Neutron RBEs for cytopenia and repopulation of stroma and hematopoietic stem cells: mathematical models of marrow cell kinetics.

The objectives of this study were to (a) extend previous bone-marrow cell kinetics models that have been published for ionizing photons to include neutron radiations, and (b) provide Relative Biological Effectiveness (RBE) values for time-specific cell killing (cytopenia) and compensatory cellular proliferation (repopulation in response to toxic injury) for neutron doses ranging from 0.01 to 4.5 Gy delivered uniformly over one minute, hour, day, week, and month. RBEs for cytopenia of a cell lineage were based on ratios of protocol-specific doses that determined the same cell population nadir, whereas the RBEs for repopulation of a lineage were based on the ratios of protocol-specific doses that corresponded to the same total number of cells killed over the radiation treatments, and which should be replaced for long-term survival of the animal. Time-dependent RBEs were computed for neutron exposures relative to the effect of 60Co gamma rays given as a prompt dose. By the use of these RBE factors, low or variable dose rates, dose fractionations given over long periods of time, and different protocols involving several radiation qualities can be converted realistically, and by standard convention, into an equivalent dose of a reference radiation comprised of x or gamma rays given either as a pulse or at any other reference dose rate for which risk information based on epidemiological or animal dose-response data are available. For stromal tissues irradiated by fission neutrons, time-dependent RBEs for cytopenia were computed to range from 4.24 to 0.70 and RBEs for repopulation varied from a high of 6.88 to a low of 2.24. For hematopoietic stem cells irradiated by fission neutrons, time-dependent RBEs for cytopenia were computed to range from 5.02 to 0.22 and RBEs for repopulation varied from a high of 5.02 to a low of 1.98. RBEs based on tissue-kerma-free-in-air would be about twofold lower for isotropic cloud or rotational exposure geometries because marrow dose from isotropic neutron fields suffer factor-of-two greater attenuation than the gamma doses from gamma photons. For certain doses and dose rates, the RBE values computed for compensatory cellular proliferation clearly demonstrate the behavior that is commonly referred to as an inverse dose-rate effect, i.e., protraction of exposure may-under certain conditions-increase the magnitude of the dose response. Furthermore, because of non-linear rates for repair and repopulation, the highest RBEs are not necessarily found for the lowest doses nor the lowest RBEs always found at the highest doses.