Prognosis of non-small-cell lung cancer patients with positive pleural lavage cytology

Transcription

1 Interactive CardioVascular and Thoracic Surgery 20 (2015) doi: /icvts/ivv047 Advance Access publication 11 March 2015 ORIGINAL ARTICLE THORACIC Cite this article as: Nakao M, Hoshi R, Ishikawa Y, Matsuura Y, Uehara H, Mun M et al. Prognosis of non-small-cell lung cancer patients with positive pleural lavage cytology. Interact CardioVasc Thorac Surg 2015;20: a Prognosis of non-small-cell lung cancer patients with positive pleural lavage cytology Masayuki Nakao a, *, Rira Hoshi b, Yuichi Ishikawa c, Yosuke Matsuura a, Hirofumi Uehara a, Mingyon Mun a, Ken Nakagawa a and Sakae Okumura a Department of Thoracic Surgical Oncology, Cancer Institute Hospital, The Japanese Foundation for Cancer Research, Tokyo, Japan b Department of Cytology, Cancer Institute Hospital, The Japanese Foundation for Cancer Research, Tokyo, Japan c Department of Pathology, Cancer Institute Hospital, The Japanese Foundation for Cancer Research, Tokyo, Japan * Corresponding author. Cancer Institute Hospital, Ariake, Koto-ku, Tokyo , Japan. masayuki.nakao@jfcr.or.jp (M. Nakao). Received 30 September 2014; received in revised form 15 January 2015; accepted 23 January 2015 Abstract OBJECTIVES: Positive pleural lavage cytology (PLC) is considered as a precursor condition of pleural dissemination (PD) or malignant pleural effusion (PE), and one of the poor prognostic factors in surgically resected non-small-cell lung cancer (NSCLC) patients. Although PD and PE are classified as M1a, PLC does not contribute to the tumour, node and metastasis (TNM) classification of the Union Internationale Contre le Cancer. This study aimed to evaluate the prognostic effect of positive PLC status in surgically resected NSCLC patients compared with PD and/or PE. We also aimed to consider the contribution of positive PLC status to the TNM classification. METHODS: We reviewed 1572 consecutive patients with completely resected NSCLC, and analysed the relationship between PLC status, other clinicopathological factors and prognosis. The survival rates of 45 patients with PD and/or PE were also investigated. RESULTS: Positive preresection PLC ( pre-plc) status was observed in 56 patients. Pre-PLC status was significantly associated with other clinicopathological factors. Positive pre-plc patients exhibited a worse 5-year overall survival (50.6%) compared with negative pre-plc patients (78.0%), but better survival than PD and/or PE patients (21.0%). Prognosis of positive pre-plc patients was equal to that of pt3, negative pre-plc patients; survival equality was observed when patients were stratified according to pn0, pn1 and pn2. CONCLUSIONS: Positive pre-plc had the significant prognostic effect in surgically resected NSCLC patients. However, it is not a contraindication for surgical resection, unlike PD and/or PE. Our data suggest that positive pre-plc should be classified as pt3 in next TNM classification. Keywords: Lung cancer Surgery Pleural effusion Pleural dissemination INTRODUCTION The first report of pleural lavage cytology (PLC) in non-small-cell lung cancer (NSCLC) was published by Spjut et al. [1] in Since then, although there have been some differences regarding the sampling and specimen handling methods used, many studies have suggested that PLC status is a significant prognostic factor in surgically resected NSCLC patients [2 13]. We previously assessed the prognostic effect of positive PLC status, which focused on recurrence patterns in early-stage disease in 2007 [2]. The presence of cancer cells in pleural lavage indicates the spread of microscopic intrapleural cancer resulting from visceral pleural invasion, vascular invasion or lymphatic permeation of the primary lesion. It is considered to be the precursor of pleural dissemination (PD) or malignant pleural effusion (PE). PD and PE are classified as M1a in the tumour, node and metastasis (TNM) classification of the Union Internationale Contre le Cancer (UICC) because of their strong effect on patient survival [14 16]. They are grouped in stage IV, and are considered to be contraindications for surgical resection. However, PLC status does not contribute to the TNM classification, and the appropriate management of positive PLC patients is controversial. This study aimed to evaluate the prognostic effect of positive PLC in surgically resected NSCLC patients compared with PD and/ or PE. We also aimed to consider the contribution of positive PLC status to the TNM classification. PATIENTS AND METHODS Patients A total of 1572 consecutive NSCLC patients were enrolled in this retrospective study. They underwent complete surgical resection ORIGINAL ARTICLE The Author Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.

2 778 M. Nakao et al. / Interactive CardioVascular and Thoracic Surgery at our institution without any preoperative anticancer treatment between January 1991 and December Complete resection was defined as either lobectomy or more extensive resection with systematic ipsilateral hilar and mediastinal lymph node dissection through thoracotomy or thoracoscopic procedure, and an absence of residual cancer both macroscopically and histologically. In addition, 45 NSCLC patients with PD and/or PE were also investigated to compare the survival probability rates. A total of 23 patients of clinical (c) Stage I, 8 patients of cstage II and 11 patients of cstage III were included. Because PD and PE were found unexpectedly during an operation and the disseminated nodules were localized, patients underwent lobectomy or more extensive resection and resection of disseminated nodules for macroscopically complete resection during the same period. Three patients of cstage IV also underwent macroscopically complete resection. Patients who underwent exploratory thoracotomy or only partial resection of the main lesions were excluded. Adjuvant chemotherapy was performed, according to patient s general condition and will, in a part of patients with stage II III disease using platinum-based doublet regimens and a part of patients with stage IB disease using oral UFT. All patients were followed up on an outpatient basis at a minimum of every 6 months by physical check-up, chest computed tomography and laboratory testing and every 12 months by systemic examination until 5 years after surgery. We reviewed the medical records of all patients for the following clinicopathological factors: age, gender, cstage, pathological T and N (pt and pn) classification, histological type, pleural invasion, PLC status and adjuvant chemotherapy. Histopathological evaluation Histological classification of the resected specimens was determined according to the World Health Organization International Histological Classification of Tumors [17]. Clinical and pathological stage was determined according to the UICC TNM Classification for Lung and Pleural Tumors [14, 15]. Pleural invasion was defined by detailed evaluation of sliced specimens stained using the Elastica van Gieson method, and was classified as follows: Pl0, tumour within the subpleural lung parenchyma or superficial invasion into the pleural connective tissue beneath the elastic layer; Pl1, tumour invasion beyond the elastic layer; Pl2, tumour invasion to the pleural surface; and Pl3, tumour invasion to any part of the parietal pleura. Lymphatic permeation and vascular invasion were excluded from the analysis because of a lack of data in many patients. Pleural lavage cytology Immediately after thoracotomy, the pleural cavity was washed with 100 ml of physiological saline solution before any surgical manipulation. The surgeon avoided touching the pleural surface to obtain only desquamated cells. The fluid was carefully removed, and then centrifuged at 1500 rpm for 5 min. The resulting sedimented material was stained using the Papanicolaou method, and the results of the cytological examination were divided into three categories: negative, suggestive and positive. Only samples with positive findings were defined as positive preresection PLC ( pre-plc). In cases with pleural adhesions, they were performed after removing adhesions by necessity. After tumour resection and systematic lymph node dissection, the pleural cavity was washed with ml of physiological saline solution before wound closure. Fifty millilitres of the saline wash solution was removed, and evaluated in the same manner for pre-plc. Only samples with positive findings were defined as positive post-resection PLC (post-plc) [2]. Statistical analysis The Fisher s exact test was used to analyse the correlation between PLC status and other clinicopathological factors. Cox proportional hazard regression model was used to identify statistically significant differences in survival and to estimate hazard ratios and 95% confidence intervals. Survival curves were plotted using the Kaplan Meier method, and the statistical significance of differences among subgroups was determined using the log-rank test. The end-point for analyses was overall survival, which was measured from the date of surgery to the date of death by any cause. The last follow-up observation was censored when the patient was alive or lost to the follow-up. Cancer recurrence was divided into three categories according to the site of initial recurrence: locoregional, pleural and distant. Locoregional recurrence was defined as disease in the ipsilateral lung, and hilar and mediastinal lymph nodes. Pleural recurrence was defined as disease within the hemithoracic pleura, and distant recurrence was defined as any other disease. All P-values were two-sided, and P < 0.05 were considered to indicate a statistically significant difference. We used the statistical analysis software (Dr SPSS II for Windows, Standard Version 11.0, SPSS, Inc., Chicago, IL, USA) for all analyses. Clinicopathological and survival data were collected based on a retrospective chart review and no personally identifiable information was included. Data collection and analysis were approved and the need for obtaining informed consent from each patient was waived by the Institutional Review Board. RESULTS Positive prepleural lavage cytology and clinicopathological features The clinicopathological characteristics of the 1572 NSCLC patients are given in Table 1. The pre-plc status was assessed in all 1572 patients, and a positive result was observed in 56 (3.6%). The post-plc status was assessed in 1135 patients because it was not a mandatory procedure in early phase of the study period, and a positive result was observed in only three patients. Because of the lack of data from many cases and the small number of positive patients, post-plc status was excluded from subsequent analyses. The correlations between pre-plc status and other clinicopathological factors are given in Table 2. Positive pre-plc was significantly correlated with a higher pt (P < 0.001) and pn (P < 0.001) classification, as well as the presence of pleural invasion (P < 0.001). Although adenocarcinoma was the major histology of positive pre-plc patients, the same trend was also observed in negative pre-plc patients. Effect of prepleural lavage cytology on survival The median length of the overall follow-up for the censored patients was 5.6 years. The correlations between pre-plc status and disease recurrence are given in Table 3. Recurrence was

3 M. Nakao et al. / Interactive CardioVascular and Thoracic Surgery 779 Table 1: Clinicopathological characteristics of the patients Table 2: Correlations between pre-plc status and other clinicopathological factors Patients (N = 1572) n % Age (years) Median (range) 65 (16 86) Sex Male Female cstage I II III pstage I II III Histological type Adenocarcinoma Squamous cell ca Adenosquamous ca LCNEC Large cell ca Others Pleural invasion pl pl pl pl Pre-PLC Positive Negative Post-PLC a Positive Negative Adjuvant Yes chemotherapy No a Data not available for 437 patients. LCNEC: large cell neuroendocrine carcinoma; PLC: pleural lavage cytology. observed more frequently (P < 0.001) and the rate of pleural recurrence at the initial site was higher (P < 0.001) in pre-plc-positive patients. There were no significant differences in the sites of recurrence other than pleural recurrence between groups (P = 0.133). The prognostic significance of positive pre-plc status and other clinicopathological features for overall survival is presented in Table 4. Univariate analysis revealed that patient survival was significantly associated with pre-plc status, as well as age, gender, pt classification, pn classification, histological type and pleural invasion. The positive pre-plc group had a relatively worse 5-year overall survival rate (50.6%) than the negative pre-plc group (78.0%). Multivariate analysis was performed among pre-plc status and other three factors; pt classification, pn classification and pleural invasion. They were the prognostic factors investigated in univariate analyses, and had close correlation with pre-plc status. The result showed that pre-plc status was not an independent predictor of poor outcome (P = 0.289). Pre-PLC Negative (N = 1516) Positive (N = 56) P-values Age Sex Female Male cstage I II III pt classification T T T T4 8 0 <0.001 pn classification N N N <0.001 Histological type Adenocarcinoma Others Pleural invasion pl pl pl pl <0.001 Adjuvant chemotherapy Yes No PLC: pleural lavage cytology. Table 3: Correlations between pre-plc status and initial recurrent sites Pre-PLC Negative (N = 1516) Positive (N = 56) P-values Recurrent cases <0.001 Pleural recurrence <0.001 Other sites recurrence a Locoregional recurrence 67 2 Distant recurrence ORIGINAL ARTICLE a Data not available for 10 patients. Positive prepleural lavage cytology and pleural dissemination or malignant pleural effusion survival Survival curves of the positive and negative pre-plc groups are shown in Fig. 1. To compare the survival probability rates, the survival curve of 45 patients with PD and/or PE was interposed. The positive pre-plc group exhibited better 5-year overall survival rate compared with the PD and/or PE group (50.6 vs 20.9%, P = 0.006). Incorporating pre pleural lavage cytology into tumour, node and metastasis classification Figure 2 shows the survival curves of positive pre-plc patients according to their pt classification. Because of the small number of patients, pt1 patients were assessed together with pt2. The 5-year overall survival rates were 59.4% for 39 patients with pt1 2, compared with 31.9% in 17 patients with pt3 tumours. Because

4 780 M. Nakao et al. / Interactive CardioVascular and Thoracic Surgery Table 4: Prognostic significance of pre-plc status and other clinicopathological factors for overall survival Number Univariate analysis Multivariate analysis 5-ys (%) P-values HR 95% CI P-values Age (years) <0.001 Sex Female Male <0.001 pt classification T T < <0.001 pn classification N N < <0.001 Histological type Adenocarcinoma Others < Pleural invasion pl pl < Pre-PLC Negative Positive < Adjuvant chemotherapy Yes No ys: 5-year overall survival rate; PLC: pleural lavage cytology; HR: hazard ratio; CI: confidence interval. Figure 1: Overall survival curves of the positive pre-plc, negative pre-plc and PD/PE groups. PLC: pleural lavage cytology; PD: pleural dissemination; PE: malignant pleural effusion. there were no statistically significant differences in survival between the two groups of positive pre-plc patients (P = 0.082), we considered they could be grouped in the same pt classification. Figure 3 shows a comparison of the survival curves of positive and negative pre-plc patients stratified by pt classification. The 5-year overall survival rates of the negative pre-plc patients were 88.9% in pt1, 73.9% in pt2 and 54.3% in pt3. The prognoses of the three groups of negative pre-plc patients were significantly different (pt1 vs pt2, P < 0.001; pt2 vs pt3, P < 0.001; pt1 vs pt3, P < 0.001). The overall survival rate of the positive pre-plc group was equal to that of pt3 negative pre-plc patients (P = 0.583). The results of the same comparisons stratified according to pn classification are shown in Fig. 4. The overall survival rates of positive pre-plc patients were also equal to those of pt3 negative Figure 2: Overall survival curves of positive pre-plc patients according to pt classification (pt1-2 vs pt3). pre-plc patients in pn0 (P = 0.596), pn1 (P = 0.774) and pn2 (P = 0.459) status. DISCUSSION There have been many previous studies that have suggested poor outcome of positive PLC NSCLC patients. The incidence of positive pre-plc findings ranged % in previous reports, with most studies reporting an incidence of 5% [2 13]. The pre-plc status was found to be positive in 3.6% of cases in the present study, which is reasonable. Although the study period was long, difference in the incidence of a positive finding between periods was within the acceptable range, 4.4% from 1991 to 2002 and

5 M. Nakao et al. / Interactive CardioVascular and Thoracic Surgery 781 Figure 3: Overall survival curves of positive and negative pre-plc patients stratified by pt classification. Figure 4: Stratified analysis of comparisons between the positive pre-plc group and pt3 negative pre-plc patients in pn0, pn1 and pn2 status. 2.8% from 2003 to Some previous studies also suggested that positive post-plc findings have prognostic importance, with an incidence of % [7, 8]. However, positive post-plc status was extremely rare in the current study, and was detected in only three patients. This may be because of the manner in which we assessed post-plc status. We sampled 50 ml of the ml physiological saline solution used to wash the pleural cavity. In some positive pre-plc cases, we also infused hypotonic cisplatin or distilled water into the pleural cavity as adjuvant therapies. Although this did not contribute to overall survival as we reported previously [2], it may have affected the low frequency of positive post-plc cases. In the present study, we aimed to clarify the prognostic differences between positive pre-plc and PD and/or PE status. Although a positive PLC status is considered to be a precursor condition for PD and/or PE, the differences between these conditions have not been assessed in detail, previously [7, 10]. We enrolled 45 surgically resected patients with PD and/or PE during the same period, excluding individuals who underwent exploratory thoracotomy or only partial resection of the main lesions. Although they did not undergo histological complete resection, they were the most appropriate group to compare the survival probability with the positive PLC group because almost the same operative procedure was performed. Our results suggested that positive pre-plc patients had a much better 5-year overall survival (50.6%) than the PD and/or PE group (21.0%). These findings suggest that a positive PLC status is not equal to PD and/or PE; thus, it is not a contraindication for surgical resection. We performed exploratory analysis to incorporate pre-plc status into the TNM classification. Our results suggested that a positive pre-plc status was a significant prognostic factor in univariate analysis, but was not an independent predictor of poor outcome in multivariate analysis. We speculate that a positive pre-plc status is closely correlated with other clinicopathological factors and should be incorporated into one of the existing TNM factors. Because there were no significant differences in survival between the pt1 2 and pt3 groups in positive pre-plc patients, we considered they could be grouped in the same pt classification. We propose that positive pre-plc status should be classified as pt3 on the basis of survival equality. Although a small number of cases were included in each group, our proposal was verified by additional analyses according to pn status. Lim et al. [9] suggested that it be appropriate to upstage patients with a positive pre-plc status by one T classification. Their recommendation is consistent with the current proposal, whereby pre-plc status should be incorporated into the T factor. Although their study included a large number of patients from multiple institutions and had strong statistical power, there is a matter of data accuracy. Because there are no established methods to sample PLC and handle the specimens or diagnostic criteria, the incidence of positive PLC results varies among institutions. Although the current study scale is smaller, it has the benefit of including high-quality data from a unified procedure performed at a single institution. Kaneda et al. [4] suggested that a precise diagnosis of positive pre-plc status as PL3 (=T3) was appropriate. They performed a survival comparison between positive pre-plc patients and each of the PL0 3 groups. Although the assessment process differed, their results were consistent with those of the current study. Although the results of previous reports have slightly varied in terms of frequency and prognosis, data consistently revealed that positive pre-plc patients have a worse prognosis than negative pre-plc patients. However, the treatment of these patients in ORIGINAL ARTICLE

6 782 M. Nakao et al. / Interactive CardioVascular and Thoracic Surgery clinical practice remains unclear. Our results revealed a prognostic difference between positive pre-plc and PD and/or PE patients. Surgical resection may contribute greatly to improve their prognosis. Because the pre-plc status can be obtained at thoracotomy, neoadjuvant therapies may be difficult to indicate. Curative surgery and systemic adjuvant treatments would be the only ways to improve patient prognosis. It is important to establish a routine PLC sampling method and unified diagnostic criteria to allow the accumulation of precise data regarding positive pre-plc patients and accurately assess the contribution of a positive PLC status to the TNM classification. A large amount of data that are prospectively collected following these methods and criteria will clarify the features and prognosis of patients with a positive PLC status. CONCLUSIONS A positive pre-plc status had a significant prognostic effect in surgically resected NSCLC patients. However, positive pre-plc patients had a much better 5-year overall survival rate compared with those with surgically resected PD and/or PE. Therefore, positive PLC is a precursor but not equal to PD and/or PE, and surgical resection may help improve patient prognosis. We propose that positive pre-plc status should be classified as pt3 in the next TNM classification. Conflict of interest: none declared. REFERENCES [1] Spjut HJ, Hendrix VJ, Ramirez GA, Roper CL. Carcinoma cells in pleural cavity washings. Cancer 1958;11: [2] Satoh Y, Hoshi R, Ishikawa Y, Horai T, Okumura S, Nakagawa K. Recurrence patterns in patients with early stage non-small cell lung cancers undergoing positive pleural lavage cytology. Ann Thorac Surg 2007;83: [3] Higashiyama M, Oda K, Okami J, Maeda J, Kodama K, Takenaka A et al. Prognostic value of intraoperative pleural lavage cytology for lung cancer without carcinomatous pleuritis: importance in patients with early stage disease during long-term follow-up. Eur J Cardiothorac Surg 2009;35: [4] Kaneda M, Yokoi K, Ito S, Niwa H, Takao M, Kondo R et al. The value of pleural lavage cytology examined during surgery for primary lung cancer. Eur J Cardiothorac Surg 2012;41: [5] Kawachi R, Nakazato Y, Masui K, Takei H, Koshi-ishi Y, Goya T. Clinical significance of pleural lavage cytology for non-small cell lung cancer: is surgical resection valid for patients with positive pleural lavage cytology? Interact CardioVasc Thorac Surg 2009;9: [6] Taniguchi Y, Nakamura H, Miwa K, Adachi Y, Fujioka S, Haruki T et al. Prognostic significance of pleural lavage cytology after thoracotomy and before closure of the chest in lung cancer. Interact CardioVasc Thorac Surg 2009;9: [7] Aokage K, Yoshida J, Ishii G, Enatsu S, Hishida T, Nishimura M et al. The impact on survival of positive intraoperative pleural lavage cytology in patients with non-small-cell lung cancer. J Thorac Cardiovasc Surg 2010; 139: e1241. [8] Kotoulas C, Lazopoulos G, Karaiskos T, Tomos P, Konstantinou M, Papamichalis G et al. Prognostic significance of pleural lavage cytology after resection for non-small cell lung cancer. Eur J Cardiothorac Surg 2001;20: [9] Lim E, Clough R, Goldstraw P, Edmonds L, Aokage K, Yoshida J et al. Impact of positive pleural lavage cytology on survival in patients having lung resection for non-small-cell lung cancer: an international individual patient data meta-analysis. J Thorac Cardiovasc Surgery 2010; 139: [10] Okada M, Sakamoto T, Nishio W, Uchino K, Tsuboshima K, Tsubota N. Pleural lavage cytology in non-small cell lung cancer: lessons from 1000 consecutive resections. J Thorac Cardiovasc Surg 2003;126: [11] Shintani Y, Ohta M, Iwasaki T, Ikeda N, Kanou T, Tomita E et al. Intraoperative pleural lavage cytology after lung resection as an independent prognostic factor for staging lung cancer. J Thorac Cardiovasc Surg 2009;137: [12] Li YN, Shi HZ, Liang QL, Yang HB, Huang GM. Prognostic significance of pleural lavage cytology in patients with lung cancer: a meta-analysis. Lung Cancer (Amsterdam, Netherlands) 2008;60: [13] Saso S, Rao C, Ashrafian H, Ghaem-Maghami S, Darzi A, Athanasiou T. Positive pre-resection pleural lavage cytology is associated with increased risk of lung cancer recurrence in patients undergoing surgical resection: a meta-analysis of 4450 patients. Thorax 2012;67: [14] Goldstraw P. Staging Manual in Thoracic Oncology. Denver, CO: IASLC, 2009, [15] International Union against Cancer. In: Sobin LH, Wittekind CH (eds). TNM Classification of Malignant Tumours, 7th edn. New York, NY: Wiley-Liss, 2009, [16] Postmus PE, Brambilla E, Chansky K, Crowley J, Goldstraw P, Patz EF Jr et al. The IASLC Lung Cancer Staging Project: proposals for revision of the M descriptors in the forthcoming (seventh) edition of the TNM classification of lung cancer. J Thorac Oncol 2007;2: [17] Travis WD, Brambilla E, Muller-Hermelink HK. Pathology and Genetics of Tumours of the Lung, Pleura, Thymus and Heart. Lyon, France: IASLC Press, 2004,