Good Afternoon Dr. Jones
My Calendar & Email
May
6, 2000
Weather Here
Weather Here
|
|
| ||
|
May
6, 2000 Weather Here | 2:45 a.m. | |
|
Ronald B. Kuppersmith,
MDa
Houston
Abstract
We retrospectively reviewed the charts of the first 28 head and neck patients at our institution who were treated with IMRT. All had head and neck neoplasms, including squamous cell carcinoma, adenoid cystic carcinoma, paraganglioma, and angiofibroma. Total radiation doses ranged from 1,400 to 7,100 cGy, and daily doses ranged from 150 to 400 cGy/day. A quality assurance system ensured that computer-generated dosimetry matched film dosimetry in all cases. For midline tumors, this system allowed us to decrease the dose to the parotid glands to less than 3,000 cGy. The incidence of acute toxicity was drastically lower than that seen with conventional radiotherapy delivery to similar sites. This is the first report of the application of IMRT strictly to head and neck neoplasms. We discuss the indications, technique, and initial results of this promising new technology. We also introduce the concept of the Simultaneous Modulated Accelerated Radiation Therapy boost technique, which has several advantages over other altered fractionation schemes.
Introduction
In conventional radiotherapy, standard treatment protocols have evolved for various disease sites based on clinical experience, reproducibility, and duration of treatment. The radiation dose delivered to the tumor and surrounding tissue can be modified by adjusting the fixed treatment fields, beam energies, the use of electrons, the weight of different beams, and the use of wedges, boluses, and tissue compensators. Unfortunately,
the standard treatment planning process is based on trial and error, and
modification of a specific treatment plan can require hours of computation
without any guarantee that the resultant plan will be satisfactory. Furthermore,
conventional radiotherapy utilizes a relatively small number of fixed
fields to deliver a homogenous radiation dose to the tumor and surrounding
structures. When radiation directed to normal tissues exceeds the tissues’
tolerance, complications may occur. Intensity-modulated beam radiotherapy (IMRT) is a new technology that is based on arc rotation of the beam through 270º (figure 1) and a collimator that has 40 small fields measuring 1 cm2 (figure 2). During rotation, the configuration of the collimator can be adjusted every 5º, thereby allowing for the delivery of highly conformal doses to a tumor while sparing the adjacent normal structures. IMRT offers several advantages over conventional radiotherapy: its ability to treat a target within a target, its ability to escalate the dose to the tumor while sparing critical structures, its use of 3-D computers for optimized treatment planning, and its ability to re-treat previously irradiated patients. Never before has the radiation oncologist been able to write a dose prescription for the avoidance of normal tissue to such a degree. After a thorough evaluation of the system, IMRT was first used clinically in the United States on March 21, 1994, at the Baylor College of Medicine in Houston.1-9 This article is a report of that series of procedures.
In preparation for immobilization, patients were placed under general anesthesia. A neurosurgeon then placed screws in the vertex of the skull. A talon (docking device) was attached to the screws, completing the immobilization and readying the patient for the planning CT and subsequent daily treatments. The degree of reproducibility in positioning patients for later treatments was excellent: within 1 mm. For treatment planning purposes, physicians obtained a high-resolution, contrast-enhanced CT at 3-mm intervals with 3-mm thickness. The target and normal structures were identified, and data were entered into the computerized treatment planning system. The normal structures that are typically identified in head and neck patients are the orbit, lens, retina, optic nerve, optic chiasm, brainstem, spinal cord, brain, mandible, parotid glands, and lacrimal glands. In cases of tongue cancer, the tongue was immobilized with a stent that was specifically tailored to the patient’s anatomy. Glottic lesions were treated with slightly larger fields to compensate for potential movement during therapy. After the target
and normal structures were identified, the computerized treatment planning
system performed "inverse planning" and calculated optimal doses.
(Planning is called "inverse" because rather than placing the
radiation field first and then looking at the dose distribution around
the tumor and normal structures, the IMRT system defines the tumor and
normal structures and assigns optimum doses first. Then the computer system
determines the best treatment field.) As part of inverse planning, doses
of radiation were assigned to the tumor and normal structures, and the
computer defined the pattern of delivery of these doses. Dose optimization
was accomplished by assigning weighted values to the target(s) and their
surrounding structures according to their relative importance. (Targets
are usually given priority over normal adjacent tissues when weights are
assigned, but not always. If, for example, the optic chiasm is surrounded
by the target tumor and the patient cannot accept the risk of a loss of
vision, the optic chiasm will be assigned a higher weight.) The computer
planning system then optimized the treatment plan by 1) minimizing the
amount of radiation that was delivered to the surrounding structures,
as ranked by their relative importance (weighted score) while 2) maximizing
the dose to the target(s). Dose optimization was determined by calculating the "cost function." In simplified terms, the cost function is the sum of the amount of target structure that will be irradiated at less than the prescribed dose plus the amount of normal structure that will be irradiated at more than the limiting dose, weighted by their assigned priority. The objective is to minimize the cost function. Dose modulation was accomplished with the special dynamic computer-controlled multivane collimator (figure 2). The apertures of the 40 small "beams" on the collimator are calibrated in increments of 10% at every 5º of arc, as determined by intensity distribution patterns generated by the computer program. After the planning
was completed, extensive quality assurance with film dosimetry was performed
to document dosimetry. Therapy was then initiated. The daily setup time
was typically 5 to 10 minutes, and the duration of treatment ranged from
12 to 15 minutes for patients who received two or three arcs, up to as
long as 30 minutes for eight arcs.
Results Twenty-eight patients with head or neck neoplasms were treated between March 1994 and April 1997. Patient characteristics and treatment details are shown in table 1 (for patients who had a history of radiotherapy) and table 2 (for patients who received primary therapy). Acute toxicity was graded according to RTOG criteria, from grade 1 (mild) to grade 4 (severe), and found to be minimal for the ear, eye, salivary glands, and skin. In general, grade 3 (moderate) complications were confined to patients who were treated with a full dose that covered a large volume of mucosal membrane within the oral cavity and oropharynx in an accelerated fractionation scheme. (Full doses were administered to 18 patients who had had no previous radiotherapy.) These complications were clearly the result of the aggressive treatment and were no different from those seen after accelerated treatment with standard techniques. Two patients required feeding tubes due to poor oral intake that was related to mucositis. Otherwise, treatment was extremely well tolerated. The skull screws were well tolerated by all but 1 patient, who experienced a wound infection at his screw site. Analysis of the dose delivered to the parotid glands revealed an exceptional degree of tissue sparing, as significant portions of these glands received doses that are considered to be within their tolerance limit (figure 3). Parotid doses were determined for both the contralateral and ipsilateral glands in 12 patients. Parotid glands that were in the target area were excluded from the tissue-sparing analysis. In patients who had midline tumors, the parotid doses were not typically identified during treatment planning and these, too, were not included in the analysis because they received minimal doses. Although long-term outcome data have not yet matured, thus far only 1 of 20 (5%) definitively treated patients has demonstrated a local failure.
Case Studies
Case 1 A 67-year-old man with a history of a poorly differentiated T4N0M0 nasopharyngeal carcinoma—with invasion into the middle cranial fossa in the area of the foramen ovale, the petrous apex, right cavernous sinus, and portions of the clivus—was first treated in 1994. His initial therapy consisted of one cycle of chemotherapy with 5-fluorouracil and cisplatin (which had to be discontinued because of neurologic toxicity) and conventional radiation therapy (6,700 cGy delivered to the area of primary disease and 6,000 cGy to the draining lymphatics). Radiation complications included xerostomia, retinal neovascularization, and skin changes on the neck. An MRI performed in December 1995 showed that his disease had progressed to the point that he was no longer considered to be a good candidate for conventional radiation, and he was still not a good candidate for chemotherapy. He was referred for IMRT. The man received 4,000 cGy (in 20 fractions of 200 cGy each) at the primary site. Radiation doses were minimized at the critical structures that were identified during treatment planning: the temporal lobes of the brain (1,320 cGy), the lens of each eye (273 cGy), the orbits (250 cGy, 327 cGy), the optic chiasm (172 cGy), the brainstem (509 cGy), and the optic nerves (311 cGy). The patient tolerated IMRT extremely well and manifested no evidence of acute toxicity.
Case 2 A 50-year-old man had a 4-year history of palatal discomfort. An excisional biopsy of the soft palate lesion revealed a well-differentiated adenoid cystic carcinoma of the cribriform type. The tumor exhibited perineural invasion. CT examination of the pharynx and skull base revealed no evidence of disease in the pterygopalatine fossa or in the foramen rotundum. The patient was referred for evaluation by radiotherapy, and he elected to undergo IMRT. The patient received 6,000 cGy (in 25 fractions of 240 cGy each) at the primary site and another 5,000 cGy (25 fractions of 200 cGy) extending along the nerve pathway at risk back to Meckel’s cave. Minimal doses were delivered to the critical structures: the mandible (1,413 cGy), lenses (650 cGy), optic chiasm (176 cGy), optic nerves (1,202 cGy), retina (677 cGy), brainstem (1,246 cGy), spinal cord (1,195 cGy), right parotid (1,186 cGy), and left parotid (1,534 cGy). The patient tolerated IMRT well, although during treatment he developed moderate mucositis (RTOG grade 3), mild xerostomia (grade 1), and mild dysphagia (grade 1). At 14 months post-therapy, he had no evidence of disease or late complications of radiotherapy.
Case 3 A 62-year-old man had a 2-month history of a progressive sore throat on the left. On examination, he was noted to have a 4-cm ulcerative lesion of the soft palate and anterior tonsillar pillar. Incisional biopsy revealed a moderately differentiated squamous cell carcinoma. Based on the physical exam, CT, triple endoscopy, and chest x-ray, the patient was staged as T3N0M0. After a multidisciplinary evaluation, he decided to undergo IMRT. The patient received 6,000 cGy (25 fractions of 240 cGy) at the primary site and 5,000 cGy (25 fractions of 200 cGy) in the cervical lymphatics to control possible microscopic disease. Again, doses were minimized at the critical structures: spinal cord (1,977 cGy), brainstem (669 cGy), left parotid gland (2,277 cGy), right parotid gland (5,226 cGy), mandible (3,587 cGy), and orbits (73 cGy). The patient completed his radiation as scheduled, but he developed confluent mucositis (RTOG stage 3) and experienced a weight loss of 15 lbs during treatment, which necessitated the placement of a feeding tube.
Case 4 A 72-year-old man gave a history of a left-sided glomus jugulare, which extended from the base of his skull to the internal auditory canal and which involved the jugular foramen, carotid canal, and hypoglossal canal. The condition resulted in deficits of cranial nerves VIII through XII. An MRI to evaluate the lesion showed the presence of a pituitary macroadenoma (2 ¥ 1.5 ¥ 1 cm). Multidisciplinary evaluation revealed that the pituitary macroadenoma was nonsecreting, and that both lesions would be treated by IMRT. The patient received 4,500 cGy at the pituitary adenoma and the glomus jugulare (25 fractions of 180 cGy). Doses were minimized at critical structures: optic chiasm (3,981 cGy), optic nerve (1,286 cGy), orbits (541 cGy), brain (388 cGy), brainstem (1,418 cGy), and spinal cord (895 cGy). The patient tolerated radiation extremely well, with no evidence of acute toxicity. After the completion of therapy, the patient’s ability to swallow improved. An MRI 2 years after therapy revealed a dramatic shrinkage of the pituitary adenoma, but the glomus tumor remained unchanged.
Discussion
With conventional technology, the radiation oncologist’s ability to avoid depositing full radiation doses in nearby normal tissues is limited. IMRT allows physicians to plan and deliver selective doses that conformally avoid normal critical structures that either lie adjacent to or are surrounded by the tumor. The four specific cases described here illustrate this point with reference to the parotid gland. Conventional radiotherapy technology frequently delivers large radiation doses to most of the entire parotid glands, which often leads to permanent xerostomia and adversely affects the patient’s quality of life. The ability to spare the parotids while depositing tumoricidal doses to the primary target lessens the severity of xerostomia without jeopardizing control of the tumor. The ability to conformally avoid normal structures makes dose escalation to the tumor a more feasible option. With conventional radiotherapy, planning treatment is an exercise in trial and error. It is also time consuming and does not always produce the best solutions. With IMRT, fast, powerful computers can identify and evaluate many "iterations" (combinations of delivery intensity) within a reasonable time and use mathematical models to find optimal solutions closest to the radiation oncologist’s prescription. IMRT’s ability to shorten the overall duration of treatment in head and neck cancer with its accelerated functional radiation has improved local control by l5%.11-15 The biologic explanation is that IMRT inhibits tumor cell repopulation. Conventional accelerated radiation therapy has been typically administered in more than one fraction per day. Some of the different schemata are listed in table 3; they include the Massachusetts General Hospital accelerated split course,12 the M.D. Anderson concomitant boost,13 and the Continuous Hyperfractionated Accelerated Radiotherapy (CHART).14 One exception to twice-a-day fractionation is the Polish regimen, also called Continuous Accelerated Radiotherapy (CARE), in which patients are treated once a day every day of the week, including weekends.16 With our own SMART boost regimen, patients undergo IMRT only once a day, but they still receive a higher-than-conventional dose at the primary target (ie, a site of visible and/or palpable disease) along with a conventional fraction at the secondary target (ie, a region at risk for microscopic disease). This regimen is more convenient and is likely to be more cost-effective than twice-a-day dosing. In conventional radiotherapy, we frequently deal with whole-organ tolerance. When only a portion of an organ is irradiated, its tolerance is likely to increase. Through the accumulation of clinical and dose volume data regarding new radiotherapy technologies, we are likely to improve our understanding of partial-organ tolerance and perhaps significantly increase the radiation dose to tumors. IMRT has made feasible new approaches to radiotherapy in head and neck neoplasms. Clinical data are continuously being accumulated, and it is hoped that they will demonstrate an improvement in cure rates and a reduction in morbidity.
References 1. Woo
SY, Sanders M, Grant W, Butler EB. Does the "peacock" have anything
to do with radiotherapy [editorial]? 2. Butler
EB, Woo SY, Grant W, Nizin PS. Clinical realization of 3D conformal intensity
modulated radiotherapy [letter]. 3. Carol
M, Grant WH, Bleier AR, et al. The field-matching problem as it applies
to the Peacock three dimensional conformal system for 4. Lee
AG, Woo SY, Miller NR, et al. Improvement in visual function in an eye
with a presumed optic nerve sheath meningioma after treatment with 5. Woo
SY, Grant WH, Bellezza D, et al. A comparison of intensity modulated conformal
therapy with a conventional external beam stereotactic 6. Grant
W. Experience in intensity modulated beam delivery. In: Mackie TR, Palta
JR, eds. Teletherapy: Present and Future. Madison, Wis.: 7. McGary
JE, Grant WH, Woo SY, Butler EB. Reporting and analyzing dose distributions:
A concept of equivalent uniform dose [comment]. 8. Butler
EB, Woo SY, Grant WH. Intensity modulated therapy and inhomogenous dose
to the tumor: A note of caution [letter]. 9. Sternick ES, ed. The Theory and Practice of Intensity Modulated Radiation Therapy. Madison, Wis.: Advanced Medical Publishing, 1997. 10. Cox JD,
Stetz J, Pajak TF. Toxicity criteria of the Radiation Therapy Oncology
Group (RTOG) and the European Organization for Research and 11. Peters LJ,
Ang KK, Thames HD Jr. Accelerated fractionation in the radiation treatment
of head and neck cancer: A critical comparison of different 12. Wang CC. Improved local control for advanced oropharyngeal carcinoma following twice daily radiation therapy. Am J Clin Oncol 1985;8:512-6. 13. Ang KK,
Peters LJ, Weber RS, et al. Concomitant boost radiotherapy schedules in
the treatment of carcinoma of the oropharynx and nasopharynx. 14. Saunders MI, Dische S, Grosch EJ, et al. Experience with CHART. Int J Radiat Oncol Biol Phys 1991;21:871-8. 15. Parsons JT, Mendenhall WM, Cassisi NJ, et al. Hyperfractionation for head and neck cancer. Int J Radiat Oncol Biol Phys 1988;14:649-58. 16. Maciejewski
B, Skladowski K, Pilecki B, et al. Randomized clinical trial on accelerated
7 days per week fractionation in radiotherapy for head a
The Bobby R. Alford Department
of Otorhinolaryngology and Communicative Sciences, Baylor College of Medicine,
Houston. Reprint requests:
Brian Butler, MD, Department of Radiation Oncology, Baylor College of
Medicine, One Baylor Plaza, Houston, TX 77030.
[ APRIL TOC ] [ ENTKnow Members Page ] [ ENT Journal ] |
Procedure 
