Long-term safety and efficacy of subcutaneous pasireotide in patients with Cushing’s disease: interim results from a long-term real-world evidence study

The interim results from this international real-world study support the favorable long-term benefit–risk profile of pasireotide in patients with CD, which has been previously demonstrated in a Phase III clinical trial [10, 14]. The primary objective of this ongoing study is to document the long-term safety and tolerability profile of pasireotide in patients with CD in a real-world setting. Notably, more than 25% of patients who participated in our study had received over 3 years of treatment with pasireotide when including treatment exposure prior to study entry (maximum treatment duration of 9.9 years).

Half (50%) of the patients experienced a drug-related AE and 14% of patients experienced a drug-related SAE, the primary endpoint of the study, from study start. Most (70%) drug-related AEs were mild to moderate in severity and were consistent with the known safety profile of pasireotide [9], the most common being hyperglycemia-related AEs, nausea and diarrhea. AEs and SAEs were less common in patients who were treated with pasireotide prior to study entry than in those who initiated pasireotide at first study visit, and fewer prior users of pasireotide discontinued because of AEs than new users (15% and 32%, respectively). The lower rate of drug-related AEs and SAEs in prior versus new users may be explained by a potential selection bias: patients who tolerated pasireotide were more likely to continue pasireotide treatment (prior-use group). Furthermore, the most common AEs generally occur shortly after initiation of therapy.

Gastrointestinal AEs are a known class effect of somatostatin analogues. During our real-world study, a higher proportion of patients in the new-use group experienced AEs of nausea (32% vs. 8%) and diarrhea (16% vs. 9%) compared with the prior-use group, potentially reflecting an improvement in gastrointestinal tolerance over time. Compared with the previous Phase III study of twice-daily pasireotide in previously pasireotide-naïve patients, fewer patients in the new-use group of our study had a recorded AE of nausea (32% vs. 52%) or diarrhea (16% vs. 58%) [9]. The proportion of patients who discontinued because of any AE was slightly higher in the new-use group in our real-world study compared with the 12-month Phase III study (32% vs. 16%). Differences in the frequency of AEs and safety-related discontinuation rates between the two studies may result from numerous factors, including differences in duration of treatment/observation, the higher frequency of scheduled visits and safety assessments in the clinical trial setting, and differences in patient populations.

Hyperglycemia-related AEs have been frequently observed in clinical studies of pasireotide (57–77% of patients) [9, 15‐17]. Increases in blood glucose levels predominantly occur soon after initiation of pasireotide, are more common in patients with pre-existing diabetes mellitus or impaired glucose tolerance, and can be effectively managed with appropriate antidiabetic medication soon after onset [9, 18]. Hyperglycemia associated with pasireotide results from the suppression of insulin secretion and decreased incretin response and is reversible upon treatment discontinuation [18‐20]. In our study, hyperglycemia-related AEs that were suspected to be related to study drug were reported in < 25% of participating patients, and in a greater proportion of new users than prior users of pasireotide (45% and 16%, respectively). In addition, few patients experienced grade 3 hyperglycemia-related AEs, which were almost exclusively limited to new users (13% vs. 1%; no patients experienced a grade 4 event). Each patient who experienced a grade 3 hyperglycemia-related AE during the study had a prior history of diabetes. Results from this study indicate that pasireotide-associated hyperglycemia can be effectively managed with appropriate antidiabetic treatment in the clinical practice setting, as shown by the low rate (4%) of patients in this real-world evidence study discontinuing for this reason. The lower occurrence of hyperglycemic events in prior-use patients may have resulted from the appropriate management of pasireotide-associated hyperglycemia with antidiabetic agents prior to enrollment, as well as the improvement in hypercortisolism. In addition, patients with uncontrolled hyperglycemia on pasireotide may have been less likely to continue treatment as part of the study. It is important that blood glucose levels are monitored soon after the start of pasireotide treatment so that appropriate intervention can be taken promptly if these levels rise [21]. As decisions on the management of hyperglycemia were made by the local investigator, it was not possible to assess the effect of specific therapeutic approaches on blood glucose levels in this real-world study. A Phase IV study is currently ongoing to evaluate the effects of incretin-based therapy compared with insulin on glycemic control at 16 weeks in patients with CD or acromegaly (http://​ClinicalTrials.​gov NCT02060383).

In a previous Phase III study, pasireotide provided rapid and sustained reductions in mUFC and normalized levels in ~ 20% of enrolled patients with CD at months 6 and 12 [9]. In this ongoing real-world study, normal mUFC levels were achieved by 50% of new users who had an evaluable assessment within the first month of pasireotide treatment. Importantly, this high response rate was maintained for up to 9 months of treatment; however, few of these patients had evaluable assessments after this point. Future analyses from this study should allow us to explore the long-term response to pasireotide in new-use patients. Direct comparison of response rates for new-use patients in our study and the previous Phase III trial is not possible. Long-term control of mUFC levels was demonstrated by a significant number of patients who were receiving pasireotide prior to study start: 65% of these patients had mUFC ≤ ULN at baseline, compared with 82% and 67% of patients after a further 12 and 24 months of treatment, respectively. It is important to note that response rates are reported for patients with evaluable assessments in this study. As a result, response rates may be overstated as patients who discontinued treatment because of an unsatisfactory therapeutic effect (16% and 15% of patients in the new- and prior-use groups, respectively) were not included in the analyses. Although patients were permitted to receive concomitant medications for CD (most commonly cabergoline), the majority of patients (76% at month 12 and 80% at month 24) received pasireotide as monotherapy during the study. The higher response rates seen for the prior-use than the new-use group likely reflect the fact that patients with a favorable response to pasireotide would be more likely to continue receiving treatment in this study.

While resection of the ACTH-secreting pituitary adenoma by transsphenoidal surgery is first-line treatment for most patients with CD [5], multimodal treatment is frequently required in order to control hypercortisolism in patients with persistent or recurrent disease [1, 5]. In this real-world study, almost one-quarter of patients had received prior pituitary irradiation, which is generally considered third-line treatment in CD [22]. Interestingly, fewer patients had received prior pituitary irradiation in the new-use group (10%) than in the prior-use group (27%). As patients in the new-use group generally had a more recent diagnosis of CD than patients in the prior-use group (mean time since diagnosis: 43 vs. 83 months, respectively), it is possible that patients and physicians have been more willing in recent years to use medical therapy over radiotherapy for patients in whom surgery has failed, is not feasible or has been refused.

Our study has several limitations that we acknowledge. Owing to its observational nature, a therapy protocol and visit schedule are not imposed and, as such, specific efficacy assessments and study visits are not mandatory. So far, this has resulted in almost half of the patients who entered our study not having mUFC assessments at baseline and many having missing values at subsequent time points. In addition, some patients received concomitant medications for CD during this real-world study. Taken together, these factors may have complicated overall data interpretation. For the same reason, an accurate report of changes in clinical signs of hypercortisolism and health-related quality of life was not formally performed in the study. Nevertheless, a key strength of this study is the large number of patients receiving pasireotide therapy and monitored in real-world clinical practice.

In conclusion, the interim results from this observational study demonstrate that pasireotide provides rapid and sustained control of mUFC and offers acceptable tolerability in real-world clinical practice, with no new safety signals emerging over time. Notably, a lower incidence of hyperglycemia was observed in prior- compared with new-use patients, suggesting that pasireotide-associated hyperglycemia did not worsen over time if appropriately managed after onset. These findings support pasireotide as a long-term therapeutic option that can be utilized to provide sustained control of hypercortisolism for patients with CD.