1 Project overview

1.1 Background

There is limited data on real-world outcomes associated with use of prescription cannabidiol (CBD) therapy. The purpose of this study was to describe the use of emergency healthcare services, time to first emergency healthcare service use, and change in anti-seizure medication (ASM) use over the first 12-months of prescription CBD therapy.

1.2 Methods

This was a single-center, prospective cohort study of patients prescribed CBD by a neurology clinic provider and fulfilled through the center’s specialty pharmacy from January 2019 through April 2020. Patients enrolled in a clinical trial for CBD or for whom the insurance approval process was not completed by the center’s specialty pharmacy were excluded. Descriptive statistics were used to summarize the data with results presented as medians with interquartile ranges (IQR). Pediatric age was defined as less than 18 years.

1.3 Outcomes

Clinical Characteristics

Demographic characteristics of study population (age, sex, race, ethnicity, insurance type, height/weight)
Prior AED therapies trialed (number, type)
Previous non-pharmacological treatments trialed (dietary, surgical, VNS, DBS)
Number of drug-drug interactions at baseline
Obtainment of baseline HFP
Route of administration
Number and type of concurrent AED therapies upon Epidiolex initiation
Number and type of drug-drug interactions present upon Epidiolex initiation
Outcome of drug interaction management upon Epidiolex initiation

Treatment Outcomes

Liver function lab results over the first 12-months of Epidiolex therapy
Adverse events reported during the first 12-months of Epidiolex therapy
Describe the use of emergency healthcare services over the first 12-months of Epidiolex therapy
Time to first emergency healthcare service use during the first 12-months of Epidiolex therapy
Change in ASM use over the first 12-months of Epidiolex therapy
Evaluate adherence, discontinuation rates, and reasons for non-adherence and discontinuation of Epidiolex during the first 12-months post-initiation

2 Exclusions

	N	N=160
Exclusions
Reason for exclusion	160
Participation in clinical trial		1.2% ( 2)
Receiving medication through manufacturer		0.6% ( 1)
Receiving from external specialty pharmacy		10.6% ( 17)
VSP did not complete PA process		2.5% ( 4)
No exclusion criteria met. Patient should be INCLUDED in the study.		85.0% (136)
N is the number of non-missing values. Numbers after proportions are frequencies.

3 Descriptive statistics

3.1 Demographics

	N	Adult N=44	Pediatric N=92	Combined N=136
Demographics
Age at start of prescription therapy	136	21.2 27.5 43.9 (32.8 ±13.3)	4.6 10.0 13.8 ( 9.5 ± 5.1)	8.6 13.8 20.9 (17.1 ±13.9)
Patient age category	136
Adult		100% (44)	0% ( 0)	32% (44)
Pediatric		0% ( 0)	100% (92)	68% (92)
Gender	136
Female		57% (25)	47% (43)	50% (68)
Male		43% (19)	53% (49)	50% (68)
Race	136
White		86.4% ( 38)	83.7% ( 77)	84.6% (115)
Black or African American		11.4% ( 5)	10.9% ( 10)	11.0% ( 15)
Other Asian		0.0% ( 0)	1.1% ( 1)	0.7% ( 1)
Decline to Answer		2.3% ( 1)	0.0% ( 0)	0.7% ( 1)
Unknown		0.0% ( 0)	4.3% ( 4)	2.9% ( 4)
Insurance type	136
Commercial		22.7% (10)	19.6% (18)	20.6% (28)
Medicare		45.5% (20)	0.0% ( 0)	14.7% (20)
Medicaid		31.8% (14)	72.8% (67)	59.6% (81)
Tricare		0.0% ( 0)	7.6% ( 7)	5.1% ( 7)
Other		0.0% ( 0)	0.0% ( 0)	0.0% ( 0)
Height (cm)	122	153 164 173 (163 ± 15)	102 130 147 (126 ± 30)	120 146 163 (139 ± 31)
Weight (kg)	136	49 62 76 (67 ±24)	17 29 38 (34 ±23)	24 37 60 (44 ±28)
Route of administration	136
By mouth (PO)		93.2% ( 41)	78.3% ( 72)	83.1% (113)
G-tube		6.8% ( 3)	18.5% ( 17)	14.7% ( 20)
Other		0.0% ( 0)	3.3% ( 3)	2.2% ( 3)
a b c represent the lower quartile a, the median b, and the upper quartile c for continuous variables. x ± s represents X ± 1 SD. N is the number of non-missing values. Numbers after proportions are frequencies.

3.1.1 Age, height and weight

3.1.1.1 Pediatric

3.1.1.2 Adult

3.2 Medications

3.2.1 Previous ASM

	N	Adult N=44	Pediatric N=92	Combined N=136
Previous ASM by age group
Number of previous medications	136	8.0 11.5 14.0 (11.5 ± 4.0)	5.0 6.0 8.0 ( 6.7 ± 3.1)	5.0 7.0 11.0 ( 8.2 ± 4.1)
Number of previous medications	136
2		0.0% ( 0)	4.3% ( 4)	2.9% ( 4)
3		0.0% ( 0)	7.6% ( 7)	5.1% ( 7)
4		2.3% ( 1)	12.0% (11)	8.8% (12)
5		4.5% ( 2)	19.6% (18)	14.7% (20)
6		0.0% ( 0)	12.0% (11)	8.1% (11)
7		11.4% ( 5)	13.0% (12)	12.5% (17)
8		9.1% ( 4)	7.6% ( 7)	8.1% (11)
9		6.8% ( 3)	2.2% ( 2)	3.7% ( 5)
10		4.5% ( 2)	10.9% (10)	8.8% (12)
11		11.4% ( 5)	3.3% ( 3)	5.9% ( 8)
12		13.6% ( 6)	3.3% ( 3)	6.6% ( 9)
13		2.3% ( 1)	1.1% ( 1)	1.5% ( 2)
14		11.4% ( 5)	2.2% ( 2)	5.1% ( 7)
15		4.5% ( 2)	0.0% ( 0)	1.5% ( 2)
16		6.8% ( 3)	0.0% ( 0)	2.2% ( 3)
17		4.5% ( 2)	0.0% ( 0)	1.5% ( 2)
18		4.5% ( 2)	1.1% ( 1)	2.2% ( 3)
21		2.3% ( 1)	0.0% ( 0)	0.7% ( 1)
a b c represent the lower quartile a, the median b, and the upper quartile c for continuous variables. x ± s represents X ± 1 SD. N is the number of non-missing values. Numbers after proportions are frequencies.

3.2.2 Previous non-pharmacological treatments trialed

	N	Adult N=44	Pediatric N=92	Combined N=136
Previous non-pharm
Previous non-pharmacological treatments trialed	68
Deep brain stimulation (DBS)		3.6% ( 1)	0.0% ( 0)	1.5% ( 1)
Ketogenic diet		10.7% ( 3)	55.0% (22)	36.8% (25)
Surgery		0.0% ( 0)	7.5% ( 3)	4.4% ( 3)
Surgery, Ketogenic diet		0.0% ( 0)	5.0% ( 2)	2.9% ( 2)
Surgery, Vagal nerve stimulation (VNS)		21.4% ( 6)	5.0% ( 2)	11.8% ( 8)
Surgery, Vagal nerve stimulation (VNS), Ketogenic diet		14.3% ( 4)	5.0% ( 2)	8.8% ( 6)
Vagal nerve stimulation (VNS)		21.4% ( 6)	10.0% ( 4)	14.7% (10)
Vagal nerve stimulation (VNS), Ketogenic diet		28.6% ( 8)	12.5% ( 5)	19.1% (13)
N is the number of non-missing values. Numbers after proportions are frequencies.

3.2.3 Drug-drug interactions at baseline

	N	N=136
Interactions
Number of drug-drug interactions	136	1.0 1.0 3.0 (1.7 ±1.5)
Number of drug-drug interactions	136
0		19.9% (27)
1		33.8% (46)
2		20.6% (28)
3		12.5% (17)
4		7.4% (10)
5		3.7% ( 5)
6		2.2% ( 3)
a b c represent the lower quartile a, the median b, and the upper quartile c for continuous variables. x ± s represents X ± 1 SD. N is the number of non-missing values. Numbers after proportions are frequencies.

3.2.4 Concurrent medications at start

	N	N=136
Concurrent medications
Number of concurrent medications	136	2.0 3.0 4.0 (2.7 ±1.3)
Number of concurrent medications	136
0		2.9% ( 4)
1		14.7% (20)
2		27.9% (38)
3		25.0% (34)
4		24.3% (33)
5		3.7% ( 5)
6		1.5% ( 2)
a b c represent the lower quartile a, the median b, and the upper quartile c for continuous variables. x ± s represents X ± 1 SD. N is the number of non-missing values. Numbers after proportions are frequencies.

3.3 Drug Drug interactions at baseline management

	N	Adult N=44	Pediatric N=92	Combined N=136
Management
Was management of a possible drug interaction performed at baseline (prior to starting prescription CBD)?	136
No		25% ( 11)	17% ( 16)	20% ( 27)
Yes		75% ( 33)	83% ( 76)	80% (109)
What type of drug interaction was managed?	109
Pharmacokinetic		9.1% ( 3)	25.0% (19)	20.2% (22)
Pharmacodynamic		42.4% (14)	36.8% (28)	38.5% (42)
Both		48.5% (16)	38.2% (29)	41.3% (45)
N is the number of non-missing values. Numbers after proportions are frequencies.

4 Previous and concurrent medications

4.1 Figure

4.2 Previous medications

4.3 Concurrent medications

5 Previous non-pharmacological treatments trialed

5.1 Totals

5.2 Intersection

Pediatric

6 Number and type of drug-drug interactions present upon Epidiolex initiation

7 Outcome of drug interaction management

7.1 Agents and outcomes

Outcomes of drug interaction management
Agent	n.agent	Outcome	n	pct.outcome
clobazam	63	other agent stopped	1	1.6%
		other agent dose change	5	7.9%
		no medication change needed, pharmacist provided counseling only	57	90.5%
benzodiazepine (such as clonazepam, diazepam, lorazepam, midazolam)	58	no medication change needed, pharmacist provided counseling only	58	100.0%
antiepileptic medication not noted above	55	other agent stopped	3	5.5%
antiepileptic medication not noted above	55	no medication change needed, pharmacist provided counseling only	52	94.5%
valproic acid	26	other agent dose change	1	3.8%
valproic acid	26	no medication change needed, pharmacist provided counseling only	25	96.2%
artisinal CBD	18	other agent stopped	18	100.0%
antihistamine (such as cetirizine, diphenhydramine, loratadine, fexofenadine)	9	no medication change needed, pharmacist provided counseling only	9	100.0%
antipsychotic	6	no medication change needed, pharmacist provided counseling only	6	100.0%
phenobarbital	5	no medication change needed, pharmacist provided counseling only	5	100.0%
antidepressant	4	no medication change needed, pharmacist provided counseling only	4	100.0%
fluoxetine		no medication change needed, pharmacist provided counseling only	4	100.0%
other		no medication change needed, pharmacist provided counseling only	4	100.0%
muscle relaxer	2	no medication change needed, pharmacist provided counseling only	2	100.0%
promethazine	2	no medication change needed, pharmacist provided counseling only	2	100.0%
carbamazepine	1	no medication change needed, pharmacist provided counseling only	1	100.0%
cyproheptadine		no medication change needed, pharmacist provided counseling only	1	100.0%
meclizine		no medication change needed, pharmacist provided counseling only	1	100.0%
mTor inhibitor (everolimus, tacrolimus)		no medication change needed, pharmacist provided counseling only	1	100.0%

7.2 Agents

Summary of agents
	Overall (N=260)
Agent
clobazam	63 (24.2%)
benzodiazepine (such as clonazepam, diazepam, lorazepam, midazolam)	58 (22.3%)
antiepileptic medication not noted above	55 (21.2%)
valproic acid	26 (10.0%)
artisinal CBD	18 (6.9%)
antihistamine (such as cetirizine, diphenhydramine, loratadine, fexofenadine)	9 (3.5%)
antipsychotic	6 (2.3%)
phenobarbital	5 (1.9%)
antidepressant	4 (1.5%)
fluoxetine	4 (1.5%)
other	4 (1.5%)
muscle relaxer	2 (0.8%)
promethazine	2 (0.8%)
carbamazepine	1 (0.4%)
cyproheptadine	1 (0.4%)
meclizine	1 (0.4%)
mTor inhibitor (everolimus, tacrolimus)	1 (0.4%)

7.3 Outcomes

Summary of outcomes
	Overall (N=260)
Outcome
no medication change needed, pharmacist provided counseling only	232 (89.2%)
other agent stopped	22 (8.5%)
other agent dose change	6 (2.3%)

7.4 PK/ PD

Characteristic	N	N = 260¹
new_di_type	260
Pharmacodynamic		184 (70.8%)
Pharmacokinetic		76 (29.2%)
¹ n (%)

8 Obtainment of baseline hepatic function panel (HFP)

8.1 Figure

9 Liver function labs during Epidiolex treatment

Characteristic	N	N = 298¹
AST result	298
High		99 (33.2%)
In range		186 (62.4%)
Low		12 (4.03%)
Not available		1 (0.34%)
ALT result	298
High		83 (27.9%)
In range		201 (67.4%)
Low		11 (3.69%)
Not available		1 (0.34%)
Undetectable		2 (0.67%)
Bili result	298
High		37 (12.4%)
In range		229 (76.8%)
Low		6 (2.01%)
Not available		6 (2.01%)
Undetectable		20 (6.71%)
¹ n (%)

10 Discontinuation

10.1 Discontinuation summary

Characteristic	N	N = 136¹
Did the patient discontinue medication	136
No		105 (77.2%)
Yes		31 (22.8%)
Time to discontinuation	31
Mean (SD)		148.0 (108.6)
Median (IQR)		117.0 (68.0 - 216.0)
Range		13.0 - 349.0
Missing		105
¹ n (%)

10.2 Figure

10.3 DC reasons

10.3.1 Reasons for DC

The denominator here is 31

value	n	pct
Major side effects/complication	12	38.7%
Common side effects	11	35.5%
No response/sub-optimal response to therapy	11	35.5%
Non-compliance	2	6.5%
NULL/Data not available	2	6.5%
Patient Decision (Not related to side effects/financial limitations)	2	6.5%
Deceased	1	3.2%

10.3.2 AEs listed as reason for DC

Here, the denominator is 16: the number of people reporting side effects as a reason for discontinuation (Major side effects/complication or Common side effects)

new	n	pct
Agitation/ irritability	4	25%
Other mood change	4	25%
Aggression	3	18.8%
Increased seizure frequency	3	18.8%
Other	3	18.8%
Somnolence	3	18.8%
Other behavioral change	2	12.5%
Self-injurious behavior	2	12.5%
Sleep disturbance	2	12.5%
Elevated transaminases	1	6.2%
Gastrointestinal upset	1	6.2%
Lethargy	1	6.2%
Musculoskeletal	1	6.2%
Other lab abnormality	1	6.2%
Rash	1	6.2%
Sialorrhea	1	6.2%
Suicidal ideation	1	6.2%

11 Adverse Events During Treatment

Denominator here is the total number of AEs reported.

new	n	pct
Total	149	-
Somnolence	44	29.5%
Gastrointestinal upset	21	14.1%
Lethargy	12	8.1%
Agitation/ irritability	9	6%
Appetite change	8	5.4%
Sleep disturbance	8	5.4%
Gait disturbance	7	4.7%
Other	7	4.7%
Aggression	6	4%
Sialorrhea	5	3.4%
Elevated transaminases	4	2.7%
Other behavioral change	3	2%
Other mood change	3	2%
Depression	2	1.3%
Mental status change	2	1.3%
Musculoskeletal	2	1.3%
Other lab abnormality	2	1.3%
Rash	2	1.3%
Dizziness	1	0.7%
Increased seizure frequency	1	0.7%

12 Adherence

12.1 Summary

Characteristic	N	N = 136¹
Patient PDC	128
Mean (SD)		0.95 (0.11)
Median (IQR)		0.99 (0.95 - 1.00)
Range		0.18 - 1.00
Reason for low adherence	8
Patient was unreachable		1 (12.5%)
Adverse effects which interrupted therapy		2 (25.0%)
Medication provided during inpatient hospitalization		1 (12.5%)
Did not start therapy at time of initiation		1 (12.5%)
Transfer to outside pharmacy prior to returning		1 (12.5%)
Medications weaned without need for further refill		1 (12.5%)
Did not take amount prescribed		1 (12.5%)
Fills	136
Mean (SD)		10.64 (3.86)
Median (IQR)		12.50 (9.75 - 13.00)
Range		1.00 - 15.00
Number of gap days	136
Mean (SD)		14.35 (34.37)
Median (IQR)		1.50 (0.00 - 16.00)
Range		0.00 - 275.00
PDC denominator	136
Mean (SD)		295.67 (107.25)
Median (IQR)		347.00 (305.00 - 356.00)
Range		0.00 - 366.00
¹ n (%)

13 Emergency Healthcare Utilization

13.1 Categorical

Characteristic	N	N = 136¹
Total number of hospitalizations	30
1		19 (63.3%)
2		9 (30.0%)
3		1 (3.33%)
5		1 (3.33%)
Total number of ER or urgent care visits	29
1		20 (69.0%)
2		7 (24.1%)
3		1 (3.45%)
4		1 (3.45%)
¹ n (%)

13.2 Continuous

Characteristic	N	N = 136
Total number of hospitalizations	30
Mean (SD)		1.5 (0.9)
Median (IQR)		1.0 (1.0 - 2.0)
Range		1.0 - 5.0
Total number of ER or urgent care visits	29
Mean (SD)		1.4 (0.7)
Median (IQR)		1.0 (1.0 - 2.0)
Range		1.0 - 4.0

13.3 Time to healthcare utilization

Characteristic	N	Medication shipped:		p-value¹	Overall, N = 136
Characteristic	N	After 3 days, N = 107	Within 3 days, N = 29	p-value¹	Overall, N = 136
time.to.hosp	30			0.5
Mean (SD)		121.0 (102.6)	140.1 (80.9)		125.4 (97.0)
Median (IQR)		113.0 (26.0 - 176.0)	98.0 (92.0 - 168.5)		103.5 (38.2 - 178.5)
Range		5.0 - 318.0	79.0 - 283.0		5.0 - 318.0
time.to.er	29			>0.9
Mean (SD)		109.7 (106.2)	111.2 (90.5)		109.9 (102.1)
Median (IQR)		79.0 (29.8 - 160.0)	69.0 (61.0 - 196.0)		69.0 (31.0 - 196.0)
Range		0.0 - 348.0	12.0 - 218.0		0.0 - 348.0
¹ Wilcoxon rank sum exact test; Wilcoxon rank sum test

14 Change in ASM Use

14.1 Categorical

Characteristic	N	N = 136¹
Number of concurrent medications	136
0		4 (2.94%)
1		20 (14.7%)
2		38 (27.9%)
3		34 (25.0%)
4		33 (24.3%)
5		5 (3.68%)
6		2 (1.47%)
Number of Concurrent AEDs at time of Study Close	105
0		4 (3.81%)
1		19 (18.1%)
2		25 (23.8%)
3		34 (32.4%)
4		17 (16.2%)
5		6 (5.71%)
Change in number of concurrent medications	105
-3		1 (0.95%)
-2		3 (2.86%)
-1		19 (18.1%)
0		73 (69.5%)
1		9 (8.57%)
¹ n (%)

14.2 Continuous

Characteristic	N	Medication shipped:		p-value¹	Overall, N = 136
Characteristic	N	After 3 days, N = 107	Within 3 days, N = 29	p-value¹	Overall, N = 136
Number of concurrent medications	136			0.5
Mean (SD)		2.7 (1.3)	2.8 (1.2)		2.7 (1.3)
Median (IQR)		3.0 (2.0 - 4.0)	3.0 (2.0 - 4.0)		3.0 (2.0 - 4.0)
Range		0.0 - 6.0	0.0 - 5.0		0.0 - 6.0
Number of Concurrent AEDs at time of Study Close	105			0.8
Mean (SD)		2.5 (1.2)	2.6 (1.4)		2.6 (1.2)
Median (IQR)		3.0 (2.0 - 3.0)	3.0 (1.0 - 3.0)		3.0 (2.0 - 3.0)
Range		0.0 - 5.0	0.0 - 5.0		0.0 - 5.0
Change in number of concurrent medications	105			0.8
Mean (SD)		-0.2 (0.6)	-0.3 (0.8)		-0.2 (0.7)
Median (IQR)		0.0 (0.0 - 0.0)	0.0 (0.0 - 0.0)		0.0 (0.0 - 0.0)
Range		-2.0 - 1.0	-3.0 - 1.0		-3.0 - 1.0
¹ Wilcoxon rank sum test

Epidiolex

Josh DeClercq

Leena Choi

Department of Biostatistics

Vanderbilt University Medical Center

2022-10-10