Primary tumors from untreated animals demonstrated an intense homogeneous BR96 target antigen staining pattern, and all were given a score of 3. In comparison, a score of 3 was only given to 6 of the 23 distant metastases in animals belonging to Group A, showing consistent CR of the primary tumor after treatment with 177Lu-DOTA-BR96 (Table
1). The metastases with reduced antigen expression showed less BR96 binding, either in specific areas or over the whole tissue section. None of the metastases completely lacked expression of the BR96 target antigen. It would have been interesting to evaluate metastases in animals not administered radioimmunoconjugate, but this would have required the removal of the primary tumors to prolong the survival of these animals to allow development of metastatic disease. In untreated rats, the tumors reach maximal tolerable volume on day 21-28, i.e. prior to earliest detected metastatic disease (Figure
1). Previous attempts to surgically remove tumors inside the abdomen have not been successful. Apart from technical difficulties, surgical removal of the primary tumor might stimulate the growth of distant metastases
. Of the same reason we have not been able to determine whether the antigen expression of the primary tumor becomes more heterogeneous with time and local progression (exceeding 25 x 25 mm) for ethical reasons.
The antigen expression was reduced in 17 of 23 of the analyzed metastases, but it was not possible to determine whether this was due to the metastatic process itself or a result of the radioimmunotherapy, or a combination of both. It is not known whether tumor cells disseminate before, or as a result of, the treatment. However, no visible metastases could be detected at time of treatment.
Since none of the metastases completely lacked the BR96 target antigen, it might be beneficial to repeat radioimmunotherapy with BR96 utilizing a radionuclide with a relatively long range, resulting in irradiation from targeted to adjacent untargeted cells by the cross-fire effect. However, such radionuclides are generally regarded as unsuitable for the treatment of single cells or cell clusters. On the other hand, studies on external beam radiotherapy have indicated that it is not necessary to irradiate all the cells in a tumor to obtain efficacy as a result of the bystander effect
In experimental studies of treatment with BR96-doxrubicin antibody conjugates in a rat brain tumor model, Muldoon et al.
 reported changes in the antigen staining pattern in residual tumors. The immunoconjugate treatment resulted in outgrowth of tumors with areas of low or no antigen staining, as well as areas with moderate to intense staining, while the untreated tumors showed uniform intense staining for BR96.
Our findings indicate that during the planning of repeated radioimmunotherapy, and/or radioimmunotherapy in combination with other antibody therapies, it is important to consider the risk of reduced antigen expression in metastases, as this may result in reduced targeting of the therapeutics.