Archives of Disease in Childhood 1994; 70: 237-240 237
Peripheral blood stem cells used to augment
autologous bone marrow transplantation
P L R Mitchell, V B Shepherd, H M Proctor, M Dainton, S D Cabral, C R Pinkerton
Abstract addition to stem cells, committed progenitor
Peripheral blood stem cells (PBSC) were cells capable of early maturation are collected
used to augment autologous bone marrow by leukapheresis. These committed progenitor
transplantation (ABMT), aiming to cells are a possible source of endogenous
hasten engraftment after high dose treat- growth factors evident before PBSC engraft-
ment in a group of heavily pretreated ment.6
patients. PBSC were obtained by leuka- After myelosuppressive chemotherapy there
pheresis during the rebound after is a massive increase in circulating progenitor
standard chemotherapy. In 11 patients cell levels.7 The accurate timing of PBSC
aged 7-17 years, high dose chemotherapy collections with this surge improves progenitor
consisted of busulphan 16 mg/kg orally cell yields.8-10 An alternative method of
with melphalan 160 mg/M2 intravenously priming for PBSC collection involves use of
for seven patients, and melphalan 200 the marrow growth factors granulocyte colony
mg/M2 intravenously alone for four. stimulating factor (G-CSF) or granulocyte-
The median number of granulocyte- macrophage colony stimulating factor (GM-
macrophage colony forming units in CSF)."
the reinfused PBSC was 3*42X 104/kg A major factor in the toxicity of high dose
(3-03-18-01) and bone marrow 12-4X104/ treatment with autologous bone marrow
kg (4 16-28-6). Neutrophil recovery to transplantation (ABMT) is the period of
B0S5XI09/1 and platelet transfusion myelosuppression, particularly in heavily pre-
independence occurred at a median of 14 treated patients. In order to achieve sustained
days (11-18) and 22 days (9-84) respec- engraftment after PBSC transplantation a
tively. In five patients the early engraft- number of PBSC harvest procedures may be
ment was transient with neutrophils required. A study in adults suggested that by
again dropping below 0-5Xl09/1 then combining PBSC transplantation and ABMT
slowly recovering. There was one toxic the number of leukaphereses could be reduced
death due to sepsis. PBSC harvesting in and a benefit from rapid, if transient, en-
these children was undertaken without graftment, was demonstrated.12 Eventual
interrupting routine chemotherapy and permanent engraftment was achieved later by
without the use of bone marrow growth the ABMT, but the combination reduced early
factors. In some patients PBSC failed to morbidity.
influence engraftment and the use of With the aim of reducing the duration of
combined chemotherapy and growth neutropenia and thrombocytopenia, PBSC
factor priming for PBSC collection may were similarly used in the current study to
give improved results. augment bone marrow rescue in a group of
(Arch Dis Child 1994; 70: 237-240) heavily pretreated patients. PBSC collections
were timed to coincide with the surge of
progenitor cells after routine chemotherapy. As
Dose escalation has become an important the PBSC transplant was not intended to
strategy in the treatment of childhood cancers replace ABMT only two collections were
and in a number of diseases attention has planned without using marrow growth factors
focused on high dose chemotherapy with in order to minimise disruption of the regular
autologous bone marrow rescue.1 Stem cells chemotherapy schedule.
Royal Marsden derived from the peripheral circulation may
Hospital, Paediatric also be used to achieve haematopoietic
Unit
P L R Mitchell reconstitution.2 3 Patients and methods
C R Pinkerton There are several potential advantages of PATIENTS AND HIGH DOSE CHEMOTHERAPY
Leukaemia Unit
transplantation using peripheral blood stem REGIMENS
V B Shepherd cells (PBSC).4 Patients with hypocellular bone Eleven patients aged 7-17 years received
H M Proctor marrow due to prolonged chemotherapy or combined PBSC transplantation and ABMT
M Dainton pelvic radiotherapy may be unsuitable for (table 1). There were five patients with
S D Cabral marrow harvesting. Sufficient progenitor cells Ewing's sarcoma, three with rhabdomyo-
Correspondence and can be obtained by PBSC collection even sarcoma, and one patient each with T cell
reprints to:
requests for
Dr Ross Pinkerton, though the marrow is hypocellular and the lymphoma, acute myeloid leukaemia, and
Department of Paediatrics, harvesting of clinically undetectable malignant angiosarcoma of the liver. High dose treatment
The Royal Marsden
Hospital, Downs Road, cells in marrow may be less likely. However, was used as consolidation of chemotherapy
Sutton, Surrey SM2 5PT. the major advantage of PBSC transplantation induced remission except for one patient with
Accepted 8 November 1993 is rapid haematopoietic reconstitution.5 In T cell lymphoma who was treated in relapse.
238 Mitchell, Shepherd, Proctor, Dainton, Cabral, Pinkerton
Table 1 Characteristics of 11 patients undergoing combined PBSC and ABMT
Case Agel
No sex Diagnosis Primary (metastases) Status* Preparation Outcomet
1 12/F Acute myeloid leukaemia CR Busulphan/melphalan FOD 24 months
2 7/F T cell lymphoma Mediastinum (bone marrow) PD Busulphan/melphalan DOD 22 days
3 17/F Ewing's sarcoma Foot (lung, lymph nodes) PR Busulphan/melphalan Toxic death 51 days
4 1 0/M Ewing's sarcoma Pelvis (lung) CR Busulphan/melphalan DOD 8 months
5 12/M Ewing's sarcoma Ileum CR Busulphan/melphalan FOD 36 months
6 16/M Rhabdomyosarcoma Pelvis (bone marrow, lymph
nodes) PR Melphalan DOD 6 months
7 9/F Ewing's sarcoma Femur (bone) CR Busulphan/melphalan FOD 17 months
8 14/M Ewing's sarcoma Pelvis (lung) CR Busulphan/melphalan FOD 18 months
9 14/M Rhabdomyosarcoma Arm (bone marrow, lymph
nodes) CR Melphalan AWD 9 months
10 16/F Angiosarcoma Liver (liver) PR Melphalan FOD 6 months
11 1 5/M Rhabdomyosarcoma Cheek Second Melphalan FOD 6 months
CR
*CR=complete remission, PD=progressive disease, PR=partial remission.
tAWD=alive with disease, DOD=died of disease, FOD=free of disease.
Patients received busulphan 16 mg/kg orally and rapidly reinfused by syringe. Patients
over four days followed by melphalan 160 received intravenous hydration during rein-
mg/M2 intravenously, or melphalan 200 mg/M2 fusion. The day ofreinfusion was designated as
given alone (table 1). day 0.
SUPPORTIVE CARE CFU-GM CULTURE
Patients were isolated in filtered air rooms, Cells were cultured in soft agar at 5 x 104
although parents had free access. No particular nucleated cells per plate to determine CFU-
procedures were followed regarding food and GM content; 10% 5637 conditioned medium
no gut sterilisation was used. Antifungal was used as growth factor. After incubation
prophylaxis comprised oral amphotericin and at 37°C for 14 days in a carbon dioxide
nystatin, or fluconazole. All patients had a incubator, the plates were scored for CFU-
Hickman line with at least two lumens. GM colonies of at least 50 cells. CFU-GM
Pneumocystis prophylaxis was commenced data were available only in those where marrow
at neutrophil recovery, with co-trimoxazole was cryopreserved.
then being continued for six months.
Patients were transfused to maintain haemo-
globin >90 g/l and platelets >20X 109/1. Results
Cytomegalovirus IgG negative patients re- PBSC harvesting was successfully carried out
ceived only cytomegalovirus negative blood in all 11 patients with no complications,
products, and all blood products were however, two patients required more than two
irradiated. collections. Most patients had PBSC harvested
at between 16 and 19 days after chemotherapy.
Table 2 outlines the levels of PBSC and
BONE MARROW AND PBSC COULECTION bone marrow progenitor cell levels reinfused,
Bone marrow was harvested and cryopreserved and the times to neutrophil and platelet
except for two patients receiving high dose recovery. Patients undergoing ABMT received
melphalan alone who were transplanted with PBSC with a median CFU-GM count of
fresh marrow. The marrow was processed 3-42X 104/kg (range 3-03-18-01). The median
to give a buffy coat concentrate and Ficoll CFU-GM count of bone marrow reinfused
separation was undertaken in some of the was 12-4X 104/kg (range 4.16-28 6).
younger children to reduce volume and red cell Reconstitution of haematopoiesis occurred
contamination. in all patients. Neutrophils reached 0 5X109/
X
Leukapheresis was carried out following at a median of 14 days (range 11-18 days),
routine myelosuppressive chemotherapy and and platelet transfusion independence was
was timed to coincide with a total white cell achieved at a median of 22 days (range 9-84
count of 2 X 109/1. It has been previously shown days). No correlation was observed between
that after such chemotherapy a total white cell days to neutrophil or platelet recovery and
count of 2x 109/1 with rapidly increasing infused levels of PBSC or bone marrow
monocyte numbers identifies the peak in nucleated or CFU-GM cells (linear regression
peripheral blood CFU-GM levels.'0 analysis for neutrophil recovery, log rank
Two leukapheresis sessions were planned. test for time to platelet transfusion independ-
The stem cell programme of the COBE ence).
Spectra cell separator was used to process Neutrophil recovery was well sustained in
1-5-2 times the estimated blood volume, four patients (cases 2, 6, 9, 11), while for two
producing about 1 mlmin of stem cell patients there was a minor dip in neutrophil
product. levels (cases 1, 4). However, the other five
PBSC and bone marrow for cryopreserva- patients showed only transient early engraft-
tion underwent controlled freezing with 5% ment, with neutrophils dropping again at day
dimethyl sulphoxide, and were stored in the 16-19 below 0 * 5 X 109/1 (figure).
vapour phase of liquid nitrogen. The cryo- One patient (case 5) with neutrophil
preserved products were thawed at the bedside recovery to 0-6X 109/l at day 14, dropped to
i. -
Peripheral blood stem cells used to augment autologous bone marrow transplantation 239
Table 2 CFU-GM levels of transplanted cells and haematological recovery
Days to recovery
Duration of chemotherapy
before harvest (months) CFU-GMX 104/kg Neutrophils Platelet Days
Case PBSC >05X109/7 transfusion in
No PBSC Bone marrow harvests PBSC Bone marrow independent hospital
1 5 5 2 6-66 28-6* 12 18 31
2 7 7 2 4-66 12 88* 14 22+ 22 (died)
3 3 4 2 18-01 18-9 14 36 51 (died)
4 3 4 2 9-79 7-6* 13 9 22
5 7 6 2 3-42* 13-6* 14 50 36
6 4 7 2 331 1053 16 25 23
7 6 3 2 303 124 17 48 23
8 1 3 3 3-38 4 16 11 11 19
9 3, 4 7 4 5 30 - 16 20 20
10 4 6 2 335 - 18 84t 58
11 3t 0t 2 3-34 94 12 15 14
*Mononuclear cell preparation using Ficoll separation.
tPlatelets remained 1C0)x 109/1 with G-CSF 5 chemotherapy. Even transient neutrophil
pug/kg/day given iintravenously from days recovery at an early stage would be advan-
19-24. However, shLe continued to be unwell tageous in reducing septic complications at a
with marked oral m ucositis and subsequently time when other treatment related toxicity
deteriorated with bacterial pneumonitis. such as mucositis and enteral toxicity are
She died of pulmoinary haemorrhage during prominent. This strategy appeared to be
artificial ventilation. successful in most of our patients.
All 11 patients Irequired broad spectrum Leukapheresis was successfully carried out
antibiotic treatment for sepsis while neutro- in these 11 paediatric patients without disrup-
penic, and all but one had World Health tion of their usual chemotherapy routine. For
Organisation grade 35-4 oral mucositis. most patients only two leukapheresis sessions
Follow up is short, but three patients with were required to collect >3*0X104/kg CFU-
Ewing's sarcoma re mained alive and free of GM.
disease at 17-36 m()nths (table 1). A further Neutrophil recovery to ¢0-0-5X109/l was
patient with Ewing's sarcoma was in complete prompt at a median of 14 days. This compares
favourably with an earlier group of similar
I
I
I'.I, i
patients receiving ABMT alone in whom the
I
median time was 18 days (range 14-27)
Case No: (unpublished data).
I:,?
2.0- I I
I, i
I
I @--
__3 Stable platelet engraftment occurred at a
1.
I@ : II
Il t7
I
i
: 5 . median of 22 days in the current group. As
expected, using bone marrow in addition to
I1
II
1.5- I
PBSC, no late graft failure was seen, but one
U-
z -
-
*
I
-.-.- 8 patient remained with platelets <20X 109/l at
./ I
I I ___-- 10 six months. However, in seven patients the
0n - early neutrophil recovery was not sustained,
0. - I
I I I
I
0 -0
I
I I
I
_ ~~~~anddropped below 0*5x109/l again in five
I I I patients due to only transient engraftment of
I the PBSC reinfusion.
Z3 I
For some of these children the addition of
0 5-
I: I I
0-5 PBSC to ABMT apparently failed to reduce
--
.1
t_ \/ */ the toxicity of the procedure, and in particular
platelet engraftment was hastened in only a
minority.
0-0- For successful transplantation using auto-
,61
I, l logous PBSC alone it has been suggested that
0 10 15 20 25 30 35 40 45 50 55 60 10-30X 104/kg CFU-GM is required in
Time (days) paediatric patients.3 However, there is con-
Patients with neutrophil counts subsequently falling below (OSX 109/. Patients receiving siderable individual variation and as little as
G-CSF were case 3 (days 19-24) and case 10 (days 10- 17, days 27-34). 0-7X 104/kg CFU-GM may be sufficient in
240 Mitchell, Shepherd, Proctor, Dainton, Cabral, Pinkerton
some cases.3 In the current study there was no likely to be little advantage to PBSC rescue
clear relationship between the stability of the compared with ABMT. High dose treatment
early neutrophil rise and the infused PBSC will be ineffective in either case and failure is
CFU-GM count, although four out of five due to chemoresistant disease in the child, not
patients with unstable early engraftment contamination of the reinfused marrow.
received 3-3-5x104/kg CFU-GM. None of
these patients had a history of bone marrow 1 Ladenstein R, Hartmann 0, Pinkerton CR. The role of
megatherapy with autologous bone marrow rescue in solid
involvement by tumour, which is known to be tumours of childhood. Ann Oncol 1993; 4 (suppl 1):
associated with slow engraftment. Likewise S45-S58.
2 Fakuda M, Kojima S, Matsumoto K, et al.
there was no clear association between the Autotransplantation of peripheral blood stem cells
duration of chemotherapy before leukapheresis mobilized by chemotherapy and recombinant human
granulocyte colony-stimulating factor in childhood
and the stability of PBSC engraftment. neuroblastoma and non-Hodgkin's lymphoma. Br J
Early platelet engraftment has been the major Haematol 1992; 80: 327-31.
3 Takaue Y. Peripheral blood stem cell autografts in children
advantage of PBSC reinfusion in adults. with acute lymphoblastic leukaemia and lymphoma:
However, in several studies patients were less updated experience. Leuk Lymphoma 1991; 3: 241-56.
4 Lowry PA, Tabbara IA. Peripheral hematopoietic stem cell
heavily pretreated than these children, a factor transplantation: current concepts. Exp Hematol 1992; 20:
likely to influence platelet engraftment in 937-42.
5 To LB, Roberts MM, Haylock DN, et al. Comparison of
particular. PBSC collections were carried out haematological recovery times and supportive care
during the rebound after chemotherapy without requirements of autologous recovery phase peripheral
blood stem cell transplants, autologous bone marrow
the use of growth factors. The use of colony transplants and allogeneic bone marrow transplants. Bone
stimulating factors before collection has now Marrow Transplant 1992; 9: 277-84.
6 Kawano Y, Takaue Y, Saito SI, et al. Granulocyte colony-
been shown to improve progenitor cell yields,"I stimulating factor (CSF), macrophage-CSF, granulocyte-
and our current practice is to utilise PBSC macrophage CSF, interleukin-3, and interleukin-6 level in
sera from children undergoing blood stem cell autografts.
obtained by combined chemotherapy and Blood 1993; 81: 856-60.
G-CSF priming for four days before and during 7 Richman CM, Weiner RS, Yankee RA. Increase in circulat-
ing stem cells following chemotherapy in man. Blood
harvest. The use of marrow growth factors add 1976; 6: 1031-9.
significantly to the cost of the procedure and 8 Craig JIO, Smith SM, Parker AC, et al. The response of
peripheral blood stem cells to standard chemotherapy for
this study demonstrates that even without their lymphoma. Leuk Lymphoma 1992; 6: 363-8.
use there is potential benefit. 9 Emminger W, Emminger-Schmidmeier W, Hocker P, et al.
The median daily increment of leukocytes during
The degree of cytoreduction achievable with haematopoietic recovery reflects the myeloid progenitor
a single course of high dose treatment is cell yield during leukapheresis in children. Bone Marrow
Transplant 1990; 5: 419-24.
limited. Although in haematopoietic malignan- 10 Fernandez JM, Shepherd V, Millar J, et al. When is the
cies this may be curative this is less likely in optimum time to harvest peripheral blood stem cells in
children following standard dose chemotherapy? Med
solid tumours. In adults PBSC reinfusion Pediatr Oncol 1993; 21: 465-9.
allows repeated sequential high dose treat- 11 Haas R, Hohaus S, Egerer G, et al. Recombinant
human granulocyte-macrophage colony-stimulating factor
ment,'3 a strategy that is feasible in the (rhGM-CSF) subsequent to chemotherapy improves
paediatric setting and could improve the poor collection of blood stem cells for autografting in patients
not eligible for bone marrow harvest. Bone Marrow
prognosis of some paediatric tumours. Transplant 1992; 9: 459-65.
The precise advantages of PBSC reinfusion 12 Gianni AM, Bregni M, Siena S, et al. Rapid and complete
haemopoietic reconstitution following combined trans-
over ABMT in children have yet to be clarified. plantation of autologous blood and bone marrow cells. A
In the young child there may be little to choose changing role for high dose chemo-radiotherapy? Hematol
Oncol 1989; 7: 139-48.
between a general anaesthetic for marrow 13 Shea TC, Mason JR, Storniolo AM, et al. Sequential cycles
harvest and several hours on a cell separator. of high-dose carboplatin administered with recombinant
human granulocyte-macrophage colony-stimulating fac-
Randomised studies evaluating engraftment tor and repeated infusions of autologous peripheral-blood
timing and procedure costs are needed. In progenitor cells: a novel and effective method for
delivering multiple courses of dose-intensive therapy.
patients with active marrow disease there is JClin Oncol 1992; 10: 464-73.