A total of 204 subjects were screened and 103 were randomized, 52 to the alirocumab group and 51 to the ezetimibe group. All 103 randomized subjects received study treatment (the safety population), but one patient from each treatment arm withdrew from treatment before any on treatment postrandomization LDL-C measurements were made (but continued in the study and had subsequent LDL-C levels obtained); the remaining 101 subjects constituted the mITT or 'on-treatment' population. Overall, 15 patients did not complete the study treatment period. The percentage of patients who discontinued the IMP during the study was similar in the alirocumab group (eight patients; 15.4%) and the ezetimibe group (seven patients; 13.7%). The percentage of patients who discontinued the IMP due to AEs was also comparable between the two treatment groups (five patients [9.6%] in the alirocumab group and four patients [7.8%] in the ezetimibe group; Table 1).
Demographics and baseline lipid values were well balanced (Table 2). The protocol called for uptitration of the alirocumab dose at week 12 if LDL-C was ≥100 mg/dl at week 8, which was prompted by a blinded interactive voice responsive system. Instead, the interactive voice responsive system was incorrectly programmed to uptitrate subjects if LDL-C was ≥70 mg/dl. This discrepancy was not known until after study database lock. A total of 14 patients were uptitrated, and of these, only one patient had an LDL-C ≥100 mg/dl at week 8.
The LDL-C was reduced in both the ezetimibe and alirocumab groups by week 4; these reductions in LDL-C between the alirocumab and ezetimibe groups were statistically significant at weeks 12 and 24. Despite uptitration of 14 alirocumab patients at week 12 to 150 mg sc. Q2W, the LDL-C values observed at both weeks 12 and 24 were similar (Figure 2). The primary end point – the percentage difference in least square (LS) mean LDL-C between the alirocumab group and the ezetimibe group at 24 weeks was -31.6% for the ITT population and -36.9% for the mITT population (p < 0.0001 for both ITT and mITT populations). LDL-C lowering in the ITT population with alirocumab was 47.2% and with ezetimibe was 15.6%. The LDL-C lowering with alirocumab was approximately 50%, as predicted for the 75-mg dose based on modeling of Phase II data, and the LDL-C lowering with ezetimibe was 17–20%, as reported in previous ezetimibe studies. The changes in lipid values in the mITT or 'on-treatment' group were better than the ITT group because mITT only included those patients and visits where the study drug was taken and laboratory work was performed within the appropriate time window for the visits. There were also statistically significant differences between alirocumab and ezetimibe changes for ApoB, non-HDL-C and total cholesterol (Figure 3). A hierarchical approach was used for analyzing the data, which resulted in statistical testing being stopped when Lp(a) did not prove statistically significant. Lp(a) was lowered 18% by alirocumab, consistent with what would be expected based on Phase II data, but was lowered 10% with ezetimibe, a result that was not expected and had not been previously reported. The cause of this decrease is unknown.
Least square mean LDL-cholesterol values for weeks 0–24.
ITT: Intent to treat; LS: Least square; Q2W: Every 2 weeks; SE: Standard error.
Percentage change from baseline in secondary lipid parameters (intent-to-treat analysis).
*p < 0.0001.
ApoB: Apolipoprotein B; HDL-C: HDL-cholesterol; ITT: Intent to treat; Lp(a): Lipoprotein (a); LS: Least square; Non-HDL-C: Non-high- density lipoprotein cholesterol; SE: Standard error; TC: Total cholesterol.
Uptitrated Versus Non-uptitrated Groups
Fourteen patients were uptitrated from 75 to 150 mg of alirocumab sc. Q2W at week 12 based on week 8 LDL-C being ≥70 mg/dL and continued this dose until week 22 (last injection); however, only one of these patients should have been uptitrated by the protocol. Thirty-two patients were not uptitrated and received alirocumab 75 mg sc. Q2W from baseline to week 22 (last injection). Only the 46 patients who achieved a week-12 visit were included in the following analysis, including the 13 inappropriately uptitrated patients. Alirocumab blood levels were equivalent at week 12 prior to study drug injection but, at week 24, the uptitrated group showed approximately double the alirocumab blood level of the non-uptitrated group. The uptitrated group had a baseline LS mean LDL-C level of 153 mg/dl (3.95 mmol/l) versus 135 mg/dl (3.49 mmol/l) for the non-uptitrated group. LDL-C levels from baseline to week 24 can be seen in Figure 4. To determine whether the nonprotocol uptitration of the 13 patients with LDL-C levels between 70 and 100 mg/dl affected the results, a recalculation of the primary end point excluding these 13 patients' data after week 12 was performed and did not yield a significantly different result (ezetimibe: -15.6% change in LDL-C; alirocumab: -44.3% change in LDL-C). In addition, the week-12 data for LDL-C before uptitration occurred showed a 19.6% decrease for ezetimibe and a 48.1% decrease for alirocumab; neither value is statistically different from the week-24 data with or without the 13 patients being inappropriately uptitrated.
Percentage change in LDL-cholesterol over time according to uptitration status in the alirocumab group intent-to-treat population, with dose uptitrated at week 12.
LDL-C: LDL-cholesterol; W: Week.
TEAEs were reported in 36 of 52 patients (69.2%) in the alirocumab group compared with 40 of 51 patients (78.4%) in the ezetimibe group (Table 3). The differences in the most frequently reported TEAEs in the alirocumab group versus the ezetimibe group were not considered clinically significant. Muscle disorders (myalgia and muscle spasms) occurred in a similar frequency within both treatment groups (3.8% in the alirocumab group and 3.9% in the ezetimibe group). Two treatment-emergent serious AEs were reported in two patients: pulmonary embolism in one patient of the alirocumab group and bone erosion in one patient of the ezetimibe group. Neither of the serious AEs was considered to be related to the IMPs by the investigator. Nine patients prematurely discontinued treatment due to a TEAE: five patients (9.6%) in the alirocumab group and four patients (7.8%) in the ezetimibe group. Six patients (11.5%) in the alirocumab group and five patients (9.8%) in the ezetimibe group had at least one allergic TEAE. One patient in the alirocumab group discontinued treatment permanently due to a potential allergic TEAE (flushing and other concomitant symptoms of fatigue, nausea and headache). One patient (1.9%) in the alirocumab group and two patients (3.9%) in the ezetimibe group had at least one local injection site reaction TEAE during the 24-week study. These reactions were mild in severity. Laboratory, ECG and vital sign abnormalities were unremarkable between the two groups. Elevated glucose (≥126 mg/dl [7 mmol/l]) was present in six of the alirocumab patients (Table 4) and one ezetimibe patients at some visit during the study. All seven patients had elevated glucose (≥100 mg/dl [5.5 mmol/l]) or HbA1c levels (≥6.5%) at screening or baseline before receiving IMP and three of the six alirocumab patients had the diagnosis of diabetes mellitus at screening. There was no pattern of change in glucose or HbA1c from screening to week 24.
Three patients in the alirocumab group treated with 75 mg Q2W had an LDL-C value <25 mg/dl (<0.65 mmol/l) and one had two consecutive LDL-C values <25 mg/dl (<0.65 mmol/l). No particular safety concern was observed in this patient. No patient had an LDL-C value <15 mg/dl (<0.39 mmol/l) at any time in the study. A total of four patients had a positive antidrug antibody (ADA) status at least once during the 24-week study period (Table 5). All had low ADA titers, which continued to decline to very low levels or become ADA negative from week 12 to the week-32 follow-up visit. There were no differences between the uptitrated and non-uptitrated groups. No safety concerns were observed in these patients; there were no neutralizing antibodies detected and the presence of ADAs did not alter the LDL-C-lowering efficacy of alirocumab.
Future Cardiol. 2015;11(1):27-37. © 2015 Future Medicine Ltd.