The relationship between the radiation dose delivered to a tumor and its effect is not completely predictable. Uncertainty in the estimation of the boron concentration in a tumor, variation in the radiation sensitivity of the tumor cells, and the complexity of the interactions between the four types of radiation comprising the boron neutron capture therapy (BNCT) dose contribute to this uncertainty. We reanalyzed the data of our previous papers to investigate the variation in radiosensitivity of tumor cells to the 10B(n,α)7Li dose: the dose generated by the reaction of thermal neutrons and 10B, hereafter the ‘boron-neutron dose’. The radiosensitivities of five tumors (EL4, SAS/neo, SAS/mp53, SCCVII and B16-BL6 melanoma) were examined. For the combination of p-boron-L-phenylalanine (BPA: C9H12BNO4) with neutron irradiation, D0, the cell survival curve for the boron-neutron dose was the smallest for the SAS/neo, followed by the EL4, SAS/mp53, SCCVII and B16-BL6 melanoma, in that order. For the combination of mercaptoundecahydrododecaborate (BSH: Na2B12H11SH) with neutron irradiation, D0 was the smallest for the EL4, followed by the SAS/neo, B16–BL6melanoma, SAS/mp53 and SCCVII, in that order. The relationships between these D0 values and the nucleocytoplasmic ratios (Xncs) or cell size indices (Xcs) obtained by histopathological microslide image were as follows: (D0 = 0.1341Xnc–1.586, R2 = 0.9721) for all tumor types with BPA-BNCT, and D0 = 0.0122Xcs–0.1319 (R2 = 0.9795) for four tumor types (all except the B16-BL6 melanoma) with BSH-BNCT. Based on these results, we proposed a new biologically equivalent effectiveness factor: the absolute biological effectiveness (ABE) factor. The ABE factor is Gy/D0. Thus, the ABE dose is the physical dose multiplied by the ABE factor, and refers to the dose needed to decrease the cell survival rate to e–ABE dose/Gy.