Download Intensive Combination Drug With Lovastatin, Probucol

16
Intensive Combination Drug
Therapy of Familial Hypercholesterolemia
With Lovastatin, Probucol, and
Colestipol Hydrochloride
Joseph L. Witztum, MD, Debra Simmons, MD, Daniel Steinberg, MD, PhD,
William F. Beltz, PhD, Robert Weinreb, MD, Stephen G. Young, MD,
Peggy Lester, RN, Nancy Kelly, RN, and Joseph Juliano, BA
Downloaded from http://circ.ahajournals.org/ by guest on June 17, 2017
Patients with familial hypercholesterolemia (FH) have had a life-long sustained elevation of
low-density lipoprotein (LDL) cholesterol levels. Consequently, there is a need to maximally
lower their elevated levels, and this usually requires lowering LDL levels more than 50%.
Because no single hypolipidemic drug will consistently produce such degrees of lowering,
combination drug therapy with two or even three agents is required to produce the desired
degree of cholesterol lowering. A prospective trial was designed to determine if combination
therapy using three hypolipidemic agents could effectively lower LDL levels in 17 severely
affected FH subjects. Colestipol hydrochloride (10 g b.i.d.), probucol (500 mg b.i.d.), and
lovastatin (20 or 40 mg b.i.d.) were given to each patient, in varying combinations, over a
25-month period. Lovastatin (40 mg/day) uniformly lowered LDL levels 36%. Probucol
lowered LDL only 14% and in a variable manner. The combination of lovastatin and probucol
lowered LDL no better than lovastatin alone. Lovastatin plus colestipol lowered LDL 52%;
probucol added as a third agent produced no further lowering. Lovastatin (80 mg/day) plus
colestipol lowered LDL 56%. Lovastatin increased high-density lipoprotein (HDL) cholesterol
levels 6%, whereas probucol decreased HDL 29%o. In all patients there was an efective lowering
of LDL levels, ranging from 40% to 70%Yo. Thus, lovastatin plus colestipol is an elfective
hypolipidemic regimen for producing marked decreases in LDL levels in FH subjects. The
addition of probucol as a third hypolipidemic agent adds little to the therapeutic regimen as
measured by lowering of LDL levels. (Circulation 1989;79:16-28)
I ndividuals with elevated plasma levels of low-
density lipoprotein (LDL) cholesterol clearly
have an increased risk for developing coroFrom the Departments of Medicine, Division of Endocrinology and Metabolism, and Ophthalmology (R.W.), University of
California, San Diego, La Jolla, California.
Supported in part by Grant HL-14197 (SCOR) from the National
Heart, Lung, and Blood Institute; by Grant PHS-RR-00827 of the
General Clinical Research Center; and by grants from the Merck,
Sharp & Dohme Research Laboratory and the Dow Chemical
Company. S.G.Y. is supported in part by a National Institutes of
Health Clinical Investigator Award; W.F.B. is supported in part
by a Whitaker Foundation Grant; and J.L.W. is an Established
Investigator of the American Heart Association.
Address for correspondence: Joseph L. Witztum, MD, Department of Medicine, M-013D, University of California, San Diego,
La Jolla, CA 92093-0613.
Received May 6, 1988; revision accepted September 12, 1988.
*All editorial decisions for this article, including selection of
reviewers and the final decision, were made by a guest editor.
This procedure applies to all manuscripts with authors from the
University of California San Diego or UCSD Medical Center.
nary artery disease (CAD). In particular, patients
with heterozygous familial hypercholesterolemia
(FH) are at increased risk for CAD, as they have
had a life-long sustained elevation of LDL cholesterol. Affected male heterozygotes have approximately a 50% chance of having a major coronary
event by age 50, and even affected female heterozygotes incur a substantially greater risk for CAD.1
Thus, vigorous attempts to lower their elevated
plasma LDL levels are indicated. In fact, one might
argue that the goal of therapy in these FH patients,
particularly in the adult, should be the achievement
of "ideal" LDL levels, which are far below the
so-called "normal" mean for the general population. Regression studies in primates,2 as well as the
results of intervention trials in humans, including the
recently completed Coronary Primary Prevention
Trial3 and the CLAS Trial,4 clearly indicate that the
beneficial effects of hypolipidemic therapy are directly
related to the extent of cholesterol lowering.
Witztum et al Drug Therapy of Familial Hypercholesterolemia
Downloaded from http://circ.ahajournals.org/ by guest on June 17, 2017
Plasma LDL cholesterol levels in patients with
familial hypercholesterolemia are almost always
well in excess of 200 mg/dl.5 If one accepts the
premise that the goal of therapy is to lower their
LDL cholesterol to levels below 130 mg/dl,6 then
therapy in such individuals needs to lower LDL
levels by more than 50%. Although occasional
patients may experience 50% drops in LDL levels
in response to a single agent, it is far more common
to see decreases of only 25-35%. Thus, in response
to bile sequestrants, LDL levels may fall 20-35%.7-10
Nicotinic acid, in the estimated 50-60% of patients
who can tolerate the required doses, may produce
decreases of 20-35%.1l,12 Probucol will produce
decreases of 10-40% in responding patients, but
many patients do not respond.13 Lovastatin, at
doses of 20-40 mg b.i.d. reduces LDL levels by
approximately 35_43%.14-18 Thus, for many if not
most patients with FH, no single drug therapy will
produce the desired 50% decrease, and combination
therapy will be required.
The bile acid binding resins, such as colestipol
hydrochloride, are believed to lower plasma cholesterol levels by increasing bile acid excretion. This
leads to the subsequent depletion of a critical intracellular pool of hepatic cholesterol and consequent
induction of hepatic LDL-receptor activity. This
increased LDL-receptor activity in turn leads to
increased removal of LDL and LDL precursors
from plasma.19-21 Lovastatin is a hypolipidemic
agent that specifically inhibits the hepatic activity of
HMG coenzyme A reductase, the rate-limiting
enzyme in cholesterol biosynthesis,22 and consequently leads to decreased hepatic cholesterol biosynthesis and further stimulation of hepatic LDLreceptor activity.23,24 Probucol is a highly lipophilic
17
agent whose hypolipidemic mechanism of action in
humans is uncertain. In rabbits, it appears to increase
LDL removal from plasma, in part by a non-LDL
receptor-mediated pathway.25 It is of particular
interest because among all known hypolipidemic
agents, it has been reported to be effective in
reducing plasma LDL concentrations in subjects
homozygous for familial hypercholesterolemia.26,27
Because these latter subjects totally lack LDLreceptor activity, it suggests that in humans probucol can lower LDL levels independently of LDL
receptor activity. In theory, colestipol, lovastatin,
and probucol could act synergistically to lower
LDL cholesterol levels.
We now report the results of a prospective trial to
determine the efficacy of lovastatin, probucol, and
colestipol used in various combinations to lower
plasma LDL levels in FH patients with marked
hypercholesterolemia. The majority of the patients
recruited for this study had previously failed to
achieve acceptable LDL cholesterol levels on various combinations of bile sequestrants, nicotinic
acid, or other hypolipidemic drugs. The results
show that even in FH patients with severe hypercholesterolemia, marked reductions in LDL cholesterol levels can be achieved by the use of lovastatin
in combination with a bile sequestrant.
Subjects and Methods
Human Subjects
Seventeen patients with severe hypercholesterolemia were recruited from the Lipid Clinic of the
Division of Endocrinology and Metabolism of the
University of California, San Diego. These 17 subjects were selected because of severe hypercholes-
TABLE 1. Clinical Characteristics of Study Subjects and Lipoprotein Values Obtained During the Control (Placebo) Period
Baseline lipoprotein values (mg/dl)
CAD!
Xanthoma/
Age/
Total triglyceride
PVD
HDL cholesterol
LDL
cholesterol
cholesterol
xanthelasma
Total
sex
Subject
42
86
+
+
438
496
1
51/M
78
50
270
205
2
60/M
80
49
408
474
3
47/M
+
130
48
244
319
4
64/M
82
46
370
308
36/M
48
56
318
258
6
24/M
114
45
446
378
7
40/M
+
34
345
344
447
8
56/M
54
124
426
504
9
56/M
+
215
32
406
481
10
36/M
120
50
218
293
11
47/M
+
193
40
254
333
12
53/F
416
72
200
355
13
61/F
197
42
460
542
14
65/F
68
91
210
315
15
63/F
125
60
504
588
16
63/F
180
32
256
324
17
52/F
CAD, coronary artery disease; PVD, peripheral vascular disease; LDL, low-density lipoprotein; HDL, high-density lipoprotein; M,
male; F, female.
18
Circulation Vol 79, No 1, January 1989
Lov
Control
400
L
Lov
Prob
Lov
Prob
Lov
+
+
+
+
Prob
Colest
Colest
Colest
Lov (80)
Total
0)
E
300
K
\LDL
-5
4-
CO
0)
~~±T+
-5
0
200
Downloaded from http://circ.ahajournals.org/ by guest on June 17, 2017
100
1
FIGURE 1. Plot ofeffects ofdrug therapy on total and
low-density lipoprotein (LDL) cholesterol levels. Each
period of this study is represented by one column and
lasted from 3 to 6 months. For each period, an
average lipid value was calculated for each individual
from two to four values obtained at monthly intervals.
The first value was obtained after 2 months of each
intervention to allow for the full expression of the
addition (or removal) of a given drug. For each period
a group mean (±SEM) was then calculated and displayed on the plot. Control, placebo for lovastatin and
placebo forprobucol; Lov, lovastatin 20 mg b. i. d. and
probucolplacebo; Prob, lovastatin placebo andprobucol 500 mg b. i. d.; Lov +Prob, lovastatin 20 mg b. i. d.
and probucol 500 mg b. i. d.; Lov +Prob + Colest, lovastatin 20 mg b. i. d. and probucol 500 mg b. i. d. and
colestipol 20 glday (usually 10 g b. i. d.); Lov + Colest,
lovastatin 20 mg b. i. d. and colestipol 20 glday; Lov
(80) + Colest, lovastatin 40 mg b. i.d. and colestipol 20
giday.
1
terolemia and failure to achieve adequate LDL
cholesterol lowering with previous hypolipidemic
therapy, which most commonly included a bile
sequestrant plus large doses of nicotinic acid. The
control, off-therapy, lipoprotein levels and pertinent clinical characteristics of the patients are listed
in Table 1. Patients were judged to have FH if they
met three of the following four criteria: 1) LDL
cholesterol level more than the 99th percentile for
their age and sex,28 2) strong family history of
hypercholesterolemia, 3) strong family history of
early coronary artery disease, or 4) presence of
tendon xanthomata. Fourteen of the 17 patients had
an unequivocal diagnosis of familial hypercholesterolemia. One subject (subject 4) had high LDL
cholesterol levels and early onset of coronary artery
disease but indeterminate family history. Three
subjects (subjects 2, 11, and 12) had positive family
histories by no xanthomas and thus were judged to
probably have FH. Secondary causes of hypercholesterolemia were excluded in all subjects by standard laboratory tests. Eight patients had existing
peripheral vascular disease, coronary artery disease, or were status post-coronary artery bypass
graft surgery and were taking a-blockers, diuretics,
or calcium channel antagonists. In these patients,
the dose of these medications was established before
entrance into this study and was not changed during
the course of the study. All patients gave informed
consent to participate in the study, which was
approved by the Human Use Study Committee of
the University of California, San Diego. A preliminary report of a part of this study was previously
presented at a symposium.29
Study Protocol
Before initiation of this protocol, patients discontinued all hypolipidemic agents for 3 months. All
patients were already on American Heart Association Phase II or Phase III diets, and a dietary review
took place during the 3 months preceding this
study, with at least one follow-up visit during each
of the subsequent drug periods during the remainder of the study. The study protocol can be appreciated from Figure 1. Patients came to the clinic on
30 occasions over a 25-month period. The study
was divided into seven periods. Each of the first five
periods was of 3 months' duration; a sixth period
lasted 6 months; and a seventh period lasted 4
months. During the first "control" period, all
patients received a double placebo, one for lovastatin and one for probucol. During the second
period, designated "lovastatin," patients received
lovastatin 20 mg b.i.d. and placebo for probucol.
During the third period, "probucol," patients
received a placebo for lovastatin and probucol 500
mg b.i.d. During the fourth period, "lovastatin plus
probucol," patients received active lovastatin (20
mg b.i.d.) and probucol (500 mg b.i.d.). During the
initial 1st year of the study, patients were blinded as
to medication as well as to their plasma cholesterol
results. Beginning with the fifth period, the blind
was broken and patients were now maintained on
lovastatin (20 mg b.i.d.) and probucol (500 mg
b.i.d.), and colestipol (10 g b.i.d.) was added.
During the sixth period, probucol was discontinued,
and lovastatin and colestipol continued. This period
was 6 months in duration to ensure that there was
no carry-over hypolipidemic effect of probucol.30
During a seventh period, colestipol was continued,
Witztum et al Drug Therapy of Familial Hypercholesterolemia
19
TABLE 2. Mean Lipid and Lipoprotein Levels During Each Period of the Drug Trial
Lipid level (mg/dl)
Total
cholesterol
404±98
286±58
345±89
268±63
220±59
Downloaded from http://circ.ahajournals.org/ by guest on June 17, 2017
Total
Period
n
LDL cholesterol
HDL cholesterol
triglyceride
1. Control
17
325 ±102
49±15
154±99
2. Lovastatin
17
208±61
53±17
126±74
3. Probucol
17
280±92
35±10
148±80
4. Lovastatin and probucol
17
206±61
39± 10
117 ±70
5. Lovastatin and probucol
15
154±53
40±10
131±75
and colestipol
6. Lovastatin and colestipol
16
234±45
157±42
49± 17
136±83
7. Lovastatin (80 mg/day) and
13
229±54
149±50
52± 16
134±67
colestipol
8. Lovastatin (80 mg/day) only
5
259 ±65
187±70
52±22
107±50
For each period an average lipid level was determined for each individual from two to four values obtained at monthly intervals and
a mean value (±SD) computed for the group. During periods one through four, all 17 patients adhered to the study protocol.
Subsequently, five subjects discontinued colestipol for varying periods and one discontinued probucol. Statistical results for the
cholesterol data are given in Figures 2-4. Plasma triglyceride levels were different from control during periods two (p<0.02), four
(p < 0.001), and eight (p < 0.05).
LDL, low-density lipoprotein; HDL, high-density lipoprotein.
Five subjects took only lovastatin (80 mg/day) at the end of the study (period B) for at least 3 months. The mean levels of cholesterol,
LDL cholesterol, HDL cholesterol, and triglycerides (±SD) during the control period for these five subjects were 396 ± 89, 309 + 100,
48+ 16, and 198 ± 136 mg/dl, respectively. These values were not different from the mean baseline values of the other 12 subjects.
and the dose of lovastatin was increased to 40 mg
b.i.d. It is important to emphasize that we did not
randomize the order of receiving probucol and
lovastatin during periods three, four, and five
because probucol is known to have a long plasma
half-life and to have residual effects for several
months after discontinuation.30 On the other hand,
after discontinuation of lovastatin, plasma lipoprotein levels return to baseline within 1 month.18
Thus, all patients received lovastatin first, then
probucol, and then the combination. During period
six when probucol was discontinued, the period
was extended for 6 months to minimize the carryover effect of probucol.30
During the first visit (visit 0), patients were entered
into the study, and a complete history and physical
examination were obtained, and lipoprotein quantification and baseline screening tests (SMA 12, T4,
CBC, urinalysis, and chest radiography) were performed. Ophthalmologic examination was performed during the control period and after 1 and 2
years of therapy and included best corrected visual
acuity, color vision (Ishihara color plates), slit-lamp
biomicroscopy, applanation tonometry, dilated fundoscopy, and visual field testing. During the entire
study, subjects were seen on a monthly basis.
During the first five periods, subjects were seen for
an extra visit 2 weeks after change of study medication to ensure absence of toxicity of the various
drug combinations. Lipoprotein quantification and
chemistry screening tests for toxicity were performed at every monthly visit. Compliance to all
phases of the study protocol was excellent as most
of the patients in this study had been followed for
many years at our Lipid Research Clinic and were
highly motivated to participate in this study. All 17
patients were followed for the entire period of
study. Adherence to lovastatin and probucol was
monitored by pill count and in general was more
than 90%. Adherence to colestipol was monitored
by self-report and the amount of resin consumed
(which we supplied). Overall adherence throughout
the study was excellent in 12 patients. One subject
(subject 17) discontinued probucol at the end of
period four because of headaches. One subject
(subject 9) discontinued colestipol after period five
but resumed 10 g/day during the final 2 months of
period seven. Four other subjects discontinued
colestipol during period seven. Thus, there were
five subjects who took only lovastatin 80 mg/day for
2 or more months during the course of the study
(referred to as period 8 in "Discussion"). Side
effects were systematically recorded at each visit
and will be reported in "Results."
Lipoprotein Quantification
Plasma was collected from each patient after an
overnight fast into tubes containing EDTA (1 mg/
ml) and promptly refrigerated. Plasma was separated within several hours and used within 3 days
for lipoprotein quantification. Total cholesterol and
triglycerides were measured by enzymatic techniques with an ABA-200 biochromatic analyzer
(Abbott Laboratories, Irving, Texas), high performance cholesterol reagent (No 236691, BoehringerMannheim Diagnostics, Indianapolis, Indiana), and
triglyceride agent (No 6097, Abbott Laboratories).
LDL and HDL cholesterol was measured according
to the standardized procedures of the Lipid Research
Clinic Manual.3' Our laboratory is under the continuous standardization program of the Centers for
Disease Control, Atlanta, Georgia.
Circulation Vol 79, No 1, January 1989
20
Lov
0
Prob
o
0
8) 10
CL
0
Lov
Lov + Prob Lov + Prob Lov + Colest Lov (80)
+ Colest
+ Colest
20
Lov + Prob
0
a)
10
0~
20
Lov + Prob Lov + Colest Lov (80)
+ Colest
+ Colest
0
4-
0
a)
30
E 30
W
0
40
CO
0
Prob
0
50
tn
40
0)C-)
50
a)C)
C2
60
60
Downloaded from http://circ.ahajournals.org/ by guest on June 17, 2017
FIGURE 2. Bar chart of effect of each drug period on
total cholesterol levels. For each drug period, the mean
percent change (± SEM) versus the control period was
calculated. All values shown were significantly different
from control period at p <0. 001. Significances of mean
differences between drug periods are indicated in the
figure. Abbreviations for individual drug periods are
explained in legend to Figure 1.
Data Analysis
Although lipid and lipoprotein analysis were done
monthly throughout the study, because of variations in time that individual drug regimens took to
achieve full effect and to allow the complete washout from a previous drug effect, we used only the
lipoprotein values obtained during the 2nd and 3rd
month of each of the first five drug periods. During
period six, lipoprotein levels were obtained monthly
from the 2nd to the 6th month. A preliminary
analysis indicated that for the group as a whole, the
lipoprotein levels obtained during months 5 and 6
were statistically similar to those obtained during
months 3 and 4, and therefore, these values were
used for each individual. For period seven, the four
values obtained at monthly intervals were used.
A multivariate, multiple linear regression model
was developed to provide a general method by which
a quantitative description of the effects of the three
drugs and their combinations could be estimated and
the significance of the different effects could be
tested. The BMDP package of statistical programs
was used for all analyses.32 A detailed description of
this model is presented in the "Appendix."
Results
All 17 patients who entered the protocol completed all phases of the study. The effects of the
various therapeutic regimens on mean plasma lipid
and lipoprotein levels are given in Table 2 and
Figure 1. The effects of the various therapeutic
regimens on the individual lipoprotein classes,
expressed as a percentage of the control value
obtained during the placebo period, are given in
Figures 2-4.
_
NS
p-
<.00lJ
FIGURE 3. Bar chart of effect of each drug period on
low-density lipoprotein cholesterol levels. For each drug
period. the mean percent change (± SEM) versus the
control period is shown. Significance levels between
periods were calculated from the multiple regression
model as explained in "Subjects and Methods." All
values were significantly different from control period at
p < 0. 001. Significances of mean differences between drug
periods are indicated in figure. Abbreviations for individual drug periods are in legend to Figure 1.
Effects on Plasma Cholesterol Levels
In response to lovastatin (20 mg b.i.d.), total
plasma cholesterol fell 29%. In contrast, probucol
lowered cholesterol only 15% in the same subjects.
The combination of lovastatin and probucol was
only slightly more effective than lovastatin alone
and clearly not additive. The addition of colestipol
hydrochloride was clearly beneficial in that total
cholesterol levels decreased further. Compared with
baseline, a mean cholesterol lowering of 46% was
achieved. When probucol was discontinued, mean
plasma cholesterol rose only slightly, -4%, indicating that on average the combination of lovastatin
1Lov
+20
a)
CL
Lov + Prob
Prob
Lov + Prob Lov
+ Colest
+
Colest Lov (80)
+ Colest
+10
-1
0
0
E
-10
-
a)
Ose
to
-20
J
-30
Lp< .001
p<
.001
NS
p <.001-Lp <.005]
FIGURE 4. Bar chart of effect of each drug period on
high-density lipoprotein cholesterol levels. For each drug
period, the mean percent change (±SEM) versus the
control period is shown. All values were significantly
different from control period at p<0.02 except for the
Lov + Colest period, which was not different. Significances of mean differences between drug periods are
indicated in figure. Abbreviations for individual drug
periods are given in legend to Figure 1.
Witztum et al Drug Therapy of Familial Hypercholesterolemia
plus colestipol was nearly as effective at lowering
total plasma cholesterol as was lovastatin, probucol,
and colestipol. In the 13 individuals who took lovastatin (80 mg/day) plus colestipol, total cholesterol
levels were only decreased 4% below levels observed
on lovastatin (40 mg/day) plus colestipol. The five
subjects who took only lovastatin (80 mg/day) had
average decrease of total cholesterol of 31%.
Downloaded from http://circ.ahajournals.org/ by guest on June 17, 2017
Effects on Plasma Triglyceride Levels
Lovastatin caused a significant (p < 0.02) decrease in plasma triglycerides, averaging an 18%
reduction. Mean plasma triglycerides were not
different on probucol alone but fell during period
four when lovastatin was again added (p <0.01).
The addition of colestipol resulted in a slight rise
in plasma triglycerides, which partly negated the
lowering effect of lovastatin. In the five subjects
who took only lovastatin (80 mg/day), at the end
of the study, plasma triglycerides were significantly reduced again (p<0.05).
Effects on LDL Levels
Lovastatin (40 mg/day) was extremely effective
in lowering LDL levels, producing an average
decrease of 36% in these severely hypercholesterolemic patients. All 17 patients experienced reductions in LDL that ranged from 15% to 51%. In
contrast, probucol alone was not nearly as effective, producing an average decrease of only 14%
(range, + 11% to -32%). In contrast to the uniform
response to lovastatin, the response to probucol
was quite variable (Table 3). For example, nine
21
subjects had LDL decreases more than 15%, three
subjects had decreases of 6-15%, and five subjects
had no change or even an increase in mean LDL
cholesterol values. When lovastatin and probucol
were combined, there was no synergistic or additive
effect observed, and the lowering of LDL could be
solely attributable to that produced by lovastatin
alone. The addition of colestipol to lovastatin plus
probucol therapy produced a substantial further
decrease in LDL levels (- 23%), and during this
period, absolute LDL cholesterol levels were
decreased an average 52% below baseline values.
On this combination, all 17 patients experienced
marked decreases in LDL, ranging from 40% to 70%
of the average values obtained during the control
period. When probucol was discontinued, the mean
LDL cholesterol level remained almost as low as
when on triple-drug therapy. Thirteen of the 17
patients were also given lovastatin (80 mg/day) while
being maintained on colestipol. The mean LDL of
these patients was 56% below their baseline value.
The five individuals who took only lovastatin (80 mg/
day) had LDL levels 40% below their control values.
Effects on HDL Levels
Lovastatin therapy alone produced a small but
consistent increase in HDL cholesterol levels that
averaged 6%. In striking contrast, probucol therapy alone produced a 29% decrease in HDL
cholesterol levels. When lovastatin was added to
the probucol therapy, during period four, this
reduction in HDL was partially compensated for,
but HDL levels were still nearly 20% below those
TABLE 3. Average LDL Cholesterol Levels Obtained During Study Periods One Through Five for Individual Subjects
Lovastatin plus
probucol plus
Lovastatin plus
probucol
Probucol
colestipol
Lovastatin
Control
(mg/dl)
(mg/dl)
(mg/dl)
Patient
(mg/dl)
(mg/dl)
182
261
322
297
1
438
97
136
228
137
2
205
126
224
329
236
3
408
124
143
167
4
244
171
135
220
306
194
5
308
136
163
188
258
154
6
183
244
341
240
378
7
117
175
249
344
218
8
280
...
357
275
9
426
173
203
300
198
406
10
108
140
238
146
218
11
161
198
242
215
254
12
139
157
166
136
200
13
288
332
312
500
460
14
104
137
196
210
145
15
238
299
418
303
504
16
184
...
220
156
256
17
LDL, low-density lipoprotein.
Shown is the average LDL cholesterol level obtained during each of the first five study periods for each individual
subject. For each individual, the value shown is the average LDL cholesterol obtained after 2 and 3 months of the
indicated therapy.
22
Circulation Vol 79, No 1, January 1989
Downloaded from http://circ.ahajournals.org/ by guest on June 17, 2017
TABLE 4. Average Drug Effects and Drug Interactions: Results of Multiple Regression Analysis
Lipid level (% change)
Total cholesterol
LDL cholesterol
HDL cholesterol
-29
-36
Lovastatin alone (40 mg/day)
+6
Probucol alone
-14
-15
-29
Addition of
Lovastatin to probucol
-22
-26
+13
Probucol to lovastatin
-6
(- 1)
-24
-17
Colestipol to lovastatin (40 mg/day)
-23
-6
-16
Colestipol to lovastatin (40 mg/day) plus probucol
-23
(-+2)
Probucol to lovastatin (40 mg/day) plus colestipol
-5*
-18
(0)
Lovastatin (40 mg/day) to lovastatin (40 mg/day)
= Lovastatin (80 mg/day)
-7
+6
(-3)
Lovastatin (40 mg/day) to lovastatin (40 mg/day)
plus colestipol
-7
(-3)
+6
-17
-23
Colestipol to lovastatin (80 mg/day)
-6
LDL, low-density lipoprotein; HDL, high-density lipoprotein.
This table gives the average drug effect on total, LDL, and HDL cholesterol levels for individual drugs and for addition of one drug
to individuals already taking one or two agents. These results are derived from a multiple regression model using log transformed data
as explained in the "Appendix." Each value represents the percent change (positive or negative) of the lipid or lipoprotein level
produced by a given drug. For example, lovastatin alone produced a 36% decrease in LDL levels in the average subject. When added
to a subject already on probucol (addition of lovastatin to probucol), it produced a further 26% decrease from LDL levels obtained on
probucol. It was not possible to calculate the effects of colestipol alone from this model, but previous results from our laboratory and
others predict a lowering of LDL cholesterol of 20-35%.7-10
All effects are significant atp<0.01 except (p)=NS, *p<0.05.
obtained at baseline. The addition of colestipol
resulted in no further change in HDL. However,
during period six, when probucol was discontinued,
HDL levels returned to control levels (4-6 months
after discontinuation of the probucol). In the 13
subjects who subsequently took 80 mg/day lovastatin plus colestipol, HDL levels again increased 6%
(compared with control values).
Reductions in Xanthomas
Nine of 17 patients had marked xanthomas, consisting of xanthelasma or tendon xanthoma or both.
Although we did not document xanthoma size by
means of xerography, in at least five individuals
there was dramatic and visible evidence of regression of xanthomata. In each case, the regression was
first observed by the patient during the fourth period
when lovastatin and probucol were being given
together. Continual regression occurred throughout
the remainder of the study. For example, subject 1,
who had extensive tendon xanthomata, had dramatic
reduction of xanthoma on his hands. This finding
was of considerable interest because no reduction
had been observed previously despite years of combined colestipol and nicotinic acid therapy.
Average Drug Effects and Drug Interactions
In an attempt to quantify an average effect on
lipoproteins produced by each drug as well as the
interaction between the different drugs, we have
analyzed the data with a linear regression model (as
described in detail in the "Appendix"). This analysis allows us to estimate the average effect on each
lipoprotein for each drug used individually and in
combination with another drug. The results are
presented in Table 4. For example, it can be seen
that lovastatin alone (40 mg/day) produced an average decrease in LDL cholesterol of 36%, whereas
probucol produced an average decrease of only
14%. The addition of lovastatin to individuals already
on probucol is predicted to lead to a further decrease
in LDL cholesterol levels of 26%; however, in the
converse situation, the addition of probucol to
subjects already on lovastatin (40 mg/day) would
produce no further decrease in LDL levels (though
total cholesterol levels would fall because of a
decrease in the HDL cholesterol levels). The model
suggests that for individuals taking lovastatin (40
mg/day), the addition of probucol is not likely to
produce further lowering of LDL levels. However,
the model predicts that the addition of colestipol to
lovastatin (40 or 80 mg/day) would cause a further
23% decrease in LDL cholesterol from levels
achieved on lovastatin alone. This magnitude of
lowering (- 23%) is similar to what literature values
predict for effects of colestipol alone.7-10 In patients
on a regimen of lovastatin (40 mg/day), increasing
the dose of lovastatin to 80 mg/day would have the
effect of lowering LDL cholesterol an additional
7%. Similarly, in subjects on lovastatin (40 mg/day)
plus colestipol, increasing lovastatin to 80 mg/day
also would add a further 7% lowering of LDL
cholesterol.
Regarding HDL cholesterol, lovastatin alone produced an average increase in HDL cholesterol of
6%, whereas probucol alone produced a 29%
decrease. The addition of lovastatin to subjects
already on probucol would be predicted to partially
reverse this decrease in HDL cholesterol levels.
Side Effects
As noted earlier, no medically important side
effects were noted with any of the medications or
Witztum et al Drug Therapy of Familial Hypercholesterolemia
Total Cholesterol
23
LDL-Cholesterol
600
500
FIGURE 5. Plot of maximal response of total and
low-density lipoprotein (LDL) cholesterol during
the combination drug therapy program. This figure
shows the control level of total and LDL cholesterol level for each individual duping the control
period and the corresponding lowest mean level
obtained during one of the drug perods. Thus,
each bar represents the maximally effective therapy
that was achieved. The mean level of total cholesterol was 404 ± 98 mg/dl before and 222 ±15 mg/dl
on therapy, for a mean decrease of44%. The mean
level of LDL cholesterol was 325 ± 102 mg/dl before
and 149±53 mg/dl during therapy, for a mean
decrease of 53%.
400
1-
300
E
200
Downloaded from http://circ.ahajournals.org/ by guest on June 17, 2017
100
0
Combination
Therapy
Combinatior
Therapy
combinations that necessitated discontinuation of
hypolipidemic drug therapy. In four patients, persistent mild liver function abnormalities (transaminase elevations <1.5-2 times that of normal)
were noted during combined drug therapy; in all
four cases, the patients were noted to consume
alcohol, in moderation, on a regular basis. For
example, subject 16 had consumed two to three
glasses of wine a day for many years and had
normal transaminase levels during periods one to
three. During period four, mild abnormalities in
SGPT and SGOT (values, 1.5-2 times normal) were
noted. When she reduced her alcohol content to
only one glass of wine a day, liver functions promptly
returned to normal despite continuation of lovastatin and other study medications. Similarly, in three
other individuals, similar mild elevations of liver
transaminases were associated with alcohol consumption because discontinuation of alcohol led to
the prompt return of liver function tests to normal
despite continuation of lovastatin and other medications. One subject developed alkaline phosphatase level 1.2 times that of normal at the end of
period two, which persisted throughout the remainder of the study. One other subject had random,
isolated elevations of transaminases that were always
less than twice that of normal. No other consistent
blood chemistry abnormalities could be attributable
to drug therapy in any of the other patients. Subject
17 reported the occurrence of a persistent mild
headache during the periods that probucol was
consumed. Within 2 days after discontinuing probu-
col during period six, the headache abated and was
not reported again during the subsequent 6-month
period of follow-up. The five individuals who discontinued colestipol did so because of gastrointestinal side effects.
Best corrected visual acuity decreased in two eyes
of two individuals from 20/20 to 20/25. One eye of
one individual was noted to have an increase in
visual acuity from 20/50 to 20/25. No changes were
noted in color vision, fundoscopy, or visual field
evaluation. Six of the patients had changes noted
with slit-lamp biomicroscopy and external examination. Three of these had slight progression of corneal
arcus in both eyes, and one patient had a new
small posterior subcapsular opacity in one eye.
Two eyes of two individuals had pigment on the
anterior capsule of the lens, and one patient had
early diffuse cortical opacities in both eyes after
two years. Initially, 11 patients were found to have
corneal arcus, and two had xanthelasma. There
was no regression in corneal arcus but xanthelasma disappeared in one patient.
Response of Individual Patients
The mean data presented above have been derived
from all 17 patients. However, in treating individual
patients, it is important to appreciate the range of
responses that can be expected. Table 3 presents
the average LDL cholesterol values obtained during
the first five periods for each study subject, and
Figure 5 displays the response of each individual
patient to the optimal combination drug therapy,
24
Circulation Vol 79, No 1, January 1989
Downloaded from http://circ.ahajournals.org/ by guest on June 17, 2017
which represents the lowest average total or LDL
cholesterol level obtained during periods five, six,
or seven. As can be appreciated from Figure 5,
during the optimal form of therapy all 17 patients
had substantial decreases in their LDL cholesterol
levels, averaging 53%, with a range from 35% to 70%
below baseline. Fifteen of the 17 patients experienced LDL cholesterol levels on combination therapy that were below the 95th percentile, and 11
patients achieved LDL cholesterol levels below the
50th percentile for age- and sex-matched controls. In
six subjects, the lowest LDL was obtained on tripledrug therapy, whereas seven subjects had their lowest on lovastatin (80 mg/day) plus colestipol.
During the period in which the lowest total cholesterol level was achieved, a mean decrease of 44%
occurred, with a range of 31% to 60%. Eleven
subjects had their lowest total cholesterol level
during the triple-drug therapy (reflecting the decrease
in HDL levels as well), whereas only four subjects
had their lowest value on lovastatin (80 mg/day)
plus colestipol. In all, nine subjects achieved total
cholesterol levels below 200 mg/dl.
In general, the data presented above indicated
that the combination of lovastatin plus colestipol
was as effective as the combination of lovastatin,
colestipol, and probucol for most individuals. However, there was individual variation, and in a few
patients (i.e., subjects 1, 4, and 8) probucol did
appear to exert a synergistic effect with other medications in lowering LDL levels.
Discussion
Several previous reports have described successful combinations of hypolipidemic agents that yielded
reductions of LDL cholesterol of 35-55%. These
combinations have usually included a bile sequestrant (either colestipol or cholestyramine) together
with nicotinic acid,33-35probucol,36'37 or neomycin.38
More recently, bile sequestrants have been combined with an inhibitor of cholesterol biosynthesis,
compactin, or lovastatin.39-41 There are only rare
reports of successful regimens that used combinations of hypolipidemic agents that did not include a
bile sequestrant.42
In the present study, we sought to test a therapeutic regimen that would allow an effective lowering of LDL levels without the necessity of using a
bile sequestrant. Although these agents are highly
effective in lowering LDL levels and in reducing the
subsequent rate of CAD,43 their use is associated
with frequent side effects and problems with longterm patient acceptance and compliance.29'43
Because lovastatin and probucol theoretically work
by different mechanisms, the hypolipidemic effect
of the two used together should be additive, and in
a preliminary study of several FH patients, this
appeared to be true.29 Therefore, we designed this
long-term prospective study involving 17 FH subjects. Although a synergistic effect of lovastatin and
probucol was seen in an occasional patient, for the
vast majority of patients the addition of probucol
added little to the therapeutic regimen, as judged by
lowering of LDL levels. In contrast, the addition of
colestipol to subjects on lovastatin (either 40 or 80
mg/day) led to a further 23% reduction in LDL
levels. Because previous reports have suggested
that, at the dose used, bile sequestrants alone lower
LDL levels 20% to 25%,7-10 our data suggest that
the bile sequestrants and lovastatin are working by
independent, or additive, mechanisms to lower LDL
levels. It is also of interest that the further addition
of lovastatin (40 mg/day) to a regimen of lovastatin
(40 mg/day) yielded only a further 7% lowering in
LDL levels. Addition of lovastatin (40 mg/day) to
someone already taking colestipol plus lovastatin
(40 mg/day) increased LDL lowering only 7% further, also suggesting that bile sequestrants and
lovastatin lower LDL by somewhat independent
modes.
Lovastatin increased HDL levels 6%, consistent
with previous reports.10"14-16 However, probucol
produced a profound depression in HDL levels,
averaging 29%, an effect that persisted despite the
simultaneous administration of lovastatin. The effect
of probucol persisted long after discontinuation of
the medication, and only after discontinuation of
therapy for 6 months did the HDL levels become
elevated again, reflecting the lovastatin effect.
Although conventional wisdom would suggest that
this profound lowering of HDL cholesterol levels is
detrimental, data from Japan indicate that probucolinduced regression of xanthoma size in patients
with homozygous FH correlates best with reduction
in HDL cholesterol levels.44 Our laboratory45,46 and
others47 have suggested that oxidative modification
of LDL increases its uptake in macrophages and
can lead to foam cell formation. Recent observations from this laboratory indicate that probucol is a
potent lipid-soluble antioxidant and that LDL isolated from patients treated with conventional doses
of probucol is protected from oxidative modification in vitro.48 Furthermore, Carew et a149 and Kita
et a150 have shown that probucol inhibits atherosclerosis in hypercholesterolemic rabbits despite any
substantial reduction in plasma cholesterol levels.
However, only further experimentation and results
of clinical trials will be able to provide guidelines as
to its potential as an antiatherosclerotic agent.
Of considerable interest was that nine of 17
patients in this study had marked xanthomas consisting of xanthelasma or tendon xanthomas or
both. Although we did not document xanthoma size
in any quantitative manner, in at least five individuals there was dramatic and visible evidence of
regression of xanthomata. This was particularly
striking in several subjects who had been followed
in this clinic for many years on combinations of a
bile sequestrant and nicotinic acid and in whom no
evidence for xanthoma regression had occurred.
Despite the fact that at the beginning of the study
eight of these subjects had clinical manifestations of
Witztum et al Drug Therapy of Familial Hypercholesterolemia
Downloaded from http://circ.ahajournals.org/ by guest on June 17, 2017
CAD or peripheral vascular disease, in no individual did worsening of disease occur over the 25
months of study, nor did evidence for CAD present
in any of the other nine subjects.
No subjects experienced any serious medical side
effect during this combination drug study, and in no
case was any hypolipidemic medication discontinued because of any serious adverse reaction.
Through periodic complete ophthalmologic examinations during the 2-year period, no visionthreatening side effects were observed. We found
no change in visual acuity, color vision, fundus
examination, or visual field test. Only minimal
changes were detected by slit-lamp biomicroscopy
and external examination. Of importance, no patient
developed a new lens opacity causing loss of visual
acuity. However, slit-lamp biomicroscopy grading
of lens opacities is a subjective examination, and
minor lens opacities may not be detected with this
technique.
The goal of therapy was to develop a welltolerated combination drug regimen for high-risk
patients that would lower total cholesterol levels to
values below 200 mg/dl and LDL cholesterol levels
to values below 130 mg/dl. Our data indicate that
lovastatin (40 mg/day) plus colestipol lowered LDL
levels 52%, and lovastatin (80 mg/day) plus colestipol decreased LDL levels 56%. These values are
consistent with other reports of the efficacy of
lovastatin and a bile sequestrant.39-41 Although substantial reductions in cholesterol and LDL values
were achieved, for many of our subjects total and
LDL cholesterol values remained above the goals
listed above. Malloy and colleagues51 recently
reported that the combination of lovastatin, colestipol, and nicotinic acid resulted in a 67% decrease in
LDL cholesterol levels in 21 FH subjects. In their
study, a mean control LDL cholesterol level of 326
mg/dl was reduced to 112 mg/dl.51 However, caution
should be exercised in using lovastatin and nicotinic
acid together as this apparently increases the possibility of lovastatin-induced myopathy.52
In a recent symposium, we summarized a number
of reports in which several different combinations of
two hypolipidemic drugs were used to produce LDL
lowering of 40-60%.29 Our current report and that of
Malloy and colleagues51 are among the first to describe
the use of three agents to treat patients with FH. The
use of probucol as a third agent, together with
lovastatin and colestipol, appears to add little as a
hypolipidemic agent for the average patient. Whether
probucol's antioxidant property will eventually advocate its use as an antiatherosclerotic drug awaits
further experimentation and clinical trials.
Acknowledgments
We wish to acknowledge the dedication and patience
of the subjects who participated in this study. Lovastatin (and placebo) were supplied by Dr. Jonathan
Tobert of Merck, Sharp and Dohme Research Labs;
probucol (and placebo) were supplied by Dr. Ralph
25
Tedeschi of Merrell Dow Pharmaceuticals; and colestipol hydrochloride was supplied by Dr. Nicholas
Andreadis of the Upjohn Company.
Appendix
Data were analyzed with multivariate, multiplelinear regressions performed with the 6R program.32
These regressions were performed on log-transformed values of plasma cholesterol, LDL cholesterol, and HDL cholesterol, and the residuals were
verified as being normally distributed. Similar analyses on untransformed data did not produce normally distributed residuals. Plasma triglycerides were
not considered in this analysis as preliminary analysis indicated that only lovastatin caused significant
changes in this variable, and the effect was relatively small. Fixed (constant) effects estimated from
the regression represent proportional changes in
plasma concentrations. Percent changes for a sum
of multiple effects were calculated with the formula
% change=100x(10s-1)
where S = sum of effects. All effects that could be
estimated with the current data were included in the
model. For each subject, only those periods for
which multiple observations were available were
analyzed.
The complete statistical model initially considered was
Yijk= + OiXi + 9Xg + (g )ixZxi +
y
wrX7r+ (wX,)iX,i +
KXK + (K0)iX,Xi + AXA + (AO)iXAXi +
(g,p)X,,X,,+ (gr6)iXxX,7xi +
(1K)XiX (gKo)iXIXKXi
+
+
(ITK)X,XK+ (TK6)iXIXKXA
+
(AK)XAXK+ (AKO)iXAXKXi +
(1rK)XSX,XK + (97K0)iXXXKXA + Eijk
where Yijk is the kth replicate for subject i during
period j. The symbols -y, Oi, (JU7p), (g7wKO)i, and so
on represent the various effects to be estimated. X
represents "independent" variables used to control
which effects are included in predicting each observation, and Eijk iS the residual error. An X has a
value of 1 if the condition indicated by its subscript
is satisfied (i.e., the medication indicated was actually taken). Otherwise, it has a value of 0. For
example, X, is set to 1 for all data from periods
during which the subject was taking lovastatin.
Similarly, X2 is set to 1 for all data for subject 2. The
nomenclature used for effects is as follows. Effect y
is the mean of all control (period one) values. The
symbol theta (0) refers to subject. Effects dependent on the subject always have an i subscript and
may be considered samples from a population of
random effects. The symbols ,, 7T, K, and A refer to
40 mg/day lovastatin, 1 g/day probucol, 20 g/day
colestipol and 80 mg/day lovastatin, respectively.
Because one cannot take 80 mg lovastatin without
taking 40 mg lovastatin, X, can only be 1 if Xg is
/,
26
Circulation Vol 79, No 1, January 1989
Downloaded from http://circ.ahajournals.org/ by guest on June 17, 2017
also 1. Thus, effects that include A must be interpreted as resulting from a second 40 mg/day dosage
of the drug and were always estimated as additive to
the effects of the first 40 mg/day. Effects with
parentheses are interactions of the variables enclosed
and are included if more than one condition is met.
In practice, additional independent variables must
be defined for each of the interaction terms, but
these are easily calculated as products of singly
subscripted variables.
Some interaction effects such as the interaction of
probucol and 80 mg/day lovastatin are not included
in the model because these combinations were
never administered. Some of the individual terms in
the above model could not be estimated with the
current data and were only calculated in conjunction with other effects. For example, because colestipol was never administered in the absence of lovastatin, the effect of colestipol alone could not be
estimated and was always confounded by its interaction with lovastatin. Thus, only the sums of
effects K+ (gK) and (K0)i + (gKO)i could be estimated.
Similarly, the interaction of colestipol and probucol
could not be separated from a possible three-way
interaction among lovastatin, probucol, and colestipol. Again, only the sums (7rK) + (PL7K) and (7K0)i +
(gunrKG)i could be estimated.
Because the 13 subjects with a period seven (80
mg/day lovastatin plus colestipol) were different
individuals than the five subjects with a period eight
(80 mg/day lovastatin only), comparisons of these
two periods must be considered an unpaired test.
This test may be interpreted as a test for an interaction between the second 40 mg/day lovastatin and
colestipol. This test showed no significance, and it
was concluded that the effect of adding colestipol to
a lovastatin regimen does not depend on whether
the subject is taking 40 or 80 mg lovastatin.
For subject 13, who had a period five but no
period six, the combination of effects (K0)13 + (gK.)13
+ (7K0)13 + (/7K0)13 had to be estimated rather than
the two combinations (K0)13 + (AKG)13 and (WK0)13 +
(,wK0)13, as was done for the other subjects. Subject 13 was therefore excluded during tests of significance on interactions between subject and colestipol and among subject, colestipol, and probucol.
The significance of each of the effects in the
above model was tested by regressing all data with
and without the effect under consideration. Results
were compared with Fisher's F test and parameters
removed from the model in a stepwise manner if
p >0.05 was obtained for all dependent variables.
Effects were removed in the following order: 1)
(A6)i, interaction between subject and 80 mg/day
lovastatin; 2) (,uO)i, interaction between subject and
40 mg/day lovastatin; and 3) (AK)j, interaction
between colestipol and 80 mg/day lovastatin. All
other effects were found to be significant for at least
one of the dependent variables. The lack of significance of the above three effects may be interpreted
to indicate that all subjects reacted similarly to the
addition of 40 mg/day lovastatin when already on 40
mg lovastatin, all subjects reacted similarly to the
first 40 mg/day lovastatin, and the effect of the
second 40 mg/day lovastatin did not depend on
whether the subject was also taking colestipol.
The effects estimated from all data using the final
model were then used to estimate the effects of
adding a drug to another regimen. For example, the
effect of adding probucol to lovastatin was estimated as Tr+ (,rr) while the effect of adding lovastatin to probucol as g + (b-r). The statistical significance of the addition of a drug was tested by
regressing the data with the appropriate sum of
effects constrained to be 0. Results were compared
to the full model with Fisher's F test. These data
are shown in Table 4.
For each lipoprotein variable measured, a group
mean percent change from the control period was
calculated for each drug period with a paired Student's t test of log-transformed data. These data are
shown in Figures 2-4.
References
1. Stone NJ, Levy RI, Fredrickson DS, Verler J: Coronary
artery disease in 116 kindred with familial type II hyperli-
poproteinemia. Circulation 1974;49:476-488
2. Wagner WD, St Clair RW, Clarkson TB, Connor JR: A
study of atherosclerosis regression in Macaca mulatta: 3.
Chemical changes in arteries from animals with atherosclerosis induced for nineteen months and regressed for fortyeight months at plasma-cholesterol concentrations of 300 or
200 mg/dl. Am J Pathol 1980;100:633-650
3. The Lipid Research Clinics Primary Prevention Trial Results
II: The relationship of reduction in incidence of coronary
heart disease to cholesterol lowering. JAMA 1984;251:365 -374
4. Blankenhorn DH, Nessim SA, Johnson RL, Sanmarco ME,
Azen SP, Cashin-Hemphill L: Beneficial effects of combined
colestipol-niacin therapy on coronary atherosclerosis and
coronary venous bypass grafts. JAA4 1987;257:3233-3240
5. Goldstein JL, Brown MS: Familial hypercholesterolemia, in
Stanbury JB, Wyngaarden JB, Fredrickson DS, Goldstein
JL, Brown MS (eds): Metabolic Basis of Inherited Disease.
New York, McGraw Hill, 1983, pp 672-712
6. Steinberg D, Witztum JL: Risk assessment and management
in daily practice, in Steinberg D, Olfesky JM (eds): Hypercholesterolemia and Atherosclerosis. New York, Churchill
Livingstone, 1987, pp 209-228
7. Witztum JL, Schonfeld G, Weidman SW: The effects of
colestipol on the metabolism of very low density lipoproteins
in man. J Lab Clin Med 1987;88:1008-1018
8. Witztum JL, Schonfeld G, Weidman SW, Giese WE, Dillingham MA: Bile sequestrant therapy alters the composition
of low density lipoproteins. Metabolism 1979;28:221-229
9. Kane JP, Malloy MJ, Tun P, Phillips NR, Freedman DD,
Williams M, Rowe JS, Havel RJ: Normalization of low
density lipoprotein levels in heterozygous familial hypercholesterolemia with a combined drug regimen. N Engl J Med
1981;304:251-258
10. Dujovne CA, Feldman EB, Frost PH, Gotto AM, Hunninghake DR, Illingworth DR, Knopp RH, Lees RS, Miller VT,
Samuel P, Stein EP, Stone NJ, Tobert JA, Zupkis RV:
Comparison of lovastatin and cholestyramine for the treatment of primary hypercholesterolemia. Arteriosclerosis 1987;
7:517A
11. Shepherd J, Packard CJ, Patsch JR, Gotto AM, Taunton
OD: Effects of nicotinic acid therapy on plasma high density
lipoprotein subfraction distribution and composition and on
apolipoprotein A metabolism. J Clin Invest 1979;63:858-867
Witztum et al Drug Therapy of Familial Hypercholesterolemia
27
Downloaded from http://circ.ahajournals.org/ by guest on June 17, 2017
12. Knopp RH, Ginsberg J, Albers JJ, Hoff C, Ogilvie JT,
Warnick GR, Burrows E, Retzlaff B, Poole M: Contrasting
effects of unmodified and time-release forms of niacin on
lipoproteins in hyperlipidemic subjects: Clues to mechanism
of action of niacin. Metabolism 1985;34:642-650
13. Glueck CJ: Colestipol and probucol: Treatment of primary
and familial hypercholesterolemia and amelioration of atherosclerosis. Ann Intern Med 1982;96:475-482
14. Tobert JA, Bell GD, Birtwell J, Kukovetz WR, Pryor JS,
Butiny A, Holmes IB, Chao Y-S, Bolognese JA: Cholesterollowering effect of mevinolin, an inhibitor of 3-hydroxy3-methylglutaryl coenzyme A reductase in healthy volunteers. J Clin Invest 1982;69:913-919
15. Illingworth DR, Sexton GJ: Hypocholesterolemic effects of
mevinolin in patients with heterozygous familial hypercholesterolemia. J Clin Invest 1984;74:1972-1978
16. Havel RJ, Hunninghake DB, Illingworth DR, Lees RS, Stein
EA, Tobert JA, Bacon SR, Bolognese JA, Frost PH, Lamkin
GE, Lees AM, Leon AS, Gardner K, Johnson G, Mellies
MJ, Rhymer PA, Tun P: Lovastatin (mevinolin) in the
treatment of heterozygous familial hypercholesterolemia.
Ann Intem Med 1987;107:609-615
17. Hoeg JM, Maher MB, Zech LA, Bailey KR, Gregg RE,
Lackner KJ, Fojo SS, Anchors MA, Bojanovski M, Sprecher DL, Brewer HB: Effectiveness of mevinolin on plasma
lipoprotein concentrations in type II hyperlipoproteinemia.
Am J Cardiol 1986;57:933-939
18. The Lovastatin Study Group II: Therapeutic response to
lovastatin (mevinolin) in nonfamilial hypercholesterolemia.
29. Witztum JL: Intensive drug therapy of hypercholesterolemia. Am Heart J 1987;113:603-609
30. Sasaki J, Tanabe Y, Saku K, Takii M, Arakawa K: Alterations in lipoprotein with discontinuation of probucol. Curr
Therap Res 1987;42:640-645
31. Lipid Research Clinics Program: Manual of Laboratory
Operations, Volume 1, Edition 2: Lipid and Lipoprotein
Analysis. US Dept of Health, Education, and Welfare publication No. (NIH) 76-628. Washington, DC, US Govern-
JAMA4 1986;256:2829-2834
19. Kovanen P, Bilheimer DW, Goldstein JL, Jaramillo JJ,
Brown MS: Regulatory role for hepatic low density lipoprotein receptors in vivo in the dog. Proc Natl Acad Sci USA
1981;78:1194-1198
20. Shepherd J, Packard CJ, Bicker S, Laurie TDV, Morgan
HG: Cholestyramine promotes receptor-mediated lowdensity lipoprotein catabolism. N Engl J Med 1980;
302:1219-1222
21. Witztum JL, Young SG, Elam RL, Carew TE, Fisher M:
Cholestyramine-induced changes in low density lipoprotein
composition and metabolism. I. Studies in the guinea pig. J
Lipid Res 1985;26:92-103
22. Alberts AW, Chen J, Kuron G, Hunt V, Huff J, Hoffman C,
Rothrock J, Lopez M, Joshua H, Harris E, Patchett A,
Monaghan R, Currie S, Stapley E, Albers-Schonberg G,
Hensens 0, Hirshfield J, Hoogsteen K, Liesch J, Springer J:
Mevinolin: A highly potent competitive inhibitor of hydroxymethylglutaryl-coenzyme A reductase and a cholesterol lowering agent. Proc NatlAcad Sci USA 1980;77:3957-3961
23. Bilheimer DW, Grundy SM, Brown MD, Goldstein JL:
Mevinolin and colestipol stimulate receptor-mediated clearance of low density lipoprotein from plasma in familial
hypercholesterolemia heterozygotes. Proc Natl Acad Sci
USA 1983;80:4124-4128
24. Grundy SM, Vega GL: Influence of mevinolin on metabolism of low density lipoproteins in primary moderate hypercholesterolemia. J Lipid Res 1985;26:1464-1475
25. Naruszewicz M, Crew TE, Pittman RC, Witztum JL, Steinberg D: A novel mechanism by which probucol lowers low
density lipoprotein levels demonstrated in the LDL receptordeficient rabbit. J Lipid Res 1984;25:1206-1213
26. Baker SG, Joffe BI, Mendelsohn D, Seftel HC: Treatment of
homozygous familial hypercholesterolemia with probucol. S
Afr Med J 1982;62:7-11
27. Yamamoto AY, Matsuzawa B, Kishino R, Hayashi R,
Hirobe K, Kikkawa T: Effects of probucol on homozygous
cases of familial hypercholesterolemia. Atherosclerosis 1983;
38. Miettinen TA: Effect of neomycin alone and in combination
with cholestyramine on serum cholesterol and fecal steroids
in hypercholesterolemic subjects. J Clin Invest 1979;
64:1485-1493
39. Mabuchi H, Sakai T, Sakai Y, Yoshimura A, Watanabe A,
Wakasusi T, Koizumi J, Takedo R: Reduction of serum
cholesterol in heterozygous patients with familial hypercholesterolemia. N Engl J Med 1983;309:609-613
40. Illingworth DR: Mevinolin plus cholestyramine therapy for
severe heterozygous familial hypercholesterolemia. Ann
Intern Med 1984;101:598-604
41. Grundy SM, Vega GL, Bilheimer DW: Influence of combined therapy with mevinolin and interruption of bile-acid
reabsorption of low density lipoproteins in heterozygous
familial hypercholesterolemia. Ann Intern Med 1985;
103:339-343
42. Hoeg JM, Maher MM, Bov E, Zech LA, Bailey KR, Gress
RE, Sprecher DL, Susser JK, Pikus AM, Brewer HB:
Normalization of plasma lipoproteins with neomycin and
niacin in type II hyperlipoproteinemia. Circulation 1984;
70:1004-1011
43. Lipid Research Clinics Program: The Lipid Research Clinics
Coronary Primary Prevention Trial Results. JAMA 1984;
251:351-364
44. Yamamoto A, Matsuuzawa Y, Yokoyama S, Funahashi T,
Yamamura T, Bun-Ichiro K: Effects of probucol on xanthomata regression in familial hypercholesterolemia. Am J
Cardiol 1986;27:29H-35H
45. Henriksen T, Mahoney EM, Steinberg D: Enhanced macrophage degradation of low density lipoproteins previously
incubated with cultured endothelial cells: Recognition by the
receptors for acetylated low density lipoproteins. Proc Natl
Acad Sci USA 1981;78:6495-6503
46. Steinbrecher UP, Parthasarathy S, Leake DS, Witztum JL,
Steinberg D: Modification of low density lipoprotein by
endothelial cells involves lipid peroxidation and degradation
of low density lipoprotein phospholipids. Proc NatlAcad Sci
USA 1984;81:3883-3887
47. Heinecke JW, Baker L, Rosen H, Chait A: Superoxidemediated modification of low density lipoprotein by arterial
smooth muscle cells. J Clin Invest 1986;77:757-761
48. Parthasarathy S, Young SG, Witztum JL, Pittman RC,
Steinberg D: Probucol inhibits oxidative modification of low
density lipoprotein. J Clin Invest 1986;77:641-644
48:157-166
28. Lipid Research Clinics Program: The Lipid Research Clinics
Population Studies Data Book, Volume 1: The Prevalence
Study. US Dept of Health, Education, and Welfare publication No. (NIH) 80-1527. Washington, DC, US Government
Printing Office
ment
Printing Office
32. Dixon WJ, (ed): BMDP Statistical Software. Berkeley,
Calif, University of California Press, 1985
33. Packard CJ, Steward JM, Morgan HG, Lorimer AR, Shepherd J: Combined drug therapy for familial hypercholesterolemia. Artery 1980;7:281-289
34. Illingworth DR, Rapp JR, Phillipson BE, Connor WE:
Colestipol plus nicotinic acid in treatment of heterozygous
familial hypercholesterolaemia. Lancet 1981;1:296-298
35. Kuo PT, Kostis JB, Moreyra AE, Hayes, JA: Familial type
II hyperlipoproteinemia with coronary heart disease: Effect
of diet-colestipol-nicotinic acid treatment. Chest 1981;
79:286-291
36. Dujovne CA, Krehbiel P, Decoursey S, Jackson B, Chernoff
SB, Pitterman A, Garty M: Probucol with colestipol in the
treatment of hypercholesterolemia. Ann Intern Med 1984;
100:477-482
37. Schonfeld G, Witztum JL, Basich P: Probucol further lowers
the serum cholesterol of colestipol treated patients with
hypercholesterolemia. Artery 1982;10:99-104
28
Circulation Vol 79, No 1, January 1989
49. Carew TE, Schwenke DC, Steinberg D: Antiatherogenic
effect of probucol unrelated to its hypocholesterolemic effect:
Evidence that antioxidants in vivo can selectively inhibit low
density lipoprotein degradation in macrophage-rich fatty
streaks and slow the progression of atherosclerosis in the
Watanabe heritable hyperlipidemic rabbit. Proc Natl Acad
Sci USA 1987;84:7725-7729
50. Kita T, Nagano Y, Yokode M, Ishi K, Kume N, Ooshima A,
Yoshida H, Kawai C: Probucol prevents the progression of
atherosclerosis in Watanabe heritable hyperlipidemic rabbit,
an animal model for familial hypercholesterolemia. Proc
NatlAcad Sci USA 1987;84:5928-5931
51. Malloy MJ, Kane JP, Kunitake ST, Tun P: Complementarity
of colestipol, niacin, and lovastatin in treatment of severe
familial hypercholesterolemia. Ann Intern Med 1987;
107:616-623
52. Grundy SM: Drug therapy: HMG CoA reductase inhibitors
for treatment of hypercholesterolemia. N Engl J Med 1988;
319:24-32
KEY WORDS * familial hypercholesterolemia * lovastatin
probucol * colestipol hydrochloride * cholesterol
.
Downloaded from http://circ.ahajournals.org/ by guest on June 17, 2017
Intensive combination drug therapy of familial hypercholesterolemia with lovastatin,
probucol, and colestipol hydrochloride.
J L Witztum, D Simmons, D Steinberg, W F Beltz, R Weinreb, S G Young, P Lester, N Kelly
and J Juliano
Downloaded from http://circ.ahajournals.org/ by guest on June 17, 2017
Circulation. 1989;79:16-28
doi: 10.1161/01.CIR.79.1.16
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1989 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is located on
the World Wide Web at:
http://circ.ahajournals.org/content/79/1/16
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally
published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further
information about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation is online at:
http://circ.ahajournals.org//subscriptions/