Hyperfractionated radiation therapy and 5-fluorouracil, cisplatin, and mitomycin-C (+/- granulocyte-colony stimulating factor) in the treatment of patients with locally advanced head and neck carcinoma.

BACKGROUND: The authors had previously reported preliminary results of a treatment regimen of concurrent hyperfractionated radiation therapy and chemotherapy in patients with locally advanced head and neck carcinoma that demonstrated both feasibility and high local control. In an attempt to reduce acute mucosal and hematologic toxicity, granulocyte-colony stimulating factor (G-CSF) was added during the second phase of this study. METHODS: Seventy patients (53 with Stage IV and 17 with Stage III disease) were entered between May 1988 and June 1995 into a Phase I/II trial of concurrent radiation therapy (74.4 gray (Gy) total dose; 1.20 Gy twice daily), 5-fluorouracil (1000 mg/m2/24 hours for 72 hours), and cisplatin (50 mg/m2) for 3 cycles with the addition of mitomycin C (8 mg/m2) in Cycle 2. G-CSF was added after the initial entry of 34 patients. RESULTS: At a median follow-up of 41 months (range, 12-80 months), 44 patients were alive with a projected median overall survival of 54 months. Grade 3/4 mucositis, observed in 65% of patients, was equally prevalent and prolonged in both G-CSF-treated (+) and G-CSF-naive (-) patients. Grade 3/4 leukopenia was present in 45% and 36% of G-CSF- and G-CSF+ patients, respectively. The 3-year locoregional control and cause specific survival rates were 68% and 75%, respectively. CONCLUSIONS: This regimen was feasible and effective but caused severe mucositis. No benefit was derived from the addition of G-CSF. This regimen deserves further modification to reduce acute mucositis toxicity yet maintain the high locoregional control rate.

Management of stage I-B, II-A, and II-B carcinoma of the cervix with high-dose-rate brachytherapy: initial results of an institutional clinical trial.

In 1989, the University of Miami began a program incorporating high-dose-rate (HDR) brachytherapy into the definitive treatment of patients with invasive carcinoma of the cervix. Patients received an average total dose to point A of 5,511 cGy (range 4,280-6,360 cGy) in an average of 57 days (range 39-84 days). An analysis of the first 24 cases found 11 FIGO Stage I-B, four Stage II-A, and nine Stage II-B tumors. At the end of all radiation therapy, 19/24 patients' tumors (79.2%) had undergone a clinical complete response (CR). With median follow-up of 26 months (range 14-63 months), three have relapsed locally, two regionally, and six in extrapelvic sites. Almost two-thirds of all failures occurred in patients with tumors >4 cm, who also took more than 8 weeks to complete their treatment. Overall 2-year actuarial survival for the entire study group is approximately 74%. A univariate analysis determined that clinical stage (P = 0.02), overall treatment time (P = 0.03), tumor size (P = 0.05), and response at the end of therapy (P = 0.005) were significant prognostic factors. Multivariate analysis showed that tumor response to therapy was the most important prognosticator of outcome (P = 0.001). Besides five cases of apical vaginal stenosis, there have been no reported chronic complications in this cohort of patients. A prospectively randomized trial is recommended to compare the efficacy of HDR vs. low-dose-rate brachytherapy in cervical carcinoma.

Conservative surgery and radiation therapy for intraductal carcinoma of the breast.

Nineteen women with intraductal carcinoma of the breast were treated with conservative surgery and radiotherapy from 1982 to 1990. All underwent excisional biopsy or wide local excision of the primary tumor. Definitive irradiation consisted of 4500 cGy in 180 cGy fractions given through tangential fields followed by a breast boost to the primary site to a total dose of 5900-6500 cGy. No patient received regional node irradiation. Median follow-up was 38 months. The five year actuarial rate of local failure was 9%. One patient failed with an infiltrating ductal carcinoma in the treated breast 31 months after initial treatment. Salvage mastectomy was performed. She remains without evidence of disease 43 months after initial treatment. Metastatic breast carcinoma has not developed in any of the patients. Cosmetic result was good to excellent in all patients. With short-term follow-up, conservative surgery and radiotherapy appear to be an acceptable alternative to mastectomy in carefully selected patients with ductal carcinoma in situ. As retrospective and randomized trials mature, the natural history of these lesions treated with conservative surgery and irradiation will be further defined.

Hyperfractionated radiation therapy and concurrent 5-fluorouracil, cisplatin and mitomycin-C in head and neck carcinoma. A pilot study.

Seventeen patients were entered into a Phase I/II trial of concurrent hyperfractionated radiation therapy (7,440 cGy total dose; 120 cGy b.i.d.) combined with constant infusion of 5-fluorouracil (5-FU) (1,000 mg/m2/24 hours for 72 hours) and cisplatin (DDP) (50 mg/m2) for a total of three cycles. Thirteen patients had Stage IV disease; three, Stage III disease; and one, Stage II hypopharyngeal disease. Thirteen of 17 patients had positive cervical lymph nodes, and the mean size of the largest lymph node was 5.5 x 5.1 cm. The patients were not treated with planned adjunctive surgery except for one patient who had a radical neck dissection for massive, rapidly growing cervical adenopathy, which recurred promptly within 1 month before the initiation of protocol therapy. After the initial six patients were entered, mitomycin-C (Mito 8 mg/m2) was added during the second cycle. All the patients completed the planned course of radiotherapy with a median dose of 7,440 cGy and a mean dose of 7,248 cGy except for two patients who died--one from toxicity and the other, suicide. The predominant toxicity was mucositis, which was grade 3/4 in 11 of 15 patients, resulting in an average interruption of radiation therapy of 12 days. Weight loss was significant and was on the average 12% of baseline weight. Hematological toxicity was mild in the 5-FU/DDP group (only one grade 3 toxicity of six) and severe in the 5-FU/DDP/Mito-treated patients (five of eight patients having grade 3/4 toxicity including one leukopenic pneumonitis death). Additional toxicity included one parapharyngeal cellulitis, which responded to antibiotics. Noncompliance with the complex regimen was only seen in three patients. One patient refused b.i.d. radiation therapy, and one patient refused further chemotherapy after the first cycle. Additionally, one patient who had a severe ethanol withdrawal reaction during the first cycle of 5-FU/DDP did not receive further chemotherapy. The complete response rate of both primary site and neck by the protocol regimen alone was 71%. However, two patients, one from each group, did undergo salvage neck dissection, and the locoregional control is currently 73%, with a mean follow-up time of 18.4 months. The feasibility of combining hyperfractionated radiation therapy with aggressive concurrent chemotherapy was demonstrated. The response and local control rate justifies the added toxicity of concurrent chemotherapy and radiation therapy.

Radiosurgery target point alignment errors detected with portal film verification.

Stereotactic radiosurgery with a linear accelerator requires an accurate match of the therapeutic radiation distribution to the localized target volume. Techniques for localization of the target volume using CT scans and/or angiograms have been described. Alignment of the therapeutic radiation distribution to the intended point in stereotactic space is usually accomplished using precision mechanical scales which attach to the head ring. The present work describes a technique used to verify that the stereotactic coordinates of the center of the intended radiation distribution are in agreement with the localized target point coordinates. This technique uses anterior/posterior and lateral accelerator portal verification films to localize the stereotactic coordinates of the center of the radiation distribution with the patient in the treatment position. The results of 26 cases have been analyzed. Alignment errors of the therapeutic radiation distribution in excess of 1 mm have been found using the portal film verification procedure.

Dose determination in high dose-rate brachytherapy.

Although high dose-rate brachytherapy with a single, rapidly moving radiation source is becoming a common treatment modality, a suitable formalism for determination of the dose delivered by a moving radiation source has not yet been developed. At present, brachytherapy software simulates high dose-rate treatments using only a series of stationary sources, and consequently fails to account for the dose component delivered while the source is in motion. We now describe a practical model for determination of the true, total dose administered. The algorithm calculates both the dose delivered while the source is in motion within and outside of the implanted volume (dynamic component), and the dose delivered while the source is stationary at a series of fixed dwell points. It is shown that the dynamic dose element cannot be ignored because it always increases the dose at the prescription points and, in addition, distorts the dose distribution within and outside of the irradiated volume. Failure to account for the dynamic dose component results in dosimetric errors that range from significant (> 10%) to negligible (< 1%), depending on the prescribed dose, source activity, and source speed as defined by the implant geometry.

Stereotactic radiosurgery: dose-volume analysis of linear accelerator techniques.

Stereotactic radiosurgery of the brain may be accomplished with a linear accelerator by performing several noncoplanar arcs of a highly collimated beam focused at a point. The shape of the radiation distribution produced by this technique is affected by the beam energy, field size, and the number and size of the arcs. The influence of these parameters on the resulting radiation distributions was analyzed by computing dose volume histograms for a typical brain. Dose volume functions were computed for: (a) the energy range of 4-24 MV x rays; (b) target sizes of 1-4 cm; and (c) 1-11 arcs and dynamic rotation. The dose volume histograms were found to be dependent on the number of arcs for target sizes of 1-4 cm. However, these differences were minimal for techniques with 4 arcs or more. The influence of beam energy on the dose volume histogram was also found to be minimal.

Computer controlled stereotaxic radiotherapy system.

A computer-controlled stereotaxic radiotherapy system based on a low-frequency magnetic field technology integrated with a single fixation point stereotaxic guide has been designed and instituted. The magnetic field, generated in space by a special field source located in the accelerator gantry, is digitized in real time by a field sensor that is six degree-of-freedom measurement device. As this sensor is an integral part of the patient stereotaxic halo, the patient position (x, y, z) and orientation (azimuth, elevation, roll) within the accelerator frame of reference are always known. Six parameters--three coordinates and three Euler space angles--are continuously transmitted to a computer where they are analyzed and compared with the stereotaxic parameters of the target point. Hence, the system facilitates rapid and accurate patient set-up for stereotaxic treatment as well as monitoring of patient during the subsequent irradiation session. The stereotaxic system has been developed to promote the integration of diagnostic and therapeutic procedures, with the specific aim of integrating CT and/or MR aided tumor localization and long term (4- to 7-week) fractionated radiotherapy of small intracranial and ocular lesions.

Optimization of high dose-rate cervix brachytherapy; Part I: Dose distribution.

Computer controlled high dose-rate (HDR) brachytherapy afterloading machines are equipped with a single, miniaturized, high activity Ir-192 source that can be rapidly moved in fine increments among several channels. Consequently, by appropriate programming of source dwell positions and times, the dose distribution can be optimized as desired. We have explored the optimization potential of this new technology for two applications: (a) cervix brachytherapy, and (b) transvaginal irradiation. Cervix brachytherapy with a gynecologic ring applicator was simulated by 48 sources of relative activities ranging from 0.17 to 1.00 that were equally distributed between the tandem and the ring. The results confirmed that the optimized distribution of physical doses are superior to those achievable with standard brachytherapy sources and applicators. For example, with five-point optimization, the relative dose-rate in the rectum was only 47% of that in point A; for standard application the dose rate was 47% higher. For transvaginal application 27 sources of relative activities between 0.07-0.79 were placed in the ring and a single source of unit strength in the tandem. Using dose distribution homogeneity as an optimization criterion, the results (+/- 2.5%) were again superior to those obtained for commonly used double ovoid (+/- 15%), linear cylinder (+/- 27%), or a "T" source (31%).

Scattering effects on the dosimetry of iridium-192.

Dosimetry calculations for iridium-192 sources generally assume that a sufficient medium surrounds both the iridium source(s) and the point of calculation so that full scattering conditions exist. In several clinical applications the iridium sources may be anatomically located so that the full scattering requirement is not satisfied. To assess the magnitude of this problem, relative measurements were made with a small ionization chamber in phantoms near air and lung-equivalent interfaces. Dose reduction caused by decreasing the volume of scattering material near these interfaces was then evaluated for a few clinical applications. The results show that reductions on the order of 8% may be expected at the interface with minimal dose reduction within the volume of the implant itself. In addition, the results indicate the verification of source strength of iridium sources in phantom require phantom dimensions determined by the source-chamber separation distance.

Improved linac dose distributions for radiosurgery with elliptically shaped fields.

Stereotactic radiosurgery techniques for a linear accelerator typically use circular radiation fields to produce an essentially spherical radiation distribution with a steep dose gradient. Target volumes are frequently irregular in shape, and circular distributions may irradiate normal tissues to high dose as well as the target volume. Improvements to the dose distribution have been made using multiple target points and optimizing the dose per arc to the target. A retrospective review of 20 radiosurgery patients has suggested that the use of elliptically shaped fields may further improve the match of the radiation distribution to the intended target volume. This hypothesis has been verified with film measurements of the radiation distribution obtained using elliptical radiation beam in a head phantom. Reductions of 40% of the high dose volume have been obtained with elliptical fields compared to circular fields without compromising the dose to the target volume.

Evaluation of heating patterns of microwave interstitial applicators using miniature electric field and fluoroptic temperature probes.

The SAR patterns were determined for four commercially available microwave (915 MHz) interstitial applicators. Values of SAR were determined using a miniature (3 mm diameter) implantable isotropic electric field probe or a custom 0.25 mm diameter fluoroptic temperature probe. These are the smallest such probes that are currently available. Similar radial variation of SAR was found at the axial position of the gap in the outer conductor for each applicator. Electric field probe measurements are much faster and avoid some of the errors caused by the rapid spatial variation of SAR with interstitial applicators. The major limitation on the electric field probe is its size; it is larger than the applicators being tested.

Stereotactic target point verification of an X ray and CT localizer.

Stereotactic radiosurgery with a linear accelerator requires the accurate determination of a target volume and an accurate match of the therapeutic radiation dose distribution to the target volume. X ray and CT localizers have been described that are used to define the target volume or target point from angiographic or CT data. To verify the accuracy of these localizers, measurements were made with a target point simulator and an anthropomorphic head phantom. The accuracy of determining a known, high contrast, target point with these localizers was found to be a maximum of +/- 0.5 mm and +/- 1.0 mm for the X ray and CT localizer, respectively. A technique using portal X rays taken with a linear accelerator to verify the target point is also described.

Conservative surgery and radiation therapy for early stage breast cancer. Can large trial experiences be reproduced in a community hospital setting?

Conservative surgery and radiation therapy were used to treat 212 patients with AJC clinical Stage I or II breast carcinoma at the Baptist Hospital of Miami. All had lumpectomy and most axillary lymph node dissection, followed by breast irradiation to a dose of 45 Gy and a boost dose of 14 to 16 Gy to the surgical bed. Median follow-up was 55 months. The five-year actuarial local control and survival rates were 94% and 86% respectively. Eighty-six percent of the patients had excellent or good cosmetic results with minimal differences between the treated and untreated breasts. Treatment-related complications were minor and infrequent. These results appear comparable to retrospective reviews at major university centers and ongoing prospective randomized trials.

Small-field stereotactic external-beam radiation therapy of intracranial lesions: fractionated treatment with a fixed-halo immobilization device.

Current techniques of stereotactic, small-field, external-beam irradiation with linear accelerators require treatment with a single fraction or only a few fractions of radiation with large single doses per fraction. Using a radiolucent halo that remained affixed to the cranium with skin-piercing screws, the authors treated 24 patients with a multifraction technique for benign and malignant brain lesions. The objective of this study was to ascertain the feasibility of maintaining the halo in place for a prolonged, multifraction course of treatment, not to assess treatment efficacy. The halo was affixed for multifraction treatments lasting 19-58 days (mean, 38.7 days; median, 40.0 days) and delivered in 16-31 fractions (mean, 24.9 fractions; median, 25.5 fractions). Two of 24 patients experienced superficial skin infection at the site of fixation, but no other significant acute or chronic toxicity attributable to the stereotactic halo was observed. The authors conclude that stereotactic, small-field, precision irradiation can be accomplished with multiple fractions as well as with a single fraction.