Correction
for
Karrison et al., J. Natl. Cancer Inst. 99 (19) 1455-1461.
Journal of the National Cancer Institute Advance Access originally published online on November 27, 2007
JNCI Journal of the National Cancer Institute 2007 99(23):1819; doi:10.1093/jnci/djm247
© Oxford University Press 2007.
Erratum
"Design of phase II cancer trials using a continuous endpoint
of change in tumor size: application to a study of sorafenib
and erlotinib in non–small-cell lung cancer" by Karrison
et al. [J Natl Cancer Inst 2007;99:1455–61 (Issue 19)].
An error occurred in the calculations for
Table 1 on the average
percent change in tumor size for the Ratain, 2006 (ref. 25 in
our paper), trial of sorafenib for renal cell cancer. When we
performed these calculations we had neglected to note that the
data reported in figure 1 of Ratain et al. were for bidimensional,
not unidimensional, changes. Consequently, the mean change in
unidimensional measurements would be about –9% rather
than –18%. This error also affects our calculations for
the Shepherd, 2005, trial (ref. 23 in our paper), because figure
1 from Ratain et al. was used to approximate the typical change
within each RECIST category. The estimated mean percent change
now becomes +3.6% rather than +10.1% for that trial. A corrected
Table 1 appears below. The implication of these changes is that
the planned clinical trial should seek to detect a smaller effect
size. This requirement is offset, however, by the smaller pooled
standard deviation (reduced from 0.346 to 0.214 on the log ratio
scale). Thus, if the true mean log ratio is 0.02 for E150/S0
(as in Shepherd, 2005), –0.06 for E150/S200, and –0.10
for E150/S400 (as in Ratain, 2006), the previously proposed
sample size of 50 patients per group will provide 85% statistical
power, using a one-sided test at the alpha=0.10 significance
level. (This calculation again incorporates a Tukey factor and
a small sample size increase in the event that nonparametric
tests are needed.)
CiteULike 
Connotea 
Del.icio.us What's this?
