Clinical characteristics and lipid lowering treatment of patients initiated on proprotein convertase subtilisin-kexin type 9 inhibitors: a nationwide cohort study.

OBJECTIVES: Given the novelty of proprotein convertase subtilisin-kexin type 9 inhibitors (PCSK9i), little is known regarding overall implementation or clinical characteristics among patients who initiate treatment. We aimed to assess the total number of patients initiated on PCSK9i along with a description of the clinical characteristics and lipid lowering treatment (LLT) of such patients.
SETTING: A register-based descriptive cohort study of patients receiving a PCSK9i in the time period from 01 January 2016 to 31 March 2017 using a cross linkage between three nationwide Danish registers. Information regarding PCSK9i prescriptions, patient demographics, concurrent pharmacotherapy, comorbidities and previous coronary procedures was identified.
RESULTS: Overall, 137 patients initiated treatment with PCSK9i in the study period from 11 in the first quarter of 2016 to 40 in the first quarter of 2017. The majority had a history of ischaemic heart disease (IHD) (67.9%) with ischaemic stroke and diabetes mellitus being present in 7.3% and 16.8% of patients, respectively. All patients initiated on PCSK9i had been previously prescribed statin treatment with atorvastatin and simvastatin being most frequently prescribed in 53% and 36% of patients, respectively. The majority of patients had received both statins and ezetimibe (94.9%) and approximately half of these patients had also received bile acid sequestrant (45.3%). Clinical characteristics mainly differed in patients receiving triple LLT compared with patients not receiving triple LLT in the regards of heart failure.
CONCLUSION: Patients treated with PCSK9i were rare, characterised by having IHD and had received various and intensive conventional LLT prior to PCSK9i initiation in agreement with current international guidelines. © Author(s) (or their employer(s)) 2019. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.

In the present study, we utilised the Danish national registries to perform a description of the characteristics of patients prescribed with PCSK9i on a national basis. We were able to include an unselected cohort of patients, with a great variety of comorbidities.

The Danish nationwide registries include data on all residents regardless of participation in the labour market, and therefore is not affected by selection bias caused by including selected age groups, hospitals or health insurance systems.

Patients with familial hypercholesterolaemia (FH), who are also eligible for PCSK9i treatment, collect their PCSK9i at a specialised hospital unit and due to the design of the present study, we were not able to identify patients with a diagnosis with FH.

Due to the observational registry design of the study, we were not able to retrieve data regarding biochemical analyses, clinical arguments for initiation of PCSK9i, LLT after initiation of PCSK9i or adverse events with LLT.

Due to a low diagnostic sensibility, we could not fully discriminate the indications for patients receiving PCSK9i.

Introduction

Atherosclerosis-related morbidity and mortality is reaching epidemic proportions worldwide and complications related to ischaemic heart disease (IHD) are the leading cause of death under 65 years of age.1 Low-density lipoprotein cholesterol (LDL-C) is a well-established and modifiable constituent in the pathogenesis of atherosclerosis and epidemiological studies have established LDL-C as one of the main risk factors for the development of atherosclerotic disease and IHD.1–3

Since the late 1980s, statins have been used to lower LDL-C levels and reduce the risk of cardiovascular events and related mortality across the spectrum of primary and secondary prevention.4–6

Randomised trials have since targeted even lower LDL-C levels achieved by intensified LDL-C lowering treatment (ie, higher dosage, more potent statins or additional ezetimibe) in patients with IHD, which has shown to further reduce the risk of cardiovascular disease (CVD).7–10

Novel treatment options to further lower the LDL-C levels are monoclonal antibodies to proprotein convertase subtilisin-kexin type 9 (PCSK9), which have shown lowering of the LDL-C levels by 40%–60% and relative reduction of cardiovascular events by 15% (CI: 0.79–0.92) when added to background statin treatment.11

The European Society of Cardiology (ESC) and The American College of Cardiology (ACC) guidelines12 13 have endorsed PCSK9i treatment in patients at high risk and very high risk of future cardiovascular events. Partly due to the high cost of PCSK9i, most countries have established national guidelines and committees to approve a prior authorisation application for medicine subsidy in the individual patient. In Denmark, PCSK9i treatment were approved for medication subsidy in patients in very high risk of future CVD (ie, patients with a history of acute coronary syndrome, acute myocardial infarction, atherosclerotic polyvascular disease or diabetes mellitus with IHD) and in patients in high risk of future CVD (ie, patients with a history of stable angina pectoris or diabetes mellitus with either target organ damage, peripheral atherosclerotic artery disease, transitory cerebral ischaemia or ischaemic cerebral infarction) who despite maximally tolerated lipid lowering treatment (LLT) required further reduction of LDL-C levels. Established cut-off LDL-C values were 3.0 mmol/L (115 mg/dL) and 3.5 mmol/L (135 mg/dL) in the very high-risk patients and high-risk patients, respectively (online supplementary appendix figure 1). Statin intolerance in these patients would also be approved for subsidy when treatment with at least three types of statin starting in low dosages titrated to maximum tolerated dosage and ezetimibe and bile acid sequestrant had been attempted prior to PCSK9i initiation. The ESC and ACC guidelines further endorse additional clinical criteria (ie, patients with a history of isolated peripheral atherosclerotic artery disease, isolated ischaemic cerebral infarction or diabetes mellitus with marked hypertension or hypercholesterolaemia) and at different LDL-C measurement cut-off values. In these patients, however, there is a lack of treat to target trials and the effects of intensified LDL-C lowering is not with thorough evidence.14 15 Given the novelty of PCSK9i and the difference in national and international guidelines, we do not know the exact prevalence, the clinical characteristics, concurrent medication or the attempts of double and triple LLT in patients initiated on PCSK9i.

In the present study, we wanted to assess the total number of patients initiated on PCSK9i in the first quarter (Q1) of 2016 to the Q1 of 2017 with a description of the clinical patient characteristics, comorbidities, medication use, concurrent LLT in these patients. Furthermore, we wanted to compare these parameters in patients attempted in triple LLT and patients in single or double LLT.

Methods

In this register-based cohort study, information regarding patient demographics, comorbidities, coronary procedures and concurrent medication use was identified using cross linkage between three different nationwide Danish registers. The Civil Registration System holds information on date of birth, sex and survival status. The Danish National Patient Register holds information on every hospital admission in Denmark since 1978, in which each hospitalisation is registered at discharge with one primary diagnosis and, if applicable, one or more secondary diagnoses according to the International Classification of Diseases, the 10th revision (ICD-10) since 1994. The Danish National Patient Register also holds information on surgeries and procedures including percutaneous coronary intervention (PCI) and Coronary Artery Bypass Grafting (CABG). Comprising data on date, quantity, strength, formulation of all prescriptions dispensed from Danish pharmacies has been accurately registered in The Danish Registry of Medicinal Product Statistics since 1995 and coded according to the Anatomical Therapeutic Chemical (ATC) classification system.

All Danish patients with a prescription for PCSK9i (ATC code C10A×13 or C10A×14) between 1 January 2016 and 31 March 2017 were included in the study cohort on the day they redeemed their prescription. Since 1 March 2017, PCSK9i were collected at specialised units at Danish hospitals and prescriptions of PCSK9i in the time period from 1 March 2017 to 31 March 2017 were not registered in the present study.

Demographic information on age, sex and vital status was identified using The Civil Registration System. Comorbidities and coronary procedures were identified using The Danish National Patient Register. Information on diabetes mellitus and hypertension was identified using The Danish Registry of Medicinal Product Statistics along with the concurrent medication (online supplementary appendix table 1).

Statistics

Categorical data is presented as counts with percentages. Statistical difference was tested using chi-squared test or Fisher’s exact test where appropriate. Continuous variables were reported as medians with IQR. Statistical difference was tested using the Wilcoxon rank-sum test. A p value <0.05 was considered significant.

Data management and statistical analyses were conducted using R statistics, V.3.4.3 (R Core Team [2017]. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL http://www.R-project.org/).

Ethics

In Denmark, retrospective register studies do not require approval from the ethics committees. The ethical legislation in Denmark does not permit the reporting of clinical patient characteristics in less than three patients. Data were made available in an anonymised format such that specific individuals could not be identified. The Danish Data Protection Agency approved this study (ref. no.: 2007-58-0015/GEH-2014-016 I-Suite no.: 02734)

Patient and public involvement

Patients and public were not involved in this research project.

Results

The total number of patients treated with PCSK9i during the total study period of 15 months was 137. The total number of patents initiating PCSK9i increased from 11 in Q1 of 2016 to 40 in Q1 of 2017 (figure 1). The overall median age was 62 (IQR: 55–68) years and 53.3% were men with no difference in number of men and women initiated (online supplementary appendix figure 2). The majority of patients had a history of IHD (68%) with 51.1% having received PCI and 24.8% having received CABG. All patients had filled a prescription of statin (100%) prior to initiation of PCSK9i with atorvastatin and simvastatin being most frequently prescribed in 57.7% and 32.1% of patients, respectively (table 1).

Figure 1

Bar plot illustrating the number of patients initiated with PCSK9 inhibitor treatment by calendar quarter. Q1 2016=01 January 2016–31 March 2016; Q2 2016=01 April 2016–30 June 2016; Q3 2016=01 July 2016–30 September 2016; Q4 2016=01 October 2016–31December 2016; Q1 2017=01 January 2017–31 March 2017.

Table 1

Clinical characteristics, demographics and medication use of patients initiated on PCSK9 inhibitors

Parameter	Study cohort
Number of patients, n	137
Age (median (IQR))	62 (55–68)
Men, n (%)	73 (53.3)
Ischaemic stroke, n (%)	10 (7.3)
Ischaemic heart disease, n (%)	93 (68)
Peripheral artery disease, n (%)	4 (3)
Chronic obstructive pulmonary disease, n (%)	5 (3.6)
Heart failure, n (%)	13 (9.5)
Chronic kidney disease, n (%)	11 (8.0)
Hypertension, n (%)	62 (45.3)
Diabetes, n (%)	23 (16.8)
Non-loop diuretics, n (%)	38 (27.7)
Loop diuretics, n (%)	22 (16.1)
Beta blockers, n (%)	61 (44.5)
Calcium channel blockers, n (%)	29 (21.2)
RAS-inhibitors, n (%)	50 (36.5)
Acetyl salicylic acid, n (%)	76 (55.5)
NSAID, n (%)	19 (13.9)
ADP-inhibitors, n (%)	38 (27.7)
Statin use, n (%)	137 (100.0)
Dosage of statins (%)
20 mg	13 (9.5)
40 mg	64 (46.7)
80 mg	60 (43.8)
Type of statin, n (%)
Atorvastatin	72 (52.6)
Fluvastatin	4 (2.9)
Pravastatin	5 (3.6)
Rosuvastatin	7 (5.1)
Simvastatin	49 (35.8)
Fibrates, n (%)	30 (21.9)
Bile acid sequestrant, n (%)	62 (45.3)
Nicotine acid, n (%)	39 (28.5)
Ezetimibe, n (%)	130 (94.9)
PCI, n (%)	70 (51.1)
CABG, n (%)	34 (24.8)

ADP-inhibitors, adenosine diphosphate inhibitors; CABG, coronary artery bypass graft; NSAID, non-steroidal anti-inflammatory drugs; PCI, percutaneous coronary intervention; PCSK9 inhibitors, proprotein convertase subtilisin-kexin type 9 inhibitors; RAS-inhibitors, renin-angiotensin system inhibitors.

The majority of patients had received statins in dosages of 80 mg (43.8%) or 40 mg (46.7%). Of the 137 included patients, 94.9% of patients had filled a prescription of ezetimibe in addition to statin and 43.1% of patients had filled a prescription of three types of LLT with bile acid sequestrants and ezetimibe in addition to statin. Few patients had filled a prescription of statin alone (2.9%) and 50% of all patients had filled a prescription of either fibrates or nicotinic acid (figure 2 and table 1).

Figure 2

Venn diagram illustrating combinations of prescribed lipid lowering treatment in 137 patients initiated on PCSK9 inhibitors.

Patients in dual LLT were more prone to have heartfailure than patients attempted in triple LLT prior to PCSK9i initiation. No other statistical difference regarding demographics, diseases, concurrent medication or coronary procedures was found between the patients receiving triple therapy compared with the patients not receiving triple LLT (table 2).

Table 2

Clinical characteristics, demographics and medication use by number of lipid lowering drugs prescribed (not triple therapy vs triple therapy) prior to initiation of PCSK9 inhibitors

Parameter	Not triple treatment	Triple treatment
Number of patients, n	78	59
Age, median (IQR)	61 (55–68)	62 (56–69)
Men, n (%)	42 (53.8)	31 (52.5)
Ischaemic stroke, n (%)	6 (7.7)	4 (6.8)
Ischaemic heart disease, n (%)	54 (69.2)	39 (66.1)
Chronic obstructive pulmonary disease, n (%)	4 (5.1)	<3 (<5%)
Heart failure, n (%)	10 (12.8)	3 (5.1)
Chronic kidney disease, n (%)	6 (7.7)	5 (8.5)
Hypertension, n (%)	34 (43.6)	28 (47.5)
Diabetes, n (%)	14 (17.9)	9 (15.3)
Non-loop diuretics, n (%)	20 (25.6)	18 (30.5)
Loop diuretics, n (%)	16 (20.5)	6 (10.2)
Beta blockers, n (%)	30 (38.5)	31 (52.5)
Calcium channel blockers, n (%)	17 (21.8)	15 (25.4)
RAS-inhibitors, n (%)	28 (35.9)	22 (37.3)
Acetyl salicylic acid, n (%)	48 (61.5)	28 (47.5)
NSAID, n (%)	12 (15.4)	7 (11.9)
ADP-inhibitors, n (%)	17 (21.8)	21 (35.6)
Dosage of statins (%)
20 mg	6 (7.7)	7 (11.9)
40 mg	40 (51.3)	24 (40.7)
80 mg	32 (41.0)	28 (47.5)
Type of statin, n (%)
Atorvastatin	46 (59.0)	33 (55.9)
Fluvastatin	<3 (<4%)	<3 (<5%)
Pravastatin	<3 (<4%)	4 (6.8)
Rosuvastatin	5 (6.4)	<3 (<5%)
Simvastatin	26 (33.3)	18 (30.5)
Fibrates, n (%)	14 (17.9)	16 (27.1)
Nicotine acid, n (%)	20 (25.6)	19 (32.2)
PCI, n (%)	38 (48.7)	32 (54.2)
CABG, n (%)	22 (28.2)	12 (20.3)

ADP, adenosine diphosphate; CABG, coronary artery bypass graft; NSAID, non-steroidal anti-inflammatory drugs; PCI, percutaneous coronary intervention; RAS, renin-angiotensin system.

Discussion

In this nationwide study, we described the prevalence and clinical and demographical characteristics of patients treated with PCSK9i over a 1-year period from 2016 to 2017. PCSK9i were prescribed in a total of 137 patients with an increase from 11 initiated patients in the Q1 of 2016 to 40 initiated patients in the Q1 of 2017. Patients were characterised by having a history of IHD and having received PCI and CABG. PCSK9i initiation was consistently clustered in patients attempted to reach maximum tolerated LLT resembled by various types and dosages of statins and patients having attempted dual LLT and half having attempted triple LLT. Patients in dual LLT were more prone to have heart failure than patients attempted in triple LLT prior to PCSK9i initiation.

We found three similar studies describing the clinical characteristics and LLT prior to PCSK9i initiation.16–18 In Rane et al and Mues et al, only 54% and 19% of patients had received two types of LLT, 24% and 26% of patients did not receive any LLT prior to initiation with PCSK9i, respectively and only 17% of patients had previously received three types of LLT. Also, in Fairman et al and Rane et al, 60% and 69% of patients, respectively, were in treatment with statin as monotherapy. The results from these studies are in discrepancy with the present study where 100% of patients had received statin treatment, 97% had received minimum two types of LLT and only 3% were in statin treatment alone. Physician reported intolerability to statin in 32% of the patients initiated on PCSK9i greatly influenced PCSK9i initiation in Rane et al. However, ezetimibe was attempted surprisingly infrequent (7.6% of patients) in contrast to the present study where ezetimibe was prescribed in 94.9% of patients. Differences of eligibility of PCSK9i in the ESC guidelines compared with the ACC guidelines could possibly also have influenced treatment regimens with LLT prior to PCSK9i initiation.12 19–21 We observed a more rigorous adherence in attempting statin treatment and additional conventional non-statin treatment prior to initiation of PCSK9i in the present study compared with similar studies and studies of patients at high risk of CVD not receiving PCSK9i.22–24 The infrequent use of rosuvastatin was primarily due to a protecting marketing patent, making rosuvastatin more expensive than atorvastatin making atorvastatin the first choice of potent statin in Denmark. Simvastatin is primarily used in the process of overcoming side effects of potent statin treatment in patients with IHD in order to reach maximum tolerated dose of LLT or as primary prophylaxis treatment of CVD. In our cohort, 36% of patients received simvastatin, which most likely indicates attempts of titrating to maximum tolerated dosage of statin when considering the strict national criteria for medicine subsidy regarding PCSK9i treatment and most patients had IHD. This could indicate persistent attempts of reaching LDL-C target goals in adherence to national clinical guidelines, and a last resort usage of PCSK9i.

We found that patients in dual LLT were more prone to have heart failure than patients attempted in triple LLT prior to PCSK9i initiation. The treating physician could possibly be susceptible to more intensive lowering of LDL-C to reduce the risk of future cardiovascular events in the more frail patients such as patients with verified heart failure. Although patients were initiated according to both national and international guidelines, perhaps certain other clinical traits might influence both physicians and committee’s approving of medicine subsidy for PCSK9i.25 Patients' demographics and concurrent medication use did notsignificantly differ in patients having received triple LLT and patients not having received triple LLT.

Despite having been endorsed by guidelines for patients with many different kinds of atherosclerotic CVD, 66% of the patients initiated on PCSK9i had a history of IHD (online supplementary appendix figure 2). The large randomised trials investigating the effects of intensified LLT7 9 10 26–29 or PCSK9i treatment11 30–37 on reducing future cardiovascular events have predominately included patients with established IHD or a recent IHD related event. The Danish national guidelines have also adapted more accommodative criteria for PCKS9i initiation in patients with a history of IHD, which explains the clustering of PCSK9i in such patients. Repeating our research in another country might show different results regarding the clinical patient characteristics, which probably will become more streamlined along with the publication of long-term randomised trials evaluating PCSK9i treatment.

Even though we observed an increasing number of patients initiated on PCSK9i, only 137 patients in total were prescribed with PCSK9i over the study period. Statins are the first-line treatment, more frequently used and have established effectiveness on improving long-term cardiovascular outcomes in both primary and secondary prevention trials without being extensively expensive compared with PCSK9i.

PCSK9i are without established long-term effectiveness on major cardiovascular events and clinical trials are still ongoing. Based on the current evidence of the effectiveness of PCSK9i on reducing major cardiovascular events, the Institute for Clinical and Economic Review (ICER) reported a number needed to treat for 5 years to avoid one major cardiovascular event (NNT5) of 28 individuals.38 While randomised clinical trials of PCSK9i have shown beneficial efficacy, they are very expensive and this may have had major influence on theoverall clinical implementation. The cost ofPCSK9i is around $14 350 per year in 2017, generating a cost-effectiveness ratio which may exceed a threshold of price pr. quality-affected life years. These economic considerations together with a possible postponed apprising of PCSK9i as a treatment option may explain why PCSK9i have been used rather infrequent when also taking into account how many patients who actually need further LLT.39 40 Economic considerations can also help explain the modified and individually formed national guidelines for PCSK9i initiation. Committing fully to the criteria for PCSK9i initiation stated in the ESC or ACC guidelines would probably not be realistic for any countries’ medical expenditures, resulting in many patients being rejected for medicine subsidy and patients with certain clinical characteristics being favored25. Ongoing trials will help establish whether further lowering of LDL-C has sufficient beneficial effects to implement PCSK9i on a wider scale and whether the cost-effectiveness of such implementation is favourable. In conclusion, conventional LLT had its core role in a first line of treatment to provide an initial assessment of LDL-C lowering by using various types of statin in various dosages with frequent attempts of double and triple LLT prior to PCSK9i initiation in patients in high or very high risk of CVD. The main usage of PCSK9i were in patients predominately with IHD in treatment with statin and ezetemibe in agreement with current national and international clinical guidelines.

Strengths and limitations of this study

In the present study, we utilised the Danish national registries to perform a description of the characteristics of patients prescribed with PCSK9i on a national basis. We were able to include an unselected cohort of patients, with a great variety of comorbidities. The Danish nationwide registries include data on all residents regardless of participation in the labour market and therefore are not affected by selection bias caused by including selected age groups, hospitals or health insurance systems.

Patients with familial hypercholesterolaemia (FH), who are also eligible for PCSK9i treatment, collect their PCSK9i at a specialised hospital unit and due to the design of the present study, we were not able to identify patients with a diagnosis of FH.

Due to the observational design of the present study, we were not able to retrieve data regarding biochemical analyses, clinical reasons for initiation of PCSK9i, LLT after initiation of PCSK9i or adverse events with statin treatment. Both statin and non-statin treatment in different dosages were applied in almost every patient in the present study, which may reflect clinical decisions of resolving intolerable adverse effects, reaching LDL-C target levels or strict adherence to guidelines. Due to a low diagnostic sensibility, we were not able to fully discriminate the indications of patients receiving PCSK9i.

We did not include a comparative cohort of patients not receiving PCSK9i; however, we were able to compare clinical characteristics, comorbidities and demographics in the present study with an observational study investigating CVD and LLT in the Danish general population.39