It is now well established that micronuclei frequency does not always rank cell lines according to radiosensitivity. There is, however, a growing interest in reconstructing cellular radiosensitivity (measured by colony assay) from concurrent micronucleus and apoptosis data. Using a variety of radiosensitive and radioresistant cell lines, we have derived a missing parameter –P _oe , the probability of cell death by other events such as small deletions, chromosome aberrations, late apoptosis and necrosis which are undetectable by micronucleus and apoptosis assays performed at a single time point. In the radioresistant glioma cell lines G120, G60, G28, G44 and G62 (SF2 ≥0.59), a characteristic threshold dose exists above which cell loss due to undetectable lesions occurs. In the radiosensitive SK-N-SH and KELLY cell lines (SF2 ≤0.43), the P _oe parameter is positive at very low doses, reaches a maximum and declines at higher doses. In the radiation resistant G28 cells, P _oe was found to be below zero for doses up to 6 Gy. In the G62, G44 and G120 cell lines, the threshold doses to induce P _oe events were 0.87, 3.04 and 3.85 Gy, respectively. Cell death by undetectable lesions is a cell-specific and time-dependent variable. Micronucleus and apoptosis assays performed concurrently and at a specific time point miss cell death due to other events and this may be the reason why reconstruction of cellular radiosensitivity from micronucleus and apoptosis data fails.