| Monoclonal Antibodies to Endotoxin: A Proposed Treatment for Hemorrhagic Shock Research performed and paper written by Amy Barcia, Ivy Lee Nip, Anne Sharp, Douglas Soderdahl and Robert Theaker at Queen's Medical Center, Cardiovascular Research Lab, Honolulu, Hawaii, Summer 1985. |
| Abstract Thirteen New Zealand white rabbits with coliforms in the gut were subjected to a hemorrhagic shock of 35 mmHg for 3 hours. Treated rabbits were resuscitated with 2.5 cc of a 20% human monoclonal antibody solution (thawed), remaining shed blood, and lactated Ringer's to achieve a mean arterial blood pressure (MAPB) within 20% of baseline levels. Control rabbits were similarly resuscitated but received 2.5 cc of a 20% fetal calf serum instead of the monoclonal antibody solution. Catheters were removed and rabbits returned to their cages until death or seven days post experimental survival. Overall, hemodynamic parameters including heart rate, femoral artery pressure, central venous pressure, and cardiac output did not differ significantly between groups. However, overall total peripheral resistance demonstrated significance (p<0.05). None of the treated rabbits survived beyond day 2 (0%); and only 2 of the 10 control rabbits survived 7 days (20%). Figures for survival rates did not demonstrate significance. Our data suggests that survival rate during hemorrhagic shock is not increased with the usage of the human monoclonal antibody solution. Despite restoration of hemodynamic parameters after severe blood loss during hemorrhagic shock, multiple system organ failure is a highly lethal and secondary complication. Evidence has been accrued that multiple system organ failure follows endotoxin poisoning. Endotoxin is a lipopolysaccharide composed of oligosaccharide side chains, a core polysaccharide and a lipid A moiety. (4) During hemorrhagic shock, the ability of the reticuloendothelial system to detoxify endotoxin is impaired, resulting in spillover of endotoxin into the systemic circulation. (1-3) In a study investigating the role of endotoxin in hemorrhagic shock, Pohlson et al. determined that the administration of antiserum with specific antiendotoxin activity in rabbits during reperfusion increased survival (6). In order to fully realize the clinical significance of their study, it is necessary to establish that antibodies are the specific components of the antiserum which neutralize the harmful effects of the endotoxin. The use of monoclonal antibodies (MCA) instead of antisera was based upon specificity of the MCA and the diminished likelihood of rejection by the immune system. If it does prove effective, MCA can be used to a greater extent and with less risk than antisera. Materials and Methods Thirteen coliform positive New Zealand white rabbits (10 control, 3 treated) were anesthetized via ketalar (ketamine hydrochloride, 50 mg/ml) subcutaneous injection and allowed to respire spontaneously. Sedation prior to shock was maintained at a minimum level with ketamine and valium (diazepam, 5mg/ml) in combined dosages of 0.2 ml through an ear vein catheter. Rabbits were secured in a supine position on a heating pad used to maintain normothermia. Femoral cutdowns were performed and all four vessels were cannulated to monitor femoral artery pressure (FAP), central venous pressure (CVP), body temperature, and cardiac output (C.O.). FAP and CVP were deteced by Gould Stratham strain-guage transducers (model p23D6) and recorded on a Hewlett-Packard 4-channel strip recorder (model 7754A). A 3 cc Foley Bardex catheter was used to obtain urine samples which were then assessed for pH, protein, glucose, ketones, bilirubin, urobilinogen, blood, and specific gravity. All lines were periodically flushed with heparinized saline (4 NIH units heparin/ml). Endotoxin levels were measured by venous blood samples. After baseline values (arterial blood gas, urinalysis, heart rate, blood pressure [pulsate and mean], cardiac output and endotoxin levels) were recorded, blood was withdrawn to obtain a mean arterial blood pressure (MABP) of 35 mmHg over 15 minutes and stored with 2 cc heparin/50 cc blood. Aliquots of blood were either withdrawn or infused to maintain a MAPBP of 35 mmHg for 3 hours. At every half hour, hemodynamic values were recorded, lines flushed, and cardiac output values obtained. Upon the completion of 3 hours, 2.5 cc of a 20% (o.5 cc + 2.0 cc saline) of fetal calf serum were infused into the 10 control rabbits (CR); and 2.5 cc of a 20% (0.5 cc + 2.0 cc saline) of thawed (stored normally at -70 degrees C) monoclonal antibody solution (obtained from Dr. Nelson Teng of Stanford University) were infused into the treated rabbits (TR). Remaining shed blood and lactated Ringer's were infused to attain a MABP within 20% of baseline or until a maximum of 100 cc had been given. The rabbits were then given a regular I.V. infusion (2x a normal rate of 4.2 cc/kg/hr) for 1 hour followed by a bolus of 3x normal rate. The thermistor and all catheters were removed and the ligated vessels were tied off. The underlying tissue and skin were sewn up with 2-0 and 3-0 silk respectively. The rabbits were then monitored twice daily or until death occurred (7 day maximum). All animals were autopsied to assess the status of the internal organs and to insure that death was not secondary to experimental manipulation. Statistical Methods: T test and Z test were utilized to assess survival data. Hemodynamic parameters were evaluated using analysis of variance methods. Results Comparison of groups: The mean weight of the control rabbit group (2.83 kg +/- 0.103 SEM). Maximum shed volume measured during the shock period (CR = 83.70 ml +/- 3.444 SEM; TR = 86.67 ml +/- 2.598 SEM) also did not significantly differ. Blood re-uptake levels for the control group (30.30 ml +/- 6.894 SEM) did not differ significantly from those for the treated group (18.00 ml +/- 5.854). Hemodynamic parameters including heart rate, FAP and CVP were analyzed. "P" values rendered suggested that overall, the differences between the control and treated groups were insignificant. Heart rate statistics demonstrated no significance between control and treated groups' values. Post hemorrhage rates were slightly lower than that of baseline readings. FAP values remained slightly below baseline levels post resuscitation. Though this was not quantitatively significant (i.e. overall), at 2.5 hours post shock, the control mean FAP value (35.10 mmHg +/- 0.348 SEM) was significantly higher than that of the treated group (33.66 mmHg +/- 0.329 SEM) (p<0.1). CVP statistics yield little variance between control and treated groups. Post resuscitation values remained above normal baseline values with a close relation between control and treated groups. C.O. values measured for both groups-- control and treated remained within a close range of each other. No overall significance was demonstrated; however, at 3 hours post shock significant variance was exemplified (CR = 202.55 +/- 24.599 SEM; TR = 439.33 +/- 162.455 SEM) (p<0.05). Overall total peripheral resistance (TPR): changeP/C.O. = (FAP - CVP)/C.O. demonstrated significance (CR = 0.205 +/- 0.018 SEM; TR = 0.150 +/- 0.035) (p<0.05); however, individual hourly computations did not produce any significant differences. Survival: 4 control rabbits survived beyond day one 940%). 2 control rabbits survived until sacrifice on day 7 (20%). 1 treated rabbit survived beyond day one. Final results showed that all three treated rabbits died before day two (0%). The human monoclonal antibodies used in this experiment did not significantly increase survival rates of rabbits suffering form hemorrhagic shock. Discussion Zieglar et al. demonstrated that the administration of antisera containing antibodies to core lipopolysaccharide improved the survival rate of patients with gram-negative bacterial sepsis. (3) Levels of endotoxin in the systemic circulation are presumed to rise following hemorrhagic shock although there is little data documenting direct measurement of endotoxin release. Pohlson induced a hemorrhagic shock of 36 mmHg for three hours in rabbits, resuscitating treated rabbits with rabbit J5 antiserum and controls with normal rabbit serum (6). Treated rabbits had a 60% survival while no control rabbit lived past the third post-experimental day. Antiserum significantly increased post-hemorrhagic shock survival. It would seem logical to assume that the active component of the antiserum used in the previous study was an antibody specific for endotoxin. This being the case, we proceeded with this experiment utilizing monoclonal antibodies in lieu of antiserum as treatment to hemorrhagic shock and the possible secondary complication. Our hypothesis was that antibodies bind to the endotoxin and neutralize the toxicity and damaging effects on the reperfused tissues and are then cleared by the immune system. However, our data does not support this apparently logical conclusion. Though at certain hours within the statistical analysis there were signs of significance, e.g. 2.5 hours post shock in the FAM (p<0.1), these were taken as flukes within the data. It appeared that these indicated differences were largely due to the small population involved. The significance demonstrated for the overall TPR was quite puzzling. This may have been due to the fact that the control group involved 10 animals as opposed to the 3 of the treated group. Most importantly, however, there was no significant statistical difference between control and treated rabbits in terms of survival. This was largely due to the fact that only three rabbits were treated with monoclonal antibodies as compared with ten controls. Three hypotheses have been formulated to explain the difference between survival results obtained by Pohlson (6) and ourselves. The inconsistency may be due to the source of the monoclonal antibodies, monoclonal antibody dosage size, or that monoclonal antibodies may not be the effective portion of antiserum. In this experiment, human antibodies were employed unlike in Pohlon's experiment (6) where rabbit antiserum was used. This carries several implications. The rabbit's immune systems may have recognized the antibodies themselves as foreign and cleared them before endotoxin neutralization occurred. Also, the antibody-endotoxin complex may have not been cleared at all. This could be due to a difference in the Fc portion of the human antibody which may be unrecognizable to the rabbits' immune systems. In addition, rabbits are polyclonal antibody formers for the same antigen, so the antiserum that Pohlson(6) used on her rabbits contained several types of antibodies to endotoxin. We administered genetically identical antibodies (monoclonal antibodies). Perhaps a combination of various clones is required for the rabbits' immune systems to recognize the antibody-endotoxin complex or to be able to completely clear it. However a larger dose of monoclonal antibodies may be able to compensate for this problem. To date, there are no means of determining the appropriateness of the monoclonal antibody dosage administered as treatment in these trials. Although this amount was advised by Dr. N. Teng of Stanford University, it may have been insufficient in dosage alone to clear the endotoxin, limiting its beneficial effects. Another explanation is that the antibodies are not the effective component of antiserum against the secondary complications of hemorrhagic shock. If it were possible to fractionate antisera, trials could be run using its other elements. This would e3stablish whether a portion of the antiserum is solely effective or if the entire antiserum is necessary. Reproducing Pohlson's work in order to verify her results may be useful in determining the efficacy of monoclonal antibodies. Variations of this experiment should entail larger doses of monoclonal antibodies, rabbit antibodies, and possibly other elements of rabbit antisera. They should be executed on a greater number of trial rabbits for a clearer representation of results. By no means should the postulated efficacy of monoclonal antibodies be disregarded in view of the expiration of three trial rabbits. A means to accurately appropriate monoclonal antibody dosages and to fractionate antisera and documented evidence that endotoxic shock is s secondary complication with hemorrhagic shock should be available before further work is done in this direction. References 1. Carey F.J., Braude A.I., Zalesky M.: Studies with radioactivity with endotoxin. III. The effects of tolerance on the distribution of radioactivity after intravenous injection of Escherichia Coli endotoxin labeled with 51Cr. J Clin Invest. 1958; 37:441. 2. Cardis D.T., Reinhold R.B., Woodruff P.W.H., Fine J.,: Endotoxemia in man. Lancet. 1972; 24:1381. 3. Zieglar E.J., et al.: Treatment of gram-negative bacteremia and shock with human antiserum to mutant E. Coli. New Eng J Med. 1982; 307;1225. 4. Zieglar E.J., McCutchen J.A.,Douglas H.,Braude A.I.:Successful treatment of human gram-negative bacteremia with antiserum against endotoxin core. Trans Assoc. Am. phys. 1981; 94:39. 5. Jacob A.I., Goldberg P.K., Bloom N., Degenshein G.A., Kozinn P.J.: Endotoxin and bacteria in portal blood. Gastroenterology. 1977; 72:1268. 6. Pohlson E.C., Zieglar E.J., McNamara J.J.:Antiserum to endotoxin: a new approach to investigate the role of endotoxin in hemorrhagic shock in rabbits. 7. Milstein, C.:Monoclonal Antibodies. Scientific American. 8. McNamara J.J., Suehiro G.T., Suehiro A., Jewett B.: Resuscitation from hemorrhagic shock. J Trauma. 1983; 23 (7):552. 9. Pohlson E.C., May M., Siri F., McNamara J.J.: Humoral factors potentiating death in primate endotoxin shock. |
| Anne P. Sharp Los Angeles CA, USA Telephone (310) 600-9247 |
 |
 |
 |
 |
 |
 |
 |
| Anne P. Sharp |
 |
 |
 |