Steven M. Larson

The pharmacology of monoclonal antibodies.

We are entering an era in which biological macromolecules can be designed to order and genetically or synthetically produced. If the experience with classical low molecular weight drugs is any guide, it will be possible in fact, easy to produce many more such agents than can be tested for effect in animals, let alone in humans. Predictive criteria are required for the process of rational design, and these must be based on an understanding of fundamental physiological and pharmacological principles. Monoclonal antibodies illustrate the point. Hundreds of new monoclonal reagents with possible clinical use are being developed every year. Each of them could be administered as whole antibody, F(ab)*, or Fab. Each could be conjugated to a toxin, a drug, a radionuclide, or a liposome containing one of those agents. Each could be class-switched to change effector functions or could be mutated in its binding site. As noted previously, the combinatorial aspects of trying every possibility are out of the question even before one has considered such current and likely future developments as ( I ) chimeric molecules combining mouse variable and human constant domains; (2) Fv (variable domain) fragments; (3) recombinants consisting of murine hypervariable segments and human variable domain framework; (4) conjugates formed by genetically grafting a linker peptide onto the molecule; and (5) enzyme-antibody or toxin-antibody chimeras. At the end of this developmental process will be ligand molecules consisting of antibody-derived binding sites grafted onto molecules designed de novo. A predictive pharmacology of biologicals must be based on theoretical and experimental information from several hierarchical levels: global (whole body), regional, local, cellular, and molecular. Here we will focus on one study of whole body pharmacokinetics, one development in regional delivery, and one issue at the local, or tissue, level.

Tumor localization of adoptively transferred indium-111 labeled tumor infiltrating lymphocytes in patients with metastatic melanoma.

Lymphoid cells infiltrating into human tumors can be expanded in vitro in medium containing interleukin-2 (IL-2). Adoptive transfer of these tumor-infiltrating lymphocytes (TIL) mediates potent antitumor effects in murine tumor models. Clinical trials to evaluate the efficacy of these cells in patients with advanced cancer are underway. We have investigated whether infused TIL labeled with indium 111 (111In) oxine can traffic and localize to metastatic deposits of tumor. Six patients with metastatic malignant melanoma who had multiple sites of subcutaneous, nodal, and/or visceral disease were the subjects of the study. The patients received cyclophosphamide 36 hours before receiving the intravenous (IV) infusion of TIL followed by IL-2 IV every eight hours. The distribution and localization of the TIL were evaluated using serial whole body gamma camera imaging, serial blood and urine samplings, and serial biopsies of tumor and normal tissue. 111In-labeled TIL localized to lung, liver, and spleen within two hours after the infusion of activity. Activity in the lung diminished within 24 hours. As early as 24 hours after injection of 111In-labeled TIL, localization of TIL to sites of metastatic deposits was demonstrated in all six patients using either imaging studies or biopsy specimens or both. 111In activity in tumor tissue biopsies ranged from three to 40 times greater than activity in normal tissue. A progressive increase in the radioactive counts at sites of tumor deposit was seen. This study shows that labeled TIL can localize preferentially to tumor, and provides information concerning the possible mechanism of the therapeutic effects of TIL.

Response of human-immunodeficiency-virus-associated neurological disease to 3'-azido-3'-deoxythymidine.

Four patients with human-immuno-deficiency-virus-associated neurological disease were treated with 3-azido-3-deoxythymidine (AZT). Three (two with chronic dementia, and one with chronic dementia and peripheral neuropathy) improved as assessed by clinical examination, psychometric tests, nerve conduction studies, and/or positron emission tomography; there was no improvement in the fourth patient who presented with paraplegia. These results support the hypothesis that certain AIDS-virus-associated neurological abnormalities are reversible by antiretroviral chemotherapy.

Radioimmunodetection and Radioimmunotherapy

The development of monoclonal antibodies that recognize tumor-associated antigens has led to significantly greater practical possibilities for producing highly specific radiolabeled antibodies for diagnosis and therapy of human tumors. A number of problems remain before this technique will be ready for routine clinical application however. Achieving the high target to background ratio that are predicted on theoretical grounds is a major challenge in cancer investigation.

Anti-murine antibody response to mouse monoclonal antibodies: Clinical findings and implications

Many patients given murine monoclonal antibody become immune to these proteins. Important factors in the development of immunity are (1) the patients ability to become immune and (2) whether whole immunoglobulin or Fab fragment was used. Circulating anti-murine antibodies increase the murine antibody plasma clearance and the liver or spleen uptake while reducing tumor targeting. Clearance rates and radioassay measured levels of anti-murine antibodies are related to serum murine/anti-murine antibody complexes. Unlabeled murine antibody displaces radiolabeled antibody from these complexes.

Effect of Phenytoin on Human Cerebral Glucose Metabolism

We used serial positron emission tomography scans with [18F]2-deoxyglucose to study the effect of phenytoin on human cerebral glucose metabolism in 10 patients with seizure disorders. Local CMRglu for each patient was measured in 10 regions of interest. EEGs were performed during each procedure to match scans for state of consciousness and exclude data from scans with ictal activity. Serial scans without a drug change were performed in six control patients. Metabolic rates were significantly lower in two cortical regions while patients were taking phenytoin. No significant changes on repeat scan were seen in the control population. Measured across all regions of interest, metabolic rates were 13% higher when patients were off phenytoin (p < 0.02).

S-[18F]Acetylcyclofoxy: a useful probe for the visualization of opiate receptors in living animals

Opiate receptor PET 3‐[18F]Acetylcyclofoxy Stereospecificity Naloxone

Toxicity, immunogenicity, and tumor radioimmunodetecting ability of two human monoclonal antibodies in patients with metastatic colorectal carcinoma.

Two human immunoglobulin M (IgM) monoclonal antibodies (MoAbs), 16.88 and 28A32, which react with cytoplasmic (28A32 and 16.88) or cell surface (28A32) determinants on human colon carcinoma cells, were administered intravenously to 26 patients with metastatic colorectal carcinoma to determine if they could localize to sites of metastatic disease, if they had any antitumor or toxic effects, and to determine whether they would elicit an antihuman MoAb response. Serial scans showed tumor uptake of radioisotope in 12 of 16 patients receiving 131I-labeled 28A32 and in nine of 12 patients receiving 131I-labeled 16.88. No antitumor effects were seen with either antibody. No antibody-related toxic effects were observed following administration of 16.88, but two patients developed localized urticarial reactions following injection with antibody 28A32. No patient developed an antibody response to 16.88. Anti-28A32 reactivity was found in five of 12 (42%) normal sera and in seven of 23 (30%) patients before receiving any antibody. Following administration of 28A32, a low titer (1:10 dilution) of anti-28A32 developed in four patients with no preexisting antibody, a decrease in the preexisting titer was seen in three other patients, the titer remained constant in one patient, and no anti-28A32 was ever detected in six patients. In most cases, anti-28A32 activity was lost at dilutions greater than 1:10 and did not appear to affect antibody half-life in the serum or whole body retention of the antibody. We conclude that these human IgM MoAbs are capable of localizing at sites of disease in vivo, are nontoxic, and are poorly immunogenic in humans. Further studies to determine the specificity of targeting and to improve the delivery of antibody to sites of tumor are indicated.

Syntheses and D2 receptor affinities of derivatives of spiperone containing aliphatic halogens

The development of a high affinity dopamine receptor ligand labeled with the positron emitting radionuclide, 18F (t 1/2 = 110 min), is of considerable interest for imaging and quantification of dopamine receptors in vivo. Derivatives of spiperone, a dopamine antagonist, labeled with 18F have been prepared, but the syntheses either proceed with inefficient fluoride utilization or involve several synthetic steps subsequent to 18F incorporation. To date, only the short-lived radioisotope of carbon, 11C (t 1/2 = 20.4 min), has been efficiently incorporated in the final synthetic step of 3-N-[11C]methyl-spiperone. 3-N-Fluoroethyl, 3-N-chloroethyl, and 3-N-bromoethyl spiperone derivatives are prepared by alkylation of spiperone with the appropriate 2-tosyloxy ethyl halide. In addition, alpha-fluorospiperone, containing fluorine alpha to the butyrophenone carbonyl, has been prepared. The 3-N-haloethyl spiperones display high affinity for dopamine receptor in vitro. Incorporation of [18F]fluoride during the final synthetic step yields a high affinity, 18F-labeled dopamine receptor-binding ligand.

Clinical use of a monoclonal antibody to bombesin-like peptide in patients with lung cancer.

Small cell lung cancer (SCLC) is one of four major types of lung cancer and constitutes roughly 25% of all the new cases. Unfortunately, despite major improvements in overall patient survival as a result of combination chemotherapy, over 90% of all SCLC patients die of their tumor.’,’ More effective treatments are clearly needed. Drug development programs based on empiric screening of various compounds have been successful for treatment of certain tumors, but not of lung cancer, despite massive investments of time and money.’*‘ Many cancer investigators feel that therapeutic progress for lung cancer will be made by exploiting the increased understanding of tumor cell biology in order to use rationally based therapies. The observation that a monoclonal antibody which binds to gastrin-releasing