Ghresearchsociety.org

The Journal of Clinical Endocrinology & Metabolism 89(7):3099 –3102 Copyright 2004 by The Endocrine Society CONSENSUS STATEMENT
Biochemical Assessment and Long-Term Monitoring in
Patients with Acromegaly: Statement from a Joint
Consensus Conference of The Growth Hormone
Research Society and The Pituitary Society
Acromegaly is associated with significantly increased curate quantification of previously undetectable levels, in- morbidity and mortality. As a consequence, treatment of the cluding nadir and glucose-suppressed GH levels. These as- disease is indicated in almost all cases once the diagnosis is says are of value in diagnosis and follow-up treatment established. Studies published between 1970 and 1988 re- because they allow better definition of the neurosecretory ported standardized mortality rates in patients with acro- properties of GH, particularly in characterizing the lower megaly to be 1.6 –3.3; more recent studies, published in the limit of spontaneous and suppressed GH secretion. They past 3 yr, reported the rate to be lower, ranging from 1.3–1.8.
have permitted diagnosis of mild and subtle manifestations However, when disease activity is controlled in patients with of acromegaly and improved critical evaluation of therapeu- acromegaly, the relative mortality risk is reduced toward normal. Changes brought about by improvements in assay The epidemiological studies linking GH levels measured methodology for GH and IGF-I mean that hormone levels by RIA to mortality need to be interpreted in the context of reported from these retrospective analyses cannot be simply the new and more sensitive assays. These studies had iden- converted by formula to draw conclusions about the out- tified that treatment to target GH thresholds varying be- come of acromegaly assessed biochemically using current tween 2 and 5 ␮g/liter was associated with improvement in the mortality rate to nearer that of the general population.
For these reasons The Growth Hormone Research Society Comparative studies have demonstrated that GH levels and The Pituitary Society formed a joint program committee quantified by current assays are lower than those measured and invited international experts to address the current sta- by RIAs. There is no simple conversion factor between the tus of both biochemical assessment and long-term monitor- two types of assays, but it would appear that the target ing in patients with acromegaly at a consensus conference threshold may be lowered severalfold.
held in Feldafing, Germany, in April 2003.
Regardless of methodology, an optimal assay should present information on precision and sensitivity across the Biochemical assessment of the patient with acromegaly expected physiological and pathological ranges. The sensi- Biochemical evaluation of the patient with possible acro- tivity limit of the assay should be less than 0.1 ␮g/liter, with megaly status includes measurements of serum concentra- an interassay coefficient of variation less than 15%. Under tions of IGF-I and GH and study of the neurosecretory reg- ideal conditions any assay should be validated with a normal ulation of GH secretion through dynamic testing (1). The range for suppressed GH levels after an oral glucose load.
analysis of serum GH and IGF-I concentrations is limited by The pituitary somatotrophs as well as somatotroph pitu- the lack of standardization and diverse technical problems itary adenomas secrete various isoforms of GH, with the monomeric 22-kDa isoform being the most abundant (ϳ50%)in the circulation. As biological activity is not confined to the 22-kDa form of GH, but is also mediated by other isoforms, Previous epidemiological studies have shown that GH e.g. the 20-kDa form, assays developed to specifically mea- levels in acromegaly are a prognostic indicator of mortality.
sure 22-kDa GH isoform are not necessarily advantageous to Since the early 1960s, GH measurement has been the cor- other GH assays. For any GH assay, information on which nerstone of the biochemical evaluation of acromegaly. The GH isoforms are recognized by its antibodies is desirable.
measurement of GH has evolved from polyclonal RIAs of The absence of adequate standardization of GH assays limited sensitivity to today’s two-site monoclonal antibody, limits comparisons of results between different laboratories.
nonisotopic assays with enhanced sensitivity, allowing ac- Discordance between laboratories should be minimized byadoption of a common recombinant reference preparation,the use of appropriate matrix conditions, and participation in Abbreviations: MRI, Magnetic resonance imaging; OGTT, oral glu- external quality control programs. To interpret the results of GH measurement, treating physicians must have knowledge JCEM is published monthly by The Endocrine Society (http://www.
of relevant assay characteristics (e.g. specificity and GH-bind- endo-society.org), the foremost professional society serving the en-
docrine community.
ing protein interference). The availability of highly purified J Clin Endocrinol Metab, July 2004, 89(7):3099 –3102 recombinant human GH in the standards used facilitates the to be defined for the specific GH assay used, indicates per- sisting impaired neuroregulation of GH secretion. This hasbeen found to be associated with a higher risk of recurrence The availability of IGF-I assays has improved the diagno- sis and management of acromegaly. The serum IGF-I level is Early postoperative assessment of treatment outcome a reliable indicator of GH status in acromegaly and is the best Biochemical evaluation is necessary for the critical assess- biochemical marker of clinical disease activity. The physiol- ment of therapeutic outcome. Clinical benefits of surgery ogy of the regulation of IGF-I and its binding proteins is may be seen rapidly (within days). Postoperative timing of complex. The biochemical diagnosis of acromegaly is diffi- the evaluation of the GH and IGF-I status is influenced by the cult in states of physiologically high GH production, such as patient’s clinical response to surgery and by local practice.
This is usually undertaken in conjunction with the evaluation Some of the factors that contribute to the shortcomings of pituitary function. Early assessment may provide limited of GH measurement also apply to IGF-I assays. Problems information on operative outcome, but formal evaluation include inadequate age-adjusted normative data, lack of should be performed 3 months postoperatively. Stabilization standardization, susceptibility to interference from binding of serum IGF-I levels usually occurs within 3 months after proteins, and the lack of a pure international reference prep- surgery, but may, on rare occasions, be delayed until 12 aration. A robust IGF-I assay should address these problems.
months. Preoperative medical treatment with long-acting The laboratory should participate in an external quality con- somatostatin analogs may influence the timing of postoper- trol program and provide adequate age-adjusted normative ative evaluation because of the prolonged suppressive effect data that will also allow presentation of individual IGF-I on GH of up to 3 months. Subsequent to the postoperative results as an sd score. IGF assays should be standardized to assessment, further life-long evaluation is mandatory.
improve patient management. This includes the use of acommon recombinant IGF-I standard. Until an appropriate international reference preparation is universally applied, itis essential that comprehensive reference ranges be defined Biochemical assessment. Measurement of both serum GH and for each assay. Harmony between assays is desirable, for IGF-I levels should be undertaken. Currently, data linking example, for epidemiological studies, but for the individual IGF-I and mortality are scarce. Good data exist linking ran- patient it is not necessary as long as the analytical technique dom GH levels to mortality (2). Serum GH concentrations of 2.0 ␮g/liter, determined by a traditional RIA, have been The emerging use of the GH receptor antagonist for the associated with reversal of the increased mortality of the treatment of acromegaly highlights the need for a rigorous disease. The equivalent level in modern two-site assays is IGF-I assay system, because GH measurements cannot be used to evaluate treatment efficacy. The majority of IGF-I For each GH assay, normative data for glucose-suppressed circulates as a ternary complex bound to acid-labile subunit GH concentrations are necessary for conclusions about ad- and IGF-binding protein 3, both of which are GH dependent.
equate control in individual patients. To define restoration of Measurement of these peptides offers no advantage to IGF-I normal neuroregulation of GH secretion, glucose suppres- in the assessment of disease activity, except in unusual sion of GH should be measured. This may correspond to levels as low as 0.3 ␮g/liter in a two-site assay with mono-clonal antibodies. In addition, the plasma IGF-I level should be within the age-adjusted normal range. Normal neurose-cretory dynamics, as assessed by an OGTT, may not be es- Before the availability of IGF-I assays, numerous dynamic sential for reduction of mortality risk.
tests, such as the GH response to an oral glucose tolerance Discordant values for random GH and IGF-I may be en- test (OGTT), GnRH, and TRH, were used in the diagnosis countered in up to 30% of the patients. In such cases an OGTT and follow-up of acromegaly. Of all of these dynamic tests, should be performed to properly assess a nadir GH level. In the OGTT has stood the test of time. Despite its utility, the the presence of discordant GH and IGF-I levels, therapy may lack of GH suppression in response to oral glucose is not be indicated depending on the clinical symptoms of active specific for acromegaly, because other conditions, such as acromegaly and the presence of comorbidities, such as glucose puberty, pregnancy, hepatic and renal disease, anorexia ner- intolerance, sleep apnea, hypertension, or cardiac dysfunction.
vosa, and diabetes mellitus, also cause inadequate GH sup-pression. In patients with overt diabetes mellitus the OGTT Cardiovascular and metabolic risk factors. Disease-specific met- should not be performed to diagnose acromegaly.
abolic and cardiovascular risk factors include hypertension, OGTT does not add diagnostic value when the IGF-I level glucose intolerance, hypopituitarism, and sleep apnea. These is clearly elevated, but serves for the assessment of carbo- risk factors should routinely be assessed at the time of di- hydrate intolerance. Comparing glucose-suppressed GH agnosis and during follow-up. The measurements include levels pre- and posttherapy may, however, add to the as- blood pressure, glucose metabolism (fasting blood glucose sessment of therapeutic outcome in individual patients. In- and hemoglobin A1c at a minimum), evaluation of residual dependently of IGF-I normalization after treatment, failure pituitary function, and evaluation of sleep apnea.
of GH to fall into the normal post-OGTT GH range, which has Cardiac abnormalities are prevalent in acromegaly. There J Clin Endocrinol Metab, July 2004, 89(7):3099 –3102 is a poor correlation of cardiac dysfunction with GH status.
apy. For this reason, at the present time it is recommended For this reason even patients with mild disease may be at that pituitary tumor size is closely monitored with MRI every risk. Appropriate cardiac evaluation is warranted when clin- 6 months during the first year of therapy and annually ically indicated. Awareness of additional established and modifiable risk factors, such as dyslipidemia, smoking, andobesity, also applies in acromegaly. Active management of persisting, modifiable cardiovascular and metabolic risk fac-tors should be included as a therapeutic goal.
Concerns about possible increased mortality from cere- brovascular disease in patients who have been treated with Follow-up recommendations in treated acromegaly. Surgery is external pituitary irradiation as well as the introduction of usually the first-line treatment for acromegaly. If surgery new medical therapies have narrowed the indication for does not achieve satisfactory disease control, further therapy Conventional multiple dose radiotherapy induces hypo- pituitarism frequently, so pituitary function should be as- Postsurgery. After surgery, GH and IGF-I should be measured sessed annually. The effects of single dose radiosurgery are as described above. If found to be normal, GH and IGF-I more rapid in onset and therefore should be evaluated at should continue to be monitored at least at annual intervals 6-month intervals. Pituitary hormone replacement therapy is lifelong. Recurrences may occur at any time and have been commenced according to good endocrine practice. In the documented in up to 10% of patients within the first 15 yr.
patient with cured acromegaly, GH replacement may be Biochemical or clinical evidence of recurrence necessitates mag- considered if symptoms of GH deficiency occur, and IGF-I is netic resonance imaging (MRI) evaluation of the pituitary.
below the normal age-adjusted normal range. However, datademonstrating the long-term benefit and safety of GH re- placement in such patients are scarce. In this context the use Dopamine agonists. These include cabergoline, bromocriptine, of IGF-I is recommended in the diagnosis of GH deficiency pergolide, or quinagolide. They provide adequate biochem- because stimulation tests may not be reliable in the diagnosis ical control in a minority of patients with acromegaly. It may take 3 months to achieve maximal suppression of GH and The decline in GH is exponential after conventional ra- IGF-I. If treatment is successful measurements should be diotherapy but fairly slow, so efficacy should be assessed at repeated annually. Side-effects may include nausea, occa- 2 yr and annually thereafter. After single dose radiosurgery, sional vomiting, orthostatic hypotension (particularly at ini- on the other hand, the effects on GH and IGF-I can be as- tiation of therapy), constipation, a Raynaud-like phenome- sessed after 1 yr. If the patient is receiving concomitant med- ical therapy, the assessment of residual disease activity canbe achieved either by dose reduction or discontinuation for Somatostatin analogs. These include formulations of oct- at least 3 months. For patients taking dopamine agonists, 1 reotide and lanreotide. The achievement of satisfactory GH month off treatment is sufficient for assessing residual ac- and IGF-I levels occurs in up to 60% of patients, but is tivity. Evaluation of tumor size by MRI is recommended in inversely related to pretreatment GH levels. Dose titration is patients with persistent disease after radiotherapy.
indicated. Once stable control is achieved, measurements canbe repeated annually. The inhibition of insulin secretion by somatostatin analogs may cause temporary and reversibledeterioration in glucose tolerance, so fasting blood glucose Controversy exists over the development of colonic polyps and subsequent progression to colonic cancer in acromegaly.
doubt persists, an OGTT for the assessment of glucose tol- Most studies do not support these concerns. Therefore, erance can be performed. Gall stones occur on somatostatin colonoscopy should be performed according to conventional analogs because of loss of gall bladder motility and other guidelines for screening colonic cancer. This should mean mechanisms, but only rarely does their development cause that colonoscopy is undertaken in patients with acromegaly symptoms, so routine gall bladder ultrasonography is not at the age of 50 yr. Other cancers are not known to have an increased incidence in acromegaly, so screening for breastand prostate cancer should be undertaken as in the normal GH receptor antagonist. This drug suppresses IGF-I to normal in over 90% of patients, whereas circulating GH values mayrise over the first few weeks, but then plateau. For patientstaking GH receptor antagonists, only IGF-I is measured for assessment of disease activity, because measurement of en- A number of important issues have been assessed by this dogenous GH levels by conventional assays is not possible.
consensus conference. The epidemiology discussion high- After dose titration, IGF-I should be measured every 6 lighted the difficulty of defining target GH and IGF-I levels months. Abnormalities of liver function have occasionally because of changing assay methodology. As the field moves been described, so monthly monitoring of liver enzymes over forward, results for IGF-I should be reported as both absolute the initial 6 months of therapy is required and less frequently concentrations and sd scores defined against a well vali- thereafter. Occasional patients have been described whose dated, age-adjusted, normal range. As a consequence, future tumors have enlarged during GH receptor antagonist ther- studies would not be confined by changes in assay method- J Clin Endocrinol Metab, July 2004, 89(7):3099 –3102 ology. The pulsatile nature of GH secretion hinders the es- Research, London, UK), Dr. Christian J. Strasburger (Charite´ Humboldt tablishment of meaningful reference ranges for random GH University, Berlin, Germany), Dr. Michael O. Thorner (University of levels. Standardization of normal response to glucose toler- Virginia Health Sciences Center, Charlottesville, VA), Dr. Mary LeeVance (University of Virginia Health Sciences Center, Charlottesville, ance tests, however, is achievable and remains a worthy goal VA), Dr. Klaus von Werder (Schlosspark Klinik, Berlin, Germany), Dr.
for every GH assay. The use of recombinant reference prep- Pamela U. Freda (Columbia University, New York, NY), Dr. Ian Hold- arations now allows the use of mass units for reporting of GH away (Auckland Hospital, Auckland, New Zealand), Dr. Gudmundur and IGF-I values, which is strongly endorsed.
Johansson (Sahlgrenska University Hospital, Goteborg, Sweden), Dr.
In the future, we believe that studies will clarify whether Jens Otto L. Jorgensen (Aarhus Kommunehospital, Aarhus, Denmark),Dr. John Wass (Churchill Hospital, Oxford, UK), Dr. Ariel Barkan (Uni- a normal IGF-I level in treated acromegalic patients will be versity of Michigan, Ann Arbor, MI), Dr. Thierry C. Brue (Hopital de la reflected in restoration of normal standardized mortality Timone, Marseilles, France), Dr. Kazuo Chihara (Kobe University School rates. The remaining question is whether the lack of adequate of Medicine, Kobe, Japan), Dr. Annamaria Colao (Federico II University, suppression of GH in response to glucose in the face of a Naples, Italy), Dr. Wouter W. de Herder (Erasmus Medical Center,Rotterdam, The Netherlands), Dr. Saul Malozowski (NIDDK, NIH, Be- normal IGF-I level will presage recurrence or a worse thesda, MD), Dr. Mark E. Molitch (Northwestern University Medical School, Chicago, IL), Dr. Hans Orskov (Aarhus Kommunehospital, Aar- General population mortality rates have fallen, and there- hus, Denmark), Dr. Jochen Schopohl (Medizinische Klinik Innenstadt, fore the reference rate is changing against which the acro- Munich, Germany), Dr. Stephen M. Shalet (Christie Hospital, Manches- megaly mortality rate is compared. Furthermore, the risk of ter, UK), Dr. Paul M. Stewart (Queen Elizabeth Hospital, Birmingham,UK), Dr. Brooke Swearingen (Massachusetts General Hospital, Boston, death from cancer in patients with acromegaly is now ac- MA), Dr. Peter J. Trainer (Christie Hospital, Manchester, UK), Dr. Bengt knowledged not to be increased. It is possible that the ap- Anderberg (Ferring Pharmaceuticals A/S, Copenhagen, Denmark), Mrs.
parent reduction in premature death from acromegaly may Liz Baister (Novartis Pharmaceuticals UK, Camberley, UK), Dr. Keith E.
be methodological or may result from earlier diagnosis and Friend (Pharmacia Corp., Bedminster, NJ), Dr. Philip E. Harris (PfizerLtd., High Wycombe, UK), Dr. Mark L. Hartman (Eli Lilly & Co., In- improved treatment. The complications of acromegaly give dianapolis, IN), Dr. Simona Ispas-Jouron (Lilly France, Suresne, France), rise to significant morbidity. The relationship between GH Dr. Anne-Marie Kappelgaard (Novo Nordisk A/S, Bagsvaerd, Den- status and these complications and the mechanisms by which mark), Dr. Kira Weirum Knudsen (Ferring Laegemidler A/S, Copen- they arise remain to be elucidated through timely, well con- hagen, Denmark), Dr. Tomoicho Konno (Sumitomo Pharmaceuticals, Tokyo, Japan), Dr. Ernest Loumaye (Beaufor Ipsen International, Paris,France), Dr. David Musgrave (Ipsen Scandinavia A/S, Brondb, Den-mark), Dr. Cle´ment Olivier (Serono International SA, Geneva, Switzer- Acknowledgments
land), Dr. Alberto Pedroncelli (Ospedali Riuniti, Bergamo, Italy), and Dr.
Akira Shimatsu (Kyoto National Hospital, Kyoto, Japan).
Participants in the Joint Workshop by The Growth Hormone Research Society and The Pituitary Society on Biochemical Assessment and Long-Term Monitoring in Patients with Acromegaly, held in Feldafing, Ger- Received July 3, 2003. Accepted March 19, 2004.
many, April 30 to May 3, 2003, were invited by a joint program com- Address all correspondence and requests for reprints to: Christian J.
mittee with equal representation of council members from both Strasburger, M.D., Division of Endocrinology, Department of Medicine, Societies. The attendees were: Dr. Bengt-Ake Bengtsson (Sahlgrenska Campus Charite´ Mitte, D-10098 Berlin, Germany.
University Hospital, Goteborg, Sweden), Dr. Lena Carlsson (Sahlgren-ska Academy, Goteborg, Sweden), Dr. Jens Sandahl Christianssen (Aar-hus Kommunehospital, Aarhus, Denmark), Dr. David Clemmons (Uni- References
versity of North Carolina, Chapel Hill, NC), Dr. Lawrence Frohman 1. Freda PU 2003 Current concepts in the biochemical assessment of the patient
(University of Illinois, Chicago, IL), Dr. Ken Ho (Garvan Institute of with acromegaly. Growth Horm IGF Res 13:171–184 Medical Research, Sydney, Australia), Dr. Primus E. Mullis (Inselspital, 2. Holdaway IM, Rajasoorya CR, Gamble GD, Stewart AW 2003 Long term
Bern, Switzerland), Dr. Iain Robinson (National Institute for Medical treatment outcome in acromegaly. Growth Horm IGF Res 13:185–192 JCEM is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the
endocrine community.

Source: http://www.ghresearchsociety.org/files/GHRS_2006_Acromegaly.pdf