PCSK9 inhibitors in the management of hypercholesterolaemia after heart transplantation in the UK, a single centre observational study

Abstract

We looked to establish if Proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitor therapy can be safely initiated in heart transplant recipients and effectively reduce target low density lipoprotein (LDL). This prospective audit reviewed heart transplant recipients between 1st June 2019 and 1st November 2022 at Harefield Hospital in London. At baseline all patients must have attempted statin and ezetimibe therapy. All patients who remained with an LDL > 3.5 with very high cardiovascular risk or LDL > 4.0mmol/L with high risk were initiated on alirocumab injection every 2 weeks. Monitoring including biochemical analysis including immunotherapy levels, troponin, brain natriuretic peptides, electrocardiograph and echocardiogram. PCKS9i therapy was tolerated in 9/11 patients with 2 stopping treatment, one due to nausea & vomiting and one due to elevation in creatinine kinase. No adverse effects related to the heart transplant were detected and no significant change in creatinine kinase, liver function or left ventricular ejection fraction were seen. A significant reduction in LDL, total cholesterol, triglycerides was seen with LDL cholesterol reduction of 55% from 4.14 ± 0.47mmol/L at baseline to 1.86 ± 0.89mmol/L at 12 months. Estimated treatment effects constant over time expects to reduce LDL by about 2.19 to 2.77 mmol/L with 95% probability. Total cholesterol is likely to be lowered by 2.22 to 2.91 mmol/L and triglycerides by 0.42 to1.6 mmol/L with the same probability. This audit demonstrates limited safety and efficacy data of PCSK9i therapy in heart transplant recipients with elevated LDL cholesterol.

Introduction

Cardiac allograft vasculopathy has been described as the Achilles heel of heart transplantation, causing 1 in 8 deaths beyond a year after transplant¹. The mechanism behind the development of allograft vasculopathy is a complex interplay between atherosclerosis risk factors, infective and immune factors^2,3. While management of immunosuppression and antiviral suppression is key, dyslipidaemia remains one of the key determinants of allograft vasculopathy⁴.

The use of statins (HMG-CoA reductase inhibitors) remains a cornerstone of post heart transplantation therapy, associated with reduced morality, rejection, cancer as well as allograft vasculopathy⁵. In our institution our standard post transplantation protocol includes pravastatin 40 mg once a day. Ezetimibe 10 mg once a day if added if lipids remain outside of the target lipid range based on European Society of Cardiology risk profiles. The use and dose of statin post transplantation may be complicated through interactions with calcineurin inhibitors⁵ and alternatives are needed for those who are intolerant or unable to achieve a target lipid level.

Proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors are a well established novel injectable cholesterol lowering drug with safety and efficacy data that supports use with statins and in those who are statin-intolerant to achieve lower lipid levels⁶. Their safety and efficacy profile in heart transplant recipients is limited but growing, particularly with the recent publication of the EVOLVeD study showing a reduction in LDL without increased adverse events^7–13. In our institution we have established the use of PCSK9 inhibitors in heart transplant recipients with elevated low density lipoprotein (LDL) levels. We report the first UK data on use of PCSK9i in heart transplant recipients.

Methods

This prospective audit reviewed heart transplant recipients between 1 st June 2019 and 1 st November 2022 at Harefield Hospital in London. Patients received and were consented for standard of care at our institution and this was an audit of that activity. All heart transplant recipients are screened at routine 1 year follow up intervals to ensure lipid levels remained controlled. Patients whose LDL remained above 3.5mmol/L with very high cardiovascular (CV) risk or above 4.0mmol/L with high CV risk are referred to our PCSK9 inhibitor multi-disciplinary team (MDT). Cardiovascular risk was defined as per national guidelines for initiation of the PCSK9 inhibitor Alirocumab¹⁴. High risk includes any of following: acute coronary syndrome, coronary or arterial revascularisation procedures, coronary artery disease, ischaemic stroke or peripheral arterial disease. Very high risk is defined as recurrent CV events or CV events in more than 1 vascular bed. We also included heart transplantation indication of ischaemic heart disease as meeting very high CV independently.

This PCSK9 inhibitor MDT consists of medical and nursing heart transplant specialists, lipid specialists and pharmacists. Each patient’s history was reviewed to ensure that adequate titration of multiple statins and ezetimibe therapy had been attempted to maximum tolerated dose. The rejection and immunosuppression history was reviewed to ensure there was no current rejection and low risk of future rejection in context of the donor heart history. If the MDT agreed the patient could be offered PCSK9 inhibitor therapy they were then referred to the lipid specialist clinic for review and consent.

At baseline all patients are tested for routine biochemical analysis including full blood count, renal function, liver function, creatinine kinase, fasting lipids, troponin, brain natriuretic peptide as well as electrocardiograph (ECG) and an echocardiogram (Echo). All patients in this audit received Alirocumab. In our initial 3 patients the dose was commenced at 75 mg of which two were then increased to 150 mg every 2 weeks to achieve a > 60% reduction in LDL. After a change in hospital policy all remaining patients were commenced on 150 mg every 2 weeks from the onset.

The standard monitoring protocol consisted of a fasting lipid profile every 3 months for 1 year, immunotherapy levels weekly for 1 month, 1 monthly for 3 months and then every 3 months. Full biochemical analysis, ECG and Echo listed above was performed at 6 months and 12 months. All patients included in this audit were followed up for a minimum of 1 year. Our institution standard is to also report signal averaged ECGs for monitoring of rejection¹⁵ and Simpsons biplane left ventricular ejection fraction assessment on echo.

Alirocumab therapy was stopped if there were any adverse severe reactions noted including allergic reactions or side effects and any deterioration in donor heart function or fluctuant immunotherapy levels.

Statistical analysis was performed using software Stata. The comparisons are based on up to 9 degrees of freedom or fewer in some cases owing to missing observations at some comparison points. The impact of this shortage on the precision of estimates can be alleviated by pooling all observations at baseline and later. In this case, the model relating the outcome tofv

where is the outcome variable for person before treatment and is the outcome for the same person after treatment at the post-treatment periods of at the four quarterly recordings of the outcomes. In the case of half-yearly observations, . The observed or unobserved time-invariant individual effects like gender, age, comorbidities, body mass index, transplant year, past lifestyle etc. are represented by. The treatment effect can vary with the quarterly/half-yearly intervals of comparison. Random individual variations are represented by the values.

Differencing the outcomes pre and post treatment for each person removes the confounding individual effects to give the differences as

where the change in outcome . The random individual effects are assumed to have the Normal distribution.

For the cholesterol outcomes (LDL, total cholesterol, and triglycerides), the treatment is found to be effective if and more so over time if. In the case of the outcomes creatinine kinase (CK) and alanine aminotransferase (ALT), a significantly raised level () would be an unwelcome side effect but an improvement in the case of left ventricular ejection fraction (LVEF). The time interval of comparison makes no difference to the effects if does not differ from 0 significantly.

Results

In total 11 patients were included in this audit with a 1 year follow up period. Incomplete data consisted of missing cholesterol results at 12 months for one patient and missing ECG and Echo for another patient at 12 months. Of the 11 patients included the cause for heart failure was dilated cardiomyopathy in 6, ischaemic cardiomyopathy in 4 and postpartum cardiomyopathy in 1. The duration between heart transplantation and initiation of PCSK9i therapy was an average of 16.8 ± 10.0 years. In this cohort 9 patients had established coronary allograft vasculopathy with 3 having had percutaneous coronary intervention at baseline.

One patient discontinued PCSK9i therapy at 2 months due to side effects of nausea and vomiting. In another patient the PCKS9i therapy was stopped due to rising CK at 1 month. This patient had underlying chronic myopathy with starting baseline CK of 1180U/L that rose to 2082U/L leading to cessation of the intervention. In both patients the side effects resolved on stopping the medication.

The means of the cholesterol markers LDL, total cholesterol and triglycerides before and after treatment at quarterly intervals are shown in Table 1. There is a decline in these outcomes after treatment but no clear trend in the reductions over time. The fall in LDL appears sharp (Fig. 1) since all four 95% confidence intervals are entirely below that at baseline. Similarly, a strong fall in total cholesterol is indicated by three of the four 95% confidence intervals being fully below that at baseline. The reduction in triglycerides appears weaker.

Table 1.

Cholesterol markers (in mmol/L) before and after Treatment.

	LDL(mmol/l)		Total Cholesterol (mmol/l)		Triglycerides (mmol/l)
	Mean	95% CI	Mean	95% CI	Mean	95% CI
Baseline	4.14	3.67, 4.62	6.49	5.95, 7.03	2.44	1.50, 3.38
At 3 months	2.01	1.29, 2.74	4.15	3.17, 5.12	1.80	0.97, 2.63
At 6 months	1.63	0.73, 2.54	3.93	3.06, 4.80	1.90	0.81, 2.99
At 9 months	1.81	0.20, 3.41	4.15	2.18, 6.12	1.65	0.46, 2.84
At 12 months	1.86	0.97, 2.74	4.11	3.11, 5.11	2.03	1.16, 2.91

LDL low density lipoprotein.

Fig. 1.

LDL Cholesterol over time after PCSK9i administration.

Listed in Table 2 are the means of the possible side effect markers CK, ALT and the cardiac function marker LVEF before and after treatment. LVEF does not appear to alter after treatment; there is essentially no change in the mean or the confidence interval. CK appears to decrease post treatment but similarly at all change intervals. ALT appears to fall somewhat in the same manner.

Table 2.

CK, ALT and LVEF before and after Treatment.

	CK (U/l)		ALT (U/l)		LVEF (%)
	Mean	95% CI	Mean	95% CI	Mean	95% CI
Baseline	221.0	−27.6, 469.6	25.30	9.88, 40.72	61.63	57.11, 66.15
At 3 months	102.8	67.6, 138.0	15.83	11.36, 20.31
At 6 months	126.5	65.2, 187.7	15.62	12.91, 18.34	60.56	58.41, 62.70
At 9 months	104.2	18.5, 189.9	21.50	14.39, 28.61
At 12 months	106.2	64.9, 147.5	16.22	13.83, 18.62	61.02	57.20, 64.84

CK Creatinine Kinase, ALT alanine aminotransferase, LVEF left ventricular ejection fraction.

Data for 11 patients observed over 4 quarters (allowing for missing observations) are used to estimate the coefficients best fitting the data. The results (including those for CK, ALT and LVEF) appear in Table 3. The results indicate significant reduction in the LDL and total cholesterol levels where the observation quarter appears not to make any difference. However, the reduction in triglycerides appears to kick in only over time. CK, ALT and LVEF do not change significantly at the 5% level.

Table 3.

Estimated treatment effects allowing change over Time.

Outcome		p-value		p-value
LDL (mmol/l)	−2.596**	< 0.001	0.073	0.550
Total Cholesterol (mmol/l)	−2.664**	< 0.001	0.063	0.665
Triglycerides (mmol/l)	0.240	0.532	−0.775**	< 0.001
CK (U/l)	−13.199	0.631	5.463	0.694
ALT (U/l)	−2.596	0.160	0.602	0.514
LVEF (%)	−1.144	0.577	1.133	0.696

** indicates significant at the 5% level LDL low density lipoprotein, CK creatinine kinase, ALT alanine aminotransferase, LVEF left ventricular ejection fraction.

Since the time variation of the treatment effect is mostly not significant, a more precise set of estimates may be obtained by setting the effect variation over time to 0. Table 4 presents the estimated treatment effects with their p-values and the 95% confidence intervals for each outcome in the absence of effect variation over time. LDL, total cholesterol, and triglycerides decline with 95% confidence. The significant fall in triglycerides over time (seen in Table 3) is now reflected in the significant reduction in triglycerides when averaged over all time periods. As in Table 3, the treatment does not appear to change CK, ALT and LVEF at the 5% level of significance.

Table 4.

Estimated treatment effects constant over Time.

Outcome		p-value	95% confidence interval
LDL (mmol/l)	−2.478**	< 0.001	−2.767, −2.190
Total Cholesterol (mmol/l)	−2.561**	< 0.001	−2.906, −2.217
Triglycerides (mmol/l)	−1.011**	0.002	−1.599, −0.424
CK (U/l)	−4.345	0.778	−35.608, 26.918
ALT (U/l)	−1.621	0.122	−3.703, 0.461
LVEF (%)	−0.578	0.682	−3.503, 2.348

** indicates significant at the 5% level. LDL low density lipoprotein, CK creatinine kinase, ALT alanine aminotransferase, LVEF left ventricular ejection fraction.

According to the confidence interval estimates, treatment is expected to reduce LDL by about 2.19 to 2.77 mmol/L with 95% probability. Total cholesterol is likely to be lowered by 2.22 to 2.91 mmol/L and triglycerides by 0.42 to1.6 mmol/L with the same probability. Absence of a change in CK, ALT and LVEF are not ruled out at the 95% confidence level. These results appear robust allowing for the random errors in the outcomes to have non-constant variances.

Discussion

This work demonstrates in a limited capacity the safety and efficacy of PCSK9i therapy in heart transplant recipients with elevated LDL cholesterol.

Over the 1 year follow up period we identified no adverse events related to the transplant including rejection or deterioration in heart function. Two patients discontinued therapy due to side effects from PCSK9i therapy, one due to nausea and vomiting and one from elevated CK in context of chronic myopathy.

The data demonstrated a significant reduction in LDL, total cholesterol and triglycerides. In particular, LDL cholesterol reduced from 4.14 ± 0.47mmol/L at baseline to 1.86 ± 0.89mmol/L at 12 months (p < 0.001). There was no significant change in CK, ALT or LVEF in the cohort. This was assessed with pooling of baseline and later results due to low number of participants and data collection points for statistical analysis. PCKS9i therapy could potentially be a tool in patients following heart transplantation who continue to have elevated cholesterol despite optimal conventional therapy or intolerance of statins. However, similar to the wider population of high risk patients¹⁶, LDL levels still exceeded the ideal target range of < 1.4mmol/L and raise the issue of whether additional lipid lowering therapies, such as more potent statins, may still be necessary.

The optimal LDL target for heart transplant recipients is not well established. As an institution we adopt the European Society of Cardiology guidelines for managed of dyslipidaemia¹⁷ with our patient cohort included in this audit defined within the very high risk category and a target LDL of < 1.4mmol/L. It is worth noting that the post alirocumab LDL level of 1.86 ± 0.89mmol/L achieved in this audit still exceeds the ideal target of 1.4mmol/L highlighting the need for development of further efficacious lipid lowering therapies applicable to the transplant population. All patients were continued on statin therapy, typically pravastatin 40 mg once a day, regardless of LDL levels on alirocumab as they have been other determined benefits to statin use in transplantation such as reduced rejection and cancer risk⁵.

Though not included on the national guidance for initiation of Alirocumab therapy we include heart transplantation for ischaemic heart disease as its own qualifier for very high risk as the patient has already demonstrated a propensity for the atherosclerotic process¹⁸.

We routinely perform computed tomography coronary assessment every 2 years for our heart transplant recipients. We did not include this data as not all patients received coronary assessment at the correct intervals and in this audit could not account for the confounding effects of immunosuppression alteration and viral infections.

This audit has significant limitations in view of small sample size, low number of data collection points, small follow-up period and lack of hard outcome data. Thus, the statistical analysis had to be altered to pool analysis of outcomes to provide a means to assess potential effect of the intervention.

Conclusion

We hope this data will contribute to the scientific evidence base to provide a much needed therapeutic option for coronary graft vasculopathy that plagues patients post heart transplantation. We believe the next step would be to perform a larger randomised controlled trial to assess actual treatment effect and hard outcomes in this cohort.

Abbreviations

ALT

alanine aminotransferase

CV

cardiovascular

CK

creatinine kinase

ECG

electrocardiograph

Echo

echocardiogram

LDL

low density lipoprotein

LVEF

left ventricular ejection fraction

MDT

Multi-disciplinary team

PCSK9

Proprotein convertase subtilisin/kexin type 9

Author contributions

WA TK wrote the manuscript, OD TK EN CB conceptualisation and design, MT HL data interpretation, all authors reviewed the manuscript.

Data availability

The datasets generated and/or analysed during the current audit are not publicly available to avoid identification with small numbers involved but are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Disclosures

None relevant.

Ethical

Not needed. The audit was prospectively registered with out trust audit and quality improvement system.