UCLA Stroke Center

Building Clinical Trials that Will Positively Impact an Emerging Field

Jeffrey L. Saver, MD, Professor of Neurology

Director, UCLA Comprehensive Stroke Center

NIH RCTs of First Generation Neurothrombectomy Devices

• IMS 3 (NIH) » Multicenter, phase 3,

900 patients » IV TPA vs IV TPA + IA

(Merci or IA lytic or IA lytic + US), < 3h

• MR RESCUE (NIH) » Multicenter, phase 2,

120 patients » Merci Retriever vs best

medical care, 3-8h

UCLA Stroke Center

Potential Reasons for IMS 3 Nonpositivity

• Concomitant therapy » Low dose of TPA in active arm

• Confounding therapy » IA lytics > mechanical thrombectomy

• Early technology » First generation mechanical thrombectomy

• Patients without target occlusions » Spontaneous lysis, penetrators, M3s, etc » IV TPA worked too well – esp M2 MCAs

• Patients without target penumbra » No penumbral imaging selection

• Patients with high expected response treated outside of trial » Equipoise less in community before IMS 3 results known

• Infarct progression before intervention » Long door to arterial puncture times

UCLA Stroke Center

Potential Reasons for IMS 3 Nonpositivity

• Concomitant therapy » Low dose of TPA in active arm

• Confounding therapy » IA lytics > mechanical thrombectomy

• Early technology » First generation mechanical thrombectomy

• Patients without target occlusions » Spontaneous lysis, penetrators, M3s, etc » IV TPA worked too well – esp M2 MCAs

• Patients without target penumbra » No penumbral imaging selection

• Patients with high expected response treated outside of trial » Equipoise less in community before IMS 3 results known

• Infarct progression before intervention » Long door to arterial puncture times

UCLA Stroke Center

UCLA Stroke Center --Patel + Saver, Submitted

JL Saver, R Jahan, E Levy, T G Jovin, B Baxter, R Nogueira, W Clark, R Budzik, OO Zaidat, for the

SWIFT Trialists

Lancet, Aug 26, 2012

Primary Trial Endpoint Outcomes Among Randomized

Patients

Randomized Solitaire FR

N=58

Randomized Merci N=55

Non-inferiority P value1

Superiority P value1

Successful recanalization without SICH2 (Core Lab)

60.7% (34/56) 24.1% (13/54) <0.0001 0.0001

Successful recanalization study device (Core Lab)

68.5% (37/54) 30.2% (16/53) <0.0001 0.0001

Successful recanalization study device (Site Assessed)

83.3% (45/54) 48.1% (26/54) <0.0001 0.0002

Use of rescue therapy 20.7% (12/58) 43.6% (24/55) <0.0001 0.015

End of procedure successful recanalization (Site)

88.9% (48/54) 67.3% (37/55) <0.0001 0.010

End of procedure successful recanalization (Core Lab)

80.4% (45/56) 57.4% (31/54) <0.0001 0.013

1. Noninferiority by Wald’s method, superiority by Fisher’s Exact test 2. Symptomatic Intracranial Hemorrhage - Any PH1, PH2, RIH, SAH, or IVH associated with a decline in NIHSS ≥ 4 within 24hrs.

Global Disability at 90 Days (Modified Rankin Score)

12.7%

10.4%

12.7%

8.3%

10.9%

10.4%

20.0%

8.3%

21.8%

16.7%

3.6%

2.1%

18.2%

43.8%

SOLITAIREFR

MERCI

0 1 2 3 4 5 6

•CMH, p = 0.04

Hemorrhagic Transformation Outcomes

Outcomes Among Randomized Patients

Randomized Solitaire FR

N=58

Randomized Merci N=55

Non-inferiority P value1

Superiority P value1

SICH 1.7% (1/58) 10.9% (6/55) <0.0001 0.057

All ICH 17.2% (10/58) 38.2% (21/55) 0.0001 0.020

1. Fisher’s Exact

10 |

Rankin Shift

6.30%

15.10%

8.60%

12.00%

6.90%

12.90%

16.10%

9.40%

29.90%

14.10%

8.00%

2.40%

24.10%

34.10%

0.00% 20.00% 40.00% 60.00% 80.00% 100.00%

Merci™ Device

Trevo™ Device 0123456

•*Presented at European Stroke Congress 2102. Lisbon, Portugal by Dr. Raul Nogueira

Complete Recanalization Rates in Multicenter Trials

• IV TPA » Complete – 5%

• IA Lysis » Complete – 20%

• Merci/Multi-Merci » Complete – 23%

• Penumbra » Complete – 23%

Complete Recanalization Rates in Multicenter Trials

• IV TPA » Complete – 5%

• IA Lysis » Complete – 20%

• Merci/Multi-Merci » Complete – 23%

• Penumbra » Complete – 23%

• Solitaire » Complete – 51%

UCLA Stroke Center

UCLA Stroke Center

UCLA Stroke Center

Favorable outcome and mortality at 90d and onset-to-reperfusion time 480 patients from 7 studies

Mazighi M et al. Circulation 2013;127:1980-1985

Copyright © American Heart Association

Favorable Outcome

Mortality

SWIFT-PRIME: Designing for Success

Desirable Feature Design Element Highly effective device SolitaireTM FR device √

Clots to attack CTA/MRA √

Clots distinctively responsive ICA/M1 √

Brain to save Small core (ASPECTS) √

Patients who will respond Randomize all eligible √

Minimize progression Imaging to puncture optimization √

Concomitant therapy Full dose TPA √

The SOLITAIRE™ FR Revascularization Device is intended for use as adjunctive therapy to improve neurological outcomes and decrease mortality in patients experiencing an acute ischemic stroke in large intracranial vessels within 6 hours of symptom onset. The SOLITAIRE™ FR Revascularization Device is also indicated for removing thrombus from a large intracranial vessel and restoring blood flow in patients, who are ineligible for or have failed IV tPA therapy, experiencing ischemic stroke within 8 hours of symptom onset. Solitaire is a trademark of a Covidien company.

Features of Second Generation Embolectomy Trials Current

N Planned Max N

Intervention CTA/ MRA

Time TPA Imaging Status

MR CLEAN

500 500 Variable (97% SR)

+ 6 hr Y or Inel <1/3 MCA

Positive

ESCAPE 316 500 Solitaire + 12 hr Y or Inel Collat < 50%

Positive

EXTEND IA

~70 100 Solitaire + 6 hr Y RAPID Mismatch

Positive

REVASCAT ~200 690 Solitaire + 8 hr Inel or Failed

A ≥ 6/7 Pending

SWIFT PRIME

195 833 Solitaire + 6 hr Y A ≥ 6 RAPID

Pending

THERAPY 109 692 Penumbra HVS≥8mm Y < 1/3 MCA

Halt

PISTE ~40 800 Variable + 6 hr Y CT hypo ?

THRACE ~450 480 Variable + R 4h Y ?

UCLA Stroke Center

How Many of the Dominoes Will Fall?

UCLA Stroke Center

How Many of the Dominoes Will Fall?

UCLA Stroke Center

MR CLEAN

How Many of the Dominoes Will Fall?

UCLA Stroke Center

MR CLEAN

ESCAPE

EXTEND IA

How Many of the Dominoes Will Fall?

UCLA Stroke Center

MR CLEAN

REVASCAT

SWIFT PRIME

ESCAPE

EXTEND IA

How Many of the Dominoes Will Fall?

UCLA Stroke Center

MR CLEAN

REVASCAT

THRACE

SWIFT PRIME

THERAPY

ESCAPE

EXTEND IA

MR CLEAN Design

• Netherlands Hospitals • Medical therapy (w/o or w/ TPA) vs medical

therapy (w/o or w/ TPA) plus endovascular • Key entry criteria

» NIHSS ≥ 2 » Age ≥ 18 » CTA/MRA occlusion in ICA, MCA (M1/M2), or ACA

(A1/A2) » Embolectomy within 6hr » < 1/3 MCA or 100 cc

UCLA Stroke Center

MR CLEAN: Primary Outcome

•UCLA Stroke Center

Common adjusted odds ratio: 1.67 (95CI 1.21 – 2.30)

MR CLEAN: Primary Outcome

•UCLA Stroke Center

Common adjusted odds ratio: 1.67 (95CI 1.21 – 2.30)

Are We Done Yet?

•UCLA Stroke Center

Are We Done Yet?

•UCLA Stroke Center

Are We Done Yet?

•UCLA Stroke Center

Unable to return to work (mRS 2-6): 88% Permanent symptoms (mRS 1-6): 97%

Building Next Generation of Clinical Trials that Will Positively Impact an Emerging Field

Intervention Type

Special Trial Aspects

Example Comparisons

Target Patients

Reperfusion Strategies

Active Comparator

IVT+ERT vs ERTalone ICA occlusions IVT+IERT vs IVT alone M2 occlusions

Systems of Care Cluster randomization

EMS routing – PSCs first versus CSCs first

Severe deficits

Prehospital Neuroprotection

ED imaging endpoints

NA1, hypothermia, nitroglycerin vs control

EMS transported patients

Deter Reperfusion Injury

IA admin Free radical scavengers vs control

Post-successful TICI 2b/3 reperfusion

Neurothrombectomy Devices

Reperfusion 2b/3 as primary surrogate endpoint

Device A vs B Large artery occlusions

Imaging Selection

6-24h ERT vs no ERT Wake-up and late

UCLA Stroke Center

Symptoms Primary Stroke Center Neuroprotectants EMS 911

Comp Stroke Center EMS IV Lytic

Imaging

Imaging

IA Mechanical or Lytic

Angiogram

Cath Lab Reperfusion

Neuroprotectants Stroke Unit

Acute Ischemic Stroke Treatment 2.0: Fast and Furious