|
16.1
Treatment
16.11
Design considerations: The primary objectives of this double-blind
study are to compare the appetite stimulation and weight gain
among the three treatment arms. The Megace-alone treatment will
be viewed as the standard treatment due to our prior trial demonstrating
effectiveness relative to a placebo treatment arm (Loprinzi
et al.,1990). Each of the other two treatments will be compared
to the Megace arm. Megace will be judged to be superior to the
alternative treatments if appetite stimulation and/or weight
gain are significantly superior, and side effects are less frequent.
On the other hand, if one of the alternative treatments is not
found to be significantly inferior to Megace, it may be judged
to be efficacious if weight gain and appetite stimulation are
similar to that seen on Megace, and toxicities are acceptable.
16.12
Accrual: A total of 450 evaluable patients will be accrued onto
this study (150 per arm). We estimate an accrual rate of a 30
patients per month and thus expect to complete accrual in approximately
15 months. Accrual will continue one month beyond the date at
which the accrual target is met to ensure a sufficient number
of evaluable patients' data is obtained. This figure is based
on accrual from previous trials dealing with patients experiencing
anorexia/cachexia. NCCTG protocol 89-92-55 reported accrual as
high as 45 patients per month averaging 38 patients accrued monthly
over the study period. While more recent NCCTG protocols (91-92-54
and 91-92-52) have had lower accrual rates, these protocols were
more restrictive than the proposed study and so it is anticipated
that the accrual rate we will obtain more likely will resemble
that of NCCTG 89-92-55.
16.13
Power for Detecting Whether Megace is Superior to the Alternative
Treatments:
16.131
Each of the two alternative regimens will be compared to the Megace-alone
arm by conducting two two-sided tests using an alpha-error of
0.025 for each of the tests. This will result in an overall approximate
error rate of a 5% chance of falsely concluding at least one of
the two alternative regimens to be significantly different from
Megace. Primary power calculations are based on detecting differences
in appetite stimulation, although other endpoints such as weight
gain (absolute change, BMI and PIW) and side effects will also
be compared.
16.132
For the purposes of this study, clinically significant outcomes
for the two primary endpoints of appetite stimulation and weight
gain will be defined as a 15% differential in the proportion of
patients reporting appetite stimulation and a 10% differential
in the proportion of patients who are able to gain at least 10%
of baseline weight. These outcomes are both practical in terms
of improvement needed to declare one regimen better than another
and are reasonable in the light of results of our previous trials
involving Megace.
|