Normal Tissue Tolerance doses

1. Normal Tissue tolerance doses of Emami et al, (1991).

Emami et al (1991) have reviewed available clinical data and presented normal tissue tolerance doses, in terms of TD_5/5 (the normal tissue tolerance dose at 5% NTCP within 5 years after radiotherapy) and TD_50/5 (the normal tissue tolerance dose at 50% NTCP within 5 years after radiotherapy), for 1/3, 2/3 and 3/3 partial volumes of the normal tissue / organ or a reference volume (length or area) of the normal tissue/organ. The partial volume of a tissue or/and organ is presented in terms of fraction with reference to the reference volume V_ref. In some cases the reference volume of the organ is considered to be the whole volume of the organ or in some it is assumed to be a part of the tissue / organ, such as spinal cord, the tolerance doses are compiled for 5, 10 and 20 cm lengths instead of the volume and 20 cm length of the cord is taken to be the reference length, where volume represents the length. For skin, the tolerance doses are provided for the areas of 10, 30 and 100 cm², where the area of 100cm² is selected as the reference volume. 

Emami et al (1991) have designed their study to compiled 6 points normal tissue tolerance doses for 1/3, 2/3 and 3/3 partial volumes of the different organs at NTCP 5% and 50% (3 points data at NTCP = 5% and 3 point data at NTCP = 50%, thus total data points are 6) (listed in Table-1 of   Emami et al, 1991). The authors were able to compile 6 points tolerance doses only for 11 organs and could provide 5, 4, 3 and 2 points data for 2, 6, 1 and 9 organs respectively. It is mentioned in the report that due unavailability of the data in the literature, all 6 points data could not be reported for a number of organs. In some cases the data are limited to only whole volume (or reference volume) due to small size of the organs, such as optic nerve, chiasma, cauda equina, eye lens and retina. In case of rectum the reference volume is taken as 100 cm³ and tolerance doses are given only for the reference volume. There is no data provided to show the volume effect.

Results

Normal tissue tolerance data of Emami et al (1991) used to fit into the equation (1) to obtain the values of aG, k and N₀ and the results are listed in Table -1 along with the end points for the normal tissues / organs. In case of 2 points data where these parameters cannot be computed, because no attempt could be made to set correlation between NTCP and volume. Hence the value of k is set equal to zero and values of other parameters are computed. Using the values of aG, k and N₀ parameters, determined for included tissues/organs, the values of the tolerance doses for partial volumes of the organs are computed and are listed in Table-3 along with the tolerance doses compiled by Emami et al (1991). Since the parameter aG is a factor of the coefficients a and b (or a/b), so to determine the values of these coefficients, an accurate value of a/b for a tissue / organ must be known. Hence for the same, the published values of a/b, for different organs, are extracted from the literature, and are used to calculated values of a and b. The extracted values of a/b of different tissues / organs, along with their reference of the publication, and calculated values of a and b are listed in Table –5. In the calculation of the values of a and b from the factor aG for listed number of organs, it is assumed that the dose per fraction is 2Gy for the conventional treatment schedule.

Survey of the literature reveals that there is a wide scattering in the normal tissue tolerance data and there is no consensus on the issue among the practiceners. In this study suitable tolerance dose data for the organs have been extracted from the literature and combined together with Emami et al’s (1991) data to compute the values of above said parameters. Table –2 enlists the values of these parameters, i.e. aG, k and N₀, for the listed organs, for the combined tolerance data along with the source of reference. With use of the values of aG, k and N₀ parameters, from Table-2, the values of the tolerance doses for 1/3, 2/3 & 3/3 partial volumes of all listed organs are computed and are listed in Table-4 along with 95 % confidence interval with that of the computed tolerance doses. The parameter aG is used to compute the values of a and b , for combined data set of the tolerance doses for each organ, the published values of a/b for different organs, as used for Emami et al’s (1991) data, have taken into account. The extracted values of a/b of different tissues / organs, along with their reference of the publication, and calculated values of a and b are listed in Table –5.

Using the values of aG, k and N_0, from Tables- 1 & 2, 2 set of curves have been plotted between dose and computed NTCP for bladder, brain, heart and rectum for reported partial and whole volume and are shown in Figures 1-4. The solid lines of the curves are for the Emami et al’s (1991) data and broken lines are for combined data. In the curve fitting, a method of least square fit was used and the data with more than 2 points tolerance doses give all three parameters of the model. Emami et al’s (1991) data with 2 points tolerance doses does not provide the best fit of 3-parameter model and hence the values of these parameters cannot be computed. So to solve the problem for the data with 2 points tolerance doses, the volume dependent parameter k is set to zero, because there is no conclusion could be made on volume dependency of the organ, and rest of the parameters are calculated from these data.