hi speed in the bd 3 pages was dedicated to the placebo effect.
agency seems to have a lot of concern on this.
imo this might be a bigger issue than the outcome studies.
although it's very long i will just paste it here which i think
is important to discuss. opinion anyone?
5.5 Placebo Group Effects
As previously mentioned, the magnitude of the changes in several lipid and lipoprotein
parameters, as well as biomarkers of inflammation, between baseline and Week 12 in the
placebo group are rather atypical for lipid-lowering trials. These trials, including
ANCHOR, often include a several-week lead-in period to stabilize diet and concomitant
lipid-altering medications well before baseline measurements. Although even highly
statistically significant within-group changes can certainly result from factors other than
the intended experimental intervention, one concerning possibility is that the mineral oil
placebo may not be biologically inert. If this were true, the estimated treatment effects
may be biased.
Thus, the review team sought evidence that might help explain the changes observed in
the mineral oil group. These included considering the plausibility that treatment
assignment could have been unmasked due to physical differences in study drug
appearance or manufacture; reviewing the literature for mineral oil-specific effects on
lipid parameters or absorption of fat-soluble vitamins; evaluating whether the statintreated
subjects in the placebo group from MARINE demonstrated a similar pattern; and
considering elements of the ANCHOR study design that may have contributed. Finally,
the Division reviewed lipid changes observed in the placebo groups of other lipidlowering
trials.
The Chemistry, Manufacturing, and Controls data do not suggest that blinding would
have been compromised. The only difference between the active capsules and the
placebo capsules was that the drug substance (icosapent ethyl) was replaced with mineral
oil. All other formulation components and composition remained the same and were
added in an identical fashion. The submitted certificates of analysis for the AMR101 and
placebo lots used in this trial describe identical appearances of the blister packs and
capsules. Admittedly, even if study subjects were able to discern their assignment to
placebo or AMR101, it is difficult to predict what direction bias would be introduced
(e.g., how might treatment assignment influence one’s adherence to dietary instruction?).
Three studies using mineral oil as a placebo and reporting baseline and end-of-treatment
lipid values were reviewed to determine if similar changes were observed to those that
occurred in ANCHOR. 19,20,21 The population of patients studied varied greatly:
dyslipidemic women with type 2 diabetes, patients infected with HIV, and healthy
volunteers. The exposure to mineral oil placebo ranged from 10 days to 2 months with
daily doses of 6 grams or less. Despite these differences, in general, the effect of the
mineral oil placebo on lipid parameters was small. For example, after 8 weeks of
mineral oil (6g/day), the median percent change of TG from baseline in HIV-infected
patients was +1%. 22
Studies from the 1940s suggested that mineral oil may block the absorption of fat-soluble
vitamins.23,24 Articles submitted by the applicant and independent review of the available
medical literature on this issue were reviewed. Although initial studies suggested
possible malabsorption with mineral oil, subsequent studies using large volumes of
mineral oil (up to 150 mL/day) over a long period of time called these findings into
question.25,26,27,28 Of course, patients in the ANCHOR trial’s placebo group ingested far
smaller volumes of mineral oil than this as well (approximately 4 mL/day), which
weakens but does not eliminate the possibility of a local intestinal effect of mineral oil on
statin absorption.
Whether mineral oil affects statin absorption has not been formally tested to our
knowledge. The applicant submitted data regarding patients who were taking
concomitant statin therapy in the MARINE trial and who were randomized to the mineral
oil group. Only 18 patients in the mineral oil group were taking a statin. The median
percent change in LDL-C was -8% in the statin-treated mineral oil group, with large
variability (Q1 -36.0%, Q3 +30.8%); the median change was 0% in LDL-C among the 57
patients not taking statins in the mineral oil group. The applicant contends that if mineral
oil reduced statin exposure, then LDL-C should have increased after 12 weeks of
treatment, not decreased. While the reduction in LDL-C in this group is somewhat
reassuring, the small number of statin-treated patients and the large intra-subject
variability do not allow definitive conclusions from this subgroup.
Patient compliance (indirect measures): There was no dietary compliance assessment or
measurement of physical activity in the ANCHOR trial. However, indirect
measurements of diet and physical activity, i.e., weight, waist circumference, and BMI
did not demonstrate significant changes between the placebo and AMR101 treatment
groups, suggesting that physical and dietary habits between groups were not dramatically
different throughout the trial and are unlikely to have contributed to the effects observed
in the placebo group.
Regression to the mean: Subjects enrolled in ANCHOR were selected non-randomly
from a broader population of subjects of which 70% failed to be randomized. The
applicant contends that the asymmetric selection process may have contributed to a
regression-to-the mean phenomenon apparent across the lipoprotein lipids and other
biomarkers within the placebo group. If true, this would be a design element expected to
affect all treatment groups similarly and the between-group differences should provide
unbiased estimates of the treatment effects. Averaging two qualifying values separated by
one week, all following a =4-week lead-in stabilization period, should have reduced the
contribution of regression to the mean, although its possible contribution cannot be ruled
out.
Considering the 609 subjects who were excluded at the end of the screening period
because of ineligible lipid values, the majority (65%) had TG levels that were too low.
Although one cannot determine with certainty, this suggests that the study might have
been more likely to enroll patients who had “random highs” rather than “random lows,”
and if this were the case, TG levels would be expected to regress downward rather than
upward. Regarding LDL-C, most (60%) of the subjects excluded for lipid reasons had
LDL-C in range with the remainder more likely to be excluded for low LDL-C than high
LDL-C.
Therapeutic changes during Lead-in period: The applicant has put forward the
hypothesis that changes in lipid-lowering regimens and wash-out of non-statin therapy
during the lead-in period may have increased variability of TG levels after
randomization. Although this could occur, it doesn’t seem that “larger variability” would
explain the highly statistically significant changes observed in the placebo group between
baseline and Week 12.
Lipid changes in patients randomized to placebo: The applicant provided a table of
studies (see Appendix) listing the lipid changes observed in placebo-treated patients from
baseline in studies of patients with high or very high TG levels to compare with
ANCHOR. In reviewing the trajectory of lipids in a placebo group, it is important to
consider if the placebo group was on background statin therapy and if all lipid-lowering
drugs were stopped during a washout period prior to randomization, as this may affect the
degree and direction of lipid alterations. For example, if a placebo group was not on any
lipid-lowering medications, it may be reasonable to expect a worsening of lipid
parameters over time. However, if a placebo group was on statin therapy that required at
least 4 weeks of consistency, it might to reasonable to expect lipid parameters to remain
stable over time with minor fluctuations. Acknowledging the limitations of crosscomparisons,
two studies (COMBOS and FIRST) had patient populations and study
designs with lead-in periods of diet and background statin stabilization similar to
ANCHOR. The placebo groups in COMBOS and FIRST had small reductions from
baseline in TG at the 8 week and 13 week time points (-6.3% and -2.0%, respectively).
The placebo-treated patients in COMBOS also had reductions in other measured lipid
parameters. These results suggest the changes in ANCHOR are atypical, but the etiology
of this remains unclear.
News 
Market Data 
Discover 
