Surgery for Non-small Cell Lung Cancer in Younger Patients: What are the Differences?

Transcription

1 Heart, Lung and Circulation (2015) 24, /04/$ ORIGINAL ARTICLE Surgery for Non-small Cell Lung Cancer in Younger Patients: What are the Differences? Andrea Dell Amore a*, Marco Monteverde b, Nicola Martucci c, Fabio Davoli d, Guido Caroli a, Emanuela Pipitone e, Alessandro Bini a, Franco Stella a, Davide Dell Amore b, Caterina Casadio d, Gaetano Rocco c a Division of Thoracic Surgery, S.Orsola Malpighi University Hospital, Bologna, Italy b Division of Thoracic Surgery, Morgagni Pierantoni Hospital, Forlì, Italy c Division of Thoracic Surgery, Istituto Nazionale dei Tumori, IRCCS Fondazione Pascale, Naples, Italy d Division of Thoracic Surgery, University Eastern Piedmont, Novara, Italy e Department of Formative Science, University of Modena and Reggio Emilia, Modena, Italy Received 20 April 2014; received in revised form 4 June 2014; accepted 2 July 2014; online published-ahead-of-print 14 July 2014 Background Non-small cell lung cancer (NSCLC) in young adults is uncommon. The objective of this study was to evaluate the clinicopathological characteristics, outcomes and prognosis of people younger than 50 years old treated surgically for NSCLC. Methods A retrospective study was conducted using the institutional database of four thoracic surgery units to collect patients with NSCLC younger than 50 years who had undergone surgery. These patients were compared with older patients (>75-years) operated in the same institutions and in the same period. Results We identified 113 young patients and 347 older patients. Younger patients were more likely to be female, non-smokers, with fewer comorbidities. Younger patients were more likely to be symptomatic at the time of diagnosis. Risk factors for poor prognosis in younger patients were T-stage, and disease-free-interval less than 548 days. Kaplan-Meier analysis showed a lower five-year survival in older patients compared with the younger ones (66% vs 38%, p = 0.001). Conclusions In conclusion NSCLC in younger patients has some distinct clinicopathological characteristics. The overallsurvival of young patients is better than in older patients. Young patients receive more complete and aggressive treatment that could explain better survival. Further prospective studies with larger patient populations are required, to clarify the biological and genetic variance of NSCLC in younger patients. Keywords Lung cancer Younger patients Thoracic surgery Lung surgery Lung cancer prognosis Introduction Non-small-cell lung cancer (NSCLC) is the second most common cancer in men and the third in women in Italy, with about 30,384 new cases per year in men and 6784 in women [1,2]. Lung cancer is mainly a tumour of the elderly with a peak of incidence between the sixth and seventh decades of life. Only 5% to 10% of NSCLC are diagnosed in individuals younger than 50 years of age. The estimated probability of developing invasive lung cancer within a selected age is 0.03% in males and females between birth and 39 years old, 0.91% in male and 0.76% in female between *Corresponding author at: Thoracic Surgery Unit, S.Orsola Malpighi Hospital, University of Bologna, Via Massarenti 9, Bologna, Italy. Tel.: ; fax: , dellamore76@libero.it 2014 Australian and New Zealand Society of Cardiac and Thoracic Surgeons (ANZSCTS) and the Cardiac Society of Australia and New Zealand (CSANZ). Published by Elsevier Inc. All rights reserved.

2 Lung cancer surgery in the young to 59 years old. After 70 years old the probability is 6.7% in males and 5% in females [3,4]. Although various reports about lung cancer in the young have been published, there are still conflicting data on whether young people with NSCLC have a better or worse prognosis compared with the older population [5 8]. The objective of this multiinstitutional study was to evaluate the clinicopathological characteristics, outcomes and prognosis of people younger than 50 years old treated surgically for NSCLC. Materials and Methods We identified 113 patients younger than 50 years old (group- A) who had undergone intended curative lung resection for NSCLC (neuroendocrine tumours were excluded) between January 2000 and December 2012 in four different thoracic surgery units. These patients were matched to patients older than 75 years (group-b) who had undergone intended curative lung resection for NSCLC and whose details were obtained from the same institutional registries in the same period of time (347 patents). Preoperative characteristics of both populations are reported in Table 1. Preoperative assessment included careful medical history, physical examination, routine blood test, electrocardiogram and spirometry with blood gas analysis. Echocardiography and myocardial scintigraphy were performed if the patient had a history of angina or an abnormal baseline electrocardiogram. CO diffusion capacity and lung perfusion scintiscan were performed in the event of poor pulmonary function (FEV 1 < 1liter) or when a major resection was planned. The histological diagnosis of the tumour was preoperatively obtained in the majority of the patients by trans-bronchial needle aspiration (TBNA), brushing or bronchial-alveolar lavage (BAL) during flexible and/or rigid bronchoscopy, or alternatively by trans-parietal CT guided biopsy. The clinical staging process started with total-body computed tomography (CT) followed, as of 2005, by positron emission tomography (PET) even in the event of no evidence of lymph node enlargement on the CT-scan. Before 2005, all patients with enlarged mediastinal lymph nodes on CT-scan (short axis >1 cm) underwent mediastinoscopy to complete clinical staging. Bone scintigraphy was used until the advent of PET. After 2005, only patients who had PET positive mediastinal lymph nodes underwent mediastinoscopy or EBUS/EUS biopsy to complete clinical staging. Surgical lymphadenectomy was defined as follows: on the right side lymphadenectomy was defined as the removal of all the 2R, 4R, 7,8,9 lymph nodes, on the left side the 5,6,7,8,9 lymph nodes were removed. The 4L station was dissected only in the event of high suspicion of neoplastic involvement, if preoperative medistinoscopy and/or EBUS were negative. Details of the major post-operative complications were collected from the clinical report of each patient. The long-term follow-up was obtained through examination, wherever possible, or by telephone interview of the patients themselves or of a family member. In the remaining cases, follow-up was obtained from the patient s general practitioners or registry offices. Statistical Analysis Continuous variables are reported as mean SD whereas categorical variables as percentages. Comparisons between group A and B were performed using the unpaired t-test for continuous variables and the Fisher exact test for discrete variables. The variables showing a p-value <0.1 on univariate analysis were used as independent variables in multivariate analysis for younger patients. Variables related to surgical Table 1 Preoperative patient s profile. Variables Group A Group B p-value Patients N/A Age SD (25 50) SD (75 89) N/A Gender Male 66 (58) 274 (79) Female 47 (42) 73 (21) Smoking history Never 26 (23) 41 (12) Active 71 (63) 103 (30) Former 16 (14) 203 (59) COPD/asthma 9 (8) 112 (32) Low respiratory reserve 7 (6) 96 (28) Major morbidity 21 (19) 216 (62) Prior malignancy 16 (14) 89 (26) ns Symptoms at presentation 97 (86) 166 (48) Neoadjuvant Chemotherapy 21 (19) 7 (2) Neoadjuvant Radiotherapy 6 (5) 1 (0.2) 0.001

3 64 A. Dell Amore et al. treatment (i.e. resection type and extended resection) were dropped from analyses due to colinearity with tumour stage, tumour size and lymph nodes status. Stepwise Cox proportional hazards model was used to estimate hazard ratios (HR) and 95% confidence interval (95%-CI). The adjusted survival curves were calculated as a weighted average of the individual survival curves [9]. The endpoints of analyses were defined in terms of disease-free survival (DFS) and overall survival (OS). For all the analyses, two-tailed p-values of less than 0.05 were considered significant. Survival analysis was conducted according to the Kaplan-Meier method. Operative deaths were excluded from long-term statistical evaluation. Statistics were performed using SPSS version 11.0 for Windows (SPSS, Inc, Chicago, Ill) and SAS software version (9.2; SAS Institute, Cary, NC). Results Patients The younger group included 66 males and 47 females with a ratio of 1.4:1. In the older group, females represented 19% of all the population versus 42% in the younger group (p = 0.001) with a ratio of 3.8:1. Compared with group-b, the younger patients had less morbidity, better respiratory reserve and different smoking habits (Table 1). Of younger patients, 86% had at least one symptom at the time of the diagnosis, and in 92% of the cases specific diagnostic tests were performed because of symptoms. Cough with chest x-ray infiltrates, initially judged as broncopneumonia, were the most frequent symptoms (78%) followed by haemoptysis (12%) and chest pain (10%). The diagnosis of NSCLC was obtained after an average of 3.9 months (range 1-9 months). In older patients, the most frequent symptom was cough (66%), followed by chest pain (18%) and haemoptysis (16%), but 52% of these patients were completely asymptomatic at the time of diagnosis. The diagnosis of NSCLC was made in older patients after an average of 2.6 months (range months). Therefore, younger patients were more likely to be symptomatic at the time of diagnosis (86% vs 52%, p < 0.001) and the median time until presentation for treatment was significantly longer for younger patients (3.9 vs 2.6 months, p = 0.03). Younger patients were also more likely to receive neoajuvant chemotherapy (21% vs 2%, p < 0.001) and neoadjuvant radiotherapy (5% vs 0.2%, p < 0.001). Pathological Staging and Histology Pathological staging and histology of both groups are summarised in Table 2. There were no significant differences in the distribution of tumour histology between groups. Adenocarcinoma was the most common histology in both groups, followed by squamous cell carcinoma. All patients were restaged using the seventh version of the TNM [10]. The staging distribution for the two groups was significantly different. The younger group tended to be higher stage than older patients. Stage-I was more frequent in older patients (59% vs 39%). Conversely stage-iii was more Table 2 Post-operative staging and histology. Variables Group A frequent in younger patients (30% vs 14%, p < 0.001). Likewise, pn 2 -disease was most represented in younger patients (25% vs 6%, p < 0.001). Post-operative Results All patients had an open-thoracotomy approach with complete thoracic lymphadenectomy. In group-a the most frequent operation was lobectomy that was performed in 74 patients (66%). Pneumonectomy was performed in 30 patients (27%), 19 left and 11 right respectively. Bilobectomy was performed in nine patients (8%). Older patients received much less pneumonectomy (9%) and much more limited resections such as segmentectomy and wedge resection (22%). The older patients had a higher risk of postoperative complications (49% vs 20%, p < 0.001) and thus a longer in-hospital length of stay (9.8 vs 7.3 days, p = 0.03). In younger patients the most frequent complications were respiratory (prolonged air leak, atelectasis and pneumonia), while cardiac complications were the principal causes of morbidity in older patients (arrhythmias and myocardial ischaemia) followed by pneumonia and renal failure. In-hospital mortality was 1.8% (2/113 patients) in younger patients and 6.9% in older patients (24/347 patients) (Table 3). Follow-up Results Group B p-value pn stage N 0 63 (56) 240 (69) N 1 22 (20) 87 (25) N 2 28 (25) 20 (6) <0.001 Pathological TNM stage Ia 27 (24) 113 (33) Ib 17 (15) 91 (26) IIa 14 (12) 60 (17) IIb 14 (12) 35 (10) IIIa 30 (27) 45 (13) IIIb 4 (3.5) 3 (0.8) IV 7 (6.2) - <0.001 Histology Adenocarcinoma 72 (64) 158 (46) Squamous cell carcinoma 22 (46) 153 (44) Others 19 (17) 36 (10) Ns Follow-up data were complete in all patients. The mean follow-up time was SD months in younger patients without difference between groups (Table 3). At the end of the study 75 patients were still alive in group-a and 166 patients in group-b (p = 0.02). Younger patients were more likely to receive adjuvant chemotherapy compared with the elderly (52% vs 11%, p = 0.001). Five-year overall Kaplan- Meier survival was significantly higher in the younger group

4 Lung cancer surgery in the young 65 Table 3 Post-operative results, follow-up results, disease recurrence. Variables Group A Group B p-value Overall-morbidity 22 (20) 169 (49) In-hospital mortality 2 (1.8) 24 (6.9) 0.02 Mean hospital LOS (days) 7.3 3SD 9.8 4SD 0.03 Mean FU (months) SD SD Ns (Range 8 152) (Range 8 97) FU Overall Mortality 36 (32) 157 (49) 0.02 Adjuvant Therapy 58 (52) 36 (11) years overall survival (%) Disease relapse 50 (45) 122 (38) Ns Local 19 (38) 79 (65) Systemic 31 (62) 43 (35) Mean time freedom from disease (months) ,6 SD (Range 5 88) SD (Range 7 83) Ns compared with the older group (66% vs 38%, p = 0.001) (Fig. 1). The survival analyses of the younger group of patients showed the following results. The one, three and five-year overall-survival was 89%, 72% and 66% respectively (Fig. 1). There was no statistical difference between groups in terms of disease recurrence, but local recurrence was more frequent in group-b while systemic recurrence in group-a (p = 0.001) (Table 3). The five-year overall-survival for TNM stage-i was 88%, stage-ii 57% and stage-iii 49% in group-a vs 45% for stage-i, 39% for stage-ii and 18% for stage-iii in group B (p < ) (Fig. 2). Factors influencing long-term survival in young patients on univariate and multivariate analyses are reported in Table 4. Multivariate Coxregression analyses indicated that tumour size (Fig. 3) and disease-free interval less than 548 days (Fig. 4) were independent negative prognostic factors for young patients with surgically treated NSCLC. Figure 1 Overall survival for Group-A and Group-B Compared using long-rank test (p = 0.001). Discussion The prevalence of NSCLC is higher in the sixth and seventh decades of life, while the occurrence among young adults is uncommon [4,5,11]. The main risk factor for NSCLC is tobacco smoking, and the reason why this tumour is more frequent in older people is probably related to the time required for the toxin (smoke, pollution, etc) to induce genetic alterations that will ultimately lead to cancer [5,12]. Following this concept, many published studies suggest that NSCLC in young people represents a distinct clinicopathological entity with different cancer behaviour [5 7,11,13]. In order to support this statement, we compared two different groups of patients with NSCLC representing the extremes in age distribution, patients younger than 50 years old and patients older than 75 years. Indeed it has been reported that young patients are more likely to have an advanced tumour stage at diagnosis, different gender distribution (female predominance), different smoking habits, different disease presentation and different histology patterns (adenocarcinoma prevalence) [5,13 15]. In our study, we confirmed that young patients were more frequently female, they have different smoking habits (fewersmokers), were more frequently symptomatic at diagnosis, and had a higher pathological stage compared with the older group of patients. We found a prevalence of adenocacinoma in both groups without difference between them. Bryant et al [6] highlighted that younger patients presented to thoracic surgeons later than older patients. We found the same tendency; indeed, in our study, the median time until presentation for treatment was 3.9 months in younger patients versus 2.6 months in older patients (p = 0.03). In agreement with Bryant, the explanation for this is that general practitioners or physicians tend to consider the diagnosis and suspicion of cancer at the end of their diagnostic process for young patients. At the same time, young people underestimate

5 66 A. Dell Amore et al. Figure 2 Kaplan-Meier estimates of overall survival in group-a and group-b for TNM stage. symptoms and thus delay visits and are less likely to go for routine checkup compared with older people. Younger patients were thus more likely to be symptomatic at the time of diagnosis (86% vs 52%, p < 0.001). The second aspect we investigated is whether these differences between young and older patients with NSCLC could have some implication related to the prognosis. The literature contains conflicting data about the prognosis of young patients with NSCLC. In 2010, on analysing the SEER database (Surveillance, Epidemiology, and End Results), Subramamian et al [5] found that younger age was clearly associated with reduced risk of mortality due to NSCLC. Also in the study by Tian et al [14] young adults with NSCLC have a better prognosis than older patients. On the contrary, in a nested case control study by Bryant et al [6], patients younger than 45 years of age showed a five-year overall survival of 51%, statistically worse than the five-year overall survival of 62% in patients older than 45 years (p = 0.04). They concluded that this difference could be related to the delay in diagnosis and treatment, but also to the different genetic and molecular characteristics of NSCLC in the young patients, especially in those who were never smokers. However, in this study the distinction between the two groups could be partially distorted by an age-related confounding area. Indeed, patients aged between Table 4 Univariate and multivariate Cox-Regression analysis of survival in young patients (Group-A) with NSCLC. Variables DFS OS HR (95% CI) P HR (95% CI) P Univariate analysis Smoking history 2.513( ) ( ) Type of resection 2.340( ) ( ) TNM stage 3.168( ) ( ) < Disease recurrence 2.731( ) ( ) 0.01 Tumour size dichotomous 3.224( ) ( ) < pn status 2.625( ) ( ) Neoadjuvant radiotherapy 7.48( ) ( ) 0.02 Multivariate analysis * Tumour size dichotomous (1 + 2 vs 3 + 4) 2.295( ) ( ) < Disease free interval less than 548 days N/A 0.03( ) < * :adjusted for age, sex, smoking-status and tumour-stage.

6 Lung cancer surgery in the young 67 Figure 3 Kaplan-Meier estimates and adjusted estimates of overall survival of recurrence disease time (days). Dashed lines represent the unadjusted (Kaplan-Meier) survival estimates for the two groups (r-fu = 0: disease recurrence < 548 days, r-fu = 1: disease recurrence >548 days). Solid lines represent the adjusted estimates as determined from the appropriate Cox regression model. Figure 4 Kaplan-Meier estimates and adjusted estimates of overall survival of pt (dichotomous). Dashed-lines represent the unadjusted (Kaplan-Meier) survival estimates for the two groups (T 1 + T 2 vs T 3 + T 4 ). Solid lines represent the adjusted estimates as determined from the appropriate Cox regression model. 45 to 50 years old may have a biological behaviour more similar to the young defined group. The study by Kuo et al [7] compared two very different groups of patients: the very young (<40 years old) and the very old (>80 years old). In this study they showed that younger group received more aggressive treatment and had better survival. In our experience too, the higher rate of neoadjuvant and adjuvant therapy in the young group could reflect the higher proportion of advanced stages, which is the prime indication for such treatment, but also the ability of young patients to better tolerate multimodality treatment. Various studies reported similar results but with no clear effects on survival [7,11,13,16]. In our study we found a better long-term survival for younger patients: five-year overall-survival was 66% in younger patients versus 39% in the older subjects (p = 0.001, Fig. 1). The same trend is clearly evident stratifying the survival for TNM stage (Fig. 2). These results are explained by a clear patient selection-bias. Older patients are sicker, in particular with multiple cardiac and respiratory comorbidities [7,16,17]. Indeed, the difference between survival for stage-i vs stage-ii is much more evident in younger group (88% vs 57%, p < 0.01) than in older group (45% vs 39%, p = 0.06). Age alone plus these factors strongly influence life span and thus overall survival regardless the tumour stage. Moreover, in spite of a delayed diagnosis and treatment, younger patients are more likely to undergo more complete, radical, aggressive and combination modalities treatment that older patients [7,13,16]. These differences may explain the different disease recurrence patterns between groups, with a more frequent local recurrence in the older group and systemic recurrence in the younger group (p = 0.001). In our study older patients received less neoadjuvant therapy, more limited resections and less adjuvant therapy. All these factors probably balance the delayed diagnosis/treatment, the higher tumour stage and the more aggressive cancer behavior of their younger counterpart. At the univariate and multivariate analyses we found that, T-stage and disease free interval less than 18 months were factors related with a poor long-term outcome. Tian et al [14] reported the advanced stage (stage I-II vs III), and histology different from squamous cell carcinoma as risk factors for poor survival. In our study, histology had no influence on survival, but we do not have sufficient specific data to comment on this result. Mauri et al [11] reported the low performance status, the higher stage and a disease free interval of less than 12 months as risk factors for poor survival. In their, and our opinion these factors represent the multiorgan integrity of the host, the disease bulk and its aggressive malignant behaviour. The combinations of these three factors are probably the main determinants of long-term survival and treatment response. The present study has several limitations. First of all the limited number of enrolled patents, particularly in the younger group, the retrospective nature of the study, the fact that the survival analyses were not cancer related, and the lack of precise data about chemotherapy regimens and/or molecular target therapy. Conclusion We confirm that NSCLC in younger patients has some distinct clinicopathological characteristics, but we failed to establish whether the biological behaviour is unequivocally more aggressive. In our study, the overall survival of young

7 68 A. Dell Amore et al. patients is still better than in older patients, despite the delay in diagnosis, that means a more advanced preoperative stage, and thus a delay in treatment. A suspicion of lung cancer should be considered earlier in the diagnostic work-up of a young patient with pulmonary consolidation little responsive to antibiotics therapy or with haemoptysis, even in those who never smoke and especially in females. Of course, young patients receive more complete and aggressive treatment that could explain better survival. Adenocarcinoma is now the most frequent NSCLC that we find in young and even older patients. Further prospective studies with larger patient populations are required, to clarify the biological and genetic variance of NSCLC in younger people and thus to relate these results with patient prognosis and to develop new effective treatment strategies. Conflict of Interest None declared. References [1] Mangone L, Minicozzi P, Vicentini M, Gicomin A, Caldarella A, Cirilli C, et al. Key factors influencing lung cancer survival in northern Italy. Cancer Epidemiol 2013;37: [2] Inghelmann R, Grande E, Francisci S, Verdecchia A, Micheli A, Baili P, et al. Regional estimates of lung cancer burden in Italy. Tumori 2007;93 (4): [3] Sant M, Minicozzi P, Allemani C, Cirilli C, Federico M, Capocaccia R, et al. Regional inequalities in cancer persist in Italy and can influence survival. Cancer Epidemiology 2012;36: [4] Siegel R, Naishadham D, Jemal A. Cancer statistics, CA Cancer J Clin 2012;62: [5] Subramanian J, Morgensztern D, Goodgame B, Baggstrom MQ, Gao F, Piccirillo J, et al. Distinctive characteristics of non-small cell lung cancer (NSCLC) in the young. A surveillance, epidemiology and end results (SEER) analysis. J Thorac Oncol 2010;5:23 8. [6] Bryant AS, Cerfolio RJ. Differences in outcomes between younger and older patients with non-small cell lung cancer. Ann Thorac Surg 2008;85: [7] Kuo C, Chen Y, Chao J, Tsai C, Perng R. Non-small cell lung cancer in very young and very old patients. Chest 2000;117: [8] Quaglia A, Tavilla A, Shack J, Brenner H, Jassen-Heijnen M, Alemanni C, et al. EUROCARE Working Group. The cancer survival gap between elderly and middle-aged patients in Europe is widening. Eur J Cancer 2009;45: [9] Nieto FJ, Coresh J. Adjusting survival curves for confounders: A review and a new method. Am J Epidemiol 1996;143: [10] Detterbeck FC, Boffa DJ, Tanoue LT. The new lung cancer staging system. Chest 2009;136: [11] Mauri D, Pentheroudakis G, Bafaloukos D, Pectasides D, Samanthas E, Efstathiou E, et al. Non-small cell lung cancer in the young: a retrospective analysis of diagnosis, management and outcome data. Anticancer Res 2006;26: [12] Kreuzer M, Kreienbrock L, Gerken M, Heinrich J, Bruske-Hohlfeld I, Muller KM, et al. Risk factors for lung cancer in young adults. Am J Epidemiol 1998;147: [13] Inoue M, Okumura M, Sawabata N, Miyaoka E, Asamura H, Yoshino I, et al. Clinicopathological characteristics and surgical results of lung cancer patients aged up to 50 years: The Japanese Lung Cancer Registry Study Lung Cancer 2014;83: [14] Tian D, Liu H, Zhang L, Yin H, Hu Y, Zhao H, et al. Surgery for young patients with lung cancer. Lung Cancer 2003;42: [15] Skarin AT, Herbst RS, Leong TL, Bailey A, Sugarbaker D. Lung cancer in patients under age 40. Lung Cancer 2001;32: [16] Aggarwal C, Langer CJ. Older age, poor performance status and major comorbidities: how to treat high-risk patients with advanced nonsmall cell lung cancer. Curr Opin Oncol 2012;24: [17] Dell Amore A, Monteverde M, Martucci N, Sanna S, Caroli G, Stella F, et al. Early and long-term results of pulmonary resection for non-smallcell lung cancer in patients over 75 years of age: a multi-institutional study. Interact CardioVasc Thorac Surg 2013;16:250 6.