Treatment of refractory familial hypercholesterolemia by low-density lipoprotein apheresis using an automated dextran sulfate cellulose adsorption system

PREVENTWE CARDlOLO8Y

Treatment of Refractory Familial Hypercholesterolemia by Low-Density Lipoprotein Apheresis Using an Autom&d Dextran Sulfate Cellulose Adsorption System Bruce R. Gordon, MD, Sheryl F. Kelsey, PhD, David W. Bilheimer, MD, David C. Brown, MD, Peter C. Dau, MD, Antonio M. Gotto, Jr., MD, D. Roger Illingworth, MD, PhD, Peter H. Jones, MD, Susan F. Leitman, MD, James S. Prihoda, MD, Evan A. Stein, MD, PhD, Thomas N. Stern, MD, James H. Zavoral, MD, and Robert J. Zwiener, MD, for the Liposorber Study Group A subgroup of patients with familial hypercholesterolemla (FH) respond Inadequately to standard diet and drug therapy, and are therefore at high risk for the premature development or progression of coronary artery disease. This study evaluated low-density llpoprotein (LDL) cholesterol and lipoprotein (a) removal in a mu&enter, controlled trial with a new LDL spheresis procedure (Liposorb&P LA-15 System). The study comprised patients with FH who had not responded adequately to dlet and maximal drug therapy. There were 54 patients with heterozygous FH (45 randomized to treatment and 9 control subjects) and 10 with homozygous FH (all of whom recefved LDL apheresls). The study included three g-week treatment phases and a 4-week r&bound phase. Treatments were administered at I- to 14&y intervals. Mean acute reductions in ‘LDL cholesterol were 76% in heterozygous FH patlents and 81% in homozygou? ones. Time-averaged levels of LDL cholesterol were reduced 41% (243 to 143 mg/dl) in heterozygous FH patbnts and 53% (447 to 210 mg/dl) In homoxygous ones. The substantial acute reduction of lipoprotein (a) (means: 6S%, heterozygous FH; 6896, homozygous FH) has not been reported with other thdrapies. The Lipesorber LA15 System represents an important therapeutic option in FH patlents who respond inadequately to diet and drug therapy. (Am J Cardiil1662;70:101&1016)

subsetof patients with heterozygousfamilial hypercholesterolemia(FH) and virtually all with homozygous FH do not achieve adequate lowdensity lipoprotein (LDL) cholesterol control with diet and drug therapy becauseof extremely high, initial lipid levels or drug intolerance. Surgical proceduressuch as partial ileal bypass, portacaval shunt and liver transplantation, although effective for lipid lowering,1-3have been associatedwith considerable morbidity. Plasmapheresis has been used to treat patients with severe FH,4,5but causesnonspecific depletion of plasma proteins, including high-density lipoprotein (HDL) cholesterol. In contrast, LDL apheresisis an extracorporeal procedure that specifically removesapolipoprotein B-containing lipoproteins from the blood. Several LDL apheresis methods have been developed,including: (1) heparin-induced extracorporeal LDL precipitation,‘j (2) bags containing heparin-agarose,’ (3) columns containing immobilized antiapolipoprotein-B antibodies,8*9and (4) columns containing immobilized dextran sulfate.*0y11 A recently developed LDL apheresis technique uses the Liposorbefl LA-15 System (Kaneka Corporation, Osaka, Japan).12This method usesa hollow fiber cell separator and 2 disposable 150 ml columns containing dextran sulfate-cellulosebeads.The advantagesof this system include a small extracorporeal volume, nonsaturable lipid binding, disposablecolumns and absenceof sensitization to column constituents as observedin patients undergoing immunoadsorption therapy.*3 Preliminary uncontrolled studies using the Liposorber LA-l 5 System, and its precursor LA-40, which were conductedin Japan and Europe,11,12*14-22 showedit to be a safe and effective method to selectively lower LDL. The present report describesthe results from a multicenter, controlled trial of the Liposorber LA-1 5 System conducted in the United States in FH patients who were resistant to diet and maximal drug therapy.

A

METHODS From The RogosinInstitute, New York Hospital-Cornell Medical Center, New York, New York. This study was supportedby a grant from the Kaneka America Corporation, New York, New York. Manuscript received March 30, 1992; revised manuscript received and accepted June 15,1992. Addressfor reprints: Bruce R. Gordon,MD, The RogosinInstitute, 505 East 70th Street, Suite H237, New York, New York 10021. 1010

THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 70

Study cohork Patients of both sexes between the agesof 5 and 70 years who fulfilled the following criteria were eligible for the study: (1) FH heterozygotes who had LDL levels > 160 mg/dl, despite appropriate diet and maximal tolerated combination drug therapy, or who were receiving an appropriate diet and had a OCTOBER 15, 1992

documented history of drug intolerance; or (2) FH homozygotes.Diagnosis of heterozygousFH was basedon the presence of primary hypercholesterolemia, tendon xanthomas and family history of hypercholesterolemia. The diagnosis of homozygous FH included the criteria for heterozygous FH patients plus demonstration on cultured skin Iibroblasts of <20% functional LDL receptor activity. Patients were excluded from the trial for the following reasons:pregnancy; body weight <15 kg or >30% above ideal; severecardiac disorders, such as malignant arrhythmias or decompensatedcongestiveheart failure (New York Heart Association functional class IV); myocardial infarction or cerebrovascularaccident in the previous 4 months; coagulation abnormalities; uncontrollable hypertension or hypotension; severeliver fun0 tion abnormalities; diabetes mellitus; and any medical condition that in the judgment of the investigator, may interfere with safe apheresistreatment. The study was approved by the institutional review board of each participating center. All patients gave signed, informed consent. The Liposorber LA-1 5 System was granted investigational device exemption status by the Food and Drug Administration. Study design: The study protocol comprised 2 phases:a 6-week screening phase,and a 22-week study phasethat included an l&week treatment period and a 4-week rebound period. During the 6-week screening phase, all patients were stabilized on diet (American Heart Association Phase I or a diet more restricted in saturated fats and cholesterol) and maximal tolerated drug therapy. Diet and drug therapy was continued unchanged through the study period. Patients with heterozygous FH were assignedat random to diet, drug therapy and LDL apheresisor to control (diet and drug therapy) in a 41 allocation. This allocation was adopted becauselipid lowering in the treated group was expected to be so large that only a small number of control patients would be neededto demonstratestatistical sig-

nificance of LDL cholesterol reduction. Randomization was stratified by clinical center and blocked within each center. Treatments could not be blinded, but assay of lipid, hematologic, biochemical and coagulation measures was performed by a central laboratory that was unaware of the treatment group assignmentand clinical characteristics of patients. The 18-week treatment period was divided into 3 time coursesof 6 weekseach. LDL apheresiswas performed once every 2 weeks during the first 6-week course.During the secondand third 6-week courses,the treatment frequency was adjusted toward a time-averaged LDL level of 130 mg/dl in heterozygousFH patients and 160 mg/dl in homozygous ones, A 4-week rebound study followed the 18-week treatment period, during which LDL apheresiswas discontinued to observerebound in total and LDL cholesterol levels while patients received diet and drug therapy only. LDL apheresiswas performed in a continuous-flow manner using the Liposorber LA-15 System (Figure 1). Plasma separation from whole blood was performed using polysulfone hollow fibers in a 140 ml chamber with l- to - 1 and 1 half plasma volumes (50 to 70 ml/kg) processedper procedure. Sodium heparin was used as the anticoagulant. Two columns, each containing 150 ml of dextran sulfate cellulose, were used as the adsorbent for apolipoprotein B. Plasma was alternately perfused through each column, enabling regeneration of the off-line cohmm with hypertonic (0.7 M) sodium chloride solution. The processwas controlled by a computerized unit, thereby minimizing operator intervention. The plasma, after passing through the adsorbent column, was recombined with the cell stream and returned to the patient. Total extracorporeal volume was 400 ml. The columns were discarded after each treatment. For most patients, the treatment duration was 2.5 to 3.0 hours. Lipid,

lipoprotein

and apallpoprotein

determina-

tions: All lipid, lipoprotein and apolipoprotein measureRinsing-Priming Solution

Regeneration Fluid

I Pump

Blood Line

Plasma Line

+

Regeneration Line

LDL APHERESIS IN FAMILIAL HYPERCHOLESTEROLEMIA

1011

TABLE I Baseline Characteristics of All Patients* Heterozygotes

Mean age (year) Male/female Myocardial infarction (%I Angina history (%) Coronary bypass surgery !%I Coronary angioplasty (%I Cerebrovascular disease (%) Peripheral vascular disease (%) Hypolipidemlc drugs taken (%) 0 1 2 3 4 Drug received (%) Lovastatin Resin Niacin Probucol Gemfibrozil

Treatment (n = 45)

Control (n = 9)

Homozygotes (n = 10)

All Patients (n = 64)

46 29/16 14 (33) 30 (71) 23 (51) 10 (22) 5(11) 7 (16)

54 514 3 (33) 7 (78) 3 (33) 0 (0) 2(22) 4 (44)

24 317 0 (0) 5 (50) 3 (30) cl (0) O(O) l(10)

44 37127 17 (27) 42 (69) 29 (45) 10 (16) 7 (11) 12 (19)

4 10 18 11 2

(9) (22) (40) (25) (4)

0 5 2 2 0

(0) (56) (22) (22) (0)

3 (30) 2 (20) 4 (40) l(10) 0 (0)

7 17 24 14 2

(11) (27) (37) (22) (3)

37 23 22 3 2

(82) (51) (49) (7) (4)

8 (89) 3 (33) 2 (22) l(10) l(11)

7 (70) 2 (20) 3 (30) l(10) 0 (0)

52 28 27 5 3

(81) (44) (42) (8) (5)

*Includes all patients regardless of whether phases of protocol.

they completed

basehne or treatment

ments were obtained by a central laboratory (Medical Research Laboratories, Cincinnati). Total and HDL cholesterol and triglycerides were analyzed using microenzymatic procedures,as previously d&bed.23 The central laboratory maintained Centers for DiseaseControl-National Heart, Lung, and Blood Institute Part III Standardization for all 3 lipid parameters.” HDL was isolated using heparin-2 molar manganesechloride.25 LDL cholesterol was calculated using the Friedewald formula: LDL cholesterol = total - HDL cholesterol (triglycerides/5).26 Total HDL and LDL cholesterol, and triglycerides were determined immediately before and after each treatment, and during the 4-week rebound period. Estimated values of LDL cholesterol were checked at baseline, and before and after treatment at 6-week intervals by direct measurements,using the ultracentrifuge. 27Apolipoproteins Al, A2, B, E and lipoprotein (a) were measuredusing enzyme-linked immunosorbent assayprocedures28*29 twice in the 2 weeks before initial treatment, and before and after each treatment visit once every 6 weeks. The time-averaged lipid levels for heterozygousFH patients randomized to the control group were calculated as the arithmetic average of lipids measuredduring the study period. To determine time-averagedlevels for patients receiving LDL apheresis,measurementswere performed between treatments in the first and second courses,and during the 4-weekrebound period after the third course of therapy. Specimens for the rebound studies were typically sampled on days 0, 1, 2, 3, 5, 7 and 14. A simple arithmetic average of total or LDL cholesterol before and after treatment was not appropriate, because the rebound of lipid levels after LDL apheresis treatment followed a concave curve. There fore, a time-averaged value was calculated using the 1012

THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 70

area under the curve from values obtained during the rebound period. Wfkkncy of admptkm The efficiency of adsorp tion is a measureof the selectivity and adsorption specificity of the Liposorber LA-15 System for the different lipids and lipoproteins. For each treatment, regression analysiswas usedto generatea coefficient of adsorption efficiency (1) value for each lipid parameter. The following kinetic adsorption model was used V CPost = VP VP + v,, * CPre exP -11 VP + v,, 1 I where C,, is the pretreatment concentration of a parameter (mg/dl), C, is the posttreatment concentration of a parameter (mg/dl), V, is the estimatedplasma volume (ml), V,I is the extracorporeal volume + recovery solution (500 ml), V,Qis the extracorporeal volume, and V is the treated plasma volume. Theoretically, a substance with an q of 0 passes through the LA- 15 column with no adsorption, whereas a substancewith an 7 of 1 is completely adsorbedonto the column. A mean 7 value for each lipid parameter was computed by averaging the r] values of all LDL apheresistreatments. Safety parameters: For each patient, a standard history and physical examination was performed before the beginning of the study. At each visit, an interim history, body weight and vital signswere recorded.Lab oratory tests including hematology, coagulation, and biochemistry profiles, immunoglobulins and complement determinations were obtained at baseline, before the first 3 LDL apheresistreatments, and every 6 weeks thereafter. Statlstkal analysis: Statistical analyses were performed by the data coordinating center (ADK Research Corporation, Pittsburgh). Baselineforms were collected by the treatment centers and forwarded to ADK Research Corporation. Baseline data were manually entered by doubleentry verification in a data basecreated for the study on a DEC VAX 3100. Lipid and laboratory valueswere forwarded on diskettesto ADK Research Corporation from the central laboratory. SAS software (version 6, 1991, SAS Institute, Inc., Cary, North Carolina) was usedin the statistical analysis of baselineand study data. Patient characteristics and lipid values for homozygous and heterozygousFH control and treatment patients were comparedfor each group by t tests for continuous variable-s,such as lipid values, and by chi-square tests for categorical variables, such as gender. Twotailed tests were used to calculate p values. A p value <0.05 was consideredsignificant. RESULTS The clinical characteristics of all patients are listed in Table I. In all, 64 patients (37 male and 27 female, age range 8 to 67 years) enrolled in the study. The cohort included 54 patients with heterozygousFH, 45 of whom were randomized to receive LDL apheresistreatment and 9 to serve as the control group. Ten patients had homozygousFH, all of whom receivedLDL apheresis. Three homozygousFH patients had a prior surgiOCTOBER 15, 1992

cal procedure to reduce the LDL level (ileal bypass [n = 21 and portacaval shunt [n = 11). One heterozygous FH patient randomized to receive LDL apheresis did not complete the baseline phase and never received LDL apheresis.Nine heterozygousFH patients did not complete the LDL apheresisprotocol, becauseof inadeauate vascular access(n = 5) or adverseeffects (n = 4). &e control subject was found to have nonfamilial hypercholesterolemiaafter randomization. Coronary artery disease was clinically evident in most patients enrolled in the study. The mean age of onset of symptomatic coronary artery diseasewas 38 years. Cerebrovascular and peripheral vascular disease were less frequently observed. Baseline lipid values for all patients enrolled in the study are listed in Table II. Mean lipoprotein (a) levels were increased (>40 mg/dl) in all groups. Despite diet and combination lipid-lowering drug therapy, patients with FH had mean LDL levels >200 mg/dl (heterozygotes) or >400 mgjdl (homozygotes). Many patients also received concomitant medications such as calcium antagonists (30%), /3blockers (30%), nitrates (32%) and aspirin (67%). wn redudom In all, 588 ACUblipidd LDL apheresis procedures were performed in 54 patients during the courseof the study. The percent reductions in LDL cholesterol for the LDL apheresistreatments ranged from 70 to 79% in the heterozygousFH patients and from 71 to 82% in the homozygousones. The overall mean acute reductions in LDL cholesterol were >75% for both heterozygous and homozygous FH patients (Table III). The mean acute reductions in total cholesterol were 61% (heterozygous) and 70% (homozygous). In contrast, the mean acute reductions in HDL cholesterol were only 5% (heterozygous) and 13% (homozygous). Despite the acute reduction in HDL cholesterol, there was a 1 mg/dl increasein HDL cholesterol levels before treatment over time for both heterozygousand homozygous FH patients. The mean acute reductions in lipoprotein (a) of 65% (heterozygous) and 68% (homozygous) were comparable to the substantial decreasesin apolipoprotein B. As a group, homozygotes had significantly (p <0.05) greater percent reductions in total, LDL and HDL cholesterol,and TABLE Iv

apolipoproteins Al, A2 and B than did heterozygous ones. Effkieney ol admptiam LDL cholesterol had the highest q value (1.07 for heterozygotes,and 1.05 for homozygotes),indicating that it was most efficiently removed. In heterozygotes,the 7 values for lipoprotein (a) TABLE II Mean Baseline Lipids (mg/dl) on Diet and Drug Therapy Heterozygotes

(n = 45) Total cholesterol HDL cholesterol LDL cholesterol* LDL cholesterolt Triglycerides Apolipoprotein Al Apolipoprotein A2 Apolipoprotein B Apolipoprotein E Lipoprotein (a)

313 t 54 48 2 13 239 ‘t 58 239 r 56 134 -t 47 125 -+ 31 462 11 186 r 44 922 43 -+ 35

Control (n = 9) 285 52 203 193 148 139 46 153 10 64

k r 2 k + ‘2 -+ + *

Homozygotes (n = 10)

All Patients (n = 64)

506 k 39 + 447 2 447 + 99 2 100 k 38+ 341 + 12 5 47 f

339 k 47 k 266 k 259 Ir 130 f 124 k 45r 203 2 9k3 47 2

29 16 34 54 99 25 7 45 3 63

Lipid Parameter

Heterozygotes

Total cholesterol HDL cholesterol LDL cholesterolt LDL cholesterolS Triglycerides Apolipoprotein Al Apolipoprotein A2 Apolipoprotein B Apolipoprotein E Lipoprotein (a)

61 22 524 76 -c 2 73 k 4 60 r 9 14 -+ 4 15-t 7 63 -t 7 56 r 9 65 r 8

Homozygotes 70 2 13 + 81 f 78 f 59 _t 19 2 23 k 73 2 54 2 68-t

Time Average*

p Value

Difference?

40 8

320 -c 54 284 2 31 0.0794 506 2 146

220 f 41 317 2 30 0.0014 291 2 83

0.0000 0.0484

-100 2 46 +33 2 18

0.0001

-216

243 2 58 206 -t 35 0.0909 447 -t 143

143 2 39 232 k 37 0.0059 210 2 62

0.0001 0.1689

-100 + 46 +26 k 20

0.0001

-238

10

4 5 5 6 9 6 10 5 40 11

p Value* 0.0001 0.0001 0.0281 0.2032 0.5716 0.0026 0.0363 0.005 0.7245 0.3383

*Determined by analyslaof variance for tests of effects of disease group. tCalculated from Friadewaldformula.aa tDlrect measurement. Data are shown as mean + SO and are incluswe of all pabents enrolled III study regardlessof whether they completed the protocol. Abbreviations as in Table II.

Baseline

40 8

41

TABLE Ill Acute Mean Percent Reductions in Lipid, Lipoprotein and Apolipoprotein Levels

No. of Pts.

10

102 14 108 104 58 32 11 79

*Estimated from Friadewalclaquation2a tDiractly measured. Data are shown as mean -+ SD and are inclusive of all patients enrolled in study regardlessof whether they completed baselineand treatment phasesof protocol. HDL = high-density lipoprotein; LDL = low-densdy lipoprotein.

Comparison of Baseline and Time-Averaged Mean Lipid Values (mg/dl)

Total cholesterol Heterozygous FH Treatment Control p value Homozygous FH LDL cholesterol Heterozygous FH Treatment Control p value Homozygous FH

145 10 143 148 53 27 11 100 4 45

1

+ 106

z 111

*Time averagefor treatment patients calculated by area under cwve method. tCalculated as mean of all patrents’ differencesbetween study period time-averaged lipid levels and mean baseline levels. Patient number includes patients in treatment phasewith sufficient data to calculate time-averaged levels. FH = familial hypercholesterolemia; LDL = low-dens@ lipoprotein.

LDL APHERESIS IN FAMILIAL HYPERCHOLESTEROLEMIA

1013

and apolipoprotein B were similar (0.82 and 0.73, respectively). In homozygotes,the 7 values were 0.91 for lipoprotein (a) and 0.89 for apolipoprotein B. There was no evidence for HDL cholesterol removal by the columns, becausethe 7 values were -0.08 for heterozygatesand -0.06 for homozygotes.There were fewer differencesbetweenheterozygousand homozygousFH patients in q values than in acute lowering changes. The-averaged lipid changes: Forty-eight heterozygous FH patients (40 treatment and 8 control) and 10 homozygousoneshad sufficient data to determine timeaveraged changesin total and LDL cholesterol. In the 40 heterozygousFH patients, the time-averagedlevel of total cholesterol was reduced 31% (320 to 220 mg/dl), and LDL cholesterol was reduced 41% (243 to 143 mg/dl) (Table IV). In heterozygousFH patients, 42% achieved time-averaged total cholesterol levels <200 mg/dl, 75% had levels <240 mg/dl, and 50% had timeaveragedLDL cholesterol levels <130 mg/dl. In homozygous FH patients, the mean time-averagedtotal cholesterol level was reduced 42% (506 to 291 mg/dl), and the LDL cholesterol level was reduced 53% (447 to 210 mg/dl). In homozygous FH patients, 40% achieved time-averaged total cholesterol levels <240 mg/dl. Although these levels were not as low as those achievedin heterozygous FH patients, the percent reduction was greater. In control patients, LDL cholesterol increased 13% from 206 mg/dl at baseline to 232 mg/dl during the study period. The difference in time-averagedLDL cholesterol levels between the treatment and control groups was highly significant (p
0

7 Day Lowering 61% Homozygotes 53% Heterozygotes

fect of treatment frequency. The relation betweenLDL lowering and treatment frequency is shown for all patients in Figure 2. Pearsoncorrelation analysis demonstrated a significant relation betweenfrequency of therapy and reduction of time-averagedtotal and LDL cholesterol values (p
14 Day Lowering 49% Homozygotes 38% Heterozygotes o

10 20 30 LDL-C % Lowering

4o 50 l

60 70 80

0 0

90 F 100 11111111l1111111111111111111111111111111111111111111111111111111 0 2 4 6 8 10 12 14 16 18 Treatment Interval (Days)

1014

THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 70

OCTOBER 15. 1992

20

22

24

26

ment becauseof difficulty in obtaining vascular access; 4 discontinued becauseof either hypotensive reactions (n = l), headache, nausea and flushing (n = l), arrhythmia (n = 1) or dietary noncompliance (n = 1). There was no significant change in mean hematocrits during the study (baseline 42.2% for treated patients and 41.6% for controls; study period 40.5% for treated patients and 41.5% for controls). There was also no significant change in coagulation studies (baseline prothrombin time 12.5 secondsfor treated patients and 12.6 seconds for controls; study period prothrombin time 12.7 secondsfor treated patients and 13.3 seconds for controls). All other baseline hematologic, coagulation and biochemistry laboratory values also remained within the normal range throughout the study, and no significant differences in these values were found between patients receiving LDL apheresis and control subjects or between baseline and study period values. DISCUSSION The limited or invasive therapeutic options for patients with severe FH inadequately controlled on diet and maximal drug therapy have led to the development of LDL apheresis for removing apolipoprotein B-containing lipoproteins from the blood. The present study was designed to assessthe efficacy and safety of LDL apheresis,using an automated dextran sulfate cellulose adsorption system. This system was very effective in removing apolipoprotein B-containing lipoprotein particles; acute lowering of LDL cholesterol averaged 76% in patients with heterozygous FH and 81% in those with homozygous FH. The slightly greater lipid lowering observed in patients with homozygousFH most likely reflected a dilutional effect of the procedure, becauseof the smaller body weight and plasma volume of homozygousFH patients compared with those of heterozygousones. The similarity of the LDL cholesterol binding efficiency of the columns between the patient groups supports this conclusion. Repeatedlowering of LDL by the procedure had little effect on underlying cholesterol metabolism during or after LDL apheresis therapy. Comparison of the LDL cholesterol value 4 weeksafter discontinuing LDL apheresis treatment with the initial baseline level did not show an appreciable change. In contrast, the final LDL and total cholesterolvalues for the control patients showedan increasewith time. There was a minimal increasein HDL cholesterol levels before treatment during the study. A prior report demonstrated that HDLcholesterol levels may increase as a result of LDL apheresis therapy.3o Plasma exchange, an alternative treatment in refractory FH patients, reducesHDL-cholesterol levels. The acute lowering of lipoprotein (a) averaged65 to 68%. This was consistent with the acute lowering of apolipoprotein B and indicated that the apolipoprotein (a) portion of the lipoprotein (a) molecule did not interfere with lipoprotein (a) binding to the dextran sulfatecontaining columns. Reduction in lipoprotein (a) is particularly important for patients with FH, becauseof the closeassociationbetweenincreasedlipoprotein (a) levels

and risk of atherosclerosis.31Standard diet and inhibitors of 3-hydroxy-3-methylglutaryl coenzyme-A reductase do not reduce lipoprotein (a) levels.32Nicotinic acid when administered at a dosageof 4 g/day has been reported to decreaselipoprotein (a) levels by 34%~~~ LDL apheresisis the most effective method for markedly reducing lipoprotein (a) levels. Lowering of lipoprotein (a) by LDL apheresis using a heparin-induced LDL precipitation procedure was recently reported.34 However, it remains to be seenwhether reducing lipoprotein (a) also decreasesthe risk of coronary artery disease. The results of this study help provide guidelines for the optimal treatment frequency for the procedure. Therapy every other week provides a substantial reduction of time-averaged LDL cholesterol levels (approximately 40% in heterozyotesand 5m in homozygotes). This treatment frequency should produce a time-averaged LDL level of I1 30 mg/dl in heterozygotes,provided that the initial LDL level on maximal drug therapy is 1220 mg/dl. A once-weeklyinterval may be necessaryin somepatients with homozygousFH and a few with heterozygousFH to achieve desirable lipid levels. The LDL apheresisprocedure was well-tolerated by most patients. Difficulty with vascular accesswas a major problem in 5 patients. The observedadverseevents were those associatedand expected with all extracorporeal circulation procedures;hypotensionwas the most frequent event. Randomized placebo-controlled studies in patients with hypercholesterolemiahave shown that LDL lowering by diet and combination drug therapy can induce regressionof coronary lesions.35-37 Noncontrolled studies have suggestedthat LDL apheresiscan induce re gression of coronary lesions in patients with heterozygous and homozygousFH refractory to diet and drug therapy.38-40This clinical trial clearly demonstrated that the Liposorber LA- 15 Systemsafely and efficiently reduces LDL cholesterol and lipoprotein (a) levels in FH patients who have not respondedadequately to diet and maximal drug therapy, and have few therapeutic alternatives. APPENDIX The participating centers and investigators were as follows: Abbott Northwestern Hospital, Minneapolis Heart Institute, Minneapolis, Minnesota: David C. Brown, MD, JamesH. Zavoral, MD; Baptist Memorial Hospital, Memphis, Tennessee:Kenneth D. Grosshart, MD, Thomas N. Stern, MD; Christ Hospital, Cardiovascular ResearchCenter, Cincinnati, Ohio: Donald M. Black, MD, Evan A. Stein, MD, PhD; Evanston Hospital, Evanston, Illinois: Peter C. Dau, MD; Methodist Hospital and Baylor College of Medicine, Houston, Texas: Antonio M. Gotto, Jr, MD, Peter H. Jones,MD; National Institutes of Health, Department of Transfusion Medicine, Bethesda,Maryland: Susan F. Leitman, MD; Oregon Health SciencesUniversity, Portland, Oregon: D. Roger Illingworth, MD, PhD, James S. Prihoda, MD; The Rogosin Institute, New York, New York: Bruce R. Gordon, MD, Daniel M. Levine, PhD, Thomas S. Parker, PhD, Stuart D. Saal, MD, Theodore LDL APHERESIS IN FAMILIAL HYPERCHOLESTEROLEMIA

1015

I. Tyberg, MD; and University of Texas Health Science Center, Dallas, Texas: David W. Bilheimer, MD, Ricardo Uauy, MD, Robert J. Zwiener, MD.

plasmaadsorptiontherapy on familial hypercholeaterolemia.Biomarer ArrifCells 1987;15:113-124. 19. Agishi T, Kitano Y, Suzuki T, Miura A, Murakami J, Minagawa H, Ban K. Improvement of peripheral circulation by low density lipoprotein adsorption. ASAIO Tram 1989;35:349-351. 20. Rubba P, Iannuxxi A, PostiglioneA, Scarpato N. Montefusco S, GnassoA, Nappi G, CorteseC, Mancini M. Hemodynamicchangesin the peripheral circulation after repeat low density lipoprotein apheresisin familial hypercholeaterolemia. Circulation 1990;8I :610-616. 21. Berger GM, Firth JC, Jacobs P, Wood L, Marais AD, Horak A. Three REFERENCES 1. Buchwald H, Varco RL, Matts JP, Long JM, Fitch LL, Campbell GS, Pearce different schedulesof low-density lipoprotein apheresiscomparedwith plasmaMB, Yellin AE, Edminston WA, Smink RD Jr, Sawin HS Jr, Campm CT, pheresisin patients with homoxygousfamilial hypercholeaterolemia.Am J Med Hansen BJ, Tuna N, Kamegis JN, Samnarco ME, Amplatx K, Castaneda- 1990,88:94-100. 22. Thompson GR, Barbii M, Okabayashi K, Trayner I, Larkin S. PlasmaZuniga WR, Hunter DW, Bitt JK, Weber FJ, StevensonJW, Leon AS, Chalmers TC, and the PGSCH Group. Effect of partial ileal bypasssurgery on pheresis in familial hypercholaterolemia. Arteriosclerosis 1989;9(suppl I): mortality and morbidity from coronary heart dii in patientswith hypercholea- I-152-1-157. terolemiazreport of the Program on the Surgical Control of the Hyperlipidemias 23. Steiner PM, Freidel J, Bremner WF, Stein EA. Standardization of micromethodsfor plasmacholesterol,triglyceride and HDL-cholesteml with the Lipid (PGSCH). N Engl J Med 1990,323:946-955. 2. Stem EA. Mieny C, Spitz L, Sadron I, P&for J, Heimann KW, BersohnI, Clinics’ methodology.J Clin Chem Clin Biochem 1981;19:8SO. Dinner M. Portacaval shunt in four patients with homoxygoushypercholeaterol- 24. Myers GL, Cooper GR, Wimr CL, Smith SJ. The Centers for Dii Control-National Heart, Lung, and Blood Institute Lipid Standardization Pro emia. Lnncet 1975;1:832-835. gram: an approach to accurate and preciselipid measurements.Clin Lab Med 3. Bilheimer DW, Grundy SM, Starxl TE, Brown MS. Liver transplantation to 1989;9:1OS-135. provide low density lipoprotein receptorsand lower plasmacholesterolin a child with homoxygous familial hypercholeaterolemia.N Engl J Med 1984;311: 25. Wamick GR, Albcrs JJ. A comprehensiveevaluation of the heparin manganeseprecipitation procedurefor estimatinghigh-densitylipoprotein cholesterol.J 1658-1664. 4. ThompsonGR, Lowenthal R, Myant NB. Plasmaexchangein the manage- Lipid Res 1978;19:65-76. ment of homoxygousfamiliil hypercholeaterolaemia.fun& 1975;1:1208-1211. 26. Friedewald WT, Levy RI, FredricksonDS. Estimationof the concentrationof low-density lipoprotein cholesterolin plasma,without useof the preparative ultraS. Thompson GR. Miller JP, Brealow JL. Improved survival of patients with centrifuge. Clin Chem 1972;18:499-502. homoxygousfamilial hypercholeaterolaemiatreated with plasma exchange.Br 27. Lipid ResearchClinics Program. Manual of Laboratory Operations: Lipid Med J 1985;291:1671-1673. 6. &chuff-Werner P, Armstrong VW, Eiinhauer TH, Thiery J, Seidel D. Treatand Lipoprotein Analysis. Washiiton, DC US Department of Health, Education, and Welfare; Publication (NIH), 1982;75:628. ment of severe hypercholeaterolemiaby heparin-inducedextracorporeal LDL 2B. Stein EA, DiPersio L, PesceAJ, Kashyap M, Kao JT, Srivastava L, McNerprecipitation (HELP). Beirr Infiion Ther 1988;23:118-126. ney C. Enzyme-linkedhnmunoabeorbentassayof apolipoproteinA II in plasma, 7. Lupien PJ, Moorjani S, Awad J. A new approach to the managementof familial hypercholeate.rolemia: removal of plasmacholesterolbasedon the princiwith use of a monoclonal antibody. Chin Chem 198632967-971. 29. Stein E, Kreisberg R, Miller V, Mantel1 G, Washington L, Shapiro DR. ple of affhtity chromatography. Luncer 1976;1:1261-1264. 8. Stoffel W, Borberg H, Grave V. Application of specificextracorporeal removal Effect of simvastatinand choleatyraminein familii and non-familial hypercholesof low density lipoprotein in familial hypercholesterolaemia.Lance? 1981;2: temlemia. Arch Intern Med 1990;150:341-345. 30. Parker IS, Gordon BR, Saal SD, Rubm AL, Ahrens EH Jr. Plasma high 1005-1007. density lipoprotein is increasedin man when low density lipoprotein (LDL) is 9. Saal SD, Parker TS, Gordon BR, StudebakerJ, Hudgins L, Ahrens EH Jr, Rubm AL. Removal of lowdensity lipoproteins in patients by extracorporeal lowered by LDLaphersis. Proe Narl Acad Sci US A 1986;83:777-781. 31. SeedM, Hoppichler F, ReaveleyD, McCarthy S, ThompsonGR, Boerwinkle immunoadsorption.Am J hfed 1986;80:583-589. 10. Yokoyama S, Hayashi R, Kikkawa T, Tani N, Takada S, Hatanaka K, E, Utermamt G. Relation of serumlipoprotein (a) concentrationand apolipopro Yamamoto A. Specitic sorbent of apolipoprotehr B-containing lipoproteins for tein (a) phenotypeto coronary heart diseasein patientswith familial hypercholtsplasmaphereais.Arreriosclerosis 1984;4:276-282. terolemia. N Engl J Med 1990,322:1494-1499. il. Yokoyama S, Hayashi R, Satani M, Yamamoto A. Selectiveremoval of low 32. Kcstner G, Gavish D, Leopold B, Bolxano K, Weintraub MS, Breslow JL. density lipoprotein by plasmaphermisin familial hypercholesterolemia.ArmrioHMGCoA reductaseinhibitors lower LDL choleatemlwithout reducing Lp(a) sclerosis 1985;5:613-622. levels. Circularion 1989;80:1313-1319. 12. Mabuchi H, Michishita I, Takeda M, Fujita H, Koixmni J, Takeda R, 33. Carlson LA, HamstenA, Asplund A. Pronouncedlowering of semmlevelsof lipoprotein Lp(a) in hyperlipidaemicsubjectstreated with nicotinic acid. J Intern Takada S, Gonishi M. A new low density lipoprotein aphereaissystemusingtwo Med 1989;226:271-276. dextran sulfate cellulose columns in an automated column regenerating unit (LDL continuousapheresis).Atherosclerosis 1987;68:19-25. B4. Armstrong VW, SchleefJ, Thiery J, Muche R, Schuff-Werner P, Eiinhauer T, Seidel D. Effect of HELP-LDL-aphereais on serumconcentrationsof human 13. Gordon BR, Sloan BJ, Parker TS, Saal SD, Levine DM, Rubin AL. Humoral immuneresponsefollowing extracorporeal immmmadsorptiontherapy of patients lipoprotein (a): kinetic analysisof the post-treatmentreturn to baselinelevels.Ear J Clin Inuesr 1989;19:235-240. with hypercholesterolemia.Tran.@sion 1990;30:318-321. 35. Blankenhom DH, Nessim SA, Johnson RL, Sanmarco ME, Axen SP, 14. Homma Y, Mikami Y, Tamachi H, Nakaya N, Nakamura H, Goto Y. Comparisonof selectivity of LDL removalby doublefiltration anddextran-sulfate Cashm-Hemphill L. Beneficial effects of combinedcoleatipol-niacintherapy on coronary athercsclercsisand comnary venous bypassgrafts. JAMA 1987;257: cellulcsecolumn plasmaphemsis,and changesof subfractionatedplasmalipopro teins after plasmaphereaisin heteroxygousfamiliil hypercholeaterolemia.Merab3233-3240. olism 1987;36:419-425. 36. Brown G, Albers JJ, Fisher LD, SchaeferSM, Lm JT, Kaplan C, Zhao XQ, Bin BD, Fitxpatrick VF, DodgeHT. Regressionof coronary artery diseaseasa IS. Miiori A, Takahashi K, Mitamura T, Kato H, Teramoto T, Komura I, Honda Z, Toma S, Miyamoto T. Clinical evaluationof three typesof plasmapher- result of intensivelipid-lowering therapy in menwith high levelsof apolipoprotein eais in a patient with type IIa familiil hypercholeaterolemia.J C/in Apheresis B. N Engi J Med 1990;323:1289-98. 37. Kane JP, Malloy MJ, PortsTA, Phillips NR, Diehl JC, Have1RJ. Regression 198x3:209-215. 16. Frances&ii G, BusnachG, Vaccarino V, Calabreai L, Gianfranceschi G, of coronary athercsclermis during treatment of familial hypercholesterolemia Sirtori CR. Apheretic treatment of severefamiliil hypercholeaterolemia:compar- with combineddrug regimens.JAMA 1990;264:3007-3012. 38. Borbarg H, Gacxkowski A, Hombach V, Gette K, Stoffel W. Regressionof ison of dextran sulfate celluloseand double membranefdtration methodsfor low atherosclerosisin patientswith familial hypercholestemlemiaunder LDL-sphere density lipoprotein removal. Atherosclerosis 1988;73:197-202. 17. Gdaka M, Kobayashi H, So& K, Murotani N, Saito Y, Niihide T, Ycuhida sis. Prog Clin Biol Res 1988:255:317-326. S, Tani N, Takata S. Adsorption of lipoprotein containing apolipoprotein-B 39. Yokoyama S, YamamotoA, Hayashi R, Satani M. LDL-apheresis:potential procedure for prevention and regressionof atheromatousvascular lesions.Jpn through plasmaseparation for treatment of familiil hypercholsterolemia. Inr J Circ J 1987;51:1116-1122. Artif Organ 19869343-348. 40. Yamamoto A. Regressionof atherosclerosisin humans by lowering serum 18. Gdaka M, Kobayashi H, Tabata Y, SoedaK, Hayashi H, Ito S, Murotani N, Saito Y, Niihide T, Shinomiya M, Yoshida S. Long term result of LDL selective cholesterol.Atherosclerosis 1991;89:1-10.

1016

THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 70

ArrifOrgans

OCTOBER 15, 1992