The bystander effect, originating from cells irradiated in vitro, describes responses of surrounding cells not targeted by the radiation. Previously we demonstrated that the subcutaneous injection into nude mice of human adenocarcinoma LS174T cells lethally irradiated by Auger electrons from the decay of DNA-incorporated (125)I inhibits growth of co-injected LS174T cells (inhibitory bystander effect; Proc. Natl. Acad. Sci. USA 99, 13765-13770, 2002). We have repeated these studies using cells exposed to lethal doses of (123)I, an Auger electron emitter whose emission spectrum is identical to that of (125)I, and report herein that the decay of (123)I within tumor cell DNA stimulates the proliferation of neighboring unlabeled tumor cells growing subcutaneously in nude mice (stimulatory bystander effect). Similar inhibitory bystander effects ((125)I) and stimulatory bystander effects ((123)I) are obtained in vitro. Moreover, supernatants from cultures with (125)I-labeled cells are positive for tissue inhibitors of metalloproteinases (TIMP1 and TIMP2), and those from cultures with (123)I-labeled cells are positive for angiogenin. These findings call for the re-evaluation of current dosimetric approaches for the estimation of dose-response relationships in individuals after radiopharmaceutical administration or radiocontamination and demonstrate a need to adjust all "calculated" dose estimates by a dose modification factor (DMF), a radionuclide-specific constant that factors in hitherto not-so-well recognized biophysical processes.