Combined external beam and low dose rate intraluminal radiotherapy in oesophageal cancer

Radiotherapy and Oncology, 27 (1993) 7-12 © 1993 Elsevier Scientific Publishers Ireland Ltd. All rights reserved. 0167-8140/93/$06.00

7

RADION01122

Combined external beam and low dose rate intraluminal radiotherapy in oesophageal cancer Robert

J . L . C a s p e r s a, A e i l k o H . Z w i n d e r m a n b, G e r r i t G r i f f i o e n c, K e e s W e l v a a r t d, E r i k N . S e w s i n g h a, J a c o b D a v e l a a r a a n d J a n - W i l l e m

H. Leer a

aDepartment of Radiotherapy, bDepartment of Medical Statistics, CDepartment of Gastro-Enterology and dDepartment of Surgical Oncology, University Hospital Leiden, The Netherlands (Received 26 February 1992; revision received 25 October 1992; accepted 3 November 1992)

Key words: Oesophageal cancer; Radiotherapy; Brachytherapy; Palliation; Survival

Summary Thirty-five patients with oesophageal cancer were treated with external beam irradiation (50-60 Gy) followed by a boost-dose of 15-20 Gy by means of low dose rate intraluminal brachytherapy. Of the 35 patients treated 17 (48%) were pretreated with laser therapy or dilation alone. Although the intraluminal application time was long (up to 36 h) the treatment was feasable with minor acute toxicity. The palliative effect of the combined treatment was excellent; at 6 weeks post-treatment 32 of the 35 patients were able to eat solid food. Late complications were seen in six patients (17%), of which only one was severe and probably treatmentrelated. The median survival was 11 months; the 1- and 2-year survival were 42% and 10% respectively. The survival was strongly dependant on local control. Distant metastases became evident in 23% of patients. The interval between external radiotherapy and brachytherapy seemed to be critical. The results were compared with 68 historical controls. A significantly better survival was observed at 6 months. It is concluded that low dose rate intraluminal brachytherapy is a useful and feasable technique to increase the total dose for obtaining a better local control. The shortcomings are discussed and ideas for further improvement are mentioned.

Introduction To ensure a passage for food in patients with oesophageal cancer is the main aim of palliation in such patients. In 1988 we reported the results of radiotherapy in patients with oesophageal cancer [1]. It was found that the palliative effect as well as the survival was correlated with the severity of dysphagia and the total dose given. Although the passage for food could be improved and kept acceptably good in 70% of the patients, about half of them suffered from local recurrence and signs of obstruction shortly before their death. On the supposition that a higher dose could lead to a better local control, we aimed for a treatment by which a higher dose could be given to the tumour without increasing the dose in the surrounding critical organs. Based on these assumptions we changed our treatment policy by increasing the dose in patients with favourable prognostic

factors by adding an intraluminal iridium application to the external radiotherapy. In this paper we present our experience with external beam irradiation combined with intraluminal low dose rate brachytherapy in the first 35 patients.

Patients and methods From September 1988 until July 1991 a total of 126 patients with complaints of dysphagia were seen at our out-patient clinic for cancer of the oesophagus and cardia. At this clinic patients are jointly seen by the surgeon (KW), radiotherapist (RC) and the gastroenterologist (GG). Thirty patients had a cardia-carcinoma, 86 patients had an oesophageal carcinoma and in 10 patients the dysphagia originated from other causes. Four patients with oesophageal cancer were treated elsewhere. The distribution of the 82 remaining patients with

Correspondence to: R.J.L. Caspers, Department of Clinical Oncology, University Hospital Leiden, K IP, Rijnsburgerweg 10, 2333 AA Leiden, The Netherlands.

8 TABLE I

Distribution of patients according to prescribed treatment. Bad performance or distant metastases Good performance, no distant metastases Major dysphagia Minor dysphagia > 70 years < 70 years

Endoscopic treatment

30

Laser + 50 Gy + Ir

17 14 21

60 Gy + Ir

Surgery

oesophageal cancer according to their performance and advised treatment is shown in Table I. Out of the 21 patients, judged to be candidates for surgery, four were shown to have unresectable disease at operation and were subsequently treated by high dose radiotherapy (60 Gy EBRT* + 15 Gy ILBT**). So a total of 35 patients were treated with an iridium-boost after an external dose of 50 or 60 Gy. Patient characteristics are shown in Table II. All patients were in good or moderate condition and most had minor weight-loss. Table III shows tumour characteristics. The majority of the tumours were located in the lower part of the oesophagus and were of the T2-category. Apart from the four patients who were found to have unresectable tumours (T3) at time of surgery, the TNM-classification (UICC 1978) was based on clinical grounds. As expected and consistent with our former study 40% of tumours were of the adeno-type. External beam irradiation was given by 4 ortho-

gonal fields with a 6- or 8-MV photon beam. The target volume was defined by the endoscopic tumour size with a 4-cm margin in the cranial and in the caudal direction. In the transverse direction at least a 2-cm margin was used. CT scans were performed in most patients to estimate the tumour thickness. The prescribed dose was normalized at the isocenter at the intersection of the beams corresponding to the centre of the target volume (ICRU 29). If applicable a correction was made for the lower density of the lungs. The treatment consisted of 20 fractions of 2.5 Gy/4 times per week or 30 fractions of 2.0 Gy/5 times per week. The mean external beam total dose was 53.9 Gy. The boost was given by an iridium wire of standard length 14 cm and a lineic activity of 4 - 8 mCi/cm (148-296 MBq/cm). For this purpose a nasogastric applicator tube with a 3-mm external diameter was specially developed which could be remotely afterloaded by means of the microSelectron * with extended channels. The dose was specified at 10 mm from the surface of the

T A B L E II TABLE III

Patient characteristics. Tumour characteristics. Sex

Male

Female Age (mean = 70.7 years)

W H O performance

Passage-score before therapy a Weight (mean = 72.8 kg) Weight-loss (kg)

<70 71-80 > 80 0,1 2 0,1 2,3 <70 >70 .~ 5 >5

No.

%

28 7 12 18 5 28 7 17 18 18 17 21 14

80.0 20.0 34.3 51.4 14.3 80.0 20.0 48.6 51.4 51.4 48.6 60.0 40.0

apassage-score: 0, none; 1, liquid; 2, semisolid; 3, solid. *EBRT: External Beam Radiotherapy. **ILBT: Intraluminai Brachytherapy.

Site (Level a) Size (Mean = 6.2 cm) Histopathology Squamous Adeno T-category

N-category

No.

%

1,2 3,4 _<5 >5

3 32 16 19

8.6 91.4 45.7 54.3

1 2 3 0

21 14 3 26 6 33

1

2

60.0 40.0 8.6 74.3 17.1 94.3 5.7

aSite according to UICC classification 1978: 1, cervical; 2, upper thorax; 3, middle thorax; 4, lower thorax. ¢Nucletron International B.V.

TABLE IV Palliative effect of treatment. Passage-score

Solid food Semi-solid food Liquid food No passage

Initial

After ETa

After RT

After BT

5 13 13 4

6 22 7 0

25 9 1 0

32 3 0 0

aET, Endoscopictreatment; RT, external beam radiotherapy; BT, intraluminal brachytherapy.

applicator tube, corresponding to 11.5 mm of axis [5,9]. The boost dose was 15 Gy in 28 patients. In the last seven patients the boost dose was 20 Gy. The dose rate varied from 37-112 cGy/h (mean 65.1 cGy/h). The mean total dose was 70.1 Gy (range 62-79 Gy). It was our policy to perform the application 2 - 3 weeks after the end of the external beam irradiation. However because the microSelectron was not always available for this type of treatment it was not always possible to have the boost dose within this period. The interval between the external radiotherapy and the brachytherapy therefore varied from 13 to 69 days with a median of 21 days. In 80% of the cases the interval was less than 30 days. All patients were followed with a 6-week interval and at each visit an X-ray of the oesophagus with barium contrast and a chest X-ray was made. The ability to swallow was scored by the passage-score: PS0, no passage; PSI, liquids; PS2, semisolid and PS3, solid food. The body weight was determined. Follow-up investigations by CT scan, ultrasound or endoscopy were only performed in the case of suspicion of recurrent disease. The duration of follow-up ranged from 3.3 to 34.5 months; the median follow-up was 10.0 months; mean 12.2 months. The survival plots were made by the method of Kaplan and Meier. All curves were calculated from the start of radiotherapy. The dysphagia-free interval was defined as the time the patient was able to use at least semi-solid food. The local recurrence-free period was calculated from the start of radiotherapy to the first sign of regrowth. The metastases-free interval was determined by the first occurrence of distant metastases. The results were compared with our previous experiences obtained in patients with comparable tumour and patient characteristics and treated with the same dose of external beam irradiation. To determine whether the extra intraluminal dose was responsible for the observed difference a conditional logistic regression model was used.

Results Table IV shows the improvement in the passage-score

after the subsequent treatments, e.g. endoscopic treatment (ET), external beam irradiation (RT) and brachytherapy (BT). Seventeen patients with major dysphagia were pretreated with an endoscopic treatment (ET): 7 patients had dilatation and in 10 patients laser therapy was performed. The external radiotherapy was started immediately after this endoscopic treatment. As is shown, the pretreatment resulted in an improvement of the passage-score in 10 of these 17 patients. So at the start of the external radiotherapy 28 of the 35 patients were already able to use semisolid or solid food (80%). The external radiotherapy resulted in a further improvement in 19 patients. Only 1 patient still had severe dysphagia. Nine patients had minor dysphagia and 25 patients were able to use solid foods. After brachytherapy again 6 patients had an improvement of their passage-score. After completion of the full treatment all but 3 patients were able to use solid food. The barium contrast films of the oesophagus showed improvement in all patients. Although in most patients a narrowing in the contrast-column was still observable, former irregularities had disappeared as was the prestenotic dilatation. During endoscopy before the start of the intraluminal brachytherapy tumour regression was also seen in all patients; in 25 patients a partial response ( > 5 0 % regression) and in 10 patients even a complete response. A biopsy was not obtained. Seventeen patients had a local recurrence and 7 had a regional recurrence causing recurrent dysphagia in 20 patients shortly before their death. Eight patients had signs of distant metastases. Figure 1 shows the overall survival (OS) together with the dysphagia-free period (DFP) and the local recurrence-free survival (LRFS). The median survival was 11 months; the 1- and 2-year survival rate was 42% and 10% respectively. It can be seen that the curves of the dysphagia-free period and the local recurrence-free survival are almost overlapping. Although 49% of the patients still had recurrent dysphagia due to local recurrence, the curves illustrate that this event took place only shortly (2-4 weeks) before they died, which means that most patients were very well palliated during the latter part of their lives.

10 1.0C

i

~, ~

OS

--I--

DFP

0.75

=~ 25 n~

o

0.50

0.25

0.00

i

i

t

=

26

52

78

104

weeks after start of radiotherapy

Fig. 1. Overall survival (OS), dysphagia-freeperiod (DFP) and local recurrence-free survival (LRFS).

In univariate analysis, age, sex, performance, weight, initial weight-loss, tumour site, tumour size, histopathology, TN category and pre-irradiation endoscopic intervention had no significant prognostic value with respect to survival, dysphagia-free interval and local recurrence-free survival. The only factors showing prognostic value were the initial passage-score and the interval between the external radiotherapy and brachytherapy. These results were compared with 68 historical controls from our previous study [1]. These 68 patients were treated by external radiotherapy only until a total dose of 50-60 Gy was reached. No significant differences existed between the two groups with respect to age, performance status, initial passage-score, tumour site, tumour size and histopathology (t-tests). The mean total dose expressed in TDF values was 70 in the control group and 99 in the present group (P < 0.0001). Figure 2

1.0~

^

0.7E

_~ ~o5c 0.25

•

0.00

J 26

/ 52

~ 78

l i 104

weeks after start of radiotherapy

Fig. 2. Overallsurvivalafter high dose radiotherapywith and without

brachytherapy boost.

shows the survival curves for both groups. Overall no significant difference was obtained. However it is suggested that in the first 45 weeks the combined treated group did better than the control group. At 6 months after treatment the difference reached significance (p = 0.012). With a conditional logistic regression model it was found that the addition of brachytherapy was the only significant factor with respect to this survival difference at 6 months (p = 0.010). Radiation oesophagitis was observed in 46% of the patients and severe oesophagitis was seen in 5 patients (14.3%). The oesophagitis in these 5 patients required hospital admittance lasting from 1 to 2 weeks. However the symptoms were self-limiting in all but 1 case. This patient had a former history of lye-burn; in his case the radiation induced ulcerations which took several months to heal. In 6 patients late complications were seen: in 3 of them dilatation was required at 9, 11 and 15 months post-treatment because of benign stenoses without signs of recurrent tumour growth. Three other patients had fatal complications: 2 because of oesophago-pleural fistula at 3 and 11 months post-treatment and 1 because of massive bleeding at 9 months after treatment. In 2 of these patients the fatal complication occurred in the presence of a true local recurrence; in the other patient there was no evidence of recurrent tumour growth. Discussion

Palliation, which means the preservation of the food passage, is the primary goal in the treatment of oesophageal cancer [1]. There are many ways to achieve this. A method with an almost immediate effect is endoscopic intervention: dilatation, lasertherapy, photoradiation and the insertion of one of various endoprotheses. These interv.entions are not without risk; perforations can occur in up to 10% of patients [12]. Others use intraluminal brachytherapy with high activity sources to achieve fast relief of dysphagia [3,17]. However the disadvantage of all these methods is that the duration of the achieved palliation is limited. We therefore only use these techniques in patients with a short prognosis of less than 3 months. With surgical resection long-lasting palliation can be obtained [14] and there is even a chance of total cure. A 3-year survival of 30% is reported in the literature [6,7,14]. The results of surgery are not totally representative because only patients with resectable tumours at operation are taken into account, which results in a positive selection. The resection, however requires a major operation with high morbidity and, depending on the experience of the surgeon, even with a substantial percentage of fatal complications (5-10%) [7,14]. The palliative effect of external radiotherapy has been addressed in our former study [1]. The response rate

11

measured as improvement of dysphagia was 70%. After 50-60 Gy total dose the median duration of the palliative response was 10 months. The median survival in this group was 8 months. We found that the prognosis was independently related to the initial passage-score and the radiotherapy dose. The survival was much more related to the passage-score than to the tumour size while the passage-score is probably correlated with the transmural tumour growth. Although the tumour can extend over a long distance in the longitudinal direction, tumour mass is probably better reflected by the transmural extension. A similar observation was done in patients with resected tumours [14]. Yang et al. have also demonstrated that the transmural extension of the tumour has prognostic significance, and not the tumour size [ 161. In our previous study we also observed that most patients still died from or at least with signs of local tumour recurrence. The applied dose had not been enough for persistent tumour control. To raise the dose without increasing the dose in the critical surrounding tissues, we decided to add intraluminal brachytherapy to the external beam irradiation. The main purpose of this study was to test the feasibility of this combination using low dose rate brachytherapy. Low dose rate brachytherapy has the advantage of probably being safer than high dose rate [2], but has the disadvantage that the application time is long; up to 36 h in our patients. Using a specially made 3-mm wide nasogastric tube the treatment was extremely well tolerated by the patients without the use of sedation. In addition, our first experience in 35 patients indeed showed a better survival in the first 45 weeks after treatment in the combined treated group as compared to the previously treated group. This is in agreement with the findings of Sur et al. [10]. As survival and local control are likely to be closely related this improvement could be explained by a better local control due to the high radiation dose. However our local control rates, although improved with higher dosages, are still less good than in surgically treated patients. Acute toxicity was moderate and self-limiting. Late stenosis, which could easily be treated by dilatation, was seen in three cases. Only one patient (3%) died of a

References Caspers, R. J. L., Welvaart, K., Verkes, R. J., Hermans, J. and Leer, J. W.H. The effect of radiotherapy on dysphagia and survival in patients with esophageal cancer. Radiother. Oncol. 12: 15-23, 1988. Fowler, J. F. The radiobiology of braehytherapy. In: Brachytherapy HDR and LDR; Proceedings Brachytherapy Meeting: "Remote afterloading: state of the art", 4-6 May 1989, Dearborn, MI, USA, pp. 121-137. Editors: A.A. Martinez, C.G. Orton and R.F. Mould. Nucletron Corporation, Columbia, USA, 1990.

fistula without signs of tumour regrowth. This case could be considered as a treatment-related death. Low dose rate brachytherapy combined with external beam irradiation therefore seems to be feasible and safe. We realise the limitations of our conclusions since more favourable patients were offered the boost dose and the control group was treated more than 10 years ago. The only proof that a high dose is really beneficial for these patients, can be given by a randomized trial. Such a trial was conducted by Yin et al. [17]. They randomized 128 patients between external radiotherapy alone (70 Gy) and external radiotherapy combined with brachytherapy (50 Gy + 20 Gy). A numerically better survival was found at 1, 3 and 5 years in the combined treated group. The local recurrence rate was 66% in the only externally irradiated group and 40% in the combined group. The observed differences in survival were however not significant. However there was no difference in the total dose given by the two treatment arms. The ultimate goal of obtaining local control for the rest of the patient's life could not be reached with the dose range we used. However although the patients still died from local disease they were very well palliated during most of their lifespan after treatment because the dysphagia-free period was almost the same as the overall survival. Further improvements can only be expected from a better patient selection and with perhaps higher irradiation dosages. We already increased the brachytherapy dose from 15 to 20 Gy. In the seven patients treated with 20 Gy no more side effects were observed. It may be possible to increase the dose to 30 Gy, but the brachytherapy dose will finally be limited by the tolerance of the oesophageal wall [4,15]. The other limitation of intraluminal brachytherapy is the dimensions of the treatment volume; this volume is cylindrical, with a diameter of only 2 cm. Tumours with a thickness of more than 2 cm can not be treated adequately with intraluminal brachytherapy. That is why we prefer to use this technique after external radiotherapy. For a better selection we also intend to make use of endosonography [8,11,13] before and after external radiotherapy to monitor the tumour regression.

3

Hishikawa, Y., Kurisu, K., Taniguchi, M., Kamikonya, N. and Miura, T. High-Dose-Rate lntraluminal Brachytherapy (HDRIBT) for esophageal cancer. Int. J. Radiat. Oncol., Biol. Phys. 21: 1133-1135, 1991. 4 Hishikawa, Y., Shinichi, T. and Takashi, M. Esophageal ulceration induced by intracavitary irradiation for esophageal carcinoma. Am. J. Roentgenol. 143: 269-273, 1984. 5 Marinello, G., Pierquin, B., Grimard, L. and Barret, C. Dosimetry of intraluminal brachytherapy. Radiother. Oncol. 23: 213-216, 1992. 6 Orourke, I. C., Johnson, D. C., Tiver, K. W., Bull, C. A., Langlands, A. O., Feigen, M., Gebski, V. and McNeil,

12

7

8

9 10

11

12 13

R. Long-term survival after radical treatment for cancer of the oesophagus. Med. J. Aust. 154: 219-220, 1991. Schattenkerk, M. E., Obertop, H., Mud, A. J. and et al. Survival after resection for carcinoma of the esophagus. Br. J. Surg. 74: 165-168, 1987. Siewert, J. R., Holscher, A. H. and Dittler, H.J. Preoperative staging and risk analysis in esophageal carcinoma. Hepatogastroenterology 37: 382-387, 1990. Speiser, B. Variability of dose specifications in endoluminal brachytherapy. Radiother. Oncol. 23: 211-212, 1992. Sur, R. K., Singh, D. P., Sharma, S. C., Singh, M. T., Kochhar, R,, Negi, P. S., Sethi, T., Patel, F., Ayyagari, S., Bhatia, S. P. S. and Gupta, B. D. Radiation therapy of esophageal cancer - Role of high dose rate brachytherapy. Int. J. Radiat. Oncol., Biol. Phys. 22: 1043-1046, 1992. Tio, T. L., Coene, P., Jager, F. and Tytgat, G. Preoperative TNM classification of esophageal carcinoma by endosonography. Hepatogastroenterology 37: 376-381, 1990. Tytgat, G. NJ. Endoscopic therapy of esophageal cancer - Possibilities and limitations. Endoscopy 22: 263-267, 1990. Vilgrain, V., Mompoint, D., Palazzo, L., Menu, Y., Gayet, B.,

14

15

16

17

Oilier, P., Nahum, H. and Fekete, F. Staging of esophageal carcinoma - - Comparison of results with endoscopic sonography and CT. Am. J. Roentgenol. 155: 277-281, 1990. Welvaart, K., Caspers, R. J. L., Verkes, R. J. and Hermans, J. The choice between surgical resection and radiation therapy for patients with cancer of the esophagus and cardia: a retrospective comparison between two treatments. J. Surg. Oncol. 47: 225-229, 1991. Yang, Z. Y., Hu, Y. H. and Gu, X.Z. Non-cancerous ulcer in the esophagus after radiotherapy for esophageal carcinoma - - a report of 27 patients. Radiother. Oncol. 19: 121-129, 1990. Yang, Z. Y., Hu, Y. H., Yan, J. H. and Gu, X . Z . Disappearance of local tumor of esophageal carcinoma after irradiation. Int. J. Radiat. Oncol., Biol. Phys. 10: 2067-2076, 1984. Yin, W. Brachytherapy of carcinoma of the esophagus in China, 1970-1974 and 1982-1984. In: Brachytherapy HDR and LDR; Proceedings of the brachytherapy meeting "Remote afterloading: state of the art", 1989, Dearborn, MI, USA, pp. 52-56. Editors: A.A. Martinez, C.G. Orton and R.F. Mould. Nucletron Corp., Columbia, USA, 1990.