Dose-Time-Volume Relationships in Normal Tissue Response to Irradiation

Abstract

1. In contrast to early reactions, late radiation-related damage is expressed after a relatively long and dose-dependent latency with both the incidence and severity building up slowly with time and at a dose-dependent rate. The dose-response curve of late reactions is steeper than that of the early ones and this calls for more accuracy in dose delivery. 2. Most tissue responses may be expressed in terms of the underlying dose-survival curves of putative target cell(s). If we assume equal effects for equal repeated doses, a multifraction response curve may be represented by an exponential decline in survival with a slope which depends on the dose/fraction and the shape of the initial part of the single-dose survival curve. It should be remembered, however, that the kinetics of repair during an interfraction interval is both tissue and dose-dependent. 3. The initial slope of the underlying dose-survival curve is thought to be shallower in the case of late than of early reactions. Thus fractionation and reduction in the dose/fraction would result in more sparing of early reacting tissues, than of late ones. 4. In a few tissues, isoeffective total dose (TD) can be expressed as a power function of time (T) and the number fractions (N) provided that the appropriate exponents are used. In case of spinal cord damage, e.g. TD = N0.42 while ignoring any time-dependent recovery. For lung damage TD = N0.377×T0.05 for N up to 20; the time factor accounting for slow repair. 5. The linear-quadratic model (LQ) can well describe the isoeffect doses of many early reactions over a wide range of fraction sizes and grades of severity. However, the model may overestimate the tolerance of some late reacting tissues with doses per fraction <2 Gy. A time factor has been proposed to correct for time-dependent recovery processes such as slow repair or proliferation. This, however, requires a knowledge of the absolute values of the alpha or beta parameters as well as an estimate of the potential doubling time. It is,however, possible to correct for incomplete repair in case of suboptimal interfraction intervals. 6. Residual damage amounts to about 10% in case of acute skin reactions. Residual bone marrow damage may persist for a long time even after small doses and at a time when there is an apparent peripheral blood recovery. Residual damage of the late reactions in the skin, spinal cord and lung amounts to 15–50% according to the dose of the first treatment and the time of re-irradiation. The kidney recovers poorly after the first treatment and is sensitive to re-irradiation. 7. In the clinical sense, an increasing volume may increase the functional derangement, morbidity, symptoms, liability to complications ... etc resulting from a given level of response. In the radiobiological sense, the volume effect is more likely to be related to the structural and functional organization of tissues rather than to changes in cellular radiosensitivities related to volume. Tissues formed of structurally and functionally independent units arranged in parallel, are unlikely to exhibit a volume effect except in relatively small volumes where cellular migration can play a role in their repopulation. In contrast, tissues formed of units arranged in series so that damage of any unit within the treated volume would induce a response, can show a volume effect. Power law relationships are still the most practical form for expressing the volume effect. However, clinical data should be made available in order to deduce a unique power function for each specific type and degree of damage. Probabilistic models offer attractive possibilities but still have to be tested clinically.