Combined treatment of somatostatin analogues with pegvisomant in acromegaly

Meta-analyses of clinical trials showed that LA-SSAs alone normalize GH and IGF-I levels in about 50 % of patients [4]. However, due to selection bias this efficacy rate is probably an overestimation. In unselected treatment-naive patients an LA-SSA efficacy rate of 40 % seems to be more common [15, 16]. PEGV appears to be more effective, both during monotherapy as during co-treatment with LA-SSA [7, 8, 11‐14, 17]. Table 1 shows a summary of studies reporting on efficacy of acromegaly patients using LA-SSA in combination with PEGV. The efficacy rates during a period of 9 years in a large Dutch cohort are shown in Fig. 1 [12]. Because PEGV is a competitive blocker of the GH receptor (GHR), pharmacology dictates that in principle it should be possible to control IGF-I levels in all patients with acromegaly, provided that the appropriate PEGV dose is used. LA-SSAs have direct and indirect effects that result in a GH-independent decrease of IGF-I production [18, 19]. A Danish study reported that PEGV serum levels increase by 20 % when combined with LA-SSA [20]. The appropriate PEGV dose varies among acromegaly patients, the Dutch cohort reported that patients using high dose of LA-SSA needed a median weekly PEGV dose of 80 mg (range: 30–300 mg) to achieve normal IGF-I levels in 97 % of the patients [12]. Monotherapy of PEGV requires a higher cumulative weekly dose of around 130 mg to achieve a similar normalization rate after 12 months of treatment [8]. In contrast to these reports, an Italian observational study reported no difference in median required PEGV dose to normalize IGF-I levels in patients using PEGV monotherapy compared to patients treated with the combination therapy [21]. However, these groups were not similar according to severity of the disease. The PEGV monotherapy group had significant lower GH and IGF-I levels at baseline compared to the combination group in this Italian observational study [21]. Observational registries such as the ACROSTUDY™ and German Pegvisomant Observational Study (GPOS) [9, 22], in which patients using PEGV are included regardless their concomitant medication, observed lower efficacy rates around 60 %, compared with the clinical trials of PEGV monotherapy and combination treatment mentioned previously [8, 12]. In the ACROSTUDY™ PEGV was combined with LA-SSA in 23 %, with dopamine agonist in 6 %, and a combination of the three agents in 4 % of patients [9]. After 5 years the mean weekly dose was 106 mg in patients with a normal IGF-I, and 113 mg in those with an elevated IGF-I [23]. However, these studies were not designed to evaluate efficacy and dose titration, but aimed for the evaluation of safety aspects, such as rare side effects. The lower efficacy rates in the observational registration studies might be explained by the relative lower dose of PEGV. To achieve efficacy rates of more than 90 % during PEGV monotherapy, the average expected weekly dose is probably above 120–130 mg.

Escape, defined as need to increase the dose of PEGV because of an increase in IGF-I levels, was reported in 34 % of a Spanish cohort [25]. The majority of patients were easily controlled with either an increase in PEGV dose, additional medical treatment or both. Whether this increase in the necessary dose of PEGV should be called an escape is questionable, as most patients were easily controlled and remained controlled on a higher dose of PEGV. A significant number of these patients escaped from PEGV within the first 6 months after discontinuation of LA-SSA. Presumably during this period these patients were actually still receiving combination treatment, due to the long half-life of LA-SSAs. The Dutch cohort could not observe this phenomenon during combination treatment in acromegaly patients [12].

Co-administration of the highest dose of LA-SSA on top of PEGV monotherapy appears to reduce the necessary mean PEGV dose by 50 % [14], although with a high individual variability [26]. An analysis including eight patients whose mean IGF-I levels were similar during PEGV monotherapy, showed that these patients were able to reduce their PEGV dose from 131.3 ± 36.2 to 62.5 ± 16.7  mg weekly [14]. PEGV dose reduction during combination treatment might improve cost-effectiveness of medical treatment in acromegaly and may reduce injection frequency for patients. However, there are no direct studies comparing cost-effectiveness of the median required weekly PEGV dose during mono- and combination treatment.

Physicians tend to be mainly focused on biochemical parameters as GH and IGF-I levels during treatment of acromegaly. These parameters are definitely linked to a better outcome and a lower risk of morbidity and mortality [27, 28]. A recent French study has described long-term effects of PEGV on cardiorespiratory and metabolic comorbidities [29]. They reported that long-term PEGV treatment improves cardiac function measured by the left ventricular ejection fraction and the left ventricular mass index. PEGV also improved the apnea–hypopnea index. Normalized serum GH and IGF-I levels does not necessarily result in complete resolution of signs and symptoms [30, 31]. Neggers et al. reported on the results of a prospective, double blind, placebo controlled, crossover study, and demonstrated improved quality of life (QoL) in patients using LA-SSA combined with low-dose PEGV [31]. QoL was assessed by two acromegaly specific QoL-questionnaires, the AcroQoL and the PASQ. Improved QoL was observed without significant changes in IGF-I levels after addition of PEGV to LA-SSA therapy in patients with normalized IGF-I levels during monotherapy of LA-SSA. However, a study by Madsen et al. did not observe an improved QoL in patients using LA-SSA combined with a low dose of PEGV [32]. Two differences in the design of the studies might clarify these different outcomes in QoL; (1) Madsen et al. reduced the dose of LA-SSA by 50 % before adding low-dose PEGV treatment, while Neggers et al. did not change the LA-SSA dose; (2) disease specific QoL-questionnaires were used by Neggers et al. whereas Madsen et al. used QoL-questionnaires which were not disease specific.

An explanation for the observed improvement in QoL by the addition of low-dose PEGV could be a persistent systemic acromegaly disease activity during monotherapy of LA-SSA that has been hypothesized and was called ‘extra-hepatic acromegaly’ [33]. LA-SSA treatment selectively decreases hepatic-IGF-I production by around 20 % via direct and indirect mechanisms [18, 19, 34]. This might lead to an overestimation of the efficacy of LA-SSAs to normalize IGF-I via a reduction in the pathological GH secretion because of this GH-independent reduction of IGF-I levels. Therefore, GH actions on extra-hepatic tissues can remain elevated despite normalization of serum IGF-I levels. LA-SSA reduces portal insulin levels which decreases hepatic GHR expression only [34]. This results in a relatively hyper-GH-sensitive state of the other tissues. Patients report these still excessive GH actions on non-hepatic tissues such as edema, fatigue and headaches comparable to the side-effects of high-dose GH replacement therapy in GH deficient subjects. A clinical example is a Danish study which observed, despite similarly normalized IGF-I levels, that LA-SSA treatment compared with neurosurgery alone was associated with less suppressed GH levels and less symptom relief [35]. Blocking non-hepatic GH actions by low-dose PEGV could therefore be useful in treating this ‘extra-hepatic acromegaly’ [33]. Moreover, PEGV has also been shown to improve insulin resistance by several mechanisms [36‐41], which is beneficial in the presence of LA-SSAs, which are known to reduce insulin secretion [42, 43].

Recent prospective multicenter clinical trial observed tumor shrinkage in 63 % of primary treated patients with 120 mg Lanreotide Autogel administered every 28 days [44]. Tumor shrinkage was defined as clinically significant when ≥20 % tumor volume reduction was observed after 48 weeks of Lanreotide Autogel administration. While LA-SSA is successful in reducing tumor volume, previous concerns were raised whether PEGV might induce growth of the pituitary adenoma. Despite the fact that in a few cases an increase in tumor size during PEGV therapy was reported, there are no data suggesting that PEGV directly promotes tumor growth [14, 45, 46]. In the GPOS study, changes in tumor size were systematically monitored in 307 patients, from which 28 patients were treated with PEGV in combination with LA-SSA, however predominantly treated with PEGV monotherapy [47]. In eight out of 307 (2.7 %) an initial increase in tumor size was reported. In only three of these eight patients a real, but minor, increase in tumor size after PEGV treatment was observed [47]. In two of these patients a detectable rebound increase in tumor size after discontinuation of LA-SSA therapy was the probable reason of this increase in adenoma size. In a Spanish study in 75 patients, 5 (6.7 %) acromegaly patients were identified with an increase in pituitary tumor size [48]. All of these patients were pre-treated with LA-SSA and then switched to PEGV monotherapy. Noteworthy in this study is that the reference MRI was made just after LA-SSA was discontinued [48]. Therefore, the reported tumor size increases in this study may also be explained by the rebound phenomenon after cessation of LA-SSA treatment. In the Dutch cohort of acromegaly patients (n = 141) on combination therapy, growth of the adenoma has been reported in only one patient, while this growth was already observed before the addition of PEGV [12]. Moreover, during combined treatment tumor size shrinkage of more than 20 % of the largest diameter before and during combination treatment was observed in 17 % of the patients and the vast majority had a stable tumor volume. We, therefore, conclude that PEGV apparently does not influence the natural course of tumor growth, but ongoing alertness is required to monitor tumor size by repetitive pituitary imaging.

PEGV treatment is generally well tolerated and is increasingly considered to be safe [9, 49]. As LA-SSAs are on the market since 1988, more data are available on side effects. Most commonly reported are gastrointestinal symptoms such as nausea, vomiting, abdominal pain, biliary sludge or gallstones and diarrhea [50]. Abnormal glucose metabolism and injection-site reactions are also described [50]. Lipohypertrophy and hepatotoxicity are the most frequently reported adverse events during long-term PEGV treatment. Lipohypertrophy is an increased subcutaneous fat deposition around the injection sites of PEGV. It is suggested that the blockade of GH receptors causes unopposed insulin effects and, therefore, promotes lipogenesis [51]. The symptom arises in 3–15 % of the PEGV users either during monotherapy or combination treatment and appeared to be reversible by a more frequent rotation of the injection-site [12, 14, 21, 25, 52]. However, in one of the Dutch patients during combined treatment cosmetic surgery was necessary in order to reduce the significant lipohypertrophy [12]. Large observational studies as ACROSTUDY™ and GPOS, in which patients using PEGV are included regardless their concomitant medication, reported 2.2 and 7.4 % lipohypertrophy respectively.

Hepatotoxicity can occur during both PEGV monotherapy and combination therapy. Cholestasis is a common side effect of LA-SSA, as it reduces the secretion of cholecystokinin and motility of the gallbladder and induces an increase in bile concentration. This results in an increased risk in development of sludge and gallstones [53, 54]. LA-SSA often induces asymptomatic cholelithiasis, but an acute cholecystitis has been observed in only a minority of these patients [55]. The PEGV-induced elevations in hepatocellular enzymes are usually mild and self-limiting, both during monotherapy PEGV as in combination with LA-SSA [21, 52]. Transient elevated transaminases (TET) of more than three times the upper limit of normal (>3 ×ULN) were observed in 11.3–13.5 % of the patients using LA-SSA combined with PEGV [12, 21]. One Italian study compared long-term treatment of PEGV alone with PEGV in combination with LA-SSA regarding to TET [21]. Incidence of TET during monotherapy of PEGV was reported in 14.3 % of the patients, while this incidence was 11.1 % in the combined group. In the Dutchcohort (N = 19/141) all cases were transient without PEGV dose adaptation or discontinuation of the drug, except for one patient [12]. More details about TET patients are shown in Table 2. The development of TET was not PEGV dose-dependent. The ACROSTUDY™ and GPOS reported lower incidence rates of TET >3 ×ULN, 2.5 and 5.2 % respectively. The most recent ACROSTUDY™ paper, the Italian experience as it includes only patients registered in Italy (n = 341), reported an increase of TET >5 ×ULN in 0.9 % of the patients after 3 months of PEGV use and TET normalized after PEGV was withdrawn.

More frequent out-patient clinic visits and, thereby, more frequent assessments of ALT and AST might explain the observed differences in the incidence of TET, as the elevations in transaminases are transient and will pass unnoticed when follow-up intervals are wider. A previously described association between TET and Gilbert’s polymorphism (UGT1A1*28) in a Spanish study (n = 36) [58], was not confirmed by the Dutch cohort (n = 141) [12].