Int. J Radiation Oncology Biol. P/q.. Vol. 23, pp. 133-139 Printed I” the U.S.A. All tights reserved.

0360-3016/92 $5.00 + .M) Copyright 0 1992 Pergamon Press Ltd.

??Phase I/II Clinical Trial

SEQUENTIAL COMPARISON OF LOW DOSE RATE AND HYPERFRACTIONATED HIGH DOSE RATE ENDOBRONCHIAL RADIATION FOR

MALIGNANT AIRWAY OCCLUSION

M. MEHTA, M.D.,* D. PETEREIT, M.D., *,+ L. CHOSY, M.D.,? M. HARMON, M.D.,*

J. FOWLER, PH.D.,* S. SHAHABI, PH.D.,*,* B. THOMADSEN, PH.D.*,* AND T. KINSELLA, M.D.*

University of Wisconsin-Madison, Madison, WI 53192

A pilot trial (S2) was conducted at the University of Wisconsin to determine the feasibility, efficacy, and toxicity of hyperfractionated high dose rate endobronchial radiation. To avoid multiple bronchoscopies, an optimized hy- perfractionated schema was derived from the linear-quadratic model. Utilizing a single bronchoscopy, 31 patients with malignant airway occlusion received 4 Gy X 4 fractions over 2 days at 2 cm from source center using a high dose rate remote afterloader. Response and morbidity were compared to a previous trial (Sl) in which 66 patients were treated with conventional low dose rate endobronchial radiation. Response was assessed by change in perfor- mance status, symptom resolution, percent of lifetime rendered symptom-free or improved, and radiographic reaer- ation. These parameters were highly comparable between the two groups. The mean ECOG performance status improved from 2.2 to 1.8 for Sl and 2.1 to 1.6 for S2; symptom improvement or resolution was noted in 78% for Sl and 79% for S2; lifetime rendered symptom-free or improved was 54% for Sl and 57% for S2; and the overall radiographic response rate was 78% for Sl and 85% for S2. The overall incidence of fistulae was 7/101. We conclude that endobronchial radiation is an effective and safe modality for palliation, and hyperfractionated high dose rate endobronchial radiation achieves responses similar to low dose rate endobronchial radiation with a similar complication rate.

Lung cancer, Endobronchial radiation, High dose rate, Brachytherapy, Remote afterloading.

INTRODUCIION

With an annual incidence of more than 160,000 and a mortality rate of greater than 80%, lung cancer represents our greatest oncologic challenge (3). Although metastatic disease is not uncommon, failure to obtain local control continues to be a major issue. Local failure rates between 3 1% to 5 1% have been reported in several Radiation Therapy Oncology Group (RTOG) studies (15). In 300 consecutive autopsies from the Veterans’ Administration Lung Group (VALG) protocols (5), residual intrathoracic tumor was responsible for death in 75% of the patients with squamous cell carcinoma and 50% of the patients with adenocarcinoma and large cell carcinoma of the lung. Endobronchial occlusion is a common and potentially life-threatening complication, not only for patients with recurrent disease, but also at initial diagnosis. According to one estimate, 20-30% of newly diagnosed lung neo-

plasms will present with atelectasis and pneumonia due to endobronchial disease (11). Symptomatic occlusion of- ten leads to obstructive pneumonitis and hemoptysis fol- lowed by slow asphyxiation and a painful death. With a high rate of local failure using conventional therapies, an estimated 50% of patients with lung cancer will develop symptomatic endobronchial involvement. Several groups, including our own (9, 10, 12, 13, 18, 19), have demon- strated that conventional dose rate (30-60 cGy/hr) en- dobronchial radiation is highly effective in managing ma- lignant airway occlusion.

The purpose of this paper is to review an updated group of patients treated with low dose rate (LDR) endobron- chial radiation (ERT) (Sl), and to compare them with a new set of patients that were treated with high dose rate (HDR) endobronchial radiation (ERT) (S2). Parameters studied were change in Eastern Cooperative Oncology Group (ECOG) performance status, change in symptoms

Presented at the 33rd Annual Meeting of ASTRO, Washing- ton, DC, October 199 1.

* Department of Human Oncology. + Department of Medicine. * Department of Medical Physics. Reprint requests to: Minesh Mehta, M.D., Radiation Oncol-

133

ogy (K4/BlOO), University of Wisconsin Hospital, 600 Highland Ave., Madison, WI 53792. Acknowledgements-The authors wish to thank Rick Chappel for statistical consultation, Judy Smith for manuscript prepa- ration, and Debbie Pope for assistance with data acquisition.

Accepted for publication 4 October 199 1.

134 I. J. Radiation Oncology 0 Biology 0 Physics Volume 23, Number I, 1992

(dyspnea, cough, pneumonia, hemoptysis, and chest pain), percent of lifetime rendered symptom-free or improved, radiographic response, median survival, and complica- tions.

Table 1. Patient characteristics

METHODS AND MATERIALS

Low dose rate Between October 1986 and July 1990, 66 patients un-

derwent 70 LDR (Sl) endobronchial procedures. Under local anesthesia, a flexible fiberoptic bronchoscope was passed up to the point of obstruction. A 5 French poly- ethylene catheter was threaded through the accessory port of the bronchoscope and placed at least 2 cm distal to the tumor. After removal of the bronchoscope, the catheter was taped in place at the nostril. Orthogonal radiographs centered over the implant were obtained at 80 cm SAD, using an isocentric technique. Isodose distributions were calculated using treatment planning software.* The Irid- ium- 192 used in these implants was obtained as an end- to-end seed source with an activity range 37 to 48 MBq per seed.+ This usually resulted in a dose rate of 35 to 40 cGy/hr at 2 cm from the source center. The average length of the h-192 source was 8 cm. Our standard policy was to provide a 2 cm margin proximal and distal to the tumor. The end-to-end seed placement simulated a line source rather than a point source, resulting in approximately lin- ear dose fall-off. The dosimetric details have been pub- lished previously (2 1). The median implant duration was 45 hr. The median dose at 1 and 2 cm from the center of the source was 48 and 20 Gy, respectively. The maximal cord dose ranged from 1 .O to 9.0 Gy with a median dose of 4.0 Gy.

Dates No. of patients No. of procedures Male/female Median age Lung primary Metastasis Prior EBRT Prior chemotherapy Prior XRT Median XRT dose Median months since

XRT Laser

1 O/9/ 86-l/20/90 66 70

53:13

6 16(;2%) 31 (47%)

4 (6%) I7 (26%) 36 (55%)

65 Gy

High dose rate

The median age at presentation was 67 years with a male to female ratio of 53: 13. Ninety-two percent (6 1) had primary pulmonary neoplasms and 8% (5) had non- pulmonary tumors; 47% (3 1) had metastatic disease at the time of ERT; 26% (17) had received chemotherapy and 55% (36) had undergone external radiation prior to implantation with a median dose of 65 Gy and with a median time since failure of 11.5 months. Four patients had previously undergone LDR brachytherapy (Table 1). These four patients progressed inspite of LDR ERT and were re-implanted at recurrence. Twelve patients under- went Neodynium-Yttrium Aluminum Garnet laser ex- cision of the tumor prior to ERT.

Between September 1989 and October 1990,3 1 patients underwent 3 1 HDR (S2) endobronchial procedures. After September 1989, the only patients treated with LDR ERT were those who refused HDR ERT, those who were en- tered on an in-house LDR ERT protocol and the excep- tional patient with an upper lobe lesion where excessive catheter curvature precluded HDR ERT. Patients under- went a single bronchoscopy with placement of a single 6 French catheter. A 6 French catheter was necessary to facilitate easy application of the afterloading source. Pa- tients were then simulated with dummy seeds in place utilizing isocentric technique. Isodose distributions were calculated using treatment planning software.* The Irid- ium-192 used in these implants was a single 4 mm seed source with an approximate activity of 10 Ci. Utilization of this high activity source was made possible because of a remote afterloading system.* Dwell positions were .25 cm apart and the time calculated at each point depended on the dose prescribed. Because of the optimized dwell positions and dwell times, the dose fall-off was an almost perfect linear relationship from an average of 32 Gy at 1 cm to 16 Gy at 2 cm. Before the source was activated for treatment, a test cable was used. If this succeeded, treat- ment was delivered. Treatment times were on the order of minutes. As in the LDR group, 2 cm margins were used and the dose was prescribed at an average of 2 cm from the center of the implant volume.

The initial ECOG performance status and initial symptoms are presented in Table 2. The most common presenting symptoms in decreasing order of frequency were cough (67), dyspnea (63), pneumonia (28), hemop- tysis (23), and chest pain (22).

The median age at presentation was 61 years with a male to female ratio of 23:8. Seventy-seven percent (24) had primary pulmonary neoplasms and 23% (7) had non- pulmonary tumors; 74% (23) had metastatic disease at the time of HDR ERT; 19% (6) had received chemother- apy and 30% (9) had undergone external beam radiation prior to implantation. The median dose from prior ra-

LDR (Sl) HDR (S2)

11.5 12 (18%)

9/15/89-10/3/90 31 31

23:8

24 (6:7%) 23 (74%)

0 6 (19%) 9 (30%) 50 Gy

4 0

* AECL, Ontario, Canada. t BEST Industries, Springfield, VA.

* Nucletron Corporation, Columbia, MD.

HDR and LDR ERT 0 M. MEHTA et a/. 135

Table 2. Response to endobronchial radiation

A. Change in performance status

ECOG PS 0 1 2 3 4 Mean

LDR Pre 3 15 27 17 8 2.2 Post 10 29 16 7 8 1.8

HDR Pre 3 9 5 10 4 2.1 Post 7 13 4 62 5 1.6

in ECGG performance status (PS) (14) symptom reso- lution by Radiation Therapy Oncology Group (RTOG) criteria (17) percent of lifetime rendered symptom-free or improved, roentgenographic reaeration, and median survival.

Radiobiologic rationale

B. Change in symptoms for HDR group (n = 3 1)

Symptom Pre + (%) - (%) f (%)

Cough 26 73 12 15 Dyspnea 24 75 12 13 Pneumonia 14 71 14 15 Hemoptysis 10 100 0 0 Chest pain 8 75 12 13 Total 82 79

+ = Improved; - = Worse; 5 = Unchanged.

The equivalence of 4 HDR fractions of 4 Gy to a con- tinuous low dose rate exposure of 20 Gy in 40 hr (at 50 cGy/hr) was calculated from the standard linear-quadratic formula with the following assumptions:

1. The basic formulae were derived from Dale’s linear quadratic dose-effect equation (6). Our derivation was as follows:

For n fractions, of d Gy each, given at high dose rate and with intervals long enough to allow complete repair of sublethal injury (> 6 hr), the natural logarithm of the proportion of cells not sterilized, E, is given by:

C. Change in symptoms for LDR group (N = 66)

Symptom Pre + (%) - (%) + (%)

Cough 67 68 7 25 Dyspnea 63 79 10 11 Pneumonia 28 82 11 7 Hemoptysis 23 91 4 5 Chest pain 22 73 9 18 Total 203 78

E = n(ad + pd2) (1)

and for a continuous exposure longer than a few min- utes to a total dose of D,

E = (aD + g/?D2) (2)

where g =

D. Roentgenographic response

Response Partial Complete Overall

n = 14 (44%) n = 52 (74%)

HDR (%) LDR (%) 4 (28) 22 (42) 8 (57) 19 (36)

12 (85) 41 (78)

T = duration of the exposure;

or = repair constant = 0.693/T+;

diotherapy was 50 Gy with a median time since failure of 4 months (Table l), considerably shorter than the Sl group.

Ti = half-time for repair of sublethal injury. 2. The half-time for repair was assumed to be T4

= 1.5 hr.

The initial ECOG performance status and initial 3. The ratio of a//3 in equations (1) and (2) was assumed

symptoms are presented in Table 2. The most common to be 10 Gy, representing response of the tumor cells.

presenting symptoms in decreasing order of frequency 4. The ratio of a//3 = 3 was assumed to represent late-

were cough (26), dyspnea (24) pneumonia ( 14) hemop- responding tissues responsible for late complications

tysis (lo), and chest pain (8). such as ulceration or fistulae.

Patients were treated with hyperfractionated brachy- therapy. With at least a 6 hr interval, patients received BID therapy for 2 days with a total dose of 16 Gy at 2 cm. The median dose at 1 and 2 cm from the center of the source was 32 and 16 Gy, respectively. The maximum cord dose ranged from 1.0-5.0 Gy with a median cord dose of 2.5 Gy.

Equal tumor effects were defined in the two modalities, HDR and LDR, when equation (1) was made equal to equation (2) assuming that a/P = 10 Gy. This was first done for a single HDR insertion, yielding a value of 11.3 Gy for a single fraction. Although the tumor effects were equal, the late effects (cu/p = 3 Gy) were 60% higher for the single HDR dose than for 20 Gy at LDR.

For both the LDR and HDR patients, several param- eters were used to assess response. These included change

The incidence of late effects can be lowered by multiple fractions of HDR. This may also permit reoxygenation

136 I. J. Radiation Oncology 0 Biology 0 Physics

J 0

Hemoptysis Chest Pneumonia Dyspnea Cough Pain

Symptoms

Fig. 1. Percent of lifetime rendered symptom-free or improved.

and reassortment out of radioresistant phases of the cell cycles, although the magnitude of these could not be quantified. It was practical to give 2 insertions per day on each of 2 successive days, with a minimum 6-hr interval. Solving equation (1) = equation (2) for n = 4 fractions, assuming a//3 = 10 Gy, yields d = 4.1 Gy per insertion for equal tumor effects. Utilizing equation (I), it was shown that the fractionation schema that would result in equal late toxicity would have d = 3.6 Gy per insertion with a Ti of 1.5 hr and 3.8 Gy for a T+ of 2 hr. As a

(4

Volume 23, Number 1, 1992

matter of practical consideration, the 4 Gy X 4 fractions schema was selected for clinical implementation. The model predicts a possible increase in long-term compli- cations by a factor of 1.2 with this scheme compared to the LDR technique.

RESULTS

While the two groups are not entirely comparable, the primary purpose of this study was to establish the feasi- bility of hyperfractionated HDR therapy and evaluate any significant difference between HDR and LDR brachy- therapy. For the LDR group, the median follow-up was 6.5 months with an actuarial median survival of 5.5 months and actuarial 12 month survival of 1%. For the HDR group, the median follow-up was 4.5 months with an actuarial median survival of 4 months and actuarial 12 month survival of 7%.

In the LDR group, 78% of the 203 symptoms resolved or improved. Hemoptysis and pneumonia resolved or improved in 9 1% and 82%, respectively, followed by dys- pnea (79%) chest pain (73%) and cough (68%). In the HDR group 79% of the 82 symptoms resolved or im- proved. Hemoptysis resolved in lOO%, chest pain and dyspnea resolved or improved in 75%, cough in 73%, and pneumonia in 7 1%. The overall ECOG PS improved from

0.3 Fig. 2. Pre-implant (a) and postimplant (b) chest roentgenogram showing an example of complete reaeration.

HDR and LDR ERT 0 M. MEHTA et al. 137

Table 3. Complications of endobronchial irradiation

Time XRT 1CM 2CM Type (mo.) (Gy) Laser (Gy) (Gy) Location

l* TVF 3.5 63 N 32 16 LMS 2 TVF 5 72 N 43 22 RMS 3 TVF 6 60 YX4 44 20 RMS 4 TVF 9 67 N 50 20 RMS 5 TVF 9 50 Y 47.1 20 Carina 6 TEF 23 72 N 42.5 20 LMS 7 TEF 3 55 Y 58 24 LMS

* HDR patient. TVF = Tracheovascular fistula; LMS = Left main stem bron-

chus; TEF = Tracheoesophageal fistula; RMS = Right main stem bronchus.

2.2 to 1.8 for Sl, and 2.1 to 1.6 for S2. When measured as an average parameter, approximately 54% of the pa- tients’ lifetime was rendered symptom-free or symptom- improved for Sl, and 57% for S2 (Figure 1).

Fifty-two of 66 patients in S 1 and 14 of 3 1 patients in S2 had pre-therapy atelectasis or lung collapse, and there- fore were evaluable for radiographic response. The overall reaeration rate was 79% (g) in Sl and 85% (2) in S2. A partial reaeration rate of 42% and complete reaeration rate of 37% was seen in the LDR group. For the HDR group, these values were 28% and 57% (Table 2). Figure 2 illustrates a typical example of complete reaeration fol- lowing endobronchial radiation.

Post-implant pulmonary function studies and bron- choscopy were not performed on all patients principally because of patient reluctance to undergo more procedures in their advanced disease state. Of the 14 patients who underwent pulmonary function testing, the average FEV 1 improved from 1.5 to 2.1 L, and the average FVC im- proved from 2.3 to 2.9 L. Of the 15 patients undergoing post-implant bronchoscopy, 11 had a complete broncho- scopic response with no evidence of endobronchial tumor. These 15 patients came from the LDR series and represent our early effort at quantifying the one month post-implant response.

Analysis of complications is shown in Table 3. In our series of 101 ERT implants, 5 tracheovascular fistulae (TVF) and 2 tracheoesophageal fistulae (TEF) were en- countered. Any patient dying of unexplained acute he- moptysis was scored as a TVF secondary to radiation, whether proven by autopsy or not. Four TVF occurred in mainstem bronchial lesions and one in a carinal lesion. No fistulae were seen in any of the upper lobe lesions. Three of the 7 fistulae occurred in patients who had un- dergone laser therapy.

DISCUSSION Several therapeutic modalities are currently available

for malignant airway occlusion. These include external beam radiation (4, 8, 22), laser therapy (16, 24) and en-

dobronchial radiation (1, 7, 9, 10, 13, 18-2 1,23). To our knowledge, no randomized study has evaluated these mo- dalities prospectively.

External beam radiation alone achieves successful re- versal of atelectasis and pneumonitis ranging from 2 1% (4) to 61% (8). In the largest reported series, Slawson et al. (22) reported only a 23% rate of improvement in at- electasis following conventional external beam radiation in 330 patients. In Chetty et al. study (4) all of the patients who responded received more than 50 Gy, whereas no patient receiving less than 50 Gy had a favorable response. Even if patients reaerate with external beam radiation, the time required to achieve this is generally longer than ERT. Since a fair number of patients present with met- astatic disease, an immunocompromised state, or a short median survival, waiting several weeks to achieve a pal- liative result is not optimal and frequently not feasible.

Although laser therapy achieves an immediate response, the duration of response is likely to be short since a con- siderable amount of tumor, both intrinsic and extrinsic, is not treated. A recent study (4) suggested that when compared with external beam radiation, “faster palliation with fewer side effects is probably achieved with laser therapy,” but no supporting data were presented.

Endobronchial radiation is a very successful technique that delivers high dose radiation over a short interval, with few complications and a rapid, sustained response. Our series, in addition to others, supports the efficacy of ERT (1, 7, 9, 10, 12, 13, 18-21, 23). Roach et al. (18) have reviewed the current literature regarding response rates for LDR and HDR ERT. The results of eight series (1, 7, 9, 13, 18-20, 23) demonstrated 67 to 100% symp- tomatic improvement with LDR ERT and 70 to 94% with HDR therapy. These two treatment modalities are there- fore highly comparable in terms of symptomatic relief, radiographic reaeration, and percent of lifetime rendered symptom-free.

Other studies ( 1, 19) have combined endobronchial ra- diation and laser excision with comparable results. Al- though we initially used the Nd:YAG laser in 12 patients, we have abandoned this technique because of the need for general anesthesia, possible risk of intraoperative death,

Months

Fig. 3. Actuarial survival.

138 I. J. Radiation Oncology 0 Biology 0 Physics Volume 23, Number 1, 1992

apparent lack of impact on extrabronchial disease, and lack of proven superiority to endobronchial radiation alone.

The recent advent of high dose rate remote afterloaders has greatly simplified endobronchial radiation and made it very safe for personnel. This modality may be used in an outpatient setting resulting in considerable financial savings. While most of our initial HDR patients were treated as inpatients, the last 10 patients have been treated on an outpatient basis. The six French endobronchial catheter is left in place for two days and patients tolerate this with minimum discomfort. When necessary, codeine phosphate 15-30 mg p. o. q. i.d. is used as an anti- tussive. The catheter is firmly taped at the nostril and marked. Additionally, the dummy-seed catheter is used to measure the distance from the exposed catheter tip to the nostril and verified prior to each treatment. Where necessary, check radiographs or fluoroscopy is carried out. No catheter displacement has occurred. We question the use of multiple bronchoscopies in this patient population. In our series, the median survival for the LDR patients was 5.5 months and 4.0 months for the HDR patients (Figure 3). The shorter median survival for the HDR group is consistent with the fact that 74% HDR patients presented with metastasis whereas only 47% presented with metastasis in the LDR group. Since this procedure is almost always palliative, subjecting patients to multiple bronchoscopies seems less than optimal because of patient discomfort, expenditure, and quality of remaining life. This view is strengthened by the fact that our data suggest that a single bronchoscopy may suffice.

with HDR ERT with recurrences in the upper lobes. In their analysis of 38 patients, all had previous external beam radiotherapy greater than or equal to 50 Gy. The dose calculated at 1 cm was 6 Gy X 3 fractions over a 3-week period. In our group of HDR patients (N = 31), the risk of fatal hemoptysis was 3% (&). The location of this tumor was in the left main stem bronchus (Table 3). In our series, upper lobe lesions were present in 16% ($) of the HDR patients and 3 1% ($$) of the LDR patients. Fistulae did not arise from any upper lobe lesions. Our data do not concur with Bedwinek’s conclusion that HDR brachy- therapy may be associated with a high hemoptysis risk in upper lobe lesions. On the contrary, our data suggest that both LDR and HDR ERT are safe in most situations and locations, and that upper lobe lesions do not carry an increased risk of fatal hemoptysis. Note that hemoptysis as the terminal event is not an uncommon occurrence in patients with lung cancer, even without implantation.

Bedwinek et al. (2) recently reported a 46% ($$) inci- dence of fatal hemoptysis in their series of patients treated

In our present series, a large majority of patients pre- sented with metastatic disease. They were treated pallia- tively, effectively, and quickly with minimal inconve- nience and morbidity. We therefore conclude that en- dobronchial radiation is a highly effective palliative modality that provides excellent control of local symptoms for a significant duration of a patient’s remaining life. HDR ERT also offers the advantage of decreased person- nel exposure and may be less expensive if done in an outpatient setting without multiple bronchoscopies. In patients who present initially with endobronchial and lo- cally advanced disease, the role of an initial endobronchial boost followed by conventional radiotherapy is currently being investigated by our group.

REFERENCES

1.

2.

3.

4.

5.

6.

I.

Allen, M. D.; Baldwin, J. C.; Fish, V. J.; Goffinet, D. R.; Cannon, W. B.; Mark, J. B. D. Combined laser therapy and endobronchial radiotherapy for unresectable lung carcinoma with bronchial obstruction. Am. J. Surg. 150: 7 I-77; 1985. Bedwinek, J.; Petty, A.; Bruton, C.; Solfield, J.; Lee, L. High dose rate endobronchial brachytherapy and fatal pulmonary hemorrhage (Abstr.). Int. J. Radiat. Oncol. Biol. Phys. 19: 161; 1990. Boring, C. C.; Savires, T. S.; Tong, T. Cancer statistics, 199 1. 41: 19-36; 1991. Chetty, K. G.; Sassoon, C. S. M.; Viravathana, T.; Light, R. W. Effect of radiation therapy on bronchial obstruction due to bronchogenic carcinoma. Chest 95: 582-584; 1989. Cox, J. D.; Yesner, R.; Mietlowski, W.; Petrovich, Z. Influ- ence of cell type on failure pattern after irradiation for locally advanced carcinoma of the lung. From the Veterans’ Ad- ministration Group (VALG). Cancer 44: 94-98; 1979. Dale, R. G. The application of the linear-quadratic dose- effect equation to fractioned and protracted radiotherapy. Br. J. Radiol. 58: 515-528; 1985. Macha, N. N.; Mai, J.; Stadler, M.; Koch, K.; Loddenkem- per, R.; Drumhaar, D.; Schumacher, W. Neu wege der strahlentherapie des bronchialkazinomas. Dtsch. Med. Wschr. 111: 687-691; 1986.

8. Majid, D. A.; Lee, S.; Khushalani, S.; Seydel, M. The re-

9.

10.

11.

12.

13.

14.

sponse of atelectasis from lung cancer to radiation therapy. Int. J. Radiat. Oncol. Biol. Phys. 17: 847-851; 1989. Mehta, M. P.; Shahabi, S.; Jarjour, N. N.; Kinsella, T. K. Endobronchial irradiation for malignant airway obstruction. Int. J. Radiat. Oncol. Biol. Phys. 17: 847-851; 1989. Mehta, M. P.; Shahabi, S.; Jarjour, N. N.; Steinmetz, M.; Kubsad, S. Effect of endobronchial radiation therapy on malignant bronchial obstruction. Chest 97: 662-665; 1990. Minna, J. D.; Higgins, G. A.; Glatstein, E. J. Cancer of the lung. In: DeVita, V. T., Hellman, S., Rosenberg, S. A., eds. Cancer principles and practice of oncology, 2nd edition. Philadelphia: JB Lippincott Co.; 1985: 5 18. Moylan, D.; Strubler, K.; Unal, A. B.; Mohiuddin, M.; Giampetro, A.; Boon, R. Transbronchial brachytherapy of recurrent bronchogenic carcinoma: a new approach using flexible fiberoptic bronchoscope. Radiology 147: 253-254; 1983. Nori, D.; Hilaris, B. S.; Martini, N. Intraluminal irradiation in bronchogenic carcinoma. Surg. Clin. N. Am. 67: 1093- 1102; 1987. Oken, M. M.; Creech, R. H.; Tormey, D. C.; Horton, J.; Davis, T. E.; McFadden, E. T.; Carbone, P. P. Toxicity and response criteria of the Eastern Cooperative Oncology Group. Am. J. Oncol. (CCT) 5: 649-655; 1982.

HDR and LDR ERT 0 M. MEHTA et al. 139

15. Perez, C. A.; Stanley, K.; Grundy, G.; Hanson, W.; Rubin, P.; Kramer, S.; Brady, L. W.; Marks, J. E.; Perez-Tamayo, R.; Brown, G. S.; Concannon, J. P.; Rotman, M. Impact of irradiation technique and tumor extent in tumor control and survival of patients with unresectable non-oat cell car- cinoma of the lung. Cancer 50: 1091-1099; 1982.

16. Personne, C.; Colchen, A.; Leroy, M.; Vourch, G.; Toty, L. Indications and technique for endoscopic laser resections in bronchology. A critical analysis based upon 2,284 resec- tions. J. Thorac. Cardiovasc. Surg. 91: 7 10-7 15; 1986.

17. RTOG Data Managers Handbook. Philadelphia, PA: Ra- diation Therapy Oncology Group Headquarters; 1983: 4.12- 4.38.

18. Roach III, M.; Leidholdt, Jr., E. M.; Tatera, B. S.; Joseph, J. Endobronchial radiation therapy (EBRT) in the manage- ment of lung cancer. Int. J. Radiat. Oncol. Biol. Phys. 18: 1449-1454; 1990.

19. Schray, M. F.; McDougall, J. C.; Martinez, A.; Cortese, D. A.; Brutinel, M. W. Management of malignant airway

compromise with laser and low dose rate brachytherapy. Chest 93: 264-269; 1988.

20. Seagren, S. L.; Harrel, J. H.; Horn, R. A. High dose rate intraluminal irradiation in recurrent endobronchial carci- noma. Chest 88: 8 10-8 14; 1985.

2 1. Shahabi, S.; Mehta, M. P.; Wiley, A. L.; Thomadsen, B. R.; Olson, M. H. The role of computed tomography in dosi- metric evaluation of endobronchial implants. Endocu- riether. Hyperther. Oncol. 4: 187-191; 1988.

22. Slawson, R. G.; Scott, R. M. Radiation therapy in bron- chogenic carcinoma. Ther. Radiol. 132: 175- 176; 1990.

23. Speiser, B.; Spratling, L. Intermediate dose rate remote af- terloading brachytherapy for intraluminal control of bron- chogenic carcinoma. Int. J. Radiat. Oncol. Biol. Phys. 18: 1443-1448; 1990.

24. Wolfe, W. G.; Sabiston, Jr., D. C. Management of benign and malignant lesions of the trachea and bronchi with the neodynium-yttrium-aluminum-garnet laser. J. Thorac. Cardiovasc. Surg. 9 1: 40-45; 1986.