood 1991; 42 Bensinger W, Appelbaum F, Rowley S et al. Factors that 77: 1845–1850. influence collection and engraftment of autologous peripheral- 51 Villers D, Milpied N, Gaschet J et al. Alteration of the T cell blood stem cells. J Clin Oncol 1995; 13: 2547–2555. repertoire after bone marrow transplantation. Bone Marrow 43 Carabasi MH, Tilden AB, Shresta K et al. Rapid immune Transplant 1994; 13: 19–26. reconstitution in recipients of T cell depleted peripheral blood 52 Gorochov G, Debre P, Leblond V et al. Oligoclonal expansion stem cell transplants from HLA identical siblings. Blood 1997; of CD8+CD57+ T cells with restricted T-cell repertoire ␤ chain 90(Suppl.): 542a (Abstr. 2415). variability after bone marrow transplantation. Blood 1994; 83: 44 Mackall CL, Fleisher TA, Brown MR et al. Age, thymopoiesis 587–595. and CD4+ T-lymphocyte regeneration after intensive chemo- 53 Mackall CL, Hakim FT, Gress RE. T-cell regeneration: not therapy. New Engl J Med 1995; 332: 143–149. all repertoires are created equal. Immunol Today 1997; 18: 45 Bo¨hme M, Schmidt D, Radke J et al. Effect of granulocyte 245–251.