Warfarin as a new oral anticoagulant Improving stability and outcome - - PowerPoint PPT Presentation

Warfarin as a new oral anticoagulant

Improving stability and outcome of warfarin by monitoring factors II and X only Results of the Fiix-trial Páll T. Önundarson, M.D. Landspitali - The National University Hospital of Iceland and University of Iceland Faculty of Medicine

Coauthors

How to read data from randomized trials?

• Who is the real

messenger?

• Inclusion/exclusion

criteria

• Control groups
• Absolute rate vs

proportional change

• Hazard ratios

COI disclosure

• Pall Onundarson and Brynja Gudmundsdottir:

– Hold a patent or the Fiix-prothrombin time

A reminder:

Warfarin is very efficacious!

NON-VALVULAR ATRIAL FIBRILLATION

• Annual cerebral infarction rate

reduced by 79%

• No anticoag (placebo) 4.3%
• Warfarin 0.9%

SUPERIORITY VS. NON-INFERIORITY

ANNUAL CEREBRAL INFARCTION RATE IN ATRIAL FIBRILLATION 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

CTE annual rate

Placebo Warfarin 79% improvement over placebo 10 percent improvement over warfarin 20 percent improvement over warfarin 40% improvement

• ver warfarin

Ezekowitz, MD et al. NEJM 1992;327:1406-

The efficacy of VKA depends strongly on the quality of warfarin management

Efficacy and safety of dabigatran compared with warfarin at different levels of international normalised ratio control for stroke prevention in atrial fibrillation: an analysis of the RE-LY trial

Lars Wallentin, et al and on behalf of the RE-LY investigators, The Lancet 2010; 376: 975-983

Stroke and systemic embolism incidence

in each quartile of centre's mean time in therapeutic range (cTTR):

Major bleeding incidence

in each quartile of centre's mean time in therapeutic range (cTTR):

6

1990-2014: Marked improvement in warfarin control in the Western world

Onundarson PT, Einarsdottir KA, Gudmundsdottir BR. Warfarin anticoagulation in Reykjavik in 1992 vs 2006.

Int J Lab Haematol. 2008;30(5):382-9.

• International standardization of PT

reporting (INR)

• Anticoagulation Management Centers

– Specialized staff based dosing instead of private physician

• Computer software assisted dosing

– Poller et al Lancet 1998

• Self monitoring
• Self dosing

Warfarin/coumarins

a love-hate relationship

• Love

– Well studied and efficacious – Controllable anticoagulation intensity by monitoring

• Therapeutic window is well

delineated

– adjustable to personal needs (titratability)

– Compliance is measurable – Immediately reversible

• Immediately with PCC
• (Hours FFP)
• 12-16 hours vitamin K

– Cheap

• Hate

– Slow onset of effect – Hard to predict initial dose

– Mutations affect metabolism and dose size » VKORC » CYP450

– PT-INR fluctuates

in many patients leading to need for frequent testing and dose adjustments

– Serious bleeding complications – Needlestick – Work!

Can warfarin treatment be improved in 2015?

Even if warfarin is very efficacious… …managing warfarin is trouble

• Dose varies between patients

– metabolism

• Fluctuating unstable effect (INR)

– frequent dose adjustments – frequent testing

• Annual cerebral infarction rate in AF

reduced 79% compared to placebo

Ezekowitz, MD et al. NEJM 1992;327:1406-

Why does the effect of warfarin fluctuate?

Common wisdom; Drug problem

• Food interactions
• Drug interactions
• Compliance problems

Test problem: Prothrombin time?

Rephrase: Why does the INR fluctuate?

Influence of VKA on VKD factors

Activity reduced at different rate

Are the VKD factors the same?

• Antithrombotic effect?
• How accurately does the PT-

INR reflect the antithrombotic effect?

• How does the PT reflect risk
• f bleeding?

Factor T ½ (hours) II 72 VII 4 IX 18-24 X 24-48

0.1 1 10 100 20 40 60 80

FII FVII FIX FX

A

Coagulation factor activity (%) Quick Prothrombin time (seconds)

Traditional prothrombin time

Undiluted thromboplastin and CaCl2

Gudmundsdottir BR, Francis CW, Bjornsdottir A, Nellbring M, Onundarson PT. Thromb Res 2012;130:674-81

Correlation of PT/P&P with ROTEM clotting parameters in a dilute thromboplastin experiment

1 10 100 1000 1000 2000 3000

R2 = 0.27

A

Owren prothrombin time % (PP%) ROTEM CT (seconds)

1 10 100 1000 5 10 15 20 25

R2 = 0.08

B

Owren prothrombin time % (PP %) MaxVel (mm x 100/second)

1 10 100 1000 20 40 60 80

R2 = 0.11

C Owren prothrombin time % (PP %)

MCF (mm x 100) Gudmundsdottir BR, Francis CW, Bjornsdottir A, Nellbring M, Onundarson PT. Thromb Res 2012;130:674-81

ROTEM clotting time (CT), Maximum velocity (MaxVel) and maximum clot firmness (MCF) in relation to progressive reduction of individual vitamin K dependent coagulation factors in PPP

Undiluted thromboplastin Highly diluted thromboplastin

A

Coagulation factor activity (%) ROTEM MaxVel (mm x 100/second)

Gudmundsdottir BR, Francis CW and Onundarson PT, Thromb Res 2012

0.1 1 10 100 20 40 60 80

FII FVII FIX FX

A

Coagulation factor activity (%) Quick Prothrombin time (seconds)

0.1 1 10 100 2 000 4 000 6 000

FII FX FVII FIX

A

Coagulation factor activity (%) ROTEM CT (seconds)

Traditional prothrombin time

The PT-INR is not a great indicator

• f thrombin generation or of the anti-thrombotic effect of VKAs

The PT-INR associates poorly with:

IIa generation

– KE Brummel, SG Paradis, RF Branda, KG Mann Circulation 2001;104:2311-17

ROTEM clotting parameters in a dilute thromboplastin experiment

– Gudmundsdottir BR, Francis CW, Bjornsdottir A, Nellbring M, Onundarson PT. Thromb Res 2012;130:674-81

The anti-thrombotic effect depends mainly

• n reductions in FII and FX – but not FVII
• Thrombin generation correlates with FII and less

with FX but poorly with factors VII and IX activity

– Xi M, Béguin S, Hemker HC. Thromb Haemost. 1989 Sep 29;62:788-91

• Induced DIC in rabbits is prevented by reduced FII

and FX but not by reduced FVII

– A Zivelin, LV Rao and SI Rapaport. J. Clin. Invest. 92: 2131-2140 (1993).

• Monitoring warfarin with NPA

– Monitoring native FII lead to 85% reduction in major events compared to PT monitoring – Furie B, Diuguid CF, Jacobs M, Diuguid DL, Furie BC. Blood. 1990 ;75:344-9

• ROTEM experiments with low TF concentration

– Gudmundsdottir BR, Francis CW, Bjornsdottir A, Nellbring M, Onundarson

• PT. Thromb Res 2012;130:674–81.

R2 = 0.11

C Owren prothrombin time % (PP %)

MCF (mm x 100)

R2 = 0.08

B

Owren prothrombin time % (PP %) MaxVel (mm x 100/second)

R2 = 0.27

A

Owren prothrombin time % (PP%) ROTEM CT (seconds)

A

Hypothesis:

During VKA anticoagulation measuring the combined influence of only the stable FII and FX reflects clottability better than does the PT (INR) which is affected also by the short half-life FVII that has little influence on thrombus prevention but confounds VKA dosing

Fiix-PT:

Fiixing the prothrombin time

• Factor II and X (“F ii x”) deficient plasma added to pre-diluted patient sample

– corrected for all factor deficiencies other than II and X in test plasma

• Coagulation activated by undiluted thromboplastin and CaCl2 (i.e. PT reagent) and

clotting time measured

• i.e. the clotting rate is in relation to deficiency of ii and x only (Fiix-PT)

Gudmundsdottir BR, Francis CW, Bjornsdottir A, Nellbring M, Onundarson PT. Thromb Res 2012;130:674–81.

Sequential VKD- factor measurements in relation to the INR in two individuals

The short T ½ of FVII explains the difference in INR and Fiix-INR

Gudmundsdottir BR, Francis CW, Bjornsdottir A, Nellbring Moa, Onundarson PT. Critical role of factors II and X during coumarin anticoagulation and their combined measurement with a new Fiix-prothrombin time. Thromb Res 2012;130(4):674–81.

14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 20 40 60 80 100 1 2 3 4 5 6 7 8 9 10

INR FII FX FVII FiixINR

Days from warfarin initiation Coagulation factor activity percentage INR value

TESTING THE HYPOTHESIS

A DOUBLE BLIND RANDOMIZED CLINICAL TRIAL

The Fiix trial

RANNIS

(Icelandic Research Council Expert Panel Evaluation)

• This referee does not advise to fund this trial.
• If this idea ever occurred to a clotting specialist s(he) must have rejected

it before making it public because of doubts on the scientific value”

• The applicants have insufficiently considered what the adverse effects of

their approach could be for the study population.

• Moreover the theoretical and experimental background is shaky.
• … it is an ostrich’s policy to think that phenomena that remain

unobserved could not hurt the patient.

Grant application

Testing the hypothesis: The Fiix-trial

Design

• Single Center Prospective RCT

– investigator initiated – randomized double blind study – non-inferiority trial 2012-2014 – Landspitali AMS in Reykjavik

• All patients on warfarin with INR target 2-

3 > 18 yo invited to participate

–

• nly exclusions: nursing home patients,

cardioversion patients

• Randomized to monitoring with:

– Fiix-PT/INR (“Fiix-warfarin”)

• active arm

– PT/INR (“PT-warfarin”)

• control arm
• Warfarin management

– specialized nursing staff – DAWN anticoagulation software – dosing protocol designed for the PT-INR

• with maximum recommended interval 42 days

Endpoints

• Efficacy

– Total thromboembolism

• Ischemic stroke (not hemorrhagic stroke)
• TIA
• Systemic arterial embolism
• Myocardial infarction
• VTE
• Safety

– Major bleeding

– ISTH criteria

– Other clinically relevant bleeding – Non-vascular death

• Composite major vascular events
• Surrogate efficacy/convenience

parameters

– Number of tests and tests in range – Dose change frequency – TTR (Time within target range) – VGR (variance growth rate; INR fluctuation)

Treatment description:

• After exclusions, 1148 patients were
• bserved in two arms
• From March 1st 2012 to February 28 2014
• Median ITM observation 1.7 years (20

months) per patient

– 93% of observation time from individuals observed for over 1 year – None lost to follow up

Fiix-trial: Patient Characteristics

* IQR denotes interquartile (25-75%) range. There were no statistically significant differences between the two study arms in any of the listed variables.

Warfarin treatment indices

*Per protocol (on-treatment) analysis of the monitoring method **Data is shown as median (interquartile range) unless otherwise noted. ***Note: TTR denotes time within target range by Rosendaal

• method. Only patients that have 3 or more tests were included in

the TTR interval calculation and warfarin initiation periods were excluded in TTR analysis until two INRs fell within target range. ****VGR: Variance growth rate by “B1 method”, i.e. variance in INR compared to test before, size of “jumps” between tests.

Stability was improved overall

Efficacy and safety

• f Fiix-PT

Non-inferiority analysis days 1-720

Composite major vascular events

d1-720

5.0% ppy

d1-720

3.5% ppy

d181-720

5.3%

d181-720

3.3%*

Total thromboembolism (all patients)

(Stroke and SE + MI + TIA + VTE)

Intention to monitor analysis

d1-720

2.3% ppy

d181-720

1.2% ppy

d181-720

2.2%

d181-720

1.1%*

d181-720

2.2% ppy

d181-720

2.5% ppy

Major bleeding (all patients)

Per protocol analysis

Total arterial thromboembolism in atrial fibrillation patients

(Stroke and SE + MI + TIA + VTE)

Intention to monitor subgroup analysis

1.6% 2.7% 1.4%(*) 3.2%

More coming Fiix´s

TTR

in relation to vascular event occurence and warfarin monitoring method

No event Any CRVE Clinically relevant bleeding Major bleeding TE Fiix warfarin 82 79 79 79 80 PT warfarin 80 75 76 76 62

10 20 30 40 50 60 70 80 90

Time in range – TTR (%) *

Oskarsdottir AR et al: Thromboembolism and clinically relevant bleeding in relation to warfarin anticoagulation variability in patients monitored with Fiix-prothrombin time or Quick-prothrombin time. (2015, in manuscript)

Anticoagulation variation

in relation to vascular event occurence and warfarin monitoring method

No event Any CRVE Clin relevant bleeding Major bleeding TE Fiix warfarin 0.17 0.23 0.24 0.31 0.2 PT warfarin 0.21 0.35 0.31 0.59 0.5

0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65

Variance growth rate (B1) * ***

Oskarsdottir AR et al: Thromboembolism and clinically relevant bleeding in relation to warfarin anticoagulation variability in patients monitored with Fiix-prothrombin time or Quick-prothrombin time. (2015, in manuscript)

Warfarin dose variation

and major vascular events

No event Any CRVE Fiix warfarin 0.03 0.18 PT warfarin 0.03 0.12

0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2

Variance growth rate (B1)

Oskarsdottir AR et al: Thromboembolism and clinically relevant bleeding in relation to warfarin anticoagulation variability in patients monitored with Fiix-prothrombin time or Quick-prothrombin time. (2015, in manuscript)

Automated thrombin generation

ATG and VKD factor activity

ATG vs Fiix-INR or PT-INR during warfarin initiation

Fiix group, Fiix-INR vs. ETP

2 5 8 11 14 17 20 23 26 200 400 600 800 1000 1200 1400 1600 1800 2000 1 2 3 4 5

Days of treatment

Fiix group ETP Fiix group R-INR

nM x min R-INR

(Petur I. Jonsson1, Brynja R. Gudmundsdottir1, Charles W. Francis2, and Pall T. Onundarson1,3. The Fiix-prothrombin time mirrors thrombin generation during warfarin treatment. 2015, in manuscript 2015)

Dilute Fiix-PT

to measure all oral anticoagulants

Increasing thromboplastin dilutions, spiking experiments

Patient samples

Loic R. Letertre*, Brynja R. GudmundsdoTr*, Charles W. Francis†, Robert C. Gosselin‡, Mika Skeppholm§, Rickard Malmstrom¶, Stephan Moll**, Emily Hawes**,††, Suzanne Francart††, and Pall T. Onundarson*,‡‡A single test to assay warfarin, dabigatran, rivaroxaban, apixaban, unfractionated heparin and enoxaparin in plasma. Submitted, 2015

dPT A)

dFiix-PT B)

dPT C)

dFiix-PT D)

dPT E)

dFiix-PT F)

• L

• L

• w

• m

• 40
• 20

Ischemic stroke or systemic embolism

• L

• L

• w

• m

• L

• L

• w

• m

• 40
• 20

Ischemic stroke or systemic embolism or MI

• L

• L

• w

• m

Meta-analysis

NOACs including Fiix-warfarin vs PT-warfarin

Remember: TTR in warfarin controls in DOAC trials 58-66% and in TTR in Fiix-trial 80% In manuscript 2015

A reminder:

PT-warfarin is very efficacious, Fiix-warfarin more so!

NON-VALVULAR ATRIAL FIBRILLATION

• Annual cerebral infarction rate

reduced by 79%

• No anticoag (placebo) 4.3%
• Warfarin 0.9%

SUPERIORITY VS. NON-INFERIORITY

ANNUAL CEREBRAL INFARCTION RATE (%) IN ATRIAL FIBRILLATION

Ezekowitz, MD et al. NEJM 1992;327:1406-

0.86 0.66 0.77 0.78 0.80 0.56

1

PT-warfarin 79% improvement over placebo Dabigatran 150 mg Rivaroxaban Apixaban Edoxaban 60 mg Fiix-warfarin

CONCLUSIONS

• f the Fiix-trial:

COMPARED TO HIGH QUALITY (TTR 80%) PT-WARFARIN, FIIX-WARFARIN:

• IS MORE STABLE
• LOWER VGR

– High VGR associated with adverse events

• HIGHER TTR
• FEWER DOSE CHANGES
• IS CLINICALLY AT LEAST NON-INFERIOR (primary analysis)
• IS CLINICALLY SUPERIOR IN THE LONG-TERM (secondary analysis)
• long-term reduction in TE
• DID NOT INCREASE MAJOR BLEEDING
• despite lowering the long-term thromboembolic rate
• despite not being affected by FVII in the test sample.
• despite low bleeding incidence in control group

OVERALL CONCLUSION:

– A fluctuating PT-INR during warfarin treatment is partly a confounding side effect of the PT itself. – The data suggests, that if the PT is replaced with a monitoring test that is not affected by FVII such as the Fiix-PT, warfarin becomes more stable than previously assumed; a new oral anticoagulant!