408
JACC Vol. 10. No, 5
Nove mber 1 '!~ 7 :40B -4B
Safety and Efficacy of Tissue-Type Plasminogen Activator Produced by
Recombinant DNA Technology
BURTON E. SOBEL, MD, FACC
51. LOllis. Misso uri
Because of its enhanced activation of plasminogen in
association with fibrin compared with free plasminogen,
tissue-type plasminogen activat or (t-PA) evoked excitement as a potentially useful pharmacologic activator of
the fibrinolytic system. Initial studies in experimental
animals and patients demonstrated that it induced coronary thrombolysis rapidly and without concomitant,
mark ed fibrinogenolysis in contrast tostreptokinase. Large
scale clinical trials soon followed. Their results indicate
that intravenous admini stration of topA produced hy
Fibrinolysis has bee n recognized for more than 200 years ,
Perhaps surprisingly , coronary throm bolysis has a relative ly
long histor y as well , Th e current intense interest in this
mode of treatm ent of acute myocardial infarc tion reflects.
in part. improved end points of efficacy (co ronary angiogra ms that ca n be acq uired safe ly in critically ill pat ients),
increased appreciation of the seminal roles of thrombo sis in
the etio logy of transmural myocardial infarction and reperfusio n in the salvage of jeopardized tissue and the availability of activators of the fibrinolytic syste m with relative
clot selectivity such as tissue-type plasm inogen activator (tPAl only recently synthesized successfully in pharmaceu tical quantities by reco mbinant DNA techn ology (rt-PA ).
Because of the relati ve clo t selectivity oft-PA , a property
reflecting its higher affinity for plasminoge n in the fibrin
domai n than for circulating , free plasm inogen , we evaluated
human t-PA and subseque ntly rt-P A in ex perimen tal an imals
and patients with coro nary thrombosi s (1-5) . Clot lysis was
elicited promptl y without marked fibrinoge nopenia or induction of a sys temic lytic state pred isposing to bleedin g,
Subseq uently. seve ral large sca le clinica l trials wer e undertake n with rl-PA (6- 10). Some of their result s are the
subjec t of this brief status repo rt. Because this mate rial
addre sses not only publ ished results but also work in prog-
From the Washington Universit y School of Medici ne, St. Louis. Missouri.
Address for reprints: Burton E . Sobel. MD. Director , Cardiovascular
Division. Washington University School of Medicine , 660 South Euclid .
Campus Box 8086 , St. Louis, Missouri 63 110.
iD 19K7 by the American College of Cardiology
recombinant DNA technology (rt-Pa ) elicits coronary
thromboly sis in two-thirds or more of treated patients
with angiographically documented thrombotic occlusions generally without perturhing hemostatic mechanisms or inducing marked fibrinogenolysis. Bleeding is
usually confined to vascular access sites and episodes of
major bleeding are rare. Overall 6 week sur vival for
treated patients with documented acute myocardial infarction may be as high as 95%.
(J J1m Coli Cardiol 1987;10:40B-4B)
ress, some of the dat a are cited as personal communications
provided by inves tigators before completion of manu script s
already being prepared , including mater ial co mpiled by Drs.
Chesebro , Grossbard , Mueller and Passa mani.
The TIMl Trials
Phase I. The Thrombolysis in Myocardial Infarcti on
(TIM!) trial conducted by the National Heart, Lung, and
Blood Institute (NHLBI) ha s evo lved through several pha ses
to date. One (Phase I) docum en ted an almos t twofold gre ater
incidence (62 compared with 3 1% ) of angiographically confinne d coro nary thromb oly sis with rt-PA co mpared with
streptok inase in a double-blind , prospect ive . rand om se lection desig n (Fig. I) ( I I) . Sign ificant but modest improvement of global and regional ventric ular function has been
evide nt in TIMI trial patients in who m reper fusion was
induced by rl-PA within 4 hours of the onset of ches t pa in
( 12.1 3).
Mueller and her colleagues ( 14) delineated the incidence
of bleeding in several phases of the Tl AlI trial to date (FiX.
2) . The incidence of bleeding with strepto kinas e and rt-PA,
the two activators used in Phase I, was similar. For rt-PA
in all the phases of the T IMI tria l ana lyzed by Mueller et
a!. bleedi ng was confined largely to vasc ular access site s
(Fig. 3) . Bleed ing with rt-PA was generally but only loosely
related to the magnitude of decreased fibrinogen and increase d fibrinogen degr adation products. Nevertheless , the
associations support the view that laboratory criteria of a
0735-1097.187/$3.so
SOBH
lACC Vol. 10, No.5
November 1987:40B-4B
SAFETY OF rt-PA
41B
however, based in part on data from several studies and
with information obtained from computer-assisted analysis
of pharmacodynamics (17), that consumption of alpha-.
antiplasmin indicative of risk occurs with doses and durations of infusion of rt-PA exceeding theoretically defined
limits. Accordingly, the TIMI investigators and the TIMI
policy board monitored each upen label phase of the trial
80
60
---.
~
~40
Q.
o
20
o
rt -PA ( N= 1 13 )
SK (N = 1 19)
Figure 1. TIMl phaseI-efficacy. Patency ratesfor patients treated
with recombinant tissue-type plasminogen activator (rt-PA) (80
mg over 3 hours)or streptokinase (SK) (1.5 million U over I hour)
in the double-blind, random patient selection, prospective study
comprising Phase I of the TIMI trial. The number of subjects in
the two groups was 113 and 119, respectively. rt-PA was more
effective than streptokinase whether recanalization was defined by
angiography 30 (hatched bars) or 90 (solid bars) minutes after
the onset of administration of the active thrombolytic agent.
systemic lytic state portend an increased risk of bleeding,
Accordingly, the systemic lytic state seen invariably with
even subtherapeutic doses uf conventional activators is not
likely to be entirely benign.
Compared with streptokinase, rt-PA lowers plasma fibrinogen markedly less judging from results in subsets of
patients with extensive laboratory data analyzed in detail in
both TIMI Phase I (data not shown) and the European Cooperative Trial (Fig. 4) (15). Major bleeding requiring transfusion occurs less frequently with rt-PA than with streptokinase, judging from results in the European Cooperative
Trial (Fig. 5) (10,15).
Open label phases. Several open label phases of TIMI
were undertaken in part to delineate optimal dose and duration of treatment with the originally produced rt-PA and
with material available subsequently that comprised a larger
fraction of so-called single chain (suspension-culture, nonplasmin split) rt-PA, Despite some fluctuation of incidence,
major bleeding occurred within a range of 7 to 25% with
dose regimens that varied from 80 mg over 3 hours to 150
mg over 6 hours, Attempts to identify the upper bounds of
safe dosage were motivated in part by the higher earlier
recanalization rates (42 compared with 13% after 30 minutes) when 90 as opposed to 40 mg of rt-PA was given in
the first hour. Furthermore, the higher doses of rt-PA used
in TIMI open label phase E compared with those used in
Phase I were associated with substantially higher rates of
patency early (2 hours) and later ( I R to 4R hours) after the
onset of infusion (Fig, 6) (16). It was well recognized,
Figure 2. TIMI trial (total number = S50), Incidence of bleeding
expressedas the percentof patientsstudiedin thedesignatedphases
of the TIMl trial categorized as major, intracranial or total number
of episodes whether minor or major. The percent of patients manifesting each type of episode in each of the phases specified in the
trial is shown. Results in patients treated with streptokinase (SK)
in Phase I are shown in the lowest section of the figure (n =
147). Those in patients treated with recombinant tissue-type plasminogen activator (rt-PAj in Phase I (n = 143) arc represented
by the bars immediately above. In open label phases A through
E, all patients were treated with J1-PA but different doses were
employed, The dose of rt~PA in Phase I was 80 mg over 3 hours
(40. 20 and 20 mg/h, respectively). An identical dose of rt-PA
prepared by a different production procedure in which a larger
proportion of the product comprised material that had not been
cleaved by plasmin (so-calledsinglechain rt-PA) was usedin open
label phase A (n = 48). Because recanalization rates appear to
be lower on a mg/rng basis with material produced by the newer
procedures. the dose was increased to 100 rng over 3 hours (60,
20 and 20 rng/h, respectively) for open label phase B (n = 88),
It was increased further to 150 mg over 6 hours (90,20, 10, 10.
10 and 10 rng/h, respectively) for open label phase C. In each of
these phases. infusions of rt-PA were initiated only after angiographic demonstration of coronary occlusion had been acquired.
Subsequent open label phases of the TIMI trial employed administration of rt-PA without an initial angiogram and with the first
angiogram performed IR to 4R hours (phase D, n = 42) or either
2 hours or 18 to 48 hours (phase E) after the onset of administration of rt-PA in a total dose of 150 mg over 6 hours
(n = 317), Because phase E was performed in an open label
fashion with vigorous surveillance of possible bleeding complications in light of what was recognized to be a very high dose of
rt-PA. 53 of the 317 patients were treated in this phase after an
adjustment had been made to a lower dose (100 rng over 6 hours;
60,20,5,5,5 and 5 mg/h, respectively) on October 21. 1986.
E
( 3171
D
( .2 )
cs;:sl Ma jor
C
2lI2
-
Cerebra l
Tota l
l ee )
B
( 8 8)
A
( 48)
I- rt - PA
(1 .3)
I- SK
( 14 7 )
( n
)
0
20
40
60
Perc ent
80
100
428
lACC Vol. 10 . No.5
November l'IlP :40ll - ·Hl
SOBEL
SAFETY OF n ·PA
20
100
80
.r:
...
co
•
10
g 60
t::l
r&l
C'"
Transfusions
Hematomas
Bleeding
.s
C
o,
o
SK
(65)
rt ·PA
(64)
Cot h
GI
Figure 5. Hemorrha gic events within 4l:\ hours after infusion of
recombinant tissue-type plasminogen activator (rl-PA) (0.75 mg/kg
over 90 minutes) or streptokinase (SK ) (1 .5 million U over 60
minutes) in patients studied by Verstraete et al. (10) in the European Cooperative Trial. Both the incidence of major bleeding
and the overall incidence of bleeding were less among patients
treated with fl-PA.
Oth e r
Ble ed in g
Figure 3. Compilation of bleeding episodes in 344 patients studied
in TIMI Phases I . A. 8 and C. Most episodes of bleeding were
associated with vascular access sites. Additional information pertinent to bleeding in specific phases oft he TIMI trial and to specific
dose regimens particularly with respect to intracranial bleeding is
summarized in Ref, I ~. As noted in that communication, a 1.6%
incidence rate of intrac ranial bleeding (5 of 311 patients) occurred
among patients treated with 150 mg. In view of this experience,
the dose was reduced to 100 mg for patients treated subsequently
in the TIMI trial. In the remaining 53 patients studied in TIMI
open label phase E who were treated with the 100 mg dose, no
intracranial bleeding occurred. Cath ~ catheterization site: GI =
gastrointestinal tract site.
Figure 6. Patency rates. Comparison of infarc t-re lated coronary
artery patency evident angiographically 90 minutes after the onset
of infusion of recombinant tissue-type plasminogen activator (rtPAl in 429 patients with initially occluded coronary arteries documented by preinterventional angiography in Phase I (40, 20 and
20 mg of I-PA infused over 3 hours) and 2 hours or 18 to 48 hours
after the onset of infusion or rt-PA. to patients without preinterventional angiography but with criteria of evolving transmural
myocardial infarction fulfilling the TIMI entry criteria in open label
phase E of the T IMI trial (90 mg of rt-PA given in the first hour).
Results in both groups (Phase I and phase E) are compared with
the incidence of patency evident in preinterventional angiograms
available for the Phase I patients only. The higher doses of rt·PA
used in open label phase E compared with those used in Phase I
were associated with a somewhat greater rate of early patency.
rigorously and elected to reduce the maximal dose tested
(150 mg ove r 6 hours) to 100 mg given ove r 6 hours based
on experience with 311 patients in open label phase E ( 18) .
This decision was reinforced by [he lack of a detectable
decrea se of efficac y (reca naliz atio n rate for infarct-related
OP E N L ABE L E
PHA S E I
Figure 4. Hbr iuogencl ysis (after Collen et al. [151J. Changes in
plasma fibrinogen after I hour expressed as a percent of control
values with data obtained with either a precipitation (sulfite) (solid
bars) or kinetic (st r jped bars) assay after 60 minutes of treatment
with recombinant tissue-type plasminogen activator (rt -Pa) , streptokinase (SK) or neither agent (control) in 131 patients with acute
myocardial infarction. The number of patients in each group is
indicated in parentheses. Patients evaluated were among those
studied in the European Cooperative Trial (10.15).
100
80
60
~
u
J
120
~
U
L
100
c:
IV
Cl
~o
80
0
.S
.<;
:;:
60
20
c
·E
40
0
<D
o
20
..J-_ _
Pr e rt - PA
0
rt ·PA
(64)
SK
(42)
Control
(25)
N
_
Op en
143
c:::::J
90
m ~n
14 3
Not op en
2 hr
18 -048 h r
33
2 53
~ Not done
SOIlEL
SAFETY OF n-PA
)ACe Vol. 10 . I'll. 5
November Il)g7:4IlH- 4H
100
Death
rzz'2 IAI
Q52lJ Both
80
+ Ie
20
o~-­
Hospital
6 weeks
Figure 7. Mortality and morbidity throughout the hospital course
and throuehout a 6 week interval of fol low-up among 31 7 patients
treated with recombinant tissue-type plasminogen activator ([(- PA)
in open label phase E of the TIMI trial us well as the incidence
of recurrent myocardial infarction (MI) or extension of infarction
during both intervals. Episodes of intracranial bleeding (lC> were
included in the tabulation as deaths. Even with the classification
of such bleeding episodes as " mortality," survival throughout the
hospital course and throughout a 6 week inrerval of follow-upafter
infu sion of rt-PA was unusually high (94'k) for patients with documented acute transmural myocardial infarction.
arteries) with the 100 mg compared with the 150 mg dose
of n-PA over 6 hours and the virtual absence of intracranial
bleeding in patients treated with the 100 mg dose to date.
Among 400 patients treated with 120 to 170 mg of rtPA over intervals comparable with those employed in TIMI
open label E and studied in clinical trials monitored by the
Gcnentech Corporation. the incidence of intracranial bleeding within 48 hours of infusion of rt-PA was < 0.5%. Among
99R patients treated with equipotent regimens of rt-PA
produced by diverse production modes over the past several
years it was only 0.3% (19), Among analogous trials conducted in Europe involving 7H 1 patients it was 0.26% (E.
Grossbard , personal communication).
Th erapeutic failures. When intracranial bleeding and
death are considered together a~ criteria of failure of therapy
with rt-PA, only 5% of 3 17 patients given 150 mg of rt PA were adjudged to have had therapeutic failure throughout
their entire hospitalization in open label phase E of the TIMI
trial. Only 6% were adjudged to have had therapeutic failure
throughout a 6 week interval 01" follow-up (Fig. 7) (E. Passumani. personal communication), Patency occurred in 87lif
of 220 patients. judging from results of angiography 18 to
48 hours after treatment. When the dose regimen in open
label phase E was modified to 100 mg 01" rt-PA over 6 hours.
efficacy remained apparently constant. The incidence of
intracranial bleeding at this dose has been low (none among
the 53 patients treated with 100 mg in open label phase E).
43B
Side effects. Administration of rt-PA does not generally
induce an immune response. Positive results for detection
of antibodies to rt-PA by radioimmunoprecipitation assay
have been seen in only three samples from > 700 patients
tested worldwide (E. Grossbard. personal communication).
In no case were the antibodies neutralizing. In each case.
subsequent samples evaluated 3 weeks to 10 months after
treatment were negative.
Mild. febrile reactions or urt icaria have been encountered
in < 10% of TIMI patients treated. Deleterious hemodynamic effects attributable to rt-PA have not been encountered. Although as many as 40% of patients treated with rtPA have experienced nausea and vomiting. it is difficult to
determine the extent to which these and other nonspecific
phenomena are attributable to rt-PA as opposed to narcotics.
other medications or infarction in progress.
Conclusions
End point results to date from the two largest prospective.
randomized patient assignment. double-blind trials with rtPA (the TIMI trial and the European Cooperative Trial) are
concordant and compatible with those of smaller studies
with native t-PA and rt-PA as well as with those of studies
of rt-PA combined with angioplasty (20- 22). In concert.
the data indicate that: I) lntruvenou s rt-PA elicits recanalization promptly in > 70% or patients-a success rate comparable with that documented with maximal doses of intracoronary streptokinase: 2) It-PA depletes circulating fi brinogen
and elevates circulating fi brinogen degradation products much
less than does streptokinase. as reflected by results of prospective, randomized, clinical trials (1 5.23); 3) bleeding
associated with administration of rt-PA under research protocol conditions generally involving large doses of heparin
concomitantly is confined generally to vascular access sites
and is readily manageable; and 4) early treatment with rtPA is associated with remarkably low hospital and 6 week
mortality among patients with documented acute transmural
myocardial infarction.
Assi' lance in the preparation or the typescript hy Lori Dales i.,appreciated.
References
I . Bergmann SR. Fox KAA. Ter-Pouossian MM, Sohel BE (Washington
University). Collen D (University of Leuven). Clot-selective corona ry
thrombolysis with tissue-type plasminogen activator. Science IIJX3;220:
lllll-3.
2. Van de We rtF. Bergmann SR. £OfIX KAA. ct al. Coronary thrombo lysis
with intravenously administered human ti....uc-type plasm inogen artivalor produced by recombinant DNA technology. Circulation IIJiS4:hIJ:
h05 10
3 . V,Hl de Wcrf F. Ludbrook PA. Bergmann SR. ct al. Coronar y thrombolysis with tissue -type plasminogen activator in patients with evolving myocardial infarction . N Engl J Med 1 9 84;~ 10:609·-13.
448
SOBEL
SAFETY OF rt-PA
4. Collen D. Topol EJ. Tiefenbrunn AJ. et al. Coronary thrombolysis
with recombinant human tissue-type plasminogen activator: a prospective. randomized. placebo-controlled trial. Circulation 1984;70:
1012-7.
5. Tictcnbrunn AJ. Robison AK, Kumik PB, Ludbrook PA. Sobel BE.
Clinical pharmacology in patients with evolving myocardial infarction
of tissue-type plusminopen activator produced by recombinant DNA
technology. Circulation 1985:71 : 11 0-6.
6. TIMl Study Group. Special report: the Thrombolysis in Myocardial
Infarction (TIMI) trial. :-I Engl J Med 1985:31 2:932- 6.
7. Hillis LD, Borer 1. Braunwald E. et al. High dose intravenous streptokinase for acute myocardial infarction: preliminary results of a multicenter trial. J Am Coil Cardiel 1985:6:957-62.
8. Williams DO. Borer 1. Braunwald E, et al. Intravenous recombinant
tissue-type plasminogen activator in acute myocardial infarction: a
report from the NHLBI Thrombolysis in Myocardial Infarction (TIM!)
trial. Circulation 1986;73:338-46.
9. Verstraete M, Bleiteld Vi, Brower RW, ct al. Double-blind randomized trial of intravenous tissue-type plasminogen activator versus placebo in acute myocardial infarction. Lancet 1985:2:965-9.
10. Verstraete M. Bernard R. Bory M. et al. Randornised trial of intravenous recombinant tissue-type plasminogen activator versus intravenous streptokinase in acute myocardial infarction. Lancet 1985; I:
842-7 .
II . Chesebro JH. Knalterud G, Roberts R. et al. Thrombolysis in Myocardial Infarction (TIMI) Trial. Phase I: a comparison between intravenous tissue plasminogen activator and intravenous streptokinase.
Clinical findingsthrough hospitaldischarge. Circulation 1987;76:142- 54.
12. Satian RD. Sheehan FH. Markis JE. et al, Hemodynamics and structural effects of late thrombolysis (abstr). Circulation 1986:74(suppl
11 ):11-367.
13. Sheehan FH. Braunwald E, Canner P, ct al. The effect of intravenous
thrombolytic therapy on left ventricular function: a report on tissuetype plasminogen activator and streptokinase Irum the Thrombolysis
lACC Vo!' 10. No . 5
November 1987:408 - 4B
in Myocardial Infarction (TIMI Phase I) trial. Circulation 1987:75:
817-29 .
14. Mueller HS. Rao AK. FormanSA. and the T1M l lnvestigators. Thrombolysis in Myocardial Infarction CrIMI): comparative studies of coronary reperfusion and systemic fibrinogenolysis with two forms of
recombinant tissue-type plasminogen activator. J Am Coli Cardiel
1987:10:479-90.
15. Collen D. Bounameaux H, Oc Cock F. Lunen HR. Verstraete M.
Analysis of coagulation and fibrinolysis during intravenous infusion
of recombinant human tissue-type plasminogen activator in patients
with acute myocardial infarction. Circulation 1986;73:511- 7.
16. Passaman i E. Hodges M. Herman M. et a!' The Thrombolysis in
Myocardial Infarction (TIM1) Phase II Pilot: tissue plasminogen activator followed by percutaneous transluminal coronary angioplasty.
J Am Coli Cardiel 1987;10(suppl.):5I B- 64B.
17. Tiefenbrunn AJ, Graor RA. Robison AK. Lucas FV. Sobel BE. Pharmacodynamics of tissue-type plasminogen activator (t-PA) characterized by computer-assisted simulation. Circulation 1986:73: 1291 -9 .
18. Braunwald E. Knatterud GL. Passamani J::R , Robertson TL. Announcement of protocol change ill Thrombolysis in Myocardial Inturction Trial (letters). J Am Coli Cardial 1987;9:467.
19. Grossbard EB. Genentech experience with rt-PA (Activase) (letter).
J Am Coli Cardiol 1987:9:467.
20. Sobel BE. Collen D. Grossbard EB. eds. Tissue Plasminogen Activalor in Thrombolytic Therapy. New York: Marcel Dekker. 1987.
21. Sobel BE. ed. Thrombolysis and the Heart. Philadelphia: WB Saunders. 1987.
22. Collen D. Lijnen HR. Verstraete M. eds. Thrombolysis: Biological
and Therapeutic Properties of New Thrombolytic agents. Edinburgh:
Churchill Livingstone. 1985.
23. Rao AK. Pratt C. Berke A. et al. The Thrombolysis in Myocardial
Infarction Trial. Phase I: Hemorrhagic manifestations, complications.
and changes in plasma fibrinogen and fibrinolytic system. J Am Coli
Cardiol 1988;I I:(in press).