Analysis methodology
We used descriptive statistics to describe the dataset, which was
divided in 3 time periods (2008–2012, 2013–2016, and 2017–2020) for
subgroup summaries. Fisher exact test was used to compare the
characteristics among the 3 time periods. Some approvals incorporated
ongoing trials planned with adaptive designs at the time the products
were approved. We performed additional descriptive statistics by study
progressions (approved with ongoing study or completed study ) and
compared the features using Fisher exact test. To detect a trend over
time in the yearly numbers of approvals, we performed joinpoint
regression analyses via the Joinpoint Regression Program version 4.7.0.0
(National Cancer Institute and Information Management Services Inc.,
Bethesda, MD, USA). Other statistical analyses were carried out using
SAS version 9.4 (SAS Institute, Carey, NC, USA). Differences were
considered significant based on 2-sided tests if P values were
less than 0.05.