Agreed if alternative hypothesis were true.
Joking aside by BTH, it's common knowledge on this board that most clinical trials failed to reject null at interim analysis, simply because challenges due to cancer and shaky phase II data such that an inflated treatment effect assumption is used.
But I think MARQUEE has a shot at the interim. The inflation is less of an issue for Marquee, which seems to be powered with a much more conservative treatment effect than that observed from their phase II subgroup analysis, see JCO excerpt below (caveat with criticism of Cox modeling which the board statisticians will have a lot to say).
" proportional hazards model ... prespecified... used stepwise inclusion to select covariates, and the included variables were male sex, one prior chemotherapy regimen, progressive disease as best prior response, longer than 6 months since diagnosis, and EGFR wild-type status. PFS was also analyzed using central radiology review in place of investigator assessments ...
At the time of OS analysis, median follow-up was 14 months, and
105 patients had died. Median OS was 8.5 months for ET and 6.9
months for EP (HR, 0.87; 95% CI, 0.59 to 1.27; P.47; Fig 3B). The
proportional hazards model developed for the PFS analysis resulted in an adjusted OS HR of 0.87 (95% CI, 0.59 to 1.29; P.50).
Preplanned exploratory survival analyses were performed to examine
outcomes according to tumor histology and molecular characteristics.
Among patients with nonsquamous histology (n 117),
there was a trend toward benefit from ET in both PFS (HR, 0.71;
95% CI, 0.46 to 1.10; P .12; Fig 3C) and OS (HR, 0.72; 95% CI,
0.44 to 1.17; P .18; Fig 3D). Applying the proportional hazards
model generated for the ITT population to patients with nonsquamous histology in a hypothesis-generating analysis revealed an adjusted PFS HR of 0.61 (95% CI, 0.39 to 0.98; P .04) and an
adjusted OS HR of 0.58 (95% CI, 0.34 to 0.99; P.04) for patients
with nonsquamous histology. "
News 
Market Data 
Discover 
