Commentary

Acromegaly is an insidious disorder characterized by excess secretion of growth hormone (GH) generally caused by a pituitary adenoma resulting in elevations of circulating levels of GH and insulin-like growth factor (IGF-1).1 Although it is a rare disease, uncontrolled acromegaly is associated with reduced life expectancy due to severe systemic complications.2,3 Moreover, safe4 and effective5 medical treatments are now available in clinical practice that may be used as adjunctive treatments when neurosurgery,6 which remains in the majority of patients with acromegaly the first therapeutic option, fails to control the disease (in about 50% of patients).7,8 With these considerations, taken together, it is quite obvious why the definition of biochemical control of acromegaly has become one of the critical issues in the management of this disease. In this regard, the first historical step has been the Cortina Consensus Conference, which in 1999, for the first time, established general criteria for universal use.9 In fact, at that time one of the main problems for the scientific and clinical community was the wide and unacceptable variability in the criteria arbitrarily used in different settings. Moreover, the Cortina Consensus defined, for the first time, the concept that both GH and IGF-I should be “normalized” for a complete control of disease.9 In fact, in acromegalic patients GH and IGF-1 could give different but complementary information, because one parameter (GH) reflects the presence of residual adenomatous tissue and the other (IGF-1) the peripheral biological activity of the hormone secreted by the adenoma. From the clinical point of view, moreover, the normalization of GH and IGF-I have been shown to be tightly linked to normalization of the high mortality risk associated to acromegaly.10 However, in real life a relevant percentage of patients with acromegaly (up to one third), either after neurosurgery or during the course of the widely used medical treatment with somatostatin analogs,7 may show discrepant serum GH and IGF-I levels,11 ie, one parameter falls into the normal range whereas the other still remains elevated. Several mechanisms may be involved in the determinism of this discordance between GH and IGF-I in acromegalic patients including assay-related factors.12 Clearly, the timing of the biochemical assessment, after neurosurgery or for assessing the results of a given dose of a pharmacologic treatment, is a critical factor.13,14 In fact, it is well known that despite a very rapid fall of basal and after glucose load (OGTT) nadir GH, which occurs after successful neurosurgical removal of the adenoma, circulating IGF-1 levels may decline much more slowly.15 This is due to the very different half-lives of the two molecules but also to the slow restoration of normal GH neuroregulation, also depending on surrounding endocrine milieu and coexistent hypopituitarism,16 and recovery of a normal GH receptor sensitivity.15 Consequently, the most recent consensus guidelines recommend that only after a 3 to 6 month period following neurosurgery the biochemical outcome in terms of IGF-1 may be fully evaluated.14 Interestingly, in this issue of the Journal17 a study described the post-surgical outcome of serum IGF-I in 46 acromegaly patients who were retrospectively selected for a rapid (within 3 days) normalization of the GH inhibition after OGTT together with no evidence of residual adenoma tissue at magnetic resonance imaging (MRI) performed 3 to 6 months after surgery. In these cases, based on the above findings, it was decided to monitor the patients without initiating any medical treatments. Monitoring of IGF-I was initially performed quite intensively (at three month intervals for the first year after neurosurgery) and then every 6 or 12 months. The first important finding of this study is that, in all the patients evaluated, IGF-1 did normalize during follow-up.17 This confirms the clinical relevance of the early assessment of the patients after neurosurgery with the GH response to OGTT which remains a mainstay in the biochemical evaluation of acromegaly in this specific setting.9,18 Moreover, this study clearly demonstrates the clinical appropriateness and somewhat validates the novel cut-off proposed in the last consensus guidelines (GH nadir after OGTT below 0.4 ng/mL).14 In fact, this cut-off was recently made more strict based on several reports of discrepant GH and IGF-1 levels11 using the old cut-off (GH inhibition after OGTT below 1 ng/mL).9 The observation that all patients with OGTT GH inhibition below 0.4 ng/mL also underwent a normalization of IGF-1 during follow-up is strongly reassuring concerning the clinical applicability of the newly recommended criteria.14 The second relevant information given by this study17 is that in the majority of patients (59%) with early normalized GH response to OGTT IGF-1 returned to normal (and so remained at long-term follow-up) also very rapidly, ie within three months after neurosurgery. This finding is of clinical relevance since it again confirms what is recommended in the recent guidelines,14 which is that clinicians must wait at least 3 months after neurosurgery to get a correct evaluation of the binary biochemical criteria (both GH and IGF-1) and before making any decisions on the need of additional medical treatments. Obviously, this can be also considered a quite short period of time in terms of potential residual disease activity and consequent progression of complications19,20 (particularly for patients with normally OGTT suppressed post-surgical GH levels) to be safely recommended as an acceptable delay in the decision-making process. Consistent with the consensus recommendations, in the study there were some patients in whom IGF-1 normalized from the 3rd to the 6th month after surgery (the exact number of patients with this behavior cannot be exactly extrapolated by the data presentation). In fact, the last guidelines14 suggest that the observation period to correctly evaluate IGF-1 may range from 3 to 6 months after surgery. The third aspect of the paper,17 being the more controversial in terms of therapeutic decision-making, is that a relevant number of patients normalized their IGF-1 only after 6 months of follow-up, and a quarter of the overall study population obtained normal age-matched IGF-1 values more than 12 months after neurosurgery (in a few cases after 3-5 years). This is an interesting finding, although limited in absolute numbers due to the overall relatively small sample size of the study population (n. 46), since previously reflected only by few other clinical studies.21 In fact, while contemplating the possibility of a long-term (>6 months after surgery) persistent discrepancy between GH and IGF-1 values, the current guidelines do not give specific recommendations on the follow-up and management of these cases.14 The Authors did show that only high postoperative IGF-1 values were good predictors of what they call “late” (>3 months) normalization of IGF-1 during follow-up.17 This finding may be consistent with the presence of a polymorphic GH receptor in this subpopulation of patients. In fact, it is now well known from the study of Bianchi et al11 that at least half of the acromegaly, as well as the general, population does express an isoform of the GH receptor that is more sensitive to the hormone than the wild type receptor. Therefore, for a given GH value the presence of a polymorphic GH receptor does translate in higher IGF-1 levels with respect to the wild type receptor.11 We do not have information on the GH receptor polymorphism in this study population, but from previously published data we know that most of the type of discrepancy reported in the study (normal GH vs high IGF-1) that is observed after neurosurgery can be accounted by this receptor polymorphism.11 There is also no clear explanation of why patients potentially bearing the GH receptor polymorphism might undergo normalization of IGF-1 late during follow-up. However, one could hypothesize that these subjects even when returning to a physiological GH secretion22 may maintain IGF-1 in the high-normal range and therefore easily be confounded (also due to the assay variability and to the lack of sufficient normative data in specific populations from in house experiences12,23) with not fully biochemically controlled subjects. However, beside these open pathophysiological insights, the clinical message that derives from this observation is that in the presence of a physiological suppression of GH after surgery, patients with acromegaly should not undergo aggressive treatments even in the absence of a normalization of IGF-1 levels.17 Interestingly, these findings may also shed new light on the possible mechanisms underlying increased rate of normalization of IGF-1 over time of treatment with somatostatin analogs24 and on the possibility of adjuvant treatment with somatostatin analogs withdrawal in some patients with acromegaly.25,26 However, some potential drawbacks do seem to exist that may somewhat limit the wide clinical applicability of this conservative and undoubtedly apparent cost-effective indication which is based on a sort of supremacy of the post-glucose GH nadir as reliable marker of disease remission (in concomitance with the MRI absence of tumor remnant). In fact, although IGF-I finally normalized in all patients, some of them maintained high IGF-1 values for years during the follow-up.17 Therefore, besides the fact that these patients did not fully meet the binary criteria for disease control for a long time, one could argue that in the presence of discrepant hormonal values, elevated IGF-I could be more predictive for progression of complications and persistence of clinical symptoms than both random and nadir GH.27,28 Therefore, for how long could it be recommended to wait before initiating a pharmacologic treatment in a patient with acromegaly and persistently elevated for age IGF-1 (although with normal GH)? Also, are there any possible clinical or biochemical criteria that may guide toward a more conservative vs aggressive approach? Obviously, in the absence of clinical studies addressing these issues, it is not possible to answer these questions based on evidence driven data. However, based on clinical experience one may suggest that, in patients with prompt GH normalization after surgery, modestly increased IGF-1 absolute values concomitantly with a trend towards a decrease of IGF-1 over time as well as remission of symptoms and absence or good control of complications may allow the clinician to maintain a conservative approach based only on biochemical follow-up for more than 6 months (current guidelines) even if at present, and until data of the current study17 will not be confirmed by much larger experiences, there is insufficient evidence to suggest that this period can be extended over 12 to 18 months. After this time period or before, in the presence of stable and high IGF-1 and of a clinically active disease, the clinicians (endocrinologist and neurosurgeon working together in team) should consider, in the absence of reliable predictors of late IGF-1 normalization, initiating treatment with somatostatin analogs since the danger of a persistently even modestly active or subtle acromegaly (this emphasizes that therapeutic decision-making is based not only on biochemical parameters but also on an accurate clinical staging of acromegaly) may be evaluated against the risks and costs of medical therapy. In this regard, the novel message coming from the study published in this issue of Neurosurgery17 is that even in the minority of such patients in whom we initiate treatment, we have to keep in mind that spontaneous normalization of IGF-1 may occur also in the very long-term and this may be relevant in the choice of the schedule for the somatostatin analog treatment (start with low dose or tapering down with rapid normalization of IGF-1). Alternatively, since we are in the presence of apparently normalized GH secretion but a persistently increased GH bioactivity in this subset of patients, the GH antagonist pegvisomant29 may find an indication with the possibility of a progressive tapering of dose until drug withdrawal. In conclusion, data published in this issue of Neurosurgery17 are generally reassuring on the clinical effectiveness of current guidelines of biochemical assessment and follow-up of acromegaly. Caveat concerning the potential normalization of IGF-1 long-term after surgery should be taken into account, although it needs to be confirmed in large prospective studies, in designing appropriate medical treatment strategies and future guidelines. Disclosures Dr Giustina is consultant for Ipsen, Novartis and Pfizer. Dr Mazziotti received consultant fee from Novartis and received lecture fee from Ipsen. Dr Fontanella does not have conflicts of interest.