Clinical Endocrinology (2014) 81, 71–76 doi: 10.1111/cen.

12408

ORIGINAL ARTICLE

Dose-sparing and safety-enhancing effects of an IGF-I-based

dosing regimen in short children treated with growth hormone
in a 2-year randomized controlled trial: therapeutic and
pharmacoeconomic considerations
Pinchas Cohen*, Wayne Weng†, Alan D. Rogol‡, Ron G. Rosenfeld§, Anne-Marie Kappelgaard¶ and
John Germak†

*University of Southern California, Los Angeles, CA, †Novo Nordisk Inc, Plainsboro, NJ, ‡University of Virginia, Charlottesville,
VA, §Oregon Health and Science University, Portland, OR, USA and ¶Novo Nordisk A/S, Søborg, Denmark

Conclusions IGF-I-based GH dosing, targeted to age- and

Summary gender-adjusted means, may offer a more dose-sparing and poten-
tially safer mode of therapy than traditional weight-based dosing.
Context and objective Titrating the dosage of growth hor-
mone (GH) to serum levels of insulin-like growth factor-I (IGF- (Received 6 September 2013; returned for revision 10 October
I) is a feasible treatment strategy in children with GH deficiency 2013; finally revised 19 December 2013; accepted 9 January 2014)
(GHD) and idiopathic short stature (ISS). The objective was to
assess the dose-sparing effect and theoretical safety of IGF-I-
based GH therapy.
Design, setting and patients This was a post hoc analysis of a Introduction
previously described 2-year, multicenter, open-label, random-
The dosage of GH for the treatment of children with short stat-
ized, outpatient, controlled clinical trial in 172 prepubertal short
ure or growth failure has been based historically on body weight,
children [age 7
5  2
4 years; height standard deviation score
usually in the range of 25–100 mcg/kg/day in pediatric patients
(HSDS) 2
64  0
61] classified by baseline peak GH levels as
with growth disorders, depending on age and pubertal status.
GHD (<7 ng/ml) or ISS (≥7 ng/ml).
Although effective and widely used, the high cost of GH therapy
Intervention Conventional weight-based dosing of GH
and variability in treatment response has led to ongoing efforts
(0
04 mg/kg/day) (n = 34) or GH dosing titrated to an IGF-I tar-
to optimize dosing strategies to improve not only the efficacy
get of 0 SDS (IGF0T; n = 70) or an IGF-I target of +2 SDS
and cost-effectiveness of treatment, but also its long-term
(IGF2T; n = 68).
safety.1–4 GH continues to have a favourable overall safety pro-
Main Outcome Measures Change in HSDS per GH mg/kg/
file;5 however, concerns over long-term safety have been revis-
day dose (ΔHSDS/GH dose ratio) and proportion of IGF-I levels
ited5–9 and elevated serum concentrations of insulin-like growth
above +2 SDS at the end of 2 years.
factor-I (IGF-I), the presumed mediator of GH-induced somatic
Results GH dosing titrated to an IGF-I target of 0 SDS was the
growth, are associated with certain cancers.10,11
most dose-sparing treatment regimen for GHD or ISS children
Variability in response to a given dose of GH likely reflects
(meanSE ΔHSDS/GH dose ratios 48
1  4
4 and 32
5  2
8,
differences in the severity of GH deficiency (GHD) and the
respectively) compared with conventional dosing (30
3  6
6
patient’s sensitivity to treatment. Several approaches have been
and 21
3  3
5, respectively; P = 0
02, P = 0
005) and IGF2T
explored to optimize the safety and efficacy of GH therapy in
(32
7  4
8 and 16
3  2
8, respectively; P = 0
02, P < 0
0001).
children with short stature. For example, prediction-based mod-
IGF0T also resulted in the fewest IGF-I excursions above +2
els have been developed to optimize GH therapy;12–15 however,
SDS (6
8% vs 30
0% for conventional dosing; P < 0
01).
the exact contribution of prediction-derived indices to adult
height remains unclear.
Titrating the dosage of GH to serum levels of IGF-I is a feasible
Correspondence: Pinchas Cohen, USC Davis School of Gerontology, treatment strategy in children with GHD or idiopathic short stat-
University of Southern California, 3715 McClintock Avenue, Suite 103, ure (ISS).1,2,16 Nevertheless, the potential benefits pertaining to
Los Angeles, CA 90089-0191, USA. Tel.: +1 213-740-1354;
safety and the advantages of dose-sparing on cost for this method
Fax: +1 213-740-5694; E-mail: hassy@usc.edu
have yet to be determined. Therefore, we undertook a post hoc
Clinical Trial Registration: NCT00262249. analysis of a previously conducted study1,2 in which GH dose was

© 2014 John Wiley & Sons Ltd 71

72 P. Cohen et al.

titrated based on serum IGF-I levels to determine the potential information for the patient population were similar between
dose-sparing effect of this method compared with conventional treatment groups and between GHD and ISS subgroups within
weight-based dosing, as well as the theoretical effects on safety. each treatment group.2

Methods Change in HSDS after 2 years

The change in HSDS (ΔHSDS) after 2 years of treatment was
Study design and participants
previously reported.2 Briefly, the meanSE values for ΔHSDS in
The original study was a 2-year, multicenter, open-label, random- GHD children for the IGF0T, IGF2T and conventional dosing
ized, controlled clinical trial.1,2 Briefly, 172 prepubertal short chil- groups were 1
41 (0
13), 2
04 (0
17) and 1
23 (0
12), respec-
dren [age 7
5  2
4 years, height standard deviation score tively. The meanSE values for ΔHSDS in ISS children for the
(HSDS) 2
64  0
61] with low IGF-I levels were randomized in IGF0T, IGF2T and conventional dosing groups were 0
84 (0
07),
a 1:2:2 manner to 1 of 3 groups: conventional weight-based dos- 1
33 (0
09) and 0
87 (0
09), respectively. The respective
ing of GH (0
04 mg/kg/day) (n = 34); GH dosing titrated to an mean  SE values for ΔHSDS in GHD and ISS children were
IGF-I target of 0 SDS (IGF0T group; n = 70); and GH dosing significantly greater for the IGF2T group than for the IGF0T
titrated to an IGF-I target of +2 SDS (IGF2T group; n = 68). The (P < 0
001) and conventional dosing groups (P = 0
001 and
dose of GH was adjusted every 3 months based on weight (con- P < 0
001, respectively). There were no significant differences
ventional group) or, in the IGF0T and IGF2T groups, by 20% per between the IGF0T and conventional dosing groups for DHSDS.
SDS unit difference between the target and current IGF-I SDS. The DHSDS was significantly greater among GHD than ISS
Children were classified as GHD (peak GH <7 ng/ml, n = 63) or children in all treatment groups (P < 0
05).2
ISS (≥7 ng/ml, n = 102) based on GH stimulation testing at base-
line. The original study was conducted after approval by the insti-
Mean daily dose of GH
tutional review board in all centers and in accordance with the
Declaration of Helsinki. Written informed consent was obtained Mean daily doses of GH, calculated as the dose in mg/kg/day at
by the parent/legal guardian before any study procedure. the end of 2 years of treatment, were also previously reported.2
The respective mean  SE daily dose of GH in GHD and ISS
children at the end of 2 years was significantly higher for the
Statistical analysis
IGF2T group (0
091  0
017 and 0
114  0
009 mg/kg/day,
For this post hoc analysis, the 2-year change in HSDS (ΔHSDS) respectively) than for the IGF0T (0
037  0
004 and 0
032 
per mg/kg/day dose of GH used at the end of 2 years (ΔHSDS/ 0
003 mg/kg per day, respectively; P < 0
001) and conventional
GH dose ratio) was calculated and expressed in arbitrary units. dosing groups (0
041  0 mg/kg/day for both GHD and ISS;
Theoretical safety was assessed by the proportion of IGF-I mea- P = 0
002, P < 0
001, respectively). There were no significant
surements above +2 SDS at the end of 2-year treatment period. differences between the IGF0T and conventional dosing groups
For the ΔHSDS/GH dose ratio, between-group (i.e., IGF0T, for GH daily dose.
IGF2T, conventional dosing) comparisons were performed using
analysis of covariance with treatment effect, sex and baseline
Post hoc analysis of ΔHSDS/GH dose ratios
HSDS values included in the model. For comparison of the
proportion of IGF-I SDS levels above +2 at the end of 2 years, Among the three different treatment groups, the respective
Fisher’s exact test was used. mean  SE values for the ΔHSDS/GH dose ratios in GHD and
ISS children were highest in the IGF0T group (48
1  4
4 and
32
5  2
8, respectively), with significant differences compared to
Results
both the conventional dosing group (30
3  6
6 and 21
3  3
5,
respectively; P = 0
02 and P = 0
005, respectively) and the IGF2T
Study population
group (32
7  4
8 and 16
3  2
8, respectively; P = 0
02 and
The study population has previously been described.1,2 Briefly, P < 0
0001, respectively) (Fig. 1). Thus, while ΔHSDS was greater
mean  SD bone age for the study population (5
51  1
93 in the IGF2T group than in either of the other groups, the GH
years) was approximately 2 years behind their chronological age dose was also significantly higher. The resultant ΔHSDS/GH dose
(7
53  2
40 years).1 More males (n = 132; 77%) than females ratios were not only significantly lower than those for the IGF0T
(n = 40; 23%) were enrolled in the study.1 At baseline, the mean group, but were also not significantly different from those for the
HSDS for all patients was 2
64  0
61 and the mean IGF-I SDS conventional dosing group (Fig. 1).
was 3
56  1
74.1 By design, the peak stimulated GH values
were significantly lower in children classified as GHD compared
Theoretical safety based on IGF-I excursions above +2
to those classified as ISS (3
96  1
94 vs 12
87  4
77 ng/ml;
SDS
P < 0
001) and children classified as GHD had significantly lower
mean IGF-I SDS values ( 4
10  1
95 vs 3
27  1
53; Target IGF-I levels were attained at 6 months in the IGF0T
P < 0
05).2 Otherwise, the demographics and baseline group and at 9 months in the IGF2T group and remained stable

© 2014 John Wiley & Sons Ltd

Clinical Endocrinology (2014), 81, 71–76
 GH dose-sparing with IGF-I-based dosing 73

P = 0·8
Conv
60
P = 0·02 P = 0·02 IGF0T
IGF2T
50 P = 0·4

P = 0·005 P < 0·0001

40

30

20

10

Fig. 1 ΔHSDS/GH Dose Ratio After 2 Years of GH

Treatment in Children with GHD and ISS. Conv, 0
conventional weight-based dosing (GH 0
04 mg/kg/ GHD ISS
day); DHSDS/GH dose ratio, change in height
standard deviation score per mg/kg/day GH dose;
Mean (SE) Mean (SE)
GHD, growth hormone deficiency; ISS, idiopathic
short stature; IGF0T, dose titration to IGF-I target Conv 30·3 (6·6) 21·3 (3·5)
of 0 standard deviation score; IGF2T, dose titration IGF0T 48·1 (4·4) 32·5 (2·8)
to IGF-I target of +2 standard deviation score; SE,
standard error.
IGF2T 32·7 (4·8) 16·3 (2·8)

thereafter. MeanSE IGF-I SDS values for GHD children in each ≤+2 SDS
Percent IGF values > +2 SDS or ≤ +2 SDS

treatment group at the end of 2 years were 0
46  0
24 for IGF0T, 120
>+2 SDS
2
83  0
39 for IGF2T and 0
66  0
87 for the conventional dos-
100
ing group. For ISS children, the mean  SE IGF-I SDS values
were 0
05  0
24 for IGF0T, 1
93  0
38 for IGF2T and 80
0
97  0
52 for the conventional dosing group. At the end of 2
70
years, IGF-I SDS was significantly higher for the IGF2T group 60
93
than the IGF0T group (GHD, P < 0
001; ISS, P = 0
001). Never-
theless, for the combined population of GHD and ISS children, 40
while the mean IGF-I SDS was comparable between the IGF0T
20
and conventional dosing groups, the percentage of IGF-I levels 30
above +2 SDS at the end of 2 years was significantly lower in the 0 7a
IGF0T group than in the conventional dosing group (7% for IGF0T Conv
IGF0T, 30% for conventional dosing; P = 0
0083) (Fig. 2).
Fig. 2 Proportion of IGF-I Measurements Above +2 SDS by Dosing
Strategy in Children with GHD and ISS. Conv, conventional weight-
Discussion based dosing; IGF0T, dose titration to IGF-I target of 0 standard
deviation score, SDS, standard deviation score. aP<0
01 compared with
Previous analyses demonstrated the feasibility of two IGF-I- conventional dosing for proportion of IGF-I levels above +2 SDS,
based treatment regimens.1,2 Increases in growth (ΔHSDS) at 2 Fischer’s exact test.
years were comparable between the IGF-I target of the mean (0
SDS) and conventional weight-based dosing groups for both
GHD and ISS patients; the IGF-I target of upper normal (+2 proportion of IGF-I SDS values that exceeded the upper range
SDS) resulted in significantly greater ΔHSDS compared to the of normal (+2 SDS) as a theoretical measure of safety.
other two treatment groups in these patients.1,2 As the dose An effective dose-sparing strategy can save substantial health-
required to achieve an IGF-I level of +2 SDS was also signi- care costs for managed care organizations and patients receiving
ficantly higher than the other treatment groups, the current GH therapy.17 Previous pharmacoeconomic studies with GH
analysis was undertaken to compare the three treatment regi- relying on the use of decision-modelling have produced variable
mens in terms of increment in height SDS per dose as it may findings17–19 and the applicability of such methodology to real-
relate to dose- sparing potential. Furthermore, as both weight- world situations may be limited.20,21 Results from this analysis
based dosing and GH dosing targeted to the mean IGF-I SDS found that the most dose-sparing treatment regimen, based on
resulted in comparable IGF-I levels on average, a comparison analyses of ΔHSDS/GH dose ratios, for children with GHD or
between these dosing regimens was made to assess the ISS was a GH dose titrated to an IGF-I target of 0 SDS. To our

© 2014 John Wiley & Sons Ltd

Clinical Endocrinology (2014), 81, 71–76
74 P. Cohen et al.

knowledge, this is the first demonstration that a feasible dosing limit of normal would seem to be a reasonable approach in
strategy based on IGF-I target can potentially be more cost ben- terms of cost-effectiveness and more prudent in terms of safety,
eficial (based on ΔHSDS/GH dose ratio) than conventional without incurring any compromise of efficacy, especially for
weight-based dosing while having comparable efficacy (as mea- patients with GHD. Although IGF-I targets were met equally in
sured by ΔHSDS). Not only was targeting the IGF-I SDS to the patients with GHD and ISS, gains in height were significantly
mean the most dose-sparing treatment regimen, it also resulted less for ISS patients.2 This may indicate a degree of IGF-I insen-
in a significantly lower proportion of IGF-I levels above +2 SDS sitivity, as well as GH insensitivity, in the ISS patients, who may
than did conventional dosing, which could have important require a more aggressive IGF-I target.2 Also, the 2-year dura-
implications for long-term safety. tion of the study may not have been long enough for all patients
IGF-I can produce alterations in cell proliferation and apopto- to achieve optimal catch-up growth.3 As other IGF-I targets have
sis, and elevated levels of IGF-I have been linked to some can- yet to be fully examined, the optimal IGF-I target, particularly
cers in adult populations including colon, prostate and certain for non-GHD conditions, remains to be identified, and long-
types of breast cancer.10,22–24 Long-term observational studies term clinical benefits of dosing based on IGF-I targets still need
have reported on safety outcomes for children receiving GH to be demonstrated. One proposed model suggests using higher
treatment, including malignancies.5,8,25 For the most part, recent IGF-I targets (+2 to +3 SDS) to maximize height during the
results have been reassuring. Findings from the National Coop- catch-up phase followed by lower targets for maintenance.33
erative Growth Study reported no appreciable increase in de Furthermore, a general IGF-based dosing strategy would not
novo cancer [standardized incidence ratios (SIR) 1
12, 95% CI preclude also individualizing treatment based upon responsive-
0
75–1
61] and a lower than expected incidence of new-onset ness in growth.
leukaemia (SIR 0
54; 95% CI 0
11–1
58).5 However, a risk for The lack of IGF-I assay standardization and accepted norma-
second neoplasms among children treated with GH has been tive reference ranges are additional factors that may impact the
reported. A large retrospective cohort of childhood cancer survi- generalizability of the reported results. A consensus statement
vors treated with GH reported an approximate threefold higher put forth by the Growth Hormone Research Society has outlined
rate of second neoplasms than expected (SIR 3
2, 95% CI 1
9– the obstacles and steps needed to move towards a standardized
5
5) at 15 years of follow-up.25 This rate decreased somewhat process.34 Until an accepted standardized assay is available, clini-
after 32 years of follow-up, but still remained elevated (SIR 2
1, cians should utilize an assay with demonstrated reliability, col-
95% CI 1
3–3
5).8 lect and process samples appropriately and adjust to appropriate
Although a definitive link between elevated IGF-I levels associ- age and gender-matched reference norms,28 which should be
ated with GH treatment and biological end-points has not been requested from each laboratory whenever possible. When feasi-
shown, it has been recommended that IGF-I levels in individual ble, clinicians should attempt to utilize the same assay and tech-
patients should be maintained within age- and gender-based nique for a patient over time, although patient factors, such as
reference ranges.26–28 We found that in patients dosed conven- health insurance, may be a barrier.28 In addition to interassay
tionally with 0
04 mg/kg/day, the proportion of IGF-I levels variability, clinicians should also be aware of intrapatient vari-
above +2 SDS was 30%. Others have reported similar excursions. ability when interpreting IGF-I assay results, especially with
For example, 28% of pubertal patients treated with 0
7 mg/kg/ regard to borderline values.34
week had high IGF-I concentrations,29 as did 45% of SGA In conclusion, IGF-based GH dosing targeted to the age- and
patients treated with 0
057 mg/kg/day for 2 years.30 In another gender-adjusted mean (0 SDS) in GHD and ISS children
report, 17% of GHD patients had IGF-I excursions above +2 SDS resulted in a higher ΔHSDS/GH dose ratio than conventional
after 2 years even when the GH dose was based on body surface weight-based dosing, despite comparable levels of IGF-I and
area at an average dose of 1 mg/m2/day (equivalent to 0
035/mg/ ΔHSDS achieved, and may offer a more dose-sparing mode of
day).31 A GH dose-sparing effect of IGF-I-based dosing was also GH therapy than traditional weight-based GH dosing. IGF-I-
noted in a study of adult Japanese patients with GHD who were based dosing targeted to the mean SDS also decreased exposure
switched from a conventional weight-based dose regimen, with a to IGF-I levels above +2 SDS and may therefore address some of
concomitant increase in the number of patients who were main- the theoretical safety concerns related to GH treatment.
tained within the reference range for IGF-I SDS.32 The present
analysis demonstrated that targeting IGF-I to the mean (i.e., 0
Acknowledgements
SDS) significantly decreased the proportion of IGF-I measure-
ments >+2 SDS at the end of year 2 compared to conventional Funding for the statistical analysis was provided by Novo Nordisk
weight-based dosing. Decreasing the risk of exposure to high Inc. The authors wish to thank Sarah Mizne, PharmD and Joyce
IGF-I levels potentially reduces the theoretical risk of cancer and Willetts, PhD of MedVal Scientific Information Services, LLC
other adverse events related to high IGF-I levels. (Skillman, NJ), for providing medical writing and editorial assis-
There are limitations to the present analysis. It was not pro- tance. This manuscript was prepared according to the Interna-
spectively designed for the purpose of dose-sparing and was not tional Society for Medical Publication Professionals’ ‘Good
intended to provide any specific dosing target or recommenda- Publication Practice for Communicating Company-Sponsored
tions. With that said, based on our analysis, an IGF-I level Medical Research: the GPP2 Guidelines’. Funding to support the
around the mean for the population rather than at the upper preparation of this manuscript was provided by Novo Nordisk Inc.

© 2014 John Wiley & Sons Ltd

Clinical Endocrinology (2014), 81, 71–76
 GH dose-sparing with IGF-I-based dosing 75

13 Bakker, B., Frane, J., Anhalt, H. et al. (2008) Height velocity tar-
Disclosures gets from the national cooperative growth study for first-year
PC and RGR are consultants to Novo Nordisk; ADR is consul- growth hormone responses in short children. Journal of Clinical
tant to AbbVie, Novo Nordisk, SOV Therapeutics, LG Biophar- Endocrinology & Metabolism, 93, 352–357.
14 Ranke, M.B. & Lindberg, A., International Board. (2011) Predic-
maceuticals and Sanofi; WW and JG are employees of Novo
tion models for short children born small for gestational age
Nordisk, AMK is an employee and shareholder of Novo Nor-
(SGA) covering the total growth phase. Analyses based on data
disk.
from KIGS (Pfizer International Growth Database). BMC Medi-
cal Informatics and Decision Making, 11, 38.
References 15 Schonau, E., Westermann, F., Rauch, F. et al. (2001) A new and
accurate prediction model for growth response to growth hor-
1 Cohen, P., Rogol, A.D., Howard, C.P. et al. (2007) Insulin growth mone treatment in children with growth hormone deficiency.
factor-based dosing of growth hormone therapy in children: a ran- European Journal of Endocrinology, 144, 13–20.
domized, controlled study. Journal of Clinical Endocrinology & 16 Cohen, P., Rogol, A.D., Weng, W. et al. (2013) Efficacy of IGF-
Metabolism, 92, 2480–2486. based growth hormone (GH) dosing in non GH-deficient (non-
2 Cohen, P., Germak, J., Rogol, A.D. et al. (2010) Variable degree GHD) short stature children with low IGF-I is not related to
of growth hormone (GH) and insulin-like growth factor (IGF) basal IGF-I levels. Clinical Endocrinology, 78, 405–414.
sensitivity in children with idiopathic short stature compared 17 Joshi, A.V., Munro, V. & Russell, M.W. (2006) Cost-utility of
with GH-deficient patients: evidence from an IGF-based dosing somatropin (rDNA origin) in the treatment of growth hormone
study of short children. Journal of Clinical Endocrinology & deficiency in children. Current Medical Research and Opinion, 22,
Metabolism, 95, 2089–2098. 351–357.
3 Kristrom, B., Aronson, A.S., Dahlgren, J. et al. (2009) Growth 18 Christensen, T., Fidler, C., Bentley, A. et al. (2010) The cost-
hormone (GH) dosing during catch-up growth guided by individ- effectiveness of somatropin treatment for short children born
ual responsiveness decreases growth response variability in prepu- small for gestational age (SGA) and children with growth hor-
bertal children with GH deficiency or idiopathic short stature. mone deficiency (GHD) in Sweden. Journal of Medical Econom-
Journal of Clinical Endocrinology & Metabolism, 94, 483–490. ics, 13, 168–178.
4 Lee, J.M., Davis, M.M., Clark, S.J. et al. (2006) Estimated cost- 19 Takeda, A., Cooper, K., Bird, A. et al. (2010) Recombinant
effectiveness of growth hormone therapy for idiopathic short stat- human growth hormone for the treatment of growth disorders
ure. Archives of Pediatrics and Adolescent Medicine, 160, 263–269. in children: a systematic review and economic evaluation. Health
5 Bell, J., Parker, K.L., Swinford, R.D. et al. (2010) Long-term Technology Assessment, 14, 1–209, iii–iv.
safety of recombinant human growth hormone in children. Jour- 20 Hill, S.R., Mitchell, A.S. & Henry, D.A. (2000) Problems with
nal of Clinical Endocrinology & Metabolism, 95, 167–177. the interpretation of pharmacoeconomic analyses: a review of
6 Savendahl, L., Maes, M., Albertsson-Wikland, K. et al. (2012) submissions to the Australian Pharmaceutical Benefits Scheme.
Long-term mortality and causes of death in isolated GHD, ISS, The Journal of the American Medical Association, 283, 2116–2121.
and SGA patients treated with recombinant growth hormone dur- 21 Rennie, D. & Luft, H.S. (2000) Pharmacoeconomic analyses:
ing childhood in Belgium, the Netherlands, and Sweden: prelimin- making them transparent, making them credible. The Journal of
ary report of 3 countries participating in the EU SAGhE Study. the American Medical Association, 283, 2158–2160.
Journal of Clinical Endocrinology & Metabolism, 97, E213–E217. 22 Rowlands, M.A., Gunnell, D., Harris, R. et al. (2009) Circulating
7 Carel, J.C., Ecosse, E., Landier, F. et al. (2012) Long-term mor- insulin-like growth factor peptides and prostate cancer risk: a
tality after recombinant growth hormone treatment for isolated systematic review and meta-analysis. International Journal of
growth hormone deficiency or childhood short stature: prelimin- Cancer, 124, 2416–2429.
ary report of the French SAGhE Study. Journal of Clinical Endo- 23 Morgillo, F., De Vita, F., Antoniol, G. et al. (2013) Serum insu-
crinology & Metabolism, 97, 416–425. lin-like growth factor 1 correlates with the risk of nodal metasta-
8 Ergun-Longmire, B., Mertens, A.C., Mitby, P. et al. (2006) sis in endocrine-positive breast cancer. Current Oncology, 20,
Growth hormone treatment and risk of second neoplasms in the e283–e288.
childhood cancer survivor. Journal of Clinical Endocrinology & 24 Clayton, P.E., Banerjee, I., Murray, P.G. et al. (2011) Growth
Metabolism, 91, 3494–3498. hormone, the insulin-like growth factor axis, insulin and cancer
9 Sperling, M.A. (2012) Long-term therapy with growth hormone: risk. Nature Reviews Endocrinology, 7, 11–24.
bringing sagacity to SAGHE. Journal of Clinical Endocrinology & 25 Sklar, C.A., Mertens, A.C., Mitby, P. et al. (2002) Risk of disease
Metabolism, 97, 81–83. recurrence and second neoplasms in survivors of childhood can-
10 Renehan, A.G., Zwahlen, M., Minder, C. et al. (2004) Insulin- cer treated with growth hormone: a report from the Childhood
like growth factor (IGF)-I, IGF binding protein-3, and cancer Cancer Survivor Study. Journal of Clinical Endocrinology &
risk: systematic review and meta-regression analysis. Lancet, 363, Metabolism, 87, 3136–3141.
1346–1353. 26 Cohen, P., Rogol, A.D., Deal, C.L. et al. (2008) Consensus
11 Gallagher, E.J. & LeRoith, D. (2011) Minireview: IGF, insulin, statement on the diagnosis and treatment of children with idio-
and cancer. Endocrinology, 152, 2546–2551. pathic short stature: a summary of the Growth Hormone
12 Ranke, M.B. & Lindberg, A. (2010) Observed and predicted Research Society, the Lawson Wilkins Pediatric Endocrine Soci-
growth responses in prepubertal children with growth disorders: ety, and the European Society for Paediatric Endocrinology
guidance of growth hormone treatment by empirical variables. Workshop. Journal of Clinical Endocrinology & Metabolism, 93,
Journal of Clinical Endocrinology & Metabolism, 95, 1229–1237. 4210–4217.

© 2014 John Wiley & Sons Ltd

Clinical Endocrinology (2014), 81, 71–76
76 P. Cohen et al.

27 Higham, C.E., Jostel, A. & Trainer, P.J. (2007) IGF-I measure- 31 Feigerlova, E., Diene, G., Oliver, I.et al. (2010) Elevated insulin-
ments in the monitoring of GH therapy. Pituitary, 10, 159–163. like growth factor-I values in children with Prader-Willi syn-
28 Pawlikowska-Haddal, A., Cohen, P. & Cook, D.M. (2011) How drome compared with growth hormone (GH) deficiency children
useful are serum IGF-I measurements for managing GH over two years of GH treatment. Journal of Clinical Endocrinology
replacement therapy in adults and children? Pituitary, 15, & Metabolism, 95, 4600–4608.
126–134. 32 Chihara, K., Kato, Y., Kohno, H.et al. (2008) Safety and efficacy
29 Mauras, N., Attie, K.M., Reiter, E.O. et al. (2000) High dose of growth hormone (GH) during extended treatment of adult
recombinant human growth hormone (GH) treatment of Japanese patients with GH deficiency (GHD). Growth Hormone
GH-deficient patients in puberty increases near-final height: a & IGF Research, 18, 307–317.
randomized, multicenter trial. Journal of Clinical Endocrinology & 33 Park, P. & Cohen, P. (2004) The role of insulin-like growth fac-
Metabolism, 85, 3653–3660. tor I monitoring in growth hormone-treated children. Hormone
30 Cabrol, S., Perin, L., Colle, M. et al. (2011) Evolution of IGF-1 Research, 62(Suppl 1), 59–65.
in children born small for gestational age and with growth retar- 34 Clemmons, D.R. (2011) Consensus statement on the standardi-
dation, treated by growth hormone adapted to IGF-1 levels after zation and evaluation of growth hormone and insulin-like
1 year. Hormone Research in Paediatrics, 76, 419–427. growth factor assays. Clinical Chemistry, 57, 555–559.

© 2014 John Wiley & Sons Ltd

Clinical Endocrinology (2014), 81, 71–76