Biotech Values

[urche]

CAMH

99% negative predictive value would be a compelling statistic in favor of making TWA testing standard of care.

I believe you cite that statistic based on the ALPHA study reported earlier this year.
http://tinyurl.com/yoknyl

The ALPHA study (Prognostic Value of T-Wave ALternans in Patients with Heart FAilure Due to Nonischemic Cardiomyopathy) enrolled 446 consecutive patients with NYHA Class II or III non-ischemic cardiomyopathy and left ventricular ejection fraction (LVEF) less than or equal to 40%. On the primary endpoint (cardiac death and life-threatening arrhythmias) an abnormal MTWA test had a Hazard Ratio of 4.01 (p=0.002), or four times the risk than a normal MTWA test. The 12-month negative predictive value of the test was reported to be 98.7%, indicating that patients with a negative test result are at very low risk of sudden cardiac death. For patients with LVEF less than 35%, the Hazard Ratio and negative predictive value were 4.28 (p=0.004) and 99%, respectively.

Note that this study was in patients with non-ischemic cardiomyopathy. I believe that about 75% of cardiomyopathy is ischemic, at least in the developed world. As far as I know, we don't yet have phase III results of a prospective study in the ischemic cardiomyopathy population. From this PR (http://tinyurl.com/2e887j), it appears that the results will be no better than EP testing, the current standard of care.
"The positive and negative predictive values of a MTWA-directed strategy and that of an EP test alone were essentially identical," said Costantini. "This is most important, as the clinically relevant question today is not who should receive an ICD, but who is unlikely to benefit from ICD therapy. The use of a non-invasive approach to help us answer this question represents a significant step forward."


IF TWA is similar to EP testing, that is pretty good, obviously, but no where near 99% negative predictive value.
I am predicting that the negative predictive value for the typical patient with ischemic cardiomyopathy will be significantly lower, perhaps low 80's. My feeling is that will be low enough to make the test not acceptable as standard of care for a literally life and death decision.