Treatment of Middle East respiratory syndrome with a combination of lopinavir/ritonavir and interferon-ß1b (MIRACLE trial): statistical analysis plan for a recursive two-stage group sequential randomized controlled trial

Arabi, Yaseen M.; Asiri, Ayed Y.; Assiri, Abdullah M.; Aziz Jokhdar, Hani A.; Alothman, Adel; Balkhy, Hanan H.; AlJohani, Sameera; Al Harbi, Shmeylan; Kojan, Suleiman; Al Jeraisy, Majed; Deeb, Ahmad M.; Memish, Ziad A.; Ghazal, Sameeh; Al Faraj, Sarah; Al-Hameed, Fahad; AlSaedi, Asim; Mandourah, Yasser; Al Mekhlafi, Ghaleb A.; Sherbeeni, Nisreen Murad; Elzein, Fatehi Elnour; Almotairi, Abdullah; Al Bshabshe, Ali; Kharaba, Ayman; Jose, Jesna; Al Harthy, Abdulrahman; Al Sulaiman, Mohammed; Mady, Ahmed; Fowler, Robert A.; Hayden, Frederick G.; Al-Dawood, Abdulaziz; Abdelzaher, Mohamed; Bajhmom, Wail; Hussein, Mohamed A.

A detailed interim analysis plan is reported in the MIR-ACLE protocol [3] . The trial is designed as a recursive, two-stage, group sequential randomized trial. The first interim analysis will be conducted when 34 subjects (17 per group) have completed 90 days of follow-up. This is about 17.5% of the total sample size needed for the classical design (a classic two-group design requires a total of 194 subjects (97 subjects per group) to have an 80% power at a significance level of 5% using a one-sided Z test for difference in proportion to detect 20% absolute risk reduction in 90 days mortality among subjects receiving treatment (20%) compared to a control group (40%)). A Data and Safety Monitoring Board (DSMB) will be convened to review the unblinded data (efficacy and safety) and advise on continuation or termination of the trial. The determination of the stopping boundaries in the first two-stage design was calculated using the conditional power method based on the summing stagewise p values. At the first interim analysis, the DSMB will determine whether the trial should be terminated for futility or not using the following boundaries and their corresponding decisions ( Table 2) .