The Journal of Clinical Endocrinology & Metabolism Vol. 85, No. 3 939-942
Copyright © 2000 by The Endocrine Society
Treatment of Growth Hormone Deficiency in Adults
Helen Simpson and
Peter Sonksen
Department of Endocrinology
St. Thomas’ Hospital
London SE1 7EH, United Kingdom
 |
Introduction
|
|---|
"The regimen I adopt shall be for the benefit of
patients according to my ability and judgement
..." From the Hippocratic oath—c. 4th
Century B.C.
GH DEFICIENCY (GHD) in adults is
now well defined both clinically and biochemically. With the advent of
recombinant technology there is now a virtually unlimited, safe supply
of recombinant human GH for treatment of children and adults with GHD.
Anyone who has prescribed GH replacement therapy can tell anecdotal
reports of patients lives (and indeed the lives of their families) who
were completely transformed by GH replacement, and there is a wealth of
data showing GH replacement ameliorates the most prominent features of
GHD (alterations in body composition, reduced energy and work capacity,
and impaired psychological well being) over the short term by GH
replacement. However, in these days of evidence-based medicine,
especially where governments are rationing health care, it is becoming
harder to obtain funding for expensive drugs. The authors of both the
previous articles agree that the published evidence supports short-term
GH replacement in at least some adults with severe GHD. There are,
however, several questions that remain to be answered.
|
1. Are there benefits of long-term hormone replacement therapy with
GH?
|
|---|
Evidence to support long-term GH replacement is becoming
available. It is now 10 yr since the initial studies of GH replacement
and a 10-yr follow-up study of one of the original cohorts [the St.
Thomas’ cohort studied by Salomon et al. (1)] has been
published recently in this journal. Gibney et al. (2) traced
and restudied 21 of the original cohort, 10 of whom had been treated
with GH for the entire 10 yr (the reasons why only 50% continued to
take GH were mainly "local medico-political-funding" issues rather
than patient preferences. The treated and nontreated groups did not
differ measurably). The results showed that the benefits of GH in terms
of altered body composition, an improved lipid profile, and improved
psychological well being measured using the Nottingham Health Profile
were maintained and were significantly improved compared to the group
who had not received long-term GH treatment. In addition, there was
evidence indicating reduced development of atherosclerosis (as measured
by carotid intimal thickness), no increase in left ventricular wall
thickness or hypertension, and no decrease in insulin sensitivity (as
shown by measurements of fasting insulin and glucose). The conclusions
reached were that the benefits gained in the initial study were
maintained over 10 yr with no worsening of the cardiovascular status of
the GH-treated group, indeed to the contrary that the GH-treated group
showed long-term benefits from continued replacement therapy, evidence
that they benefited.
This is the first long-term study to be published, however, it only
involved a small number of patients and it is not a true prospective
randomized controlled trial but rather "randomization by NHS
lottery." True randomized long-term controlled trials will not and
cannot be done because GH replacement has been shown to offer so many
short- and medium-term advantages that no ethics committee would accept
randomization to no GH replacement and no well-informed patient would
accept to be randomized. There remains no evidence, however, that
long-term GH replacement reverses the observed increase in mortality,
especially from cardiovascular causes, or reduces the increased
fracture rate seen in GHD. These questions can only be answered by
properly designed long-term prospective studies of GH replacement. Who
is going to do these?
|
2. Is long-term GH replacement therapy safe?
|
|---|
The major concerns regarding the safety of long-term GH
replacement are those of effects on the cardiovascular system
(increased left ventricular hypertrophy and possible increased insulin
resistance, which offsets improvements in central adiposity), the
possibility of an increase in malignant tumors, and the recurrence of
pituitary tumors. Data from the 10-yr follow-up study did not indicate
any adverse effects of long-term GH treatment. Specifically, there was
no increase in vascular events in the GH-treated group, no worsening of
insulin sensitivity, no increase in left ventricular wall thickness or
hypertension, no recurrence of pituitary tumors, and no new
malignancies reported.
Although there are and never will be long-term prospective
placebo-controlled studies, there have been attempts to collect
long-term data of GH replacement. Each of the major manufacturers of GH
have initiated "postmarketing surveillance" databases to monitor
the safety of GH replacement in adults, NovoNordisk with Nordireg,
Genentech with the NCSS database, Lilly with HypoCCS, and Pharmacia
with KIMS. To date, KIMS has the largest number of patients on record
with data from 4200 patients and details from 25 different countries.
At the present time there seems to be no increase in adverse events.
There has recently been reported, however, a possible small increase in
de novo malignancies, leukemias and lymphomas, the
significance of which is unknown. This comes at a time when there is
much debate about the interpretation of epidemiological evidence
showing a link between serum insulin-like growth factor (IGF) I levels
and the prevalence of cancer of the breast and prostate in people
without GHD. In view of the fact that malignancy rates in long-standing
acromegaly are only marginally raised, it is not clear what the
significance of these epidemiological and KIMS findings is, unclear but
not unduly disturbing. It is, however, another example of the reason
why it is essential to monitor patients on GH carefully and to build
collaborative databases to accumulate longitudinal experience. Although
such individual company databases are a useful surveillance tool, it is
unlikely that this somewhat fragmented approach will ever have the
statistical power to be able to provide the long-term safety data that
we all need. What is really needed is a properly designed prospective
epidemiological study gathering information about those not receiving
GH as well as those on GH with collaboration between endocrinologists,
epidemiologists, and the pharmaceutical industry. The Growth Hormone
Research Society (GRS) is in an ideal position to facilitate this
because it has good relations with and sponsorship from all the
appropriate industries. It has already shown its ability to bring
people together in two excellent Consensus Workshops (Port Stephens in
1997 and Elat in 1999) and a workshop on "Safety" is planned for
the spring of 2000. The GRS now needs to show strong leadership in the
development of a professional collaborative epidemiological study
capable of answering the key safety issues speedily and
unambiguously.
|
3. What is the best method of monitoring clinical and biochemical
response?
|
|---|
There is no ideal marker for monitoring GH replacement. Although
changes in body composition are the most striking and consistent
finding in studies of GH replacement, changes in individual patients
(and in health) vary greatly. Attempts to achieve normal body
composition with GH replacement can result in IGF-I levels well above
the age-related normal range and clinical evidence of GH excess. IGF-I
is the most sensitive serum marker, having been repeatedly shown to be
more sensitive to GH excess than the GH-dependent peptides ALS and
IGFBP-3, as well as responding more rapidly to changes in GH dose.
However, IGF-I can still be inside the age-related normal range in
patients with clear clinical evidence of GH excess, particularly in the
face of conditions with low portal insulin concentrations
(e.g. Type 1 diabetes, malnutrition, and malabsorbtion
syndromes) or liver disease. Also IGF-I is not a marker of tissue
effectiveness to GH. It is possible that circulating GH-sensitive
markers of collagen and bone, such as procollagen 3 peptide terminal
extension peptide (P-III-P) and osteocalcin may prove more valuable in
this regard.
Until such a time that a better marker of GH replacement becomes
available, IGF-I and body composition, together with a clinical
assessment of the patient, focusing specifically on the quality of life
issues that have been shown to be most characteristic of GHD (including
energy, mood, social isolation, and self-control) taken with a
partner’s assessment where available remain the best way of monitoring
GH replacement therapy.
|
4. In view of the somatopause, until what age should GH be
given?
|
|---|
There is no evidence to suggest that elderly patients with organic
GHD should be denied GH replacement. Although there are difficulties in
diagnosing GHD using the insulin tolerance test in elderly patients who
may have silent myocardial disease, organic GHD can be distinguished
from the somatopause by the presence of pituitary pathology and by
using other provocation tests, such as arginine or GHRH (+/- GH
releasing peptide). In all patients, but particularly the elderly, GH
should be started at a low dose and titrated gradually with serum IGF-I
levels kept within the age-related normal range. Patients should be
monitored carefully for clinical evidence of GH excess because this
group is particularly susceptible to effects of GH excess.
In many ways, old age itself is similar to many features of adult onset
GHD. Aging results in an increase in body fat (particularly central
abdominal fat), loss of muscle mass, reduced strength, and reduced bone
density, together with a decrease in GH secretion. Rudman’s original
contribution to research into GH and aging in the early 1980s remains
of great importance. He was remarkably perspicacious in predicting the
importance of failing GH secretion in the changes in body composition
seen with aging and in pioneering pilot trials showed beneficial
effects of GH replacement on body composition and bone density. This
and other studies raises the important question: "Would GH
replacement be beneficial for the elderly population, in general, or in
particular those with frailty?" There is a clear need for multicenter
studies investigating long-term GH replacement in the frail elderly
with end points such as the ability to perform activities of daily
living and the ability to maintain independence. Once there is "proof
of concept" that GH replacement is able to maintain (or rebuild?)
significant amounts of lean tissues in this population, the development
of oral GH secretagogues may mean that in the future GH hormone
replacement will become similar to traditional oestrogen hormone
replacement in postmenopausal women. Obviously, much more research is
needed, but the number of older adults worldwide is increasing, and so
the burden on health care provision is increasing and this form of
hormone replacement offers a potentially powerful way of mitigating the
burden and improving the quality of life of all of us. A drug that may
help prevent falls and ameliorate frailty would be very attractive, and
cost-effective in cash-strapped health services.
|
5. Is GH replacement cost-effective?
|
|---|
Health care in the Western world as we approach the new millennium
is expensive, consuming between 6 and 12% of the gross national
product of industrialized nations. In the United Kingdom for the year
1997/1998, some £505 million of that was spent on drugs. GH
replacement is not cheap at a cost of about $5,000 (£3,500) per year,
but neither is it expensive in comparison to, say, the cost of triple
therapy for HIV-positive patients at $12,800 (£8,000) per year. Adult
GHD has prevalence of about 15 in 100,000 in the United Kingdom.
Assuming that not all of these patients tolerate or would like GH
replacement, and we estimate that 50% of them are treated with GH
replacement, then the cost would be $1.6 million (£1 million) per
year, or 0.2% of the annual National Health Service drug budget. This
cost varies between individual countries as the percentage of patients
receiving GH replacement varies considerably; for example, Sweden has a
much higher rate of GH replacement in its GH-deficient adults. It has
already been shown that GH replacement in adults leads to a significant
improvement in well-being, quality of life, energy level, and work
capacity, resulting in a proportion of patients being able to return to
work, or to work more effectively. If long-term treatment with GH is
shown, in addition, to result in a decrease in premature mortality rate
with decreased cardiovascular events and bone fractures then GH will be
most likely be shown to be cost-effective. GH replacement could become
even more cost-effective if it was introduced successfully as an
antifrailty treatment in the older general population, preventing the
morbidity and mortality associated with falls and frailty in this age
group.
It should also be mentioned that in the 10 yr follow-up study although
10 of the original patients randomized to receive GH were included in
the follow-up study only 5 of these were still on GH replacement. Some
of the other five patients were not able to continue GH because funding
was denied by their local health authority, an example of how despite
the political statements saying the contrary, in the United Kingdom
health rationing already takes place. It does seem that in many
developed countries patients with adult onset GHD are being denied
access to appropriate hormone replacement purely on the basis of
cost.
As endocrinologists we would probably all agree that there is a place
for, at the very least, a trial of GH replacement for all our patients
with GHD. In addition, there are exciting possibilities that in the
future GH may become more widely useful. It remains crucial, however,
that GH treatment is monitored closely both within the individual and
also on a larger to scale to answer some of the outstanding questions
about efficacy and safety. The existing methods of doing this are
inadequate, and the relevant industries should demonstrate their
ability to cooperate with each other and with the Growth Hormone
Research Society and set up a collaborative prospective study with
sufficient power to be able to answer these key questions within a
reasonable time frame.
|
Footnotes
|
|---|
Address correspondence and requests for reprints to: Bengt-Åke
Bengtsson, M.D., Ph.D., Research Center for Endocrinology and
Metabolism, Sahlgrenska University Hospital, S-413 45 Göteborg,
Sweden.
Received November 26, 1997.
Accepted December 9, 1999.
|
References
|
|---|
-
Salomon F, Cuneo R, Hesp R, Sonksen PH. 1989 The effects of treatment with recombinant human growth hormone on body
composition and metabolism in adults with growth hormone deficiency. N Engl J Med. 321:1797–1803.[Abstract]
-
Gibney J, Wallace JD, Spinks T, et al. 1999 The
effects of 10 years of recombinant human growth hormone (GH) in adult
GH-deficient patients. J Clin Endocrinol Metab. 84:2596–2602.[Abstract/Free Full Text]
Top
Introduction
1. Are there benefits...
