Background
Homozygous familial hypercholesterolaemia (HoFH) is a rare genetic condition characterised by markedly elevated low-density lipoprotein-cholesterol (LDL-C) levels, inadequate response to conventional drug therapy and premature-onset cardiovascular disease (CVD) [
1]. HoFH is most commonly caused by the occurrence of two LDL receptor
(LDLR) gene mutations, but can also be caused by mutations in other genes that directly or indirectly act on the LDL/LDL-receptor pathway, including the genes encoding apolipoprotein B
(apoB;
APOB), proprotein convertase subtilisin/kexin type 9
(PCSK9), and LDL-receptor adaptor protein 1
(LDLRAP1) [
1]. If untreated, patients with HoFH often die from premature CVD in adolescence or early adulthood [
2].
The aim of therapy in HoFH is to aggressively reduce LDL-C levels to prevent or delay the onset of premature CV events. The European Atherosclerosis Society (EAS) has suggested that patients with HoFH should be treated to the same LDL-C targets set for other patients requiring lipid-lowering therapy, namely < 100 mg/dL (2.6 mmol/L) for primary prevention, and < 70 mg/dL (1.8 mmol/L) in the presence of clinical atherosclerotic CVD [
1]. These targets are not achievable in the majority of HoFH patients with currently available standard therapies that require functioning LDL-receptors, such as statins, ezetimibe [
3] and inhibitors of PCSK9 [
4].
Lipid apheresis does not require functional
LDLR and can reduce LDL-C further but is invasive and not universally available. Even with the use of apheresis, targets are often not achieved [
5].
Lomitapide is an oral microsomal triglyceride transfer protein inhibitor that reduces the assembly of apoB-containing lipoproteins in the intestine and liver, and therefore does not require functioning LDL-receptors to lower LDL-C levels [
6]. Lomitapide was approved as adjunctive therapy in adult patients with HoFH following a small open-label single-arm Phase 3 study. In this study, lomitapide resulted in mean LDL-C reductions of 50.7% (Week 26; safety population; mixed-model repeated measures analysis [
7]).
We hypothesise that such large decreases in LDL-C, and in particular the attainment of the EAS recommended LDL-C target levels, may result in a lowered risk of CV events. A paper was recently published modelling the potential improvement in survival and the delay to first major adverse CV event (MACE) in treating patients with HoFH [
8]. Formal CV outcome studies to test this hypothesised benefit are not feasible in HoFH due to the relatively small number of affected patients, and ethical concerns about withholding potentially effective therapy from patients with a very severe disorder for prolonged periods of time in a double blind randomised placebo-controlled trial. Therefore, this analysis was conducted to examine the actual benefit achieved in treating patients with HoFH. The objectives of this post hoc, retrospective analysis are to determine the number of HoFH patients receiving lomitapide who reach EAS targets within 6 months, and to compare cardiovascular event rates in patients receiving lomitapide with published data.
Discussion
Lomitapide, in combination with other lipid-lowering therapies is an effective treatment for the reduction of LDL-C levels in adult patients with HoFH, allowing many of these patients to reach EAS recommended LDL-C target levels for the first time.
Most of the patients who achieved LDL-C targets did so in the first 6 months of treatment. Assessment of continuous maintenance of these LDL-C target levels is difficult in a setting where background lipid-lowering therapies, including apheresis, could be titrated upon achievement of target levels. Similarly, not all patients who achieve target during lomitapide dose titration are likely to continue on the maximum dose reached in the long-term. In our analysis, patients that reached the 100 mg/dL goal at least once, remained at goal for a median of 6 study visits suggesting that many of the patients receiving lomitapide are able to maintain the target long-term. Nevertheless, the data provided here are a reflection of the potential to reach target, rather than an indication of maintenance of LDL-C at target.
The annualized MACE rates in HoFH patients treated with mipomersen, lomitapide and evolocumab were 11.4, 2.0 and 2.1%, respectively. The pre-mipomersen exposure annualized event rate in the mipomersen trial was 26.1%.
We chose the dataset from the publication of Duell, et al. [
9] as a comparator as it reports relatively recent outcome data for HoFH patients receiving standard of care therapy before newer agents such as lomitapide, mipomersen and evolocumab became available. We were unable to locate other recent data that would have allowed us to calculate annualized event rates for HoFH patients receiving conventional lipid-lowering therapy.
One of the major limitations of our MACE rate comparison is that the lomitapide studies were open-label and did not include a control group receiving placebo. Similarly, neither the mipomersen nor the evolocumab studies include placebo arms of sufficient duration to assess CV event rates. The background event rate for patients receiving conventional lipid lowering therapy used for this comparison is therefore derived from the pre-treatment mipomersen cohort that differs in important aspects, such as apheresis usage, baseline LDL-C levels, or CVD rates at baseline from the population enrolled in the lomitapide and evolocumab studies. Indeed, the patients in the Duell study had baseline LDL-C values of 455 mg/dL, which is significantly higher than the 336 mg/dL and 324 mg/dL in the lomitapide and evolocumab studies. This may be due to underlying disease severity or due to less intensive background therapy, specifically apheresis (63 and 32% of patients in the lomitapide and evolocumab trials received apheresis at baseline, compared with none in the mipomersen studies). However, baseline LDL-C levels are likely a strong predictor of achievable LDL-C reductions and CV risk and are therefore a significant confounding factor in our analysis. Baseline CVD prevalence is another important potential confounder. In the lomitapide studies baseline CVD prevalence was 93%, it was 51% in the evolocumab study and not reported for mipomersen.
Further limitations of this post-hoc, retrospective analysis include small sample sizes, relatively short treatment durations, and the possible biases inherent in conducting retrospective evaluations of this type. Importantly, the CV event rate reported by Duell et al. [
9] could be skewed by subjects experiencing recurrent events (7 patients account for 12 events) or preferential recruitment of subjects with above average cardiovascular risk. Our study also relied on AE reporting for capture of MACE and events were not formally adjudicated. It is therefore possible that we may not have captured all events, or that some events, such as unstable angina, may have been misclassified. Furthermore, not all patients enrolled in the pivotal trial entered the lomitapide extension study, thereby introducing potential bias in the extension trial population towards a healthier ‘survivor’ population. The mipomersen treatment duration was also significantly shorter than the lomitapide exposure, and it is conceivable that longer treatment with mipomersen may have been associated with a progressive lowering of MACE rates. Thus, our findings cannot be regarded as a definitive estimation of MACE reduction with lomitapide, and can also not be used to draw any definitive conclusions about the long-term outcome associated with any one of the novel lipid-lowering therapies. It would have been preferable to combine data from multiple HoFH cohorts to obtain averaged CV event rates in patients receiving conventional lipid-lowering therapy, but we were not able to locate any data that would have allowed us to calculate annualized CV event rates.
Despite major limitations our data, along with the data for mipomersen and evolocumab, suggest that CV risk in HoFH can be reduced by pharmacotherapies that reduce LDL-C. The cohort of patients with the highest LDL-C (pre-mipomersen; LDL-C 455 mg/dL) had the highest event rates. Reducing the LDL-C in this cohort to levels comparable to those found at baseline in the lomitapide and evolocumab studies was associated with an approximate 50% reduction in MACE. The event rate, however, remained higher than that seen with lomitapide and evolocumab, likely because of prior and ongoing increased exposure of the vasculature to LDL. Baseline LDL-C levels were comparable for lomitapide and evolocumab (336 mg/dL and 324 mg/dl) and on-treatment MACE rates were remarkably similar. Although lomitapide lowered LDL-C more than evolocumab, differences in the prevalence of CVD at baseline (93% for lomitapide versus 51% for evolocumab) may have influenced CV event rates. The prognostic importance of LDL-C is reinforced by data from Raal et al., who showed that lowering LDL-C from 615 mg/dl to 452 mg/dL with conventional lipid-lowering therapy was also associated with delayed cardiovascular events and prolonged survival in patients with HoFH [
10].