It is well established that patients undergoing treatment

Transcription

1 Surgical Resection of Non Small Cell Carcinoma After Treatment for Small Cell Carcinoma W. Roy Smythe, MD, Anthony L. Estrera, MD, Stephen G. Swisher, MD, Kelly W. Merriman, MPH, Garrett L. Walsh, MD, Joe B. Putnam, Jr, MD, Ara A. Vaporciyan, MD, and Jack A. Roth, MD Department of Thoracic and Cardiovascular Surgery, The University of Texas M.D. Anderson Cancer Center, Houston, Texas Background. Development of non small cell lung carcinoma (NSCLC) in patients previously treated for small cell carcinoma (SCLC/NSCLC) is well described; however, little is known about clinical outcome. Methods. A single-institution 20-year review was performed. Patient characteristics and survival for SCLC/ NSCLC patients were compared with those for control patients matched for stage, resection, and previous malignancy. Results. One thousand four hundred four patients with small cell carcinoma were identified, and 29 underwent therapy for metachronous NSCLC: 11 of 29 patients underwent surgical resection, 10 of these 11 (90%) were stage I. Compared with surgically treated stage I NSCLC patients, SCLC/NSCLC patients were more likely to have squamous histology (70% versus 35%, p 0.026); and subanatomic resection (90% versus 17.4%, p < ). The SCLC/NSCLC patients had significantly poorer survival when compared with stage I NSCLC patients undergoing any resection (24.53 versus months, p 0.003) and stage I NSCLC patients receiving wedge resection (24.53 versus months, p 0.006). Survival was similar to NSCLC patients with a history of previous treated extrathoracic solid malignancy. Conclusions. Surgical resection for SCLC/NSCLC patients is feasible, but poorer prognosis is noted when compared with stage-matched control patients. Surgical candidates should be carefully chosen, and alternative local control modalities considered. (Ann Thorac Surg 2001;71:962 6) 2001 by The Society of Thoracic Surgeons Accepted for publication Sept 22, Address reprint requests to Dr Smythe, The University of Texas M.D. Anderson Cancer Center, Department of Thoracic and Cardiovascular Surgery, 1515 Holcombe Blvd, Box 109, Houston, TX 77030; rsmythe@mdanderson.org. It is well established that patients undergoing treatment for small cell lung carcinoma (SCLC) are at increased risk for the development of second primary tumors (SPT). In a number of published reports evaluating those surviving for at least 2 years after treatment for SCLC, up to 30% or more will develop SPT, the majority of which are of aerodigestive origin [1 4]. In many of these studies, the second primary aerodigestive malignancy most commonly encountered is carcinoma of the lung. The risk of developing a second non small cell (NSCLC) lung cancer in these patients surviving SCLC for more than 2 years is 7- to 16-fold higher than a comparable population of noncancer patients who smoke (2% to 13% per patient- year). In addition, the risk increases over time to as much as 30% at 10 years from the initial treatment date for SCLC [5, 6]. In contrast to the now predominant histologic NSCLC diagnosis of adenocarcinoma, most second primary NSCLC tumors reported in the literature in patients previously treated for SCLC are squamous cell carcinoma [7]. Although a great deal has been learned regarding the cause of second primary NSCLC after treatment of SCLC, little useful information regarding clinical outcome or truly definitive surgical resection in these patients has been available. This inconsistency is thought to result from incomplete patient follow-up, assumption of SCLC rather than NSCLC diagnosis at radiographic representation, or possibly a reluctance to treat second primary NSCLC with surgical resection because of assumed prohibitive risk factors (ie, previous thoracic irradiation) [6]. This study sought to evaluate the clinical outcome in patients that were previously treated definitively for SCLC followed by surgical treatment for a second primary diagnosis of NSCLC (SCLC/NSCLC). When compared with appropriate matched groups of patients, SCLC/NSCLC patients exhibited a poorer survival than anticipated. Material and Methods Single institutional and department retrospective data were collected for a 20-year period (1978 to 1998). Patients were included in the SCLC/NSCLC group if they had received definitive SCLC treatment, with clinical resolution of disease, and later had development of an NSCLC tumor. Non small cell lung carcinoma tumors were not anatomically restricted (ie, they were included whether or not they arose in the same anatomic lobe of the lung as the previous SCLC tumor). The majority of patients in this study were treated surgically with wedge resection alone (plus mediastinal nodal dissection or sampling) for 2001 by The Society of Thoracic Surgeons /01/$20.00 Published by Elsevier Science Inc PII S (00)

2 Ann Thorac Surg SMYTHE ET AL 2001;71:962 6 NON SMALL CELL AFTER SMALL CELL CARCINOMA 963 Table 1. Demographics of Patients Previously Treated for Small Cell Lung Carcinoma Who Had Surgical Treatment for a Secondary Diagnosis of Non Small Cell Lung Carcinoma Compared With Patients Undergoing Any Surgical Procedure for Stage I Non Small Cell Lung Carcinoma Variable NSCLC Alone (C1) SCLC/NSCLC p Value Age (y) a NS Sex (%) 36.4 male 56.6 male NS 63.6 female 43.5 female Race (%) White NS Other Tobacco Use NS (pack-years) b Operation (%) Lobectomy Wedge resection NS not significant; NSCLC non small cell lung carcinoma; SCLC small cell lung carcinoma. a Values are median range. b Values are mean range. the later NSCLC tumor and had stage I disease; therefore, three surgically treated stage I NSCLC control groups were chosen for comparison. These included a random sample of patients undergoing any resection for stage I NSCLC (C1, n 23), wedge resection alone for stage I NSCLC (C2, n 46), or wedge resection alone for a diagnosis of stage I NSCLC after previous definitive treatment for any extrathoracic nonhematologic solid malignancy (C3, n 17). The 1 patient with higher final pathologic stage disease (T3N2, stage IIIA) in the SCLC/ NSCLC group was excluded from survival and mode of resection comparisons. Survival data and curves were calculated using the Kaplan-Meier method. Continuous variables were compared by Student s t test, and categorical variables were compared with Pearson s 2, with Fisher s exact modification where applicable. A confidence level of 0.05 or lower was considered significant. All statistical analyses were performed using SPSS software (Chicago, IL). Results Study Group Characteristics From the period of 1978 to 1998, 1,404 patients with SCLC were registered and received initial therapy for that diagnosis at the University of Texas M.D. Anderson Cancer Center. Twenty-nine of these patients received some type of therapy at this institution for a new NSCLC diagnosis. A subgroup of 11 patients were identified from this group who had been treated successfully for SCLC, later developed NSCLC, and were treated with surgical resection. The demographic characteristics of these patients are shown in Table 1. All patients received chemotherapy (CAV regimen: cyclophosphamide, doxorubicin, and vincristine) as treatment for SCLC, and 10 of 11 received 45 to 50 Gy of thoracic irradiation. Squamous cell carcinoma was the most common histologic diagnosis (7 of 10, 70%), and most patients underwent wedge resection (9 of 10, 90%). The median interval between SCLC and NSCLC diagnoses was 50.4 months (mean, months). Three patients had the NSCLC tumor in the same lobe as the previous SCLC. The majority of the NSCLC were stage I (10 of 11, 91%), with 1 patient staged as IIIA (T3N2) by the American Joint Committee on Cancer TNM staging system. Median survival in the SCLC/NSCLC group was months. Overall actuarial 5-year survival in the SCLC/NSCLC group. was 27%. Although all were thought to be cured in regard to the initial SCLC diagnosis, 2 of 11 eventually succumbed to metastatic SCLC after surgical resection of their NSCLC tumors. Comparison With Control Groups Three randomly selected control groups (n 86) were chosen to attempt to compare stage, type of resection, and previous history of other malignancy: C1, patients with stage I NSCLC treated with any type of resection (n 23); C2, patients with stage I NSCLC treated by wedge resection (n 46); and C3, patients with NSCLC treated by wedge resection after successful treatment for solid nonhematologic malignancy other than SCLC (n 17). When demographic characteristics of the SCLC/ NSCLC group were compared with C1, no significant differences in age, sex, or smoking exposure history were noted. The SCLC/NSCLC patients were more likely to have squamous histology (70% versus 35%, p 0.026) and more likely to have wedge or segmental resection rather than anatomic lobectomy (90% versus 17.4%, p ). By Kaplan-Meier analysis, median survival was poorer in the SCLC/NSCLC group when compared with C1 (24.53 versus months, p 0.003; Fig 1). When compared with C2 (n 26), the median survival was still significantly different (24.53 versus months, p 0.006; Fig 2), but when compared with C3, the survival rates were similar (24.53 versus months, not significant; Fig 3). Direct comparison with the 18 of 29 SCLC/NSCLC patients not treated with surgical resection is difficult because of selection bias. Complete stage and survival data were available for 10 of 18 of these patients, with the majority presenting with stage IV NSCLC (6 of 10). The survival in these patients reflects this stage grouping with a median of 10.0 months (mean SEM, months). Comment This study sought to examine and analyze the outcome of surgical treatment of SCLC/NSCLC patients. The development of aerodigestive SPTs in patients treated successfully for SCLC is a well-known entity [1 4]. This entity may well become significantly more prevalent based on recent improvements in treatment outcome for SCLC. Turrisi and colleagues [6] have published recent results

3 964 SMYTHE ET AL Ann Thorac Surg NON SMALL CELL AFTER SMALL CELL CARCINOMA 2001;71:962 6 Fig 1. Kaplan-Meier survival analysis of patients previously treated as compared with group C1 (patients with stage I non small cell lung carcinoma undergoing any resection). The SCLC/NSCLC patients exhibited significantly poorer survival (24.53 versus months, p 0.003). of combined chemotherapy and radiation therapy for limited SCLC (n 417), with 2- and 5-year survival rates of 44% and 23% respectively, a substantial improvement over earlier figures. We identified 29 of 1,404 SCLC patients (2% raw incidence) during a 10-year period who later developed NSCLC. Approximately 10% of patients (all stages) can be expected to be alive at 2 years after treatment for SCLC. In this study, 10 of 11 patients treated surgically for NSCLC developed this diagnosis 2 or more years after treatment for SCLC. Therefore, although exact figures are not available, these data would Fig 2. Kaplan-Meier survival analysis of patients previously treated as compared with group C2 (patients with stage I non small cell lung carcinoma receiving wedge resection). The SCLC/NSCLC patients exhibited significantly poorer survival (24.53 versus months, p 0.006). Fig 3. Kaplan-Meier survival analysis of patients previously treated as compared with group C3 (non small cell lung carcinoma patients with a history of previous treated solid malignancy [prior ca] and wedge resection for non small cell lung carcinoma). Survival in the two groups was not statistically different. argue for an incidence of NSCLC in 2-year survivors of SCLC treated at this institution of at least 10%, a number compatible with other published reports [5, 7]. Although a number of studies have commented on the frequency of development of NSCLC after treatment of SCLC, little is known about treatment outcomes in these patients [5, 6]. Occurrence of NSCLC as an SPT may well become more common with advances in intensive combination therapies and improvement in survival seen in SCLC patients in the 1980s and 1990s combined with the long latency period often observed before development of NSCLC [9]. Several of the characteristics of the patients presented here correlate with what has been noted by others in the past, including a relatively long latency period and a predominance of squamous histologic findings [7, 8]. There is no clear explanation for the increased frequency of squamous cell carcinoma seen as an SPT. Previous investigators have noted that the most common combination of histologic findings when more than one is noted in these tumors is small cell and large cell, and that most small cell tumor cell lines in culture will differentiate to large cell at higher passages [10, 11]. However, a distinction has been made between combined (small cell and large cell) and mixed (small cell and other NSCLC type) tumors, and if this distinction is made, then squamous cell histology may be considered the most common histologic finding noted in the mixed type [11, 12]. Three patients in this study had NSCLC in the same lobe as the previous SCLC. Controversy has also arisen in the past regarding the concept of true second primary NSCLC versus recurrence of tumor in the bed of the previously treated SCLC, or even metastases to other locations in the lung. Yesner [10, 12] demonstrated that in 205 tumors diagnosed by biopsy as SCLC, approximately 20% of

4 Ann Thorac Surg SMYTHE ET AL 2001;71:962 6 NON SMALL CELL AFTER SMALL CELL CARCINOMA 965 treated and untreated patients tumors demonstrated a different diagnosis at autopsy, either mixed SCLC and NSCLC or NSCLC alone. Pragmatically, the significance of these point-of-origin debates is questionable, however, as the genotype of the subsequent NSCLC is likely to differ, and the differences in clinical behavior of SCLC and NSCLC tumors as well as responsiveness to various therapies is well established [13]. It is likely that the finding of early stage NSCLC disease in this study is related to a follow-up bias, ie, those with a previous diagnosis of treated SCLC would have been followed up with at least yearly radiographic examination for recurrent or new primary disease. This may have also contributed to a larger proportion of patients actually being considered surgical candidates in this study. The majority of the SCLC/NSCLC patients presented here were treated with a nonanatomic wedge resection rather than an anatomic lobectomy. Numerous studies have shown that patients treated with lobectomy, rather than wedge resections, or even anatomic segmental resections fare much better both in regard to survival as well as local recurrence [14]. The explanation for nonanatomic resections in patients presented here is uncertain, but is likely related to the issue of preoperative risk assessment and the time at which the patient was treated surgically. All 11 SCLC/NSCLC patients presented had received doxorubicin-based chemotherapy, and 10 of 11 received thoracic irradiation. This perception may also be responsible for small numbers of these patients being considered viable surgical candidates as reported in the literature. In a review discussing second primary lung carcinoma one author states that only 5% to 13% of SCLC/NSCLC patients can expect to undergo a successful surgical resection [8]. We treated 11 of 29 (38%) patients presenting at this institution in this fashion, albeit with wedge resection. In addition to concern about operative risk, 6 patients were treated before 1990, and at that time wedge resections were thought by many to be sufficient surgical therapy for early stage NSCLC. What are the risk factors for later development of NSCLC? Several authors have identified continuation of smoking as a major risk factor for development of an SPT of the lung [3, 4, 7]. In this study 4 of 11 patients were smoking at the time of NSCLC diagnosis. Finally, the survival analysis was performed comparing SCLC/NSCLC patients with a number of control groups designed to compare by stage and for resection type and history of other previous solid malignancy. As anticipated, when compared with a random sample of surgically treated patients with stage I disease (C1), a significant difference in type of resection performed was noted. Ninety percent of SCLC/NSCLC versus 17.4% of C1 patients underwent wedge resection. When survival between these two groups was compared, C1 patients fared significantly better. The literature would suggest that the type of resection performed could readily explain this enhanced survival. Martini and associates [14] demonstrated 5- and 10-year survivals in those with stage I NSCLC treated with lobectomy at 77% and 70%, respectively, versus 59% and 35% for those undergoing wedge or segmental resections (p 0.024). In an attempt to examine this variable, we compared our SCLC/NSCLC patients with a group of stage I NSCLC patients undergoing wedge resection alone (C2). Interestingly, the survival was significantly poorer in the SCLC/NSCLC group, arguing strongly against type of resection alone as the variable responsible. We then selected a control group that we felt was more comparable patients with stage I NSCLC receiving wedge resection alone and a previous history of treated nonhematologic tumor other than SCLC (C3). A comparison between these two groups demonstrated no significant difference in survival. It is reasonable to state that surgical resection may more often be possible than previously thought in these patients, as our surgical intervention rate of 38% versus the 13% or less that is reported in the literature demonstrates. Certainly the lessons recently learned in surgical treatment of emphysema would suggest that diminished pulmonary function from previous treatment (ie, thoracic irradiation) should not be an absolute impediment to resection. This is compounded by the fact that many of these patients will not be candidates for other local control modalities, such as radiation therapy because of previous treatment exposure. That being stated, the overall poor results compared with the usual expected survival in stage I patients suggests that patients should be carefully selected for surgical intervention, and consideration for alternative means of local control (radiation therapy if previous treatment contralateral, novel local control modalities such as gene therapy or newer radiation therapy modalities) should be given [15 17]. In addition, the poorer survival demonstrated for stage I NSCLC patients after curative therapy for either SCLC or any other solid malignancy argues for continued efforts directed at understanding more completely the genotypic relationship between predisposition to multiple malignancies and prognosis [18]. References 1. Heyne KH, Lippman SM, Lee JK, Lee JS, Hong WK. The incidence of second primary tumors in long-term survivors of small-cell lung carcinoma. J Clin Oncol 1992;10: Tucker MA, Murray N, Shaw EG, et al. Second primary cancers related to smoking and treatment of small-cell lung cancer. J Natl Cancer Inst 1997;89: Richardson GE, Tucker MA, Venzon DJ, et al. Smoking cessation after successful treatment of small-cell lung cancer is associated with fewer smoking-related secondary primary cancers. Ann Intern Med 1993;119: Kawahara M, Ushijima S, Kamimori T, et al. Second primary tumors in more than 2 year disease free survivors of smallcell lung cancer in Japan: the role of smoking cessation. Br J Cancer 1998;78: Johnson BE, Linnoila RI, Williams JP, et al. Risk of second aerodigestive cancers increases in patients who survive free of small cell lung cancer form more than 2 years. J Clin Oncol 1995;13: Turrisi AT III, Kim K, Blum R, et al. Twice-daily compared with once-daily thoracic radiotherapy limited small cell lung cancer treated concurrently with cisplatin and etoposide. N Engl J Med 1999;340: Johnson BE. Second lung cancers in patients after treatment for an initial lung cancer. J Natl Cancer Inst 1998;90:

5 966 SMYTHE ET AL Ann Thorac Surg NON SMALL CELL AFTER SMALL CELL CARCINOMA 2001;71: Sagman U, Lishner M, Maki E, et al. Second primary malignancies following diagnosis of small-cell lung cancer. J Clin Oncol 1992;10: Perry M. Future directions in the therapy of small cell lung cancer. Chest Surg Clin N Am 1997;7: Yesner R. Small cell tumors of the lung. Am J Surg Pathol 1983;7: Gazdar AF, Carney DN, Baylin SB. Altered morphological, biological and biochemical characteristics in long term culture and heterotransplanted tumors [Abstract]. Proc Am Assoc Cancer Res 1980;21: Yesner R. Classification of lung cancer histology. N Engl J Med 1985;312: Kelley MJ, Nakagawa K, Conrad NK, et al. Genetic analysis of second primary lung cancers in patients surviving small cell lung cancer. Clin Cancer Res 1996;2: Martini N, Bains MS, Burt ME, et al. Incidence of local recurrence and second primary tumors in resected stage I lung cancer. J Thorac Cardiovasc Surg 1995;109: Swisher SG, Roth JA, Nemunaitis J, et al. Adenovirusmediated p53 gene transfer in advanced non small-cell lung cancer. J Natl Cancer Inst 1999;91: Pirollo KF, Hao Z, Rait A, et al. p53 mediated sensitization of squamous cell carcinoma of the head, and neck to radiotherapy. Oncogene 1997;14: Quoix E, Moreau L, Oster JP. Radiochemotherapy and radiosensitization for locally advanced non small-cell bronchial cancer. Cancer Radiother 1998;2: Miller DG, Tiwari R, Pathak S, Hopwood VL, Gilbert F, Hsu TC. DNA repair and mutagen sensitivity in patients with triple primary cancers. Cancer Epidemiol Biomarkers Prev 1998;7: The Thoracic Surgery Foundation for Research and Education Harvard Executive Course The tenth offering of the Harvard Executive Course, Understanding the New World of Health Care, will be November 3 11, Applications from Harvard can be obtained in June 2001 by contacting Eleanor Brimley at (617) Alley-Sheridan Scholarship materials will be available at the same time from The Foundation. Please contact Lainie Castle at (312) , extension Research Award Applications Information and applications for Thoracic Surgery Foundation Research Grants, Research Fellowships and Career Development Awards will be available beginning August 1, To request an application, please contact Lainie Castle at (312) (phone); (312) (fax); or lcastle@sba.com ( ). TSFRE Donor Recognition Luncheon, May 2001 The Thoracic Surgery Foundation is pleased to announce that its third annual Donor Recognition Luncheon will be held at noon on Tuesday May 8, 2001 at The American Association for Thoracic Surgery s 81st Annual Meeting in San Diego, California. The Foundation will honor its New Century Society, Lifetime Members, and major individual and corporate contributors by The Society of Thoracic Surgeons Ann Thorac Surg 2001;71: /01/$20.00 Published by Elsevier Science Inc