Brain Metastasis in Resected Lung Cancer: Value of Intensive Follow-up With Computed Tomography

Transcription

1 Brain Metastasis in Resected Lung Cancer: Value of Intensive Follow-up With Computed Tomography Kohei Yokoi, MD, Naoto Miyazawa, MD, and Toshimoto rai, MD Divisions of Thoracic Surgery and Neurosurgery, Tochigi Cancer Center, Utsunomiya, Japan Background. Brain metastases are a common mode of recurrence in resected lung cancer and are usually associated with an ominous outcome. Methods. To assess the usefulness of follow-up using computed tomography of the brain for early detection and effective treatment of brain metastases, we prospectively studied 128 patients with completely resected non-small cell lung cancer. Follow-up computed tomographic scans were obtained every 2 to 6 months over 24 postoperative months in 69 patients and every 2 months for 6 postoperative months in 59. Results. Brain metastases were discovered in 11 patients (8.6%), and 7 patients were neurologically asymp- tomatic when the metastases were diagnosed. Single metastasis was found in 5 patients and multiple metastases in 6. The maximal size of all but one lesion was less than 25 mm. The median survival time and 5-year survival rate in all 11 patients with brain metastases were 10 months and 24%, respectively. Furthermore, those in 7 asymptomatic patients were 25 months and 38%, respectively. Conclusions. We consider intensive follow-up with computed tomography to be worthwhile for early detection and effective treatment of brain metastases in patients with completely resected lung cancer. (nn Thorac Surg 1996;61:546-51) fter curative resection for non-small cell lung cancer, about one third to two thirds of patients have recurrent disease depending on the stage [1-51. Brain metastases are a common mode of initial recurrence, and most cases are detected within the first 2 years after operation [6l. lthough single metastasis is occasionally treated with surgical resection and a good result is obtained [7-11], brain metastases are usually associated with an ominous outcome even after whole-brain irradiation [12]. However, these intracranial tumors are not usually diagnosed until patients become neurologically symptomatic, and little effort for early detection of the metastases has been performed. Computed tomography (CT) of the brain has a welldocumented accuracy in detecting metastatic tumors [13] and has been also reported to be of value in the preoperative staging of patients with locally advanced nonsmall cell lung cancer who were free of neurologic symptoms [14, 15]. To assess whether intensive follow-up with CT would result in an earlier diagnosis of brain metastases and thus lead to earlier and more effective treatment, we prospectively studied 128 patients who underwent complete resection of primary lung cancer. Patients and Methods Between January 1989 and September 1992, one hundred fifty-five patients with primary non-small cell lung cancer underwent surgical resection at our center. There ccepted for publication Oct 28, ddress reprints requests to Dr Yokoi, Division of Thoracic Surgery, Tochigi Cancer Center, Yohnan, Utsunomiya, Tochigi 320, Japan. were 19 patients with incomplete resection, 7 with operative or hospital death, and 1 who was lost to follow-up because of the poor performance status after operation. Excluding these 27 patients, 128 patients with complete resection constituted the study population (Table 1), and all gave informed consent for follow-up examination. Subjects consisted of 96 men and 32 women whose mean age was 64 years (range, 15 to 81 years). The surgical procedures were lobectomy in 102 patients, pneumonectomy in 12, bilobectomy in 11, and segmentectomy in 3. ll but 3 patients with segmentectomy underwent mediastinal lymph node dissection as a routine procedure [161. ll patients underwent CT of the brain before the initial operation and were certified to have no evidence of metastasis. The patients were prospectively studied by two chronologically different follow-up programs with brain CT. Follow-up CT scans were first obtained at the second postoperative month in all patients. Then, 69 patients in the early series were examined every 6 months from the sixth to 24th postoperative month and 59 patients in the late series every 2 months during the first 6 months after operation. Furthermore, when brain metastases were suspected by signs or symptoms and suspicious lesions were detected on CT, additional scans were performed more frequently. ll scans, including preoperative ones, were performed on a General Electric 9800 scanner (General Electric Medical Systems, Milwaukee, WI) with a 2-second scanning time. The brain was examined from the cranial base to the calvarium using 5- to 10-ram contiguous slices after intravenous injection of 50 ml of contrast material (iopamidol 300; Bracco Industria Chemica, Milan, Italy). The patients were also scheduled for follow-up exam by The Society of Thoracic Surgeons /96/$15.00 Published by Elsevier Science Inc SSDI (95)

2 nn Thorac Surg YOKOI ET L ;61: BRIN METSTSIS IN RESECTED LUNG CNCER Table 1. Patient Characteristics Early Series Late Series Total Characteristic (n = 69) (n - 59) (n = 128) Period of 1/1989-4/ /1989-9/1992 operation for primary tumor Histology denocarcinoma Squamous cell Large cell 3 3 denosquamous Carcinoid 2 2 Carcinosarcoma 1 1 Primary tumor stage I II Ili IIIB inations other than brain CT 1 month after operation, then every I to 3 months for 2 years, and every 6 months thereafter. Especially for the population at high risk of recurrence, checkups during the first 2 years were performed using combinations of periodic bronchoscopy, chest CT, abdominal CT or ultrasonography, and bone scintigraphy [17]. When brain metastases were discovered on CT, extracranial metastatic lesions were surveyed. Metastatic tumors of the brain were usually treated by surgical resection and radiotherapy for single lesions and radiotherapy alone for multiple lesions. Survival was measured from the date treatment of brain metastases was initiated in each patient. Results s of June 15, 1995, the median follow-up time was 39 months (range, 4 to 77 months). Of the 128 patients, 60 (46.9%) presented with evidence of recurrent lung cancer (Table 2). Brain metastases were observed in 11 patients (8.6%) as the first relapse site, 8 of whom were in the early series and 3 in the late series (Table 3). Of the 11 bronchogenic carcinomas with brain metastases, 7 were adenocarcinomas, 2 large cell carcinomas, 1 squamous cell carcinoma, and 1 adenosquamous carcinoma. Two were stage I disease and the other 9 were stage Ilia disease. Seven of the 11 patients were discovered to have brain metastases in the neurologically asymptomatic state. Metastatic tumors in 6 patients were detected by the scheduled follow-up CT. Patient 4 was examined by brain CT at the 23rd postoperative month because liver metastasis had been detected. The remaining 4 patients had neurologic signs or symptoms when brain metastases were found, and their metastases were diagnosed by an extra brain CT. Brain metastases were revealed within 12 months after operation in all but 1 patient with liver metastasis. Pa- tient 2 was suspected to have metastasis on CT at the sixth month, and patient 5, who was not examined at the sixth month, showed symptomatic metastases at the eighth month. Therefore, 9 of the patients were considered to have brain metastases by the sixth postoperative month. Of 7 asymptomatic patients, 4 had a single metastasis (Fig 1) and the other 3 had two or three metastatic lesions (Fig 2). Three of the 4 symptomatic patients had more than three metastatic tumors. The maximal diameters of brain metastases were less than 25 mm in all but 1 patient. Patients with a single metastasis were usually treated with surgical resection and radiotherapy. Patient 10 received whole-brain irradiation because the tumor occurred in the base of the frontal lobe and was nonresectable. ll patients with multiple lesions were treated with whole-brain radiation therapy. Complete responses were obtained in the 4 asymptomatic patients. Three of the 4 had a single metastasis and were treated with operation followed by radiotherapy with or without chemotherapy, and the other I received whole-brain irradiation for three small lesions. Partial responses were observed in 5 patients. The overall response rate was 81.8%. Three asymptomatic patients who had complete responses after treatment of brain metastasis survived more than 2 years. Furthermore, the 2 with a single metastasis each have remained well without recurrence, whereas the remaining 1 with multiple lesions had no further brain metastases before death. However, although the other 2 patients with asymptomatic tumors (patients 8 and 9) achieved good responses, they died of pneumonia during the treatment of brain metastases. The median survival time of all 11 patients was 10 months, and the 5-year survival rate was 24% calculated by the Kaplan-Meier method, whereas in the 7 asymptomatic patients, those were 25 months and 38%, respectively. Table 2. Mode of Recurrence in Patients With Resected Bronchogenic Carcinoma No. of Percentage Site of First Recurrence Patients of Patients Brain Brain only 10 Brain + liver 1 Lung Contralateral lung 15 Ipsilateral lung 2 Bone Other distant metastases Local recurrence Local and distant recurrences Total first relapses Total patients

3 548 YOKOI ET L nn Thorac Surg BRIN METSTSIS IN RESECTED LUNG CNCER 1996;61: C~ o E m m ~ o m m ~.~ ~'~== v o. o ~ c~r~ ~'= II o'~ ~,.~.r~ I ] l l l l l + m II.. m x Fig 1. (Patient 1.) Contrast-enhanced brain computed tomograms: () Before operation, no abnormality was found. (B) In the second postoperative month, a small enhanced mass measuring 15 mm in diameter was observed in the left parietal lobe (arrow). When &e metastasis was detected, the patient was neurologically asymptomatic and underwent surgical resection followed by focal radiotherapy and chemotherapy. He remains alive with no recurrence 67 months after treatment of the brain tumor. s II o~ ~,,-~ Comment Brain metastases are a common mode of initial treatment failure after resection of non-small cell lung cancer and occur in 6.2% to 20.3% of cases depending on the stage [3-6]. Unfortunately, most such recurrences are multiple

4 nn Thorac Surg YOKO1 ET L ;61: BRIN METSTSIS IN RESECTED LUNG CNCER Fig 2. (Patient 3.) Contrast-enhanced brain computed tomograms: () Before operation, no abnormality was seen. (B) In the sixth postoperative month, three lesions, each smaller than 15 mm, were demonstrated in the right parieto-occipital areas (arrow). t that time, the patient was neurologically normal and had whole-brain irradiah'on. He died of generally disseminated carcinoma 25 months after radiotherapy without further sign of brain metastases. and are treated by whole-brain irradiation, resulting in a short median survival time of less than 6 months [12]. few selected patients with single metastasis that usually has been detected after neurologic signs or symptoms have developed undergo surgical resections with or without radiotherapy and obtain a median survival time of 13 to 19 months and a 5-year survival rate of 13% to 28% [ Computed tomography has become the radiologic procedure of choice for the diagnosis of brain metastases. It has been reported that CT is valuable for preoperative staging of neurologically intact patients with locally advanced or non-squamous cell lung carcinoma because these patients have a high risk of metastases [14, 15]. From the results of follow-up studies [3-6], all patients undergoing curative operations were also considered a high-risk population for brain metastases. Demange and associates [9] reported, in a study of single brain metastasis from non-small cell lung cancer, that survival was better for patients in good neurologic condition at the time metastasis was discovered and thus emphasized early diagnosis. Therefore, we performed this prospective study of intensive follow-up with CT for early detection of recurrence. In our study brain metastases developed in 8.6% of patients with completely resected lung cancer; this recurrence rate was comparable with those of previous reports [3-6]. However, although metastases in previous reports were usually detected after neurologic symptoms developed, we found 63.6% of patients with brain metastases in the asymptomatic state on follow-up using CT. Moreover, the number of metastases in all asymptomatic patients was small and the maximal sizes of almost all lesions were less than 25 mm. The median survival time and 5-year survival rate of all 11 of our patients with brain metastases were comparable with those of reported results in surgically resected single brain metastasis [7-11]. Furthermore, those of 7 asymptomatic patients, despite the inclusion of patients with multiple lesions, were excellent. We believe that treatment results were improved not only by earlier detection of the metastases but also by improvement of the cure rate because 2 patients (patients 1 and 2) have survived without recurrence 67 and 42 months after treatment, respectively, and seem to be cured. Otherwise multiple metastases in patient 3 were well controlled until death and his quality of life was well maintained for a longer period. From our results, intensive follow-up with CT in patients with resected lung cancer is thought to facilitate earlier diagnosis of brain metastases and thus lead to the earlier and more effective treatment. However, there were some problems associated with the follow-up. One of the problems is that some asymptomatic patients had surgically inaccessible or multiple metastases. Moreover, notwithstanding this intensive follow-up, some patients were discovered to have recurrences after neurologic symptoms developed. Bindal and colleagues [18] recently reported that surgical treatment of multiple brain metastases in selected patients resulted in a significantly increased survival time and gave a prognosis similar to that of patients undergoing operation for single metastasis. Stereotactic radiosurgery has also been advocated as an effective treatment for surgically inaccessible or multiple

5 550 YOKOI ET L nn Thorac Surg BRIN METSTSIS IN RESECTED LUNG CNCER 1996;61: lesions [19]. Therefore, some unresectable brain metastases detected in this study were considered suitable for these new therapeutic modalities because of the small number and sizes, especially in asymptomatic patients. Magnetic resonance imaging has recently become available for diagnosing brain tumors [20], and contrastenhanced magnetic resonance imaging has been demonstrated as very sensitive for small lesions [21]. Follow-up with magnetic resonance imaging may be able to detect brain metastases in asymptomatic patients more frequently. The second problem is the duration and interval of follow-up with CT. The schedule was initially set up for the first 2 years after operation because most brain metastases are found within that period [6]. Eight patients in the early series were discovered to have brain metastases within the first 12 months, and in 7, except for patient 10, the metastases were considered to have developed by the sixth month. Consequently, in the late series, follow-up was performed more frequently during the first 6 months after operation. If the latter program had been applied to the entire study population, the same result except for patient 10 would have been obtained. ccordingly, follow-up CT every 2 months for the first 6 months after operation is considered sufficient for early detection. The cost of follow-up is the most controversial issue. This should be assessed not only in terms of the cost of examinations itself but also in terms of the improvement in quality of life, survival time, and cure rate in patients with brain metastases. From this perspective, a randomized trial comparing follow-up with and without CT is needed to assess the cost benefit. Furthermore, it is necessary that the population for such follow-up be selected based on the risk of brain recurrence. In our study no brain recurrence was found in patients with stage I or II squamous cell carcinoma. The Lung Cancer Study Group reported that patients with T1 NO squamous cell carcinoma had a very low risk of metastases [1, 6]. However, they also documented that patients with T1 NO non-squamous carcinoma and T2 NO or stage II squamous cell carcinoma had a higher probability of recurrence in the brain [1, 6]. Therefore, we recommend this follow-up using brain CT for all patients with resected lung cancer except for those with T1 NO squamous cell carcinoma. In conclusion, we consider that intensive follow-up with CT facilitates early detection and effective treatment of brain metastases in patients with completely resected lung cancer. Nevertheless, the ultimate value of this follow-up study can only be determined by controlled, randomized trials. We are grateful to the Divisions of Diagnostic Imaging and Radiation Therapy of Tochigi Cancer Center for their cooperation. References 1. The Lung Cancer Study Group (prepared by Thomas P, Piantadosi S). Postoperative T1 NO non-small cell lung cancer: squamous versus nonsquamous recurrences. J Thorac Cardiovasc Surg 1987;94: Pairolero PC, Williams DE, Bergstralh EJ, Piehler JM, Bernatz PE, Payne WS. Postsurgical stage I bronchogenic carcinoma: morbid implications of recurrent disease. nn Thorac Surg 1984;38: Martini N, Burt ME, Bains MS, McCormack PM, Rusch VW, Ginsberg RJ. Survival after resection of stage II non-small cell lung cancer. nn Thorac Surg 1992;54: The Ludwig Lung Cancer Study Group. Patterns of failure in patients with resected stage I and II non-small-cell carcinoma of the lung. nn Surg 1987;205: Yano T, Yokoyama H, Inoue T, et al. The first site of recurrence after complete resection in non-small-cell carcinoma of the lung: comparison between pn0 disease and pn2 disease. J Thorac Cardiovasc Surg 1994;108: Figlin R, Piantadosi S, Feld R, Lung Cancer Study Group. Intracranial recurrence of carcinoma after complete surgical resection of stage I, II, and III non-small-cell lung cancer. N Engl J Med 1988;318: Magilligan DJ Jr, Duvernoy C, Malik G, Lewis JW Jr, Knighton R, usman JI. Surgical approach to lung cancer with solitary cerebral metastasis: twenty-five years' experience. nn Thorac Surg 1986;42: Read RC, Boop WC, Yoder G, Schaefer R. Management of nonsmall cell lung carcinoma with solitary brain metastasis. J Thorac Cardiovasc Surg 1989;98: Demange L, Tack L, Morel M, et al. Single brain metastasis of non-small cell lung carcinoma: study of survival among 54 patients. Br J Neurosurg 1989;3: Burt M, Wronski M, rbit E, Galicich JH. Resection of brain metastases from non-small-cell lung carcinoma: results of therapy. J Thorac Cardiovasc Surg 1992;103: Macchiarini P, ngeletti C. Results and prognostic factors of surgery in the management of non-small cell lung cancer with solitary brain metastasis. Cancer 1992;70: Chatani M, Matayoshi Y, Masaki N, Inoue T. Radiation therapy for brain metastases from lung carcinoma: prospective randomized trial according to the level of lactate dehydrogenase. Strahlenther Onkol 1994;170: Lusins JO, Chayes Z, Nakagawa H. Computed tomography and radionuclide brain scanning: comparison in evaluating metastatic lesions to brain. NY State J Med 1980;80: Salbeck R, Grau HC, rtmann H. Cerebral tumor staging in patients with bronchial carcinoma by computed tomography. Cancer 1990;66: Kormas P, Bradshaw JR, Jeyasingham K. Preoperative computed tomography of the brain in non-small cell bronchogenic carcinoma. Thorax 1992;47: Yokoi K, Okuyama, Mori K, et al. Mediastinal lymph node metastasis from lung cancer: evaluation with T1-201 SPECT--comparison with CT. Radiology, 1994;192: Yokoi K, Miyazawa N, Mori K, Tominaga K. Intensive follow-up after curative resection for lung cancer. Haigan (Jpn J Lung Cancer) 1995;35: Bindal RK, Sawaya R, Leavens ME, Lee JJ. Surgical treatment of multiple brain metastases. J Neurosurg 1993;79: lexander E III, Moriarty TM, Davis RB, et al. Stereotactic radiosurgery for the definitive, noninvasive treatment of brain metastases. J Natl Cancer Inst 1995;87: Sze G, Shin J, Krol G, Johnson C, Liu D, Deck MDF. Intraparenchymal brain metastases: MR imaging versus contrast-enhanced CT. Radiology 1988;168: Russell EJ, Geremia GK, Johnson CE, et al. Multiple cerebral metastases: detectability with Gd-DTP-enhanced MR imaging. Radiology 1987;165:

6 nn Thorac Surg YOKOI ET L ;61: BRIN METSTSIS IN RESECTED LUNG CNCER INVITED COMMENTRY Yokoi and associates contend that the outcome of treatment of brain metastases from bronchogenic carcinoma has in the past been poor because of the delay in detection until symptoms have supervened. In their attempt to explore if routine postoperative scanning at frequent intervals would prove effective in early detection of such metastases they have embarked on computed tomographic scanning as the technique of choice. Today, however, they would probably choose contrastenhanced magnetic resonance imaging of the brain in preference because of the latter's proven higher sensitivity and total lack of radiation dosage. The availability of the technique would undoubtedly dictate its applicability. If one looks at using any imaging technique for routine repetitive scanning, one has to consider the likely positive yield as opposed to the cost to justify the cost logistics. It is well established that the cell type of the primary tumor to some extent determines the likelihood of metastases after curative operation, glandular and anaplastic carcinomas being more likely to metastasize than well-differentiated squamous carcinomas. Tumors in the upper lobes are more likely to metastasize to the brain than those in the lower lobes. Furthermore, tumors in stage II and IIIa are more likely to have thrown off microscopic metastases than tumors in stage I. If one looks at the patients in the series analyzed by Yokoi and associates, glandular and anaplastic carcinomas accounted for 10 of the 11 such cases of brain metastasis detected in a total of 76 patients with this cell type whereas there was only 1 well-differentiated squamous carcinoma with brain metastases in a total of 52 resected patients. Of all 11 patients in whom brain metastases developed, 10 had their primary tumors resected from the "upper lobes" if one considers the middle lobe as being developmentally and segmentally part of the upper lobe. Finally, 9 of 11 patients with metastases were in stage III at the time of resection and the other 2 had adenocarcinomas in stage I. Thus the evidence from this article shows that if one is to carry out repetitive postoperative scanning of patients who have undergone operation for non-small cell carcinoma of the bronchus, it is best done in the first 6 months at 2-month intervals in all patients who had "glandular or anaplastic" carcinomas of any stage in all "upper lobe" tumors. However, for squamous carcinomas the need for such screening can be limited to those who had their primary tumor resected in stage III. Had Yokoi and associates provided a breakdown of the numbers of each stage for individual cell type in the whole series studied, one may be able to refine the cost-effective applicability even further. Yokoi and associates are to be congratulated on undertaking the study and demonstrating that with meticulous search for asymptomatic brain metastases, prolongation of a good quality of life can be achieved even when the primary tumor was resected at a time when microscopic metastases may have occurred. Kumarasingham Jeyasingham, FRCS Department of Thoracic Surgery Frenchay Hospital Bristol BS16 1LE England