Search

There is now considerable evidence to suggest that technical innovations, 3D image-based planning, template guidance, computerized dosimetry analysis and improved quality assurance practice have converged in synergy in modern prostate brachytherapy, which promise to lead to increased tumor control and decreased toxicity. A substantial part of the medical physicist's contribution to this multi-disciplinary modality has a direct impact on the factors that may singly or jointly determine the treatment outcome. It is therefore of paramount importance for the medical physics community to establish a uniform standard of practice for prostate brachytherapy physics, so that the therapeutic potential of the modality can be maximally and consistently realized in the wider healthcare community. A recent survey in the U.S. for prostate brachytherapy revealed alarming variance in the pattern of practice in physics and dosimetry, particularly in regard to dose calculation, seed assay and time/method of postimplant imaging. Because of the large number of start-up programs at this time, it is essential that the roles and responsibilities of the medical physicist be clearly defined, consistent with the pivotal nature of the clinical physics component in assuring the ultimate success of prostate brachytherapy. It was against this background that the Radiation Therapy Committee of the American Association of Physicists in Medicine formed Task Group No. 64, which was charged (1) to review the current techniques in prostate seed implant brachytherapy, (2) to summarize the present knowledge in treatment planning, dose specification and reporting, (3) to recommend practical guidelines for the clinical medical physicist, and (4) to identify issues for future investigation.

This work was undertaken to expedite implementation of the AAPM Task Group 43 recommendations, which call for significant modifications in the way dose is calculated for interstitial sources of 192Ir, 125I, and 103Pd as well as significant changes in the dose rate constant for 125I sources. The TG43 recommendations include a new formalism for dose calculation at points defined by the radial distance, r, from the source center and the angle, theta, that such a radius makes with the source axis. For each source type, values are tabulated for the radial dose function, the anisotropy function, and the anisotropy factor. The TG43 report includes fitting functions for the radial dose function in the form of polynomials, which are poorly behaved outside the range of fitted data. No functions are offered for the anisotropy function data or the anisotropy factor data, both of which could profit from some smoothing by such functions. We have found a double exponential fit to the radial dose function that not only approximates the data adequately but also appropriately approaches zero for very large distances. The anisotropy function is conveniently fit with a form of type 1 - f(r,theta)cos(theta)e(cr), which is exactly 1 at theta=90 degrees and approaches 1 for large r (for c

Purpose: Primary unresectable and locally advanced recurrent rectal cancer presents a significant clinical challenge. Local failure rates are high in both situations. Under such circumstances, there is a significant need to safely deliver tumoricidal doses of radiation in an attempt to improve local control. For this reason, we have incorporated a new approach utilizing high dose rate intraoperative radiation therapy (HDR-IORT). Methods and Materials: Between 11/92–12/96, a total of 112 patients were explored, of which 68 patients were treated with HDR-IORT, and 66 are evaluable. The majority of the 44 patients were excluded for unresectable disease or for distant metastases which eluded preoperative imaging. There were 22 patients with primary unresectable disease, and 46 patients who presented with recurrent disease. The histology was adenocarcinoma in 64 patients, and squamous cell carcinoma in four patients. In general, the patients with primary unresectable disease received preoperative chemotherapy with 5-fluorouracil (5-FU) and leucovorin, and external beam irradiation to 4500–5040 cGy, followed by surgical resection and HDR-IORT (1000–2000 cGy). In general , the patients with recurrent disease were treated with surgical resection and HDR-IORT (1000–2000 cGy) alone. All surgical procedures were done in a dedicated operating room in the brachytherapy suite, so that HDR-IORT could be delivered using the Harrison-Anderson-Mick (HAM) applicator. The median follow-up is 17.5 months (1–48 mo). Results: In primary cases, the actuarial 2-year local control is 81%. For patients with negative margins, the local control was 92% vs. 38% for those with positive margins ( p = 0.002). The 2-year actuarial disease-free survival was 69%; 77% for patients with negative margins vs. 38% for patients with positive margins ( p = 0.03). For patients with recurrent disease, the 2-year actuarial local control rate was 63%. For patients with negative margins, it was 82%, while it was 19% for those with positive margins ( p = 0.02). The disease-free survival was 47% (71% for negative margins and 0% for positive margins) ( p = 0.04). Prospective data gathering indicated that significant complications occurred in approximately 38% of patients and were multifactorial in nature, and manageable to complete recovery. Conclusion: HDR-IORT using our technique is versatile, safe, and effective. The local control rates for primary disease compare quite well with other published series, especially for patients with negative margins. For patients with recurrent disease, locoregional control and survival are especially encouraging in patients with negative resection margins. Further follow-up is needed to see whether these encouraging data will continue.

Purpose: We have generated three-dimensional (3D) lookup tables for dosimetric analysis and optimization of high-dose rate (HDR) gynecological treatments using the Henschke applicator. The new dosimetry data have been compared with two-dimensional (2D) data currently in use. The 3D dosimetry tables have been implemented in an existing cervix treatment-planning system and have been evaluated through analysis of clinical cases. Methods and Materials: A general Monte Carlo N-Particles (MCNP) transport code was used to compute absorbed dose distributions around the intrauterine tandem and tungsten-shielded ovoid separately. The dosimetry data are represented in the x–y coordinate system for the intrauterine tandem table. The 3D table for the ovoid contains a radial dose function and an anisotropy function, as formulated in the spherical coordinate system. Absorbed dose at a spatial point is calculated by applying bilinear interpolation for the anisotropy function and linear interpolation for the radial dose function. The geometry factor for a finite line source is used. 3D dose calculations and optimization were performed for 20 treatments of 10 patients. The absorbed dose to critical structures, bladder and rectum, was compared by applying both the 2D table currently in use and the new tables. Results: The new 2D table for the intrauterine tandem yields doses different by less than 10% from those with the current table. The 3D table for the shielded ovoids shows as large as a factor of 4 reduction of dose behind the shield compared with the present 2D table. This shielding effect leads to 21.6 ± 9.3% and 20.0 ± 6.6% dose reduction at rectum and bladder, respectively, for actual treatments. Conclusion: Our analysis indicates a need for patient-specific 3D dosimetry to permit more accurate dosimetric evaluation of HDR cervix treatments using shielded applicators. We have also shown that a Monte Carlo simulation code enabled us to derive the lookup tables necessary for 3D planning.

We have developed a system for source localization and real-time planning of interstitial volume implants for intraoperative radiation therapy (IORT) using high dose rate remote afterloading techniques. Source localization is realized by using an electromagnetic tracking device, which consists of a transmitter coil, a receiver coil, and a signal processing unit, to generate the coordinates and orientation of the receiver. A drawback of the device is its sensitivity to adjacent metallic objects. Localization accuracy was evaluated in an operating room environment, where the metallic objects closest to the receiver are surgical retractors (that, incidentally, preclude radiographic localization). For achievable separation distances, we found an rms error of 0.7 mm in determining the distance between points 2 cm apart, thereby demonstrating the feasibility of the method. The receiver is mounted on a plastic block from which projects a long stylet and the transmitter is located at about 50 cm from the receiver. The stylet is inserted sequentially into source catheters to obtain the location and orientation data that serve as input to treatment planning software. The planning program optimizes source dwell time to make calculated dose conform to the dose prescribed on an ellipsoidal surface to an extent consistent with a certain level of dose uniformity inside the target volume. A least squares method is used that involves minimizing the objective function by a matrix method (nonnegative least squares). We have demonstrated that dwell time optimization can be performed in a short time and that the approach is adequate for the IORT application.

Recommendations of the American Association of Physicists in Medicine (AAPM) for the practice of brachytherapy physics are presented. These guidelines were prepared by a task group of the AAPM Radiation Therapy Committee and have been reviewed and approved by the AAPM Science Council.

Background: Orbital rhabdomyosarcoma is the most common primary malignancy of the orbit in childhood. Tumor resection and exenteration were the preferred treatment modalities for rhabdomyosarcoma. In the past 20 years, however, combined local radiation and systemic chemotherapy have shown excellent survival results. Tumor recurrence after any of the aforementioned therapies is almost always fatal. We have developed a novel treatment for recurrent disease that has resulted in long-term survival for three patients. Methods: Three patients with recurrent orbital rhabdomyosarcoma were previously treated with primary radiation and chemotherapy. At the time of recurrence, exenteration and localized brachytherapy were performed. An individually molded poly(methylmethacrylate) (Lucite; E. I. du Pont de Nemours & Co., Wilmington, Del.) device loaded with radioactive iodine seeds delivered localized high-dose radiation, 6000 cGy over 6 days, to the orbit without irradiating the brain. Results: All patients are alive and free of disease with follow-up ranging from 4 years and 4 months to 8 years and 4 months. Conclusion: A novel technique of delivering localized radiation to the orbit of three children with recurrent orbital rhabdomyosarcoma appears curative.

Purpose: The purpose of this study was to assess the feasibility of 103 Pd bracytherapy in the management of primary unresectable carcinoma of the pancreas. Methods and Materials: Between August 1988 and January 1992, 11 patients with biopsy-proven primary unresectable adenocarcinoma of the pancreas were treated with 103 Pd brachytherapy during laparotomy. The median age was 66 (range 57–70). The most common presenting symptoms were weight loss (eight patients), pain (six patients), and nausea/vomiting (four patients). Less common symptoms were jaundice (two patients), early satiety (two patients), and ascites (one patient). All patient underwent laparotomy and surgical staging. Eight patients had T3N0M0 disease, two patients had T3N1M0 disease, and one patient had T3N1M1 disease. The surgical procedure performed was biliary bypass in six patients, biopsy only in four patients, and gastric bypass in one patient. The average tumor dimension was 4.0 cm. The median activity, matched peripheral dose (MPD) and implanted volumes were 95.3 mCi, 124.4 Gy, and 33 cm 3 , respectively. The median initial dose rate was 0.21 Gy per hour. Five patients received postoperative external beam radiation therapy (median 45 Gy) and seven patients received chemotherapy postoperatively. The median follow-up was 7 months (range 1–19). Results: The median survival for the entire group of patients was 6.9 months. Ten of 11 patients have died, with 1 patient presently alive and receiving chemotherapy for metastatic disease to the liver, but without local progression radiographically. Five of 11 patients (45%) were locally controlled, defined as either a complete response or freedom from progression at the site of the implant as evaluated by computed tomography scan. In the other six patients, the median time to local progression was 6.9 months. Five patients developed distant metastases (four liver, one subcutaneous nodule). Two patients failed in regional sites (one omentum, one paraaortic lymph node). Four of 11 patients (36%) developed acute postoperative complications that included one gastric outlet obstruction, one duodenal perforation, and two with sepsis. One of 11 patients (9%) developed a late complication of radiation enteritis 5 months after implantation. The median survival for patients experiencing complications was 1.7 months as compared to 8.4 months for the patients who did not develop a complication ( p = 0.10). Pain relief was obtained in five out of six (83%) for the patients presenting with pain for a median duration of 24 weeks. Local control did not appear to be related to the MPD, dose rate, implanted volume, treatment with external beam irradiation, or the use of chemotherapy. Patients were more likely to develop a complication if the MPD was greater than 115 Gy (four out of six patients) as compared to those whose MPD was less than 115 Gy (one out of five patients) p = 0.12). Conclusions: Because there was no improvement in median survival over conventional modalities, and the complication rate was high; we do not recommend 103 Pd brachytherapy as a component of the treatment of unresectable adenocarcinoma of the pancreas.

Background. Long-term survival and possible cure can be achieved in selected patients after resection of liver metastases from a colorectal primary tumor. In this study, the authors used radioactive implants as adjunctive therapy for patients who had incomplete resections of liver metastases. Methods. From 1979 to 1990, 17 implants were performed in 12 patients (median age, 61 years). Brachytherapy was performed for microscopically positive margins in 7 implants and for gross residual disease in 10 implants. Volume implants were performed for gross residual disease using 125I seeds, with a median matched peripheral dose of 160 Gy. Most of the implants for microscopic residual disease were planar implants with 125I suture seeds, with a median treatment dose of 150 Gy. Results. There were no serious postoperative complications. Overall median survival was 18 months, and 42% of patients survived 2 years or more. Extrahepatic metastases occurred in 83% of patients (10 of 12). Intrahepatic failure close to the implanted site occurred in 42% of patients (5 of 12), and intrahepatic failure anatomically remote from the implant site occurred in 25% (3 of 12). Conclusions. This approach is feasible and achieves reasonable local control of residual disease after resection of hepatic metastases from colorectal cancer. The anatomic pattern of failure supports efforts to intensify local therapy and to address the high rate of distant metastases. Cancer 1994; 73:1800–4.

Purpose: To quantitate the influence of dose inhomogeneity on brachytherapy efficacy. Methods and Materials: A computed tomography-based system of planning, implementation and evaluation was used to generate tumor-specific dose-volume histograms of eight permanent 125 I implants of prostate cancers. The radiobiological effect was then assessed, voxel by voxel, in terms of the biologically effective dose and the associated cell inactivation. The overall cell survival of the entire target volume was then computed. To evaluate the influence of inhomogeneity, the dose-volume histogram was modified in an iterative fashion, with the corresponding surviving fraction calculated after each step. Specifically, the volume in the highest dose bin was combined with that in the next bin to give a new frequency distribution from which cell survival was generated. Tumor control probability (TCP) was also used as an endpoint, using the same iterative procedure. Results: Doses 20–30% higher than D99 (the dose that covered 99% of the target volume) contributed to additional cell inactivation, but still higher doses did not further increase cell kill. With homogeneous irradiation at D 99 as a reference, we defined the inhomogeneity enhancement factor as the ratio of the biologic effective dose of the actual implant to that of the reference dose distribution. The calculated enhancement factors were inversely dependent on tumor potential doubling time (Tp), about 1.2 – 1.3 for a Tp of 30 days, and between 1.3 and 1.7 if Tp=10 days, with higher values for implants with low D99. Dose inhomogeneity enhanced TCP. For implants with high control probabilities, doses significantly higher (> 20%) than the D99 value did not further enhance the tumor control probabilities. In contrast, for implants with relatively low tumor control and D 99 values, the control probability continued to increase with doses significantly higher than D 99 , up to a dose of 2 × D 99 . The underlying reasons were the incorporation of patient "population averaging" in the calculation and the saturation of tumor control dose response at about 120 Gy. Conclusion: Dose heterogeneity in implants increased tumor cell kill and local control probability, although doses > 20% higher than the prescription dose is wasted. The increase the beneficial effect of dose inhomogeneity may be greatest when most needed.

Purpose: A large number of medical centers have recently instituted the use of High Dose-Rate Afterloading Brachytherapy (HDRAB). There is wide variation in treatment regimens, techniques, and dosimetry being used and there are no national standard protocols or guidelines for optimal therapy. Methods and Materials: The Clinical Research Committee (CRC) of the American Endocurietherapy Society (AES) met to formulate consensus guidelines for HDRAB in cervical, endometrial, and endobronchial tumors. Conclusion: Each center is encouraged to follow a consistent treatment policy in a controlled fashion with complete documentation of treatment parameters and outcome including efficacy and morbidity. Until further clinical data becomes available, the linear quadratic model can be used as a guideline to formulate a new HDR regimen exercising caution when changing from a Low Dose Rate (LDR) to a HDRAB regimen. The treatments should be fractionated as much as practical to minimize long term morbidity. As more clinical data becomes available, the guidelines will mature and be updated by the Clinical Research Committee of the AES.

Today's imaging and computer, technology has made significant advances in brachytherapy optimization and evaluation possible. Optimization may be of source strength or of source position (or both). An objective function of achieved versus desired doses is minimized, a global minimum being associated with an analytic solution and a local minimum with an iterative solution. Optimization of source strengths is shown useful in planning low-dose-rate-brachytherapy either of cervical cancer, using a remote afterloader with pneumatically loaded cesium 137 sources, or of recurrent endometrial cancer, using a vaginal cylinder and iridium 192 seeds in ribbons. Source position optimization, on the other hand, is exemplified by permanent percutaneous implants for prostate cancer or by temporary stereotactic implants, for brain tumors with iodine 125 sources of low and high strength, respectively. Evaluation of the true minimum peripheral dose generally requires, that three-dimensional images be available. Dose-volume histograms serve to describe both dose uniformity within the treatment volume and the extent to which normal tissues are being irradiated. With respect to the future, deconvolution and simulated annealing techniques are being explored for brachytherapy optimization, and we may expect greater emphasis on the biologically effective dose in brachytherapy evaluation.

The effect of eye plaque materials (gold backing and silastic seed-carrier insert) on the dose distribution around a single 125I seed has been measured, using cubic lithium fluoride thermoluminescent dosimeters (TLDs) 1 mm on an edge, in a solid water eye phantom embedded in a solid water head phantom. With an 125I seed (model 6711) positioned in the center slot of the silastic insert for a 20-mm plaque of the design used in the collaborative ocular melanoma study (COMS), dose was measured at 2-mm intervals along the plaque central axis (the seed's transverse axis) and at various off-axis points, both with and without the COMS gold backing placed over the insert. Monte Carlo calculations (MORSE code) were performed, as well, for these configurations and closely the same geometry but assuming a large natural water phantom. Additional Monte Carlo calculations treated the case, both for 20- and 12-mm gold plaques, where the silastic insert is replaced by natural water. Relative to previous measurements taken in homogeneous medium of the same material (without the eye plaque), the dose reduction found by both Monte Carlo and TLD methods was greater at points farther from the seed along the central axis and, for a given central-axis depth, at larger off-axis distances. Removal of the gold backing from the plaque did not make measurable difference in the dose reduction results (10% at 1 cm).

We compare the relative efficacy of high dose rate (HDR) and low dose rate (LDR) irradiation in gynecological implants, using formulations based on the linear-quadratic (alpha-beta) model. We consider both acute reaction and late effect as endpoints in evaluating equivalence for HDR and LDR treatments, at Point A (as defined in the text) and at other locations. We define a therapeutic factor gamma as an index for relative efficacy, with gamma1 indicating a gain in substituting HDR for LDR. It is found that gamma1 for most clinical situations, although a modest decrease (of about 15%) in dose to critical organs for HDR would raise gamma above unity.

Two-dimensional relative dose distributions have been measured around 125I brachytherapy seeds. The two seed models studied, models 6711 and 6712, were manufactured by the 3M Company. Silicon detectors immersed in water phantoms were used to measure the dose. A computerized data acquisition system that controlled the radial position of the diode and the angular rotation of the seed, as well as a manually controlled system were used to collect and store the data. Our results show that the two seed models have relative dose distributions which are quite similar; however, the absolute dose distributions are sufficiently different to warrant separate look-up tables for the two seed models. Additionally, our results are compared with dose distribution data previously obtained for the model 6711 seed.

The study evaluates the effect of the locally recurring tumor on the incidence of metastatic disease in early stage carcinoma of the prostate. The probability of distant metastases was studied in 679 patients with Stage B-C/NO carcinoma of the prostate treated at MSKCC between 1970 and 1985 (median follow-up of 97 months). Patients were staged with pelvic lymph node dissection and treated with retropubic 125I implantation. The actuarial distant metastases free survival (DMFS) for patients at risk at 15 years after initial therapy was 37%. Cox proportional hazard regression analysis of covariates affecting the metastatic outcome showed that local failure, used in the model as a time dependent variable, was the most significant covariate, although stage, grade, and implant volume were also found to be independent variables. The relative risk of metastatic spread subsequent to local failure was 4-fold increased compared to the risk without evidence of local relapse. The 15-year actuarial DMFS in 351 patients with local control was 77% compared to 24% in 328 patients who developed local relapses (p < 0.00001). The relation of distant spread to the local outcome was observed regardless of stage, grade, or implant dose. Even stage BI/NO-Grade I patient with local control showed a 15-year actuarial DMFS of 82%, compared to 22% in patients with local relapse; p < 0.00001). The median local relapse-free survival (LRFS) in the 268 patients with local recurrences who did not receive hormonal therapy before distant metastases were detected was 51 months, compared to a median of 71 months for DMFS in the same patients (p < 0.001), consistent with the possibility that distant dissemination may develop secondary to local failure. Furthermore, distant metastases in patients with local control, apparently already existing as micrometastases before treatment, were detected earlier (median DMFS of 37 months) than in patients with local relapse (median DMFS of 54 months; p = 0.009). These data suggest that the existence and re-growth of local residual disease in localized prostatic carcinoma promotes an enhanced spread of metastatic disease, and that early and complete eradication of the primary tumor is required if a long term cure is to be achieved, although the clinical expression of secondary metastases may not become apparent for 6.5 years or more in one-half of the patients.

Dose measurements using LiF thermoluminescent dosimeters (TLD) have been performed for single 103Pd seeds (model 200) at the center of a solid water phantom. TLD cubes 1 mm on an edge were used for measurements from 1 mm to 1 cm at 1-mm intervals. The cubes were centered along transverse and longitudinal axes and along radial lines from seed center at 10 degrees increments. TLD chips of dimension 3.1 X 3.1 X 0.89 mm were used at distances of 2, 2.5, 3, and 4 cm at 15 degrees angular intervals. Data are presented as the product of distance squared and dose rate per unit source strength, plotted versus distance and angle. At 1 cm from seed center along the transverse axis this product was found to be 0.88 cGy cm2 mCi-1h-1. A dose-rate table in polar coordinates has been formulated for use with multiseed dose distribution calculations. Comparison with data of Meigooni et al. [Endocuriether./Hyperthermia Oncol. 6, 107-117 (1990)] shows general agreement for distances of 2 cm or greater. A comparison of our transverse axis data with Russell's calculated values (Theragenics Internal Report, 4 November 1984) for an ideal point source of 103Pd shows very good agreement except at distances less than 0.5 cm, where differences are attributable to the extended source effect in the actual seed.

Dose rates in water have been determined for the two types of {sup 125}I seed currently used in brachytherapy. The need for such determinations became evident when water/air ratios measured with a silicon diode were found to be lower than expected. Extensive measurements using lithium fluoride thermoluminescent dosimeters (TLD's) have been performed in a solid water phantom, at distances from 0.1 to 10 cm from the seed center and at angular increments of 10{degree}, 15{degree}, or 30{degree} within a plane through the seed axis. Dose calibration of the TLD's was accomplished by irradiation in air with {sup 125}I seeds of the same type and of strengths traceable to a calibration at the National Institute of Standards and Technology (NIST). Relative calibration of TLD's was monitored by irradiation, in an oven-type x-ray machine, of control dosimeters simultaneously and all dosimeters intercurrently with the {sup 125}I irradiations. Values obtained for the dose rate constant, i.e., dose rate per unit air-kerma strength at 1 cm on the transverse axis, were 0.853 and 0.932 cGy h{sup {minus}1} U{sup {minus}1} (1.08 and 1.18 cGy h{sup {minus}1} mCi{sup {minus}1}) for the 6711 and 6702 seeds, respectively. Measured data were supplemented with Monte Carlo-calculated relative dose ratemore » data generated using the MORSE code. These calculations used 100 energy groups from 10 to 35.4 keV and involved energy collection bins ranging from 0.025 to 1.2 cm on an edge. Normalized at 1 cm, transverse axis calculated data are not significantly different from measured data (ours or cited literature) at distances either {lt}2.5 or {gt}8 cm. Normalized at different distances along the transverse axis, our off-axis calculated and measured distributions agree closely at all angles but differ from literature measured distributions at small ({le}1 cm) distances and, for small angles, increasingly at larger distances ({ge}5 cm).« less

The current trend toward the use of 125I seed rather than 60Co applicators in the treatment of ocular tumors stems largely from the extent to which photons from the two radionuclides are attenuated differently by plaque material on the one hand and by intraocular tissue on the other. The much lower energies of 125I photons from seeds in a rimmed gold plaque not only permit near-total elimination of radiation dose in orbital tissue adjacent to the plaque but also allow significant dose reduction to critical structures within the eye, relative to either 60Co plaque or proton beam treatment (Fairchild 1984). Although other photonemitting radionuclides have been used in ophthalmic applicators (Luxton et al. 1988a), we will focus attention here on the physical characteristics of 60Co and 125I applicators because they have been the most widely used and because they illustrate well the pertinent differences between “high energy” and “low energy” photons for this type of brachytherapy.

Purpose: Retroperitoneal sarcomas represent a formidable challenge to the treating oncologist due to their location, large size, and poor prognosis. The purpose of this study was to determine if the addition of high-dose-rate intraoperative radiation therapy (HDR-IORT) to surgery and external beam radiotherapy (EBRT) would improve the outcome in these patients. Methods and Materials: Thirty-two patients with retroperitoneal soft tissue sarcoma were prospectively treated according to a protocol that included maximal tumor resection, HDR-IORT, and postoperative EBRT when feasible. Twelve patients presented with primary and 20 with locally recurrent disease. The tumors were high-grade in 20 patients and low-grade in 12 patients. Complete gross resection was achieved in 30 patients. HDR-IORT was given to a dose of 12–15 Gy. Additional EBRT was given to 78% of patients to a dose of 45–50.4 Gy. The two patients with gross residual disease received an additional I-125 permanent implant to a median peripheral dose of 140–160 Gy. The median follow-up was 33 months (range 1–77 mo). Results: The 5-year actuarial local control rate for the whole group was 62%. For patients with primary disease, the local control rate was 74% compared to 54% in patients with recurrent disease ( p = 0.4). The overall 5-year distant metastasis-free survival rate was 82%. In patients with high-grade tumors the rate was 70% vs. 100% in those with low-grade tumors. This difference was statistically significant, p = 0.05. The 5-year disease-free and overall survival rates were 55% and 45%, respectively. The most common type of post-treatment complication was gastrointestinal obstruction (18%) followed by fistula formation (9%), peripheral neuropathy (6%), hydronephrosis (3%), and wound complication (3%). Conclusions: We are encouraged by the favorable local control rate and the acceptable morbidity with this new technique applied to a challenging patient population.

It has been clearly documented that complete resection plus radiation therapy can locally control the overwhelming majority of soft-tissue sarcomas of the extremity and superficial trunk (1-18). In this regard, the radiation-therapy approach can be either by external beam, brachytherapy, or a combination of both (1-18). The value of adjuvant brachytherapy over surgery alone for high-grade lesions has been demonstrated in a prospective randomized trial at Memorial Sloan Kettering Cancer Center (MSKCC) (19,20). Suit (2,3,9,12) has demonstrated excellent local control for low-grade lesions using surgery with external-beam irradiation (EBRT). Other investigators have successfully employed a combination of adjuvant EBRT and brachytherapy (8,10,11,17,18). Local control rates between 80 and 100% have been reported when both surgery and radiotherapy are properly integrated.

Intraoperative irradiation by high-dose-rate remote-afterloading brachytherapy (HDRIORT) requires, in addition to capabilities in surgery and radiation oncology, specialized physical equipment and facilities, appropriate computer software for treatment planning and, as an integral part of the interdisciplinary team, physicists trained in dose planning, dose delivery, and quality-assurance procedures. The rationale for HDR-IORT is related, in large part, to the physical advantages it affords of confining the therapeutic dose to a highly localized target and sparing normal structures either by moving them away or shielding them. These advantages may substantially offset the radiobiological disadvantage associated with a single-fraction treatment (see Chapter 2). Although only a few institutions are currently performing HDR-IORT, sufficient experience has been acquired to establish its feasibility with respect to such physical factors as available source strength, suitable applicators, and acceptable treatment durations.

We have performed extensive computational and experimental dosimetry of the Henschke applicator with respect to high dose-rate 192 Ir brachytherapy using a GAMMAMED remote afterloader. Our goal was to generate clinically useful two- and three-dimensional look-up tables. Dose measurements of the Henschke applicator involved using TLD chips placed in a polystyrene phantom. Monte Carlo simulations were performed using the MCNP code. The computational models included the detailed geometry of 192 Ir source, tandem tube, and shielded ovoid. The measured dose rates were corrected for the dependence of TLD sensitivity on the distance of measurement points from the source. Transit dose delivered during source extension to and retraction from a given dwell position was estimated by Monte Carlo simulations, and a correction was applied to the experimental values. For the applicator tandem, the ratio of dose rates obtained by MCNP to those measured by TLD chips ranges from 0.92 to 1.10 with an average of 0.98 and a standard deviation of 0.02. The measured and calculated dose rates at 1 cm on the transverse axis are 1.10 cGy U −1 h −1 . For the shielded ovoid, the ratio ranges from 0.88 to 1.16 with an average of 1.00 and a standard deviation of 0.07. Causes of the discrepancy between the Monte Carlo and TLD results were identified. We found that the combined uncertainty of measured dose rates due to these causes is 5.6% for the applicator tandem and 8.4% for the shielded ovoid. Therefore, the results of the Monte Carlo simulation are considered to have been validated by the measurements within the uncertainty involved in the calculation and measurements.

The radiobiological efficacy of high dose rate (HDR) intracavitary implant relative to that of low dose rate (LDR) implants has been represented quantitatively by the therapeutic factor lamba defined as [EDRtum/ERDorgan]HDR/[ERDtum/ERDorgan]LDR where ERD is the extrapolated response dose. The ERDs for the tumor and a critical organ are calculated using the linear-quadratic model and depend on the values of alpha/beta ratio, the tumor and normal tissue repair time constants (mu 1 and mu 1), number (N) of HDR fractions, the dose per fraction (d), and the fractional tumor and organ dose (ft and fl) relative to the dose at the reference point. We have studied the variation of gamma with d, mu t, mu l, ft, and ft and have derived specific conditions for which lamba can have a value equal to or greater than unity leading to a therapeutic gain in HDR. It is found that lamba is less than unity for commonly used parameter values, specifically when mu t is assumed to be equal to mu l. However, if mu t is greater than mu l, lamba can have a value greater than unity for many possible values of ft and fl.

PURPOSE To correlate the pattern of failure and subsequent survival with the technical accuracy of stereotactic brain tumor implantation. METHODS AND MATERIALS The patterns of failure of 47 consecutive patients with primary or recurrent gliomas after stereotactic implantation delivering 60 Gy via removable high activity 125I sources were reviewed. When the tumor was covered at all levels by the chosen isodose distribution, the implant was considered to be "ideal." If the coverage was not complete, a numerical description of the volume of tumor outside the isodose was assigned. Criteria for "adequate" and "inadequate" implants were defined. Standard radiographic criteria, with pathologic confirmation in 26 cases, were used to categorize the patterns of failure into the following components: central, peripheral, distant (within the brain parenchyma), leptomeningeal, and spinal. A peripheral failure was scored as being in the "direction of error" when the prescribed isodose did not cover the tumor volume and the subsequent tumor progression was in this region. Survival was calculated from the date of implantation. RESULTS Of 47 cases examined, 72% had an element of central and/or peripheral failure and 23% had a component of distant or meningeal failure. Among the patients with "adequate" or "inadequate" ("nonideal") implants who had a component of peripheral failure, only 19% were in the "direction of error." All patients with technically "inadequate" implants progressed in both the central and peripheral region. Among the groups who had "ideal," "adequate," and "inadequate" implants; 37%, 70%, and 75%, respectively, underwent reoperation [p = not significant (NS)]. Patients who underwent reoperation had a longer median survival than those who did not; 521 days vs. 298 days, respectively (p = 0.035). For patients with "nonideal" implants, a median survival of 470 days was found for patients undergoing reoperation vs. 184 days for those who did not (p = 0.016). CONCLUSIONS (a) Patients with "inadequate" implants failed in both the central and peripheral region in all cases. This pattern, while less common in those with "ideal" or "adequate" implants, occurred in the majority of cases. (b) The technical excellence of the implant had no impact on survival. (c) Patients with "nonideal" implants were more likely to have reoperation than those with "ideal" implants, and this intervention was associated with a significant survival advantage.

Purpose : To determine the relative biological effectiveness (RBE) of I-125 and Pd-103 relative to Co-60. Methods and Materials : A cell line REC :ras, derived from rat embryo cells, was used. Cells in exponential or plateau phase were irradiated at dose rates of about 0.07 Gy/h and 0.14 Gy/h. To circumvent the interface effect, cells were grown and irradiated on membranes made of cellulose acetate, which has an effective Z of 7.5. I-125 and Pd-103 seeds were placed in a custom designed template that yielded a homogeneous dose distribution in the plane of the cell culture. The dose rates of irradiation were measured by calibrated thermoluminescence dosimetry (TLD) chips. Results and Conclusions : Our measurements yielded an RBE of about 1.4 for I-125 at dose rates of about 0.07 Gy/h, and an RBE of about 1.9 for Pd-103 at dose rates of about 0.07 Gy/h and 0.14 Gy/h. The RBE of I-125 is similar to those measured by other investigators, the RBE for Pd-103 is being reported for the first time.

The relatively precise placement of brachytherapy sources afforded by stereotactic frames for brain implants is not generally achievable for other sites, which lack the fixed geometry of the cranium and its contents. An exception is a source-containing rigid mold that delivers brachytherapy when inserted securely in a surgical defect. A technique has been developed in which an acrylic mold of the region to be treated is suspended in a demountable aluminum box, which is then filled to a suitable level with dental stone to form a casting that supports the mold and that can be removed intact. First, the box is aligned on a mill table and a ball mill is used to drill three parallel holes in the acrylic mold, with precisely known locations and depths and as widely separated as possible. The spherical air cavities that result from plugging these holes with ball-milled acrylic rods become reference markers in subsequent computed tomography (CT) scans. After optimum CT-coordinate locations have been planned for 125I seeds in catheters, they are transformed to mill coordinates using a matrix developed from the known marker coordinates in the two systems. Catheter holes are then drilled with the mold in the reassembled casting and box. The method has been used to treat both recurrent maxillary cancer and recurrent orbital rhabdomyosarcoma.

The relative dose distribution around the 103Pd model 200 implant seed was measured with a computerized data acquisition system employing a p-n junction silicon diode immersed in a water phantom. Data are acquired in polar coordinates by computer control of (1) the diode distance from the seed center and (2) the rotation angle of seed about a transverse axis. Transverse axis data are compared with thermoluminescent dosimeter (TLD) measurements and a Monte Carlo calculation by others.

103Pd is being substituted for 125I in permanent implants for which it is desired to deliver a higher initial dose rate while maintaining readily achieved radiation protection. We have constructed a nomograph to assist in determining both the total seed strength required and the appropriate needle spacing for 103Pd implants. We have calculated the "matched peripheral dose" (MPD), that is, the dose for which the isodose contour volume is equal to the target volume, for 64 125I and 13 103Pd actual implants as if 103Pd had been used for all of them, employing a computer lookup table based on single-seed dose distribution measurements in solid water. The calculated data were used to obtain a least-squares fit to a linear relationship between the logarithm of the total seed strength for a given MPD and the logarithm of the average dimension, da (cm). We found that, for a nominal MPD of 11,500 cGy, total seed strength (in mCi) is given by 3.2 da2.56. A 103Pd nomograph has been constructed on the basis of this power function relationship. Our nomographic guide for planning 103Pd implants calls for total seed strength to increase significantly faster as a function of target volume average dimension than is the case for 125I. This nomograph will facilitate the application of 103Pd seeds in permanent implants.

Our approach to planning stereotactic 125I brachytherapy of brain tumors has involved least-squares optimization of individual seed positions within the target contour, followed by repeated combining of seeds from nearest-neighbor catheters in order to achieve an acceptably low number of catheters and an acceptable-separation of entry points. In one option, the catheters diverge from an extra-cranial point that can be close to the skull if all catheters are to be placed through a small craniectomy to treat a larger-diameter target. In another option, catheters converge toward a point beyond the target, to facilitate perpendicularity at the skull surface if a separate opening is to be drilled for each catheter. In either case, the fact that seed orientations are known, permits including anisotropy in dose calculations. Trial seed locations are constrained to a target region defined on a 1-mm mesh, both in the initial optimization of single-seed catheters and in subsequent combinations followed by tune-up optimizations. In the optimization process, sum-of-squares contributions are weighted more heavily when the dose rate is lower than the target dose rate; the weighting imbalance falls short of keeping all target points above the target dose rate and requires targeting on a dose rate about 25% higher than the desired minimum dose rate.

The tumor control probabilities (TCPs) for 11 transperineal permanent implants of the prostate with 125I seeds have been computed, based on target-specific volume-dose histogram data. Prostate dose-response of this modality is converted from that of external beam therapy based on biologically effective dose derived from the α β model. The calculated TCPs for these transperineal prostate cases are compared with the results of 679 previous retropubic implants, providing interesting insight about these two techniques. Further refinement and validation of this model await long-term followup results of sufficient number of patients.

A computerized tomography-based 3-dimensional evaluation scheme, using semi-automatic seed localization from transverse computerized tomography-images, has been developed for permanent implants. The treatment isodose contour is generated at each scan plane and overlaid on the corresponding target contour for coverage visualization. Volume and surface dose histograms are generated for the target and neighboring normal structures. Dose inhomogeneity within the target is also estimated from the full-width at half maximum of the differential volume dose histogram. The matched peripheral dose is calculated from the ellipsoidal volume approximation for the target. The present evaluation method has been applied here to 10 actual permanent 1–125 prostatic implantations. Coverages by 150 Gy and 100 Gy levels are found to range from 78–96% and 91–99% of the target volume, respectively. The average matched peripheral dose is found to be about two times the average peripheral dose (the dose encompassing 99% target) and about three times the average minimum peripheral dose (the dose encompassing 100% target). The dose inhomogeneities within the target volume, in these 10 cases, range from 220–420 Gy.

A computerized planning procedure has been developed for CT-guided transperineal prostate implants. The segment for custom planning of perineal needle orientations allows placement of I-125 seeds in the entire prostate gland while avoiding the pubic bones. Least-squares optimization (LSO) is used to obtain the seed-loading pattern along the needles. The optimized seed distribution produces a better fit between treatment and target volumes than that obtained from our previous manual technique. Also, the present semi-automatic technique reduces planning time by about a factor of 10 compared to that of the manual approach.

From 1968 to 1987, 123 consecutive patients with nonmetastatic choroidal melanoma were treated with cobalt-60 plaques. One hundred sixteen patients were followed up for a mean of 3.8 years. Twenty patients had local failure, and 14 patients had distant failure. Complications included 32 cataracts, and seven enucleations were required. Local recurrence did not correlate with tumor height, tumor volume, dose, or dose rate. Increased volume (P = .004) and height (P = .01) correlated with increased rates of distant metastases. Dose adjusted for volume did not correlate with the rate of metastases.

From 1973 to 1980, 1,427 cancer patients were treated by the Brachytherapy Service, Radiation Oncology Department of Memorial Sloan-Kettering Cancer Center (MSKCC) with I-125 permanent interstitial implantation. Concern has been expressed with regard to the possible toxic effects of the low average photon energy (28 kev) of I-125 on bone, secondary to its increased absorption in these tissues because of the photoelectric effect. In an attempt to address this concern with respect to cortical bone, we reviewed pertinent data on those patients whose site of implant was at close proximity to osseous tissue in the pelvic, chest wall and head and neck regions. The review included 74 sites in 58 patients having no prior history or clinical evidence of any pathologic or metabolic bony disorder. Long term (greater than 1 year) follow-up was available in over 50% of the cases, with 10 patients followed more than 4 years. Local control rate for the 74 implanted sites was 93%. The implanted activity ranged from 1.6 to 48.8 mCi and the implanted volume from 0.5 to 152 cm3. At this follow-up range, there appears to be no excessive osseous toxicity either clinically or radiologically with I-125 permanent implantation.

The most useful formulation of the volume--dose histogram for brachytherapy is one which suppresses inverse square law effects while preserving proportionality between volume and the area under the curve. These objectives are met by plotting the distribution of volume per unit -3/2 power of dose rate versus the -3/2 power of dose rate; such a distribution is constant for a single point source. Adding sources results in the formation of a peak, which provides a graphical indication of dose rate uniformity. Evaluation of other peak parameters such as width, position, and contained volume, in relation to treatment dose rate, permits a quantitative and clinically relevant volume--dose assessment of interstitial implants.

One hundred twenty-four patients were treated with advanced recurrent head and neck cancer for palliation with radioactive permanent Iodine-125 (125I) implants. Complete regression occurred in 71% of the 118 lesions for which evaluation was possible and greater than 50% regression occurred in 18%; no meaningful regression occurred in 11%. Local recurrence of cancer was subsequently seen in 21% of the lesions which had regressed completely, in 55% of those which had regressed incompletely, and in 100% of those which had not regressed. The incidence of serious complications was 5.5%. Overall, in 64% of the instances the implanted lesions remained controlled until the patient's death, usually due to progression of cancer elsewhere in the body. It is concluded that permanent 125I implants offer useful palliation to the patient with recurrent head and neck cancer with a minimum of toxicity and inconvenience. Because of their low toxicity even after prior full-course external radiation therapy, the authors are currently investigating their use as planned adjunct to external radiation therapy and chemotherapy in the initial definitive management of patients with locally advanced head and neck cancer.

While cervical applicators with shielded ovoids are used widely in brachytherapy, we know of no system for calculating dose distributions for them. For shielded sources, because of a lack of symmetry and because of a rapid variation of dose as a function of position relative to the source, extensive measured data in three dimensions are required. In the method we have developed, the dose at a given point from a source in a shielded ovoid is calculated by multiplying the dose from an unshielded source by the “effective attenuation factor” of the shields. The latter quantity is obtained by linear-interpolation in a three-dimensional table generated from measurements described in an earlier paper. The unshielded-source dose is calculated as the product of source strength, time of implant, distance-dependent geometry factor and a tabulated quantity called the “relative dose rate factor”. Relative dose rate factor is obtained by dividing measured dose rate by the product of geometry factor and source strength. Division by the geometry factor reduces the amount of data required with respect to accuracy in linear-interpolation. Input localization data must include not only the position of the end points defining the source but also a third reference point to define the orientation of the shields.