Prognosis of lung cancer resection in patients with previous extra-respiratory solid malignancies

Transcription

1 European Journal of Cardio-Thoracic Surgery Advance Access published February 7, 2013 European Journal of Cardio-Thoracic Surgery (2013) 1 5 doi: /ejcts/ezt031 ORIGINAL ARTICLE Prognosis of lung cancer resection in patients with previous extra-respiratory solid malignancies Pierre Benoit Pagès a, Pierre Mordant a, Aurélie Cazes b, Bertrand Grand a, Christophe Foucault a, Antoine Dujon c, Françoise Le Pimpec Barthes a and Marc Riquet a, * a b c Department of Thoracic Surgery, Georges Pompidou European Hospital, Paris-Descartes-University, Paris, France Department of Pathology, Georges Pompidou European Hospital, Paris-Descartes-University, Paris, France Department of Thoracic Surgery, Cedar Surgical Centre, Bois Guillaume, France * Corresponding author. Department of General Thoracic Surgery, Georges Pompidou European Hospital, 20 rue Leblanc, Paris, France. Tel: ; fax: ; marc.riquet@egp.aphp.fr (M. Riquet). Received 4 October 2012; received in revised form 18 December 2012; accepted 3 January 2013 Abstract OBJECTIVES: Non-small-cell lung cancer (NSCLC) following pulmonary or pharyngolaryngeal malignancies has been widely studied, but only a few articles have focussed on lung cancers following other solid malignancies. Our purpose was to compare the characteristics and prognosis of patients with NSCLC according to the medical history of the extra-pulmonary and extra-pharyngolaryngeal solid malignancy. METHODS: Patients who underwent surgery for NSCLC from January 1980 to December 2009 in two French thoracic centres were reviewed. We compared patients with no history of cancer (Group 1) and patients with a history of extra-pulmonary and extrapharyngolaryngeal solid malignancy (Group 2). RESULTS: There were 4992 patients: 4603 (92%) in Group 1 and 389 (8%) in Group 2. In comparison with Group 1, Group 2 showed an increasing incidence over the last 3 decades (2 8%), an older population (65.9 vs 61 years, P < 0.001), a higher proportion of women (34 vs 18%, P < 0.001), non-smokers (20 vs 10%, P < 0.001), adenocarcinomas (53 vs 40%, P < 0.001), T1 (16 vs 14%, P = 0.047) and second nodule in the same lobe (4 vs 2%, P < 0.001). The overall survival was not significantly different between the two groups (P = 0.09). In multivariate analysis, older age, male gender, pneumonectomy, higher T, higher N, incomplete resection and history of extra pulmonary extra pharyngolaryngeal solid malignancy were significantly associated with a worse prognosis. CONCLUSIONS: Despite an earlier diagnosis, a history of extra-pulmonary and extra-pharyngolaryngeal solid malignancy is associated with a worse prognosis in patients with NSCLC undergoing surgical resection. Overall survival is particularly low after a history of bladder and upper gastrointestinal malignancies. Keywords: Second primary lung cancer Extra-pulmonary malignancy Overall survival INTRODUCTION Multiple primary cancers involving the lung were first reported by Cahan et al. [1] in Sochocky et al. [2] then reported a frequency of 8.6% among the patients with lung cancer. Lung cancer commonly occurs after previous pharyngolaryngeal or lung cancer, and these situations have been commonly studied in the past [3 5]. Massard et al. [6] published the first article focussing on lung cancers occurring after other system malignancies in This study included 55 patients, but some of them presented a history of leukaemia, whose pathology and management are very far from those of solid malignancies. More recently, Hofmann et al. [7] reported 163 patients with primary lung cancer and previous extra-pulmonary malignancy, but only a minority of these patients underwent a curative resection. In the absence of previous large series, our purpose was to further analyse the results of the surgical resection of lung cancer in patients with a history of extra-pulmonary and extrapharyngolaryngeal solid malignancy. MATERIAL AND METHODS The clinical records of patients who underwent surgery for NSCLC from January 1980 to December 2009 in Georges Pompidou European Hospital (Paris) and Cedar Surgery Centre (Bois Guillaume) were prospectively collected and retrospectively analysed. The preoperative diagnosis workup included chest X-ray, bronchoscopy, computed tomography (CT) scan of the chest, spirometry, lung perfusion scan and a thorough search for distant metastases including PET-scan in recent years. Mediastinoscopy was performed to exclude N3 disease and to confirm N2 involvement in patients included in various induction therapy protocols depending on different referring centres. The Author Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.

2 2 P.B. Pagès et al. / European Journal of Cardio-Thoracic Surgery N3 disease and distant metastases precluded surgery. The staging system was the seventh edition of the International Staging System for NSCLC [8]. Patients were divided in two groups according to their previous oncological history. Group 1 included patients with no history of other cancer. Group 2 included patients with a history of previous extra-pulmonary and extra-pharyngolaryngeal solid malignancy. A second malignancy was considered metachronous when it was diagnosed and treated before discovering the lung cancer, and synchronous when both malignancies were diagnosed at the same time. The malignancies were considered as two primaries when the histology was different or after immunochemistry confirmation when histology was similar. Doubtful cases were considered as potential metastases and then excluded. There were 4992 patients: 4603 (92%) in Group 1 and 389 (8%) in Group 2. The study was approved by our Thoracic Surgery Society Ethic Committee (CERC-SFCTCV) that waived need for informed consent. Follow-up information was obtained from the hospital case records, from a questionnaire completed by either the chest physician or the general practitioner or from death certificates. Postoperative mortality was defined as any death occurring during the first month after surgery or during the initial hospital stay if longer. The main outcome was the overall survival, defined as the time interval between the date of lung surgery and the date of death or the last follow-up visit for censored patients. The average follow-up duration was 72.8 ± 48 months. Actuarial survival curves were estimated by the Kaplan Meier method. Statistical comparisons between survival distributions were made using the log-rank test. Multivariate analysis was performed using the Cox proportional hazards model for overall survival analysis. Univariate analysis used the following outcome variables: gender, age, type of surgical resection, histology, type of N involvement, and history and site of previous solid malignancy. All data analyses were conducted with the twosided test: a P value was considered as statistically significant when <0.05. The statistical software used for the analysis was SEM (Anticancer Centre Jean Perrin, Clermont-Ferrand, France) [9]. RESULTS Demographics of the populations are given in Table 1. The number of patients presenting with a history of extra-pulmonary and extra-pharyngolaryngeal solid neoplasia increased from 2 to 8% over the last 3 decades. In comparison with Group 1, Group 2 showed an older population (65.9 vs 61 years, P < 0.001), a higher proportion of women (34 vs 18%, P < 0.001) and nonsmokers (20 vs 10%, P < 0.001). Among patients in Group 2, 73 NSCLC (19%) were synchronous and 316 (81%) were metachronous with the first cancer. For metachronous cancer, the time interval between the first and second cancer was 87 ± 79 months (range months). The frequency of induction therapy was similar between the groups. Surgical outcomes did not significantly differ between the groups: postoperative mortality was similar (3.6%) and postoperative complications were not significantly different (22 vs 25%, P = 0.28). Histology and pathology of the populations are provided in Table 2. In comparison with Group 1, Group 2 showed a higher proportion of adenocarcinomas (53 vs 40%, P < 0.001), a higher proportion of T1 tumours (27 vs 26%, P = 0.047), a higher proportion of N0 extension (59 vs 52%, P = ), and a higher proportion of Table 1: Patient characteristics and management (n = 4992) Group 1 (n = 4603) Group 2 (n = 389) P-value Time period (98%) 29 (2%) P < (93%) 137 (7%) (92%) 223 (8%) Demographics Female 841 (18.2%) 132 (34%) P < Mean age (years) 61 ± ± 9.4 P < Smoker 4140 (90%) 313 (80%) P < Management Induction therapy 731 (16%) 60 (15%) 0.81 Exploratory thoracotomy 209 (5%) 9 (2%) P < Sub-lobar resection 256 (6%) 44 (11%) (wedge, segmentectomy) Lobectomy (lobe, 2557 (55%) 261 (67%) bilobe and sleeve) Pneumonectomy 1581 (34%) 75 (19%) (including completion) Complete lymphadenectomy 4067 (88%) 358 (92%) Percentage of not available data concerning smoking habits is similar in both groups (P = 0.46). Table 2: Patient histology and pathology Group 1 (n = 4603) Group 2 (n = 389) P-value Histology Adenocarcinoma 1859 (40%) 207 (53%) P < Squamous cell 2129 (46%) 127 (33%) Undifferentiated 373 (8%) 38 (10%) large cell Adenosquamous 119 (3%) 7 (2%) Others 123 (3%) 10 (2%) Tumour Tx 72 (2%) 4 (1%) In situ 14 (1%) 0 (0%) T0 71 (2%) 3 (1%) T1a 643 (14%) 63 (16%) T1b 536 (12%) 43 (11%) T2a 1627 (35%) 166 (43%) T2b 431 (9%) 34 (9%) T3 919 (20%) 58 (15%) T4 268 (6%) 18 (5%) Other nodule In the same lobe 75 (2%) 16 (4%) P < In another lobe 49 (1%) 13 (3%) Nodes Nx 117 (3%) 3 (1%) N (52%) 230 (59%) N1 867 (19%) 76 (20%) N2 1 station 787 (17%) 48 (12%) N2 2 stations 426 (9%) 32 (8%) second nodule in the same lobe (4 vs 2%, P < 0.001). As a consequence of the smaller T size and less-frequent N involvement, pneumonectomy was performed less in Group 2 than in Group 1 (Table 1). Postoperative mortality was similar in both groups,

3 P.B. Pagès et al. / European Journal of Cardio-Thoracic Surgery 3 despite a slightly higher rate of postoperative complications in Group 2. The sites of the previous cancers observed were: bladder (n = 52, 13.4% of Group 2, 1% of the whole population), colorectum (CR, n = 61, 16 and 1%), prostate (n = 57, 15 and 1%), breast (n = 74, 19 and 2%), kidney (n = 37, 17 and 1%), uterus (n = 25, 6% and 1%), skin (n = 25, 6 and 1%), upper gastrointestinal (GI) tract (n = 29, 7 and 1%), thyroid gland (n = 7, 2 and <1%), testis (n = 6, 2 and <1%), melanoma (n = 2, 1 and <1%) and liver, ovary and vagina (n = 1 each, <1 and <1%). The patients characteristics according to the site of the previous malignancy are summarized in Table 3. In univariate analysis, 5- and 10-year overall survival were not significantly different between Groups 1 and 2 (45 and 29% vs 40 and 22%, respectively, P = 0.091) (Fig. 1). The median survivals for Groups 1 and 2 were 47 and 42 months, respectively. Among the 4603 patients included in Group 1, 3309 patients died: 1653 (50%) of NSCLC, 175 (5.3%) of another cancer, 630 (19%) of a non-malignant disease and 851 (27.7%) from unknown reasons (non-significant). Among the 389 patients included in Group 2, 266 patients died: 128 (48%) of NSCLC, 23 (9%) of another cancer, 50 (19%) of a non-malignant disease and 65 (24%) from unknown reasons. Long-term survival did not significantly differ between synchronous and metachronous cancers (5-year survival 33 vs 42%, respectively, P = 0.15). In multivariate analysis, older age, male gender, pneumonectomy, higher T, higher N, incomplete resection, and history of extra-pulmonary and extrapharyngolaryngeal solid malignancy were significantly associated with a worse prognosis (Table 4). When deciphering the respective roles of previous cancers in the adverse prognosis of Group 2 patients, we found three subgroups of previous cancers that impact the prognosis (Tables 5 and 6). Within Group 2, the overall survival was high in the case of breast or uterus malignancies (OR 1.13; 95% CI when compared with Group 1 used as a reference), intermediate in case of skin, kidney, colorectum, and prostate malignancies (OR 1.28; 95CI ), and low in the case of bladder and upper GI malignancies (OR 1.45; 95CI , P < 0.001). DISCUSSION Studying the characteristics and prognosis of lung cancer resection in patients with previous extra-pulmonary and extrapharyngolaryngeal solid malignancy, we found that despite an earlier diagnosis and lesser resection performed, long-term survival was significantly worse than in the overall surgical population. Although the natural history of NSCLC as a second cancer remains unknown, the epidemiology of cancer survivors is widely reported. In the USA, improvements made in early detection and curative treatments have led to a threefold increase in the number of cancer survivors since 1971 [10]. This number is still growing by 2% every year, which shows that the proportion of patients undergoing NSCLC resection after previous extra-pulmonary and extra-pharyngolaryngeal solid malignancy has progressively increased over the last decades [7, 11]. This group of patients includes a higher proportion of female, an older age and a higher proportion of non-smokers. Cancer incidence is increasing with age, explaining that advanced age is found in the group with a previous malignancy, as previously Table 3: Main characteristics of patients included in Group 2 according to the site of previous malignancy Organ Kidney (n = 37) Bladder (n = 52) CR (n = 61) Prostate (n = 57) Breast (n = 74) Uterus Skin Upper GI (n = 29) Time period a (3%) 4% 9 (17%) 32% 2 (3%) 7% 0 1 (1%) 4% 5 (20%) 18% 5 (20%) 18% 5 (17%) 18% (30%) 9% 22 (42%) 17% 24 (39%) 19% 12 (21%) 9% 27 (37%) 21% 10 (40%) 8% 13 (52%) 10% 9 (31%) 7% (67%) 12% 21 (41%) 10% 35 (58%) 17% 45 (79%) 22% 45 (62%) 22% 10 (40%) 5% 7 (28%) 3% 15 (52%) 7% Histology AdK 20 (54%) 16 (31%) 27 (44%) 31 (54%) 55 (74%) 14 (56%) 10 (40%) 17 (59%) SCC 12 (32%) 26 (50%) 30 (49%) 14 (25%) 8 (11%) 8 (32%) 10 (40%) 9 (31%) Others 5 (14%) 10 (19%) 4 (7%) 11 (19%) 11 (15%) 3 (12%) 5 (20%) 3 (10%) Tumour T1 8 (22%) 16 (31%) 19 (31%) 16 (28%) 21 (28%) 6 (24%) 6 (24%) 7 (24%) T2 20 (54%) 26 (50%) 29 (46%) 33 (58%) 39 (53%) 12 (48%) 14 (56%) 15 (52%) T3 + T4 9 (24%) 10 (19%) 13 (21%) 7 (13%) 13 (18%) 7 (28%) 4 (16%) 7 (24%) Nodes N0 19 (51%) 34 (65%) 34 (56%) 37 (65%) 45 (61%) 16 (64%) 15 (60%) 17 (59%) N1 8 (22%) 11 (21%) 14 (23%) 7 (13%) 14 (19%) 5 (20%) 4 (16%) 5 (17%) N2 10 (27%) 7 (14%) 13 (21%) 12 (21%) 14 (19%) 4 (16%) 5 (20%) 7 (24%) Postoperative course Deaths 1 (2.7%) 1 (1.9%) 2 (3.3%) 2 (3.5%) 3 (4.1%) 1 (4%) 0 (0%) 0 (0%) Complications 12 (32%) 20 (39%) 13 (21%) 18 (32%) 15 (20%) 3 (12%) 4 (16%) 10 (35%) AdK: adenocarcinoma; SCC: squamous cell carcinoma. a Percentages in brackets represent the proportion in each site (total of the column = 100%). Percentages in italics represent the proportion in each time period (total of the line = 100%).

4 4 P.B. Pagès et al. / European Journal of Cardio-Thoracic Surgery described by Quadrelli et al. [11]. Finally, the oncological history can potentially explain the higher rate of non-smoking patients in this group [12]. Patients presenting a history of malignancy undergo treatment at earlier stages of their lung cancer compared with those with no history of previous cancer. This situation suggests that the Figure 1: Overall survival according to the group. Table 4: Variables Uni- and multivariate analysis Univariate Multivariate P-value P-value Odds ratio 95% confidence interval Age >62 vs <61 P < P < Male vs female P < Pneumonectomy vs P < P < other T2 vs T1 P < P < T3 + T4 vs T1 P < P < N1 vs N0 P < P < N2 vs N0 P < P < R1 + R2 vs R0 P < P < Group 2 vs Group P < follow-up of patients who have survived a first cancer allows the screening of a second malignant tumour earlier in its development. In the literature, up to 50% of patients operated on for an NSCLC after a previous malignancy presented a pathological stage I[11] when compared with 30% in the overall population. The proportions of adenocarcinomas and squamous cell carcinomas differ according to the study, with a majority of adenocarcinomas in some [11, 13], and a majority of SCC in others [6, 14]. This discrepancy may be explained by the heterogeneity of the population studied, and by the difficulties in classifying some lung tumours as primary or metastatic localizations [15, 16]. Early diagnosis of NSCLC has been recently reported to improve overall and cancer-related survivals in a large screening trial including patients with smoking habits and no history of previous malignancy [17]. However, the link between early diagnosis and survival benefit was not confirmed in our study. On the contrary, patients with a previous malignancy have a worse prognosis than those without, despite this earlier diagnosis. This finding is also debated, as some studies did not find any influence of the oncological history over the prognosis of NSCLC [18, 19]. Furthermore, the 5-year survival of patients with previous malignancy is estimated at around 40%, even in cases of synchronous tumours, suggesting that concomitant diagnoses of pulmonary and extra pulmonary tumours should not rule out sequential resections. Similarly, in most solid malignancies, the suspicion of a unique lung metastasis should not rule out the surgical resection of both tumours [3, 11, 13, 20]. The worse prognosis associated with previous malignancies hides a heterogeneous situation according to the site of the previous tumour, as previously stated [21]. Interestingly, when Massard et al. [6] reported 55 patients with a history of previous extra-pulmonary and extra-cervical malignancies, three subgroups were described: tobacco-induced cancers (kidney, bladder and oesophagus), hormone-dependant cancers (breast, prostate and female genital) and miscellaneous cancers (skin, colon and leukaemia). However, this classification did not lead to prognostic implications, perhaps because tobacco-related cancers were underestimated, and currently include mouth, pharynx, larynx, lung, oesophagus, stomach, liver, pancreas, kidney, urinary bladder, colorectum and uterine cervix [22]. Moreover, prostate cancer had nowadays a better prognosis than breast or cervix cancers [21]. Our prognostic analysis led to a different classification, with higher survival in cases of breast or uterus malignancies, intermediate in case of skin, kidney, colorectum, and prostate malignancies, and low in cases of bladder and upper GI malignancies. This classification suggests that long-term survival is driven by the prognosis of both previous malignancy and NSCLC, with no localization overwhelming the other. Table 5: Overall survival of patients included in Group 2 according the site of previous malignancy Organ Kidney (n = 37) Bladder (n = 52) CR (n = 61) Prostate (n = 57) Breast (n = 74) Uterus Skin Upper GI (n = 29) Median survival (months) 5-year survival (%) 41% 29% 41% 41% 49% 56% 43% 26% 10-year survival (%) 21% 13% 26% 0% 28% 30% 29% 9% CR: colorectum; Upper GI: upper GI tract.

5 P.B. Pagès et al. / European Journal of Cardio-Thoracic Surgery 5 Table 6: Review of the literature n (% of patients) Survival Prognostic factors Massard et al. [6], year survival stage I 47% 5-year survival stage II 31% 5-year survival stage IIIa 17% Koppe et al. [19], year survival 61% Hx better than no Hx Liu et al. [13], (0.6%) 3-year survival 25%, median 14 months LC first better than other cancer first Duchateau et al. [14], (17%) 5-year survival 6.8% Hx better than no Hx Hofmann et al. [7], year survival 12.7% Furak et al. [18], (4.7%) 5-year survival 38%, median 52 months DFI > 36 months better than <36 months Aguilo et al. [22], (11.2%) 5-year survival 9.4%, median 9 months Quadrelli et al. [11], (8.2%) 5-year survival 65.3% This study, (8%) 5-year survival 40%, median 42 months No Hx better than Hx Hx: history of previous cancer; LC: lung cancer; DFI: disease-free interval. This study has several limitations, including the retrospective analysis, bicentric setting, and the lack of information regarding the management of previous solid malignancies. However, the large number of patients in both groups allows powerful comparisons of the study and control groups. With respect to these limitations, this study shows that despite earlier diagnosis, patients with a history of previous extrapulmonary and extra-pharyngolaryngeal solid malignancy have a worse prognosis after surgical resection of NSCLC, when compared with those with no history of malignancy. Conflict of interest: none declared. REFERENCES [1] Cahan WG. Multiple primary cancers, one of which is lung. Surg Clin North Am 1969;49: [2] Sochocky S. Primary carcinoma in lung associated with primary malignancies in other systems. Br J Clin Pract 1977;31:52 6. [3] Jayaprakash V, Cheng C, Reid M, Dexter EU, Nwogu CE, Hicks W et al. Previous head and neck cancers portend poor prognoses in lung cancer patients. Ann Thorac Surg 2011;92: [4] Chuang SC, Scelo G, Tonita JM, Tamaro S, Jonasson JG, Kliewer EV et al. Risk of second primary cancer among patients with head and neck cancers: a pooled analysis of 13 cancer registries. Int J Cancer 2008;123: [5] Deleyiannis FW, Thomas DB. Risk of lung cancer among patients with head and neck cancer. Otolaryngol Head Neck Surg 1997;116: [6] Massard G, Ducrocq X, Beaufigeau M, Elia S, Kessler R, Hervé J et al. Lung cancer following previous extrapulmonary malignancy. Eur J Cardiothorac Surg 2000;18: [7] Hofmann HS, Neef H, Schmidt P. Primary lung cancer and extrapulmonary malignancy. Eur J Cardiothorac Surg 2007;32: [8] Rami-Porta R, Crowley JJ, Goldstraw P. The revised TNM staging system for lung cancer. Ann Thorac Cardiovasc Surg 2009;15:4 9. [9] Kwiatkowski F, Girard M, Hacene K, Berlie J. [Sem: a suitable statistical software adaptated for research in oncology]. Bull Cancer 2000;87: [10] Travis LB. The epidemiology of second primary cancers. Cancer Epidemiol Biomarkers Prev 2006;15: [11] Quadrelli S, Lyons G, Colt H, Chimondeguy D, Silva C. Lung cancer as a second primary malignancy: increasing prevalence and its influence on survival. Ann Surg Oncol 2009;16: [12] Warren GW, Kasza KA, Reid ME, Cummings KM, Marshall JR. Smoking at diagnosis and survival in cancer patients. Int J Cancer 2012; doi: /ijc [13] Liu YY, Chen YM, Yen SH, Tsai CM, Perng RP. Multiple primary malignancies involving lung cancer-clinical characteristics and prognosis. Lung Cancer 2002;35: [14] Duchateau CSJ, Stokkel MPM. Second primary tumors involving nonsmall cell lung cancer: prevalence and its influence on survival. Chest 2005;127: [15] Ye J, Hameed O, Findeis-Hosey JJ, Fan L, Li F, McMahon LA et al. Diagnostic utility of PAX8, TTF-1 and napsin A for discriminating metastatic carcinoma from primary adenocarcinoma of the lung. Biotech Histochem 2012;87:30 4. [16] Bishop JA, Sharma R, Illei PB. Napsin A and thyroid transcription factor-1 expression in carcinomas of the lung, breast, pancreas, colon, kidney, thyroid, and malignant mesothelioma. Hum Pathol 2010;41: [17] National Lung Screening Trial Research Team, Aberle DR, Adams AM, Berg CD, Black WC, Clapp JD, Fagerstrom RM et al. Reduced lung-cancer mortality with low-dose computed tomographic screening. N Engl J Med 2011;365: [18] Furák J, Troján I, Szöke T, Tiszlavicz L, Eller J, Lázár G. Lung cancer as a second primary malignant tumor: prognostic values after surgical resection. Interact CardioVasc Thorac Surg 2008;7:50 3. [19] Koppe MJ, Zoetmulder FA, van Zandwijk N, Hart AA, Baas P, Rutgers EJ. The prognostic significance of a previous malignancy in operable non-small cell lung cancer. Lung Cancer 2001;32: [20] Adebonojo SA, Moritz DM, Danby CA. The results of modern surgical therapy for multiple primary lung cancers. Chest 1997;112: [21] Siegel R, DeSantis C, Virgo K, Stein K, Mariotto A, Smith T et al. Cancer treatment and survivorship statistics, CA Cancer J Clin 2012;62: [22] Aguiló R, Macià F, Porta M, Casamitjana M, Minguella J, Novoa AM. Multiple independent primary cancers do not adversely affect survival of the lung cancer patient. Eur J Cardiothorac Surg 2008;34: