Targeted radionuclide therapy: theoretical study of the relationship between tumour control probability and tumour radius for a32P/33P radionuclide cocktail

As revealed by previous theoretical studies, targeted radionuclide therapy (TRT) that relies on a single beta-emitting radioisotope is likely to be inappropriate for clinical scenarios such as disseminated malignancy. For a patient with a vast number of tumours and metastases of largely differing sizes a high level of therapeutical efficiency might be achieved only for a restricted range of tumour sizes. This is due to the limited range of beta-electrons in human tissue, essentially causing the therapeutical impact to vary tremendously with tumour size. The dependence of curability on the tumour dimension is expected to be significantly altered if a radionuclide cocktail, consisting of a long-range and a short-range beta-emitter, such as (32)P and (33)P, is involved in the treatment. In this study, a radiation transport simulation was performed, using the MCNP4c2 Monte Carlo code, in order to investigate the relationship between tumour control probability (TCP) and tumour size, associated with concurrent use of (32)P and (33)P. Two different models of intratumoural distribution of cumulated activity were taken into account. One simulated an ideal radionuclide uptake in tumour tissue and the other referred to a limited radiotracer penetration. The results were examined in comparison to tumours targeted with pure (32)P, (33)P and (131)I. For both uptake scenarios a considerable reduction of the overall variation of TCP and thus an increasing chance of achieving tumour cure was observed for tumour sizes ranging from microscopic dimensions up to macroscopic diameters, if the targeted radionuclide treatment relies on a (32)P/(33)P cocktail. It was revealed that particular attention has to be given to the ratio of the (32)P and (33)P specific cumulated activities (SCA) in the tumour, since this is a significant determinant of the resulting behaviour of tumour control probability as the tumour diameter varies. This study suggests that a 32P/33P approach is more applicable to diseases that involve a variety of tumours and metastases differing in size.