Streptokinase Resistance: When Might Streptokinase Administration Be Ineffective?

Br Heart J 1992;68:449-53 449 Streptokinase resistance: when might streptokinase administration be ineffective?

Maurice B Buchalter, Ganesh Suntharalingam, Ian Jennings, Catherine Hart, Roger J Luddington, Ronjon Chakraverty, S Kim Jacobson, Peter L Weissberg, Trevor P Baglin

Abstract patients after 24 months. Retreatment Objective-(a) To develop an assay for with streptokinase is likely to be sub- streptokinase resistance. (b) To deter- optimal even after 24 months. The fibrin mine the prevalence of streptokinase plate lysis assay detects resistance in resistance in patients presenting with patients with normal concentrations of acute myocardial infarction for the first streptokinase antibodies. Streptococcal time. (c) To determine the prevalence of infection is associated with a high streptokinase resistance in patients after incidence of streptokinase resistance. exposure to streptokinase or streptococcal infection. (Br Heart J 1992;68:449-53) Design-Open, prospective. Patients-30 healthy volunteers. 40 Thrombolysis has become an essential com- patients admitted to the coronary care ponent of the management of an acute unit at Addenbrooke's Hospital with sus- myocardial infarction. Large controlled trials pected acute myocardial infarction, 12 have shown that streptokinase compares patients 12 months after streptokinase favourably with other fibrinolytic agents in treatment, eight patients 24 months terms of efficacy, side effects, and cost.'-5 It is after streptokinase treatment, and sera therefore likely to remain the thrombolytic from 12 patients with raised anti- agent of choice in the United Kingdom for the streptolysin 0 (ASO) titres. foreseeable future. Because of its antigenic Methods-Three assays were used; a nature, however, the presence of neutralising dilution neutralisation a,ssay, an enzyme antibodies in some patients may reduce its linked immunosorbent assay (ELISA) effectiveness. Rapid development of immuno- for immunoglobulin G (IgG) anti- globulin G (IgG) antibodies to streptokinase streptokinase antibodies, and an in vitro has been reported,6 and antibodies have been fibrin plate lysis assay. All measurements detected up to eight months after treatment.7 were performed on venous blood Current recommendations advise against samples. retreatment with streptokinase within six Results-Neutralisation and IgG months and immediately after a streptococcal antibody titres were positively infection.89 The prevalence of antistrepto- correlated. Mean (SEM) Clinical antistreptokinase antibodies in the general population Pharmacology Unit, kinase concentrations in the 30 controls and in patients presenting to the coronary care School of Clinical were 87 (10) U/ml (neutralisation assay) unit for the first time is unknown, and their Medicine, University and 28 (6 3) U/ml (ELISA). Correspond- to neutralise a standard of Cambridge potential dose of M B Buchalter ing concentrations in patients before streptokinase has not been rigorously G Suntharalingam streptokinase were 68 (6-1) U/ml and 18 evaluated. R Chakraverty (4 5) U/ml with a mean fibrin plate assay To test whether the of anti- P L Weissberg development 117 (7-1)% that of controls. Resistance to bodies is associated with significant strepto- Clinical Microbiology streptokinase was detectable in and Public Health one kinase resistance we have developed an in Laboratory, patient after 72 hours and in all patients vitro fibrin plate lysis assay and correlated this Addenbrooke's by day 10. By day 10 concentrations were with both functional and immunological Hospital, Cambridge 4388 (919) 773 (109) and 17 concentrations of S K Jacobson U/ml, U/ml, streptokinase antibodies. A (5 4)%. At both 12 and 24 months resis- normal range for neutralising activity was Department of Haematology, tance was present in 75% of patients. established in a group of normal subjects. Addenbrooke's Similarly 66% of high ASO titre sera Patients admitted to a coronary care unit with Hospital, Cambridge showed resistance. The fibrin plate lysis suspected myocardial infarction were inves- I Jennings detected R J Luddington assay significantly reduced tigated on arrival and, to determine how soon T P Baglin streptokinase dependent fibrinolysis in and for how long neutralising activity was C Hart vitro in the absence of raised total con- present, blood samples were taken early (three Corrcspondencc to centrations of antistreptokinase anti- to ten days) and late (12 and 24 months) after M B Buchalter, Dcpartrnent of Clinical bodies. streptokinase treatment. We also tested sera Pharmacology, Conclusions-The prevalence of strep- from patients with raised antistreptolysin 0 F and G Block, Addenbrookcs Hospital, tokinase resistance in patients present- (ASO) titres for streptokinase neutralising Cambridgc CB2 2QQ. ing with their first myocardial infarction activity to confirm whether recent strepto- Acccpted for publication is low. Resistance develops early after coccal infection precluded administration of 21 April 1992 treatment and is still present in 75% of streptokinase. 450 Buchalter, Suntharalingam, Jennings, Hart, Luddington, Chakraverty, Jacobson, Weissberg, Baglin

Patients and methods lyse the clot was used to calculate the strepto- PATIENTS AND CONTROLS kinase neutralising titre. Patient samples were Over a six week period, venous blood samples diluted in normal plasma until clot lysis was were taken before administration of strepto- achieved. The neutralising activity (U/ml) was kinase, from 40 patients admitted to the calculated as:- highest concentration of strep- coronary care unit of Addenbrooke's Hospital tokinase failing to lyse clot x dilution of test for the first time with a suspected acute myo- plasma in normal plasma/10. A concentration cardial infarction. Further samples were taken > 2 SDs above the mean ofthe normal controls on days three to four and on the day of was defined as raised. discharge (days seven to 10) from 15 patients who had received 1-5 million units of intra- FIBRIN PLATE LYSIS ASSAY venous streptokinase (Streptase, Hoechst). Fibrinogen (Diagen) was dissolved in None of the patients had previously received imidazole buffer to a concentration of650 mg/l. streptokinase. Venous blood was also taken Fibrin plates were prepared by clotting 10 ml from (a) 12 patients who had received strepto- fibrinogen solution with 10 U bovine thrombin kinase 12 months previously, (b) eight patients (Diagen) and 25 mM calcium chloride in petri who had received streptokinase 24 months dishes (Sterilin). Streptokinase was added to previously, (c) 12 patients with raised anti- the plasma samples to achieve a concentration streptolysin 0 titres, (d) 30 healthy normal of 625 U/ml. This was the calculated plasma controls with no documented history of strep- concentrations of streptokinase that would tococcal infection. All patients gave informed result from a dose of 1-5 million units to a 70 kg verbal consent, and the study was approved by subject with a packed cell volume of 45%. the local ethics committee. Twenty five p1 of each plasma sample were placed on a fibrin plate and incubated at 37°C MEASUREMENT OF STREPTOCOCCAL ANTIBODIES for 24 hours. The area of lysis was calculated as Immunoglobulin G antibodies to streptokinase the product of two diameters at 900 to each were measured by an enzyme linked immuno- other. The mean area of lysis of the 30 normal sorbent assay (ELISA) incorporating solid controls was 1394 mm2. All areas oflysis caused phase streptokinase. Neutralising antibodies by patients plasma with added streptokinase were measured by a dilutional clot lysis assay. were then reported as percentages of this value. (a) ELISA assay: 100 p1 of 1000 U/ml A reduction in streptokinase dependent in vitro streptokinase (Streptase, Hoechst) was added fibrinolysis >2 SDs below the mean normal to the wells of a microtitre plate (Nunc, activity was regarded as abnormal. Denmark). Plates were sealed, incubated over- night at room temperature, and washed three ASO TITRES times in 0-01 mol/l phosphate buffered saline Titres of ASO were measured by a standard (PBS) and 0-5 ml/l Tween. One hundred p1 of sheep red cell haemolysin assay (Wellcome plasma, diluted one in 40 with phosphate Diagnostics). buffered saline (PBS)/0- 1% bovine serum albumin, was then added to each well and STATISTICAL ANALYSIS incubated for two hours at room temperature. Comparisons between study groups and the After three washes the wells were incubated normal controls were performed with the Wil- with 100 p1 peroxidase-conjugated rabbit anti- coxon's ranking test for unpaired data (Mann- human IgG (Dako) diluted 1/5000 in PBS/ Whitney test). The same test was used for Tween for two hours at room temperature. comparisons in the group tested on several days After washing, 100 p1 substrate solution (3 mg/ as the numbers tested on each day varied. ml orthophenylenediamine in 0-01 mol/l citrate phosphate buffer, with 0 01% hydrogen perox- ide was added and the reaction stopped by the Results addition of 150 p1 sulphuric acid. Absorbance The mean age of the 40 patients (28 men) at 492 nm was read with a Titertek Multiscan admitted to the coronary care unit was 58 plate reader. A standard curve was derived (range 38-71) years. All patients received 1-5 from doubling dilutions ofnormal plasma from million units of streptokinase by a one hour 1/20, assigned 100 U/ml and patient samples intravenous infusion. None of the patients were sufficiently diluted to enable a concentra- suffered an allergic reaction. Venous blood tion of antibody to be found from the standard samples were taken from 15 patients on day 3 curve. The concentration of antibody in the and on the day of discharge (days 7/8 in seven patient sample was then multiplied by the patients and days 9/10 in eight patients). appropriate dilution factor. A concentration The ELISA assay and the neutralisation titre > 2 SDs above the mean of the normal controls assay produced comparable results and were was defined as raised. positively correlated (r = 0-55; p < 0-001). (b) Neutralisation assay: streptokinase Overall neither assay was correlated with the neutralising activity was measured by a method fibrin plate lysis assay. Between days 1 and 10 similar to that previously reported.7 Strepto- after streptokinase, however, a reduction (more kinase was diluted in saline to give a range of than 2 SD below mean normal) in fibrin plate concentrations from 500 U/ml to 2000 U/ml. lysis was associated with antistreptokinase IgG Twenty p1 of each dilution was added to 200 p1 concentrations above 155 U/ml and a neutral- plasma and a fibrin clot formed by the addition isation titre above 375 U/ml. The table shows of 1 U bovine thrombin (Diagen). The highest the normal ranges and the concentrations concentration of streptokinase that failed to found in each patient group for each assay. Streptokinase resistance: when might streptokinase administration be ineffective? 451

Mean (SEM) in each study groupfor each assay

Serial values in treated patients (n = 15) Patients Patients Normal Admitted at 12 at 24 High ASO controls patients Day 0 Day 3/4 Day 7/8 Day 9/10 months months titre sera Method (n = 30) (n = 40) (n = 15) (n = 15) (n = 7) (n = 8) (n = 12) (n = 8) (n = 12) Neutralising titre 87 - 68 375*** 2053*** 4388*** 150*** (U/ml) (10 0) (6.1) (127) (818) (919) (32-3) - - ELISA IgG 28 17 18 38 311*** 773*** 120*** 140** 177** antistreptokinase (6-3) (3-4) (4-5) (30-7) (133) (109) (24-9) (27-6) (44-3) (U/ml) Fibrin platelysis 100% - 117% 108% 34%*** 17%*** 43%*** 40%** 50%** assay as % of mean normal (4 0%) (7.1%) (14 1%) (15 6%) (5-4%) (9 7%) (9.1%) (10 1%) **p < 0-01; ***p < 0 001.

Fibrin plate lysis in the 30 normal controls being given streptokinase (mean age 56, range (mean age 49, range 34-64 years, 10 women) in 36-70 years, two women), only three had a the absence of streptokinase was less than 2% raised neutralisation titre, but six had raised of that in the presence of streptokinase. IgG concentrations. All six of these had None of the patients admitted to the reduced fibrin plate lysis values (range coronary care unit had raised concentrations of 10%-35%). An additional three patients, IgG antistreptokinase antibodies on admission. however, with IgG concentrations within the In 15 of these patients neutralisation titres and normal range also had impaired fibrinolysis fibrin plate lysis values were also measured. (10%, 32%, 38%). Only three patients had no None had evidence of either neutralising evidence of streptokinase resistance by any antibodies or impaired in vitro fibrinolytic method. responses to streptokinase. Of the eight patients studied at 24 months By days 3/4 (n = 15) the neutralising titre (mean age 53, range 43-62 years, two women), had risen above the normal range (>2 SDs six had reduced fibrin plate lysis (range above the mean) in nine patients (range 200 to 3%-24%). Five of these six had raised IgG 2000 U/ml). Eight of these, however, showed concentrations. no increase in antistreptokinase IgG by ELISA Of the 12 sera from patients with raised and no reduction in fibrinolytic activity in the ASO titres, six had raised ELISA IgG concen- fibrin plate lysis assay. One patient had a rise in trations and all of these, plus two with neutralising titre to 2000 U/ml and an ELISA concentrations within the normal range, had IgG of 437 U/ml. This was associated with a reduced fibrin plate lysis values (range reduction of in vitro fibrinolysis to 6% of 9%-64%). Figure 3 shows the individual and control values. mean fibrinolytic responses ofeach ofthe study By days 7/8 (n = 7) the neutralisation titre groups. was above the normal range in all patients and IgG antibodies were above normal in all but two. Only the patients with raised IgG Discussion antibodies had significantly reducedfibrinolytic None of the patients admitted with acute responses. By day 10 (n = 8) all patients had myocardial infarction had raised concentra- raised antibody concentrations by both assays tions of antistreptokinase antibodies or in vitro with reduced fibrinolytic responses to a mean fibrinolytic resistance to streptokinase. The value of 17% (range 1% to 54%) of controls. prevalence of antistreptokinase antibodies in The day 7-10 results were significantly different patients presenting to the coronary care unit for from the normal control and day 0 results for the first time is therefore low and most should each of the three assays (p < 0-001). Figures 1 and 2 show the IgG antistreptokinase concen- 300 - trations and the fibrinolytic responses over the

10 day study period. - Of the 12 patients studied 12 months after 250

200 Figure 1 1500 - Antistreptokinase ._4 (Anti-Sk) IgG antibody a) 150- concentrations measured * in 1200 * C by ELISA over the 10 day . period after 1 5 million *C 100 units of intravenous E * D0 I * L 0 streptokinase in 15 (D9oo 0 patients (allpatients were 50- 0 tested on day 0, days 3/4 0 and days 7/10 and many 9 A points '? 600- 0 vI I - overlap). 0 Horizontal line indicates c 0 2 4 6 8 10 mean (+2 SDs) for 0 Day normal control group. 3o0_J- I Figure 2 Fibrin plate lysis (%) relative to mean normal result over the 10 days after 1-5 million units of intravenous streptokinase in 15 patients (all patients were 0 -* tested on day 0, days 314 and days 7/10 and many points 0 2 4 6 8 10 overlap). Horizontal lines indicate mean (2 SDs) for Day normal control group. 452 Buchalter, Suntharalingam, Jennings, Hart, Luddington, Chakraverty, Jacobson, Weissberg, Baglin

Figure 3 Individual caused reduced fibrin lysis in the absence of values (point) and mean raised streptokinase antibodies by either assay. values (bar) offibrin plate Thus a normal IgG concentration > 10 days lysis in each of the study 200. groups. Valuesfor many after treatment does not necessarily ensure a patients overlap. normal fibrinolytic response. This discrepancy

a) between antibody concentrations and impaired * fibrinolysis has not been explained. One a C possibility is that there is reduction in total -0 100 s antibody concentrations over time but main- _ * D | tenance of high affinity antibodies. A study of o- * the concentrations of antistreptokinase *- * immunoglobulin subclasses after treatment *. -, | is being undertaken to examine this - *. . phenomenon. 0 Normals 0 3 37 7101012 24 High To understand fully the clinical relevance of I,-% - ASO reduced in vitro fibrinolysis a comparison Days Months between the measurements made in this study and a clinical marker ofeffective thrombolysis is needed. At present no reliable non-invasive marker exists. In a recent study, however, respond to a standard dose of streptokinase. Massel et al showed that the thrombolysis This is despite the fact that many of them will induced by streptokinase in an external jugular have encountered streptococci in the past. vein thrombosis model in rabbits was sig- By days three to four there had been a slight nificantly impaired by giving streptokinase to increase in neutralising antibody concentra- the rabbits one month previously.'0 The tions; however, in only one patient was this pretreated animals had raised IgG antistrep- substantial enough to cause a reduction in tokinase titres. This study suggests that there is streptokinase induced fibrinolysis. Thus a close correlation between in vitro resistance retreatment with streptokinase within 72 hours and impaired thrombolysis in vivo. of the initial dose is likely to be successful in The results of our study suggest that up to most patients but may result in impaired 75% ofpatients treated once with streptokinase thrombolysis in some. By day seven most will -be resistant to further treatment with patients had streptokinase resistance and by streptokinase containing.agents after two years. day 10 all patients had resistance. This time This suggests that the outcome from retreat- course suggests a primary rather than ment with streptokinase containing agents is secondary antibody response to streptokinase. likely to be unpredictable for a period begin- Patients studied 12 months after a dose of ning four to seven days after streptokinase streptokinase had varying concentrations of doseage and lasting for more than two years. antibodies. Nevertheless, nine (75%) of them Other data suggest that this period may last for still had significantly impaired fibrin plate lysis. at least four years." To identify patients who Of the eight patients studied at 24 months six would benefit from repeated treatment with (75%) still had impaired fibrin plate lysis. streptokinase, antibody concentrations or Therefore, roughly three quarters of patients preferably fibrin plate lysis, could be measured might be expected to have some degree of routinely at regular intervals after first treat- resistance to streptokinase, even after 12 or 24 ment. Such information, however, would have months. Further information on the effects of to be available immediately on admission with a streptokinase resistance on drug readministra- further infarct. The only way to ensure this tion is urgently required. Similarly, roughly would be for the patients to carry this informa- two thirds of patients with serological evidence tion themselves. Unfortunately, none of the of recent streptococcal infection are likely to tests used in this study could be performed have impaired thrombolysis with streptokinase rapidly enough at the bedside to be used on (eight of 12 patients). The precise temporal admission. We believe that titrating the dose relation between streptococcal infection and upwards on the basis ofthe results ofan in vitro streptokinase resistance also warrants further test is an impractical approach that will only study. increase the antigenic load and the risk of an A reduction of in vitro fibrinolysis might adverse immunological reaction. reasonably be expected to predict a decrease in In conclusion, streptokinase is likely to thrombolytic activity in vivo. However, it was remain a first line agent for thrombolytic not possible in this study to determine to what treatment in acute myocardial infarction. The extent the two are correlated. In the 10 days prevalence of streptokinase resistance at first after a dose of streptokinase raised concentra- presentation is low unless there is evidence of tions of antistreptokinase antibody measured recent streptococcal infection. The best by the neutralisation assay were not necessarily predictor ofstreptokinase resistance is the fibrin associated with reduced in vitro fibrinolysis. By plate lysis assay. Resistance can develop within contrast, a raised ELISA IgG titre was always 72 hours, although most patients seem to have a associated with reduced fibrinolysis. In these primary antibody response developing over patients an increase in IgG antibodies is likely seven to 10 days. All patients are resistant after to predict impaired thrombolysis. Serum from 10 days and roughly 75% remain resistant after five patients, (three 12 months after treatment 24 months. Retreatment with thrombolytic and two with a high ASO titre) however, agents containing streptokinase is inadvisable Streptokinase resistance: when might streptokinase administration be ineffective? 453

until the full clinical relevance and time course 4 The International Study Group. In-hospital mortality and clinical course of 20,891 patients with suspected acute of streptokinase resistance is established. myocardial infarction randomised between alteplase and streptokinase with or without heparin. Lancet 1990; 336:71-5. PLW is a British Heart Foundation Senior Research Fellow. 5 The Third International Study of Infarct Survival (ISIS 3). Meeting of the American College of Cardiology. Atlanta: Am Coll Cardiol, 1991. 6 Lynch M, Littler WA, Pentecost BL, Stockley RA. Anti- 1 Sheehan FH, Braunwald E, Canner P, et al. The effect of streptokinase titres after intravenous streptokinase (Itr). intravenous thrombolytic therapy on left ventricular func- Lancet 1990;693:534. tion: a report on tissue type plasminogen activator and 7 Jalihal S, Morris GK. Antistreptokinase titres after streptokinase from the thrombolysis in myocardial infarc- intravenous streptokinase. Lancet 1990;335:184-5. tion (TIMI Phase I) trial. Circulation 1987;75:817-29. 8 ABPI data sheet compendium 1990-9 1. Datapharm publica- 2 White HD, Rivers JT, Maslowski AH, et al. Effect of tions 676: (Data sheet Streptase). intravenous streptokinase as compared with that of tissue 9 ABPI data sheet compendium 1990-91. Datapharm publica- plasminogen activator on left ventricular function after tions 782: (Data sheet Kabikinase). first myocardial infarction. N Eng JMed 1989;320:817-21. 10 Massel D, Turpie AGG, Cairns JA, Ofosu FA, Buchanan 3 Gruppo Italiano per lo Studio della Sopravvivenza nell, MR. Previous streptokinase (SK) therapy inhibits sub- Infarto Miocardico. GISSI-2: a factorial randomised trial sequent SK thrombolysis. Circulation 1991;84:I1-467. of alteplase versus streptokinase and heparin versus no 11 Elliot JM, Cross DB, Cederholm-Williams S, White HD. heparin among 12,490 patients with acute myocardial Streptokinase titers 1 to 4 years after intravenous infarction. Lancet 1990;336:65-71. streptokinase. Circulation 1991;84:II -1 16.