RBE values for colo-rectal injury after caesium 137 gamma-ray and neutron irradiation. I. Single doses

The incorporation of the concept of minimum RBE (RBEmin) into the linear-quadratic model and the potential for improved radiobiological analysis of high-LET treatments

PURPOSE

The formulation of relative biological effectiveness (RBE) for high linear energy transfer (high-LET) radiation treatments is revisited. The effects of changed production of sub-lethal damage with varying LET is now considered via the RBEmin concept, where RBEmin represents the lower limit to which RBE tends at high doses per fraction.

MATERIALS AND METHODS

An existing linear-quadratic formulation for calculating RBE variations with fractional dose for high-LET radiations is modified to incorporate the twin concepts of RBEmax (which represents the value of RBE at an effective dose-per-fraction of 0 Gy) and RBEmin.

RESULTS

Fits of the model to data showed RBEmin values in the range of 0.1- 2.27. In all cases the raw data was a better statistical fit to the model which included RBEmin, although this was only very highly significant in one case. In the case of the mouse oesophagus it is shown that, if change in the beta-radiosensitivity coefficient with LET is considered as trivial, an underestimation > 5% in RBE can be expected at X-ray doses of 2 Gy/fraction if RBEmin is not considered. To ensure that the results were not biased by the statistical method used to obtain the parameter values relevant to this analysis (i.e., using fraction-size effect or Fe-plots), an alternative method was used which provided very similar correlation with the data.

CONCLUSIONS

If the production of sublethal damage is considered independent of LET, there will be a risk that non-corrected evaluation of RBE will lead to an over- or under-estimate of RBE at low doses per fractions (the clinically relevant region).

Modeling volume effects of experimental brachytherapy in the rat rectum: uncovering the limitations of a radiobiologic concept

PURPOSE

To analyze the significance of volume effects in experimental brachytherapy, based on modeling normal tissue complication probability.

METHODS AND MATERIALS

Experimental brachytherapy in the rat rectum was based on an eight-step 2.5-mm step size source configuration for 192Ir, afterloaded into an unshielded polystyrene applicator. Volume effects were studied using a half-circumferential lead-shielded applicator and a shorter (two-step) source configuration. The main end point was rectal stenosis.

RESULTS

Rectal stenosis was always caused by a radiation ulcer. With the shielded configuration, single-dose ED50 (50% incidence of rectal stenosis) increased from 23 Gy to 36.5 Gy. Single-dose ED50 for the short configuration was 77.9 Gy. The data showed a reasonable fit to a three-parameter version of the biophysical model described by Jackson et al. (1995). This model assumes that organs consist of a large number of radiobiologically independent subunits and that radiation causes a complication if the fraction of the organ damaged is greater than its functional reserve. The fraction of the organ damaged is calculated summing over fractions of the organ damaged at each dose level. The calculated mean functional reserve (nu50) of the rat rectum, assuming a cumulative functional reserve distribution in the group of experimental rats, was 0.53.

CONCLUSIONS

The volume effect observed within small brachytherapy volumes agreed well with clinical experience of large tolerance doses in contact X-ray therapy. However, the nu50 value was comparable to the high functional reserve value reported for liver. Experimental volume effects probably reflect repair processes originating in the areas adjacent to small radiation fields of brachytherapy more than the radiobiologic characteristics of the cells in the irradiated volume.

Murine strain differences in the volume effect and incidence of radiation-induced colorectal obstruction

PURPOSE

Interindividual variation in the level of normal tissue damage after radiotherapy has been clinically observed. Murine models have suggested that there may be a genetic component to the variation in susceptibility of different radiation-induced normal tissue complications. Currently, there are no experimental data available describing interstrain differences in the "volume effect" for irradiated normal tissues, such as the colorectum. The aims of this study are to determine if there are strain differences in: 1. the incidence of colorectal obstruction; and 2. the volume effect, after irradiation of the colorectum using two mouse strains that are known to vary in their susceptibility for developing pulmonary fibrosis.

METHODS AND MATERIALS

Various lengths (5.2 to 22.9 mm) of the colorectum of male C57B1/6 and C3Hf/Kam mice were irradiated with a single dose (30 Gy) of 137Cs gamma rays. Also, various doses (20 to 35 Gy) were given to a single length (22.9 mm) of colorectum. The incidence of obstruction was determined as a function of length and dose at 6 months after irradiation. The Threshold Probability model was fit to the length-response data.

RESULTS

C57B1/6 mice developed colorectal obstruction at significantly higher incidence than C3Hf/Kam mice at all lengths after a single dose of 30 Gy. In addition, the data showed a strain difference in the threshold length of colorectum that had to be irradiated before obstructions were observed.

CONCLUSION

Strain differences in the incidence of radiation-induced colorectal obstruction were observed, consistent with previous studies that showed a strain difference in radiation-induced pulmonary fibrosis. The presence of a threshold length of colorectum that was different for the two strains is consistent with the concept that there may be a critical threshold amount of colorectal tissue that can tolerate a high dose without complication, and that the dimensions of the threshold may vary among individuals.

Chronic radiation damage in the rat rectum: an analysis of the influences of fractionation, time and volume

PURPOSE

Analysis of four different sets of experiments performed by the G.S.F. group in Munich investigating the late tolerance of the rat rectum to external or intracavitary irradiation.

MATERIAL AND METHODS

The endpoint was late rectal stenosis in female Wistar rats. The raw data were fitted to the linear-quadratic model by means of a likelihood maximization method (Direct Analysis). The model was altered to allow for repopulation, incomplete repair, and varying irradiated lengths of the rectum.

RESULTS

Fractionation sensitivity was high or intermediate (alpha/beta ratio values [95% confidence limits] ranging from 2.67 [0.86, 4.80] to 6.65 [2.21, 11.73] Gy). Significant repopulation occurred when treatments were longer than 5 days (Dprolif equal to 0.61 [0.20, 1.47] and 1.08 [0.58, 1.90] Gy/day, in fractions of 4 Gy). Another interpretation is that radiosensitivity changed during treatment. Repair half-time estimates ranged between 1.84 [1.52, 2.34] and 5.02 [2.83, 21.7] h. Finally, the present analysis indicated that the smallest surviving compartment capable of tissue rescue was about 1/50 to 1/100 of a 1 cm high cylinder of the rectum wall.

CONCLUSIONS

The radiobiological features of late stenosis in the rats are consistent with combined injuries of early and late responding components of the rectal wall. This raises some concerns about the possible danger of hyperfractionated treatments, where the beneficial impact of fraction size reduction may be obviated for interfraction intervals that are too short. Also, accelerated irradiation may result in more late complications because of increased early reactions.

Early and late injuries in mouse rectum after fractionated X-ray and neutron irradiation

PURPOSE

to assess mouse rectum tolerance to fractionated X-ray and neutron irradiation.

MATERIALS AND METHODS

doses per fraction ranged between 0.25 and 35 Gy for X-rays, 0.05 and 12 Gy for neutrons. Neutron top-up doses were added when the fractionated irradiation was given in fractions less than 2 Gy of X-rays or 0.35 Gy of neutrons in order to bring the damage into the detectable range. The early endpoints were the nadir of weight loss occurring within the first 2-3 weeks following irradiation and lethality by 2 months. The late endpoints were the peak of weight reached at maturity of the mice, the proportion of short feces in the daily fecal output at 10 months and lethality by 12 months. The linear-quadratic (LQ) model was fitted to the data (direct "one-step" analysis) and the estimated parameters were used to calculate relative biological effectiveness (RBE) values.

RESULTS

alpha/beta ratio estimates were for X-rays: 19.9 Gy [95% confidence limits: 15.2, 27.0] for weight nadir. 13.4 Gy [9.3, 19.5] for early lethality, 6.4 Gy [3.6, 11.0] for peak weight, and 6.9 Gy [4.2, 10.8] for late lethality, for neutrons 19.9 Gy [9.5, 61.0] for peak weight. The fecal-deformity data were poorly fitted by the LQ model. The RBE was slightly higher for acute endpoints than for the late ones when X-ray fraction sizes were equal to or larger than 10 Gy. However, the change in RBE with decreasing X-ray dose per fraction was much steeper for the late endpoints, so that it became equal to or even higher than for acute reactions at doses per fraction of 5 Gy or less.

CONCLUSION

Our results were consistent with those obtained from previously published studies using the same experimental system but larger doses per fraction.

Development of a shielded 241Am applicator for continuous low dose rate irradiation of rat rectum

This paper describes the development of sources, applicators, and techniques that can be used to irradiate rat rectums with continuous irradiation at dose rates of interest in brachytherapy, either with the full circumference of the rectum irradiated, or with half of the circumference shielded from the radiation. The system uses encapsulated 241Am sources, to irradiate rat rectum with 60 keV photons continuously at a dose rate of up to 50 cGy/hr. Details of the design and fabrication of the 241Am sources, the rectal applicator, the dosimetry of the system, and the protocols for preparing and irradiating the rats, and for detecting early rectal injury using histological examination of irradiated rectum are presented. Highly effective shielding (attenuation factors as low as 0.04) of half of the circumference of the rat rectum was achievable. Unidirectional 241Am irradiators for intracavitary brachytherapy offer a unique tool for examining the effects of shielding a portion of the circumference of the rat rectum, on the radiation tolerance of the rectum.

The effect of cis-diamminedichloroplatinum(II) on radiation damage in mouse rectum after fractionated irradiation

The influence of cis-diamminedichloroplatinum(II) (c-DDP) on radiation injury in the rectum of mice was investigated after single dose and fractionated irradiation. Mice were exposed to single doses, or 2, 4 or 8 fractions of X-rays given daily (or twice a day for the 8 fractions) with or without 8 mg/kg of c-DDP. The incidence of rectal stenosis and of anal discharge were scored and a direct analysis of the data for linear-quadratic dose-dependence of damage was performed with correction for censoring. For each endpoint, c-DDP did not significantly change the dose-response curves after fractionated irradiation. Subtile modifications in the coefficients of the linear-quadratic relationship were observed. The drug appeared to increase the alpha term by a factor of 2.3 and the beta term by a factor of 1.5, but the differences did not reach statistical significance. For the rectal stenosis endpoint, the alpha/beta was 4.4 Gy after irradiation alone and 6.9 Gy after combination treatment. Again the differences were not significant. These data suggest that c-DDP did not reduce the repair capacity in the mouse rectum during fractionated irradiation. Combined modality therapy with c-DDP and radiation would therefore not be expected to cause an increase in late damage in the large bowel.

Adhesion formation in experimental chronic radiation enteropathy

We have studied the late changes associated with radiation enteropathy in mice over a period of 224 days following single or split doses of gamma radiation delivered to the total abdomen (TAI). We focused on the importance of adhesion formation as a cause of strictures and gut-associated deaths following TAI. Gut-associated peritoneal adhesions were found in mice 2-7 months after receiving 13.5-17.5 Gy TAI and appeared to constitute the most consistent serious late effect of irradiation. There was a good correlation between adhesion formation and death for both the single and split-doses of radiation. Adhesions primarily involved the large gut, normally near the cecum. They appeared to result from serosal breakdown and were the major cause of partial gut obstruction. Submucosal fibrosis was present but seemed to be a comparatively minor cause of strictures. Local lymphoid hyperreactivity was also seen following TAI and may have contributed to the late sequelae. The complexity of the pathogenesis of chronic radiation enteropathy was indicated by finding three successive waves of non-scheduled deaths following TAI. The first wave (28-70 days) was not related to adhesion formation and may have been due to localized failure of mucosa to regenerate after irradiation with consequent ulceration. The second wave (98-140 days) occurred over the period when adhesion formation and fibrosis were most marked. In the third wave (168-224 days), the additional complication of fluid exudation was seen. Further experimentation is obviously needed to better define the complex pathogenesis of radiation enteropathy with dose and time after radiation but our data strongly support a multifactorial causation with an important role for adhesion formation in the disease complex.

Radiation tolerance of the rat rectum to fractionated X-rays and pi-mesons

A 2.5 cm segment of the rectum of female F344 rats was irradiated with up to 10 fractions of X-rays or pions. Transient signs of acute proctitis were followed by chronic rectal injury starting at about 2 months. Recto-vaginal fistulas were frequently observed in animals with severe chronic injury. Two patterns of chronic injury were observed, an early type consisting of deep ulcers and fistulas, and a late type characterized by vascular injury, fibrosis and mucosal cysts. In a 4-fraction X-ray experiment, the influence of a low-residue diet was compared with a regular diet, showing no significant differences in pathology or survival. Isoeffective doses were determined for the occurrence of severe rectal injury at 250 days. Fitting the data to an LQ-model yielded an alpha/beta ratio of 13 Gy for pions, and 6.5 Gy for X-rays. Increasing the overall treatment time from 9 days to one month gave a significant rise in isoeffective doses for chronic injury. This suggests that the more delayed types of injury did not develop independently from the acute mucosal changes. The RBE of pions for rectal injury at 250 days was 1.2 for single doses, increasing to about 1.4-1.5 at a dose per fraction of 4 Gy. These values are similar to those obtained in other acute and late responding tissues.

Early and late damage in the mouse rectum after irradiation and cis-diamminedichloroplatinum(II)

The effects of radiation and cis-diamminedichloroplatinum(II) (c-DDP) in the mouse rectum were assessed using two functional endpoints (anal discharge/diarrhoea and body weight changes), one lethal endpoint (obstructive rectal stenosis) and a microscopic endpoint (semiquantitative histological scoring). After irradiation, anal discharge, diarrhoea and lethal rectal stenosis were found to be the result of submucosal fibrosis with a secondary mucosal ulcer due to mechanical damage by the faeces. Weight loss at both early and late times after irradiation seemed to be related to epithelial cellular depletion. c-DDP, when given in combination with X-rays, did not enhance the incidence of late anal discharge, diarrhoea or rectal stenosis. Using an arbitrary semiquantitative scale for measuring the degree and extent of late rectal injury, c-DDP sometimes seemed to postpone the expression of radiation damage in the rectum. This effect, however, might also be the result only of interexperimental variation. An increase by c-DDP in the early body weight loss after irradiation was also observed which could be explained by an additive effect in the rectal epithelium.

The r.b.e. of fast fission neutrons (2 MeV) for chronic radiation damage of the large bowel of rats after single dose and fractionated irradiation

Late damage in the rectum of rats was studied after local irradiation with 2 MeV fast fission neutrons and the results were compared with those for 300 kV X-rays. A similarity between latency times, and between clinical, gross pathological and histological appearances of the rectal obstruction after the two types of radiation was observed. The r.b.e. evaluated from ED 50/200 days values was 1.8 for single doses, 2.1 for two fractions and 3 for five fractions and these are similar to r.b.e.s from other investigations of chronic intestinal radiation damage. Using linear quadratic analysis, a limiting r.b.e. of 5 was obtained. The beta-values are similar for both radiation qualities. Because of the high gamma-contamination (25 per cent) in the RENT I beam a possible correction of the r.b.e. for neutrons only, assuming interaction, is discussed.

Protection of acute and late radiation damage of the gastrointestinal tract by WR-2721

WR-2721 was investigated for its protective effect against acute and late damage produced by irradiation of the esophagus, small intestine, and colon of mice. The microcolony assay was used to measure the acute response of the small intestine and the colon, and an LD50 assay (within the 28- to 42-day time range) was used to measure acute esophageal damage. A dose of WR-2721 at 400 mg/kg, injected 30 min prior to irradiation, resulted in a protection factor (PF) of 1.6 against radiation damage in these three regions of the gastrointestinal tract. Lethality and histology scores were applied to determine late radiation damage to the rectum, at times ranging from 3 to 15 months after irradiation. Deaths occurred after doses of 20 Gy and above throughout the postirradiation period. WR-2721 increased the survival of mice; the PF calculated from the LD50 values was 1.5. PFs of animal survival did not vary during the observation period. Histological studies showed evidence of ulceration, fibrosis, and vascular changes as late radiation damage. WR-2721 protected against radiation-induced histological damage with a PF of 1.3. There was no qualitative difference between the types of histological damage observed in the group undergoing only irradiation and the group treated with WR-2721. Biochemical measurements of fibrosis by hydroxyproline determination of collagen 16 months after irradiation showed an increase in collagen per milligram wet weight of rectal tissue in all irradiated groups, but no increase in the amount of collagen per 5 mm segment of the rectum. Thus it appears that the apparent fibrosis is a result of atrophy rather than collagen accumulation. We conclude that WR-2721 is indeed effective at protection against late damage from large single doses of radiation to the rectum as measured histologically and also improves the long-term survival of the mice, although the target cells for this damage are not known.

Dose-response for bone regeneration after single doses of 60Co irradiation

The Bone Growth Chamber (BGC) methodology was used to establish a dose-response relationship for regeneration of mature bone tissue after irradiation of 5, 8, 11, 15 and 25 Gy single dose 60Co. The BGC, which is a titanium implant, was inserted in the proximal tibial metaphyses, bilaterally, of a rabbit immediately following local irradiation to one tibia. Each animal thus served as its own control. During a healing period of 4 weeks, the two canals penetrating the implant became filled with more or less newly formed bone. At the end of the healing period, the implants were removed and taken apart and the newly formed bone was collected and its volume measured by microradiography and microdensitometry. It was found that in the dose range of 5 to 8 Gy bone regeneration was reduced by about 20% as compared to non-irradiated controls. Between 8 and 11 Gy, there was a critical range in that a small increase in dose resulted in a greatly reduced bone formation. At 11 Gy and above, the depression in bone formation, as compared to non-irradiated controls, was about 65 to 75%.

Some short-term cell kinetic effects of ionizing radiation on mouse bladder urothelium

Early morphological changes and the pattern of reactive proliferation of the hairless mouse urinary bladder urothelium after irradiation are reported. Groups of female hairless mice were anaesthetized with sodium pentobarbital and irradiated over the bladder region with 0, 10, 20 and 30 Gy. Control groups were sham-treated. Short-term cell kinetic changes were monitored using incorporation of tritiated thymidine and flow cytometry. Only minor radiation-induced alterations in the cell kinetic pattern were recorded, and no significant histomorphological changes were seen. However, a marked increase in the thymidine incorporation was seen in the control animals on the first day after anaesthesia. Radiation proctitis induced early deaths in the 30 Gy irradiated animals. The present results are in accordance with commonly accepted radiobiological theories, but not in agreement with results previously published by others.

Large bowel stenosis in rats after fractionated local irradiation

After local irradiation of the distal large bowel rats may develop fatal large bowel stenosis within 200 days. The incidence rises steeply with dose. The LD50 for single dose irradiation is 20 Gy. The tolerance of the large bowel to local irradiation depends strongly on dose fractionation. Extending the intervals between the fractions and decreasing the dose per fraction both have a pronounced dose-sparing effect.

Late effects of irradiation in mouse jejunum

The response of mouse jejunum at intervals up to 1 year after single 'priming' doses of X-rays has been assessed by crypt survival after retreatment with single doses of X-rays and morphometric analysis of changes in the intestinal submucosa. The first 'priming' dose was given as a single dose to the whole abdomen. To assess crypt survival, groups of mice were retreated to the whole body with a range of test doses 2, 6 or 12 months later, while other groups of mice were given only the priming doses. These data were compared to crypt survival in mice not previously irradiated. The crypt dose-survival curves in mice re-irradiated at all three intervals after priming irradiation were displaced to higher doses in pre-treated than in non-pre-treated mice and were characterized by higher D0 values. Misonidazole given before the test exposure reversed this effect so that the dose survival curve for crypts in pre-treated mice were superimposed on that for mice not previously irradiated, suggesting that the increase in isoeffect dose and the change in the D0 in previously exposed mice was due to crypt hypoxia. Quantification of the area of the submucosa showed that its area was increased at all three times after the priming doses and was a result of collagen deposition and oedema. Thus, the hypoxia in the crypts was probably secondary to these changes. Deaths began at 6-7 months after priming irradiation and were due to intestinal obstruction and stenosis. Thus, as in other tissues, two phases of injury can be assayed in the intestine of experimental animals.

RBE values and repair characteristics for colo-rectal injury after caesium 137 gamma-ray and neutron irradiation. II. Fractionation up to ten doses

Early and late colo-rectal damage in mice have been assessed after 137Cs gamma irradiation and 3 MeV neutrons given as 1,2,5 or 10 fractions. Damage was measured by early changes in body weight, the late production of short faecal pellets and the pattern of lethality after irradiation. The data have been analysed in terms of the time course of expression of damage, fractionation effects and the RBE for neutrons over a wide range of doses per fraction (0.5-12.5 Gy neutrons, 3.5-33.5 Gy gamma rays). An initial epithelial denudation led to an early loss of weight, maximal at 11-17 days after irradiation. A dose-dependent weight reduction persisted over the animals' life-time. Deaths after localised pelvic gamma irradiation were progressive with no sharp demarcation between early or late phases of injury. The time course for lethality was qualitatively similar after neutrons. Beyond six months the rectum became constricted by fibrosis and a higher proportion of small faecal pellets was observed. At 6-15 months relatively shallow dose-response curves were obtained for this change. The sparing effect of fractionation was marked for the gamma-irradiated mice and almost absent after neutrons. A very high repair increment (11 Gy) was seen with two gamma-ray fractions of 20 Gy. At lower doses per fraction the proportion of each gamma-ray fraction recovered was 50-69% for all assays, i.e., similar to that for other normal tissues. There was a slight enhancement in the sparing effect for the late compared with the early assays over the lower dose range. The RBE was strongly dependent on dose per fraction because of the lack of reparable damage after neutrons. The RBE for both early and late effects was 5.0 at a neutron dose per fraction of 1 Gy. Extrapolation of the RBE data to lower doses, using the linear quadratic model, predicts a higher RBE for late (7.4-12.7) than for early damage (5.7-8.5) if gamma-ray doses below 5 Gy are used.

Cesium: a bibliography update

A brief outline of both industrial and medicinal uses is given. This is followed by a survey of literature on cesium salts for the period between January 1981 to May 1984. The bibliography searched indicates an interesting pharmacological and behavioral effect for Cs-salts in addition to its well known physiological and itopic effects. A pattern of clinical trials suggests the potential of cesium salts in certain cancer therapies, affective disorders, tumor imaging, radiotherapy and certain cardiovascular usages. The need for continued probing into the biological activity of this alkalimetal is suggested.