Overcoming the nephrotoxicity of radiometal-labeled immunoconjugates: improved cancer therapy administered to a nude mouse model in relation to the internal radiation dosimetry.

BACKGROUND

Elevated renal uptake and extended retention of radiolabeled antibody fragments and peptides is a problem in the therapeutic application of such agents. However, cationic amino acids have been shown to reduce renal accretion. The aims of the current study were to evaluate whether this methodology would benefit therapy with yttrium 90 (90Y)-labeled antibody fragments (Fab, F(ab)2), to establish the relationship between radiation dosimetry and observed biologic effects, and to compare the antitumor efficacy of antibody fragments with that of whole immunoglobulin (Ig)G.

METHODS

The maximum tolerated dose (MTD) and the dose-limiting organ toxicity of 90Y-labeled anti-carcinoembryonic antigen (CEA) MN-14 monoclonal antibodies (Fab, F(ab)2, and IgG) were determined in nude mice bearing GW-39 human colon carcinoma xenografts. The mice were treated with or without kidney protection by administration of D-lysine, with or without bone marrow transplantation (BMT), or with combinations of each. Toxicity and tumor growth were monitored at weekly intervals after radioimmunotherapy. Dosimetry was calculated from biodistribution studies using 88Y-labeled antibody. Three different dosimetric models were examined: 1) taking solely self-to-self doses into account, using S factors for 90Y in spheroids from 0.1 to 1 g; 2) correcting for cross-organ radiation; and 3) using actual mouse anatomy as represented by nuclear magnetic resonance imaging with a three-dimensional internal dosimetry package (3D-ID).

RESULTS

The kidney was the first dose-limiting organ with the use of Fab fragments. Acute radiation nephritis occurred at injected activities > or = 325 microCi, and chronic nephrosis at doses > or = 250 microCi. Activities of 200 microCi were tolerated by 100% of the animals (i.e., the MTD). Application of lysine decreased the renal dose by approximately fivefold, facilitating a 25% increase in the MTD (to 250 microCi), because myelotoxicity became dose-limiting despite red marrow doses of less than 5 gray (Gy). By using BMT and lysine, the MTD could be doubled from 200 to 400 microCi, where no biochemical or histologic evidence of renal damage was observed (kidney dose, < or = 40 Gy). With injected activities of > or = 325 microCi without kidney protection, and with a hepatic self-to-self dose of only 4 Gy, rising liver enzymes were observed, which could be explained only by cross-organ radiation from radioactivity in the kidneys (in the immediate neighborhood of the right kidney up to > or = 150 Gy). The MTD of F(ab)2 fragments could be elevated only by a combination of BMT and lysine. With IgG, the bone marrow alone was dose-limiting. Tumor dosimetry correlated well with antitumor effects; Fab was more effective than F(ab)2, which was consistent with its more favorable dosimetry, and it may also be more effective than IgG due to its higher dose rate and more homogenous distribution. Dosimetry Model 1 was insufficient for predicting biologic effects. Model 2 seemed to be more accurate, accounting for interorgan crossfire. However, Model 3 showed an additional substantial contribution to the red bone marrow dose due to crossfire from the abdominal organs.

CONCLUSIONS

These data show that radiation nephrotoxicity is an important effect of cancer therapy with radiometal-conjugated antibody fragments or peptides. However, this effect can be overcome successfully with the application of cationic amino acids, which substantially increase the anti-tumor efficacy of radiometal-labeled immunoconjugates. For understanding the biologic effects (e.g., liver toxicity) of 90Y in a mouse model, accounting for cross-organ radiation is essential. Further studies with radiometal-conjugated monoclonal antibody fragments and peptides are necessary to determine the MTD, dose-limiting organs, antitumor effectiveness, and nephroprotective effects of cationic amino acids in humans.