|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2001 American Society for Clinical Oncology Factors Affecting the Risk of Brain Metastases After Definitive Chemoradiation for Locally Advanced Non–Small-Cell Lung CarcinomaFrom the Department of Radiation Oncology and the Division of Medical Oncology, Department of Medicine, Hospital of the University of Pennsylvania, Philadelphia, PA. Submitted July 31, 2000; accepted October 30, 2000. Address reprint requests to Theodore J. Robnett, MD, Department of Radiation Oncology, 2 Donner Bldg, Hospital of the University of Pennsylvania, 3400 Spruce St, Philadelphia, PA 19104; email: robnett{at}xrt.upenn.edu
PURPOSE: As therapy for locally advanced non–small-cell lung carcinoma (NSCLC) improves, brain metastases (BM) may become a greater problem. We analyzed our chemoradiation experience for patients at highest risk for the brain as the first failure site. METHODS: Records for 150 consecutive patients with stage II/III NSCLC treated definitively with chemoradiation from June 1992 to June 1998 at the University of Pennsylvania were reviewed. Most patients (89%) received cisplatin, paclitaxel, or both. All had negative brain imaging before treatment. Posttreatment brain imaging was performed for suspicious symptoms. Incidence of BM was examined as a function of age, sex, histology, stage, performance status, weight loss, tumor location, surgery, radiation dose, initial radiation field, chemotherapy regimen, and chemotherapy timing. RESULTS: Crude and 2-year actuarial rates of BM were 19% and 30%, respectively. Among pretreatment parameters, stage IIIB was associated with a higher risk of BM (P < .04) versus stage II/IIIA. Histology alone was not significant (P < .12), although patients with IIIB nonsquamous tumors had an exceptionally high 2-year BM rate of 42% (P < .01 v all others). Examining treatment-related parameters, crude and 2-year actuarial risk of BM were 27% and 39%, respectively, in patients receiving chemotherapy before radiotherapy and 15% and 20%, respectively, when radiotherapy was not delayed (P < .05). On multivariate analysis, timing of chemotherapy (P < .01) and stage IIIA versus IIIB (P < .01) remained significant. CONCLUSION: Patients with later stage, nonsquamous NSCLC, particularly those receiving induction chemotherapy, have sufficiently common BM rates to justify future trials including prophylactic cranial irradiation.
THE BRAIN IS A common site of metastasis in patients with lung cancer. The rate of brain metastasis (BM) is highest in small-cell lung cancer, and multiple randomized trials and a larger meta-analysis have suggested that the problem is common enough after chemotherapy, with or without radiation, that routine prophylactic cranial irradiation (PCI) in patients with no other discernible disease is justified.1-3 Rates of brain metastasis in non–small-cell lung cancer (NSCLC) are lower than those for small-cell lung cancer, but the problem of BM after treatment is significant. Recent data show that survival in locally advanced NSCLC is improved by the addition of chemotherapy to radiotherapy and/or surgery.4-6 Since the acceptance of chemoradiation as the standard of care for patients able to tolerate such therapy,7 there has been little information reported to document the change in natural history of BM after chemoradiation. Improved survival in patients receiving chemoradiation versus those receiving radiation therapy alone may increase the opportunity for BMs to manifest and become a threat to quality of life (QOL), survival, or both. To this end, we reviewed our institutional experience of patients receiving chemoradiation for NSCLC in an attempt to find prognostic and treatment-related failure patterns associated with BM as the first site of failure.
The University of Pennsylvania radiation oncology electronic database was queried for lung cancer patients treated with chemoradiation between June 1992 and June 1998. The starting date for this study was selected to correspond to the time period when chemoradiation replaced radiation therapy (XRT) alone in the management of most patients with good performance status stage III NSCLC in our institution. Also included were five patients with clinical stage II disease who did not have mediastinal lymph node sampling but who were suspected to have more aggressive disease by virtue of the large volume of hilar and/or peribronchial lymph nodes pathologically involved with disease. Routine computed tomography, magnetic resonance scanning of the brain, or both were completed before initiation of treatment. Posttreatment scanning was only performed in the event of CNS signs or symptoms, such as new onset seizure, visual field changes, nausea or vomiting, motor or sensory changes, new onset headaches, confusion, or cranial nerve deficits. This series did not include patients treated with XRT alone, because in the time period of this study, chemoradiotherapy was the standard treatment of stage III lung cancer. XRT alone was generally reserved for patients with poorer performance status. This series also excluded patients treated with chemoradiation for palliative intent. A total of 150 consecutive patients were identified for this study. Patients were analyzed on an intent-to-treat basis, hence the completion of planned treatment was not required for inclusion in the analysis. Patient characteristics are listed in Table 1. All patients had histologically confirmed NSCLC clinical stages II, IIIA, or IIIB by the 1997 staging.8 The median age was 63 years; 44% of the patients were female. Patient characteristics are outlined by treatment sequence in Table 2.
All patients received definitive radiotherapy using conventional (1.8 to 2 Gy) fraction size; anteroposterior/posteroanterior fields encompassed the gross tumor and regional nodes (including at minimum the ipsilateral hilum and mediastinum) to 40 to 45 Gy. The ipsilateral supraclavicular nodes were typically included in these initial fields. Patients with inoperable NSCLC received additional boost XRT (off spinal cord) to gross disease with a minimum 2-cm margin to a total dose of 60 to 66 Gy. Twenty-one patients received chemotherapy both before and during XRT; for the purposes of this report, they are considered to be in the sequential group, because their treatment plan incorporated intentional delay of thoracic radiation. Chemotherapy was delivered concurrently in 59% of patients and sequentially in 35% of patients (6% unknown). The decision to deliver chemotherapy sequentially versus concurrently was multifactorial. Patients who were highly symptomatic from local-regional disease usually started both radiotherapy and chemotherapy concurrently, unless prevented by logistical issues. For other patients, the decision regarding sequential versus concurrent therapy was individualized, based largely on the assessment of the patient’s perceived ability to tolerate the increased rigors of concurrent therapy. Surgical patients with definitive intent either before or after chemoradiation were included in this study. It has not been routine to offer both radiotherapy and chemotherapy after resection of node-positive NSCLC in our institution; however, selected patients with adverse pathologic findings and excellent performance status were given adjuvant chemoradiotherapy if they strongly desired it. Twenty patients received postoperative chemoradiotherapy for pathologic stage II/IIIA disease; the final radiotherapy dose was 50 to 60 Gy for these patients. Patients with stage IIIA NSCLC planned for subsequent resection were thought to belong to a favorable subpopulation with nonbulky (limited N2) disease. These patients received only a limited boost to a total dose of 45 to 54 Gy involving the same boost volume described for inoperable disease or no boost at all. Sixty-five patients underwent thoracotomy and attempted resection. The crude risk of BM as the first site of failure after treatment was calculated. For actuarial purposes, the time of control in the brain was calculated from the time of diagnosis. Patients suffering progression at non-CNS sites and/or death from any cause were considered censored for actuarial BM rate at the date of failure. The following pretreatment factors were evaluated by univariate analysis for possible impact on the outcome of BM: age, sex, histology (squamous v nonsquamous cell), stage (II/IIIA v IIIB), performance status, weight loss (< 10% v > 10% of prior body weight), tumor location (left v right, upper v middle/lower lobe), and nodal status (N0/N1 v N2). Treatment-related factors analyzed included whether surgery was performed, radiation dose, initial radiation field size, chemotherapy regimen, and timing of chemotherapy (sequential v concurrent). Chemotherapy drugs were analyzed as platinum-based drugs versus non–platinum-based drugs. The log-rank test was used for univariate analysis. Multivariate analysis was performed using a logistic regression model9 containing all variables that attained or trended toward univariate statistical significance. Further analysis of restriction models employed multiple criteria to assess reliability. The mean follow-up for all patients in this series is 15 months. The mean follow-up for surviving patients is 18 months.
The median overall survival was 14.5 months. Of the 150 patients treated, 29 were subsequently diagnosed with BM (19%). The 2-year actuarial incidence of BM was 30% for the entire population. Median time to diagnosis of BM from date of cancer diagnosis was 9.3 months, with a range of 1.9 to 21.9 months. Of the 29 cases that recurred first in the brain, 17 patients (59%) died from BM or had BM as the only site of relapse at the time of death. An additional seven patients (24%) died with BM and metastases at other sites; in these cases, the cause of death was unclear. Five patients (17%) with BM as the first relapse site died of metastases at another site.
Univariate Analysis
Examining treatment-related factors, the crude risk of BM was 27% in patients receiving chemotherapy before radiotherapy and 15% in patients who received concurrent chemoradiation (P < .03). Actuarial rate of BM as a function of chemotherapy sequence is shown in Fig 2. The actuarial rate of BM at 2 years was 39% in patients receiving chemotherapy first and 20% in patients receiving concurrent chemoradiation (P < .05). When any delay in radiation therapy was examined (delay of radiation because of either initial surgery or chemotherapy), the difference in BM incidence continued to be statistically significant (P < .02). The 2-year actuarial rate of BM with any delay in radiation therapy was 36% versus 15% if patients received radiation therapy as part of initial treatment. As listed in Table 2, there was a slight imbalance between the two groups; patients receiving concurrent chemoradiation were more likely to have undergone surgery and less likely to have received taxane-based chemotherapy. However, no difference in the incidence of BM was found between patients receiving definitive chemoradiation versus those receiving trimodality therapy (P = .65) or those receiving platinum- versus taxane-based chemotherapy (P = .72).
Other pretreatment factors, including age, sex, performance status, weight loss, tumor location, and nodal status as well as treatment-related factors (initial XRT field size, XRT dose, chemotherapy drugs, surgery) were not associated with a univariate risk of BM.
Multivariate Analysis
The incidence of BM in modern patients receiving definitive chemoradiation (without PCI) for locally advanced NSCLC has been estimated to be between 12% and 28%.10-13 Our data are within this range and, given the relatively short median follow-up time of patients, probably underestimate the incidence of BM. Even so, subgroup analysis shows that certain populations are at particularly high risk of brain metastasis. Nonsquamous histologies, particularly adenocarcinoma, have previously been found to produce CNS metastases more frequently than squamous cell carcinoma,14-16 and our data reveal this trend as well. Further, our series suggests a 50% increase in crude incidence of BM when comparing II/IIIA patients (18%) with those with IIIB disease (27%). The end point of BM as the first site of failure was chosen because this cohort would most likely benefit from improved therapy addressing the brain. Such improved therapy would almost certainly improve QOL. Because brain imaging was not routinely performed in our cohort after the completion of whole brain radiation, we are unable to assess what the salvage rate of whole-brain radiation was after first recurrence in the brain. Other studies suggest that at least one half of the patients who relapse in the brain die of uncontrolled disease after salvage with radiation therapy.17-20 Such treatment may prolong survival as well. The CNS has long been acknowledged as a common site of relapse for NSCLC, but recent studies employing aggressive chemoradiation have shown increasing rates of BM.21,13 Eberhardt et al22 observed a 46% crude rate of BM in their first 28 patients treated with chemoradiation followed by thoracic surgery, although these were not necessarily the first sites of relapse. A review of the Radiation Therapy Oncology Group (RTOG) database comparing patients who received XRT alone with chemoradiation patients found that "brain metastasis was altered little by the addition of chemotherapy, but other distant metastases were significantly decreased..."16,p.505 These two observations taken together suggest a change in the natural course of NSCLC when chemoradiation is used instead of XRT alone. Although chemotherapy may not change the incidence of BM per se, the decrease in metastasis at other sites in combination with an increased median survival with chemoradiation may cause a relative increase in the incidence of brain metastases. If this is the case, then a change in patterns of therapy for this population may be in order. Of the four randomized trials of PCI in patients with locally advanced NSCLC, all have shown a decreased incidence of BM, but none have shown an improvement in survival.23-26 A possible explanation for the negative randomized trials of PCI may be that the rates of local-regional and/or non-CNS distant failure were so high that BM lacked the opportunity to manifest themselves. Future studies of PCI for NSCLC should limit enrollment to patients treated with both chemotherapy and local therapy. Another possible explanation for the negative outcomes of earlier NSCLC PCI trials may be caused by the inclusion of patients not at particularly high risk for BM. The most likely way a survival or QOL advantage may be shown is by selecting a patient population with a high propensity for BM, and our data indicate that patients with IIIB, nonsquamous histology disease fit this description. The argument against embarking on PCI trials focuses on the relative radioresistance of NSCLC compared with small-cell lung cancer, thus requiring a whole-brain XRT dose that would be excessively neurotoxic. However, the experience of Stuschke et al13 suggests that a dose of 30 Gy in 2-Gy fractions was both well tolerated and effective, with a decrease in the risk of BM by almost four-fold. The finding that delay of radiotherapy is associated with an increase in BM for NSCLC is similar to the findings of the randomized trial of Murray et al27 for small-cell lung cancer. In that trial, patients had a higher risk of BM in the arm receiving delayed thoracic XRT. These findings are consistent with the hypothesis that early, aggressive locoregional and systemic therapy permits better control of the disease, which is also supported by the comparison of RTOG patients treated with XRT with or without chemotherapy.16 Our data showed that patients in whom radiotherapy is delayed—either because of induction chemotherapy or surgery—had an increased risk of BM. Because surgery and radiotherapy are both local treatments, this finding cannot be easily ascribed to a single hypothesis. One possible explanation is that patients who underwent surgery first are likely to have had more limited local-regional disease and better performance status than other patients. These patients are unlikely to suffer local-regional failure and thus have a greater opportunity to develop BM. A second explanation is that while surgery is indeed more effective than XRT in controlling gross, local disease, XRT may be more effective in eradicating occult regional disease (eg, supraclavicular nodes), which can serve as a nidus for metastases to the brain. If this disease is in part responsible for seeding the brain, then these two modalities may not be interchangeable. Our findings are limited to those of any retrospective analysis, and there may very well be some validity in the idea that patients who received chemotherapy before radiation therapy had poor prognostic features that rendered them ineligible for concurrent chemoradiation. It should be noted, however, that the prognostic factor of chemotherapy sequencing remained a significant parameter under multivariate analysis, along with overall stage, and the significance of chemotherapy sequencing was strengthened by the inclusion of pretreatment- and treatment-related parameters in the model. Our results contradict those found in a secondary analysis of the Japanese randomized trial of sequential versus concurrent chemoradiation, in which the latter group had a higher risk of BM.28 Certainly more study of this topic is warranted, and future studies of PCI should include both patients treated with sequential or concurrent chemo-XRT. Our data and those recently reported by the RTOG29 suggest trials should focus on patients with nonsquamous histology. Such PCI studies should also be limited to those patients who receive systemic chemotherapy, although restriction to a taxane- versus platinum-based regimen in not paramount. Patients on such studies should be those most likely to benefit from control of the brain. Therefore, enrollment should be limited to patients thought to be in clinical local control with at least a partial response after chemoradiotherapy by computed tomography scan, good performance status, and an unremarkable restaging work-up. These patients may then be stratified by stage, whether definitive surgery was performed, and timing of radiation therapy (early or delayed), with randomization to PCI in 2-Gy daily fractions to 30 Gy versus no PCI. Primary end points should include intracranial control, overall survival, and QOL measures. In conclusion, our data suggest that the risk of BM as the first site of failure in patients with locally advanced NSCLC treated with chemoradiation is sufficiently high to justify prospective trials of PCI. Several subgroups have been identified to have a particularly high risk, but these results will require validation in additional studies.
Presented in part at the Eighty-Second Annual Meeting of the American Radium Society, April 1-5, 2000, London, England, and in part at the Thirty-Sixth Annual Meeting of the American Society of Clinical Oncology, May 20-13, 2000, New Orleans, LA.
1. Arriagada R, LeChevalier T, Borie F, et al: Prophylactic cranial irradiation for patients with small-cell lung cancer in complete remission. J Natl Cancer Inst 87: 183-189, 1995 2. Gregor A, Cull A, Stephens RJ, et al: Prophylactic cranial irradiation is indicated following complete response to induction therapy in small cell lung cancer: Results of a multicentre randomised trial. Eur J Cancer 33: 1752-1758, 1997
3.
Auperin A, Arriagada R, Pignon J-P, et al: Prophylactic cranial irradiation for patients with small-cell lung cancer in complete remission. N Engl J Med 341: 476-484, 1999 4. Dillman RO, Seagren SL, Propert KJ, et al: A randomized trial of induction chemotherapy plus high-dose radiation vs. radiation alone in stage III non-small cell lung cancer. N Engl J Med 323: 940-45, 1990[Abstract]
5.
Sause WT, Scott C, Taylor S, et al: Radiation Therapy Oncology Group (RTOG) 88-08 and Eastern Cooperative Group (ECOG) 4588: preliminary results of a phase III trial in regionally advanced, unresectable non-small cell lung cancer. J Natl Cancer Inst 87: 198-205, 1995
6.
Jeremic B, Shibamoto Y, Acimovic L, et al: Hyperfractionated radiation therapy with or without concurrent low-dose daily carboplatin/etoposide for stage III non-small cell lung cancer: a randomized study. J Clin Oncol 14: 1065-1070, 1996 7. ASCO: Clinical practice guidelines for the treatment of unresectable non-small-cell lung cancer. J Clin Oncol 15: 2996-3018, 1997[Abstract] 8. AJCC : American Joint Committee on Cancer (AJCC) Cancer Staging Manual ( ed 5 ). Philadelphia, PA, Lippincott-Raven Publishers, 1997 9. Kleinbaum DG, Kupper LL, Muller KE: Applied regression analysis and other multivariate methods ( ed 2 ). Belmont, CA, Duxbury Press, 1988 10. Machtay M, Seiferheld W, Komaki R, et al: Is prolonged survival possible for patients with supraclavicular node metastases in non-small cell lung cancer treated with chemoradiotherapy? : Analysis of the radiation therapy oncology group experience. Int J Radiat Oncol Biol Phys 44: 847-853, 1999[Medline] 11. Komaki R, Scott C, Ettinger D, et al: Randomized study of chemotherapy/radiation therapy combinations for favorable patients with locally advanced inoperable nonsmall cell lung cancer: Radiation Therapy Oncology Group (RTOG) 92-04. Int J Radiat Oncol Biol Phys 38: 149-155, 1997[Medline] 12. Skarin A, Jochelson M, Sheldon T, et al: Neoadjuvant chemotherapy in marginally resectable stage III M0 non-small cell lung cancer: Long-term follow-up in 41 patients. J Surg Oncol 40: 266-274, 1989[Medline]
13.
Stuschke M, Eberhardt W, Pottgen C, et al: Prophylactic cranial irradiation in locally advanced non-small-cell lung cancer after multimodality treatment: Long-term follow-up and investigations of late neuropsychologic effects. J Clin Oncol 17: 2700-2709, 1999 14. Cox JD, Yesner R, Mietlowski W, et al: Adenocarcinoma of the lung: Recent results from the veterans administration lung group. Am Rev Respir Dis 120: 1025-1029, 1979[Medline] 15. Line DH, Deeley TJ: The necropsy findings in carcinoma of the bronchus. Br J Dis Chest 65: 238-242, 1971[Medline] 16. Cox JD, Scott CB, Byhardt RW, et al: Addition of chemotherapy to radiation therapy alters failure patterns by cell type within non-small cell carcinoma of lung (NSCCL): Analysis of radiation therapy oncology group (RTOG) trials. Int J Radiat Oncol Biol Phys 43: 505-509, 1999[Medline] 17. Borgelt BB, Gelber R, Kramer S, et al: The palliation of brain metastases: Final results of the first two studies by the Radiation Therapy Oncology Group. Int J Radiat Oncol Biol Phys 6: 1-9, 1980[Medline] 18. Kurtz JM, Gelber R, Brady LW, et al: The palliation of brain metastases in a favorable patient population: A randomized clinical trial by the Radiation Therapy Oncology Group. Int J Radiat Oncol Biol Phys 7: 891-895, 1981[Medline] 19. Mandel L, Hilaris B, Sullivan M, et al: The treatment of single brain metastasis from non-oat cell lung carcinoma: Surgery and radiation versus radiation therapy alone. Cancer 58: 641-649, 1986[Medline] 20. Montana GS, Meachan WF, Caldwell WL: Brain irradiation for metastatic disease of lung origin. Cancer 29: 1477-1480, 1972[Medline] 21. Law A, Daly BMM: High incidence of isolated brain metastases (CNSMets) following complete response (CR) in advanced non-small cell lung cancer (NSCLC): A new challenge. Lung Cancer 18: 65, 1997 (abstr 248) 22. Eberhardt W, Wilke H, Stuschke M, et al: Preoperative chemotherapy and chemoradiotherapy based on hyperfractionated accelerated radiotherapy followed by surgery in locally advanced (LAD) inoperable NSCLC stages IIIA and IIIB: The value of PCI. Lung Cancer 18: 65, 1997 (abstr 249) 23. Miller TP, Crowley JJ, Mira J, et al: A randomized trial of chemotherapy and radiotherapy for stage III non-small cell lung cancer. Cancer Therapeutics 1: 229-236, 1998
24.
Cox JD, Stanley K, Petrovich Z, et al: Cranial irradiation in cancer of the lung of all cell types. JAMA 245: 469-472, 1981 25. Russell AH, Pajak TE, Selim HM, et al: Prophylactic cranial irradiation for lung cancer patients at high risk for development of cerebral metastasis: Results of a prospective randomized trial conducted by the radiation therapy oncology group. Int J Radiat Onco Biol Phys 21: 637-643, 1991 26. Umsawasdi T, Valdivieso M, Chen TT, et al: Role of elective brain irradiation during combined chemoradiotherapy for limited disease non-small cell lung cancer. ase non-small cell lung cancer. J Neurooncol 2: 253-259, 1984[Medline]
27.
Murray K, Coy P, Pater JL, et al: Importance of timing for thoracic irradiation in the combined modality treatment of limited-stage small-cell lung cancer. J Clin Oncol 11: 336-344, 1993
28.
Furuse K, Fukuoka M, Kawahara M, et al: Phase III study of concurrent versus sequential thoracic radiotherapy in combination with mitomycin, vindesine, and cisplatin in unresectable stage III non-small-cell lung cancer. J Clin Oncol 17: 2692-2699, 1999 29. Komaki R, Seiferheld W, Curran W, et al: Sequential vs. concurrent chemotherapy and radiation therapy for inoperable non-small cell lung cancer (NSCLC): Analysis of failures in a phase III study (RTOG 9410). Int J Radiat Oncol Biol Physics 48: 113, 2000 (abstr 5)
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2001 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
