Interstitial Irradiation for Unresectable Carcinoma of the Lung

Transcription

1 THE ANNALS OF THORACIC SURGERY Journal of The Society of Thoracic Surgeons and the Southern Thoracic Surgical Association VOLUME 20 NUMBER 5 NOVEMBER 1975 Interstitial Irradiation for Unresectable Carcinoma of the Lung Basil S. Hilaris, M.D., Nael Martini, M.D., Mostafa Batata, M.D., and Edward J. Beattie, Jr., M.D. ABSTRACT From 1963 to 1971, 105 patients with histologically proved cancer of the lung were explored at Memorial Hospital and underwent interstitial implantation using encapsulated sources of radon 222 (53 patients) or iodine 125 (52 patients). These lung cancers were considered unresectable because of extension of the disease into the mediastinum with fixation or invasion of the major vessels, trachea, and esophagus or chest wall involvement. No apical lesions, which have a better prognosis, are included in this review. Sixty-nine patients had epidermoid cancer, 24 had adenocarcinoma, and the remaining 12 had various other histological types. All patients were staged according to the criteria proposed by the American Joint Committee using the TNM definitions (standing for tumor, nodes, and metastasis). Local control was obtained in 8 of 10 patients (80% with clinical Stage I and I1 unresectable cancers of the lung and in 44 of the 95 (46%) with clinical Stage 111 lung cancer. The two-year survival was 50% for Stages I and I1 and 7% for Stage 111. Five patients have survived for five years or more. The complications, disease-free interval, local recurrences, distant metastases, and survival are presented and indications for this type of therapy outlined. I nterstitial implantation of radioactive material at thoracotomy was done for the first time in 1933 by Evarts Graham in the United States. After completion of a left pneumonectomy for bronchogenic carcinoma, he inserted seven radon seeds into the bronchovascular stump [41. In 1937 Ormerod [8] reported from England the implantation of radon seeds through a broncho- From the Departments of Radiation Therapy and Surgery, Memorial Sloan-Kettering Cancer Center, New York, N.Y. Supported in part by U.S. Public Health Service Grant no. EC-00113, Bureau of Radiological Health, Rockville, Md. Presented at the Eleventh Annual Meeting of The Society of Thoracic Surgeons, Montreal, Que., Canada, Jan , Address reprint requests to Dr. Hilaris, Department of Radiation Therapy, Memorial Sloan- Kettering Cancer Center, 1275 York Ave., New York, N.Y

2 HILARIS ET AL. scope in 67 patients with unresectable cancer of the lung; the longest survival at the time his article was published was four and a half years after treatment. The first interstitial implant for unresectable cancer of the lung was reported in 1941 by Binkley [2] from Memorial Hospital in New York City, using radon 222 (222Rn) seeds. From 1941 to 1955, lung implants were performed by the thoracic surgeons themselves when unresectable tumor was encountered. In 1955 at Memorial Hospital U. K. Henschke introduced iridium 192 (ls2ir), one of the artificial radionuclides. Since then, all implants have been done by the radiotherapy service. From 1941 to 1961, 272 patients at Memorial Hospital were treated at thoracotomy by interstitial implantation. Eight of these 272 patients (3%) survived five years or longer [3, 5, 61. Although the results of interstitial irradiation at thoracotomy have been encouraging, this form of treatment has been restricted to a very few institutions for two reasons: first, the technique of interstitial implantation is more difficult to learn than x-ray therapy and requires both knowledge of radiotherapy and surgical skill; second, interstitial implantation using high-energy radionuclides such as 222Rn, gold 198, and lg21r poses a radiation exposure problem to the operator as well as to the patient and everyone in contact with him. The introduction of low-energy iodine 125 ( 1251) gamma-ray-encapsulated sources in 1965 at Memorial Hospital removed some of these obstacles to interstitial irradiation [7]. Iodine 125 has a low-energy gamma emission (30 kv) and a long half-life (60 days). The long half-life allows these sources to be stored in the hospital and makes interstitial implantation available on short notice, simplifying the problem of source availability. The low gamma energy has resulted in a substantial reduction in radiation exposure. Technical refinements over the last decade have made interstitial implantation more accurate. Finally, the introduction of computers in medicine has improved our ability to determine the dose distribution within the implanted volume. Materials and Methods From 1963 to 1971 inclusive, a total of 594 patients with lung cancer underwent exploratory thoracotomy at Memorial Hospital. These represent 43% of all cases of primary lung cancer seen during that period. Resection of the tumor was carried out in 356 patients (60%); 105 (18% of the explored patients) had masses considered unresectable and were treated at thoracotomy by interstitial implantation. The remaining patients had a biopsy only and then were treated postoperatively by external irradiation, chemotherapy, or a combination of both. The 105 patients who had radioactive substances implanted at thoracotomy are the subject of this report. Fifty-two had 1251 implanted and 53 received 222Rn. No superior sulcus tumor is included in this review. There were 90 men and 15 women, a ratio of 6: 1. The age ranged from 32 to 8 1 years with an average of 69 years and a median of 63. Sixty-seven percent of the tumors (70/105) were epidermoid carcinoma, 19% (20/105) were adenocarcinoma, 6% (6/105) oat cell carcinoma, and 4% (4/105) bronchiolar carcinoma. There 492 THE ANNALS OF THORACIC SURGERY

3 Interstitial Irradiation fm Pulmonary Carcinoma was 1 patient with a giant and spindle cell carcinoma, 1 with a large cell anaplastic carcinoma, and 3 with unclassified carcinoma. All patients were staged retrospectively according to the clinical classification proposed by the American Joint Committee for Cancer Staging and End Results Reporting [ll. Six patients were classified in Stage I, 4 in Stage 11, and 95 in Stage 111. All patients were followed for at least three years after treatment. Patients were selected for interstitial irradiation only if there was a proved histological diagnosis of carcinoma of the lung. The majority of the 105 patients had disease still localized within the thorax that could not be removed by operation. Thirty-seven patients had extension of the disease into the mediastinum with fixation or invasion of the major vessels, the trachea, or the esophagus; 32 had either extensive chest wall involvement or pleural effusion; 24 had multiple matted hilar or mediastinal lymph node metastases; 10 with clinical Stage I or I1 lesions were considered poor surgical risks for resection because of their restricted pulmonary reserve or poor cardiac status; finally, 6 were known preoperatively to have multiple pulmonary lesions or a single distant metastasis and were explored solely for palliative implantation. The main objective of implantation was to treat the primary tumor and its gross extension, if any, into the mediastinum. External irradiation to the mediastinum or the primary site was added only if all gross tumor was not implanted, if there were positive mediastinal nodes, and if the implant dose distribution was considered unsatisfactory. The amount of external irradiation varied from 3,000 rads every three weeks to 4,000 rads in four weeks. It was given through opposing anterior and posterior fields with a supervoltage x-ray or cobalt gamma-ray beam. Ninety-four of the 105 patients had a diagnostic bronchoscopy. The tumor was seen and biopsied in 34% of the patients. Abnormal findings of distorted or compressed bronchi without tumor were noted in 36% of the patients; findings were normal in the remaining 30%. Implantation Technique All interstitial implantations at thoracotomy were done through a posterolatera1 incision, the usual approach when resection of a lung cancer is contemplated. This approach has been found very satisfactory, permitting good exposure of the tumor to be implanted and allowing inspection of the hilar area and mediastinum. It is unsatisfactory, however, if the tumor extends to the contralateral hilum. The implantation technique has been previously described in detail [5,6]; it consists basically of three steps. 1. The first is to determine the amount of radioactivity that should be implanted. This is done by estimating the three dimensions of the implanted volume, obtaining the average dimension, and multiplying it by an empirically derived factor, which for lz5i is 5. The derived value indicates the number of millicuries of lz5i required to deliver a minimum tumor dose within the implanted volume. VOL. 20, NO. 5, NOVEMBER,

4 HILARIS ET AL. 2. The second step is insertion of hollow 17 gauge stainless steel needles, 15 cm long, into the tumor mass. The needles are placed parallel to each other about 1 cm apart, starting from the periphery of the tumor and working toward the center. Care is taken to avoid the pulmonary vessels and bronchus if the hilum is being implanted. This is accomplished by palpating the needle tips with one hand under the hilum. If possible, the needles are placed in such a manner that a margin of about 1 cm beyond the tumor is maintained. 3. The third step is actual insertion of the radioactive sources into the tumor through each of the preinserted needles. Complications Five of the 105 patients (5%) died within the first 30 days after operation. Three of these patients had been implanted with Iz5I and 2 with 222Rn; the difference in mortality between these two groups is not statistically significant (Xz < 2.71). An additional 17 of the 105 patients (16%) developed early or late complications, some of the latter being the cause of death. Six of these patients had received lz51 (11% of 52 patients) and 11 had 222Rn (20% of 53 patients). This difference is not statistically significant (X < 2.71). The complications in the group of patients implanted with lz5i were septicemia (3 patients), empyema (2 patients), and lung abscess (1 patient). The complications in the group implanted with 222Rn were pulmonary hemorrhage (4 patients), empyema (2 patients), bronchopleural fistula (1 patient), lung abscess (2 patients), and transient esophagitis (2 patients). Results SURVIVAL Survival was calculated from the date of the exploratory thoracotomy, which corresponds closely to the date of diagnosis. Ten of the 105 patients (9.5%) have survived three or more years after implantation. Figure 1 shows survival according to the clinical stage of the disease. Stages I and I1 are grouped together because of the small number of patients in each category. The median survival for Stages I and I1 is 24 months compared with 6.5 months for patients with Stage I11 disease. The three-year survival is 40% for Stages I and I1 and about 7% for Stage 111. Figure 2 shows the overall survival according to location and size of the primary tumor. The median survival for intrapulmonary tumors less than 3 cm in diameter (TI) is 12 months; for intrapulmonary tumors 3 cm or greater in diameter (Tz), 8.5 months; and for the remainder of the lesions (T3), 6.5 months. The three-year survival is 17%, lo%, and 776, respectively. Figure 3 shows overall survival according to the status of the regional lymph nodes as determined clinically prior to thoracotomy. The median survival in the absence of lymph node involvement (No) is 8 months; with positive ipsilateral hilar nodes (Nl), 7 months; and with positive mediastinal nodes (N2), 6 months. The corresponding three-year survival is 17%, 8%, and 5%. 494 THE ANNALS OF THORACIC SURGERY

5 Interstitial Irradiation for Pulmonary Carcinoma loor FIG. 1. Survival according to clinical stage of disease prior to thoracotomy Years after implantation Figure 4 shows survival based on the presence (MI) or absence (M,) of distant metastases. The median survival is 7 months in both groups. The three-year survival in the absence of distant metastases is 10%; there were no three-year survivors when distant metastases were present. LOCAL CONTROL Tumor response was determined from chest roentgenograms taken every 2 months, or more often as indicated. Local control was defined as complete regression of the implanted pulmonary shadow with no evidence of regrowth or new pulmonary lesions at the time of the last follow-up or at death. Local control was obtained in 50% (52/105) of all implanted patients (Table 0 ) - e. a I I I VOL. 20, NO. 5, NOVEMBER,

6 HILARIS ET AL. 1 FIG. 3. Survival according to.status of regional nodes as determined clinically prior to thoracotomy. (No = no lymph node involvr- I 1 2 mat; N, = ipsilateral hilar node involvement; Years after implantation N2 = mediastinal node involvement.) 1). Tumors treated by implantation of lz5i showed local control in 63% (33/52) of the patients as compared with 36% (19/53) when 222Rn sources were used. This difference is statistically significant ( X2 = 6.93). Local control was obtained in 5 of the 6 patients with clinical Stage I disease, 3 of the 4 with clinical Stage 11, and 44 of the 95 with clinical Stage I11 unresectable cancer of the lung. In Stage 111 patients, implantation with lz5i resulted in better local control (26/44, or 59%) than implantation with 222Rn (18/51, or 35%). This difference is statistically significant ( X2 = 4.37). TREATMENT FAILURES An analysis of treatment failures by stage and site is shown in Table 2. Approximately one-fourth of all treated patients had persisting or recurrent - m >.- e - u) w Fn FIG. 4. Survival based on the presence or absence of distant metastases prior to thoracotomy. (M, = no distant metastases; MI = distant metastases pesent.) Years after implantation 496 THE ANNALS OF THORACIC SURGERY

7 Interstitial Irradiation fur Pulmonary Carcinoma TABLE 1. LOCAL CONTROL OF TUMOR IN 105 PATIENTS WHO RECEIVED RADIOACTIVE IMPLANTS No. of Patients Disease Treated Controlled Stage I 6 5 (83%) Stage I1 4 3 (75%) Stage (46%) Total (50%) disease in the primary site at the time of this report; 3% had nodal metastases and 39% had distant metastases with no evidence of disease in the primary site; 26% had disease in the primary focus as well as distant metastases. The remaining 8% had no evidence of disease. Distant metastases were detected during the course of the disease in 66 patients; the site of metastasis is known, however, in 49 patients. Table 3 lists the site of metastases in these 49 patients according to the frequency of their occurrence. Brain, liver, and bone were the most frequent metastatic sites. Comment In our institution less than half of all patients seen with untreated lung cancer are surgically explored, and many of these patients have unresectable tumors at thoracotomy. Interstitial therapy is a more effective method than postoperative external irradiation to treat the local disease in these unresectable cancers. There are four reasons for this. 1. A higher dose is delivered to the tumor by an implant than by external irradiation. 2. The radiation effect is precisely localized by inserting the radioactive source within the tumor. 3. The radiation dose is readily adaptable to the tumor shape and falls TABLE 2. TREATMENT FAILURES FOLLOWING IMPLANTATION OF RADIOACTIVE SOURCES No. of Failure (No. of Patients) Patients More Stage Treated T N M Than One I I Total T = primary tumor; N = regional or distant nodal metastasis; M = distant metastasis. VOL. 20, NO. 5, NOVEMBER,

8 HILARIS ET AL. TABLE 3. SITE OF DISTANT METASTASES IN 49 PATIENTS Site No. of Patients % Brain Liver Bone Spine Opposite lung Skin Eye ~ ~~ rapidly outside the implanted tumor; damage to surrounding normal lung is therefore less than with external irradiation. 4. The treatment time is much shorter; a permanent implant requires only a single procedure, while a complete course of external irradiation usually requires five to six weeks. Interstitial implantation, in our experience, has a greater curative potential in patients with small to moderate-sized lung tumors that are well defined and easily accessible so that they can be adequately implanted. Tumors larger than 6 cm in diameter, multiple lesions, or the presence of regional lymph node metastases preclude curative treatment by implantation; in selected cases, however, quick and effective palliation may be obtained by an interstitial implant. There is little to gain by implanting highly anaplastic tumors since these respond to moderate amounts of irradiation easily given by external beam. R efmences 1. American Joint Committee for Cancer Staging and End Results Reporting. A Clinical Staging System for Carcinoma of the Lung, Chicago, September Binkley, J. S. Role of surgery and interstitial radon therapy in cancer of superior sulcus of lung. Acta Union Znt Cancer 6:1200, Cliffton, E. E., Henschke, U. K., and Selby, H. H. Treatment of cancer of the lung by interstitial implantation. Cancer 11:9, Graham, E. A., and Singer, J. J. Successful removal of entire lung for carcinoma of bronchus. JAMA 101:1371, Henschke, U. K. Interstitial implantation in the treatment of primary bronchogenic carcinoma. Am J Roentgen01 79:981, Henschke, U. K., Hilaris, B. S., Mahan, G. D., and Wright, F. E. Interstitial Implantation of Radioactive Seeds during Thoracotomy. In W. L. Watson (Ed),Lung Cancer. St. Louis: Mosby, Pp Hilaris, B. S., Henschke, U. K., and Holt, J. C. Clinical experience with long half-life and low energy encapsulated radioactive sources in cancer radiation therapy. Radiology 91:1163, Ormerod, F. C. The pathology and treatment of carcinoma of the bronchus. J Laryngol Otol 52:733, THE ANNALS OF THORACIC SURGERY

9 Interstitial Irradiation for Pulmonary Carcinoma Discussion DR. DARRELL D. MUNRO (Montreal, Que., Canada): I find discussing this paper a very formidable task because of the techniques involved. Praise does wonders for the sense of hearing, and with this in mind, I sincerely commend the authors on their presentation. My job is all the more difficult because, as far as I know, our group in Montreal has absolutely no experience with interstitial irradiation for unresectable carcinoma of the lung, and I strongly suspect that most of the audience is in a similar position; in fact, the authors stress that this form of treatment has been restricted to a very few institutions, for reasons they have outlined. How, then, can one offer any constructive criticism if the experience with this form of treatment is very limited or nil? Justifiably, one cannot. Hence my commendation to the authors for presenting this supplementary form of treatment to us. It should stimulate our thinking. There is no subject, however complex, which, if studied with patience and intelligence, will not become more complex. I would suggest that we set aside materials, judgments, and implantation technique and go straight for the jugular : patient selection for this technique and the results of treatment. This paper concerns the majority of 105 patients with unresectable cancer of the lung still localized within the thorax. Of these, 24 had extensive regional node involvement and 6 were known to have multiple lesions or distant metastases; in the latter group there were no three-year survivors. Furthermore, external irradiation was added in some situations, including all instances of positive mediastinal nodes. These qualifications remove 30 patients, leaving 75 treated by implantation technique solely. May we assume that localized interstitial irradiation is very much less an immunosuppressing factor than external irradiation? If so, it seems reasonably logical to suggest that selection of these patients should satisfy the following requirements: (1) patients receiving interstitial irradiation alone should show evidence of an inherent ability to have their tumors controlled to localized areas of growth and invasion; and (2) patients so treated should have a relatively low total tumor load. The local irradiation technique used in conjunction with the operation will probably do very little to augment the recognized temporary immunosuppression from operation alone. Therefore, if you really examine the material presented in this paper, the patients are a highly selected group in whom localized invasive direct extension from the primary tumor for technical reasons precluded a curative resection. In other words, this technique is being employed in carcinoma of the lung in which there has been a particular balance of forces between the primary tumor and the patient host. The immunological interplay between a patient s disease process and his immune mechanisms now is looming large on the horizon, and one can t help but think that patient selection was therefore limited to those individuals whose tumor, for some unknown reason, had managed to stay confined to local extension and invasion without distant metastases. It follows, then, that if it were possible technically to resect such tumors, the prognosis would be improved, and that, failing this, the destruction of active malignant cells by other means would lead to some survival. The results that startled me are the 5 patients who have survived for five years or more and the 50% two-year survival for Stage I and I1 disease. You are well aware that carcinoma of the lung is one of the signal cancers of the National Cancer Institute, so the sharp end-point of death or survival really measures the efficacy of any treatment program. One wonders why these 5 patients lived and why there was a 50% survival for Stages I and 11. Can we honestly attribute these results to the implantation radiation effect, or is it possible that other factors, as yet unknown and not understood but suspected, were brought into play as a result of the interstitial irradiation technique? Many facts presented in this paper certainly stimulate interest, alert us to other possibilities, and, I would hope, produce intense discomfort in a great many of us who, contrary to the authors, continue to utilize obsolete, outdated, and inadequate methods of treating unresectable tumors. Despite all advances, lung cancer remains a national disaster, and we had better look to new techniques for some answers. The authors of this paper have attempted to treat the incurable. Most of us cure only those who are treatable. VOL. 20, NO. 5, NOVEMBER,

10 HILARIS ET AL. DR. MARTINI: I wish to thank Dr. Munro for his interesting remarks. The treatment of choice for localized cancer of the lung remains surgical resection. The purpose of our presentation today was to offer yet another form of treatment for the local control of lung cancer. Interstitial implantation with radioactive sources is now in its fourth decade of use at Memorial Hospital. It is not meant to substitute for adequate resection when resection is feasible. However, nearly one-half of the patients judged unresectable at the time of thoracotomy are considered so not because of massive tumor involvement, but because of the tumor s precarious location, making resection difficult or hazardous, or because of limitations imposed by the patient s marginal pulmonary reserve. It is in these instances that implantation becomes an invaluable form of treatment. The procedure adds, on the average, 45 minutes to one hour to the operating time. It does not increase hospital stay, and it converts an exploratory thoracotomy into a therapeutic procedure. When satisfactory implantation is carried out, no further therapy for local control may be necessary. When implantation is judged inadequate, supplementary external radiation therapy is given. For a disease that continues to be a leading cause of cancer death in men, we include another effective means of treatment in the armamentarium. 500 THE ANNALS OF THORACIC SURGERY