Silicon diodes as detectors for backscatter radiation in photon fields.

We investigated the use of unencapsulated silicon semiconductor detectors for backscatter radiation detection. The results were compared with Monte Carlo (MC) calculations modelling the experimental set-up. A special diode was manufactured, which was designed so that it allowed the positioning of different materials in close contact with the detector surface. Polymethylmethacrylate (PMMA), Pb, Ti and Fe (stainless steel) were used as backscatter materials. The diode signal was measured by integrating the current when irradiating the diode with an equal photon fluence obtained from a medical Co-60 source. When compared to the signal with PMMA as backscatter material the increase in signal was 21%, 27% and 73% for Ti, Fe and Pb, respectively. This is in reasonable agreement with the MC calculations, when taking the effective measurement depth in the Si diode detector into account.

Anastomotic breaking strength and healing of anastomoses in rat intestine with and without chronic radiation damage.

OBJECTIVE: To assess the influence of chronic radiation damage on anastomotic healing in the small bowel in rats. DESIGN: Controlled laboratory study. SETTING: University hospital, Sweden. MATERIAL: 90 male Sprague-Dawley rats. INTERVENTIONS: A short segment of the distal ileum was exteriorised and irradiated with a single dose (experimental group, n = 45) or exposed only (control group, n = 45). Twenty weeks later resection and anastomosis were done within this segment using 7/0 polypropylene. MAIN OUTCOME MEASURES: The anastomotic breaking strength, the amount of perianastomotic hydroxyproline, and the number of anastomotic complications. RESULTS: The breaking strength and the amount of perianastomotic hydroxyproline were higher in the irradiated than in the non-irradiated group. In contrast, anastomotic complications were significantly more common in irradiated animals. CONCLUSION: Anastomotic complications in irradiated intestine are not related to the amount of perianastomotic collagen or to breaking strength.

External beam radiotherapy of localized prostatic adenocarcinoma. Evaluation of conformal therapy, field number and target margins.

The purpose of the present study was to identify factors of importance in the planning of external beam radiotherapy of prostatic adenocarcinoma. Seven patients with urogenital cancers were planned for external radiotherapy of the prostate. Four different techniques were used, viz. a 4-field box technique and four-, five- or six-field conformal therapy set-ups combined with three different margins (1-3 cm). The evaluations were based on the doses delivered to the rectum and the urinary bladder. A normal tissue complication probability (NTCP) was calculated for each plan using Lyman's dose volume reduction method. The most important factors that resulted in a decrease of the dose delivered to the rectum and the bladder were the use of conformal therapy and smaller margins. Conformal therapy seemed more important for the dose distribution in the urinary bladder. Five- and six-field set-ups were not significantly better than those with four fields. NTCP calculations were in accordance with the evaluation of the dose volume histograms. To conclude, four-field conformal therapy utilizing reduced margins improves the dose distribution to the rectum and the urinary bladder in the radiotherapy of prostatic adenocarcinoma.

Field displacement during external radiotherapy in prostatic adenocarcinoma treated with radioactive 198Au implants and external irradiation.

The purpose of this work was to study displacement error and internal movements of the prostate during external beam radiotherapy. Verification films in the frontal (n = 194) and lateral (n = 64) portals were investigated in 14 patients treated with radioactive 198Au implants. Displacement errors of two implants were investigated. In seven patients, filling of the rectum and the bladder with contrast medium or isotonic saline was performed during CT investigation for planning purposes to detect movements of the prostate. Most (95%) of the displacement errors were less than 10 mm in the frontal portal and less than 15 mm in the lateral portals. No correlation to the patient's weight was found. The displacement errors were randomly distributed. The spatial relations between the implants were not altered during the treatments. Small movements of the prostate were observed. To conclude, the positioning system employed at present (laser) can be sufficient for the margins used (2 cm). In lateral portals, however, the system did not have the ability to detect a possible systematic displacement error from simulator to accelerator. The intention is to decrease the margins to 1 cm, which will necessitate a better positioning system.

A general solution to charged particle beam flattening using an optimized dual-scattering-foil technique, with application to proton therapy beams.

This paper describes a dual-scattering-foil technique for flattening of radiotherapeutic charged particle beams. A theory for optimization of shapes and thicknesses of the scattering foils is presented. The result is a universal optimal secondary-scatterer profile, which can be adapted to any charged particle beam by a simple scaling procedure. The calculation of the mean square scattering angle of the beam after passing through the scattering foils is done using the generalized Fermi-Eyges model for charged particle transport. It is shown that the fluence profile in the plane of interest can be made flat to better than 1% inside a predefined beam radius provided the shaped secondary scatterer has the universal radial thickness profile. The thicknesses of the two foils are optimized to minimize the total energy loss. The theory has been tested experimentally in an 180 MeV clinical proton beam. The measured distributions agree well with the calculations.

Combined systemic chemotherapy and irradiation of muscle-invasive transitional cell carcinoma of the urinary bladder.

A phase II study is presented, which encompasses the period June 1987 until July 1993, and includes 53 patients with muscle-invasive bladder cancer T2-4b who, due to age and/or poor health (37 cases) or primarily extensive lesions (18 cases), were considered inoperable and for whom treatment with neoadjuvant chemotherapy (cisplatin/methotrexate/leucovorin rescue) and radical irradiation was planned. The total number of intended chemotherapy courses could be delivered without undue toxicity to 46 patients (83%) and 44 subsequently underwent radiotherapy: this modality was, by and large, well tolerated. The primary transurethral resection and chemotherapy produced an objective response in 62% of the 53 patients and in 75% of the 44 evaluable patients. The combined programme produced an objective response in 83% of the 37 evaluable patients, 71% in the 44 patients who completed the combined programme and in 59% of the total group of 53 patients. The follow-up ranged from 3 to 62 months. Radiotherapy increased the total objective response rate, proving effective in approximately 50% of patients who did not respond to chemotherapy. The results of this study are regarded as promising and pave the way for a phase III trial.

Estramustine potentiates the radiation effect in human prostate tumor transplant in nude mice.

In this study, we have investigated the combined effect of estramustine treatment and external beam radiation on human prostatic cancer tumor cells (DU 145) transplanted in nude mice. The treatment was given according to two different schedules. In the first treatment regimen, estramustine was administered intraperitoneally (i.p.) intermittently for 20 days. The radiation therapy, which was started on day 9, was given with 6 Gy fractions during an 11-day-long period to a total dose of 36 Gy. The combination treatment (estramustine + radiation) resulted in a significant tumor growth retardation as compared to the control group. This pronounced effect was seen neither with radiation alone nor with estramustine alone. In order to further extend the radiation treatment time, a second therapy regimen was employed. In this part of the study, estramustine was administered i.p. intermittently for 26 days. The radiation therapy, which was started on day 6, was given with 4 Gy fractions during a 21-day-long period to a total dose of 40 Gy. Under these conditions, a significant tumor growth retardation was disclosed, when comparing the combination treatment (estramustine + radiation) with radiation alone. The tumors were analyzed for content of necrosis and proliferative activity. The largest proportion of necrosis was seen in the combination (estramustine + radiation) treatment group. Also, the tumors from this group expressed a decreased proliferative activity. The data indicate that estramustine acts as a radiosensitizing agent in human prostatic cancer cells in vivo. The radiosensitizing properties of the drug encourage further studies with respect to clinical application.

Renal function after autologous bone marrow transplantation.

Seventy-two out of 102 consecutive patients autografted for various hematologic and lymphoid malignancies had a relapse-free survival of greater than 6 months after autologous bone marrow transplantation (ABMT) and were evaluated for long-term effect of the treatment on the renal function. The myeloablative therapy included total body irradiation (TBI) in a single fraction of 7.5 Gy in 41/72 patients. Mean glomerular filtration rate (GFR) showed a significant decrease (p less than 0.01) and serum creatinine and serum urea an increase (p less than 0.05) 6 months after ABMT. Twelve of 72 patients (17%) developed renal dysfunction defined as greater than 25% decrease in GFR, in most cases accompanied by hematuria and proteinuria. Onset was 3-6 months after ABMT. Some patients have later improved considerably, but others continue to deteriorate in renal function. The single most important risk factor for renal dysfunction after ABMT was irradiation. Renal damage was most frequent in lymphoma patients conditioned with BEAC (carmustine [BCNU], etoposide, cytarabine, cyclophosphamide) followed by irradiation, suggesting that this drug combination might have potentiated the toxicity of irradiation. Nephrotoxic antibiotics probably contributed to renal damage in individual cases. Young age did not appear to be a risk factor. Our data indicate that combined treatment with BEAC and TBI should be used with caution and that renal function should be monitored in all patients after bone marrow transplantation to detect any new toxicity patterns of the various conditioning regimens currently used.

Local failure in patients treated with radiotherapy and multidrug chemotherapy for small cell lung cancer.

Fifty-three patients with small cell carcinoma of the lung were treated with chemotherapy and radiotherapy, 40 Gy in the chest tumour. Intrathoracic failure occurred in 89% of the cases with extensive disease and in 60% of those with limited disease. Since 86% of all failures were localized within the target volume, one can conclude that in most cases the radiation dose was too low for eradication of the tumour. The treatment technique resulted in dose inhomogeneities of more than +/- 5% in 45% of the cases. The high local failure rate might indicate the need of improved radiotherapy, in the first place higher radiation dose. However, 82% of the patients with limited disease and local failure and 50% of those without local failure also developed distant metastases. This might indicate that the curative potential of improved thoracic radiotherapy probably is limited. Besides, lethal treatment toxicity affected particularly patients in whom local cure had been achieved, indicating the difficulty of increasing the treatment intensity without increasing the lethal toxicity in potentially curable cases.

General specifications for silicon semiconductors for use in radiation dosimetry.

Silicon semiconductor detectors used in radiation dosimetry have different properties, just as e.g. ionisation chambers, affecting the interaction of radiation with matter in the vicinity of the sensitive volume of the detector, e.g. wall materials, and also the collection of the charges liberated in the detector by the radiation. The charge collection depends on impurities, lattice imperfections and other properties of the semiconductor crystal. In this paper the relevant parameters of a silicon semiconductor detector intended for dosimetry are reviewed. The influence of doping material, doping level, various effects of radiation damage, mechanical construction, detector size, statistical noise and connection to the electrometer is discussed.

Patient dose measurements in photon fields by means of silicon semiconductor detectors.

Semiconductor detectors based on p-type silicon and designed for in vivo measurement of entrance dose at the reference point from photon radiation fields, are described. To estimate the absorbed dose at the reference point from measurements with a thin detector, field-size dependent correction factors must be applied to the reading, as the shape of the dose buildup curve varies with field size. To decrease or avoid field-size dependent correction factors, the detector can be covered with a buildup cap. The presence of such a detector will cause perturbation of the radiation field. Therefore, the design of a detector, irrespective of its type, intended for patient dosimetry involves a compromise between minimizing the radiation field perturbation and minimizing field-size dependent correction factors. Detectors with three different buildup caps were designed to cover the energy range from cobalt-60 to 16-MV x rays. The three different detector types were investigated with respect to their signal dependence on field size, field perturbation, and directional dependence. A summary of radiation damage effects on sensitivity, and of sensitivity variation with temperature is also presented.

Characteristics of a selectively shielded p-Si detector in 60Co and 8 and 16 MV roentgen radiation.

A p-silicon semiconductor detector with a filter of wolfram powder mixed with epoxy, that entirely covered the back of the detector, was investigated and compared with ionization chambers and an unshielded semiconductor detector. Relative depth and profile distributions obtained as signals from the semiconductors were compared with corresponding dose distributions measured with the ionization chambers in 60Co, 8 and 16 MV roentgen radiation of different field sizes. It was found that relative signal distributions from the shielded semiconductor detector agreed, within 1 per cent of the maximum signal, to the depth dose curves and that the relative signal in profile distributions also agreed, within 1 mm or 1 per cent of the signal at the central axis, as compared with dose measurements with a cylindrical, thimble ionization chamber. The relative signal in the building-up region was compared with a plane parallel ionization chamber with a deviation corresponding to a position of less than 1 mm.

Characteristics of a p-Si detector in high energy electron fields.

Comparison of depth ionization distributions from a silicon semiconductor detector and depth dose curves from a plane parallel ionization chamber show that a semiconductor detector of p-type is well suited for relative electron dosimetry in the energy range of 6 to 20 MeV in Ep,0. Maximum deviations of the order of 1.5 per cent and of 1 mm were obtained down to a phantom depth of about 1 mm. The directional dependence of the detector was about 4 per cent.

Selective shielding of a p-Si detector for quality independence.

p-silicon semiconductor detectors were partially shielded with lead filters of various geometries with the purpose to minimize the quality dependence in assessments of depth dose distributions in large fields of cobalt radiation. It is shown that a lead shielding of the detector, with the front side excluded, reduced the signal error from about 3 to less than one per cent of the maximum signal as compared with an ionization chamber. Shielding was also done by encapsulating the detector in a mixture of wolfram powder and epoxy resin, with similar results. The signal received at the surface was not affected by this filter geometry. The directional dependence was investigated and was far from uniform for the wolfram-shielded detector. This will however not cause any practical problems when the detector is oriented towards the radiation source.

Radiation damage induced dose rate non-linearity in an n-type silicon detector.

Depth dose measurements in continuous cobalt radiation and pulsed roentgen rays were performed with nonirradiated and preirradiated detectors made of n-type silicon. A change in the relative signal at 15 cm depth of 5 to 10 per cent was found in pulsed roentgen ray fields when the detector was radiation damaged. Further experiments showed that the preirradiated detector had a superlinear dose response characteristic at high dose rates. A theoretic model was worked out and the non-linearity is explained by the properties of the recombination centers created during the preirradiation. The recombination centers are also responsible for the sensitivity drop after irradiation. At low dose rates in continuous radiation the non-linearity effect is not observed, which is in accordance with the theory. The theory in conjunction with our and other experimental results suggests that a p-silicon detector will remain linear also after heavy preirradiation.

Variance measurements with two semiconductor dose detectors.

Variance measurements were performed with pairs of semiconductor detectors of different ionization volumes in continuous and pulsed radiation fields. By a statistical analysis the relative variance of the detector was separated from that of the accelerator. The relative variance of the detector signal was shown to be correlated to the detector volume and to the dose per pulse. From the measurements the number of events per pulse within the detector volume was estimated and the number of pulses needed to have a reading with a given precision was estimated.

On the theory of an intravenous isotope method for cerebral blood flow measurements.

The theoretical basis of the intravenous isotope method has been examined with particular emphasis on the physiological significance of the area of the curves obtained and the meaning of the peak activity. It is concluded that the area is proportional to the volume of distribution of the indicator within the monitored volume, and inversely related to the cardiac output. The peak activity will be a measure of the fractional blood flow to the brain. It will be affected, however by the extension of the input bolus to the brain but also by the transit time within the brain itself. An unexpected low cardiac output, for instance, will lead to an over-estimation of the cerebral blood flow as derived from the peak activity even though the concomitant prolongation of the input bolus will reduce the error. The theory thus suggests that central circulatory variables will have an influence on the curves obtained and thus also will influence the validity of the method. The validity can be substantially improved by (1) the determination of the input bolus by, for instance, a gamma detector focused over the aortic arch, and (2) by the determination of the cardiac output; this can theoretically be carried out if a true intravascular indicator is used and the investigation is expanded to include the determination of the systemic blood volume.

The validity of an intravenous isotope method for cerebral blood flow measurements. An experimental study.

The intravenous isotope method has been examined. The method uses 99Tc as the indicator with recording of the radioactivity with two gamma-dectors placed bilaterally over the skull and oriented over the bifurcation of the middle cerebral artery. 65% of the total curve obtained was found to be generated by activity passing in the ipsilateral internal carotid artery, 25% in the contralateral internal carotid artery and less than 10% by activity passing in the two external carotid arteries. The counting rate efficiency for an intracerebral location was 19.4 +/- 1.8 times the count rate from the same activity placed at 1 m distance in air which in turn was 5 cps/muCi. The variability in efficiency from patient to patient was thus estimated at +/- 9%. The duration of the input bolus had a marked influence on the vurve, which will be of importance since it showed a rather large variability from patient. Using the peak activity as an index of the cerebral blood flow the intravenous method showed a surprisingly good correlation with the 133Xe wash-out method; the correlation coefficient was 0.93 and the variability at slightly subnormal values estimated at +/- 15%. It is concluded that the method can be used in the case of limited demands on the accuracy of the cerebral blood flow determinations.