Copeptin and high sensitive Troponins Discussion of NEJM publications on sensitive Troponins BRAHMS GmbH, August 2010.

Copeptin and high sensitive Troponins

Discussion of NEJM publications on sensitive Troponins

BRAHMS GmbH, August 2010

• What Copeptin can do• Translating results into work-up changes (simplified)

• What sensitive Troponin assays can do• Translating results into work up changes (simplified)

• The issues of current trial results evaluating sensitive Troponin (New England Journal, 2009)

Outline

2

Simplified patient work up in the ED

3

Only a small proportion of chest pain patients are diagnosed with AMI

Source: crude average from Reichlin et al./ Keller et al., NEJM 2009

STEMI 10%

NSTEMI 10%

4

5Source: Reichlin et al. J Am Coll Cardiol 2009;54:60-8

Incremental value of Copeptin for rapid rule out of acute myocardial infarction

Combination of Copeptin / Troponin for early rule out of AMI

Copeptin cut off: 14pmol/L

Troponin cut off

NPV (negative predictive value) = 99.4%!

6Source: Reichlin et al. J Am Coll Cardiol 2009;54:60-8

Patient work up with insensitive Troponin -1

100 patients Tn < cut off Tn > cut off sum

non-AMI 80

NSTEMI 10

STEMI 10

sum 100

7Source: Reichlin et al. J Am Coll Cardiol 2009;54:60-8

Patient work up with current Troponin -2

100 patients Tn < cut off Tn > cut off sum

non-AMI 78 2 80

NSTEMI 10

STEMI 10

sum 100

8Source: Reichlin et al. J Am Coll Cardiol 2009;54:60-8

100 patients Tn < cut off Tn > cut off sum

non-AMI 78 2 80

NSTEMI 3 7 10

STEMI 10

sum 100

Patient work up with current Troponin -3

9Source: Reichlin et al. J Am Coll Cardiol 2009;54:60-8

100 patients

Tn < cut off

Tn > cut off

sum

non-AMI 78 2 80

NSTEMI 3 7 10

STEMI 10 10

sum 81 19 100

19 patients Tn positive: Highly suspicious for AMI, start treatment or confirmatory diagnosis 17 with AMI

81 patients Tn negative: Unclear on admission if NSTEMI to be re-tested

Tn re-test <cut off no AMI 78

Tn re-test >cut off NSTEMI 3

Summary:•81 need to be re-tested!!•after re-testing: 22 diagnosed with AMI, 2 false positives

Patient work up with insensitive Troponin -4

10Source: Reichlin et al. J Am Coll Cardiol 2009;54:60-8

11

100 patients

Tn < cut off, Copeptin <

cut off

Tn < cut off, Copeptin >

cut off

Tn > cut off sum

non-AMI 2 80

NSTEMI 7 10

STEMI 10 10

sum 19 100

Patient work up combining Troponin and Copeptin -1

Source: Reichlin et al. J Am Coll Cardiol 2009;54:60-8

12

100 patients

Tn < cut off, Copeptin <

cut off

Tn < cut off, Copeptin >

cut off

Tn > cut off sum

non-AMI 60 18 2 80

NSTEMI 0 3 7 10

STEMI 10 10

sum 60 21 19 100

60 patients Tn & Copeptin neg.: very unlikely to have AMI NPV 99.4% rule out

21 patients Tn neg, Copeptin pos.: unclear on admission if NSTEMI to be re-tested

Tn re-test <cut off no AMI 18

Tn re-test >cut off NSTEMI 3

Copeptin in combination with Troponin reduces re-testing by 2/3:•60 immediately ruled out, accordingly time-to-intervention reduced

At the core: % NSTEMITn works on AMICopeptin on non-AMI

Patient work up combining Troponin and Copeptin -2

Source: Reichlin et al. J Am Coll Cardiol 2009;54:60-8

13

100 patients Troponin Troponin + Copeptin

Troponin re-tests 81 21

Copeptin tests 0 100

non-AMI declared AMI(false positives)

2 2

Temporary admittance (6h) for re-testing

81 21

missed AMI (false negatives )

0 0

Comparing work-up of Copeptin + Troponin with Troponin alone

14

Combining Troponin and Copeptin has potential to reduce costs significantly

Break even point: temporary admittance = Copeptin

100 patients Troponin Troponin + Copeptin

Assumed costs per patient

Troponin re-tests 81 21 $ 20

Copeptin tests 0 100 $ 20

Non-AMI declared AMI(false positives)

2 2 $ 1000

Temporary admittance (6h) for re-testing

81 21 $ 100

Missed AMI (false negatives )

0 0 $ 2000

Potential savings per patient:

-- $ 53

Costs are based on assumptions and serve as an example only

15

What sensitive Troponin assays can do

Which cut off to use?99th percentile: 95% sensitivity, 80% specificity

99th percentile

Source: Reichlin et al. J Am Coll Cardiol 2009;54:60-8

16

100 patients

Tn < cut off

Tn > cut off

sum

non-AMI 64 16 80

NSTEMI 1 9 10

STEMI 10 10

sum 65 35 100

35 patients Tn positive: highly suspicious for AMI, start treatment or confirmatory diagnosis 19 with AMI

65 patients Tn negative: unclear on admission if NSTEMI to be re-tested

Tn re-test <cut off no AMI 64

Tn re-test >cut off NSTEMI 1

Change over insensitive Tn: See next slide

Patient work up with sensitive Troponin

Source: Reichlin et al. J Am Coll Cardiol 2009;54:60-8

17

100 patients

Tn < cut off

Tn > cut off

sum

non-AMI 64 16 80

NSTEMI 1 9 10

STEMI 10 10

sum 65 35 100

Change over insensitive Tn:• 16 re-tests less: 81 - 65• 2 NSTEMI identified earlier : 9 - 7 • 14 additional false positives!! 16 - 2

Patient work up with sensitive Troponin(see slide 16)

100 patients

Tn < cut off

Tn > cut off

sum

non-AMI 78 2 80

NSTEMI 3 7 10

STEMI 10 10

sum 81 19 100

Patient work up with insensitive Troponin(see slide 10)

Comparison between sensitive Troponin and insensitive Troponin

18

99th percentile

Quite similar: 99th percentile 88% sensitivity, 92% specificity

What can sensitive Troponin do?

Source: Keller et al., NEJM 36;9, 2009

19

100 patients

Tn < cut off

Tn > cut off

sum

non-AMI 74 6 80

NSTEMI 1 9 10

STEMI 10 10

sum 75 25 100

25 patients Tn positive: Highly suspicious for AMI, start treatment or confirmatory diagnosis 19 with AMI

75 patients Tn negative: Unclear on admission if NSTEMI to be re-tested

Tn re-test <cut off no AMI 74

Tn re-test >cut off NSTEMI 1

Patient work up with sensitive Troponin

Change over insensitive Tn: See next slide

Source: Keller et al., NEJM 36;9, 2009

100 patients

Tn < cut off

Tn > cut off

sum

non-AMI 74 6 80

NSTEMI 1 9 10

STEMI 10 10

sum 75 25 100

20

Change over insensitive Tn:• 6 re-tests less: 81 - 75• 2 NSTEMI identified earlier : 9 - 7 • 4 additional false positives!! 6 - 2

Patient work up with sensitive Troponin(see slide 19)

100 patients

Tn < cut off

Tn > cut off

sum

non-AMI 78 2 80

NSTEMI 3 7 10

STEMI 10 10

sum 81 19 100

Patient work up with insensitive Troponin(see slide 10)

Comparison between sensitive Troponin and insensitive Troponin

21

100 patients Insensitive Troponin

Sensitive Troponin

Insensitive Troponin + Copeptin

Sensitive Troponin + Copeptin

assumed costs p.p.

Troponin re-tests

Copeptin tests

non-AMI declared AMI(false positives)

Temporary admittance (6h) for re-testing

missed AMI (false negatives )

Potential savings per patient:

Combining Troponin and Copeptin has potential to reduce costs significantly

99th percentile: 10% NSTEMI undetected

22

Finding the perfect cut-off for sensitive Troponin

Best specificity at high sensitivity

Best sensitivity at high specificity

99th percentile

Best sensitivity at high specificity 20% NSTEMI undetected

Best specificity at high sensitivity 4 out of 5 tests positives are false positives!

Yet another issue...

23

NSTEMI: Troponin plays a key role in AMI definition

New Definition of AMI (since 2000)

Criteria for acute Myocardial InfarctionI. Detection of rise and / or fall of cardiac biomarkers (preferably Troponin) with at least one value above the 99th percentile of the upper reference limit (URL) together with evidence of myocardial ischemia with at least one of the following:

- Clinical symptoms of ischemia

- ECG change indicative of new ischemia

- Imaging evidence of new loss of viable myocardium

II. Sudden, unexpected cardiac death

III. Pathological findings of an acute myocardial infarction in autopsy

Source: Thygesen et al. Universal Definition of Myocardial Infarction; J Am Coll Cardiol 2007; 50(22):2137-2195

24

The (fundamental) issue of NEJM trial results

Troponin plays a key role in NSTEMI definition; for the discussed data sets, therefore

more sensitive Tn detects less sensitive Tn

no additional NSTEMI observable that are sensitive Tn negative

25

t0 t1

Diagnosis (insensitive Tn)

NSTEMI

NSTEMI

other

NSTEMI

cut off Tn

pat. 1

pat. 3

pat. 4

pat. 5

Tn at baseline

The (fundamental) issue of NEJM trial resultssimplified simulation for insensitive Tn assays

pat. 2 other

Results of Gold Standard Diagnosis using insensitive Tn assays

Comparing insensitive Tn at baseline with Gold Standard Diagnosis

26

t0 t1

Diagnosis (insensitive Tn)

NSTEMI

other

NSTEMI

cut off Tn

Tn at baseline

Tn sens at baseline

No change

The (fundamental) issue of NEJM trial resultssimplified simulation for insensitive vs. sensitive Tn assays

cut off Tn

Part I: patient profiles that remain unchanged !

NSTEMIpat. 1

pat. 4

pat. 5

pat. 2

pat. 3

other

27

t0 t1

Diagnosis (insensitive Tn)

NSTEMI

other

other

cut off Tn

pat. 6

pat. 7

pat. 8

pat. 9

Tn at baseline

Tn sens at baseline

cut off Tn

Part II: additional patient profiles!

other

No additional NSTEMI observable that are

sensitive Tn negative

improves Tn sens performance

worsens Tn sens performance if cut off is chosen too lowpat. 10 other

The (fundamental) issue of NEJM trial resultssimplified simulation for insensitive vs. sensitive Tn assays

28

The (fundamental) issue of NEJM trial results

Troponin plays a key role in NSTEMI definition; for the discussed data sets, therefore

more sensitive Tn detects less sensitive Tn

no additional NSTEMI observable that are sensitive Tn negative

What are the consequences if sensitive Troponin is used for definition of AMI?

29

t0 t1

Diagnosis (sensitive Tn)

NSTEMI

NSTEMI

NSTEMI

cut off Tn

pat. 6

pat. 7

pat. 8

pat. 9

Tn sens at baseline,

new diagn.

The (fundamental) issue of NEJM trial resultssimplified simulation for future Gold Standard Diagnosis

cut off Tn

Part III: Gold Standard Diagnosis using sensitive Tn

other

Tn sens at baseline, old diagn.

Diagnosis (insensitive Tn)

NSTEMI

other

other

other

Results not predictable

pat. 10 other NSTEMI

30

The (fundamental) issue of NEJM trial results

Troponin plays a key role in NSTEMI definition, for current data sets, therefore more sensitive Tn detects less sensitive Tn no additional NSTEMI observable that are sensitive Tn negative

What are the consequences if sensitive Troponin is used for definition of AMI definition? applied to data set from Keller et al.*: 30% NSTEMI

* Keller et al. Copeptin Improves Early Diagnosis of Acute Myocardial Infarction; J Am Coll Cardiol 2010; 55(19): 2096-106

31

The (fundamental) issue of NEJM trial results

Troponin plays a key role in NSTEMI definition, for current data sets, therefore more sensitive Tn detects less sensitive Tn no additional NSTEMI observable that are sensitive Tn negative

What are the consequences if sensitive Troponin is used for definition of AMI definition? applied to data set from Keller et al.*: 30% NSTEMI

Advantage Copeptin: Copeptin is unrelated to Tn/AMI definition different pathological pathway very low risk that results are biased ROC curves may look similar, but information is not

32

Summary

The Copeptin algorithm

A useful and easy-to-use algorithm in chest pain:

negative Troponin as AMI rule out (insensitive or sensitive Tn)

together with a

negative Copeptin as AMI rule out (reason: low number of AMI)

33

Conclusions

The Copeptin algorithmA useful (but simplified) algorithm in chest pain:

• high Troponin as AMI rule in (current or sensitive Tn)• low Copeptin as AMI rule out

Reason: low number of AMI

Sensitive Troponin results are not easy to interpret• first value still needs to be verified by serial measurement after 6 hours• sensitive Troponin causes false positive values• many areas where sensitive Troponin is not available (general

practitioner, hospitals without high throughput labs etc.)

Simplified work-up – quicker diagnosis possible with Copeptin

~5%

~15%

~80%

~5%

~15%

~20-40%

~40-60%

Need to wait for 2nd Tn

2nd Tn