Several radionuclides used in medical imaging emit Auger electrons, which, depending on the targeting strategy, either may be exploited for therapeutic purposes or may contribute to an unintentional mean absorbed dose burden. In this study, the virtues of 12 Auger electron-emitting radionuclides were evaluated in terms of cellular S values in concentric and eccentric cell-nucleus arrangements and by comparing their dose-point kernels. Methods: The Monte Carlo code PENELOPE was used to transport the full particulate spectrum of 67 Pt, and 201 Tl by means of event-by-event simulations. Cellular S values were calculated for varying cell and nucleus radii, and the effects of cell eccentricity on S values were evaluated. Dosepoint kernels were determined up to 30 μm. Energy deposition at DNA scales was also compared with an α emitter, 223 Ra. Results: PENELOPE-determined S values were generally within 10% of MIRD values when the source and target regions strongly overlapped, that is, S(nucleus←nucleus) configurations, but greater differences were noted for S(nucleus←cytoplasm) and S(nucleus←cell surface) configurations. Cell eccentricity had the greatest effect when the nucleus was small, compared with the cell size, and when the radiation sources were on the cell surface. Dose-point kernels taken together with the energy spectra of the radionuclides can account for some of the differences in energy deposition patterns between the radionuclides. The energy deposition of most Auger electron emitters at DNA scales of 2 nm or less exceeded that of a monoenergetic 5.77-MeV α particle, but not for 223 Ra. Conclusion: A single-cell dosimetric approach is required to evaluate the efficacy of individual radionuclides for theranostic purposes, taking cell geometry into account, with internalizing and noninternalizing targeting strategies. The therapeutic rationale for molecularly targeted radiotherapy is the selective delivery of a radionuclide to tumor cells via a targeting moiety, thereby enhancing the therapeutic index of the agent. Several Auger electron-emitting radionuclides have been proposed for molecularly targeted radiotherapy of small metastases and disseminated cancer cells, with some promising clinical results (1-3). These radionuclides are well suited to serving as molecularly targeted radiotherapy agents because of the extremely short range in matter (nanometers to a few micrometers) of the low-energy, intermediate-linear-energy-transfer (LET) Auger and Coster-Kronig electrons they emit (4). These electrons account for high energy deposition in the immediate vicinity of the decay site, and because of their short range, irradiation of normal neighboring cells is limited, thus reducing nonspecific radiotoxicity. In addition, radiation emitted during the nuclear decay can be exploited for imaging purposes either with SPECT (in the case of g rays in the energy range of 70-360 keV or Bremsstrahlung imaging for pure b 2 emitters) or PET (in the case of annihilation photons), thus making Auger electron-emitting ra...