Bone Marrow Transplantation (2012) 47, 684–693
& 2012 Macmillan Publishers Limited All rights reserved 0268-3369/12
www.nature.com/bmt
ORIGINAL ARTICLE
Growth hormone treatment impact on growth rate and final height of
patients who received HSCT with TBI or/and cranial irradiation in
childhood: a report from the French Leukaemia Long-Term Follow-Up
Study (LEA)
F Isfan1,2,3, J Kanold1,2,3, E Merlin1,2,3, A Contet4, N Sirvent5, E Rochette1,2,3, M Poiree5, D Terral6,
H Carla-Malpuech6, R Reynaud7, B Pereira8, P Chastagner4, MC Simeoni9, P Auquier9, G Michel10
and F Deméocq1,2,3,10
1
CHU Clermont-Ferrand, Centre Régional de Cancérologie et Thérapie Cellulaire Pédiatrique, Hôpital Estaing, Clermont-Ferrand,
France; 2INSERM-CIC 501, Clermont-Ferrand, France; 3Clermont Université, Univ Clermont1, Faculté de Médecine, ClermontFerrand, France; 4CHU Nancy, Service d’Onco-Hématologie et Immunologie Pédiatrique—Hôpital Brabois, Nancy, France; 5CHU
Nice, Service d’Hématologie Oncologie Pédiatrique, Hôpital Archet, Nice, France; 6CHU Clermont-Ferrand, Pédiatrie Générale et
Multidisciplinaire, Hôpital Estaing, Clermont-Ferrand, France; 7CHU Marseille, Service de Pédiatrie Multidisciplinaire—Hôpital
d’Enfants de la Timone, Marseille, France; 8CHU Clermont-Ferrand, DRCI, Biostatistics Unit, Clermont-Ferrand, France;
9
Marseille Université, Faculté de Médecine, Service de Santé Publique, Marseille, France and 10CHU Marseille, Service de Pédiatrie
et d’Hématologie Pédiatrique—Hôpital d’Enfants de la Timone, Marseille, France
The literature contains a substantial amount of information about factors that adversely influence the linear
growth in up to 85% of patients undergoing haematopoietic SCT (HSCT) with TBI and/or cranial irradiation
(CI) for acute leukaemia (AL). By contrast, only a few
studies have evaluated the impact of growth hormone
(GH) therapy on growth rate and final height (FH) in
these children. We evaluated growth rates during the preand post-transplant periods to FH in a group of 25
children treated with HSCT (n ¼ 22), TBI (n ¼ 21) or/and
CI (n ¼ 8) for AL and receiving GH therapy. At the start
of GH treatment, the median height Z-score was 2.19
(3.95 to 0.02), significantly lower than at AL diagnosis
(Po0.001). Overall height gain from start of GH
treatment to FH was 0.59Z (2.72 to 2.93) with a
median height Z-score at FH of 1.35 (5.35 to 0.27).
This overall height gain effect was greater in girls than in
boys (P ¼ 0.04). The number of children with heights in
the reference population range was greater after than
before GH therapy (P ¼ 0.07). At FH the GVHD and GH
treatments lasting o2 years were associated with shorter
FH (P ¼ 0.02 and 0.05). We found a measurable beneficial
effect of GH treatment on growth up to FH.
Bone Marrow Transplantation (2012) 47, 684– 693;
doi:10.1038/bmt.2011.139; published online 4 July 2011
Correspondence: Professor J Kanold, Pédiatrie, Centre Régional de
Cancérologie et de Thérapie Cellulaire Pédiatrique, Hôpital Estaing,
C.H.U., B.P. 69, 1, place Lucie-Aubrac, 63003 Clermont-Ferrand,
France.
E-mail: [email protected]
Received 14 February 2011; revised 18 May 2011; accepted 26 May 2011;
published online 4 July 2011
Keywords:
irradiation
height; growth hormone; children; HSCT;
Introduction
The growth and growth hormone (GH) deficiency have
been observed in children after haematopoietic SCT
(HSCT) with TBI1–5 and those previously undergoing
central nervous system irradiation (CI) to treat acute
leukaemia (AL).1–3,5–7 Conditioning regimen chemotherapy
and irradiation are often superimposed on the ‘standard’
therapies, which may compromise pubertal development
and final height (FH) in their own right.8 Also, with the
increasing number of survivors and duration of follow-up,
patients may experience a significant loss of height
potential. The literature contains a number of studies
about factors that adversely affect growth in children
treated for AL (age at the time of AL treatment, irradiation
dose and fraction size, and so on).3,6,9–14 By contrast, only a
few studies have evaluated the impact of GH replacement
therapy on growth rate and FH in these children.1,2,7,15–17 A
better understanding of the factors influencing response to
GH will allow the identification of patients who can benefit
from appropriate hormonal replacement therapy. This is
important given that currently fewer than half of the GH
deficiency (GHD) children after HSCT or CI receive GH
therapy.6,18
We evaluated growth rates during the pre- and posttransplant periods to FH in a group of 25 children treated
with HSCT, TBI or/and CI for AL and subsequently
receiving supplemental GH therapy. The impact of GH
treatment on FH was of particular interest.
GH treatment and final height after HSCT
F Isfan et al
685
Patients and methods
Study design
This study was based on an analysis of growth, pubertal
development and hormonal status data collected prospectively as part of an ongoing French long-term follow-up
programme for patients treated for AL in childhood.
Details of the whole programme, known as ‘LEA’ (for
‘Leucémie de l’Enfant et de l’Adolescent’) have already
been described elsewhere.19 Briefly, this programme was
initiated in 2003 to make a prospective evaluation of the
long-term health status, quality of life and socioeconomic
status of childhood leukaemia survivors enrolled in
treatment from 1980 to the present in the participating
centres. At the end of 2009, the cohort comprised 944
patients. In application of French law, this long-term
follow-up programme was filed with the Comité consultatif
sur le traitement de l’information en matie`re de recherche
dans le domaine de la santé (French data privacy authority
governing healthcare research) and approved by the CNIL
(French national data privacy authority). Written informed
consent was obtained from the patients and their parents or
legal guardians for programme participation. In the entire
cohort of 944 patients there were 257 patients who had had
HSCT. Among these 188 had had TBI (14 had also had
previous CI) and 12 had had only CI. Among no
transplanted patients 121 had had CI. It makes a total of
321 irradiated patients. In the entire cohort of 944 patients
there were 54 patients with identified GHD. A total of 35
patients were treated with GH and 25 had achieved FH at
the time of their last visit. This study was restricted to those
patients who met all the following inclusion criteria: (1) had
undergone HSCT, including TBI or/and cranial irradiation
(CI), (2) had GHD and had received supplemental GH
therapy and (3) had achieved FH at the time of their last
visit. Of the 944 children included in the LEA programme
25 met study entry criteria and are described here.
Patients
Twenty-five children with AL (21 lymphoblastic and
4 myeloid) diagnosed at a median age of 4.4 years (range
1.9–12.3) underwent treatment according to the French
multicenter protocols applicable at the time of diagnosis
(from June 1980 to March 2001). In addition to multiagent
chemotherapy, eight children had received CI at target
doses ranging from 18 to 24 Gy (including the spine in
seven patients). In six children the CI was given as
‘prophylaxis’ in the first 6 months of treatment (in four
of them it was years before HSCT/TBI administration).
In two children CI was given as treatment for a CNS
relapse 4 and 6 years, respectively, after AL diagnosis (in
one of them CI was given shortly before HSCT/TBI). Three
children were given CI only without HSCT/TBI (Table 1).
Twenty-two children had received HSCT at a median age
of 6.7 years (range 2.4–12.7). Seventeen patients underwent
HSCT/TBI before and 5 after the age of 10 years. Seven
patients underwent HSCT/TBI in first and 15 in second or
subsequent CR. The median times from AL diagnosis to
HSCT/TBI were 0.38 years (range 0.31–0.42) for patients
in first CR and 2.17 years (range 0.96–4.46) for those in
second or subsequent CR. HSCT preparative regimens
depended on diagnosis, protocols in use at the time of
transplantation, phase of patient disease and donor type.
Patients were conditioned with CY (n ¼ 3), melphalan þ /
cytarabine (n ¼ 18) plus TBI (n ¼ 21).20 TBI was given as 2Gy doses twice a day for a total of six doses (n ¼ 17), five
doses (n ¼ 2) or seven doses (n ¼ 1). One patient received a
total of 6 Gy in six fractions. One patient (number 21)
received misulban/melphalan conditioning regimen without
TBI. Two boys (Nos 2 and 10) received, respectively, 16
and 24 Gy testicular irradiation as treatment for testicular
leukaemia and five others (Nos 1, 3, 7, 12 and 16) 4–6 Gy
testicular boost irradiation at the time of TBI. Eight
children were treated with autologous and 14 with
allogeneic HSCT. Thirteen patients developed GVHD
graded as described previously.21 Acute GVHD grade II
or more occurred in two children and extensive chronic
GVHD in three children. All of them were given steroid
therapy. Median patient age at the last time point was 22.5
years (range 18.5–36.0).
Methods
Height evaluation. Height was measured before and
during AL treatment and at least annually thereafter until
18 years of age or more. Height was measured using a
Harpenden stadiometer (Holtain Ltd, Crymych, UK) and
was expressed as height-for-age-and-sex Z-score, calculated
as height minus mean height for age and sex divided by the
s.d. of height for age and sex. This Z-score value allows
data from boys and girls to be compared without
distinction. Height-for-age-and-sex Z-scores were calculated using the French sequential study of growth. These
values were compared with the healthy reference population used in the French growth charts, which have normal
reference values for stature by age and sex for children aged
0–18 years.22 A normal range is from 2.00Z to 2.00Z, a
negative value indicates below-age-expected height, and a
positive value indicates above-age-expected height.
Height-for-age-and-sex Z-scores were calculated at four
key time points: AL diagnosis (Z1), first irradiation (HSCT/
TBI or CI, whichever came first; Z2), start of GH treatment
(Z4) and FH (Z6). The intermediate time points representing the HSCT/TBI (Z2a), CI (Z2b), GHD diagnosis (Z3) and
end of GH treatment (Z5) were also analysed. Height at
each time point was defined by its associated height-forage-and-sex Z-score. Changes in height-for-age-and-sex
Z-scores between key time points were also calculated (DZ).
FH was defined as the tallest height measured at patient
age 18 years or older, and when height velocity was o1 cm/
year. For three patients who were older than 18 years
when they started their annual follow-up in the LEA longterm follow-up programme and for whom we did not
have the 18th birthday height but only programme
inclusion height we arbitrarily took their 20th birthday as
the time of FH.
GH testing. In LEA programme, all children with a fall in
growth velocity were screened for GHD. All patients are
given an annual evaluation of basal plasma insulin-like
growth factor-I concentration. GH secretion in response to
Bone Marrow Transplantation
Bone Marrow Transplantation
M
F
F
M
M
F
M
F
M
F
M
F
M
F
F
M
M
F
F
F
M
F
F
F
M
3
15
19
17
1
9
8
14
10
22
2
23
24
6
13
7
12
5
20
18
16
4
11
21
25
Median
4.9
11.8
4.1
12.3
2.0
5.6
7.6
4.4
3.4
3.1
7.0
2.3
3.9
8.1
2.4
8.6
2.6
3.5
4.4
1.9
5.8
7.8
6.4
5.5
2.9
4.4
Age
(years)
1.95
0.02
1.00
1.25
0.41
1.27
0.60
1.86
3.43
2.00
1.78
0.55
0.97
0.65
0.59
0.77
0.03
0.65
0.68
0.00
0.16
0.53
0.39
0.44
0.59
0.16
Height
Z-score
115
146
104
140
86
115
120
109
110
100
111
85
104
127
90
124
91
98
100
83.5
113
120
113
108
95
109
Height
(cm)
Diagnosis
ALL
AML
ALL
ALL
ALL
ALL
ALL
ALL
ALL
ALL
ALL
ALL
ALL
AML
AML
ALL
ALL
ALL
ALL
ALL
ALL
ALL
AML
ALL
ALL
a
–
–
–
–
–
–
–
4.7P24
3.7P18
3.3P24
7.2P24
2.5P24
4.1P30
–
–
–
–
–
–
–
–
–
–
11.6R24
6.8R24
4.4
Age
(years)
CI
–
–
–
–
–
–
–
1.39
3.22
1.59
1.88
0.97
0.69
–
–
–
–
–
–
–
–
–
–
1.35
1.00
0.14
Height
Z-score
5.2
12.1
4.5
12.7
2.4
6.0
8.0
6.5
6.5
6.3
11.4
–
–
9.0
3.4
9.6
3.8
5.0
6.9
4.8
8.7
10.9
9.9
11.8
–
6.7
Age
(years)
1.67
0.36
0.89
1.33
0.90
1.12
0.65
0.30
2.54
0.42
2.10
–
–
0.40
2.29
1.04
0.97
0.05
1.07
0.46
0.29
1.02
0.71
1.68
–
0.32
Height
Z-score
Age (Bone age)
(years)
9.5 (10.5)
14.7 (11.0)
10.4 (10.8)
15.9
7.3 (7.0)
12.8 (11.5)
15.1 (13.0)
13.4 (11.0)
14.2 (14.5)
15.2 (11.0)
12.6 (11.5)
10.6 (8.0)
10.8 (10.0)
12.7
11.1
13.5 (12.5)
15.0 (14)
12.6 (12)
13.2 (9)
9.1 (7.5)
14.2
16.2
14.0 (12.0)
13.7 (11)
9.3 (7.0)
13.2 (11.0)
allo/MRCR1
allo/MisRfirst line
autoCR1
alloCBCR1
autoCR1
autoCR1
autoCR1
allo/MRCR2
autoCR2
allo/MRCR2
autoCR2
–
–
allo/MUDCR2
allo/MUDCR2
allo/MUDCR2
allo/MRCR2
allo/MUDCR2
allo/MRCR2
allo/MRCR2
allo/MRCR2
allo/MRCR2
autoCR2
autoCR2
–
0.02
2.19
1.08
3.86
1.84
1.15
1.81
2.88
0.65
3.53
2.45
2.37
1.30
2.89
1.44
2.63
1.75
2.35
3.95
2.08
1.60
2.41
2.53
2.79
1.64
2.19
Height
Z-score
GH treatment start
Type/donor
status
HSCT/TBI
5.8
2.7
5.5
1.0
8.9
1.6
3.0
3.1
2.5
3.3
5.0
4.9
5.9
3.9
4.1
2.0
1.0
4.0
3.7
6.4
3.7
1.1
1.8
3.6
7.5
3.7
GH treatment
duration (years)
1.65
0.04
1.46
4.23
1.07
1.46
0.40
2.36
0.27
1.20
2.90
0.57
0.57
1.64
1.18
5.35
3.32
1.82
1.02
2.27
0.40
1.82
1.11
0.04
1.35
1.35
Height
Z-score
165
163
155
149
168
155
172
150
176
157
157
160
171
154
156
142
155
153
158
151
172
153
157
163
166
157
Height
(cm)
Final
Abbreviations: ALs ¼ acute leukaemias; CI ¼ craniospinal irradiation; FH ¼ final height; GH ¼ growth hormone; HSCT ¼ hematopoietic SCT; MisR ¼ mismatched related; MR matched related;
MUD ¼ matched unrelated donor.
P
prophylactic CI; R for relapse CI, followed by nos. representing CI doses (Gy).
a
Cranial irradiation only.
Sex
Clinical and statural data of 25 children treated for ALs in childhood and who subsequently received supplemental GH therapy
Patient no.
Table 1
GH treatment and final height after HSCT
F Isfan et al
686
GH treatment and final height after HSCT
F Isfan et al
687
GH supplemental therapy. All the patients received GH
treatment with recombinant human GH at 0.4–0.7 IU/kg/
week, administered as s.c. injection 4–7 days every
week, corresponding to 0.033 mg/kg body wt/day (0.1 IU/kg
body wt/day).
Statistical analysis. All data are reported as medians
(range). The characteristics of patients by age, sex,
diagnosis, transplant and irradiation type were compared
in univariate analysis using the Kruskal–Wallis test for
continuous variables and Fisher’s exact test for categorical
factors. Growth in these patients was quantified by plotting
the median height Z-scores at each key time point. Height
Z-scores were compared across patient subgroups using
two-way repeated measures analysis of variance (and the
Friedman test when appropriate) followed by a Tukey–
Krammer test to compare differences between and within
groups. This analysis was completed with linear mixed
models (multivariate analysis) to assess the impact of AL
treatment and GH therapy on growth over time. Change in
height Z-score from AL to GHD diagnosis and from GH
treatment start to FH was evaluated. Factors considered as
possibly contributing to this change in height Z-score were
also examined. These factors included age at different time
points, sex, primary diagnosis, exposure to CI and/or
HSCT, type of HSCT, GVHD and duration of GH
treatment. The effect of GH therapy on growth was then
tested, after adjusting for patient characteristics. Interactive
effects with GH therapy were also investigated. Linear
mixed models allow the comparison of change in height
Z-scores between covariate groups over time. The parameters in this regression model were estimated by the
method of maximum likelihood and the Wald statistic was
used to test the equality of growth rates. The statistical
analyses were performed using the STATA statistical
software, version 10.0 (StataCorp, College Station, TX,
USA); Po0.05 was considered statistically significant.
Results
Growth before GH therapy
Figure 1 shows the median height Z-scores over time for all
the patients. At the time of AL diagnosis (Z1), heights were
normal in 24 of the 25 patients (96%) and one patient was
taller than the normal reference population. The patients
older than 5 years at diagnosis were significantly shorter
than those who were younger than 5 (median Z-score
0.44 vs 0.62, respectively; P ¼ 0.01).
When the AL treatment started (Z1) height Z-scores
started to decrease. At the time of first irradiation (Z2)
median Z-score was 0.36 (range 1.88 to 3.22, not
significantly different from Z1, P ¼ 0.34). However, at the
time of GHD diagnosis (Z3) it was 2.12 (range 3.98 to
0.5, significantly lower than Z1, Po0.001).
Median loss in height Z-score between AL diagnosis and
first irradiation (DZ2Z1) was 0.25 (range 1.59 to 1.75).
In those patients who were irradiated in first CR, median
height was nearly the same at first irradiation (HSCT/TBI
or CI) as at AL diagnosis (n ¼ 13, median DZ2Z1 ¼
0.28). However, it was clearly decreased in those
irradiated in second or subsequent CR (n ¼ 12, median
DZ2Z1 ¼ 0.45). At the time of first irradiation normal
height was still seen in 23/25 children (92%), and the two
remaining patients were still taller than the normal
reference population.
Median loss in height Z-score between HSCT/TBI or CI
and GHD diagnosis (DZ3Z2) was 1.73 (range 5.63 to
0.25) and was significantly more marked than the loss
observed from AL diagnosis to first irradiation (DZ2Z1).
However, considering median Z-score decrease rate (Z loss
per year) there was no difference before (0.38 Z/year,
range 3.41 to 1.78) or after (0.39 Z/year, range 1.01 to
0.07) the first irradiation (HSCT/TBI or CI; P ¼ 0.44).
Hence, the overall growth rate decrease from AL diagnosis
Height-for-age Z-score
pharmacological stimuli was evaluated after documented
growth failure, namely a loss of growth rate 42 s.d. or if
insulin-like growth factor-I level was o2 s.d. according to
age and sex. Patients showing slower growth rate but with
normal or borderline GH secretion underwent repeat tests
after 6 months until a definite result was obtained.
Spontaneous production and peak responses to at least
two stimulation tests were considered for each patient.
These tests were insulin tolerance tests (0.05 UI/kg i.v.),
GH-releasing hormone infusion tests (80 mg, Somatoreline
i.v., Choay/Sanofi, Gentilly, France) or propanolol glucagon tests (0.25 mg/kg oral propanolol and 1 mg glucagon
i.m.). Blood was sampled every 30 min and a secretion
curve was plotted.
The cutoff level for diagnosis of GHD was considered
to correspond to a peak level after stimulation o20 mUI/L
(10 ng/mL). The tests were performed at least 6 months
away from any antileukaemic treatment.
All patients were also tested for thyroid function.
Luteinizing hormone, follicle-stimulating hormone and
either estradiol or testosterone were assayed.
6
5
4
3
2
1
0
-1
-2
-3
-4
-5
-6
Z1 Z2
0
Z3
Z4
Z5
Z6
5
10
15
Time from AL diagnosis (years)
Figure 1 Height-for-age Z-scores over time in 25 children treated with
HSCT/TBI or/and CI and who subsequently received supplemental GH
therapy. The boxes represent the 25th and 75th quartiles; the band near the
middle of the box is the median. Whiskers extend to the upper and lower
adjacent values that are the 9th percentile and the 91st percentile. Data not
included between the whiskers are plotted. Six key time points (median
time) are represented: Z1, AL diagnosis (the last untreated measurement);
Z2, HSCT/TBI or CI whichever occurred first; Z3, GHD diagnosis; Z4, GH
treatment start; Z5, GH treatment end; Z6, FH.
Bone Marrow Transplantation
GH treatment and final height after HSCT
F Isfan et al
688
Height-for-age Z-score
Growth during and after GH therapy
Patients were diagnosed with GHD at a median of 12.4
years (range 6.8–15.6) and 6.2 years (range 2.1–11.7) after
AL diagnosis and 4.6 years (range 1.7–11.6) after first
irradiation (HSCT/TBI or CI). The median time from
GHD diagnosis to start of GH therapy was 0.6 years (range
0.2–3.4 years).
At the time patients started GH treatment, chronological
age was 13.2 years (range 7.3–16.2 years) and bone age,
available in 20 of the 25 patients (80%), was 11 years (range
7.0–14.5 years). Bone age was within 2 years of chronological age for 12 patients and was 42 years lower for age
in 8 patients. Median Z-score (Z4) was 2.19 (range 3.95
to 0.02), significantly oZ1 (Po0.001). Fourteen children
(56%) had heights 42Z below the mean at GH treatment
start. The patients older than 10 years at the time of GH
treatment start were shorter than those younger than 10
(P ¼ 0.01).
GH treatment lasted for a median 3.7 years (range 1.0–
8.9). Median Z-score recovery during GH treatment
(DZ5Z4) was 0.43 (range 1.10 to 2.91) and the median
recovery rate was 0.12 Z/year (range 1.31 to 0.78).
In 17 children height Z-scores increased whereas in 8
children it continued to decrease after the start of GH
treatment. Similar proportions of patients in the ‘responding’ group and in the ‘nonresponding’ group received CI,
allo-HSCT and had steroid therapy as treatment for
GVHD.
After discontinuation of GH treatment Z-score increase
was observed in 18 children (catch-up continued in 14 and
started in the other 4). Median recovery rate was not
significantly slower after GH treatment discontinuation:
0.03 Z/year (range 0.29 to 0.41) than ‘on GH treatment’
6
5
4
3
2
1
0
-1
-2
-3
-4
-5
-6
Z1 Z2
No GVHD
0
Figure 2
Z3 Z4
Z5
GVHD
5
10
15
Time from AL diagnosis (years)
Z6
(P ¼ 0.17). Overall recovery from start of GH treatment to
FH (DZ6Z4) was 0.59Z (range 2.72 to 2.93). This overall
recovery effect (DZ6Z4) was stronger in girls (P ¼ 0.04;
Figure 2).
Patients who received allo-HSCT/TBI and patients who
had GVHD tended to have lower height Z-scores at the
time of GH deficit diagnosis (Z3). This pattern persisted in
time, and the difference became statistically significant
from the start of GH treatment (Z4) until FH (Z6; P ¼ 0.05
for HSCT/TBI and 0.03 for GVHD; Figure 2).
FH
The median cumulative change in height Z-score from AL
diagnosis to FH was 1.94. (range 4.58 to 0.40) with
median Z-score at FH of 1.35 (range 5.35 to 0.27). After
GH supplemental treatment FH were in the reference
population range in 19 of the 25 patients (76%), whereas
six children had a FH that remained below the 2Z-score.
However, the proportion of children with heights in the
reference population range was greater after than before
GH therapy (19/25 vs 11/25, P ¼ 0.07).
At FH, the GVHD and GH treatment lasting o2 years
were significantly associated with a shorter FH in
univariate analysis (P ¼ 0.02 and 0.05, respectively).
In the multivariate analysis, there was no covariate
significantly associated with a lower FH.
Toxicity after GH therapy
None of the GH-treated patients developed any adverse
effects requiring discontinuation of GH. None had a
relapse. In all, 8 secondary neoplasms occurred in seven
patients during GH treatment (two patients) or after GH
treatment had been discontinued (five patients) for 4–16
years. Among these seven patients, three developed
papillary thyroid carcinoma, one papillary thyroid carcinoma and malignant peripheral nerve sheath tumour, one
renal carcinoma, one melanoma and one basal cell
epithelioma. In two patients who had thyroid carcinoma
while on GH treatment the treatment was not stopped.
Other complications or late effects observed after GH
treatment were exostosis in one patient and benign
schwannoma in one patient (Table 2).
Height-for-age Z-score
to GH treatment start was 0.38 Z/year (range 0.89 to
0.08). At the time of GHD diagnosis only 12/25 (48%)
patients had normal heights and none was taller than the
normal reference population.
In univariate analysis, height loss from AL diagnosis to
GHD diagnosis (DZ3Z1) was greater in the girls
(P ¼ 0.007; Figure 2) and in patients younger than 5 years
at the time of CI (P ¼ 0.02). There was no significant
correlation between HSCT/TBI, GVHD and height loss
from AL diagnosis to GHD diagnosis (DZ3Z1).
6
5
4
3
2
1
0
-1
-2
-3
-4
-5
-6
Z1 Z2
Z3 Z4
Boys
0
Z5
Z6
Girls
10
5
15
Time from AL diagnosis (years)
Height-for-age Z-scores over time in 25 children treated with HSCT/TBI or/and CI and who subsequently received supplemental GH therapy.
Six key time points: see Figure 1. Height loss from AL diagnosis to GHD diagnosis (DZ3Z1) was greater (P ¼ 0.007) and treatment effect (DZ6Z4) was
stronger in girls (P ¼ 0.04). Height Z-scores were significantly lowers from points Z4 to FH in patients with GVHD (P ¼ 0.034).
Bone Marrow Transplantation
GH treatment and final height after HSCT
F Isfan et al
689
Table 2
AL relapses and second malignancies occurring on and after GH treatment
Relapse
CCSS, Sklar et al.26
GH pts
Leung et al.17
GH pts
Control
Bakker et al.15
GH pts
Control
Sanders et al.
GH pts
Control
This study
GH pts
?
Osteochondroma/
exostoses
Second malignancy
?
15/361 (4.1%)
0/43 0%
?
2/43 (4.6%)
8/544 (1.4%)
?
16/544 (2.9%)
1/23 (4.3%)
7/23 (30.4%)
2/23 (8.6%)
6/43 (14%)
2/43 (4.6%)
1/43 (2.3%)
12/42 (28.5%)
6/42 (14.3%)
1/48 (2.1%)
5/48 (10.4%)
8/48 (16.6%)
0/25 0%
1/25 (4%)
7/25a (28%)
GH treatment not associated with an increased risk for relapse of the original
malignancy. No. secondary malignancies was significantly higher than expected
and was significantly associated with the administration of GH
One Sclerosing sweat duct ca of the scalp (4 months post-GH),
one myelodysplastic syndrome (2 months post-GH)
?
One Osteosarcoma (5.3 years after GH start), one papillary thyroid ca
(3.5 years after GH start)
Schwannoma
1
0/42 0%
4 pts on GH, 1 pt 3.3 years post-GH, 1 pt 9 years post-GH two brain tumor, one
basal cell ca, one sarcoma, one squamous ca, one thyroid ca
One brain tumor, three basal cell ca, one sarcoma, one squamous ca, one
thyroid ca, one lymphoma
2 pts on GH, 5 pts 4–16 years post-GH four thyroid ca, one MPNST, one renal
ca, one melanoma, one basal cell epithelioma, one benign schwannoma
Abbreviations: AL ¼ acute leukemia; CCSS ¼ Childhood Cancer Survivor Study; ca ¼ carcinoma; GH ¼ growth hormone; MPNST ¼ malignant peripheral
nerve sheath tumor.
a
8 s malignancies in 7 pts.
Discussion
Although the majority of studies on the long-term effect of
AL treatment during childhood have examined the
leukaemia-treatment-related factors that have a role in
the FH outcome (HSCT, TBI, CI), little is known about the
efficacy or adverse effects of GH replacement therapy in
these children.
In this study, data on 25 patients with AL, who had
HSCT/TBI or CI, and who had reached their FHs, were
analysed. An overall decrease of 1.94 in FH Z-score value
was found from AL diagnosis to FH. The height loss in
our patients was comparable to that reported in the few
studies published to date and reporting FH after GH
treatment.1,2,6,7,23
Height loss before start of GH treatment
A decrease in the rate of growth in children after HSCT/
TBI or CI seems to be the result of an interaction of
different factors related to the effects of chemotherapy,
irradiation and steroid therapy (for example, damage to the
epiphyseal growth plate) as well as multiple endocrinological causes (GHD, delayed or early puberty and hypothyroidism).
Before the onset of GH therapy, a significant decrease in
height Z-score (2.19) was observed in our study, slightly
more than reported in previous studies, where s.d.s. ranged
from 1.2 to 21,2,16,17,23,24 (Table 3). In our study height
loss from AL diagnosis to GHD diagnosis was greater in
girls and in patients younger than 5 years at the time of
cranial irradiation. This is in line with other data indicating
that girls17,23 and younger patients1,7 lose more s.d.s./Z
before the start of GH treatment. Conversely, in the study
of Sanders et al.,1 boys were shorter than girls at the last
evaluation before GH therapy.
Response to GH treatment
Only few data are available on the response to substitution
GH therapy on growth in heavily treated children with AL
(Table 3). There are two types of studies: (1) those
comparing height in treated vs untreated groups of patients
and (2) those seeking factors affecting response to GH
substitution in treated patients.
Besides problems linked to the fact that these are all
retrospective, nonrandomised and small-series studies, two
major difficulties arise in discussing these earlier findings.
First, GHD diagnostic criteria differ among studies and not
all the treated patients had biochemically proven GHD
Second, the changes in height are more often evaluated
from the therapeutic key points (HSCT or TBI or CI) to
FH rather than from start of GH treatment to FH. When
no information is available on the changes in height after
start of GH therapy no valid conclusions can be drawn
regarding the effect of GH therapy.
GH was generally administered at 20–30 mg/kg/day
(0.4–0.7 IU/kg/week). Treatment duration was 3–4 years
in most studies6,7,15,16 with 44 years17,23 or o1 year25 of
GH substitution in some others. Treatment duration was
found to be a predictive factor for height loss in the study
of Frisk et al.7 In our study in which GH substitution was
stopped when growth velocity was o1 cm/year (it was
assumed that there would be no further benefit from
Bone Marrow Transplantation
GH treatment and final height after HSCT
F Isfan et al
Bone Marrow Transplantation
Abbreviations: F ¼ female; FH ¼ final height; GH ¼ growth hormone; HSCT ¼ hematopoietic SCT; M ¼ male; s.d.s. ¼ standard deviation score.
1.35
1.35 M, 1.33 F
+0.59
2.19
1.75 M, 2.39 F
25
Present study
3.8
10
Frisk et al.7
4.5
47
Leung et al.17
3.7
?
Es.d.s. at ALL
diagnosis
1.6
1.9 M, 1.4 F
(E+1.2)
2.2 M, 1.75 F
?
?
0.06
1.3 M, 2 F
1.6
1.5 M, 1.3 F
1.2
F shorter than M
?
7
42
Dai et al.23
Sanders et al.1
6
4.5
7
Chemaitilly
et al.16
3.3
(E0)
+0.1
3.2
11
1.5
0.75 M, 1.7 F
2.0
1.7
2.0 M, 1.6 F
2.0
The height loss seems to be reduced by the GH treatment when the TBI is
performed between 4 and 8 years
Positive effect of GH therapy on height s.d.s. The estimated effect of GH therapy
was +1.1 s.d.s. after 5 years with no significant differences between sexes
Treatment with GH maintained height s.d.s. from the onset of GH therapy until
attainment of FH. The change in height from TBI to FH for individuals treated
with GH was less impaired compared with that observed in those not treated with
the difference did not achieve statistical significance
No evidence of a real catch-up in growth was observed following GH treatment
GH therapy was associated with significantly improved final height in children
younger than 10 years at HSCT, but did not impact the growth of older children
The s.d. scores for adult height improved and approached the height s.d. scores at
the time of diagnosis of ALL
Beneficial effect of GH treatment on growth up to FH. There was any significant
difference in height loss after HSCT between those who had received GH and those
who had not P ¼ 0.22
?
6
Couto-Silva
et al.24
Bakker et al.15
27
1.2
2.5 M, 3.4 F
Author’s conclusions or comments
Z-score or
s.d.s. at FH
DZ-score or s.d.s.
from GH treatment
start to FH
Height Z-score
or s.d.s. at GH
treatment start
GH treatment
duration
(years)
No. of
pts
References
Table 3
Changes in Z-score or s.d. score for height between start of GH treatment and FH reported in seven studies
690
continuing the treatment in these patients), there seemed to
be a trend towards better height growth in children treated
for longer than 2 years.
The response to GH treatment varies. In our study there
was an ‘on-treatment’ increase in height in 68% vs 85% of
patients in the study of Bakker et al15 (during the first year
of treatment).
Like ours, most studies on the effect of GH therapy after
AL treatment have shown an increase in growth rate after GH
treatment was started compared with the pre-GH-treatment
rate. Observed growth recovery rates during GH treatment
were comparable in our study to that in Frisk et al.7 ( þ 0.12 Z
and þ 0.18 s.d.s./year) and estimated to be greater in Bakker
et al.15 ( þ 0.35 s.d.s. in the first year of treatment).
Generally three scenarios are observed in patients with
replacement GH therapy: (1) significant catch-up growth, for
example, up to þ 1.2 s.d.s. in Leung et al.17 and þ 0.59Z in
our study; (2) a maintained height s.d.s./Z-score15,16,23 and
(3) ‘only’ a height s.d.s./Z-score loss reduction.1,2,24
In our study, the overall recovery effect from start of GH
therapy to FH was stronger in girls. The similar pattern of
recovery was observed by Dai23 in a small series of seven
patients. This observation is also in agreement with Sanders
and Cohen, who showed that girls had a better height
growth from GHD diagnosis to FH than boys.1,6 Conversely, for Frisk et al.7 there was no significant difference
in height loss after BMT between boys and girls. The
differences in height growth between girls and boys are not
readily explained, suggesting that the reliability of this
finding should be verified in specific studies.
An interaction of the GH treatment effect with age is
expected. A significant interaction of the GH treatment
effect with age at HSCT was observed by Sanders et al.1
GH treatment resulted in 0.86 s.d. increase in FH in
children with HSCT before the age of 10 years, whereas
no significant effect of GH therapy was found in children
receiving HSCT after the age of 10 years.
In our study we failed to confirm this. However, it should
be emphasised that patient age at GH treatment is directly
linked to patient age at AL treatment. The younger patients
are at the time of AL treatment (growth failure), the
younger they can be expected to be at the time of GH
treatment (growth recovery). Therefore, younger patients
are at greater risk of growth damage, but can also be
expected to show better response to GH therapy and gain
an additional s.d. or Z-score. This height gain could offset
height loss so that the younger children who received GH
would have a relative FH similar to the older children.
Nevertheless, neither we could not act on patient age at
diagnosis nor on the damage made by AL treatment, the
point is not to lose time at initiation of GH treatment.
In future studies, with a larger cohort of patients
(continuing LEA programme ) and more detailed information of stage of puberty we may be able to define more
precisely the subset of ‘older’ patients for whom GH
substitution is still of great benefit.
FH
Data on FH achieved after HSCT/TBI/CI in response to
GH therapy are very limited. In our study the median FH
GH treatment and final height after HSCT
F Isfan et al
691
was 1.35Z after an increase of 0.59Z from start of GH
treatment to FH. Also, at FH the number of children with
height in the reference population range was greater than
before GH therapy (76% vs 44%). Therefore, we can
conclude that there is a positive effect of GH treatment on
FH. A similar conclusion was reached by Bakker et al.15 (23
patients), with an estimated effect of GH therapy of
þ 1.1 s.d. after 5 years. In Leung et al.17 (47 patients) after a
median of 4.5 years of GH therapy there was a significant
catch-up growth of about þ 1 s.d., resulting in adult heights
in the normal range for most of the patients.
Four studies have reported on the impact of GH
treatment on FH by comparing patients who had received
GH and those who had not. In the first study by Cohen
et al.6 FH in 28 patients treated with GH was not statistically
different from that of the remaining 153 untreated patients.
Also, 77% of patients (treated and untreated) who attained
their FHs reached normal heights. In the second study by
Frisk et al.,7 although at FH there was no significant
difference in height loss between those who had received GH
(10 patients) and those who had not (6 patients), the authors
support the use of early GH treatment in children with
decelerating growth rate and low GH levels. In the third
study by Sanders et al.,1 GH-treated patients (42 patients)
lost 0.06 s.d., whereas untreated ones (48 patients) lost
0.53 s.d. from GHD diagnosis to FH. In a fourth study by
Chemaitilly et al.16 the change in height for individuals
treated with GH (seven patients) was less impaired than that
observed in those not treated (five patients).
Toxicity of GH treatment
Overall incidence of second malignancies in LEA cohort
was 3.6 and 8.9% in LEA HSCT patients. It should be
noted that all patients who entered this long-term follow-up
programme benefited from better screening and reporting
than those who did not. In our study, 7 of 25 patients
(28%) developed a second malignancy during or after the
GH treatment had been discontinued. As in other studies
we found that of these, papillary thyroid carcinoma was
one of the most common. (Table 2)
Published data from a Childhood Cancer Survivors
Study cohort show that patients treated with GH had a
threefold increased risk of developing a second neoplasm
compared with untreated survivors.26,27 Among 361 survivors of childhood cancer who had GH treatment, 15 had
second malignancies (4.1%). In the 119 AL survivors
treated with GH, 4 developed a second solid neoplasm
(3.4%): 2 osteosarcoma, 1 astrocytoma and 1 glioma. No
case of secondary leukemia was described.27
In the study of Sanders et al,1 6/42 GH-treated patients
(14%) developed a second neoplasm, 4 during and 2 after
the completion of GH therapy. In the non-GH-treated
group the incidence of secondary malignancy was 16.6%
(8/48 patients). The types of malignancies observed were
similar to those recorded after childhood transplantation.
There was no significantly higher incidence of second
neoplasm in a treated vs untreated group.
In 43 children with AL diagnosis and GH treatment Leung
et al.17 described two patients who developed a secondary
malignancy (a sclerosing sweat duct carcinoma of the scalp
and a myelodysplastic syndrome). There was no statistical
evidence in this study that GH replacement therapy was
associated with increased risk of secondary malignancy. In
our study the number of patients with second malignancies
was slightly higher than in the other studies. However, it
should be compared with the incidence of second malignancies in non-GH treated patients from the LEA cohort. As
we have not performed a matched controlled study on our
cohort, we can only speculate that this might be explained by
a relatively long period of follow-up (cumulative incidence
increases with time) and/or AL-related-treatment intensity.
Also, 25 patients with GHD described here are those who
had the ‘most aggressive’ antileukaemia treatment (TBI,
HSCT, GVH treatment). Therefore, GH treatment is not the
only factor that impacts on the higher incidence of second
malignancies in these patients.
Evidently, both these data and ours indicate a need for
continued surveillance of childhood cancer survivors
treated with GH.
Conclusion
These retrospective results are based on a small heterogeneous sample and therefore should be interpreted with
caution. We found a measurable beneficial effect of GH
treatment on growth up to FH, a finding already reported.
As catch-up growth was observed in our patients we
consider that a timely start of appropriate hormonal
replacement therapy could help to reduce the negative
impact of the AL-related treatment on the growth of
paediatric patients. Therefore, in conclusion to this study,
we propose that all children who are high-risk patients
(HSCT, TBI, CI) should routinely undergo once a year
testing of GH secretion in response to pharmacological
stimuli and endocrine function.
GH secretion testing once a year
GH deficiency
Yes
GH
treatment
GH secretion
testing / 6 months
No
Growth velocity
decrease
≥ 2SD / year
Endocrine
function
No
Watchful
waiting
Yes
Randomization
Final height
Figure 3 Algorithm proposed for a growth follow-up management in
children after intensive AL treatment in the context of a future larger,
randomized and long-term study. In this algorithm the choice between
treatment and watchful waiting will be determined by based on clinical
(growth velocity) and biological parameters (GH status but also endocrine
function). Patients will be screened for GHD once a year. Those with
biochemically proven GHD will be treated. Those without biochemically
proven GHD and normal growth rate will be re-tested next year. Those
without biochemically proven GHD and decreased growth rate (X2 s.d./
year) could be randomized for treatment or waiting group.
Bone Marrow Transplantation
GH treatment and final height after HSCT
F Isfan et al
692
Moreover, in the context of a future larger, randomized,
long-term study, the approach adopted could be to
annually screen patients for GHD. Those with biochemically proven GHD will be treated. Those without biochemically proven GHD and normal growth rate will be retested next year. Those without biochemically proven GHD
and decreased growth rate (X2 s.d./year) could be randomized for treatment or waiting group. Endocrine function
should be taken into account (Figure 3).
Conflict of interest
The authors declare no conflict of interest.
Acknowledgements
The cooperation of all participating teams in the LEA
programme was greatly appreciated: Hématologie Pédiatrique
APHP St Louis-Paris, Hématologie Pédiatrique APHP R
Debré—Paris (Pr A Baruchel), Oncologie Pédiatrie APHP
Trousseau—Paris (Pr G Leverger), Institut d’Hématologie et
d’Oncologie Pédiatrique HCL—Lyon (Pr Y Bertrand), Hémato
oncologie pédiatrique et greffe de moelle CHU—Nancy (Pr P
Chastagner, Pr Bordigoni), CRCTCP CHU—Clermont-Ferrand
(Pr F Deméocq), Hémato oncologie pédiatrique CHU—Nice
(Dr N Sirvent), Hémato pédiatrique APHM La Timone—
Marseille (Pr G Michel) Oncologie Pédiatrie CHU—Grenoble
(Pr D Plantaz), Unité d’hémato-oncologie et greffes de moelle
CHU-Rennes (Dr Virginie Gandemer). We thank Pr Pascal
Auquier, Marie Claude Simeoni and their staff: EA3279 ‘Qualité
de Vie et Maladies Chroniques’ Service de Santé Publique et
Epidémiologie—Unité d’aide méthodologique à la Recherche
clinique et épidémiologique, Marseille AP-HM, for database
support.
References
1 Sanders JE, Guthrie KA, Hoffmeister PA, Woolfrey AE,
Carpenter PA, Appelbaum FR. Final adult height of patients
who received hematopoietic cell transplantation in childhood.
Blood 2005; 105: 1348–1354.
2 Sanders JE. Growth and development after hematopoietic
cell transplant in children. Bone Marrow Transplant 2008; 41:
223–227.
3 Bernard F, Bordigoni P, Simeoni MC, Barlogis V, Contet A,
Loundou A et al. Height growth during adolescence and final
height after haematopoietic SCT for childhood acute leukaemia: the impact of a conditioning regimen with BU or TBI.
Bone Marrow Transplant 2009; 43: 637–642.
4 Huma Z, Boulad F, Black P, Heller G, Sklar CA. Growth in
children after bone marrow transplantation for acute leukemia. Blood 1995; 86: 819–824.
5 Brauner R, Adan L, Souberbielle JC, Esperou H, Michon J,
Devergie A et al. Contribution of growth hormone deficiency
to the growth failure that follows bone marrow transplantation. J Pediatr 1997; 130: 785–792.
6 Cohen A, Rovelli A, Bakker B, Uderzo C, Van Lint MT,
Esperou H et al. Final height of patients who underwent bone
marrow transplantation for hematological disorders during
childhood: a study by the Working Party for Late Effects–
EBMT. Blood 1999; 93: 4109–4115.
Bone Marrow Transplantation
7 Frisk P, Arvidson J, Gustafsson J, Lönnerholm G. Pubertal
development and final height after autologous bone marrow
trans plantation for acute leukemia. Bone Marrow Transplant
2004; 33: 205–210.
8 Davies HA, Didcock E, Didi M, Ogilvy-Stuart A, Wales JK,
Shalet SM. Growth, puberty and obesity after treatment for
leukaemia. Acta Paediatr Suppl 1995; 411: 45–50.
9 Sanders JE, Pritchard S, Mahoney P, Amos D, Buckner CD,
Witherspoon RP et al. Growth and development following
marrow transplantation for leukemia. Blood 1986; 68:
1129–1135.
10 Borgström B, Bolme P. Growth and growth hormone in
children after bone marrow transplantation. Horm Res 1988;
30: 98–100.
11 Thomas ED, Clift RA, Hersman J, Sanders JE, Stewart P,
Buckner CD et al. Marrow transplantation for acute
nonlymphoblastic leukemic in first remission using fractionated or single-dose irradiation. Int J Radiat Oncol Biol Phys
1982; 8: 817–821.
12 Brauner R, Fontoura M, Zucker JM, Devergie A, Souberbielle
JC, Prevot-Saucet C et al. Growth and growth hormone
secretion after bone marrow transplantation. Arch Dis Child
1993; 68: 458–463.
13 Thomas BC, Stanhope R, Plowman PN, Leiper AD. Growth
following single fraction and fractionated total body irradiation for bone marrow transplantation. Eur J Pediatr 1993; 152:
888–892.
14 Leiper AD, Stanhope R, Lau T, Grant DB, Blacklock H,
Chessells JM et al. The effect of total body irradiation and
bone marrow transplantation during childhood and adolescence on growth and endocrine function. Br J Haematol 1987;
67: 419–426.
15 Bakker B, Oostdijk W, Geskust RB, Stokvis-Brantsma WH,
Vossen JM, Wit JM. Growth hormone (GH) secretion and
response to GH therapy after total body irradiation and
haematopoietic stem cell transplantation during childhood.
Clin Endocrinol 2007; 67: 589–597.
16 Chemaitilly W, Boulad F, Heller G, Kernan NA, Small TN,
O’Reilly RJ et al. Final height in pediatric patients after
hyperfractionated total body irradiation and stem cell transplantation. Bone Marrow Transplant 2007; 40: 29–35.
17 Leung W, Rose SR, Zhou Y, Hancock ML, Burstein S,
Schriock EA et al. Outcomes of growth hormone replacement
therapy in survivors of childhood acute lymphoblastic
leukemia. J Clin Oncol 2002; 20: 2959–2964.
18 Holm K, Nysom K, Rasmussen MH, Hertz H, Jacobsen N,
Skakkebaek NE et al. Growth, growth hormone and final
height after BMT: possible recovery of irradiation-induced
growth hormone insufficiency. Bone Marrow Transplant 1996;
18: 163–170.
19 Michel G, Bordigoni P, Simeoni MC, Curtillet C, Hoxha S,
Robitail S et al. Health status and quality of life in long-term
survivors of childhood leukaemia: the impact of haematopoietic stem cell transplantation. Bone Marrow Transplant 2007;
40: 897–904.
20 Bordigoni P, Esperou H, Souillet G, Pico J, Michel G, Lacour
B et al. Total body irradiation-high-dose cytosine arabinoside
and melphalan followed by allogeneic bone marrow transplantation from HLA-identical siblings in the treatment of
children with acute lymphoblastic leukaemia after relapse
while receiving chemotherapy: a Societe Francaise de Greffe de
Moelle study. Br J Haematol 1998; 102: 656–665.
21 Vogelsang GB, Lee L, Bensen-Kennedy DM. Pathogenesis and
treatment of graft-versus-host disease after bone marrow
transplant. Annu Rev Med 2003; 54: 29–52.
22 Sempé M, Pédron G, Roy-Pernot M-P. Auxologie, méthode et
séquences. Théraplix: Paris, 1979, pp 124.
GH treatment and final height after HSCT
F Isfan et al
693
23 Dai QY, Souillet G, Bertrand Y, Galambrun C, Bleyzac N,
Manel AM et al. Antileukemic and long-term effects of two
regimens with or without TBI in allogeneic bone marrow
transplantation for childhood acute lymphoblastic leukemia.
Bone Marrow Transplant 2004; 34: 667–673.
24 Couto-Silva AC, Trivin C, Esperou H, Michon J, Baruchel A,
Lemaire P et al. Final height and gonad function after total
body irradiation during childhood. Bone Marrow Transplant
2006; 38: 427–432.
25 Clement-De Boers A, Oostdijk W, Van Weel-Sipman MH,
Van den Broeck J, Wit JM, Vossen JM. Final height and
hormonal function after bone marrow transplantation in
children. J Pediatr 1996; 129: 544–550.
26 Sklar CA, Mertens AC, Mitby P, Occhiogrosso G, Qin J,
Heller G et al. Risk of disease recurrence and second
neoplasms in survivors of childhood cancer treated with
growth hormone: a report from the Childhood Cancer
Survivor Study. J Clin Endocrinol Metab 2002; 87: 3136–3141.
27 Ergun-Longmire B, Mertens AC, Mitby P, Qin J, Heller G, Shi
W et al. Growth hormone treatment and risk of second
neoplasms in the childhood cancer survivor. J Clin Endocrinol
Metab 2006; 91: 3494–3498.
Bone Marrow Transplantation