Department of Medicine, Columbia University College of Physicians and Surgeons, New York, New York 10032
Address all correspondence and requests for reprints to: Pamela U. Freda, M.D., Department of Medicine, Columbia University College of Physicians and Surgeons, 630 West 168th Street, New York, New York 10032.
Since their introduction into clinical practice, somatostatin analogs have been the medical therapy of choice for the treatment of acromegaly. Therefore, considerable data exist on their use for this purpose. This review summarizes the literature on somatostatin analog therapy of acromegaly and describes its efficacy in terms of the main goals for the treatment of acromegaly, i.e. normalization of hormone levels, GH and IGF-I, relief of the signs and symptoms of acromegaly, and pituitary tumor shrinkage. The side effects of somatostatin analogs and their role as primary therapy for acromegaly are also discussed. The review focuses on the data with the depot analogs because these formulations are most likely to be chosen in clinical practice.
Biology of somatostatin and somatostatin analogs
The native peptide somatostatin is widely distributed in the central nervous system and peripheral tissues (1, 2, 3). Somatostatin is processed from its precursor into its two biologically active forms, somatostatin-14 and somatostatin-28 (2). Native somatostatin has diverse physiological actions, including its role as a central nervous system neurotransmitter and neuromodulator, regulatory hormone in the gastrointestinal tract and pancreas, and inhibitor of GH and TSH release in the pituitary (2, 3). In vitro, native somatostatin retains its inhibitory effect on GH secretion in many GH-secreting tumors, and this led to the development of analogs of somatostatin for clinical use in the treatment of acromegaly (4). The two analogs of somatostatin available for clinical use are the cyclic octapeptides octreotide (Dphe-cys-phe-Dtrp-lys-thr-cys-thr-ol) and lanreotide (Dnal-cys-tyr-Dtrp-lys-val-cys-thr) (1, 5, 6, 7). Octreotide is the only analog currently available for clinical use in the treatment of acromegaly in the United States. Lanreotide has been used extensively in Europe for the treatment of acromegaly. Certain properties of these analogs, including their greater potency for GH suppression and longer half-life after peripheral administration than somatostatin, make them advantageous to the native hormone for the treatment of acromegaly. For example, octreotide is 45 times more potent at suppressing pituitary GH secretion than native somatostatin-14 (8) and has a half-life of 2 h when given sc (9).
Somatostatin analogs and native somatostatin elicit their biological effects by activating somatostatin receptors. There are five distinct somatostatin receptors, types 15 (10). They are all G protein-coupled receptors, but differ in their regulatory/signaling pathways, tissue distribution, and the affinity to which the somatostatin analogs bind to them (6, 10). Relative to native somatostatin, the analogs have a narrowed spectrum of receptor activity that allows for greater specificity of GH suppression. Octreotide and lanreotide have greatest affinity for receptor subtypes 2 and 5, with their affinity for subtype 2 being about 10-fold higher than for subtype 5 (1). Receptor subtypes 2 and 5 are those through which endogenous somatostatin suppression of GH occurs (11) and are also the predominant types of somatostatin receptors found in GH-secreting pituitary tumors (12, 13, 14, 15). A significant percentage of GH-secreting pituitary tumors are resistant to octreotide and lanreotide, and this may be explained in part by variable tumoral expression or reduced receptor density of subtypes 2 or 5 on the tumors of these patients (16).
Somatostatin analogs: pharmacodynamics
The first preparation of somatostatin analog available for clinical use was sc-administered octreotide. After sc injection, serum octreotide levels rise within 30 min and then fall over the next few hours. The maximal suppressive effect of octreotide on GH levels occurs between 2 and 6 h after the injection (17). GH levels rise between injections given every 8 h, but with continued treatment this rise is lessened (9). Importantly also, in somatostatin analog responsive patients, the suppressive effect on GH secretion does not wane with time. The typical dose of octreotide in its sc form is 100250 µg thrice daily, but doses up to 1500 µg over a 24-h period can be given (18).
Octreotide is also available in a long-acting release (LAR) preparation, octreotide LAR, in which octreotide is enclosed in microspheres of a slowly biodegrading polymer that allows for prolonged release of drug (19). After the injection of octreotide LAR, octreotide levels rise briefly, corresponding to release of octreotide from the surface of the microspheres. Octreotide levels then fall and begin to rise again about 714 d later after the injection and remain elevated for an average of 34 d (20). Steady-state conditions are usually achieved after two to three injections. GH levels fall and rise in response to the changing octreotide levels (20, 21). The usual starting dose of octreotide LAR is 20 mg with titration down to 10 mg or up to 30 or 40 mg, based on the response of GH and IGF-I levels. The manufacturer recommends monthly drug administration, allowing for an overlap of injections that will maintain sufficiently high octreotide levels. However, the effect of LAR may be prolonged enough to maintain GH and IGF-I suppression up to 6 or 8 wk after an LAR injection, allowing lengthening of the dosing interval beyond every 4 wk in some patients (22, 23).
Lanreotide is also available in a slow release (SR) preparation where lanreotide is encapsulated in microspheres of biodegradable polymer. After injection of lanreotide SR, lanreotide levels remain elevated for about 11 d (24). Thus, lanreotide has a shorter duration of suppression of GH levels than octreotide LAR, which requires lanreotide injections to be given at 1014 d intervals, but the interval may need to be decreased or can be increased in some patients (22).
Criteria for assessing efficacy of somatostatin analogs
The efficacy of somatostatin analogs for the treatment of acromegaly needs to be assessed by a number of important criteria. First, efficacy should be assessed in terms of biochemical control of GH and IGF-I levels. In most studies, GH control has been defined as GH levels obtained at random or from a mean of eight hourly samples of less than 2.0 or 2.5 µg/liter and in a fewer number of studies as glucose- suppressed GH levels of less than 1.0 µg/liter. It is important to note that with newer highly sensitive GH assays, GH levels less than 2.5 µg/liter do not necessarily ensure normalization of IGF-I levels and biochemical control in all patients (25). However, epidemiological data have shown that a GH level less than 2.5 µg/liter (when GH is measured by polyclonal RIA) is associated with a reduction of the mortality in acromegaly (26, 27), and therefore this is the GH cutoff used in most somatostatin analog studies. Efficacy should also be assessed as normalization of IGF-I level that is increasingly recognized as an essential criterion for cure of acromegaly (28, 29). Normalization of IGF-I level has been associated with a reduction of the excess mortality in acromegaly (30). Because most patients with acromegaly have macroadenomas at diagnosis, another crucial aspect of somatostatin analog therapy for acromegaly is their effect on pituitary tumor size. Other important aspects of somatostatin analog therapy for acromegaly are their effect on the clinical signs and symptoms of acromegaly, the side effects of the analogs, and their efficacy as primary therapy.
Biochemical control
Although most patients treated with somatostatin analogs will have some fall in GH levels, many fewer patients have persistent suppression and normalization of GH levels. In data combined from many studies of patients treated with depot somatostatin analogs as adjunctive therapy, GH levels were suppressed in 56% of patients treated with LAR (19, 20, 31, 32, 33, 34) and 49% of those treated with lanreotide SR (Refs. 7, 24, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44 ; Table 1). Considerable variability does exist among the data from different studies (Table 1
). This variability may reflect the wide ranges in number of patients studied (8149 patients) and in the duration of treatment (636 months). Analysis of data from many studies shows that IGF-I levels with long-term somatostatin analog therapy were normalized in 66% of those treated with octreotide LAR (19, 20, 31, 32, 33, 34) and 48% of patients who received lanreotide therapy (7, 24, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45). The percentage of patients achieving IGF-I normalization varied among studies (Table 1
). It is important to point out that approximately 90% of patients in the trials with octreotide LAR were preselected for octreotide responsiveness. In most cases, this was determined as acute suppression of GH to less than 5 µg/liter after sc injection of octreotide. By contrast, only about 10% of patients were reported to be similarly selected in lanreotide studies. The extent to which the selection bias in some studies influenced the reported efficacy figures is unknown.
|
A number of studies have compared the efficacy of sc octreotide and octreotide LAR. These studies, for the most part retrospective, have reported a similar efficacy for the two preparations (36, 39, 43, 44, 50). In clinical practice, however, it is expected that greater patient compliance with the depot form will result in better control with this formulation.
The efficacy of somatostatin analogs in combination with the dopamine agonists bromocriptine or cabergoline has been examined in a few small studies. Overall, most studies, although not all, have shown that 1020% of somatostatin analog-resistant patients have some further suppression of GH and/or IGF-I levels with the addition of a dopamine agonist to somatostatin analog therapy (51, 52, 53, 54, 55). The comparative efficacy of a somatostatin analog combined with bromocriptine vs. cabergoline has not been assessed.
An improvement in the signs and symptoms of acromegaly occurs overall in 6474% of patients treated with depot analog therapy (7, 20, 34, 37, 39, 41, 42, 44, 45). Studies have reported improvements to varying degrees in headache, soft tissue swelling, arthralgia, carpal tunnel syndrome, snoring, hyperhidrosis, fatigue, and malaise. With depot somatostatin analog therapy, as has been shown for sc octreotide, more patients subjectively report improvement in the signs and symptoms of acromegaly than the number of patients whose GH/IGF-I levels normalize. Symptomatic improvement is likely due to lowering without complete normalization of GH/IGF-I levels in such patients. Objective improvement in the clinical manifestations of acromegaly is also an important aspect of treatment of acromegaly. Manifestations of cardiovascular disease, the leading cause of mortality in these patients, can improve with somatostatin analog therapy. A number of studies have shown improvements with somatostatin analog therapy in cardiac structure and function, including left ventricular mass index, left ventricular hypertrophy, and ejection fraction (56, 57, 58, 59, 60). Normalization of IGF-I has also been associated with improvement in cardiac performance on octreotide (61). An improvement in sleep apnea can also occur in some patients with acromegaly treated with octreotide (62).
Tumor shrinkage
Another very important component of the efficacy of somatostatin analog therapy for acromegaly is its effect on tumor shrinkage (Table 2). Overall, in combined data from patients receiving either the long-acting analogs or sc octreotide as adjunctive therapy, about 30% of patients had tumor shrinkage (19, 24, 31, 32, 33, 35, 37, 40, 41, 42, 43, 59, 63, 64, 65). Tumor size reduction was most often reported to be between 20 and 50%. Tumor shrinkage data were available from trials in which approximately 90% of patients treated with octreotide LAR and about 10% of those treated with lanreotide SR were preselected for octreotide responsiveness. As a result, we do not know whether the tumor shrinkage statistics would differ in the unselected overall population of patients with acromegaly treated with somatostatin analogs. Fewer than 1% of tumors were reported to enlarge on somatostatin analogs, suggesting that clinically apparent tumor growth may be slowed with this therapy even in patients whose GH and IGF-I levels are not normalized. However, other factors, including prior radiotherapy and follow-up of not more than 34 yr in the studies to date, could have played a role in the apparent lack of tumor growth reported in patients treated with somatostatin analogs.
|
Another important, but somewhat controversial use of somatostatin analogs is as primary therapy for acromegaly. Most of the data reported on primary somatostatin analog therapy of acromegaly are from patients treated with sc octreotide (n = 188 of 227, 83%), and only a small number of reported patients have received primary treatment with lanreotide SR (n = 22) or octreotide LAR (n = 17). Only about 6% of patients in these trials were reported to be preselected for octreotide responsiveness. Overall, when the data from all analogs are combined, IGF-I normalized in 60% of patients, and GH was suppressed in 50% of patients who received primary somatostatin analog therapy (Table 1) (32, 35, 36, 39, 44, 66, 67, 68, 69, 70, 71). In general, these figures are similar to those in studies where the analogs are used as adjunctive therapy. In one retrospective analysis of patients treated with octreotide, GH and IGF-I levels were suppressed to the same degree in a group of patients who received octreotide as primary therapy and in a group who had failed transsphenoidal surgery and who received octreotide as adjunctive therapy (71).
The effect of primary therapy on tumor shrinkage is also clearly very important. Studies assessing tumor shrinkage with primary somatostatin analog therapy have reported 48% of patients to have some tumor shrinkage (Table 2; Refs. 19, 32, 36, 37, 39, 44, 66, 67, 69, 70, 71, 72, 73, 74). In general, most tumor shrinkage was between 20 and 50% (32% of patients). It should be noted, however, that primary somatostatin analog therapy cannot be relied upon to produce enough tumor shrinkage to relieve visual compromise due to optic chiasm compression in patients with acromegaly (36).
There are a number of important considerations for the use of somatostatin analogs as primary therapy for acromegaly. The presence of visual or neurological compromise is almost always an indication for surgery because analog therapy may not provide enough shrinkage to relieve pressure on the optic chiasm. Tumor size and degree of invasiveness are very important considerations. If surgery is performed by an experienced surgeon, microadenomas have a high rate of surgical cure, up to 8590%, far better than the success rate of somatostatin analog therapy (30, 75). Again, in the hands of an experienced surgeon, small, well circumscribed macroadenomas also have a cure rate of 6570% (75). For large and/or invasive tumors, surgery is not likely to be curative, and thus a 5060% normalization rate of IGF-I with analog therapy is greater than that achieved with surgery in many of these patients. However, consideration needs to be given to how not removing the tumor mass could impact on the options for further therapy. For example, radiotherapy may be limited if the residual tumor is near the optic chiasm and medical therapy may be inadequate if significant tumor growth occurs. The patients surgical risk is a consideration, and in elderly patients or others with a high risk, primary medical therapy may be more strongly considered. The risks of transsphenoidal surgery itself in experienced hands are very small. Clearly, the experience of the pituitary surgeon is crucial, and whether or not one is available at a particular center plays a major role in the endocrinologists recommendations. All of these considerations for the use of primary somatostatin analog therapy in acromegaly clearly need to be weighed on an individual patient basis.
Given greater patient compliance and at least equivalent efficacy of the depot formulation to sc octreotide, the long-acting depot analogs are the formulations of choice for use in clinical practice. Octreotide LAR is in clinical use in the United States and requires less frequent dosing than lanreotide SR. Short-acting sc octreotide still has a role in assessing patient tolerability to somatostatin analogs where it is given as a 2-wk course before initiation of depot analog therapy. The sc octreotide may also be useful when rapid lowering of GH is needed, but the short octreotide test of acute GH suppression is not adequately predictive of long-term efficacy to warrant need for this test in clinical practice (76).
Another question that is frequently raised about the use of somatostatin analogs in acromegaly is their role as preoperative therapy. Some studies have not demonstrated an improved surgical remission rate in patients who receive preoperative somatostatin analog therapy (68, 72), but others have reported a benefit to pretreatment (67, 70). Another retrospective analysis found improved clinical conditions, including preoperative blood pressure, and shorter hospitalization to be associated with use of preoperative octreotide (66).
Depot analogs: side effects
The side effects of depot somatostatin analogs are similar to those of short-acting octreotide (Table 3; Refs. 7, 35, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 59). The most common side effects are gastrointestinal with up to half of patients initially experiencing diarrhea, nausea, or abdominal discomfort, but in most patients these are transient. New gallstones were reported to occur with long-acting analog therapy, most often within the first year of therapy, in about 15% of patients. Additional patients may develop gallbladder sludge or microlithiasis with long- acting somatostatin analog use. Most patients remain asymptomatic. The routine use of surveillance gallbladder ultrasonograms in asymptomatic patients is not felt to be necessary, but symptomatic patients should be managed as clinically appropriate. In general, side effects do not limit therapy, but a number of studies reported 35% dropout rates, in one study up to 25% (38), with depot somatostatin analog therapy for a variety of reasons, including lack of efficacy or side effects.
|
In summary, clinically available somatostatin analogs control GH or IGF-I excess in about 5060% of patients whether used as primary or secondary therapy. Signs and symptoms of the disease improve in most patients. Tumor shrinkage occurs with somatostatin analogs used as adjunctive therapy in about 30% of patients and with their use as primary therapy in about 48% of patients. The shrinkage in most patients is greater than 20%, but less than 50% of tumor size. The pros and cons of primary somatostatin analog therapy for acromegaly need to be considered carefully in each individual patient. Future directions for somatostatin analog therapy of acromegaly include the possible development for clinical use of somatostatin receptor selective analogs that have been shown in in vitro studies to have greater affinity for receptor subtypes 2 and/or 5 than octreotide or lanreotide and could more effectively suppress GH in some patients (77, 78, 79, 80). In clinical trials, there is also a longer-acting preparation of lanreotide that would allow for monthly dosing of this analog (81). In the future, there may also be investigations into the combined use of somatostatin analogs and the GH receptor antagonist, pegvisomant. This combined therapy may be useful in normalizing IGF-I levels with pegvisomant in patients who are not completely responsive biochemically to somatostatin analogs, but who have rapidly growing tumors that seem to benefit from some suppressive effect of the analog on tumor growth (82).
Acknowledgments
I thank Dr. Sharon Wardlaw for helpful critique of the Xmanuscript and Mr. John Sowinski for assistance with manuscript preparation.
Footnotes
This work was supported by NIH Grant K08-DK02561 to P.U.F. Results from this work were presented in part in Symposium S582 at the 83rd Annual Meeting of The Endocrine Society, Denver, Colorado, June 23, 2001.
Abbreviations: LAR, Long-acting release; SR, slow release.
Received December 19, 2001.
Accepted March 22, 2002.
References