October 31st, 2013

Does OPTIMIZE Reveal the Optimal DAPT Time Frame?


In the OPTIMIZE trial of 3119 patients with stable coronary artery disease or low-risk ACS treated with zotarolimus-eluting stents, 3 months of dual antiplatelet therapy (DAPT) was noninferior to 12 months for net adverse clinical and cerebral events (composite of all-cause death, MI, stroke, or major bleeding) without significantly increasing the risk of stent thrombosis. Tell us:

  • In what percentage of your PCI cases do you use a zotarolimus stent?
  • Are the results unique to the zotarolimus-eluting DES or applicable to other second generation (i.e., everolimus-eluting) DES?
  • Does this study persuade you to administer DAPT for 3 months rather than 12 months to low-risk patients treated with a DES?

To read more about OPTIMIZE, click here for our news story.


3 Responses to “Does OPTIMIZE Reveal the Optimal DAPT Time Frame?”

  1. Fahim H Jafary, MD, FACC, FSCAI says:

    The Endeavor stent used in this trial was jokingly called the “diet coke” version of a DES because it still had a greater late loss than other DES at the time. These results can’t be extrapolated to the current DES because theoretically they may still take longer to fully bury the stent struts under neointima. Hence I won’t elect to use DAPT for 3 months at present. On the other hand, it is reassuring that no “badness” was seen with just 3 months of DAPT and these data are congruent with other studies demonstrating that interruption of DAPT isn’t necessarily a bad thing. As everyone knows, the one year DAPT recommendation was never based on any solid evidence.

  2. The conclusion that 3 month DAPT is no worse (noninferior) than 12 month DAPT cannot be supported by the data reported in OPTIMIZE for the following reasons:
    1. The noninferiority margin fixed as 2.7% ARD assuming expected active control event rate of 9%. This translates into a risk ratio margin of 1.3 (9+2.7/9 = 11.7/9). However, the observed active control event rate was 5.8%. So, the risk ratio margin becomes inflated to 1.47 (5.8+2.7/5.8). Had the investigators appropriately fixed the margin as risk ratio, they would have failed to establish noninferiority as the 95% upper bound exceeds 1.3 risk ratio.

    2. Typically, it is standard practice to establish noninferiority with respect to an efficacy endpoint, and demonstrate an ancillary advantage for the new therapy (bleeding). When one uses CVD, MI, stent thrombosis as the primary efficacy endpoint, noninferiority is not established (using the risk ratio margins) and there is no bleeding advantage.
    The use of NACCE is not an acceptable regulatory endpoint for noninferiority trials.

    3. By convention, noninferiority trials require higher power (>90%) than the 80% used in this trial.
    Coupled with the fact that this was an open label trial, the low power challenges the reliability of the conclusions. Low power increase the likelihood of both type 1 and 2 errors.

    Until the results of the properly designed DAPT trial become available, the most effective strategy to minimize DAPT is to look for a reason not to deploy DES. Most patients with current generation BMS will never have restonosis, so why not give them the benefit of doubt with BMS without the hassle of prolonged DAPT. For those whose lesion characteristics require DES (no more than 30-40% of patients), I still give up to 1 year of DAPT unless the bleeding risk is prohibitive.

  3. Terrific comments Sanjay – these insights are hard to derive from the headlines. And Happy Diwali to you and any of our readers who are celebrating.