Stromal antigen targeting by a humanised monoclonal antibody: an early phase II trial of sibrotuzumab in patients with metastatic colorectal cancer.

BACKGROUND: A novel immunological approach to colon cancer therapy is the antibody targeting of the fibroblast activation protein (FAP), which is highly expressed by stroma cells of this tumour. Unconjugated sibrotuzumab (BIBH 1), which is a humanised version of the murine anti-FAP mAb F19, was investigated for its anti-tumour activity, safety and pharmacokinetics. PATIENTS AND METHODS: Patients with metastatic colorectal cancer received weekly intravenous infusions of unconjugated sibrotuzumab at a dose of 100 mg over 12 scheduled weeks. The study was implemented as an open-label, uncontrolled, multicentre trial. RESULTS: 25 patients were enrolled. Patients had one or more measurable lesions, predominantly liver lesions, at baseline. At least 8 repeated weekly infusions of sibrotuzumab in 17 evaluable patients did not result in complete or partial remission. Rather, ongoing tumour progression was noted in all patients except for 2 patients with stable disease. However, progressive disease was also observed post-study in these 2 patients who received 1 and 6 additional infusions, respectively, of sibrotuzumab. Sibrotuzumab exhibited 2-compartment pharmacokinetics with a dominant terminal phase and t1/2 beta = 5.3 +/- 2.3 days. Adverse drug reactions (rigors/chills, nausea, flushing and one incidence of bronchospasm) were observed in 5 patients. Of the 24 patients given 2 or more infusions of sibrotuzumab, antibodies against sibrotuzumab were found in 3 patients (12.5%) after 4-12 infusions. CONCLUSIONS: Sibrotuzumab was well tolerated and safe. The minimal requirement for the continuation of this exploratory trial, of at least one complete or partial remission, or equivalently, of 4 patients with stable disease, was not met.

Pre-clinical evaluation of a methotrexate-albumin conjugate (MTX-HSA) in human tumor xenografts in vivo.

Methotrexate covalently bound to human serum albumin in a 1:1 molar ratio (MTX-HSA) is a new macromolecular drug which is currently being studied in phase I clinical trials by the German Association for Medical Oncology (AIO) Phase I/II study group. Previous studies have shown that MTX-HSA differs favorably from unbound MTX in terms of plasma half-life time, tumor accumulation of albumin and uptake mechanisms into cancer cells. To achieve optimal drug efficacy, repeated treatment cycles were necessary. To evaluate the anti-tumor activity of MTX-HSA and MTX in pre-clinical in vivo models, we selected 7 solid human tumor xenografts growing s.c. in nude mice and administered drug either i.p. or i.v. weekly for 3 weeks. The maximal tolerated dose (MTD) of MTX-HSA in nude mice was 12.5 mg/kg given i.p. on days 1, 8 and 15, whereas the MTD for free MTX was 100 mg/kg given i.v. MTX-HSA was significantly more active (p > 0.01) than MTX in 3 models. In the soft tissue sarcoma SXF 1301, MTX-HSA effected complete remission/cure after a single injection, whereas free MTX resulted in short-lasting, partial tumor regression. In the prostate-cancer model PRXF PC3M, MTX-HSA produced growth inhibition of 92.8% of control or an optimal test/control (T/C) of 7.2% compared to a T/C of 20.8% for MTX (p = 0.05). In the osteosarcoma model SXF 1410, optimal T/C values were 10.2% and 14.5%, respectively (p = 0.025). In lung cancers LXFE 409 and LXFL 529, bladder cancer BXF 1258 and breast cancer MAXF 449, both compounds were inactive. The improved therapeutic effects seen in 3 xenograft models under MTX-HSA treatment are promising and might be due to specific accumulation of the compound in solid tumors owing to their enhanced permeability and retention effect. Thus, clinical development of MTX-HSA will continue and sarcomas as well as prostate cancers will be included as potential target tumors for upcoming clinical phase II trials.

Albumin-based drug carriers: comparison between serum albumins of different species on pharmacokinetics and tumor uptake of the conjugate.

Albumin-based drug carrier systems have been developed in the field of chemotherapy to improve the passive tumor targeting properties of anti-cancer drugs. Usually, serum albumins of different species are used as carrier proteins, mostly of bovine (BSA), human (HSA) or rat (RSA) origin. The resulting albumin conjugates are often tested for anticancer activity in heterologous tumor models. No data is available whether the choice of the albumin species might influence the pharmacokinetics or the tumor uptake rates of the conjugates in vivo. Residualizingly ([111In]DTPA) radiolabeled RSA, BSA or HSA were administered to Walker-256 carcinoma-bearing rats. No significant difference was found in the absolute or the weight-adjusted tumor uptake rates of the three albumin tracers. The tumors were the major catabolic sites accumulating 14-18% of the injected dose (ID). Low hepatic uptake rates were determined for all albumins (below 100% ID). Minor differences were found for hepatic uptake in favor of the autologous RSA (5.8% ID) versus HSA (6.9%) and BSA (8.0%). These differences might have occurred during the commercial preparation or the radiolabeling of the different batches. In addition, there are structural differences between the three albumins, which might have contributed, despite high sequence homologies above 70% for RSA, HSA and BSA. These minor differences in the distribution patterns of RSA, HSA or BSA might not decisively influence the results of drug targeting experiments in rats. For further studies with albumin conjugates, HSA was chosen as drug carrier in rodent animal models when considering later human use. In rats or nude mice multiple injections of various HSA-drug conjugates were well tolerated without signs of allergy or anaphylaxis.

Methotrexate-albumin conjugate causes tumor growth delay in Dunning R3327 HI prostate cancer-bearing rats.

Based on the rationale of a preferred albumin uptake by tumors, conjugates comprising of rat serum albumin (RSA) as a drug carrier and of methotrexate (MTX) as chemotherapeutic drug were prepared. For a comparative study of MTX-RSA and MTX we chose a slow growing Dunning R3327 HI prostate cancer model. In a radiopharmacologic study blood kinetics and the tumor and organ distribution pattern of residualizingly labeled MTX-RSA were determined, and were found to be similar to that of residualizingly labeled RSA. The MTD was established for Copenhagen rats at a total four injections of 2 mg/kg MTX or MTX-RSA administered at days 0, 4, 8 and 12. Tumor volume measurements and tumor removal showed a small non-significant growth delay in the MTX treatment group, suggesting MTX resistance for the Dunning R3327 HI prostate carcinoma. In contrast, treatment with MTX-RSA resulted in a significant (50%) growth inhibition of the Dunning R3327 HI tumor. The cellular mechanisms responsible for MTX resistance in Dunning HI tumor cells is not known. The improved therapeutic effects seen during MTX-RSA treatment in this slow growing adenocarcinoma might be a result of prolonged tumor exposure time and an altered cellular uptake by a lysosomal route. MTX-albumin conjugates have shown antitumor activity exceeding that of MTX in several tumor xenografts in nude mice, including human prostate cancer. The recently initiated clinical development of MTX-human serum albumin will be continued and cancer of the prostate will be included as a potential target tumor during further clinical phase II testing.

Phase I trial of methotrexate-albumin in a weekly intravenous bolus regimen in cancer patients. Phase I Study Group of the Association for Medical Oncology of the German Cancer Society.

Methotrexate-albumin conjugate (MTX-HSA) is a novel human albumin-based prodrug conjugate of methotrexate (MTX). A low MTX loading rate provided optimal tumor targeting and therapeutic efficacy during preclinical testing. The objectives of this first Phase I study of MTX-HSA were to determine dose-limiting toxicity (DLT) and maximum tolerated dose (MTD) in a weekly regimen. Seventeen cancer patients who were no longer amenable to standard treatment were enrolled and were evaluable for DLT. Up to eight injections were performed in weekly intervals. Dose escalation was as follows: 20, 40, 50, and then 60 mg/m2 MTX-HSA (based on the amount of MTX bound to albumin). Additional MTX-HSA courses were feasible in case of tumor response. DLT (mainly stomatitis, Common Toxicity Criteria grade 3) occurred, beginning at the 50 mg/m2 dose level after repeated administrations; in one case, thrombocytopenia was dose-limiting. Two events of DLT occurred at the 60 mg/m2 dose level within the first two administrations. Mild anemia, transaminitis, and one case of skin toxicity were found. No significant leukopenia, nausea, renal toxicity, or other toxicities were observed. MTX-HSA was well tolerated. Drug accumulation occurred on the weekly schedule. The half-life of the drug was estimated to be up to 3 weeks. Tumor responses were seen in three patients: (a) a partial response was seen in one patient with renal cell carcinoma (response duration, 30 months, ongoing); (b) a minor response was seen in one patient with pleural mesothelioma (response duration, 31 months, ongoing); and (c) a minor response was seen in one patient with renal cell carcinoma (response duration, 14 months until progression). Poststudy treatment was administered at 2-4-week intervals. No signs of toxicity or drug accumulation were seen. Altered pharmacological properties of MTX-HSA such as plasma half-life, tumor targeting, or intracellular metabolism might have contributed to these responses. The MTD for weekly administration was 4 x 50 mg/m2 MTX-HSA during short-term treatment. A regimen with MTX-HSA injections of 50 mg/m2 every 2 weeks was recommended for a further clinical Phase I study.

Antitumor activity of methotrexate-albumin conjugates in rats bearing a Walker-256 carcinoma.

We have recently reported that albumin accumulates in solid tumors and serves there as a source of nitrogen and energy. Methotrexate-albumin conjugates [MTX(I)-RSA] derivatized at a molar ratio of 1:1 differ favorably from original MTX in terms of plasma presence and tumor uptake. The purpose of this study was to evaluate the therapeutic efficacy of these novel conjugates in a comparative study with low m.w. MTX is Sprague-Dawley rats bearing a Walker-256 carcinoma. The maximum tolerated dose (MTD) for MTX and MTX(I)-RSA was determined (2 mg/kg based on MTX injected on days 1, 3 and 8). The tumor-bearing rats received injections of either the MTD or MTD/2 of MTX, MTX-albumin or mixtures containing the MTD/2 or MTD/4 of both MTX and MTX-albumin. No toxic side effects were observed. Cure rate and tumor growth retardation were slightly better for the conjugate compared with MTX alone in the MTD group (16 complete remissions vs. 14 of 20 rats). The best results were achieved for the combination treatment with MTX and MTX-albumin, with complete remission in all 20 rats. In conclusion, MTX-albumin conjugates show therapeutic activity in vivo without toxic side effects. Additive effects were observed for a combination of MTX-albumin and MTX. These effects might be caused by the much longer tumor exposition time of the conjugate in conjunction with a different route of uptake (pinocytosis for MTX-albumin vs. folate receptors for MTX).

Pharmacokinetics of methotrexate-albumin conjugates in tumor-bearing rats.

Linking chemotherapeutic drugs to a macromolecular carrier system may enhance tumor targeting, reduce toxicity and overcome drug resistance mechanisms. As an elementary model to evaluate the pharmacological properties of macromolecular drug carrier systems we chose rat serum albumin (RSA) for carrier and methotrexate (MTX) as antineoplastic drug. The conjugation procedure yielded conjugates with an approximate 1:1 molar loading rate (MTX(1)-RSA). In the first part of the study a residualizing [111In]DTPA protein label was used for mapping in vivo the catabolic sites of the native carrier protein and of the MTX(1)-RSA drug conjugate in Walker 256 carcinosarcoma bearing rats. The tumor accumulation was about 14% of the injected dose for the RSA and MTX(1)-RSA tracers after 24 h. Tracer entrapment by organs with an active mononuclear phagocyte system was low (liver below 7% and spleen below 1.5% of the injected dose after 24 h). The 1:1 conjugation of MTX to RSA did not decisively alter the pharmacokinetic properties nor the tumor or tissue distribution of the native carrier protein RSA. In the second part of the study the different properties of the MTX(1)-RSA conjugate were compared with MTX in vivo. About 2 mg MTX/kg body weight either of the drug conjugate or of the original drug were injected after being additionally spiked with radiolabeled tracers. Plasma concentrations were simultaneously determined by immunological and radioactive means. After 24 h about 12% MTX(1)-RSA was found in circulation compared to 0.03% MTX. Favorable tumor accumulation rates of about 14% were achieved for MTX(1)-RSA versus 0.04% for MTX. About 45-fold more of the injected dose of [3H]MTX accumulated in the liver as compared to the tumor (1.5 versus 0.03%) after 24 h. Conjugation of MTX to RSA reversed this ratio in favor of the tumor to 1:1.4 (13.6 versus 9.6%). In conclusion, the potential therapeutic benefit of the MTX(1)-RSA conjugate lies in its very long tumor exposure time and its improved tumor accumulation rate compared to conventional MTX. In addition the conjugation to albumin might enhance the therapeutic effects over those achieved by long-term continuous infusion of MTX, as MTX(1)-RSA enters the cells by a different uptake mechanism. This might also help to circumvent MTX resistance mechanisms, such as a reduction in folate receptor numbers or impaired MTX polyglutamylation.

The loading rate determines tumor targeting properties of methotrexate-albumin conjugates in rats.

Albumin dominates the plasma proteins in man. Following our observation that albumin turnover in rodent tumors is markedly increased, we will present evidence that albumin can be employed as an efficient carrier for targeting cytostatic agents like methotrexate (MTX) into tumors. The considerable discrepancy in the molecular weight of MTX (454 Da) and albumin (about 67,000 Da) tempted researchers to load multiple drug molecules on one carrier molecule. It was supposed that the optimal therapeutic efficacy of MTX protein conjugates could be achieved by increasing the number of the molecules of MTX attached to the carrier. In this paper we will show that only loading rates of close to 1 mol of the cytostatic drug MTX/mol of albumin offer optimal conditions for targeting MTX-albumin conjugates into rodent tumors. Conjugates bearing 5, 7, 10 and 20 molecules of MTX on average showed considerable alterations in the HPLC profiles of the conjugates compared to albumin. Conjugates carrying 5-20 mol MTX, tagged with a residualizing radiolabel, were efficiently trapped by the liver before reaching the tumor. The tumor uptake rates of these conjugates declined dramatically with an increasing molecular load of the cytotoxic drug linked to albumin. Competition experiments with maleylated bovine serum albumin and fucoidan revealed that scavenger receptors present on the cells of the liver monocyte macrophage system were involved in this process. For further preclinical and clinical studies, we chose MTX-albumin conjugates, derivatized at a molar ratio of 1:1. These conjugates enjoy the same favorable tumor targeting properties like albumin, e.g. high tumor uptake rates, low liver uptake rates and a very long biological half-life.

Pharmacokinetic properties of LMW-heparin-tyramine fractions with high or low affinity to antithrombin III in the rat.

We have recently presented evidence that a macrophage scavenger receptor-mediated pathway is responsible for the hepatic uptake of unfractionated heparins and low-molecular-weight heparins (LMWHs) in the rat. The same receptor-mediated pathway was partially responsible for the removal of oxidized low-density lipoprotein. Unfractionated and fractionated LMWHs exert their anticoagulatory effects predominately by reversibly binding to the plasma glycoprotein antithrombin III. In this study LMWHs modified by endpoint attachment of tyramine were radiolabeled and their fractions with low or high affinity to AT-III studied in vivo in rats. The high-affinity fraction was predominately cleared from the circulation by a hepatic uptake mechanism. About 25% of the injected high-affinity tracer material was recovered, whereas only about 8% of the low-affinity material was found in the liver after 180 minutes. Blocking the scavenger receptor-mediated liver RES uptake mechanism by maleylated bovine serum albumin led to a considerable decline in liver uptake (9 versus 25%). The low-affinity material was rapidly eliminated into the urine after 180 minutes. About 45% of the low-affinity material was excreted versus 23% of the high-affinity material. Tight binding to AT-III prevented LMWH-tyramine from being rapidly cleared from the circulation via the kidneys into the urine; instead, the scavenger receptor-mediated hepatic uptake mechanism seemed to be more dominant in removing material with high affinity to AT-III from blood.

Dosage, anticoagulant, and antithrombotic effects of heparin and low-molecular-weight heparin in the treatment of deep vein thrombosis.

We have performed a prospective, randomized, controlled trial comparing continuous intravenous unfractionated heparin with twice-daily subcutaneous (s.c.) high-dose low-molecular-weight (LMW) heparin in the initial treatment of 50 patients with acute proximal deep vein thrombosis. In this article we analyze the relationship between the dosage of the heparins, the anticoagulant effects on aPTT, and thrombin and factor Xa inhibition to the improvement of the Marder score after a 10-day treatment period. Improvement of the Marder score was observed in about 70% of patients without regard to administration of unfractionated or LMW heparin. Patients in both treatment categories were divided into two groups, namely, those who showed an improvement of the Marder score and those who did not. In the group of patients with unfractionated heparin and regression of thrombus size the mean dosage was 33,000 U/day, whereas the mean dosage was 37,000 U/day in the patients with status idem of the Marder score after the 10-day treatment period. Thrombin clotting time values were in contrast to the dosage. Patients with regression of thrombosis showed higher thrombin clotting time values compared with those with status idem. These results were also seen with aPTT and the Heptest coagulation assay, but the differences between the two groups were less pronounced. No differences between these two groups of patients were seen or detected with the S2222 chromogenic anti-factor Xa method. Patients receiving 2 x 12,000 IU s.c./day LMW heparin did not show these differences, the dosage being adjusted by the anti-Xa levels, ranging from 0.6 to 1.0 U/mL 4 hours after the s.c. injection. The groups of patients categorized as to improvement or not of the Marder score did not show differences in the daily dose. The anti-Xa activity was higher in patients with regression of thrombosis compared with patients without regression. The other coagulation parameters did not show any relation to the clinical outcome of thrombus regression. The relationship between the change of the Marder score at day 10 and the anticoagulant effect on the different coagulation systems correlated weakly for patients receiving unfractionated heparin. The highest correlation was found for the improvement of Marder score and thrombin inhibition in the heparin group with r = 0.42. For LMW heparin no correlation could be detected. Heptest coagulation values were in the same range for patients receiving unfractionated and LMW heparin. In contrast to the chromogenic anti-Xa assay, aPTT, thrombin clotting time, and prothrombin time values differed substantially in the two treatment regimens. Treatment of recent deep vein thrombosis with unfractionated heparin profits from laboratory monitoring, whereas monitoring of the anticoagulant effect during the treatment with s.c. LMW heparin does not influence the outcome on thrombus regression.

Enhanced albumin uptake by rat tumors.

Albumin dominates the nitrogen and energy resources in blood. However, only limited data is available on its accumulation and catabolism by tumors. This was caused by the lack of suitable radiolabels for long-term follow-up of protein catabolism in vivo. Conventional radiolabels like radioiodine are metabolically unstable. After lysosomal degradation diffusible tracer residues are rapidly released from catabolic sites, Tumors with high metabolic activity evade detection. To study the uptake of rat serum albumin (RSA) by tumors a conventional radioiodine label and two residualizing radiolabels were chosen. It is known that residualizing I-131-tyramine-deoxisorbitol and In-111-DTPA protein labels remain trapped at catabolic sites after lysosomal degradation of their carrier proteins. We were able to show by scintigraphy and after organ removal that a Walker-256 carcinosarcoma with a turner size of about 5% of the body weight accumulated more than 20% of the initially injected dose of a In-111-DTPA-RSA within 24 h. Tumor uptake rates for albumin exceeded those of the kidneys by about 4 times, and those of the liver by about 3 times. It was estimated that about one out of two albumin molecules trapped by an ovarian-342 tumor must have been degraded during 72 h. High uptake and degradation rates would make albumin an alternative nitrogen and energy source for these tumors. Although an unfavorable time-frame limits the use of residualizingly labeled albumin for scintigraphic tumor diagnosis in man, albumin might be an interesting carrier for delivering covalently attached chemotherapeutic agents into tumors by an alternative lysosomal route.

Comparison of immunological and functional assays for measurement of soluble fibrin.

Various assays have been developed for quantitation of soluble fibrin or fibrin monomer in clinical plasma samples, since this parameter directly reflects in vivo thrombin action on fibrinogen. Using plasma samples from healthy blood donors, patients with cerebral ischemic insult, patients with septicemia, and patients with venous thrombosis, we compared two immunologic tests using monoclonal antibodies against fibrin-specific neo-epitopes, and two functional tests based on the cofactor activity of soluble fibrin complexes in tPA-induced plasminogen activation. Test A (Enzymun-Test FM) showed the best discriminating power among normal range and pathological samples. Test B (Fibrinostika soluble fibrin) clearly separated normal range from pathological samples, but failed to discriminate among samples from patients with low grade coagulation activation in septicemia, and massive activation in venous thrombosis. Functional test C (Fibrin monomer test Behring) displayed good discriminating power between normal and pathological range samples, and correlated with test A (r = 0.61), whereas assay D (Coa-Set Fibrin monomer) showed little discriminating power at values below 10 micrograms/ml and little correlation with other assays. Standardization of assays will require further characterization of analytes detected.

The injection of heparin prolongs the plasma clearance of oxidized low density lipoprotein in the rat.

There is evidence that oxidized-LDL plays an important role in atherogenesis. We now report on the in vivo interaction between unfractionated heparin and oxidized LDL in rats. The recovery rates of the native LDL particles ranged between 75% and 85% of the injected dose. Heparin did not interfere with the clearance rates of native LDL. After administration of radioactive labeled oxidized-LDL particles, 26% of the material was measured in circulation after 5 minutes, 8% after 20 minutes, and 3% after 60 minutes. After injection of heparin 2 minutes prior to oxidized-LDL tracer particles, 44% of the tracer was found in blood after 5 minutes, 23% after 20 minutes, and 9% after 60 minutes. Oxidized-LDL tracer particles disappeared from blood with an alpha half-life of 5 minutes and a beta half-life of 7.5 minutes. After receptor blocking with unfractionated heparin the alpha half-life of the oxidized-LDL tracer was prolonged to 17.5 minutes and the beta half-life to 27.5 minutes. These results indicate that heparin molecules of a comparatively small molecular weight competed the scavenger receptor mediated uptake of oxidized-LDL particles in vivo. Oxidized-LDL particles are known to mediate their pro-atherosclerotic activity in part by stimulating smooth muscle cell proliferation by a scavenger receptor-mediated pathway. It can be speculated, if heparins interfere with the uptake of oxidized-LDL, heparins might thus in part exert their known antiatherosclerotic properties.

Complexes of a modified low-molecular-weight heparin with protamine are predominantly cleared by a macrophage scavenger receptor-mediated process in rats.

Neutralization of heparin with its antidote protamine is associated with side effects such as pulmonary hypertension. The pharmacodynamic effects of protamine treatment are well documented. However, little is known about metabolic fate of heparin-protamine complexes. Twenty Wistar rats received a 131I-labeled low-molecular-weight heparin tracer intravenously. Four groups of animals were formed: a control group receiving the tracer, a second group receiving protamine after tracer application, a third group receiving maleylated bovine serum albumin (mal-BSA) prior to tracer and protamine injections to inhibit scavenger receptors of the reticuloendothelial system, and the last group receiving preformed heparin-protamine aggregates. All animals were examined by scintigraphy. Blood and tissue samples were analyzed for radioactivity. Protamine injection 2 min after heparin tracer application lead to a rapid decline in blood radioactivity. The radioactivity in the liver increased from 17% for the control to 43% after protamine application. Injection of mal-BSA prior to protamine prevented tracer accumulation in the liver and increased urine excretion (34% versus 20%). In vitro preformed heparin-protamine precipitates were rapidly trapped in the liver. We present evidence that, like the polyanionic heparin, polyanionic heparin-protamine complexes are phagocytosed by a scavenger receptor-mediated mechanism by macrophages, predominantly in the liver. The amount, the size, and the charge density of the complexes might trigger a mediator release from macrophages leading to phenomena such as pulmonary hypertension.