e13102 Background: Zalutumumab (HuMax-EGFr, clone-2F8), a fully human IgG1 EGFR antibody, currently in phase III clinical trials, was previously demonstrated to have a dual mechanism of action. First, it blocks EGFR signaling by preventing ligand binding and inducing receptor downmodulation. This mechanism was found to be effective in established xenograft tumors at doses giving full receptor occupancy. Second, we observed in vitro induction of antibody-dependent cell-mediated cytotoxicity (ADCC), which was found to be already maximally active at very low, sub-saturating antibody concentrations. Because the contribution of ADCC to the antitumor effects of therapeutic cancer antibodies in solid tumors remains unclear we studied the importance of ADCC induction by zalutumumab in xenograft models. Methods: We prepared human EGFR antibodies with different functionalities for comparison with zalutumumab to: 1) clone 2F8-N297Q a mutant of zalutumumab, which has a nonfunctional Fc part regarding interaction with Fc receptors, 2) clone-018, a non-blocking human EGFR mAb, for which we observed ADCC in vitro as only mechanism of action and 3) clone-018-N297Q which displayed no in vitro activity. Results: Therapeutic treatment of established A431 tumor xenografts with these antibodies showed efficacy for zalutumumab and clone 2F8-N297Q only, indicating that blocking of signaling is crucial in this condition. In contrast, when given on the day of tumor induction, both zalutumumab and clone-018, but not clone-018-N297Q, were effective in preventing A431 tumor xenografts. In an experimental xenograft model, zalutumumab and clone 018 could completely prevent outgrowth of lung metastases, even when given at a low, nonsaturating dosing. Conclusions: Our experiments indicate that blocking of EGFR signaling is required for effectiveness of zalutumumab in established tumors, requiring doses giving full receptor occupancy. We conclude that induction of ADCC represents an additional, independent mechanism, which seems insufficient for reducing primary solid tumors, but is likely to have an important contribution in preventing metastases. Author Disclosure Employment or Leadership Position Consultant or Advisory Role Stock Ownership Honoraria Research Funding Expert Testimony Other Remuneration Genmab Genmab Genmab

Glutaraldehyde is frequently used for the chemical modification of hemoglobin to improve the characteristics for use as a blood substitute. Polymerization of hemoglobin prolongs the vascular retention by increasing the molecular size. There are different ways to perform the cross-linking reaction, and this determines the stability and the oxygen affinity of the final product. In our laboratory we developed a glutaraldehyde-polymerized hemoglobin solution (PolyHbXI) with a high degree of intramolecular cross-linking and lowered oxygen affinity. In preclinical safety studies, PolyHbXI was generally well tolerated. However, in three animal models (rats, rabbits and rhesus monkeys), we observed transient hemorrhagic disorders. We further analyzed this side effect in a histopathological study with rats. Hemorrhagic lesions were observed in several tissues (with predilection for the intestinal wall) two days after administration of PolyHbXI, and the lesions were resolved after one week. Microscopically, the bleedings were characterized as “small vessel vasculitis” with a neutrophil infiltration. Platelet numbers, bleeding time and APTT values were normal. This leads to the hypothesis that endothelial cell injury plays a central role in the pathophysiology. By comparing different modified hemoglobin preparations we found evidence that the toxic factor is the result of the cross-linking procedure with glutaraldehyde.

The serpins C1 esterase inhibitor (C1Inh), antithrombin (AT), α1‐antitrypsin (α1AT) and α2‐antiplasmin (α2AP) are known inhibitors of coagulation factor XIa (FXIa). Although initial studies suggested α1AT to be the main inhibitor of FXIa, we recently demonstrated C1Inh to be a predominant inhibitor of FXIa in vitro in human plasma. The present study was performed to investigate the plasma elimination kinetics of preformed human FXIa–FXIa inhibitor complexes injected in rats. The amounts of complexes remaining in circulation were measured using enzyme‐linked immunosorbent assays. The plasma half‐life time of clearance (t1/2) was 98 min for FXIa–α1AT complexes, whereas it was considerably shorter, i.e. 19, 18 and 15 min for FXIa–C1Inh, FXIa–α2AP and FXIa–AT complexes, respectively. Thus, due to this different plasma t1/2, preferentially FXIa–α1AT complexes may be detected in clinical samples. Furthermore, measuring FXIa–FXIa inhibitor complexes in patient samples may not help to clarify the relative contribution of the individual serpins to inactivation of FXIa in vivo

Modification of hemoglobin (Hb) by crosslinking and polymerization results in an improved oxygen release capacity and a prolonged vascular retention time. Modification improves the efficacy and prevents certain side effects. It eliminates leakage of Hb through the kidneys and accumulation in the tubuli. Another important issue is the degree of purification of Hb solutions. Traces of membrane fragments may cause immunogenic and thrombogenic side effects. To determine the contamination with erythrocyte membrane fragments, we developed assays for glycophorin-alpha and phospholipids. Special models were evaluated for testing the maximum allowable level of membrane contamination. As an in vitro model for thrombogenicity we used confluent monolayers of human umbilical vein endothelial cells. These cells were incubated with Hb solutions and subsequently tested on tissue factor (TF) procoagulant activity. TF was tested by the factor VII-catalyzed activation of factor X. The lower detection limit of this assay for endotoxin was 0.5 ng/ml. Hb did not cause any tissue factor expression even after prolonged incubation. No cooperation was found within endotoxin. As an in vivo test on thrombogenicity we developed a guinea pig model in which we can follow the generation of fibrinopeptide A (FPA). This is one of the most sensitive markers for thrombin activation in vivo. When slightly contaminated Hb solutions (phospholipid content 2 nmol/ml) were infused in the presence of factor Xa at a dose (9 micrograms/kg) which in itself did not induce FPA generation, we observed an increase in FPA levels in the plasma from 1.2 +/- 0.4 ng/ml to 5.2 +/- 0.7 ng/ml. Factor Xa is used to mimic a stressed clinical condition with activated coagulation.(ABSTRACT TRUNCATED AT 250 WORDS)

C1-inhibitor is the only known inhibitor of the classical pathway of complement and the major inhibitor of the contact pathway of coagulation. Like other serine proteinase inhibitors, C1-inhibitor can exist in three conformations, ie, the native, the proteinase-complexed, and the proteolytically inactivated form. Here we studied the plasma elimination kinetics of these three forms of human C1-inhibitor in rats. The clearance of the complexed form of C1-inhibitor appeared to be the most rapid and depended in part on the proteinase involved (observed plasma t1/2 was 20 minutes for C1s-C1-inhibitor, 32 minutes for kallikrein-C1-inhibitor, and 47 minutes for beta XIIa-C1- inhibitor), whereas that of native C1-inhibitor was the slowest (observed plasma t1/2 4.5 hours). Inactivated C1-inhibitor was cleared with an apparent plasma t1/2 of 1.6 hours. Thus, the short plasma t1/2 of complexed relative to native C1-inhibitor explains why in patients only low concentrations of C1-inhibitor complexes may be observed despite activation of the contact and/or complement systems.

In the present study we investigated the mechanism of prolongation of the plasma retention of free hemoglobin by polymerization. Polymerization of intramolecularly crosslinked hemoglobin with glutaraldehyde yields a mixture of large polymers, small polymers and monomers. In exchange transfusion experiments in rats we analyzed plasma samples by gel filtration to determine the clearance of polymers of different size. A positive correlation was found between polymer size and vascular retention. Furthermore, the clearance of large polymers appeared to be highly dose-dependent: after 20% and 70% exchange transfusions, we observed for large polymers a plasma half-life of 12 and 26 hours, respectively, whereas the half-life for 64kD monomers was 4 hours in both cases. The degradation of hemoglobin was followed by measuring the bilirubin excretion. The infused heme was recovered as bilirubin within 72 hours. The delay between the disappearance of free hemoglobin from the plasma and the recovery as bilirubin was about six hours and was not affected by polymerization or dose. We conclude that polymerization prevents the operation of certain clearance mechanisms, while still allowing a route of clearance that is easily saturated. The intracellular degradation of heme into bilirubin is not affected by the modifications of hemoglobin and is not easily saturated.

In the present study we investigated the therapeutic action of antithrombin III (AT III) in taurocholate-induced experimental pancreatitis with high lethality in rats. High-dose AT III treatment greatly improved the survival rate not only when given as pretreatment but also when given 2 hr after induction. No favorable effect on survival rate was observed on administration after 5 hr. Both intravascular and intraperitoneal AT III administration locally restored decreased AT III levels in the peritoneal cavity and increased plasma AT III to supranormal levels. The primary pancreatic insult seemed to be unaffected by the treatment, because neither the rise in plasma lipase nor the development of ascites or the extension of the pancreatic necrosis were diminished. Because heparin pretreatment of the rats was also effective, the mechanism of the beneficial action was probably mediated by inhibition of the proteases of the coagulation cascade, thereby preventing intravascular coagulation in the pancreas and distant organs and subsequent systemic complications. The high efficacy of AT III treatment in this experimental model may stimulate clinical studies evaluating the efficacy of AT III treatment in an early stage of acute pancreatitis.

The formation and reduction of extracellular methemoglobin (metHb) in plasma was studied in vivo in conscious rats after isovolemic exchange transfusions with polymerized hemoglobin solutions. After exchange transfusions of 40 and 70% of the blood volume with hemoglobin solutions, containing less than 6% methemoglobin, the methemoglobin level remained below 15%, whereas exchange transfusions of greater than 90% resulted in an increase in the metHb level to about 30% after 24 hours. The reduction of metHb was studied after exchange transfusions with fully oxidized hemoglobin. A gradual decrease in the metHb level to 20% was observed after exchanges of 5 or 40%. A higher exchange transfusion (70%) resulted also in a decrease in the metHb level but only to approximately 40% in 24 hours. In another series of experiments the reduction of metHb was studied in vitro with isolated human erythrocytes. Incubation of the erythrocytes in the presence of oxidized polymerized hemoglobin (3 g%) resulted in a decrease in the percentage of metHb from 91% to 64%. In the presence of 0.2 mM ascorbic acid the metHb level declined to 22%, suggesting a synergistic effect. These results indicate (1) that there is a potent reducing mechanism present in blood that can reduce extracellular oxidized polymerized hemoglobin and (2) that isolated erythrocytes have a large capacity to reduce extracellular metHb, and may also play an important role in the reduction of extracellular metHb in vivo.

Hemoglobin modified by intramolecular cross-linking with 2-nor-2-formylpyridoxal 5'-phosphate (NFPLP) has potential application as an oxygen-carrying plasma expander with improved vascular retention time (threefold) and oxygen-transporting properties compared to native hemoglobin. Under some conditions a further prolongation is required. In this study polymerization of the purified modified hemoglobin (HbNFPLP) with glutaraldehyde was investigated. The influence of the degree of polymerization on the iso-oncotic concentration and the viscosity of the polymerized Hb-NFPLP (polyHbNFPLP) was investigated with three products polymerized to an increasing extent. Exchange transfusions in rats followed by gel filtration analyses of plasma samples provided information on the vascular retention time of four separate polymer fractions (i.e., monomers, dimers, trimers/tetramers, and polymers). The vascular retention time of these fractions correlated with their size. For lightly and highly polymerized HbNFPLP we found fivefold and sevenfold increases, respectively, compared to the retention time for native hemoglobin. This finding and the different shapes of the clearance curves suggested a different clearance mechanism from the vascular system for the monomer and polymer fractions. By polymerizing HbNFPLP the oxygen affinity was enhanced (oxygen half-saturation pressure shifted from 45 to 22 mm Hg), and it appeared to be independent of the degree of polymerization. Because hemoglobin was undetectable in the urine of the rats after the exchange transfusions, no accumulation of polyHbNFPLP will occur in the tubules of the kidney.(ABSTRACT TRUNCATED AT 250 WORDS)

Modified hemoglobin solutions have potential application as plasma expanders with oxygen-transporting capacity. In a previous study it was found that modification of hemoglobin by intramolecular cross-linking with 2-nor-2-formylpyridoxal 5'-phosphate (NFPLP) improves the vascular retention time by a factor of three, and it also improves the oxygen-transporting properties. In the present study we investigated in rats how, after exchange transfusion of a clinically relevant dose, the modified hemoglobin (HbNFPLP) was distributed in the body compared with how the unmodified hemoglobin was distributed. By using a new technetium 99m labeling technique, we found in a scintigraphic study that accumulation of hemoglobin in the kidneys was greatly diminished by the intramolecular cross-linking with NFPLP. These findings were confirmed by light-microscopic observations after diaminobenzidine staining. It was concluded that the impairment of kidney function caused by blockade of the tubuli is not to be expected from HbNFPLP. In the liver and spleen, where the free HbNFPLP is possibly eliminated, some accumulation of 99mTc label was observed, but the major part of the extravascular label was diffusely spread throughout the body. This led to the conclusion that important accumulation of undegraded HbNFPLP does not occur in the liver and spleen. Rapid appearance of both hemoglobin and HbNFPLP in the lymph showed that cross-linking with NFPLP does not prevent the distribution of hemoglobin over the interstitial space in the first hours after administration. However, pharmacokinetic analysis demonstrated that transcapillary transfer contributes only to a limited extent to the disappearance from the circulation. During 24-hour infusions of HbNFPLP, a steady state with a constant plasma concentration was easily reached. The latter experiment indicated that the eliminating system does not become saturated during prolonged administration of large doses of HbNFPLP.

Radiolabeled hemoglobin may be a useful tool in the study of the body distribution of hemoglobin solutions developed as plasma expanders with oxygen-transporting capacity. The present investigation compares the suitability of two radiolabeling techniques for hemoglobin. 125I labeling of hemoglobin with Iodogen as iodinating agent caused major changes in the chromatographic behaviour and an accelerated plasma clearance of the labeled hemoglobin in rats. A recently developed two-step procedure for 99mTc labeling gave better results. The label had only minimal influence on the chromatographic behaviour of hemoglobin. In vivo, no free label occurred in the circulation and no transfer of the label to other plasma proteins took place. The plasma clearance of 99mTc-labeled hemoglobin in rats was slowed. However, this could be explained entirely by diminishing glomerular filtration, probably by inhibition of the dissociation of the hemoglobin molecule into dimers. The plasma clearance of hemoglobin modified by intramolecular cross-linking, which prevents dissociation of the molecule into dimers and thus excretion by the kidney, was not influenced by the label. We conclude that the 99mTc labeling procedure is suitable for in vivo distribution studies of hemoglobin when it is taken into account that the urinary excretion is underestimated. For cross-linked hemoglobin, which is more promising as plasma expander, no such restriction exists.

Intravenous administration of certain immunoglobulin preparations may cause severe adverse reactions, especially in immunodeficient patients. These reactions are generally assumed to be related to the anticomplementary activity of the preparations, caused by IgG aggregates. Because the exact mechanism of the adverse reaction is unknown, we investigated the reactions induced in anesthetized rats on rapid intravenous administration of different human immunoglobulin preparations. The most conspicuous observation was a severe, long-lasting hypotension, induced by standard immunoglobulin preparations (for intramuscular use), which appeared to be independent from the concomitant complement and neutrophil activation. The long-lasting hypotension was not related to the presence of prekallikrein activator, which induced a transient hypotensive reaction only after sensitizing the rats to bradykinin. The reactions appeared to be associated with IgG aggregates. It was found that certain aggregates induced mainly complement activation, whereas others had mainly a hypotensive effect or no effect at all. It was concluded that the rat model provides a sensitive and reproducible test system for vasoactive properties of immunoglobulin preparations for intravenous administration that cannot be predicted from in vitro measurements, such as anticomplementary activity, aggregate content or prekallikrein activator activity. It is suggested that the test may also be used for other plasma products for intravenous administration.

The usefulness of hemoglobin solutions as plasma expanders with oxygen-carrying capacity is limited by a high oxygen affinity and a rapid clearance from the circulation. A large amount of the hemoglobin is cleared by the kidneys, because the hemoglobin can pass the glomeruli after dissociation into dimers. This dissociation can be prevented by cross-linking the beta chains with 2-nor-2-formylpyridoxal 5'-phosphate (NFPLP), a modification that also diminishes the oxygen affinity. In the present study, the vascular retention of modified hemoglobin (HbNFPLP) compared with unmodified hemoglobin (Hb) was investigated in rats and rabbits by replacing half the blood volume with a mixture of Hb and HbNFPLP (7 gm/100 ml). The amount of free hemoglobin in the circulation was determined from the plasma concentration, corrected for the decrease in plasma volume. The decrease in plasma volume was calculated from the increase in hematocrit (in the rats, 30% to 40% in 3 hours). The ratio Hb/HbNFPLP was determined by high-performance anion-exchange chromatography. The half-disappearance times for HbNFPLP and Hb were found to be 3 hours and 1 hour in rats and 7 hours and 2.5 hours in rabbits, respectively. In the rats, one third of the unmodified Hb was found in the urine 5 hours after the exchange, against only 5% of the HbNFPLP. In rats without kidneys, the ratio of Hb/HbNFPLP in the circulation remained constant. The results demonstrate that intramolecular cross-linking of hemoglobin with NFPLP prevents excretion by the kidneys, but does not influence other clearance mechanisms.

From these liver perfusions with Hb and Hb/HbNFPLP solutions the following conclusions can be drawn: In spite of the chemical modification of the hemoglobin molecule, no rheological differences are seen. All parameters measured were sensitive to hypoxia induced by a decrease in perfusion flow rate. The NFPLP-induced decrease in oxygen affinity was reflected in a higher venous PO2. These in-vivo observations are in agreement with the in-vitro measured oxygen dissociation curves. The difference in PO2 did not result in a change in the other oxygen-sensitive parameters in this model under the chosen conditions. Possible causes for these observations are: the level of hypoxia was too low the oxygen supply in the perfusions with the modified hemoglobin solutions was lower than the oxygen supply in the perfusions with normal hemoglobin. Whether or not this observation is due to an intrinsic property of the modified hemoglobin molecule remains to be established.