0360-3016/91 $3.00 + .oO Copyright 0 1991 Pergamon Press plc

Iru. J. Rndrotion Oncology Rio/. Phys., Vol. 20, pp. 1215-1218 Printed in the U.S.A. All rights reserved.

??Original Contribution

RADIOTHERAPY

OF Tl GLOTTIC

CANCER

WITH 6 MeV X RAYS

YASUYUKI AKINE, M.D.,* NOBUHIKO TOKITA, M.D.,* TAKASHI OGINO, M.D.,* IWAO TSUKIYAMA, M.D.,* SUNAO EGAWA, M.D.,* MASAHISA SAIKAWA, M.D.,+ WAICHIROH OHYAMA, M.D.,+ TAKASHI YOSHIZUMI, M.D.+ AND SATOSHI EBIHARA, M.D.+ The NationalCancerCenter Hospital,Tokyo, Japan We treated 154 patients with Tl glottic carcinomawith 6 MeV X rays through 16 cm2parallel-opposingopen fields on a free set-up delivering a median dose of 67 Gy in 63 weeks. Observed and relative 5-year survival rates for all patients were 87% and lOO%, respectively. The local control rate at 5 years was 89%. Of 18 patients who clinically had local recurrence, 17 were salvaged by a secondary treatment. There were no complications requiring medical or surgical attention. A tendency toward increasing local control rates with increasing total doses was observed in the range between 57.5 Gy and 72.5. No significant correlation was found between local control rates and field size, daily dose, or the technique used. A tendency toward a lower local control rate was noted for patients whose anterior commissures were grossly involved, however, it is not known if this could be attributed to the use of 6 MeV X rays. The results are comparable to those obtained with @‘Coas reported in the literature. It is concluded that 6 MeV X rays on a free set-up delivering 65-70 Gy in 64-7 weeks can be used satisfactorily for the treatment of early glottic carcinoma. Glottic carcinoma, Tl, Radiotherapy, 6 MeV X rays, Radiotherapy technique.

INTRODUCrION

to 83 with a median value of 63, and there were only 6 women in this series. X rays from a 6 MeV linear accelerator (SAD = 100 cm) were used for 151 patients. Eight of these patients also received part of their treatment with a 6oCo teletherapy unit (SAD = 80 cm). Only three patients were treated with 6oCo alone. Patients were placed in the supine position without immobilization devices during irradiation. One hundred sixteen patients were treated with lateral opposing fields without a bolus or a wedge filter. Of the patients 110 were irradiated with 6 MeV X rays alone. Twenty-two patients received treatments through a single lateral field. Two patients were treated with anterior oblique fields with wedge filters, and one with anterolateral right-angle fields with wedge filters. Thirteen patients were treated with a combination of these fields. The field size was 4 cm X 4 cm for 131 patients (Fig. 1). Nine patients were treated through fields ranging from 9 to 12 cm2, and for the remaining 14 patients, the field size was 25 cm2. The field size remained unchanged throughout the treatment period in the majority of patients.

Although radiotherapy has been the treatment of choice for patients with early glottic carcinoma, there has been no consensus as to the most appropriate radiotherapy technique. These patients are preferably treated with 6oCo or 4 MeV X rays since the use of higher energy X rays may underdose the lesion because of lack of electronic equilibrium (2). As a result, there have been few reports on the treatment of early glottic carcinoma in which photons with energy higher than that of 6oCo are used. In 1962 we started using 6 MeV X rays for the treatment of these patients, and we established the current protocol in 1967. The present study compares our treatment protocol and treatment results to those reported in the literature.

METHODS AND MATERIALS During the period from 1967to 1983, 154 patients with T 1 (International Union against Cancer, 1989(6))invasive squamous cell carcinoma of the glottis were treated with radiotherapy. Of the 154, 124 patients had Tla lesions and the rest had Tl b lesions. Patient age ranged from 33

* Dept. of Radiation Therapy. of Head and Neck Surgical Oncology. Reprint requests to: Y. Akine, M.D., Department of Radiation Therapy, The National Cancer Center Hospital, 5-l-l Tsukiji, Chuou-ku, Tokyo, Japan.

Supported in part by Grant-in-Aid for Cancer Research 35 from the Japanese Ministry of Health and Welfare. Accepted for publication 14 December 1990.

+ Dept.

1215

l-

1216

I. J. Radiation Oncology 0 Biology 0 Physics

June 199 1, Volume 20, Number 6

lated with the method of Kaplan and Meier (7), and their differences were examined with both the generalized Wilcoxon test and the log rank test. The statistical analysis was done with a statistical analysis program.* RESULTS

Fig. 1. A radiogram taken with a linear accelerator showing 4 X 4 cm2 radiation field used at our institution.

A group of 108 patients received a daily fraction dose of 2 Gy. Thirty-seven patients received a daily dose of 2.4 Gy. Others received daily doses of 1.9 Gy (seven patients), and one patient each received 2.2 Gy or 3.1 Gy daily. The dose was specified at the mid-point without tissue inhomogeneity corrections and the distribution of total doses is shown in Figure 2. All patients were treated 5 times a week; 109 patients were treated both fields daily, and the remaining 45 were treated only one field daily. Data on vital status for all patients were obtained for a minimum of 5 years or until their deaths. Observed and relative survival rates for all patients were calculated with the actuarial method (1). Local control rates were calcu7

t

Observed and relative 5-year survival rates were 87% and lOO%, respectively. The local control rate at 5 years was 89% for all patients. One of 19 patients who failed was found to have the glottic carcinoma on autopsy. Of 18 patients who clinically developed local recurrence, 14 were treated with total laryngectomy, 3 with partial laryngectomy, and 1 with cryosurgery. One of the 18 patients who clinically failed died of disseminated laryngeal cancer. The remaining patients were salvaged with the secondary treatments. Of the 19 patients who failed, 3 had glottic carcinoma on the same side of the vocal cord as the initial one 5 years or longer after the initial treatments. No patient had a complication requiring medical or surgical treatment. Local control rates by subgroups according to the total dose given are shown in Table 1. Although there was no statistical significance, a tendency toward increasing local control rates with increasing total doses in the range between 57.5 Gy and 72.5 Gy was observed. There was a tendency toward lower local control rate for patients with the anterior commissure being grossly involved as compared to that for patients without it (Table 2). Patients were subgrouped according to main characteristics of their tumors in appearance as follows: granulation-like, which had resemblance to granulation tissue and was reddish in color-48 patients; hyperkeratotic, which had fine granular surface and whitish in color-85; endophytic-2; superficial-7; necrotic-2; polypoid- 10. Subgroups of patients, whose tumors were other than granulation-like or hyperkeratotic, were not analyzed because of their small numbers of patients. The subgroup hyperkeratotic had a tendency of lower local control rates as compared to that of granulation-like (p = 0.11, the generalized Wilcoxon test, p = 0.05, the log rank test), which continued to decrease after 5 years (86% at 5 years, 77% at 10 years). Statistical analysis revealed that there

Table 1. Local control rates by total doses

Fig. 2. Number of patients plotted as a function total doses.

* SAS-SAS Institute, Inc., Cary, NC, USA.

Total dose (GY)

Number of patients

57.5> 62.5>, 257.5 67.5>, 262.5 72.5>, 867.5 272.5

1 29 49 66 9

Local control rate at 5 years 0.80 0.91 0.93 0.78

l/l + + t +

0.08 0.04 0.03 0.14

Radiotherapy of T I glottic cancer 0 Y. Table 2. Relationship between local control rates and parameters

Parameters

Number of patients

Local control rates at 5 years (%)

27 127

81 91

0.06

48 85

96 85

0.11

86 68

93 84

0.08

124 30

90 84

0.5 1

47 44 33

86 95 89

0.36

39 108 7

91 89 86

0.96

14 131 9

77 90 88

0.15

110 44

90 85

0.25

Anterior commissure Grossly involved No gross involvement Appearance of the tumor Granulation-like Hyperkeratotic Age 632 >63 T Tla Tlb Size of tumor in T I a 313 213 l/3 Daily dose >2 Gy 2 GY 2 Gy> Field size 25 cm* 16 cm* 16 cm*> Technique Parallel opposing with 6 MeV x rays Others * The generalized

Wilcoxon

p*

test.

was no correlation between local control rates and age, substage, size of the tumor, daily dose, size of the field, or therapy technique used (Table 2).

ISCCCSES 0 1

100 90

3 4 5

60 40 20

i

ao

1217

AKINE et al.

DISCUSSION With regard to our treatment protocol, questions arise as to the use of 6 MeV X rays to irradiate a treatment volume defined by a field size of 4 X 4 cm* through parallel-opposing open fields. A radiograph taken with a linear accelerator shows that this portal adequately covers the vocal cords (Fig. 1). Harwood et al. reported a higher local recurrence rate for patients treated with 6oCo through a field size of 25 cm* or smaller (4). When 6oCo was used, the field size was defined on a 50% isodose line and if the field was defined on a 90% isodose line, the treatment volume would be smaller. Fletcher and Goepfert recommend inclusion of the cricoid and at least one tracheal ring in the radiation portal for lesions involving the anterior commissure (3). In this case the portal size would exceed 4 X 4 cm*. The present study shows that the results are comparable to those reported in the literature using larger portal sizes. Opinions differ concerning the significance of the involvement of the anterior commissure by tumors on local control rates (3, 8). A lower control rate for patients with these tumors was thought to result from reduced doses to this region by some investigators (9). The present study shows a tendency of lower local control rate for patients with the anterior commissure being grossly involved. It is yet be determined whether this could be attributed to the use of 6 MeV X rays. When parallel opposing open fields are used, the dose inhomogeneity occurs at anterior and posterior extremities in the treatment volume because of the anatomical contour of the neck. Computed dose distribution maps show a very appropriate dose distribution of 6 MeV X rays when compared to that of 6oCo (Figs. 3a, 3b). Henk used a pair of anterior oblique fields covering the thyroid and cricoid cartilages for early glottic carcinoma (5). Fletcher and Goepfert used open fields for the first half and wedged

ISCCOSES I

0 1 2 3 4

100

5

20

I% 60 40

Fig. 3. Dose distribution maps for 6 MeV X rays (a) and 6oCo (b) generated incorporates computer tomography data.

by a treatment

planning

system which

1218

I. J. Radiation Oncology 0 Biology0 Physics

fields for the second half with an exclusion of the arytenoid cartilage at the dose of 50-55 Gy for lesions involving the posterior third of the cord (3). In our series, the arytenoid cartilage was included in the field during an entire course of irradiation. A prominent feature of 6 MeV X rays is the longer distance of secondary electrons to attain an electronic equilibrium in comparison with photons of lower energies. Inadequate build-up of electrons may be found beneath the skin of the neck and at the surface of the vocal cord. The vocal cords are located on the surface of an air cavity in the soft tissue. Epp and associates reported that the presence of an air gap of a 4 cm dimension and absence

June 1991, Volume 20, Number 6

of side or end walls caused an appreciable deficiency in dose at the distal surface of the air cavity when 10 MeV X rays were used (2). Although the dose reduction at the surface of the vocal cords is conceivable, the local control obtained in our series is comparable to those reported elsewhere (3, 4, 10). Since we observed a tendency toward increasing local control rates with increasing total dose in a range between 57.5 Gy and 72.5 Gy, a total dose of 65-70 Gy in 6.5-7 weeks is thought to be appropriate for early glottic cancer. The use of 6 MeV X rays on a free set up is an effective alternative to conventional treatment techniques using lower energy photons.

REFERENCES 1. Cutler, S. J.; Ederer, F. Maximum utilization of the life table method in analyzing survival. J. Chron. Dis. 8:699712; 1958. 2. Epp, E. R.; Boyer, A. L.; Doppke, K. P. Underdosing of lesions resulting from lack of electronic equilibrium in upper respiratory air cavity irradiated by 10 MV X-ray beams. Int. J. Radiat. Oncol. Biol. Phys. 2:613-6 19; 1977. 3. Fletcher, G. H.; Goepfert, H. Larynx and pyriform sinus. In: Fletcher, G. H., ed. Textbook of radiotherapy, 3rd edition. Philadelphia, PA: Lea & Febiger; 1980:330-363. 4. Harwood, A. R.; Hawkins, N. V.; Rider, W. D.; Bryce, D. P. Radiotherapy of early glottic cancer-I. Int. J. Radiat. Oncol. Biol. Phys. 5473-476; 1979. 5. Henk, J. M. Cancer of the head and neck. In: Hope-Stone, H. F., ed. Radiotherapy in clinical practice. London: Butterworth & Co.; 1986:93- 123.

6. Hermanek, P.; Sobin, L. H., eds. TNM classification of malignant tumours. Geneva: International Union Against Cancer; Berlin: Springer-Verlag; 1987. Kaplan, E. L.; Meier, P. Nonparametric estimation for incomplete observation. J. Am. Stat. Assoc. 53:457-48 1; 1958. Kirchner, J. A. Cancer at the anterior commissure of the larynx. Arch Gtolaryng. 9 1:524-525; 1970. Meyer-Breiting, E.; Burkhardt, A. Management of laryngeal cancer and histopathological inference. In: Meyer-Breitinb, E.; Burkhardt, A., eds. Tumours of the larynx. Berlin: Springer-Verlag; 1988: 115-141. 10. Wang, C. C. Treatment of glottic carcinoma by megavoltage radiation therapy and results. Am. J. Roentgenol. 120: 157163; 1974.