Radiolocalisation of an anti-CEA Monoclonal Antibody (FO23C5) and its Fragments in a Colon Carcinoma Xenograft Model

A new monoclonal antibody designated FO23C5 against a protein component of carcinoembryonic antigen (CEA) has been developed. A xenograft system of human colon cancer was used to compare the intact monoclonal IgG with its fragments ((Fab’)2 and Fab) and with an established anti-CEA antibody (MAb35) and the antibody AUA1 raised against the colon carcinoma cell line. We demonstrate that FO23C5 compares well with the existing anti-CEA antibody and with AUA1, and that F(ab’)2 fragments perform best in achieving optimal tumour to normal tissue ratios compared with intact IgG and Fab fragment.

A paediatric telecardiology service for district hospitals in south-east England: an observational study

OBJECTIVES

To compare caseloads of new patients assessed by paediatric cardiologists face-to-face or during teleconferences, and assess NHS costs for the alternative referral arrangements.

DESIGN

Prospective cohort study over 15 months.

SETTING

Four district hospitals in south-east England and a London paediatric cardiology centre.

PATIENTS

Babies and children.

INTERVENTION

A telecardiology service introduced alongside outreach clinics.

MEASUREMENTS

Clinical outcomes and mean NHS costs per patient.

RESULTS

266 new patients were studied: 75 had teleconsultations (19 of 42 newborns and 56 of 224 infants and children). Teleconsultation patients generally were younger (49% being under 1 year compared with 32% seen personally (p = 0.025)) and their symptoms were not as severe. A cardiac intervention was undertaken immediately or planned for five telemedicine patients (7%) and 30 conventional patients (16%). However, similar proportions of patients were discharged after being assessed (32% telemedicine and 39% conventional). During scheduled teleconferences the mean duration of time per patient in sessions involving real-time echocardiography was 14.4 min, and 8.5 min in sessions where pre-recorded videos were transmitted. Mean cost comparisons for telemedicine and face-to-face patients over 14-day and 6-month follow-up showed the telecardiology service to be cost-neutral for the three hospitals with infrequently-held outreach clinics (1519 UK pounds vs 1724 UK pounds respectively after 14 days).

CONCLUSION

Paediatric cardiology centres with small cadres of specialists are under pressure to cope with ever-expanding caseloads of new patients with suspected anomalies. Innovative use of telecardiology alongside conventional outreach services should suitably, and economically, enhance access to these specialists.

MT1-MMP and MMP-2 mRNA expression in human ovarian tumors: possible implications for the role of desmoplastic fibroblasts

Expression of activated MMP-2 (72 kDa type IV collagenase) is highly associated with the malignant phenotype in adenocarcinomas, but predominant expression of the mRNA appears to be in stromal cells. MT1-MMP (membrane type 1-matrix metalloproteinase) is implicated in tumor-epithelial cell surface activation of latent pro-MMP-2, indicating a mechanism for tumor-stromal interaction in invasion. We determined the relative mRNA distribution of these MMPs in human ovarian tumors with a view to analyzing potential variations in the epithelial-mesenchymal interactions dictating ovarian tumor cell spread. In situ hybridization using 35S-labeled riboprobes was used to analyze 33 human ovarian tumors and mouse xenografts of human ovarian (DOV 13, SKOV3) and breast (MCF 7) tumor cell lines known to express MT1-MMP and MMP-2. MMP-2 mRNA was expressed in 31 of 33 and MT1-MMP mRNA was expressed in 29 of 33 tumor cases. MMP-2 mRNA was predominantly expressed in desmoplastic fibroblasts and in the subepithelial stroma. MT1-MMP mRNA showed some colocalization with MMP-2 in stromal cells. Neoplastic epithelial cell labeling for MT1-MMP mRNA was present in borderline and malignant tumors but not in benign tumors, and was invariably less than stromal labeling. Xenografts of DOV 13, SKOV 3, and MCF 7 cells showed some stromal localization of MMP-2 mRNA and weak labeling of DOV 13 cells. There was variable labeling for MT1-MMP mRNA in the neoplastic cells only. The colocalization of MT1-MMP and MMP-2 mRNAs in ovarian carcinoma stroma supports the view that MT1-MMP is closely associated with MMP-2 expression and function. It suggests that either additional mechanisms are involved in regulating MMP-2 activation at the tumor cell surface, or more intriguingly, that desmoplastic fibroblasts may be the primary mediators of extracellular matrix remodeling with respect to this system.

Intraperitoneal yttrium-90-labeled monoclonal antibody in ovarian cancer

From March 1987 to March 1988, a phase I to II study was carried out in 25 patients with ovarian cancer. They received escalating doses of intraperitoneally (IP) administered yttrium-90 (Y-90)-labeled monoclonal antibody, HMFG1, against a tumor cell-surface antigen. Myelosuppression prevented an escalation of the administered Y-90 activity above 25 mCi. Y-90-labeled antibody was absorbed from the peritoneal cavity into the circulation. Maximum blood Y-90 activity was observed 40 hours after the IP injection with a mean of 21% of the injected activity (range, 14.2% to 26.4%) in the circulation. The radiation dose the bone marrow received from circulating Y-90-labeled antibody (the blood radiation dose) was calculated by applying the Medical Internal Radiation Dose (MIRD) formulation to the measured Y-90 activity in patients blood. Myelosuppression occurred following calculated blood radiation doses to bone marrow of only 10 to 30 cGy. The excessive myelosuppression following such modest radiation doses from circulating Y-90-labeled antibody could be explained by the uptake of Y-90 by bone. In an attempt to reduce bone absorption of Y-90, seven patients received an intravenous (IV) infusion of EDTA (Sinclair Pharmaceuticals Ltd, Godalming, United Kingdom). This increased the urinary excretion of Y-90 from a mean of 11.1% to 32.3% of the injected activity (P = .0001). Fourteen patients had assessable tumor at laparoscopy. Tumor regression was observed in one patient, and palliation of ascites in a further patient.

Intraperitoneal radio-localization of tumors pre-targeted by biotinylated monoclonal antibodies

We describe 2-step and 3-step strategies for intraperitoneal tumor radio-localization by means of monoclonal antibodies (MAbs). Nude mice bearing intraperitoneal human colon carcinoma tumors were injected i.p. with biotinylated MAb AUAI, followed 24 hr later by radioiodinated streptavidin (2-step). The uptake of radioactivity in tumor and normal tissues was measured 4 hr after injection of radioactive compound. A 3-step strategy consisted in administering biotinylated antibody, cold avidin after 24 hr and 111In-labelled biotin after a further 4 hr; mice were then killed 2 hr later. Tumor localization of intraperitoneally-administered biotinylated antibody and direct targeting of radioactive streptavidin to biotinylated antibody bound to tumor sites were demonstrated using immunohistochemistry and autoradiography. Our results show that (i) the 2-step approach increased the percentage of radioactivity uptake by tumor with respect to directly labelled antibodies (24% vs. 6%) and improved the tumor/non-tumor ratio; (ii) the 3-step approach allowed faster blood clearance of the radioactive probe (111In-biotin) and yielded high tumor/non-tumor ratios. "Pre-targeting" methods appear to have advantages over the conventional 1-step approach with directly radiolabelled antibody.

Enhancement of monoclonal antibody uptake in human colon tumor xenografts following irradiation

Indium-111-labeled AUA1 tumor-associated monoclonal antibody raised against an antigen of colon adenocarcinoma was used to evaluate the effect of ionizing radiation on antibody uptake by the LoVo adenocarcinoma cell line grown as a xenograft in nude mice. Tumors were exposed to single doses of external X-irradiation of between 400 and 1600 cGy followed, 24 h later, by administration of specific or nonspecific antibody. Animals were sacrificed 3 days after antibody administration. At doses higher than 400 cGy, tumor uptake with both specific and nonspecific antibody was significantly increased. No difference in changes in tumor volume was observed between the groups receiving irradiation and the controls. Specific antibody uptake by tumors was always significantly higher than nonspecific having an approximate 4-fold binding advantage. Vascular permeability and the vascular volume of irradiated and control tumors was measured 24 and 72 h after irradiation, using iodine-125-labeled nonspecific antibody and labelling of the red blood cells in vivo with 99mTcO4. At doses higher than 400 cGy, vascular permeability in the tumor 24 h after irradiation was significantly increased (P less than 0.05), while the vascular volume decreased (P less than 0.001) compared to control values. However at 72 h after irradiation there was no difference between treated and control groups. The results obtained in this study suggest a potential value of external irradiation to increase monoclonal antibody uptake by tumors governed mainly by the increased vascular permeability of the tumor vasculature soon after the irradiation exposure.

Radiolocalisation of an anti-CEA monoclonal antibody (FO23C5) and its fragments in a colon carcinoma xenograft model

A new monoclonal antibody designated FO23C5 against a protein component of carcinoembryonic antigen (CEA) has been developed. A xenograft system of human colon cancer was used to compare the intact monoclonal IgG with its fragments (Fab')2 and Fab) and with an established anti-CEA antibody (MAb35) and the antibody AUA1 raised against the colon carcinoma cell line. We demonstrate that FO23C5 compares well with the existing anti-CEA antibody and with AUA1, and that F(ab')2 fragments perform best in achieving optimal tumour to normal tissue ratios compared with intact IgG and Fab fragment.

Localization of monoclonal antibody AUA1 and its F(ab')2 fragments in human tumour xenografts: an autoradiographic and immunohistochemical study

The mouse monoclonal antibody (MAb) AUA1, when applied on LoVo tumour sections, reacts by staining all tumour cells, on their cell surfaces. To investigate the accessibility of these sites to antibody when the tumour is present as a solid mass in vivo, subcutaneous xenografts of LoVo were first prepared in nude mice. The mice were then injected intravenously with either 125I-labelled AUA1, 125I AUA1 F(ab')2 or with 125I-labelled HMFG2 (negative control antibody). Animals were killed at various time intervals. Gross and micro-autoradiography as well as immunohistochemistry were performed on tissue samples of tumour and control organs. The in vivo injected antibody, in contrast to that studied in vitro, was localized only, as detected by autoradiography, on a thin layer of tumour cells adjacent to the vascularized stroma. On microscopically small tumour islands the antibody penetration was complete. Most of the radioactivity was on the cell surfaces, as seen on in vitro immunostaining. With intact antibody, similar autoradiographic results were obtained at days 1, 3 and 6. With F(ab')2 fragments there was deeper penetration into the tumour at days 1 and 3, though less radioactivity was found; by day 6 the activity had greatly decreased. Radioactivity in the control organs was limited to the blood pool. Negative control antibody HMFG2 showed no localization on the tumour cells. These results were not due to differences in antigenic expression of the tumour cells but reflect the problem of accessibility of antigenic sites in vivo.

Antibody-guided localization of intraperitoneal tumors following intraperitoneal or intravenous antibody administration

Intraperitoneal tumors from a human cancer cell line (LoVo) were established in nude mice by i.p. inoculation of a single cell suspension. Two preparations of the same monoclonal antibody, radiolabeled with 125I and 131I were injected i.p. and i.v. into the same animals. Localization was assessed by dissection and counting the activity in tumors and normal tissues. Tumor/tissue ratios 1 h after i.p. injection of antibody were approximately 50 times higher than after i.v. administration. This i.p./i.v. advantage fell to around 4 by 8 h and was just greater than 1 by 24 h. This effect was observed with both specific and nonspecific antibody, indicating that it is due to the route of administration. However, the absolute amounts of antibody bound to tumors depended on the specificity of the antibody. Twenty % of the injected dose of specific antibody was bound per gram to tumor 1 to 2 h after i.p. injection, falling to 10%/g by 24 h and remaining at this level up to 5 days after antibody administration. In contrast, less than 10%/g of nonspecific antibody was detected in tumors after 1 h; this fell rapidly to normal organ levels of less than 5%/g by 8 h. This study demonstrates a major advantage when administering radiolabeled monoclonal antibodies i.p. for targeting intraperitoneal tumors.

Enhancement by gamma-interferon of in vivo tumor radiolocalization by a monoclonal antibody against HLA-DR antigen

Athymic nu/nu (nude) mice bearing s.c. human breast tumors were treated systemically with recombinant human gamma-interferon. These tumors were phenotypically negative for HLA-DR prior to therapy, but after 4 days of treatment, 80% of the cells expressed this antigen in vivo as assessed by immunoperoxidase (F. R. Balkwill et al., Eur. J. Cancer Clin. Oncol., in press, 1986). A radioiodine-labeled murine monoclonal antibody (TAL-1B5) against HLA-DR specifically localized to the tumors in recombinant human gamma-interferon-treated but not in control mice. An isotype-identical murine monoclonal antibody that did not react with control or recombinant human gamma-interferon-treated tumors did not show any specific localization. These results demonstrate that specific localization to tumors of radio-labeled monoclonal antibodies to HLA-DR can be facilitated by systemic therapy with gamma-interferon.

Diffusion of oxygen through the mouse ear

Oxygen tension differences across the mouse ear have been measured polarographically under conditions of no blood flow. For some experiments the ear was split into two by cleavage along the central cartilage plate, and the diffusion of oxygen measured in both directions across these asymmetrical preparations. Measurements were also made on ears from which the stratum corneum had been removed by stripping with Sellotape. It was possible to relate these results to a simple multi-layer diffusion model. The main barrier to diffusion of oxygen resides in the stratum corneum, whose permeability is estimated to be 1 . 2 X 10(-8) ml O2 atm-1 cm-1 S-1. The permeability of the rest of the ear is 4 . 7 X 10(-7) ml O2 atm-1 cm-1 S-1. The inhibition of tissue respiration by the local injection of solutions of sodium amytal, potassium cyanide and other substances reduced the oxygen gradients by factors of between 3 and 7. Cooling the ear from room temperature to 0 degree C reduced the gradients by a factor of about 4.