The Patient Journey with PCSK9 Inhibitors in Community Practice

Abstract

BACKGROUND:

Trials have demonstrated that proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors are effective as an adjunct to statin therapy, but access and cost issues have limited their use in community practice.

OBJECTIVE:

To better understand patients’ experiences when trying to obtain, fill, and use PCSK9 inhibitor therapy in community practice.

METHODS:

We conducted a patient survey to evaluate patient experiences with PCSK9 inhibitors including medication initiation, indication for treatment, insurance approval status, medication persistence, and reason for discontinuation. The survey was emailed to 4,740 adults who used a patient access support program.

RESULTS:

Overall, 1,327 of 4,740 adults completed the survey (28.0% response rate). Of those, 75.0% were 60+ years old, 52.8% were male, and 92.4% were white. At the time of PCSK9 inhibitor prescription, 70.2% were not on a statin (with 84.4% of those not on a statin reporting statin intolerance). Overall, 74.6% of patients found the drug approval process to be “somewhat” or “very” burdensome. Among n=1,216 patients who initiated treatment, 33.7% discontinued by the time of the survey, with 50.0% taking the drug for 1–6 months. Patient out-of-pocket costs were the leading reported reason for discontinuation.

CONCLUSIONS:

The majority of PCSK9 inhibitor users in community practice were not on a statin, presumably due to statin intolerance. The drug approval process and costs continue to be strong reasons for lower initiation of PCSK9 agents, as well as higher discontinuation rates.

Introduction

In 2015, two proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors, alirocumab and evolocumab, were approved for patients with atherosclerotic cardiovascular disease (ASCVD) or heterozygous familial hypercholesterolemia (HeFH) who required additional lipid-lowering beyond maximally tolerated statin therapy. Subsequently, two large cardiovascular outcomes trails have demonstrated the efficacy of PCSK9 inhibitors in reducing cardiovascular disease events.^1,2 Nevertheless, in the year following the approval of PCSK9 inhibitors, their use was low. Furthermore, only half of PCSK9 inhibitor prescriptions received payer approval, and up to a one-third of approved prescriptions were never filled. The reasons for these low rates are due in part to insurance barriers,³ which lead to high rates of payer rejections and increased out-of-pocket expenses for patients.⁴ To date, there is limited information on patients’ experiences when trying to obtain, fill, and use PCSK9 inhibitor therapy in community practice.

Methods

Patient population

We conducted a survey of all patients who had participated in a company-sponsored patient support program for PCSK9 inhibitor therapy; when agreeing to participate in the portal, these patients consented to being contacted for research purposes. The support program provided patients with information about therapy and provided assistance with medication initiation. A total of 4,740 adults who consented to being contacted and had a valid email addresses were sent a link to complete an online survey between March 7, 2017 and November 2, 2017.

Among adults who consented to be contacted for research purposes and had a valid phone number, a further unique subset of adults was randomly selected to participate in a phone-based survey. The phone-based survey was conducted given the anticipated lower response rate for the email-based survey compared with what expected using a telephone contact, as well as potential differences in the population of patients willing to respond to a phone vs. email-based survey. The goal sample size for the phone survey was 500 adults. A centralized call center at the Duke Clinical Research Institute contacted eligible adults between August 23, 2017 and December 20, 2017 until a final sample size of 500 were reached. Since calls were made in “batches,” the final sample size was 503 adults.

Survey characteristics

The online survey included participant self-report of demographics, clinical characteristics, PCSK9 inhibitor use and approval status, drug persistence behaviors, and patient beliefs about the medication. Patient beliefs about the medication were assessed using concepts from the Injection-Treatment Acceptance Questionnaire, including all questions from the domains “perceived efficacy” and “injection self-efficacy” and a subset of questions of the domains “acceptance of side effects,” “injection convenience,” and “overall acceptance.”^5,6

The phone survey included similar questions to the online survey with two exceptions: 1) additional questions were added to clarify the presence of ASCVD and the timing of statin discontinuation (if applicable) based on preliminary results from the online survey; and 2) questions in the online survey addressing the burden of the approval process and patient beliefs about drug efficacy, tolerability, and acceptability were shortened on the phone survey due to time constraints.

Participants provided informed consent (electronic for the online survey, verbal for the phone survey), and the study received institutional review board approval from Duke University.

Analysis

Analyses were stratified by survey type and comparisons of responses between the two survey types were performed to assess representativeness of the sample. Patient characteristics were compared between groups based on use and approval status. In descriptive analyses, categorical variables were summarized using percentages and compared using Chi-squared tests. Patients who had missing data to questions about medication prescription, use status, or approval status were excluded from the final analyses. All statistical analyses were performed using SAS version 9.4 (SAS, Cary, NC).

Results

Of the 4,740 patients contacted via email, 1,023 (21.6% click rate) clicked on the email link to access the survey, of whom 85.9% (n=879) completed the consent process. After reading the consent, 55 patients declined to participate, leaving 824 who completed the survey, for an email response rate of 17.4%. In the phone survey, a total of 752 adults were contacted, of whom 503 (66.9% response rate) agreed to participate in the survey (Supplemental Figures 1 and 2). Table 1 describes the patient characteristics of all patients who consented to take the survey. There were slight differences in the demographic characteristics of adults who took the survey via email and phone (Table 1), with those completing the survey via email being slightly younger, more likely to report being White or Caucasian race, having government insurance (e.g. Medicare, Medicaid), higher education, and higher income. There were also differences in the reported indication for treatment.

Table 1.

Characteristics of survey respondents overall and by survey type

Patient characteristics	Overall (n=1327)	Email consent (n=824)	Phone consent (n=503)	p-difference
Age				0.0007
<40	21 (1.7%)	17 (2.2%)	4 (0.9%)
40-59	290 (23.3%)	190 (24.4%)	100 (21.5%)
60-79	852 (68.5%)	537 (68.9%)	315 (67.7%)
More than 80	81 (6.5%)	35 (4.5%)	46 (9.9%)
Sex (% male)	52.8	54.0	50.9	0.29
Race				0.009
Caucasian/White	1132 (92.4%)	721 (94.1%)	411 (89.5%)
Black	46 (3.8%)	20 (2.6%)	26 (5.7%)
Other	47 (3.8%)	25 (3.3%)	22 (4.8%)
Education				0.0004
High school or less	193 (20.1%)	105 (16.5%)	88 (27.2%)
At least some college	350 (36.5%)	239 (37.6%)	111 (34.4%)
Graduate degree	415 (43.3%)	291 (45.8%)	124 (38.4%)
Household income				<0.0001
<$30,000	224 (18.1%)	89 (11.4%)	135 (29.5%)
$30,000-$69,999	753 (60.9%)	490 (63.0%)	263 (57.4%)
>$70,000	259 (21.0%)	199 (25.6%)	60 (13.1%)
Household income				0.0004
<$30,000	193 (20.1%)	105 (16.5%)	88 (27.2%)
$30,000-$69,999	350 (36.5%)	239 (37.6%)	111 (34.4%)
>$70,000	415 (43.3%)	291 (45.8%)	124 (38.4%)
Insurance type				0.02
Private	746 (56.5%)	446 (54.3%)	300 (60.0%)
Government	462 (35.0%)	312 (38.0%)	150 (30.0%)
None	87 (6.6%)	50 (6.1%)	37 (7.4%)

^*In the phone survey, ASCVD was assessed by asking patients about a history of stroke, heart attack, coronary stent, or heart bypass. In the email survey, this was broadened to include coronary artery disease, angina, abdominal aortic aneurysm, and transient ischemic attacks.

^†Phone survey only

^‡Email survey only

ASCVD, atherosclerotic cardiovascular disease; HeFH, heterozygous familial hyperlipidemia; PCSK9, proprotein convertase subtilisin/kexin type 9 inhibitors

Patients could have been referred to the patient support program from which we identified the sample without having a prescription for a PCSK9i. Of adults who consented to take the survey, n=21 in the email survey and n=34 in the phone survey reported never being prescribed a PCSK9 inhibitor (or did not remember being prescribed one) and were excluded from subsequent analyses. The final sample size of adults prescribed a PCSK9 inhibitor who completed the survey was 1,269 (803 email and 466 phone).

Patients prescribed PCSK9 inhibitors

Table 2 lists characteristics of adults prescribed a PCSK9 inhibitor who completed a survey. Most adults were older than 60 (75.0%), reported White or Caucasian race (92.4%), and had at least some college education or a graduate degree (81.9%). There was a slight majority of males (52.9%) and adults with private insurance (59.0%). The leading patient-reported indication for therapy was ASCVD (n=459, 36.2%) or ASCVD and HeFH (31.2%); however, 14.1% of patients did not know their indication, had missing data, or reported their indication as “other.” While cardiologists were the most common prescribing provider type, primary care physicians were responsible for 12.9% of all PCSK9 prescriptions.

Table 2.

Characteristics of patients prescribed PCSK9 inhibitor

Patient characteristics	All patients prescribed PCSK9 inhibitors (n=1269)*
Age
<40	21 (1.7%)
40-59	290 (23.3%)
60-79	851 (68.5%)
More than 80	81 (6.5%)
Sex (% male)	657 (52.9%)
Race
Caucasian/White	1131 (92.4%)
Black	46 (3.8%)
Other	47 (3.8%)
Education
High school or less	224 (18.1%)
At least some college	752 (60.9%)
Graduate degree	259 (21.0%)
Household income
<$30,000	193 (20.2%)
$30,000-$69,999	350 (36.5%)
>$70,000	415 (43.3%)
Insurance type
Private	745 (59.0%)
Government	462 (36.6%)
None	30 (2.4%)
Insurance medication coverage
Full	824 (66.9%)
Partial	356 (28.9%)
None	51 (4.1%)
Clinical characteristics†
HeFH only	235 (18.5%)
ASCVD only	459 (36.2%)
ASCVD and HeFH	396 (31.2%)
Other/no response /missing	179 (14.1%)
Currently taking any statin	376 (29.8%)
Currently taking high-intensity statin	190 (15.5%)
Not taking statin	884 (70.2%)
Currently taking ezetimibe	236 (19.3%)
Not taking statin
Phone: Reason for not taking statin‡
Side effects	284 (85.8%)
Other	37 (11.2%)
Never took a statin	10 (3.0%)
Email: Reason for not taking statin§
Side effects	462 (84.3%)
Stopped when starting PCSK9 inhibitor	44 (8.0%)
Other	39 (7.1%)
Never took a statin	3 (0.6%)
Prescribing doctor
Cardiologist	963 (76.3%)
Endocrinologist	93 (7.4%)
Primary care	163 (12.9%)
Other	44 (3.5%)
Patient involvement in approval¶
Completing paperwork	276 (23.3%)
Getting paperwork to provider	441 (37.2%)
Calling insurer	340 (28.7%)
None of the above	84 (6.6%)
Burden of approval process#	(n=496)
Not at all burdensome	126 (25.4%)
Somewhat burdensome	199 (40.1%)
Very burdensome	171 (34.5%)

^*134 respondents were prescribed PCSK9 inhibitors but had subsequent missing data or did not know their approval status

^†In the phone survey, ASCVD was assessed by asking patients about a history of stroke, heart attack, coronary stent, or heart bypass. In the email survey, this was broadened to include coronary artery disease”, angina, abdominal aortic aneurysm, and transient ischemic attacks.

^‡Phone survey only

^§Email survey only

^¶Patient involvement options are not mutually exclusive.

^#The question of if patients stopped statin after initiating PCSK9 inhibitors was not asked on phone survey to prevent potentially prompting patients to stop therapy inappropriately after seeing high rates of this on the first email survey.

ASCVD, atherosclerotic cardiovascular disease; HeFH, heterozygous familial hyperlipidemia; PCSK9, proprotein convertase subtilisin/kexin type 9 inhibitors

Overall, statin utilization was low; only 29.8% of those surveyed were taking a statin, and 15.5% were taking a high-intensity statin (defined as atorvastatin 40 mg or 80 mg, and rosuvastatin 20 mg or 40 mg). Similarly, very few (19.3% overall) of those prescribed a PCSK9 inhibitor were taking ezetimibe. Very few adults (n=13) reported never trying a statin; the majority of adults who were not on a statin had previously tried one but discontinued due to side effects (84.9%) (Table 1).

PCSK9 inhibitor initiation and self-reported persistence and adherence

Figure 1 demonstrates the breakdown of treatment status at the time of the survey. Of the 1,263 adults with data on current PCSK9 inhibitor utilization, n=1,216 (96.3%) subjects had received at least one dose of PCSK9 inhibitor. Among the 1,216 who initiated therapy, n=779 (61.3%) were still on treatment at the time of the survey while n=428 (33.7%) had discontinued therapy by the time they completed the survey.

Figure 1. Patient response to survey.

Patient response to survey contact and PCSK9 inhibitor prescription PCSK9, proprotein convertase subtilisin/kexin type 9 inhibitor

Figure 2 shows patient-reported reasons for never initiating PCSK9 inhibitor therapy (among n=47 adults who were prescribed, but never received therapy) or discontinuing therapy (among n=428 adults who initiated but discontinued treatment). In both groups, the leading patient-reported reasons for lack of PCSK9 inhibitor utilization was lack of insurance approval (54.3% of those who never initiated therapy and 34.4% of those who stopped therapy), followed by high costs to the patient (39.1% of those who never initiated therapy and 38.2% of those who stopped therapy). Among those who discontinued treatment, 24.4% reported discontinuing due to perceived side effects. Rates of discontinuation were similar among those with a reported history of statin intolerance (20.4%) and those without statin intolerance (26.8% discontinued PCSK9 inhibitor).

Figure 2. Reason for not currently using PCSK9 inhibitors.

Reason for not currently using PCSK9 inhibitors and whether the drug was discontinued or never received. PCSK9, proprotein convertase subtilisin/kexin type 9 inhibitor

Among patients who started a PCSK9 inhibitor, slightly more than half (64.1%, n= 779) were still taking their prescribed PCSK9 inhibitor at the time of the survey. Table 3 shows drug adherence and persistence behavior by patient PCSK9 inhibitor use status. There were no significant differences between those currently on therapy and those who had discontinued therapy based on age, race, or insurance type. Nevertheless, those on therapy were more likely to be male (57.0% vs. 45.3%, p=0.0005), report a graduate degree (22.9% vs. 16.8%, p=0.02), and have a higher household income. Among current users, 85.3% had been on therapy for more than 6 months. Of those who discontinued therapy, 50.0% stopped the medication after 1–6 months, while 42.6% stopped after ≥6 months. Patients currently on therapy reported missing a dose or having a delay to dose of more than 1 week significantly more than those who had discontinued therapy (28.2% of those on therapy vs. 9.3% of those who discontinued, p=<0.0001) or forgotten to take their medication on the day it was due (36.7% vs. 22.2%, p=<0.0001). While many reported skipping a prescription fill due to price (17.7% of those on therapy vs. 49.4% of those who discontinued), few reported intentionally spacing out doses to make the prescription last longer (6.1% vs. 4.8%, p=0.37).

Table 3.

Patient persistence and adherence with PCSK-9 inhibitors by use status

Patient persistence and adherence*	On therapy (n=779)	Discontinued therapy (n=428)	p-value
Age			0.13
<40	12 (1.5%)	6 (1.4%)
40-59	193 (24.8%)	86 (20.2%)
60-79	521 (67.1%)	297 (69.7%)
More than 80	45 (5.8%)	36 (8.5%)
Sex (% male)	443 (57.0%)	193 (45.3%)	0.0005
Race			0.10
Caucasian/White	706 (91.0%)	380 (89.8%)
Black	25 (3.2%)	21 (5.0%)
Other	28 (3.6%)	19 (4.5%)
Education			0.02
High school or less	128 (16.8%)	89 (21.0%)
At least some college	461 (60.3%)	264 (62.3%)
Graduate degree	175 (22.9%)	71 (16.8%)
Household income			<0.0001
<$30,000	97 (12.6%)	92 (22.0%)
$30,000-$69,999	211 (27.4%)	123 (29.4%)
>$70,000	288 (37.4%)	107 (25.5%)
Insurance type			0.97
Private	459 (59.3%)	257 (60.2%)
Government	290 (37.5%)	155 (36.3%)
None	9 (1.2%)	6 (1.4%)
Insurance medication coverage			0.43
Full	512 65.9%)	281 (66.0%)
Partial	222 (28.6%)	119 (27.9%)
None	29 (3.7%)	22 (5.2%)
Clinical characteristics			0.01
HeFH only	138 (17.7%)	87 (20.3%)
ASCVD only	288 (37.0%)	149 (34.8%)
ASCVD and HeFH	269 (34.5%)	115 (26.9%)
Other/no response /missing	84(10.8%)	77 (18.0%)
Months using drug			<0.0001
Less than 1 month	2 (0.3%)	31 (7.4%)
1-6 months	111 (14.4%)	210 (50.0%)
>6 months	658 (85.3%)	179 (42.6%)
Medication adherence (% yes)
Missed or delayed doses by 1 week	218 (28.2%)	39 (9.3%)	<0.0001
Forgotten to take dose	283 (36.7%)	93 (22.2%)	<0.0001
Intentionally spaced doses	47 (6.1%)	20 (4.8%)	0.37
Skipped filling due to price	130 (17.7%)	203 (49.4%)	<0.0001
None of the above	309 (39.8%)	164 (39.0%)	0.78

^*9 respondents answered the question about receiving therapy, but had subsequent missing data

ASCVD, atherosclerotic cardiovascular disease; HeFH, heterozygous familial hyperlipidemia; PCSK9, proprotein convertase subtilisin/kexin type 9 inhibitors

Table 4 compares patient-reported beliefs about disease risk and PCSK9 inhibitors between those who were still on therapy and those who had discontinued or did not initiate treatment for the subset of patients who completed the email survey. Of email responses, the majority of both groups reported confidence in drug efficacy, reporting they were “quite” or “very confident” in the drug treating the condition (89.6%, p=0.0001), though those on therapy did have higher overall confidence in treatment effectiveness than those who discontinued. Similarly, those currently on therapy had higher beliefs in the safety of PCSK9 inhibitors and acceptability of therapy; 82.6% of current users found therapy side effects “acceptable” or “very acceptable” compared to 55.7% of former users (p≤0.0001), and 94.0% vs. 84.8% reported the drug to be overall “acceptable” or “very acceptable” (current vs. non-users, respectively, p=0.002). Drug administration was well tolerated; 92.9% overall reported it was “easy” or “very easy” to give the injection.

Table 4.

Patient beliefs about PCSK9 inhibitors by use status^*

Patient beliefs	Overall (n=520)	On therapy with approval (n=406)	Discontinued/did not initiate with approval (n=114)	p-value
Drug effectiveness
Confidence in drug treating condition				<0.0001
Not at all confident	6 (1.2%)	0 (0.0%)	6 (6.0%)
A little confident	8 (1.6%)	6 (1.5%)	2 (2.0%)
Somewhat confident	37 (7.5%)	29 (7.4%)	8 (8.0%)
Quite confident	85 (17.3%)	68 (17.4%)	17 (17.0%)
Very confident	355 (72.3%)	288 (73.7%)	67 (67.0%)
Injection effectiveness in treating condition				0.003
Not at all effective	5 (1.2%)	1 (0.3%)	4 (4.1%)
A little effective	8 (1.8%)	4 (1.2%)	4 (4.1%)
Somewhat effective	26 (6.0%)	19 (5.6%)	7 (7.2%)
Quite effective	68 (15.7%)	49 (14.5%)	19 (19.6%)
Very effective	327 (75.4%)	264 (78.3%)	63 (65.0%)
Drug safety
Acceptability of side effects				<0.0001
Very unacceptable	42 (8.6%)	18 (4.7%)	24 (23.1%)
Unacceptable	18 (3.7%)	4 (1.0%)	14 (13.5%)
Neither acceptable nor unacceptable	53 (10.8%)	45 (11.7%)	8 (7.7%)
Acceptable	114 (23.3%)	94 (24.4%)	20 (19.2%)
Very acceptable	262 (53.6%)	224 (58.2%)	38 (36.5%)
Drug delivery
Acceptability of pain at injection site				0.009
Very unacceptable	19 (3.8%)	15 (3.7%)	4 (3.9%)
Unacceptable	9 (1.8%)	3 (0.8%)	6 (5.9%)
Neither acceptable nor unacceptable	45 (9.0%)	35 (8.7%)	10 (9.8%)
Acceptable	131 (26.0%)	102 (25.4%)	29 (28.4%)
Very acceptable	299 (59.4%)	246 (61.4%)	53 (52.0%)
Confidence in ability to give injection				0.01
Not at all confident	3 (0.6%)	3 (0.8%)	0 (0.0%)
A little confident	5 (1.0%)	4 (1.0%)	1 (1.0%)
Somewhat confident	13 (2.6%)	11 (2.8%)	2 (2.0%)
Quite confident	59 (11.8%)	37 (9.3%)	22 (21.6%)
Very confident	422 (84.1%)	345 (86.3%)	77 (75.5%)
Ease of giving injection				0.17
Very difficult	1 (0.2%)	1 (0.3%)	0 (0.0%)
Difficult	8 (1.6%)	6 (1.5%)	2 (2.0%)
Neither difficult nor easy	27 (5.4%)	25 (6.2%)	2 (2.0%)
Easy	86 (17.1%)	62 (15.5%)	24 (23.5%)
Very easy	381 (75.8%)	307 (76.6%)	74 (72.6%)
Drug acceptability
Overall acceptability of treatment				0.0001
Very unacceptable	15 (3.0%)	9 (2.3%)	6 (5.7%)
Unacceptable	6 (1.2%)	1 (0.3%)	5 (4.8%)
Neither acceptable nor unacceptable	19 (3.8%)	14 (3.5%)	5 (4.8%)
Acceptable	98 (19.4%)	72 (18.1%)	26 (24.8%)
Very acceptable	366 (72.6%)	303 (75.9%)	63 (60.0%)

^*Email survey only

PCSK9, proprotein convertase subtilisin/kexin type 9 inhibitors

Insurance approval and patient process experience

Not all of those who were on therapy or had discontinued therapy received insurance approval. Among adults who had started PCSK9 inhibitor therapy and for whom approval status was known, the approval rate was 74.9%. Of those still on therapy, n=631 (81.0%) reported receiving approval, n=95 (12.2%) reported not receiving approval, and approval status was missing or unknown in n=53 (6.8%). Of those who never initiated or discontinued therapy (n=475), n=205 (43.2%) had received approval, n=204 (42.9%) had not received approval, and approval status was unknown for n=66 (13.9%). Characteristics of patients according to approval status are described in Supplemental Table 1. Those who received approval were older and more likely to report government insurance and an indication of ASCVD than those who did not receive approval. There was no difference in education or income by approval status, nor was statin utilization different among those who received approval and those who did not.

The majority of patients were engaged in the prior authorization process in some capacity, with the most cited method being getting paperwork to/from the doctor or pharmacy (37.2%). There were similar rates of participation in the prior authorization process across approval and utilization status (Table 1). The majority of all groups found the process to be “somewhat” or “very” burdensome (74.6%).

Discussion

While the provider experience with PCSK9 inhibitor therapy has been documented,⁷ the patient experience (including the approval process and use) has not previously been well described. In this survey, we identified first that PCSK9 inhibitor utilization has been largely confined to patients with statin intolerance, rather than patients with high on-statin LDL-C levels. Next, we reconfirmed that cost and lack of insurance approval decreased patient initiation of therapy. Finally, we identified that even among patients who receive therapy, persistence is suboptimal, with discontinuation driven by both cost and insurance approval as well as patient-perceived side effects.

Large outcomes trials have demonstrated the safety and efficacy of PCSK9 inhibitors in reducing the risk of cardiovascular events when added to a background of statin therapy,^1,2 and at the time of the study, the Food and Drug Administration label states that PCSK9 inhibitors are to be used in addition to maximally-tolerated doses of statin.⁸ Yet despite existing evidence and indication, the majority of patients in this study who had been prescribed a PCSK9 inhibitor were not on a statin and 84.4% of those patients reported a statin intolerance. Whether these patients meet strict criteria for statin “intolerance” is unknown, but at minimum it appears that the majority of those who are being prescribed a PCSK9 inhibitor self-reported being statin intolerant. Our rate of statin utilization (29.8%) was slightly lower than what was shown by Rane et al., who reported that only 40.5% of patients prescribed PCSK9 inhibitors were currently using a statin by physician report, and similar to what was found in a large national analysis of pharmacy claims data, where 76.9% of those prescribed PCSK9 inhibitor were not on a statin.^4,9 Importantly, these prescriptions were not inappropriate: those who are statin intolerant are at high risk of recurrent events^10,11 and PCSK9 inhibitors have been shown to be effective in lowering low-density lipoprotein cholesterol (LDL-C) in those with statin intolerance.^12,13 The demand for lipid-lowering therapy options for statin intolerant patients may have driven uptake given the lack of previously available therapeutic options; however, efforts to increase uptake of PCSK9 inhibitor therapy should now focus on those with ASCVD who are at high risk for recurrent events and persistently high LDL-C despite maximally tolerated therapy, as studied in the trials.

Furthermore, there was a small subset of patients in our study who reported stopping statins following PCSK9 inhibitor initiation. Providers should continue to emphasize to patients that the benefits of PCSK9 inhibitor therapy are in addition to statins and encourage continued statin therapy while on PCSK9 inhibitors. Continuing statin therapy after starting PCSK9 inhibitors is also important to maintain insurance coverage for therapy; many insurance companies require documentation of LDL-C lowering on PCSK9 inhibitors. If patients stop statin therapy after starting PCSK9 inhibitors, their apparent LDL-C lowering on PCSK9i will be reduced and they may not be able to obtain re-authorization for ongoing therapy. Our study also found a low rate of ezetimibe use in those prescribed PCSK9 inhibitors (19.3%). This survey did not further evaluate the reason for lack of ezetimibe use. However, at the time of the survey guidelines did not recommend using ezetimibe prior to initiating PCSK9 inhibitor, and payer requirements for ezetimibe as a step-through therapy also varied.¹⁴ Whether the recommendation to try ezetimibe first in the latest lipid guidelines leads to an increase in ezetimibe use in those prescribed PCSK9 inhibitors remains to be seen.

At the time of the survey, we found that even among patients who were enrolled in a company-sponsored patient portal with access to support for navigating the prior authorization process, only 66.9% of patients prescribed PCSK9 inhibitors had received insurance approval. Lack of payer approval was a leading reason that patients did not start or discontinued therapy; however even among those who received approval and were still on therapy, most patients reported the approval process to be “burdensome” or “very burdensome.”

As shown in prior studies, a key barrier to patient utilization of PCSK9 inhibitors was cost, which was cited by nearly 40% of those who discontinued or did not initiate treatment. Patient out-of-pocket costs are strongly correlated with primary non-adherence, or lack of patients’ filling a prescription even after approval. This survey reveals that costs also appear to impact patient long-term persistence to treatment. There is hope in the fact that the recently announced 60% decrease in price of both drugs may translate into reduced out-of-pocket costs for those with government insurance.^15,16 For those that are still unable to afford the prescription, improving access to patient assistance programs may help improve long-term adherence to PCSK9 inhibitor therapy.

Largely due to cost and lack of insurance approval, one in three adults who initiated a PCSK9 inhibitor discontinued therapy, with 24.4% of those who discontinued treatment citing side effects from PCSK9 inhibitor therapy as the reason (representing around 8% of those who initiated therapy). This number was slightly higher than what was seen in the Further Cardiovascular Outcomes Research With PCSK9 Inhibition in Subjects With Elevated Risk (FOURIER) and Evaluation of Cardiovascular Outcomes After an Acute Coronary Syndrome During Treatment With Alirocumab (ODYSSEY OUTCOMES) trials, in which 628 of 13,784 (4.6%) and 362 of 9,462 (3.8%), respectively, discontinued therapy due to adverse events.^1,2 Importantly, in both trials, similar numbers of participants in the placebo arm also discontinued due to adverse events (4.2% [n=581] of 13,780 in FOURIER, and 3.7% [n=347] of 9,462 in ODYSSEY OUTCOMES),^1,2 which may suggest that much of the intolerance to PCSK9 inhibitors in clinical trials may be related unrelated to the active therapy or may reflect a “nocebo effect” as seen commonly in statin clinical trials. Our study was not designed to obtain detailed information on patient-perceived side effects, nor was it designed to determine whether patient-reported side effects were due to a true causal link vs. a nocebo effect. However, it is important to note that studies evaluating treatment acceptance suggest a high level of drug acceptance^5,6,17 and multiple studies have shown that evolocumab and alirocumab have been well tolerated by patients.^18,19 Clinicians should monitor patients for both real and perceived side effects and help determine if side effects may be related to PCSK9 inhibitors. Both patients and providers should report any potential PCSK9 inhibitor-related side effects to the Food and Drug Administration’s MedWatch program. Nevertheless, despite the side effect rates reported in this survey, most patients found PCSK9 inhibitors acceptable and believe that they are effective. Even patients who had discontinued therapy reported high rates of confidence in drug efficacy and tolerability. Identifying and addressing barriers to long-term persistence to PCSK9 inhibitors is critical to achieve the population health benefits of the medication; in clinical trials, the event rates between the treated and placebo arms did not separate for the first 6–12 months. ^1,2

This study has several limitations. First, the subsample of patients who chose to enroll in the patient portal may not be representative of the overall population. For example, patients less satisfied with their experience with PCSK9 inhibitors or who did not feel the medication was effective may have been less likely to participate in the patient portal or to participate in the survey. Second, indication for therapy was self-reported in this survey, which is inherently subject to recall bias and may not have been well understood by patients at the time of the survey. Next, this study identified patients referred to a company-sponsored support program for alirocumab. As a result, the results likely apply more to patients prescribed alirocumab. Because patients may be prescribed one brand and switched to another, or prescribed both pending insurance approval, we asked patients about therapy as a class and did not evaluate differences by therapy type. Importantly, this survey was conducted prior to the results of the FOURIER and ODSYSSEY OUTCOMES trials, after which new agreements with payers have been undertaken with the goal to lower the bar to approval. In addition, the cost of both PCSK9 inhibitors have been significantly decreased since the time of this survey. Finally, arrangements between the manufacturers and health plans have changed since the survey. For example, Express Scripts, Inc., which is a large pharmacy benefits manager in the United States, announced they will pass a portion of their rebates to the patients who are effectively lowering out-of-pocket expenses.²⁰ Also, companies have worked with payers to lower the prior authorization requirements for providers. Further exploration is needed to better understand the impact of these changes on patient initiation and persistence to PCSK9 inhibitors.

Conclusion

In conclusion, in contrast to what was studied in trials, PCSK9 inhibitors appear to be prescribed more commonly in adults with statin intolerance than those on high-intensity statins. Efforts to expand utilization of PCSK9 inhibitors should focus on high-risk adults with persistently high LDL-C, despite statin therapy. Lack of insurance approval and high out-of-pocket costs are consistent barriers to both initiation and long-term persistence to PCSK9 inhibitor therapy. Continued efforts between payers and pharmaceutical companies to decrease access barriers will be critical to ensure the maximal effectiveness of PCSK9 inhibitors to reduce the burden of ASCVD.

Highlights.

• 70.2% of patients were not on a statin at the time of PCSK9 inhibitor prescription

• 74.6% of patients found the drug approval process to be burdensome

• Most common reason for PCSK9 inhibitor discontinuation was out-of-pocket costs