PII S0360-3016(99)00253-9CLINICAL INVESTIGATION EsophagusMULTI-INSTITUTIONAL RANDOMIZED TRIAL OF EXTERNALRADIOTHERAPY WITH AND WITHOUT INTRALUMINAL BRACHYTHERAPYFOR ESOPHAGEAL CANCER IN JAPAN T OMOHIKO O KAWA ,M.D.,*T AKUSHI D OKIYA ,M.D.,†M ASAMICHI N ISHIO ,M.D.,‡Y OSHIOH ISHIKAWA ,M.D.,§K OZO M ORITA ,M.D.,࿣AND J APANESE S OCIETY OF T HERAPEUTIC R ADIOLOGYAND O NCOLOGY (JASTRO)S TUDY G ROUP*Department of Radiology and Oncology,Tokyo Women’s Medical University,Tokyo,Japan;†Department of Radiology,Tokyo Medical Center,Tokyo,Japan;‡Department of Radiology,Sapporo National Hospital,Hokkaido,Japan;§Health and WelfareDepartment,Hyogo Prefectural Government,Hyogo,Japan;and ࿣Department of Radiation Oncology,Aichi Cancer Center,Aichi,JapanPurpose:With the aim of improving the results of treatment of esophageal cancer,we designed this multi-institutional,randomized trial to establish the optimal irradiation method in radical radiation therapy for esophageal cancer by clinically evaluating external irradiation alone and in combination with intraluminal brachytherapy.Methods and Materials:The study population consisted of patients with squamous cell carcinoma who were expected to be successfully treated with radical radiation therapy.The patients who could be given intraluminal brachytherapy at the end of external irradiation of 60Gy were stratified into 2groups.Patients assigned to receive external irradiation alone received boost irradiation of 10Gy/week on a schedule similar to the previous one,and with the same or smaller irradiation field.Intraluminal brachytherapy was performed,as a rule,with the reference dose point set at a depth of 5mm of the esophageal submucosa,and a total of 10Gy was irradiated at a daily dose of 5Gy,on a once-weekly schedule with low-dose-rate or high-dose-rate brachytherapy equipment.Results:A total of 103patients were registered,94of whom were analyzable,with 8ineligible,and 1for whom complete information was unavailable.The overall cumulative survival rate was 20.3%at 5years.The cause-specific survival rate was 31.8%at 5years.The cause-specific survival rate at 5years was 27%in the external irradiation alone group and 38%in intraluminal brachytherapy combined group.There was no significant difference between the 2groups (p ؍0.385).However,in the patients with 5cm or less tumor length,the cause-specific survival rate was 64%at 5years in the intraluminal brachytherapy combined group,which showed a significant improvement over 31.5%in the external irradiation alone group (p ؍0.025).In the patients with Stage T1and T2disease,cause-specific survival rates tended to be better in the intraluminal brachytherapy combined group than in the external irradiation alone group (p ؍0.088).In the patients with more than 5cm tumor length or Stage T3–4disease,there were no significant differences between the two groups by treatment methods (p ؍0.290).The incidence of early and late complications did not differ according to whether intraluminal brachytherapy was used.Conclusion:For the purpose of establishing the usefulness of intraluminal brachytherapy,further prospective randomized studies are necessary to evaluate the efficacy in tumors with short length and those with shallow invasion,or to assess the usefulness of intraluminal brachytherapy,as additional irradiation in large advanced tumors have been shown to have disappeared by diagnostic imaging after chemoradiotherapy with 60Gy/6w external irradiation.©1999Elsevier Science Inc.Esophageal cancer,Radiotherapy,Intraluminal Brachytherapy.INTRODUCTIONTo improve the results of treatment of patients with esoph-ageal cancer,it is important to achieve good local control.Because the esophagus is adjacent to highly radiation-sen-sitive organs such as the lungs,bone marrow,etc.,it is difficult to irradiate tumors with high doses.Although irra-diation techniques have improved as a result of advances in treatment planning equipment and irradiation equipment,there are still some patients in whom radical radiation therapy within doses that the bone marrow and lung can tolerate is difficult with external irradiation alone.On the other hand,intraluminal brachytherapy allows high-dose irradiation of esophageal cancer,with little exposure ofReprint requests to:Tomohiko Okawa,M.D.,Department of Radiology and Oncology,Tokyo Women’s Medical University,8-1Kwada-cho,Shinjuku Ku Tokyo,162-8666,Japan.Tel:03-3353-8111;Fax:035269-7355.Acknowledgments —We wish to express our sincere thanks to the medical institutions and physicians who kindly contributed to this investigation by registering cases and recording case report forms.Accepted forpublication 28April 1999.Int.J.Radiation Oncology Biol.Phys.,Vol.45,No.3,pp.623–628,1999Copyright ©1999Elsevier Science Inc.Printed in the USA.All rights reserved0360-3016/99/$–see front matter623adjacent organs at risk.However,intraluminal brachyther-apy,in which the dose sharply declines with the distance from the radiation source,is suggested to be a useful tech-nique for tumors with relatively shallow invasion.With the aim of improving the results of treatment of esophageal cancer,we designed this study to establish the optimal irradiation method in radical radiation therapy for esopha-geal cancer by clinically evaluating external irradiation alone and in combination with intraluminal brachytherapy.METHODS AND MATERIALSThe study population consisted of patients with squamous cell carcinoma who were expected to be successfully treated with radical radiation therapy.The patients entered into the study satisfied all of the following criteria as a rule:(a)patients who had received no prior treatment;(b)patients with squamous cell carcinoma which had primarily devel-oped in the intrathoracic esophagus,18–24cm from incisor;(c)the tumor was 10cm or less in length;(d)no distant metastasis;(e)aged 80years or less;(f)no serious compli-cation;(g)performance status (PS)was 3or less;(h)no active double/multicentric cancer;(i)patients whose pre-treatment laboratory test data fulfilled the normal condi-tions.We performed the pre-and post-treatment work-up of barium swallow,endoscopy,and CT scan in all cases,and in some cases,endoesophageal ultrasound (EUS).We used TNM (UICC,1987)for staging.The patients entered the study after informed consent to participate had been ob-tained from the patient or his or her family.We evaluated whether intraluminal brachytherapy (ILBT)was able to perform after 60Gy by barium swallow and endoscopic findings.Patients were allocated 2groups,external radiotherapy (ERT)alone or ERT with ILBT.About two-thirds of patients were entered into this trial at each institution.After the stratification,the central statistical board randomly assigned the patients to receive external irradiation alone or in combination with intraluminal brachytherapy,using a telephone call or facsimile;and treatment after an external irradiation of 60Gy was per-formed according to their assignment.After radiotherapy,maintenance chemotherapy with etoposide (25mg/day,3cycles every 2weeks)was performed in each institution (Fig.1).Radiation therapy was started with external irradiation with linac X-rays or 60Co ␥-rays,and the patients received,as a rule,an external irradiation equivalent to 60Gy/6weeks at a daily dose of 2Gy,on a 5-day weekly schedule.After 60Gy irradiation,either external irradiation or intralu-minal brachytherapy was performed according to the as-signment.Patients assigned to receive external irradiation alone received boost irradiation of 10Gy/week on a sched-ule similar to the previous one,and with the same or smaller irradiation field.Intraluminal brachytherapy was performed,with the reference dose point set at a depth of 5mm of the esophageal submucosa,because all of the institutions used this reference point.A total of 10Gy was irradiated at a daily dose of 5Gy,on a once-weekly schedule with low-dose-rate (LDR)or high-dose-rate (HDR)brachytherapy equipment.Definition of the HDR or LDR (50–150cGy/h)is according to ICRU Report 38.Ra and Cs were used in LDR and Ir was used in HDR.Diameter of the intra-esophageal brachytherapy applicator is 1cm with balloon,which surface touches the esophageal tumor surface after insertion.The survival rate was calculated by the Kaplan-Meier method,counting from the first irradiation day to May 1997,as the final follow-up day.The log-rank test was used to test for significant differences,and p Ͻ0.05was judged to be significantly different.The generalized Wil-coxon test was also used if necessary.Fisher’s exact prob-ability was used for testing between the two groups.RESULTSThis study was performed as a multi-institutional coop-erative study designated by The Japanese Society for Ther-apeutic Radiology and Oncology (JASTRO).The registra-tion of patients was conducted between May 1,1991and May 31,1995,at 22medical institutions in various regions of Japan.A total of 103patients were registered,94of whom were analyzable,with 8ineligible,having more than 10cm in tumor length and neoadjuvant chemotherapy be-fore radiotherapy cases,and 1for whom complete informa-tion was unavailable.The follow-up period ranged from 10to 82months,the median being 24months.The median follow-up period in the survivors was 48months,with 4alive for 5years or longer.The characteristics of the analyzable patients were as follows:73men and 21women,aged 41–83years (median 73),PS 0in 39patients,1in 34,2in 15,and 3in 6.Five patients exceeded 80years of age,but were included in the analysis because they fulfilled all of the other selection criteria.The main location of the lesions was upper thoracic in 11patients,middle thoracic in 59,and lower thoracic in 24,with middle thoracic accounting for 62.8%.The major axis of the lesion was 5.5cm in median length,ranging from 0.8to 10cm.In the TNM classification,the T-stage was T1in 20patients,T2in 25,T3in 36,T4in 10,and Tx in 3;N-stage was N0in 68patients,N1in 23,and Nx in 3.The response at the time of registration,(i.e.,the response to the external irradiation of 60Gy)was a complete response(CR)Fig.1.Randomization and treatment schedule.624I.J.Radiation Oncology●Biology●Physics Volume 45,Number 3,1999by barium swallow and endoscopicfindings in28patients, partial response(PR)in59,no change(NC)in4,and unknown in3.Fifty-one patients received external irradia-tion alone,and43received external irradiation with intralu-minal brachytherapy.The characteristics or response to the 60Gy external irradiation did not differ between the exter-nal irradiation-alone group and the intraluminal brachyther-apy combined group.(Table1).Among the94patients who completed radiation therapy, the response at1month after radiotherapy was evaluated as CR in49,PR in43,NC in1,and unknown in1;the response rate(CRϩPR)being97.9%,and the CR rate 52.1%.The median duration of response,taking all relapses and metastases into consideration by every1-month fol-low-up study,was134days(27–971days).Overall cumulative survival rate with actuarial method, and cause-specific survival rate in the analyzable patients are presented in Fig.2.Cause-specific survival rate was corrected by handling the14patients who died of other diseases under no-cancer-bearing conditions as“discontin-ued”at the time of death.The overall cumulative survival rate was32.9%at2years and20.3%at5years.The cause-specific survival rate was40.7%at2years and31.8% at5years.The complete response rate at the completion of irradia-tion was49%for external irradiation alone and55.8%for intraluminal brachytherapy combined.Intraluminal brachy-therapy made no difference in tumor response(pϭ0.327).The cause-specific survival rate was27%in the external irradiation alone group and38%in intraluminal brachytherapy combined group.There was no significant difference between the two groups(pϭ0.385,Fig.3). However,in the patients with5cm or less tumor length,the cause-specific survival rate was74.6%at2years and64% at5years in the intraluminal brachytherapy combined group,which showed a significant improvement over the 39.4%and31.5%in the external irradiation alone group(pϭ0.025,Fig.4).In the patients with Stage T1and T2 disease,the relapse-free and cause-specific survival rates tended to be better in the intraluminal brachytherapy com-bined group than in the external irradiation alone group (pϭ0.088,Fig.5).In the patients with more than5cm tumor length or stage T3-4disease,there were no significant differences between the two groups by treatment methods(cause specific:pϭ0.290,relapse-free:pϭ0.743).In the combined group of external irradiation and intralu-minal brachytherapy,28institutions used LDR and15HDR brachytherapy.The results of tumor response,survival rate, and complications were not different in the two brachyther-apy types.The complications that occurred from start of treatment to 6months after the completion of treatment were regarded as “early complications,”and those noted thereafter,as“late complications.”Special attention was paid to pain,nausea/ vomiting,appetite loss,and general fatigue.All types of complications were counted in patients with more than one complication during the observation period,and the grade included in the analysis was the most severe one during that period.Findings clearly judged to result from the aggrava-tion of the underlying disease or from other complications were not included.In the early period,pain,nausea/vomit-ing,appetite loss,and general fatigue were recorded as subjective manifestations.Early complications of grade3or more were found in8patients,with an incidence of5.9% for external irradiation alone and11.6%for intraluminal brachytherapy combined.The incidence of early complica-tions did not differ between the two groups(pϭ0.266). Late complications of Grade3or more were found in8.9% of the patients receiving external irradiation alone(ulcer2, pneumonitis1)and8.3%of those receiving combined in-traluminal brachytherapy(ulcer2,stenosis1),showing no difference according to the irradiation method(pϭ0.625).Two patients underwent an operation for esopha-geal stenosis after radiation therapy,one treated with exter-nal irradiation alone and the other treated with intraluminal brachytherapy combined.Table1.Patient characteristics according to external irradiation with/without intraluminal brachytherapy Characteristic ERT ERTϩILBT Total no.5143 Male:Female40:1133:10 Median age(years)7275 Range41–8350–82 Performance status02118 12113 269 333 Tumor locationLU65LM3029EI159 Tumor lengthMedian 6.35 Range0.8–102–10Ϲ5cm2420Ͼ5cm2723 TNM stageT1137T21213T31917T455Tx21N03830N11310Nx03 Effect at60GyCompleteresponse1414 Partial response3326No change13N/A30 ERTϭexternal radiotherapy;ILBTϭintraluminal brachyther-apy;IUϭupper intrathoracic esophagus;IMϭmiddle intratho-racic esophagus;EIϭlower intrathoracic esophagus;N/Aϭcomplete information not available.625RadiotherapyϮintraluminal brachytherapy for esophageal cancer●T.O KAWA et al.These results showed that the incidence of early and late complications did not differ whether intraluminal brachy-therapy was used.DISCUSSIONIn the treatment of unresectable advanced esophageal cancer,radiation therapy is reported to have a 5-year sur-vival rate of about 10%(1,2).Such unfavorable treatment results may be caused by the inclusion of patients who have large tumors,distant metastases,or poor medical condition in the cases of indication for radiotherapy.In our prospec-tive randomized study,we excluded patients with distant metastases,stratified the enrolled patients in advance by prognostic factors such as sex and tumor length,and then assigned the patients who could receive intraluminal brachytherapy after 60Gy external irradiation to either of the two treatment groups.Intraluminal brachytherapy as part of the radiation therapy for esophageal cancer has been reported to be useful,because its combined use results in the improvement of primary response and survival rate (3–6);however,all of these reports are retrospective studies.In addition,because these studies included all patients having CR or PR after ERT,and who could receive intraluminal brachytherapy,it is undeniable that the selection of patients was greatly biased,which makes it difficult to simply eval-uate the treatment results compared with external irradiation cases.In other words,it seems likely that favorable resultswere obtained,because the subjects were limited to those who responded to external irradiation and who could be equipped with an applicator for intraluminal brachytherapy.The statistical analysis of all of the patients enrolled in our study revealed no significant differences in the primary response at the end of treatment,or the survival rate be-tween the external irradiation alone group and the intralu-minal brachytherapy combined group.However,in rela-tively small tumors with shallow invasion (tumor length of 5cm or less,T1or T2stage),the intraluminal brachyther-apy combined group yielded significantly better ly,it is suggested that the tumor with short length and shallow invasion is a proper indication for intraluminal brachytherapy after external irradiation.This is because intraluminal brachytherapy,in which the dose sharply de-clines with the distance from the radiation source,enables a uniform irradiation of small tumors with shallow invasion but does not irradiate large tumors with deep invasion (7–9).Recently,it has been reported that in superficial (namely,stage T1)tumors,the 2-year survival rate was higher with intraluminal brachytherapy combined than with external irradiation alone (10),and that even intraluminal brachy-therapy alone is expected to be effective in small superficial tumors (11).In Japan,the following schedule has been generally used as the optimal irradiation method for radical radiation ther-apy in combination with intraluminal brachytherapy for esophageal cancer:after the patient shows a remarkable response as potentially CR or PR at 50ϳ60Gy external irradiation,intraluminal brachytherapy is employed as boost therapy by LDR irradiation at 10ϳ20Gy/2ϳ3fractions using the mucosa plane as the reference dose point (12,13).Hishikawa et al.have reported that HDR intraluminal brachytherapy following 60Gy external irradiation is at a high risk of causing fistula formation unless the dose is 20Gy or less (14).Yorozu et al.have reported that,in view of esophageal ulcer,HDR intraluminal brachytherapy is safe up to a total dose of 16Gy (4Gy/fraction,twice-weekly schedule)after 50Gy external irradiation (15).The high incidence of complications with intraluminal brachytherapy may be explained by the fact that the areas close to the radiation source are irradiated at an extremely high dose,while the dose sharply declines in some areas whichareFig.2.Overall survival curves of esophagealcancer.Fig.3.Cause-specific survival by treatmentmethod.Fig.4.Cause-specific survival (tumor length Ϲ5cm).626I.J.Radiation Oncology●Biology●Physics Volume 45,Number 3,1999distant,even if slightly,from the radiation source.Calling the area given more than twice as high dose as the reference dose point “hyperdose sleeve,”Marinello et al.have de-scribed that the tissue included in this area is at a high risk of developing radiation complications such as necrosis (16).In our study,8.3%of the patients developed late complica-tions of Grade 3or more,but the incidence did not differ between the external irradiation alone group and the intralu-minal brachytherapy combined group.This may be because “hyperdose sleeve”resulting from intraluminal brachyther-apy was limited,with the dose set at 10Gy in total (5Gy/fraction)at a depth of 5mm under the esophageal submucosa.As the definitive treatment for tumors 10cm or less in length,the guidelines presented by the American Brachytherapy Society suggest that after 45–50Gy irradia-tion,intraluminal brachytherapy at a total dose of 10Gy with two fractions in HDR or a single irradiation of 20Gy in LDR brachytherapy (17).As the palliation of advanced esophageal cancer,Sur et al.reported that dose of optimalbrachytherapy without external radiation ranges between 16Gy in two fractions and 18Gy in three fractions given a week apart (18).In our study of radiotherapy with curative intent,after 60Gy external irradiation,10Gy/2fractions,for both LDR and HDR,were added to make a total dose of 70Gy.As a result,intraluminal brachytherapy proved ef-fective for small and shallow tumors.Therefore,in small and shallow tumors,we consider it appropriate to achieve complete response using 50ϳ60Gy external irradiation and then performing intraluminal brachytherapy at 10ϳ15Gy/2ϳ3fractions.In this case,it is very important that the total dose should not exceed 70Gy.For large and deep tumors,(tumor length more than 5cm,T3or T4stage),for multi-modal treatment including chemotherapy,it is necessary to achieve complete tumor response and to improve the sur-vival rate.For the purpose of establishing the usefulness of intraluminal brachytherapy,prospective randomized studies seem necessary to evaluate the efficacy in tumors with short length and those with shallow invasion,or to assess the usefulness of intraluminal brachytherapy,as additional ir-radiation in large advanced tumors have shown to have disappeared by diagnostic imaging after chemoradiotherapy with 60Gy/6W external irradiation.Multidisciplinary stud-ies (19–23)on the combination of hyperfractionation radio-therapy and various anticancer drugs are also valuable.Finally,reference dose point is a problem to be solved in intraluminal brachytherapy.In general,the reference dose point is set at a depth of 5mm of the submucosa;however,some institutions use the mucosal surface or a certain dis-tance from the center of the radiation source.The interna-tional harmonization of reference dose point is urgently needed,because it is essential for comparison of treatment results between different institutions,and for conducting multi-institutional cooperative studies.REFERENCES1.Okawa T,Kita M,Tanaka M,et al .Results of radiotherapy for inoperable locally advanced esophageal cancer.Int J Radiat Oncol Biol Phys 1989;17:49–54.2.Nishio M,Morita K,Yamada T,et al .National survey of radiotherapy for esophageal cancer (in Japanese).J Jpn Soc Cancer Ther 1992;27:912–924.3.Hishikawa Y,Kurisu K,Taniguchi M,et al .High-dose-rate intraluminal brachytherapy for esophageal cancer:10years experience in Hyogo College of Medicine.Radiother Oncol 1991;21:107–114.4.Hareyama M,Nishio M,Kagami Y,et al .Intracavitary brachytherapy combined with external-beam irradiation for squamous cell carcinoma of the thoracic esophagus.Int J Radiat Oncol Biol Phys 1992;24:235–240.5.Caspers RL,Zwinderman AH,Griffioen G,et al .Combined external beam and low dose rate intraluminal radiotherapy in oesophageal cancer.Radiother Oncol 1993;27:7–12.6.Moni J,Armstrong JG,Minsky BD,et al .High dose-rate intraluminal brachytherapy for carcinoma of the esophagus.Dis Esophagus 1996;9:123–127.7.Poplin E,Fleming T,Leichman L,et al .Combined therapies for squamous-cell carcinoma of the esophagus,a SouthwestOncology Group Study (SWOG-8037).J Clin Oncol 1987;5:622–628.8.Seydel HG,Leichman L,Byhardt R,et al .Preoperative radi-ation and chemotherapy for localized squamous cell carci-noma of the esophagus:A RTOG Study.Int J Radiat Oncol Biol Phys 1988;14:33–35.9.Coia LR,Engstrom PF,Paul A.Nonsurgical management of esophageal cancer:Report of a study of combined ra-diotherapy and chemotherapy.J Clin Oncol 1987;5:1783–1790.10.Okawa T,Tanaka M,Kita-Okawa M,et al .Superficial esoph-ageal cancer:Multicenter analysis of results of definitive ra-diation therapy in Japan.Radiology 1995;196:271–274.11.Hishikawa Y,Kurisu K,Taniguchi M,et al .Small,superficial esophageal carcinoma treated with high-dose rate intracavitary irradiation only.Radiology 1989;172:267–270.12.Ikeda M,Ando C,Ishikawa T,et al .Guideline for radiother-apy in the treatment of esophageal cancer (in Japanese).Jpn J Clin Cancer 1987;33:1001–1019.13.Yin W.Radiotherapy of carcinoma of oesophagus in China.Chin Med J 1977;110:289–293.14.Hishikawa Y,Kurisu K,Taniguchi M,et al .High-dose-rateFig.5.Cause-specific survival (T1,2cases).627Radiotherapy Ϯintraluminal brachytherapy for esophageal cancer●T.O KAWA et al.intraluminal brachytherapy(HIDRIBT)for esophageal cancer.Int J Radiat Oncol Biol Phys1991;21:1133–1135.15.Yorozu A,Dokiya T,Ogita M,et al.Dose-response relation-ship of intracavitary irradiation with a new applicator follow-ing external beam irradiation for esophageal cancer(in Japa-nese).J Jpn Soc Cancer Ther1996;31:427–436.16.Marinello G,Pierquin B,Grimard L,et al.Dosimetry ofintralumina brachytherapy.Radiother Oncol1992;23:213–216.17.Gaspar RE,Nag S,Herskovic A,et al.American Brachyther-apy Society(ABS)consensus guidelines for brachytherapy of esophageal cancer.Int J Radiat Oncol Biol Phys1997;38:127–132.18.Sur RK,Bernard DM,Victor CL,et al.Fractionated high doserate intraluminal brachytherapy in palliation of advanced esophageal cancer.Int J Radiat Oncol Biol Phys1988;40:447–753.19.Calais G,Dorval E,Louisot P,et al.Radiotherapy with highdose rate brachytherapy boost and concomitant chemotherapyfor stages IIB and III esophageal carcinoma:Results of a pilot study.Int J Radiat Oncol Biol Phys1997;38:769–775. 20.Herskovic A,Marts K,Al-Sarraf M,et bined chemo-therapy and radiotherapy compared with radiotherapy alone in patients with cancer of the esophagus.N Engl J Med1992;326:1593–1598.21.Bates BA,Detterbeck FC,Bernard SA,et al.Concurrentradiation therapy and chemotherapy followed by esophagec-tomy for localized esophageal carcinoma.J Clin Oncol1996;14:156–163.22.Jeremic B,Shibamoto Y,Acimovic L,et al.Acceleratedhyperfractionated radiation therapy and concurrent5-fluorou-racil/cisplatin chemotherapy for locoregional squamous cell carcinoma of the thoracic esophagus:A phase II study.Int J Radiat Oncol Biol Phys1998;40:1061–1066.23.Poplin EA,Jacobson J,Herskovic A,et al.Evaluation ofmultimodality treatment of locoregional esophageal carci-noma by Southwest Oncology Group9060.Cancer1996;78: 1851–1856.628I.J.Radiation Oncology●Biology●Physics Volume45,Number3,1999。