Sarepta Therapeutics : EP.149 Iff et al. Delay in DMD prog with eteplirsen longer time to LOA v SOC

EP.149 Presented at the 2021 World Muscle Society Virtual Congress, September 20−24, 2021

Delay in Duchenne Muscular Dystrophy Progression With Eteplirsen: Longer Time to Loss of Ambulation Versus Standard of Care

Joel Iff,1 George Bungey,2 Abby Paine,2 Baoguang Han,1 Heather Gordish-Dressman,3 Erik Henricson,4 Craig McDonald,4 and the Eteplirsen and CINRG Duchenne Natural History Study Investigators

1Sarepta Therapeutics Inc, Cambridge, MA, USA; 2DRG Abacus, Part of Clarivate, London, UK; 3Children's National Hospital, George Washington University School of Medicine and Health Sciences, Washington DC, USA; 4University of California, Davis, CA, USA

Please scan QR code for full study details

Objective

To estimate the treatment

benefit of eteplirsen vs standard of

care (SOC) for time to loss of ambulation (LOA) in patients with Duchenne muscular dystrophy (DMD) using a post hoc analysis of individual patient-level data

Key Takeaway

LOA is significantly delayed in

CONCLUSIONS

• Eteplirsen treatment was associated with increased median age at loss of ambulation by 2.7 years (15.7 vs 13.0 years for eteplirsen vs SOC)
• Eteplirsen treatment was associated with a statistically significant 47% risk-reduction of LOA vs SOC across the lifespan
• Median age at LOA in the SOC group was similar to the 13.40 years identified in a broader population of patients from CINRG1
• Results are robust to the inclusion of all genotyped CINRG patients who were ambulatory at baseline in SOC group

patients treated with

eteplirsen vs SOC

METHODS

Data sources

RESULTS

Baseline characteristics at study entry: Base case analysis

• Eteplirsen patients were significantly older than SOC at the start and end of study
• Overall, corticosteroid use was similar between the treatment groups

Age at LOA: Base case analysis
	Kaplan-Meier of age at LOA	Median age at LOA and Cox model results
		Median age
	100%		Cox		P
		at LOA
	90%		model	95% CI
		(K-M estimate),	value
	80%	Treatment	HR
survival	years (95% CI)
70%	Eteplirsen	15.7 (12.7-NE)
60%	0.53	0.30-0.93	0.027
SOC	13.0 (12.1-15)
50%
	40%	CI=confidence interval; HR=hazard ratio; K-M=Kaplan-Meier; NE=not evaluable

• Eteplirsen clinical trials: Studies

201/202/405, Study 203, Study 204,

and Study 301/PROMOVI

• SOC: Placebo arm of the DEMAND III

trial, Leuven Neuromuscular Reference

Center Registry, Telethon Italian DMD

Registry, and the Cooperative

International Neuromuscular Research

Group Duchenne Natural History

(CINRG DNH)

Inclusion criteria

Characteristic	Eteplirsen
(n=118)
Race, n (%)
White	100 (84.7)
Black or African	3 (2.5)
American
Pacific Islander	2 (1.7)
Asian	9 (7.6)
Other	4 (3.4)

SOC/Placebo

(n=113)

1. (61.1)

1. (0.9)

0

1. (14.2)

1. (2.7)

Cumulative	30%	SOC (27 events)	Eteplirsen treatment was associated with a statistically
20%
10%	Eteplirsen (23 events)	significant 47% risk reduction of LOA vs SOC across the
		lifespan, translating to ~21% longer in ambulation
	0%

0	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	• Time to LOA was significantly longer in the eteplirsen
Patients at risk over time:					Age (years)										treatment group
																• Tests of proportion hazards assumption suggest
SOC 113	113	113	113	113	112	110	96	79	63	48	35	21	11	6	5	3	1	1	0
Eteplirsen 118	118	118	118	118	118	116	109	98	88	65	48	34	21	16	10	5	3	2	0	assumption is valid (e.g., P=0.86 for Schoenfeld residual)
• Maximum study follow-up times were 8.1 years for eteplirsen and 8.9 years for SOC			• Median age at LOA in SOC group was similar to the
		13.40 years identified in a broader population of
• Median study follow-up times were 1.8 years for eteplirsen and 1.1 years for SOC				patients from CINRG1

• Amenable to exon 51 skipping
• Receiving treatment with eteplirsen or SOC/placebo
• Receiving steroids for ≥30% of the study
• Ambulatory at baseline visit

Analyses

• Base case analysis: SOC group includes only exon 51 skip-amenable patients
• Sensitivity analysis: SOC group includes all genotyped CINRG patients who were ambulatory at baseline, excluding skip exon-44 and del_3-7

Definition of LOA

• LOA was defined according to a

combination of 10-meter walk/run time

≥30 s and 6-minute walk distance = 0 m,

Unknown	0
Baseline age, years	8.68 (2.42)
Age at last study visit,	10.73 (2.74)
years

Total time on treatment	748 (440)
during study, days

Corticosteroid regimen, n (%)

Prednisone or	53 (44.9)
prednisolone (daily)
Deflazacort (daily)	28 (23.7)
Prednisone or
prednisolone	13 (11.0)

(intermittent)
All others (including	24 (20.3)
unknown)
Treatment exposure,	241.8
patient years

24 (21.2)a

7.85 (2.30)

9.73 (2.57)

687 (589)

1. (15.9)

1. (41.6)

17 (15.0)

31 (27.4)

0

Age at LOA: Sensitivity analysis

Kaplan-Meier of age at LOA

event	100%
90%
80%
of	70%
60%
free
50%
Proportion	40%			Eteplirsen (23 events)
10%
	30%			SOC (91 events)
	20%
	0%
	0	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20	21	22

Age (years)

Patients at risk over time:

SOC 278 278 278 278 278 277 268 245 208 171 138 103

79

59

41

29

16

10

7

4

1

1

0

Median age at LOA and Cox model results

	Median age	Cox		P
			at LOA
			model	95% CI
Treatment	(K	-	M estimate),	value
years (95% CI)	HR
Eteplirsen	15.7 (12.7-NE)	0.62	0.39-0.99	0.045
SOC	14.1 (13.0-15.0)

CI=confidence interval; HR=hazard ratio; K-M=Kaplan-Meier; NE=not evaluable

• Time to LOA was significantly longer in the eteplirsen

treatment group

• Tests of proportion hazards assumption suggest

assumption is valid (e.g., P=0.66 for Schoenfeld residual)

• Median age at LOA in SOC group was similar to the 13.40

years identified in a broader population of patients from

CINRG1

or inability to complete the tests

Please scan QR code for additional study details

All values are mean (SD) unless otherwise noted. aRace data were not available for any patients in the Leuven and Telethon SOC studies.

Eteplirsen 118 118

118

118

118

118

116

109

98

88

65

48

34

21

16

10

5

3

2

0

0

0

0

REFERENCES

1. McDonald C, et al. Lancet. 2018; 391:451-461.
2. Mendell JR, et al. Ann Neurol. 2016;79(2):257-271.
3. Mendell JR, et al. J Neuromuscular Dis. 2021;doi:10.3233/JND-200548.

ACKNOWLEDGMENTS & DISCLOSURES

The authors and Sarepta Therapeutics, Inc., thank the patients and their families for their participation in the studies. This study was funded by Sarepta Therapeutics, Inc. Editorial support was provided by Kristin M. Allan, PhD, and was funded by Sarepta Therapeutics, Inc. Disclosures: JI and BH are employees of Sarepta Therapeutics, Inc. and may own stock/options in the company. GB was an employee of DRG Abacus at the time of the study. AP is an employee of Zedediah Consulting and partner of DRG Abacus. HG-D is the co-founder of TRiNDS, LLC. EH reports consulting fees (Sarepta Therapeutics, Inc.). CM reports consulting (Astellas/Mitobridge, Bristol Myers Squibb, Capricor, Catabasis Pharmaceuticals, Edgewise Therapeutics, Eli Lilly, Epirium Bio [formerly Cardero Therapeutics], Gilead, Halo Therapeutics, Italfarmaco, Novartis, Pfizer, Prosensa, PTC Pharmaceuticals, Santhera Pharmaceuticals, and Sarepta Therapeutics, Inc.), research funding, principal investigator, and speaking fees (Sarepta Therapeutics, Inc.).

Presented previously at the 2021 Muscular Dystrophy Association Virtual Clinical & Scientific Conference, March 15−18, 2021.

BACKGROUND

• Eteplirsen is indicated to treat Duchenne muscular dystrophy (DMD) in patients with genetic mutations amenable to exon 51 skipping
• Previous analyses have shown that eteplirsen is associated with significant and clinically meaningful delays in time to loss of ambulation (LOA)2,3
• Additional data now allow for a more comprehensive analysis of a larger number of eteplirsen-treated patients

PATIENT SELECTION

DEFINITION OF LOA

• Loss of ambulation (LOA) was defined according to a combination of 10-meter walk/run time ≥30 s and 6-minute walk distance = 0 m (or inability to complete the tests)
• • For patients with both outcomes available, both outcomes had to be satisfied to indicate LOA
  • In the eteplirsen trial datasets, a rate-limiting cell value of 30 s was recorded by clinicians if the patient failed the test
  • In the CINRG dataset LOA was confirmed by ensuring the variable measuring velocity to complete 10m walk run = 0 m/s
• Time to wheelchair use was used for Study 405, as 10-meter walk/run time was not available
• • Time to LOA based on this definition aligned with time to LOA based on 10-meter walk/run time for the 2 patients in Study 405 who had lost ambulation during the 201/202 study
• Outcomes were checked at prior and subsequent visits to LOA event to prevent confounding of missing data/fractures

STATISTICAL ANALYSIS

• Kaplan-Meiercurves were constructed from the patient data sets to provide a visual representation of the proportion of patients who experienced LOA or were censored over time (i.e., did not experience an event before the end of the study, were lost to follow-up, or withdrew)
• A Cox proportional hazards model was used to calculate a hazard ratio to compare the difference in treatment effect between eteplirsen and SOC over time

Eteplirsen-treated patients			Study 301/	SOC - Base case analysis				SOC - Sensitivity analysis
Study 201/202/405	Study 203	Study 204	BIOMARIN/DEMAND III	Leuven	Telethon
PROMOVI		BIOMARIN/DEMAND III
(NCT01396239/	(DMD114044/	NMRC	Italian DMD
NCT01540409)	(NCT02420379)	(NCT02286947)	(NCT02255552)	registry	registry	CINRG DNH	(DMD114044/
			NCT01254019)	NCT01254019)
Study total
n=12	n=33	n=24	n=109	Study total	n=186	n=87	n=97	n=440	Study total	n=186
		n=7		n=30		n=2	n=71	n=85	n=381
					n=2
Amenable to exon	n=12	n=26	n=24	n=79	Amenable to exon 51	n=184	n=16	n=12	n=59		n=184
51 skipping
				skipping
					n=124					n=124
Receiving eteplirsen	n=12	n=26	n=24	n=79	Receiving	n=60	n=16	n=12	n=59	Receiving	n=60
		n=2	n=1		SOC/placebo			n=1	n=16	SOC/placebo
Receiving corticosteroids	n=12	n=24	n=23	n=79	Receiving	n=60	n=16	n=11	n=43	Receiving
corticosteroids at	corticosteroids at	n=60
at study visit
				study visit		n=3	n=2	n=12	study visit
			n=20
Ambulatorya at baseline					Ambulatorya at baseline	n=60	n=13	n=9	n=31		n=60
with ≥1 follow-up visit with	n=12	n=24	n=3	n=79	with ≥1 follow-up visit with
relevant outcome data					relevant outcome data
		Total N=118				Total N=113

aAmbulatory defined as 10-meterwalk-run time <30 secs / 6-minute walk distance > 0. CINRG DNH=The Cooperative International Neuromuscular Research Group Duchenne Natural History; NMRC=Neuromuscular Reference Center.

Leuven	Telethon
NMRC	Italian DMD	CINRG DNH
registry	registry
n=87	n=97	Study total	n=440
n=71	n=85		n=25
		Excluding patients
n=16	n=12	with known exon 44	n=415
		mutations or del_3-7a	n=195
n=16	n=12	Receiving SOC (corticosteroids	n=220
at study visit), with baseline
	n=1	and follow-up data	n=24
n=16	n=11	Include only	n=196
genotyped patients
n=3	n=2
n=13	n=9

Total N=278

Presented at the 2021 World Muscle Society Virtual Congress, September 20−24, 2021