- Email: [email protected]
Clinical applicability of staging small cell lung cancer according to the seventh edition of the TNM staging system
Lung Cancer 81 (2013) 65–70
Contents lists available at SciVerse ScienceDirect
Lung Cancer journal homepage: www.elsevier.com/locate/lungcan
Clinical applicability of staging small cell lung cancer according to the seventh edition of the TNM staging system Byung Woo Jhun a , Kyung-Jong Lee a , Kyeongman Jeon a , Gee Young Suh a , Man Pyo Chung a , Hojoong Kim a , O Jung Kwon a , Jong-Mu Sun b , Jin Seok Ahn b , Myung-Ju Ahn b , Keunchil Park b , Joon Young Choi c , Kyung Soo Lee d , Joungho Han e , Sang-Won Um a,∗ a
Division of Pulmonary and Critical Care Medicine, Department of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea Division of Hematology-Oncology, Department of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea c Department of Nuclear Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea d Department of Radiology and Center for Imaging Science, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea e Department of Pathology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea b
a r t i c l e
i n f o
Article history: Received 2 January 2013 Received in revised form 18 February 2013 Accepted 5 March 2013 Keywords: Small cell lung cancer UICC 7th TNM stage Two-stage Survival Prognosis Applicability
a b s t r a c t The two-stage system of limited and extensive disease has been widely employed for small cell lung cancer (SCLC). However, the International Association for the Study of Lung Cancer has proposed that the TNM classification should be incorporated into clinical practice. The purpose of this study was to evaluate the applicability of the Union for International Cancer Control (UICC) 7th TNM staging system to SCLC. We retrospectively reviewed the medical records of consecutive patients with newly diagnosed histologically proven SCLC between March 2005 and January 2010. Patients who had other concurrent malignancies or had combined-type SCLC were excluded. We assessed overall survival (OS) according to the T descriptor, N descriptor, M descriptor, and TNM stage grouping. In total, 320 SCLC patients were included. Median age was 65 years and 286 patients (89.4%) were male. Median OS was 12.7 months. There were no significant differences in OS according to the T descriptor (P = 0.880). However, there were significant differences in OS according to the N (P < 0.001) and M (P < 0.001) descriptors and TNM stage grouping (P < 0.001). Hazard ratios for OS, adjusted for known prognostic factors, differed significantly according to the N and M descriptor, and TNM stage grouping, but not according to the T descriptors. The UICC 7th TNM staging system may contribute to a more precise prognosis in SCLC patients. Further studies are required to evaluate the applicability of the TNM staging system to SCLC. © 2013 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Small cell lung cancer (SCLC) accounts for 15–20% of all lung cancer cases [1,2]. It generally has a rapid doubling time, a high growth fraction, and early development of hematogenous metastasis [3]. Only about one-third of patients have limited disease confined to the chest at initial presentation. SCLC is highly responsive to initial treatment with cytotoxic agents or radiation; however, because of the disease’s high recurrence rate and aggressive behavior, the survival rate over the past decades has not increased in line with expectations [4,5]. With currently available treatments, median
∗ Corresponding author at: Division of Pulmonary and Critical Care Medicine, Department of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, 50 Irwon-dong, Gangnam-gu, Seoul, 135-710, Republic of Korea. Tel.: +82 2 3410 3429; fax: +82 2 3410 6956. E-mail addresses: [email protected], [email protected] (S.-W. Um). 0169-5002/$ – see front matter © 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.lungcan.2013.03.005
survival ranges from 16 to 26 months even in patients with limited disease [6,7]. Without treatment, median overall survival (OS) ranges from 2 to 4 months [8]. The precise determination of the prognosis is therefore important in choosing an adequate treatment modality and determining treatment strategies for individual patients. Clinical stage is known to be closely correlated with prognosis [9,10]. The two-stage system introduced by the Veterans Affairs Lung Study Group has been routinely used to classify SCLC cases as limited or extensive disease. Nevertheless, this two-stage system is problematic when applied to patients with extremely limited disease [11] or with locoregionally advanced disease, such as those with contralateral hilar and supraclavicular nodes [12–15] or malignant effusions [16]. Some studies have suggested that a separate analysis of prognosis, rather than a two-stage categorization, may be beneficial. For example, Kreisman et al. reported that SCLC presenting as a lung mass without nodal involvement had superior survival rates compared with the general category of limited SCLC [11]. Shepherd et al. showed that a subgroup of
66
B.W. Jhun et al. / Lung Cancer 81 (2013) 65–70
patients within limited-stage disease with no mediastinal nodal involvement had better median survival than patients with mediastinal lymphadenopathy [17,18]. In addition, Zhu et al. recently demonstrated significant survival differences within both the limited-stage and extensive-stage groups according to metabolic parameters measured by PET/CT [19]. Although the patient groups were categorized by metabolic parameters, rather than using the TNM staging system, the data suggest that a separate analysis of prognosis in both limited-stage and extensive-stage disease may be possible. Moreover, in recent studies by the International Association for the Study of Lung Cancer (IASLC), evaluation of the applicability of the seventh edition of the TNM system to SCLC revealed the prognostic significance of TNM stages [20,21]. However, there are few data regarding the applicability of the Union for International Cancer Control (UICC) 7th TNM staging system to SCLC in clinical practice. The purposes of the present study were to evaluate the applicability of the UICC 7th TNM staging system to SCLC and to evaluate whether the TNM staging system can provide additional prognostic information compared with the two-stage system.
2. Materials and methods 2.1. Study patients and data collection We reviewed the medical records of all adult patients who were newly diagnosed with histologically proven SCLC between March 2005 and January 2010 at Samsung Medical Center, a referral hospital in Seoul, South Korea. Patients with other concurrent malignancies or combined-type SCLC (SCLC with squamous cell carcinoma and/or adenocarcinoma) were excluded from the study. Data were collected retrospectively for clinical manifestations, laboratory and radiological tests, cyto-histopathology, and survival, including physical examination, laboratory investigations, chest radiography, contrast-enhanced CT scans of the chest and upper abdomen, integrated PET/CT scans, magnetic resonance imaging of the brain, and cyto-histopathological results of transthoracic needle biopsy or bronchoscopic biopsy. Patients who received only best supportive care were excluded in the survival analyses. All study patients who had been classified by the two-stage system were restaged according to the UICC 7th TNM staging. The Institutional Review Board at the Samsung Medical Center approved the study protocol. Informed consent was waived because of the retrospective nature of the study. 2.2. Two-stage and TNM staging systems 2.2.1. Two-stage In the current study, limited-stage disease was defined as disease confined to the ipsilateral hemithorax that could be encompassed within a tolerable radiation port. Thus, ipsilateral/ contralateral mediastinal lymph nodes, ipsilateral/contralateral supraclavicular lymph nodes, recurrent laryngeal nerve involvement, and superior vena cava obstruction were all included. Extensive-stage disease was defined as disease beyond the ipsilateral hemithorax, including malignant pleural or pericardial effusion and hematogenous metastasis. Patients with tumors that could not be included within a tolerable radiation field were also considered to have extensive stage disease [22]. 2.2.2. TNM staging system All study patients were restaged according to seventh edition of the TNM staging system [23]. Restaging was performed with consideration of all available clinical information,
radiological findings on contrast-enhanced CT and integrated PET/CT, and cyto-histopathological data. 2.3. Survival analysis in relation to the TNM staging system OS was defined as the time interval from histopathological diagnosis to death from any cause or the end of follow-up. The last follow-up date was January 31, 2012. We compared the OS of the study patients in relation to the T descriptor (Nany , Many ), N descriptor (Tany , Many ), M descriptor (Tany , Nany ), and TNM stage grouping. 2.4. Integrated 18F-FDG PET/CT acquisition and data analysis After the patients fasted for at least 6 h before their PET/CT examinations, they received an intravenous injection of 370 MBq of 18F-FDG and then rested for 45 (43–47) min before undergoing imaging. Image acquisition was performed using an integrated PET/CT device (Discovery LS, GE Healthcare, Milwaukee, WI) that consisted of an Advance NXi PET scanner and an 8-slice Light Speed Plus CT scanner. The 18F-FDG PET data were evaluated using the maximum standardized uptake value (SUVmax ). We used the region of interest analysis tools included with the scanner to calculate the SUVmax over the primary tumor after correction for the injected dose of 18F-FDG and the patient’s weight. 2.5. Statistical analysis The data are presented as number (percentage) or median (interquartile range [IQR]). To compare the OS in relation to each TNM descriptor, Kaplan–Meier analysis was performed and survival curves were compared by log-rank testing. Cox proportional hazard models for OS were also created. To compare the prognostic usefulness of the TNM staging system and the two-stage system, Kaplan–Meier analysis was performed in relation to the two-stages (limited and extensive) and the TNM stage groupings 1 (I + II, III, IV) and 2 (stages I, II, III, IV); the survival curves were compared by log-rank testing and P-values were corrected using Bonferroni’s method to counteract the problem of multiple comparisons. The patient population was also subdivided by median values for clinico-laboratory characteristics that have been demonstrated to be prognostic factors: age, Eastern Cooperative Oncology Group (ECOG) performance status (PS), lactate dehydrogenase (LDH) level, albumin level, and maximum standardized uptake (SUVmax ). Univariate survival rate analysis was performed using the Kaplan–Meier method, with comparisons of survival curves by log-rank testing. To assess the possible independent effects of the T descriptor (model 1), N descriptor (model 2), M descriptor (model 3), TNM stage grouping (models 4 and 5), and two-stage (model 6) on OS, multivariate analysis was performed with a Cox proportional hazard model using variables with P-values <0.10 in the univariate analyses; P-values were corrected using Bonferroni’s method to counteract the problem of multiple comparisons. Values of P < 0.05 were considered to indicate statistical significance. All statistical analyses were performed using PASW 20.0 (SPSS Inc., Chicago, IL). 3. Results 3.1. Patients In total, 362 patients with histologically proven SCLC were identified during the study period. Of these, 42 patients were excluded: 21 had combined-type SCLC and 21 had other concurrent malignancies. Consequentially, 320 patients with SCLC were included in the analysis and 23 patients who received only best supportive care were excluded in the survival analyses. Characteristics of the study patients are shown in Table 1. The median age was 65 years (59–71
B.W. Jhun et al. / Lung Cancer 81 (2013) 65–70
67
Table 1 Characteristics of the study patients. Characteristic
Number (%) or median (IQR)
Number of patients Age, years Male/female Smoking Current smoker Ex-smoker Never smoker Overall survival, months ECOG performance status 0, 1 ≥2 Weight loss Lactate dehydrogenase level, IU/L Albumin, g/dL PET/CT scan SUVmax Pleural effusion Pericardial effusion Brain metastasis at diagnosis Treatment Definitive CCRT Definitive chemotherapy 2 (0.6) Palliative chemotherapy Palliative radiotherapy Palliative chemotherapy and radiotherapy 100 (31.3) Best supportive care CR during study period at any time
320 65 (59–71) 286 (89.4)/34 (10.6) 229 (71.5) 69 (21.6) 22 (6.9) 12.7 (7.5–23.2) 251 (78.4) 69 (21.6) 93 (29.1) 431 (357.8–549.3) 3.8 (3.5–4.1) 254 (79.4) 10.5 (8.3–12.5) 75 (23.4) 9 (2.8) 66 (20.6) 91 (28.4) 91 (28.4) 13 (4.1)
23 (7.2) 34 (10.6)
IQR, interquartile range; ECOG, eastern cooperative oncology group; SUVmax , maximum standardized uptake, there is missing data in 66 cases for SUVmax ; PET, positron emission tomography; CT, computed tomography; CCRT, concurrent chemoradiotherapy; CR, complete remission.
years), and 286 patients (89.4%) were men. Median OS was 12.7 months (7.5–23.2 months). No patients were lost to follow-up. Two hundred ninety-seven patients (92.8%) received definitive or palliative chemotherapy with or without radiotherapy, and 23 patients (7.2%) received best supportive care alone. No patient underwent surgery for SCLC. Most patients had a relatively good performance status. The treatment strategies in our institution were consistent during the study period. Most commonly used chemotherapy regimen was etoposide/cisplatin in 160 patients (50.0%) followed by irinotecan/cisplatin in 41 patients (12.9%), etoposide/carboplatin in 37 patients (11.6%), irinotecan/carboplatin in 27 patients (8.4%), and so on. In ninety-one patients who received definitive RT, radiotherapy was delivered as 2 Gy once daily to the median of 52.5 Gy (IQR 45.0–52.5). In 13 patients who received palliative RT for lung, spine, and brain, a median of 27.0 Gy (IQR: 20.0–30.0) was irradiated. 3.2. Survival in relation to the TNM staging system OS comparisons among the TNM descriptors are shown in Table 2. The T category did not show a definite correlation with OS, and differences between survival curves according to T category were not significant (P = 0.880; log-rank test). Hazard ratios (HRs) for OS between T stages were not significant. A higher N category was associated with poorer OS, with significant differences between survival curves according to N category (P < 0.001; log-rank test); HR for OS between N1 and N2 or N2 and N3 differed significantly, but not between N0 and N1. As the M category increased, OS became poorer, and the survival curves according to M category differed significantly (P < 0.001; log-rank test). HR for OS between the M0 and M1a stages differed significantly, but not between M1a and M1b. Table 3 presents OS comparisons for the two-stage system and TNM stage groupings. For the two-stage system, extensive-stage
Fig. 1. A, Kaplan–Meier overall survival (OS) curve for all SCLC patients according to two-stage grouping. B, Kaplan–Meier OS curve according to TNM grouping (I + II, III, IV). C, Kaplan–Meier OS curve according to TNM grouping (I, II, III, IV), P, P-value from log-rank test.
disease predictably showed significantly poorer OS compared with limited-stage disease (P < 0.001, Fig. 1). For the TNM stages, OS tended to decrease as TNM stage grouping increased. Owing to the relatively small numbers of patients with early-stage disease (IA, IB, IIA, and IIB), the stages were considered as three (I + II, III, and IV) or four (I, II, III, and IV) groups. Increased stage within each stage grouping was associated with poorer OS, and the differences between the survival curves were significant (P < 0.001 for each; log-rank test, Fig. 1). HRs for OS between two (limited and extensive) and three (I + II, III, and IV) stage groupings differed significantly (Table 3). There were also significant differences between III and IV in four (I, II, III, and IV) stage grouping (Table 3). Compared with the two-stage system, TNM stage groupings can provide additional prognostic subgroups in patients with SCLC. 3.3. Univariate survival analysis in relation to clinical characteristics Univariate analysis of OS was performed according to the median values for the following clinico-laboratory characteristics, which have been identified as prognostic factors: age, ECOG PS, LDH level, albumin level, and SUVmax (Table 4). There were significant
68
B.W. Jhun et al. / Lung Cancer 81 (2013) 65–70
Table 2 Overall survival in small cell lung cancer patients according to TNM descriptors of the 7th edition of the TNM staging system. Category T descriptor (Nany , Many )
N descriptor (Tany , Many )
M descriptor (Tany , Nany )
T0 T1 T2 T3 T4 N0 N1 N2 N3 M0 M1a M1b
No .of patient (%)
Median (IQR) OS months
43 (14.5) 60 (20.2) 108 (36.4) 55 (18.5) 31 (10.4) 11 (3.7) 30 (10.1) 115 (38.7) 141 (47.5) 122 (41.1) 40 (13.5) 135 (45.4)
14.7 (9.2–18.2) 19.5 (9.1–27.6) 13.0 (7.1–22.8) 12.2 (8.4–25.5) 11.9 (8.9–18.8) 39.0 (13.7–58.4) 23.7 (18.4–30.0) 14.8 (9.6–25.0) 10.3 (7.1–16.2) 21.8 (11.5–31.0) 11.5 (6.9–19.7) 11.1 (7.4–15.6)
Comparison
HR (95% CI)
P*
T1 vs. T0 T2 vs. T1 T3 vs. T2 T4 vs. T3
0.6 (0.3–1.0) 1.5 (0.9–2.3) 1.1 (0.7–1.6) 1.1 (0.7–1.7)
N1 vs. N0 N2 vs. N1 N3 vs. N2
3.2 (0.8–14.3) 1.8 (1.0–3.1) 1.7 (1.2–2.3)
0.135 0.042 <0.001
M1a vs. M0 M1b vs. M1a
2.5 (1.6–3.8) 1.2 (0.8–1.8)
<0.001 0.999
0.084 0.140 0.999 0.999
IQR, interquartile range; OS, overall survival; HR, hazard ratio; CI, confidence interval; *P-value for HR, relative to preceding row.
Table 3 Overall survival in small cell lung cancer patients according to two-stage and 7th edition of the TNM stage grouping. Category Two-stage
Limited Extensive I + II III IV I II III IV
TNM stage grouping
TNM stage grouping
No .of patient (%)
Median (IQR) OS months
118 (39.7) 179 (60.3) 23 (7.7) 98 (33.0) 176 (59.3) 6 (2.0) 17 (5.7) 98 (33.0) 176 (59.3)
22.4 (11.5–31.7) 11.5 (7.4–15.9) 26.9 (23.2–39.7) 18.9 (10.1–29.2) 11.2 (7.4–15.7) 43.7 (36.1–62.2) 24.3 (20.7–31.1) 18.9 (10.1–29.2) 11.2 (7.4–15.7)
Comparison
HR (95% CI)
P*
extensive vs. limited
2.9 (2.1–3.9)
<0.001
III vs. I + II IV vs. III
2.7 (1.3–5.5) 2.4 (1.8–3.4)
<0.006 <0.001
II vs. I III vs. II IV vs. III
3.1 (0.6–21.8) 1.7 (0.8–3.8) 2.5 (1.7–3.5)
0.999 0.324 <0.001
IQR, interquartile range; OS, overall survival; HR, hazard ratio; CI, confidence interval; *P-value for HR, relative to preceding row.
differences in OS in relation to age (P = 0.006), ECOG PS (P < 0.001), LDH level (P < 0.001), and albumin level (P = 0.003). However, there was no difference in OS in relation to SUVmax (P = 0.829). 3.4. Multivariate survival analysis in prognostic models
4. Discussion
To assess the possible independent effects of the T descriptor (model 1), N descriptor (model 2), M descriptor (model 3), TNM grouping (model 4 and 5), and two-stage (model 6) on OS, we performed multivariate analyses adjusted for age, ECOG PS, LDH level, and albumin level, all of which had P-values <0.10 in the univariate analyses (Table 5). The N descriptor (model 2), M descriptor (model 3), three TNM stage (I + II, III, and IV) grouping (model 4), and twostage (model 6) remained significant in multivariate analyses, and the estimated hazard ratio increased with advancing N, M, and TNM
Table 4 Univariate analysis of overall survival in relation to clinical characteristics. Characteristic Age, years <65 ≥65 ECOG PS 0, 1 ≥2 LDH, IU/L <430 ≥430 Albumin, g/dL ≥3.8 <3.8 SUVmax <10.5 ≥10.5
No. of patients
Median OS months
146 151
15.8 11.5
244 53
14.8 8.7
151 146
18.8 10.0
168 129
15.3 11.2
122 124
14.2 14.6
stage groupings (Table 5). The T descriptor (model 1) and four TNM stage (I, II, III, and IV) grouping (model 5) were not significant in the multivariate analysis (Table 5).
P* 0.006
<0.001
<0.001
0.003 0.829
OS, overall survival; ECOG, Eastern Cooperative Oncology Group; PS, performance status; LDH, lactate dehydrogenase; SUVmax , maximum standardized uptake, there is missing data in 51 cases for SUVmax ; *P-value from log-rank test, relative to preceding row.
In the current study, we showed that the seventh edition of the TNM staging system can provide additional prognostic information compared with the two-stage system, even after adjustment for other known prognostic factors. HRs for OS in relation to TNM stage grouping (I + II, III, and IV) were significantly higher for more advanced TNM stage groups, when adjusted for age, ECOG PS, LDH level, and albumin level. However, the HRs did not differ significantly in four TNM stage (I, II, III, and IV) grouping, probably because of the relatively small numbers of patients with early-stage disease. HRs for OS between M categories, N1 and N2, or N2 and N3 remained significant, when adjusted for age, ECOG PS, LDH level, and albumin level (Table 5). The lack of a significant difference in the hazard ratio between N0 and N1 stages may be due to the relatively small numbers of patients with N0 or N1 disease, given that survival curves according to N category differed significantly (P < 0.001; log-rank test). However, HRs for OS between T categories were not significant. These results are supported by a few previous studies by IASLC on the applicability of the TNM staging system to SCLC. Shepherd et al. reported that both the sixth and seventh editions of the TNM staging system could more precisely predict survival rates in SCLC, and these authors indicated that stratification by stage (I–III) should be incorporated in clinical trials [20]. Ignatius et al. reported that compared with the sixth edition, the seventh edition of the TNM staging system resulted in better discrimination of survivals in SCLC patients [21]. One of the remarkable aspects of the current study is that after we adjusted for other known prognostic factors to evaluate the applicability of TNM staging system more practically, the correlation between TNM stage and SCLC prognosis remained (Table 5). Therefore, our data may have clinical significance and
B.W. Jhun et al. / Lung Cancer 81 (2013) 65–70
69
Table 5 Multivariate analysis for overall survival in prognostic models. Characteristic
Model 1 HR (95% CI)
Age, years 1 <65 ≥65 1.2 (1.0–1.5) ECOG PS 1 0, 1 ≥2 2.1 (1.5–2.9) LDH, IU/L 1 <430 ≥430 1.8 (1.4–2.3) Albumin, g/dL ≥3.8 1 <3.8 1.1 (0.9–1.5) T descriptor 1 T0 0.6 (0.3–1.1) T1 1.4 (0.9–2.3) T2 1.0 (0.7–1.3) T3 1.2 (0.6–2.1) T4 N descriptor NA N0 N1 N2 N3 M descriptor NA M0 M1a M1b TNM stage grouping I + II NA III IV TNM stage grouping NA I II III IV Two-stage NA Limited Extensive
Model 2
Model 3
Model 4
Model 5
Model 6
P*
HR (95% CI)
P*
HR (95% CI)
P*
HR (95% CI)
P*
HR (95% CI)
P*
HR (95% CI)
P*
0.176
1 1.4 (1.1–1.8)
<0.001
1 1.2 (0.9–1.6)
<0.001
1 1.2 (1.0–1.6)
0.098
1 1.3 (1.0–1.7)
0.089
1 1.2 (1.0–1.5)
0.195
<0.001
1 1.9 (1.3–2.6)
<0.001
1 1.9 (1.3–2.7)
<0.001
1 1.9 (1.3–2.6)
<0.001
1 1.8 (1.3–2.5)
<0.001
1 1.9 (1.3–2.6)
<0.001
<0.001
1 1.6 (1.3–2.1)
<0.001
1 1.7 (1.2–2.3)
<0.001
1 1.6 (1.2–2.1)
<0.001
1 1.6 (1.2–2.1)
<0.001
1 1.6 (1.2–2.1)
<0.001
0.395
1 1.1 (0.9–1.5)
0.324
1 1.1 (0.8–1.5)
0.638
1 1.1 (0.9–1.4)
0.458
1 1.1 (0.9–1.4)
0.470
1 1.1 (0.9–1.5)
<0.298
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
0.100 0.196 0.999 0.999 1 3.9 (0.9–16.3) 1.9 (1.0–3.3) 1.4 (1.0–2.0) NA
NA
NA
NA
0.084 0.030 0.039 1 1.9 (1.2–3.1) 1.4 (0.9–2.2) NA
NA
NA
0.002 0.102 1 2.8 (1.3–6.0) 2.2 (1.5–2.9) NA
NA
0.004 <0.001 1 3.6 (0.6–22.0) 1.8 (0.8–3.8) 2.2 (1.6–3.0) NA
NA 0.999 0.180 <0.001 1 2.6 (2.0–3.4)
<0.001
All models are adjusted for age, ECOG PS, LDH, and albumin; Model 1, UICC 7th T descriptor. Model 2, UICC 7th N descriptor. Model 3, UICC 7th M descriptor. Model 4, UICC 7th three TNM stage grouping (I + II, III, and IV). Model 5, UICC 7th four TNM stage grouping (I, II, III, and IV). Model 6, Two-stage; HR, hazard ratio; NA, not applicable; *P-value for HR, relative to preceding row.
may provide more precise information for evaluating prognosis in SCLC using the TNM staging system. Our data for the T descriptor did not show a significant correlation between T category and OS, unlike the findings of some previous studies [11,20,21,24]. However, there have also been studies with results similar to ours. For example, Lim et al. evaluated survival in 59 limited-stage SCLC patients who underwent surgical resection or adjuvant chemotherapy, and found no clear survival difference according to T category [25]. Angeletti et al. evaluated the influence of T and N stages on long-term survival in 49 SCLC patients who underwent surgical resection or adjuvant chemotherapy. They showed that survival was significantly influenced by N, but not T category [26]. The contradictory findings for the T category may be partly explained by the characteristics of the tumors, which usually present with bulky lymphadenopathy and without a lung tumor or conglomerated lesion [27,28], making the accurate assessment of T category difficult. Further studies on the applicability of the TNM staging system in T0 SCLC without distant metastasis are needed. The roles of PET/CT scan in the staging of SCLC are still controversial. Some authors insisted that PET/CT scan might enable more accurate clinical staging [29–31]. However, PET/CT scan could have a false positive result to define mediastinal and distant metastases. SUVmax can also be variably interpreted by several factors including patient glucose level and variability in scanners. However, in
our institution, the protocol for PET/CT scan was consistent during study period. In the current study, 34 (10.6%) patients were restaged from limited to extensive due to the detection of distant metastasis based on PET/CT scan and the managements were changed from concurrent chemoradiotherapy to chemotherapy. The most common sites of metastasis was bone (n = 14), followed by chest wall or contra-lateral lung (n = 7), distant lymph nodes including intra-abdominal or axillary lymph nodes (n = 5), liver (n = 4), adrenal gland (n = 2), breast (n = 1), and brain (n = 1). Of these, 4 patients were pathologically confirmed (axillary lymph node = 2, bone = 1, breast = 1). Thirty metastatic lesions were not pathologically confirmed due to high risk of complications at the physician’s discretion. Interestingly, differences in OS according to SUVmax were not statistically significant in the present study, unlike in previous studies showing the relevance of metabolic parameters measured by PET/CT to prognosis of SCLC [19]. However, one report suggested the unsuitability of SUVmax as a prognostic factor. Van der Leest et al. evaluated the clinical role of SUVmax in 75 SCLC patients [32]. OS and disease-free survival were significantly different according to SUVmax only in patients with stage IV disease and who underwent chemotherapy; when all patients were considered, OS did not differ significantly between those with high and low SUVmax values (P = 0.81). Furthermore, there was no difference in OS according to high and low SUVmax among all treated
70
B.W. Jhun et al. / Lung Cancer 81 (2013) 65–70
patients (P = 0.342) or among all non-treated patients (P = 0.291) [32]. The current study has some limitations. First, relatively few early-stage, especially stage I and II, patients were included, as most SCLC patients initially present with advanced disease because of the tumor’s aggressiveness. The more severe cases were admitted to our tertiary referral center. The correlation between TNM stage and OS in SCLC in our study might have been stronger if more early-stage disease cases had been included. Second, not all suspected malignant fluid was confirmed by cytology or biopsy in this study. Among 75 patients who had suspected malignant pleural or pericardial effusion (Table 1), 26 with distant metastases did not undergo cytopathological examination. Among the remaining 49 patients, only 10 were found to have malignant cells by cytopathological examination; the other 39 patients were accepted as having malignant pleural effusion on the basis of bloody exudates that could not be explained by other reasons or clinical judgment that most pleural effusions with lung cancer are due to tumor. Finally, there was no direct comparison of two-stage system and TNM staging system in this study. Therefore, it cannot be ascertained that the TNM staging system is superior to the two-stage system.
[9]
[10]
[11] [12]
[13] [14]
[15]
[16]
[17]
5. Conclusion [18]
The UICC 7th TNM staging system can contribute to a more precise prognosis in patients with SCLC. Further prospective validation studies are required to confirm the applicability of this TNM staging system to SCLC in clinical practice.
[19]
[20]
Conflict of interest statement All authors have no financial disclosures or personal relationships with other people or organisations that could inappropriately influence their work.
[21]
[22]
Acknowledgments [23]
This study was supported by the Samsung Biomedical Research Institute (CA92092). All authors have no financial disclosures or personal relationships with other people or organisations that could inappropriately influence their work. References [1] Govindan R, Page N, Morgensztern D, Read W, Tierney R, Vlahiotis A, et al. Changing epidemiology of small-cell lung cancer in the United States over the last 30 years: analysis of the surveillance, epidemiologic, and end results database. J Clin Oncol 2006;24:4539–44. [2] Janssen-Heijnen ML, Coebergh JW. The changing epidemiology of lung cancer in Europe. Lung Cancer 2003;41:245–58. [3] Elias AD. Small cell lung cancer: state-of-the-art therapy in 1996. Chest 1997;112:251S–8S. [4] Murray N, Turrisi 3rd AT. A review of first-line treatment for small-cell lung cancer. J Thorac Oncol 2006;1:270–8. [5] Hann CL, Rudin CM. Management of small-cell lung cancer: incremental changes but hope for the future. Oncology (Williston Park) 2008;22:1486–92. [6] Johnson BE, Bridges JD, Sobczeck M, Gray J, Linnoila RI, Gazdar AF, et al. Patients with limited-stage small-cell lung cancer treated with concurrent twice-daily chest radiotherapy and etoposide/cisplatin followed by cyclophosphamide, doxorubicin, and vincristine. J Clin Oncol 1996;14:806–13. [7] Turrisi 3rd AT, Kim K, Blum R, Sause WT, Livingston RB, Komaki R, et al. Twicedaily compared with once-daily thoracic radiotherapy in limited small-cell lung cancer treated concurrently with cisplatin and etoposide. N Engl J Med 1999;340:265–71. [8] Foster NR, Mandrekar SJ, Schild SE, Nelson GD, Rowland Jr KM, Deming RL, et al. Prognostic factors differ by tumor stage for small cell lung cancer: a pooled
[24]
[25]
[26]
[27]
[28] [29]
[30]
[31]
[32]
analysis of North Central Cancer Treatment Group trials. Cancer 2009;115: 2721–31. Albain KS, Crowley JJ, LeBlanc M, Livingston RB. Determinants of improved outcome in small-cell lung cancer: an analysis of the 2,580-patient Southwest Oncology Group data base. J Clin Oncol 1990;8:1563–74. Osterlind K, Hansen HH, Hansen M, Dombernowsky P. Mortality and morbidity in long-term surviving patients treated with chemotherapy with or without irradiation for small-cell lung cancer. J Clin Oncol 1986;4:1044–52. Kreisman H, Wolkove N, Quoix E. Small cell lung cancer presenting as a solitary pulmonary nodule. Chest 1992;101:225–31. Urban T, Chastang C, Vaylet F, Mathieu M, Leclerc P, Paillotin D, et al. Prognostic significance of supraclavicular lymph nodes in small cell lung cancer: a study from four consecutive clinical trials, including 1,370 patients. “Petites Cellules” Group. Chest 1998;114:1538–41. Argiris A, Murren JR. Staging and clinical prognostic factors for small-cell lung cancer. Cancer J 2001;7:437–47. Hoffman PC, Golomb HM, Bitran JD, DeMeester TR, Cohen L, Griem ML, et al. Small-cell carcinoma of the lung: a five-year experience with combined modality therapy. Cancer 1980;46:2550–6. Sagman U, Maki E, Evans WK, Warr D, Shepherd FA, Sculier JP, et al. Small-cell carcinoma of the lung: derivation of a prognostic staging system. J Clin Oncol 1991;9:1639–49. Micke P, Faldum A, Metz T, Beeh KM, Bittinger F, Hengstler JG, et al. Staging small cell lung cancer: Veterans Administration Lung Study *Group versus International Association for the Study of Lung Cancer—what limits limited disease? Lung Cancer 2002;37:271–6. Shepherd FA, Ginsberg RJ, Haddad R, Feld R, Sagman U, Evans WK, et al. Importance of clinical staging in limited small-cell lung cancer: a valuable system to separate prognostic subgroups. The University of Toronto Lung Oncology Group. J Clin Oncol 1993;11:1592–7. Shepherd FA, Ginsberg RJ, Feld R, Evans WK, Johansen E. Surgical treatment for limited small-cell lung cancer. The University of Toronto Lung Oncology Group experience. J Thorac Cardiovasc Surg 1991;101:385–93. Zhu D, Ma T, Niu Z, Zheng J, Han A, Zhao S, et al. Prognostic significance of metabolic parameters measured by (18)F-fluorodeoxyglucose positron emission tomography/computed tomography in patients with small cell lung cancer. Lung Cancer 2011;73:332–7. Shepherd FA, Crowley J, Van Houtte P, Postmus PE, Carney D, Chansky K, et al. The International Association for the Study of Lung Cancer lung cancer staging project: proposals regarding the clinical staging of small cell lung cancer in the forthcoming (seventh) edition of the tumor, node, metastasis classification for lung cancer. J Thorac Oncol 2007;2:1067–77. Ignatius Ou SH, Zell JA. The applicability of the proposed IASLC staging revisions to small cell lung cancer (SCLC) with comparison to the current UICC 6th TNM Edition. J Thorac Oncol 2009;4:300–10. Stahel RGR, Havemann K. Staging and prognostic factors in small cell lung cancer: a consensus report. Lung Cancer 1989;5:119–26. Groome PA, Bolejack V, Crowley JJ, Kennedy C, Krasnik M, Sobin LH, et al. The IASLC Lung Cancer Staging Project: validation of the proposals for revision of the T, N, and M descriptors and consequent stage groupings in the forthcoming (seventh) edition of the TNM classification of malignant tumours. J Thorac Oncol 2007;2:694–705. Uno T, Sumi M, Sawa Y, Teshima T, Hara R, Ikeda H, et al. Process of care and preliminary outcome in limited-stage small-cell lung cancer: results of the 1995–1997 patterns of care study in Japan. Int J Radiat Oncol Biol Phys 2003;55:626–32. Lim E, Belcher E, Yap YK, Nicholson AG, Goldstraw P. The role of surgery in the treatment of limited disease small cell lung cancer: time to reevaluate. J Thorac Oncol 2008;3:1267–71. Angeletti CA, Macchiarini P, Mussi A, Basolo F. Influence of T and N stages on long-term survival in resectable small cell lung cancer. Eur J Surg Oncol 1989;15:337–40. Forster BB, Muller NL, Miller RR, Nelems B, Evans KG. Neuroendocrine carcinomas of the lung: clinical, radiologic, and pathologic correlation. Radiology 1989;170:441–5. Pearlberg JL, Sandler MA, Lewis Jr JW, Beute GH, Alpern MB. Small-cell bronchogenic carcinoma: CT evaluation. AJR Am J Roentgenol 1988;150:265–8. Bradley JD, Dehdashti F, Mintun MA, Govindan R, Trinkaus K, Siegel BA. Positron emission tomography in limited-stage small-cell lung cancer: a prospective study. J Clin Oncol 2004;22:3248–54. Fischer BM, Mortensen J, Langer SW, Loft A, Berthelsen AK, Petersen BI, et al. A prospective study of PET/CT in initial staging of small-cell lung cancer: comparison with CT, bone scintigraphy and bone marrow analysis. Ann Oncol 2007;18:338–45. Azad A, Chionh F, Scott AM, Lee ST, Berlangieri SU, White S, et al. High impact of 18F-FDG-PET on management and prognostic stratification of newly diagnosed small cell lung cancer. Mol Imaging Biol 2010;12:443–51. van der Leest C, Smit EF, Baas J, Versteijlen RJ, van Walree N, Hoogsteden HC, et al. SUV(max) during 18FDG-PET scanning in small cell lung cancer: Similar information as in non-small cell lung cancer? Lung Cancer 2012;76:67–71.
Contents lists available at SciVerse ScienceDirect
Lung Cancer journal homepage: www.elsevier.com/locate/lungcan
Clinical applicability of staging small cell lung cancer according to the seventh edition of the TNM staging system Byung Woo Jhun a , Kyung-Jong Lee a , Kyeongman Jeon a , Gee Young Suh a , Man Pyo Chung a , Hojoong Kim a , O Jung Kwon a , Jong-Mu Sun b , Jin Seok Ahn b , Myung-Ju Ahn b , Keunchil Park b , Joon Young Choi c , Kyung Soo Lee d , Joungho Han e , Sang-Won Um a,∗ a
Division of Pulmonary and Critical Care Medicine, Department of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea Division of Hematology-Oncology, Department of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea c Department of Nuclear Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea d Department of Radiology and Center for Imaging Science, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea e Department of Pathology, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea b
a r t i c l e
i n f o
Article history: Received 2 January 2013 Received in revised form 18 February 2013 Accepted 5 March 2013 Keywords: Small cell lung cancer UICC 7th TNM stage Two-stage Survival Prognosis Applicability
a b s t r a c t The two-stage system of limited and extensive disease has been widely employed for small cell lung cancer (SCLC). However, the International Association for the Study of Lung Cancer has proposed that the TNM classification should be incorporated into clinical practice. The purpose of this study was to evaluate the applicability of the Union for International Cancer Control (UICC) 7th TNM staging system to SCLC. We retrospectively reviewed the medical records of consecutive patients with newly diagnosed histologically proven SCLC between March 2005 and January 2010. Patients who had other concurrent malignancies or had combined-type SCLC were excluded. We assessed overall survival (OS) according to the T descriptor, N descriptor, M descriptor, and TNM stage grouping. In total, 320 SCLC patients were included. Median age was 65 years and 286 patients (89.4%) were male. Median OS was 12.7 months. There were no significant differences in OS according to the T descriptor (P = 0.880). However, there were significant differences in OS according to the N (P < 0.001) and M (P < 0.001) descriptors and TNM stage grouping (P < 0.001). Hazard ratios for OS, adjusted for known prognostic factors, differed significantly according to the N and M descriptor, and TNM stage grouping, but not according to the T descriptors. The UICC 7th TNM staging system may contribute to a more precise prognosis in SCLC patients. Further studies are required to evaluate the applicability of the TNM staging system to SCLC. © 2013 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Small cell lung cancer (SCLC) accounts for 15–20% of all lung cancer cases [1,2]. It generally has a rapid doubling time, a high growth fraction, and early development of hematogenous metastasis [3]. Only about one-third of patients have limited disease confined to the chest at initial presentation. SCLC is highly responsive to initial treatment with cytotoxic agents or radiation; however, because of the disease’s high recurrence rate and aggressive behavior, the survival rate over the past decades has not increased in line with expectations [4,5]. With currently available treatments, median
∗ Corresponding author at: Division of Pulmonary and Critical Care Medicine, Department of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, 50 Irwon-dong, Gangnam-gu, Seoul, 135-710, Republic of Korea. Tel.: +82 2 3410 3429; fax: +82 2 3410 6956. E-mail addresses: [email protected], [email protected] (S.-W. Um). 0169-5002/$ – see front matter © 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.lungcan.2013.03.005
survival ranges from 16 to 26 months even in patients with limited disease [6,7]. Without treatment, median overall survival (OS) ranges from 2 to 4 months [8]. The precise determination of the prognosis is therefore important in choosing an adequate treatment modality and determining treatment strategies for individual patients. Clinical stage is known to be closely correlated with prognosis [9,10]. The two-stage system introduced by the Veterans Affairs Lung Study Group has been routinely used to classify SCLC cases as limited or extensive disease. Nevertheless, this two-stage system is problematic when applied to patients with extremely limited disease [11] or with locoregionally advanced disease, such as those with contralateral hilar and supraclavicular nodes [12–15] or malignant effusions [16]. Some studies have suggested that a separate analysis of prognosis, rather than a two-stage categorization, may be beneficial. For example, Kreisman et al. reported that SCLC presenting as a lung mass without nodal involvement had superior survival rates compared with the general category of limited SCLC [11]. Shepherd et al. showed that a subgroup of
66
B.W. Jhun et al. / Lung Cancer 81 (2013) 65–70
patients within limited-stage disease with no mediastinal nodal involvement had better median survival than patients with mediastinal lymphadenopathy [17,18]. In addition, Zhu et al. recently demonstrated significant survival differences within both the limited-stage and extensive-stage groups according to metabolic parameters measured by PET/CT [19]. Although the patient groups were categorized by metabolic parameters, rather than using the TNM staging system, the data suggest that a separate analysis of prognosis in both limited-stage and extensive-stage disease may be possible. Moreover, in recent studies by the International Association for the Study of Lung Cancer (IASLC), evaluation of the applicability of the seventh edition of the TNM system to SCLC revealed the prognostic significance of TNM stages [20,21]. However, there are few data regarding the applicability of the Union for International Cancer Control (UICC) 7th TNM staging system to SCLC in clinical practice. The purposes of the present study were to evaluate the applicability of the UICC 7th TNM staging system to SCLC and to evaluate whether the TNM staging system can provide additional prognostic information compared with the two-stage system.
2. Materials and methods 2.1. Study patients and data collection We reviewed the medical records of all adult patients who were newly diagnosed with histologically proven SCLC between March 2005 and January 2010 at Samsung Medical Center, a referral hospital in Seoul, South Korea. Patients with other concurrent malignancies or combined-type SCLC (SCLC with squamous cell carcinoma and/or adenocarcinoma) were excluded from the study. Data were collected retrospectively for clinical manifestations, laboratory and radiological tests, cyto-histopathology, and survival, including physical examination, laboratory investigations, chest radiography, contrast-enhanced CT scans of the chest and upper abdomen, integrated PET/CT scans, magnetic resonance imaging of the brain, and cyto-histopathological results of transthoracic needle biopsy or bronchoscopic biopsy. Patients who received only best supportive care were excluded in the survival analyses. All study patients who had been classified by the two-stage system were restaged according to the UICC 7th TNM staging. The Institutional Review Board at the Samsung Medical Center approved the study protocol. Informed consent was waived because of the retrospective nature of the study. 2.2. Two-stage and TNM staging systems 2.2.1. Two-stage In the current study, limited-stage disease was defined as disease confined to the ipsilateral hemithorax that could be encompassed within a tolerable radiation port. Thus, ipsilateral/ contralateral mediastinal lymph nodes, ipsilateral/contralateral supraclavicular lymph nodes, recurrent laryngeal nerve involvement, and superior vena cava obstruction were all included. Extensive-stage disease was defined as disease beyond the ipsilateral hemithorax, including malignant pleural or pericardial effusion and hematogenous metastasis. Patients with tumors that could not be included within a tolerable radiation field were also considered to have extensive stage disease [22]. 2.2.2. TNM staging system All study patients were restaged according to seventh edition of the TNM staging system [23]. Restaging was performed with consideration of all available clinical information,
radiological findings on contrast-enhanced CT and integrated PET/CT, and cyto-histopathological data. 2.3. Survival analysis in relation to the TNM staging system OS was defined as the time interval from histopathological diagnosis to death from any cause or the end of follow-up. The last follow-up date was January 31, 2012. We compared the OS of the study patients in relation to the T descriptor (Nany , Many ), N descriptor (Tany , Many ), M descriptor (Tany , Nany ), and TNM stage grouping. 2.4. Integrated 18F-FDG PET/CT acquisition and data analysis After the patients fasted for at least 6 h before their PET/CT examinations, they received an intravenous injection of 370 MBq of 18F-FDG and then rested for 45 (43–47) min before undergoing imaging. Image acquisition was performed using an integrated PET/CT device (Discovery LS, GE Healthcare, Milwaukee, WI) that consisted of an Advance NXi PET scanner and an 8-slice Light Speed Plus CT scanner. The 18F-FDG PET data were evaluated using the maximum standardized uptake value (SUVmax ). We used the region of interest analysis tools included with the scanner to calculate the SUVmax over the primary tumor after correction for the injected dose of 18F-FDG and the patient’s weight. 2.5. Statistical analysis The data are presented as number (percentage) or median (interquartile range [IQR]). To compare the OS in relation to each TNM descriptor, Kaplan–Meier analysis was performed and survival curves were compared by log-rank testing. Cox proportional hazard models for OS were also created. To compare the prognostic usefulness of the TNM staging system and the two-stage system, Kaplan–Meier analysis was performed in relation to the two-stages (limited and extensive) and the TNM stage groupings 1 (I + II, III, IV) and 2 (stages I, II, III, IV); the survival curves were compared by log-rank testing and P-values were corrected using Bonferroni’s method to counteract the problem of multiple comparisons. The patient population was also subdivided by median values for clinico-laboratory characteristics that have been demonstrated to be prognostic factors: age, Eastern Cooperative Oncology Group (ECOG) performance status (PS), lactate dehydrogenase (LDH) level, albumin level, and maximum standardized uptake (SUVmax ). Univariate survival rate analysis was performed using the Kaplan–Meier method, with comparisons of survival curves by log-rank testing. To assess the possible independent effects of the T descriptor (model 1), N descriptor (model 2), M descriptor (model 3), TNM stage grouping (models 4 and 5), and two-stage (model 6) on OS, multivariate analysis was performed with a Cox proportional hazard model using variables with P-values <0.10 in the univariate analyses; P-values were corrected using Bonferroni’s method to counteract the problem of multiple comparisons. Values of P < 0.05 were considered to indicate statistical significance. All statistical analyses were performed using PASW 20.0 (SPSS Inc., Chicago, IL). 3. Results 3.1. Patients In total, 362 patients with histologically proven SCLC were identified during the study period. Of these, 42 patients were excluded: 21 had combined-type SCLC and 21 had other concurrent malignancies. Consequentially, 320 patients with SCLC were included in the analysis and 23 patients who received only best supportive care were excluded in the survival analyses. Characteristics of the study patients are shown in Table 1. The median age was 65 years (59–71
B.W. Jhun et al. / Lung Cancer 81 (2013) 65–70
67
Table 1 Characteristics of the study patients. Characteristic
Number (%) or median (IQR)
Number of patients Age, years Male/female Smoking Current smoker Ex-smoker Never smoker Overall survival, months ECOG performance status 0, 1 ≥2 Weight loss Lactate dehydrogenase level, IU/L Albumin, g/dL PET/CT scan SUVmax Pleural effusion Pericardial effusion Brain metastasis at diagnosis Treatment Definitive CCRT Definitive chemotherapy 2 (0.6) Palliative chemotherapy Palliative radiotherapy Palliative chemotherapy and radiotherapy 100 (31.3) Best supportive care CR during study period at any time
320 65 (59–71) 286 (89.4)/34 (10.6) 229 (71.5) 69 (21.6) 22 (6.9) 12.7 (7.5–23.2) 251 (78.4) 69 (21.6) 93 (29.1) 431 (357.8–549.3) 3.8 (3.5–4.1) 254 (79.4) 10.5 (8.3–12.5) 75 (23.4) 9 (2.8) 66 (20.6) 91 (28.4) 91 (28.4) 13 (4.1)
23 (7.2) 34 (10.6)
IQR, interquartile range; ECOG, eastern cooperative oncology group; SUVmax , maximum standardized uptake, there is missing data in 66 cases for SUVmax ; PET, positron emission tomography; CT, computed tomography; CCRT, concurrent chemoradiotherapy; CR, complete remission.
years), and 286 patients (89.4%) were men. Median OS was 12.7 months (7.5–23.2 months). No patients were lost to follow-up. Two hundred ninety-seven patients (92.8%) received definitive or palliative chemotherapy with or without radiotherapy, and 23 patients (7.2%) received best supportive care alone. No patient underwent surgery for SCLC. Most patients had a relatively good performance status. The treatment strategies in our institution were consistent during the study period. Most commonly used chemotherapy regimen was etoposide/cisplatin in 160 patients (50.0%) followed by irinotecan/cisplatin in 41 patients (12.9%), etoposide/carboplatin in 37 patients (11.6%), irinotecan/carboplatin in 27 patients (8.4%), and so on. In ninety-one patients who received definitive RT, radiotherapy was delivered as 2 Gy once daily to the median of 52.5 Gy (IQR 45.0–52.5). In 13 patients who received palliative RT for lung, spine, and brain, a median of 27.0 Gy (IQR: 20.0–30.0) was irradiated. 3.2. Survival in relation to the TNM staging system OS comparisons among the TNM descriptors are shown in Table 2. The T category did not show a definite correlation with OS, and differences between survival curves according to T category were not significant (P = 0.880; log-rank test). Hazard ratios (HRs) for OS between T stages were not significant. A higher N category was associated with poorer OS, with significant differences between survival curves according to N category (P < 0.001; log-rank test); HR for OS between N1 and N2 or N2 and N3 differed significantly, but not between N0 and N1. As the M category increased, OS became poorer, and the survival curves according to M category differed significantly (P < 0.001; log-rank test). HR for OS between the M0 and M1a stages differed significantly, but not between M1a and M1b. Table 3 presents OS comparisons for the two-stage system and TNM stage groupings. For the two-stage system, extensive-stage
Fig. 1. A, Kaplan–Meier overall survival (OS) curve for all SCLC patients according to two-stage grouping. B, Kaplan–Meier OS curve according to TNM grouping (I + II, III, IV). C, Kaplan–Meier OS curve according to TNM grouping (I, II, III, IV), P, P-value from log-rank test.
disease predictably showed significantly poorer OS compared with limited-stage disease (P < 0.001, Fig. 1). For the TNM stages, OS tended to decrease as TNM stage grouping increased. Owing to the relatively small numbers of patients with early-stage disease (IA, IB, IIA, and IIB), the stages were considered as three (I + II, III, and IV) or four (I, II, III, and IV) groups. Increased stage within each stage grouping was associated with poorer OS, and the differences between the survival curves were significant (P < 0.001 for each; log-rank test, Fig. 1). HRs for OS between two (limited and extensive) and three (I + II, III, and IV) stage groupings differed significantly (Table 3). There were also significant differences between III and IV in four (I, II, III, and IV) stage grouping (Table 3). Compared with the two-stage system, TNM stage groupings can provide additional prognostic subgroups in patients with SCLC. 3.3. Univariate survival analysis in relation to clinical characteristics Univariate analysis of OS was performed according to the median values for the following clinico-laboratory characteristics, which have been identified as prognostic factors: age, ECOG PS, LDH level, albumin level, and SUVmax (Table 4). There were significant
68
B.W. Jhun et al. / Lung Cancer 81 (2013) 65–70
Table 2 Overall survival in small cell lung cancer patients according to TNM descriptors of the 7th edition of the TNM staging system. Category T descriptor (Nany , Many )
N descriptor (Tany , Many )
M descriptor (Tany , Nany )
T0 T1 T2 T3 T4 N0 N1 N2 N3 M0 M1a M1b
No .of patient (%)
Median (IQR) OS months
43 (14.5) 60 (20.2) 108 (36.4) 55 (18.5) 31 (10.4) 11 (3.7) 30 (10.1) 115 (38.7) 141 (47.5) 122 (41.1) 40 (13.5) 135 (45.4)
14.7 (9.2–18.2) 19.5 (9.1–27.6) 13.0 (7.1–22.8) 12.2 (8.4–25.5) 11.9 (8.9–18.8) 39.0 (13.7–58.4) 23.7 (18.4–30.0) 14.8 (9.6–25.0) 10.3 (7.1–16.2) 21.8 (11.5–31.0) 11.5 (6.9–19.7) 11.1 (7.4–15.6)
Comparison
HR (95% CI)
P*
T1 vs. T0 T2 vs. T1 T3 vs. T2 T4 vs. T3
0.6 (0.3–1.0) 1.5 (0.9–2.3) 1.1 (0.7–1.6) 1.1 (0.7–1.7)
N1 vs. N0 N2 vs. N1 N3 vs. N2
3.2 (0.8–14.3) 1.8 (1.0–3.1) 1.7 (1.2–2.3)
0.135 0.042 <0.001
M1a vs. M0 M1b vs. M1a
2.5 (1.6–3.8) 1.2 (0.8–1.8)
<0.001 0.999
0.084 0.140 0.999 0.999
IQR, interquartile range; OS, overall survival; HR, hazard ratio; CI, confidence interval; *P-value for HR, relative to preceding row.
Table 3 Overall survival in small cell lung cancer patients according to two-stage and 7th edition of the TNM stage grouping. Category Two-stage
Limited Extensive I + II III IV I II III IV
TNM stage grouping
TNM stage grouping
No .of patient (%)
Median (IQR) OS months
118 (39.7) 179 (60.3) 23 (7.7) 98 (33.0) 176 (59.3) 6 (2.0) 17 (5.7) 98 (33.0) 176 (59.3)
22.4 (11.5–31.7) 11.5 (7.4–15.9) 26.9 (23.2–39.7) 18.9 (10.1–29.2) 11.2 (7.4–15.7) 43.7 (36.1–62.2) 24.3 (20.7–31.1) 18.9 (10.1–29.2) 11.2 (7.4–15.7)
Comparison
HR (95% CI)
P*
extensive vs. limited
2.9 (2.1–3.9)
<0.001
III vs. I + II IV vs. III
2.7 (1.3–5.5) 2.4 (1.8–3.4)
<0.006 <0.001
II vs. I III vs. II IV vs. III
3.1 (0.6–21.8) 1.7 (0.8–3.8) 2.5 (1.7–3.5)
0.999 0.324 <0.001
IQR, interquartile range; OS, overall survival; HR, hazard ratio; CI, confidence interval; *P-value for HR, relative to preceding row.
differences in OS in relation to age (P = 0.006), ECOG PS (P < 0.001), LDH level (P < 0.001), and albumin level (P = 0.003). However, there was no difference in OS in relation to SUVmax (P = 0.829). 3.4. Multivariate survival analysis in prognostic models
4. Discussion
To assess the possible independent effects of the T descriptor (model 1), N descriptor (model 2), M descriptor (model 3), TNM grouping (model 4 and 5), and two-stage (model 6) on OS, we performed multivariate analyses adjusted for age, ECOG PS, LDH level, and albumin level, all of which had P-values <0.10 in the univariate analyses (Table 5). The N descriptor (model 2), M descriptor (model 3), three TNM stage (I + II, III, and IV) grouping (model 4), and twostage (model 6) remained significant in multivariate analyses, and the estimated hazard ratio increased with advancing N, M, and TNM
Table 4 Univariate analysis of overall survival in relation to clinical characteristics. Characteristic Age, years <65 ≥65 ECOG PS 0, 1 ≥2 LDH, IU/L <430 ≥430 Albumin, g/dL ≥3.8 <3.8 SUVmax <10.5 ≥10.5
No. of patients
Median OS months
146 151
15.8 11.5
244 53
14.8 8.7
151 146
18.8 10.0
168 129
15.3 11.2
122 124
14.2 14.6
stage groupings (Table 5). The T descriptor (model 1) and four TNM stage (I, II, III, and IV) grouping (model 5) were not significant in the multivariate analysis (Table 5).
P* 0.006
<0.001
<0.001
0.003 0.829
OS, overall survival; ECOG, Eastern Cooperative Oncology Group; PS, performance status; LDH, lactate dehydrogenase; SUVmax , maximum standardized uptake, there is missing data in 51 cases for SUVmax ; *P-value from log-rank test, relative to preceding row.
In the current study, we showed that the seventh edition of the TNM staging system can provide additional prognostic information compared with the two-stage system, even after adjustment for other known prognostic factors. HRs for OS in relation to TNM stage grouping (I + II, III, and IV) were significantly higher for more advanced TNM stage groups, when adjusted for age, ECOG PS, LDH level, and albumin level. However, the HRs did not differ significantly in four TNM stage (I, II, III, and IV) grouping, probably because of the relatively small numbers of patients with early-stage disease. HRs for OS between M categories, N1 and N2, or N2 and N3 remained significant, when adjusted for age, ECOG PS, LDH level, and albumin level (Table 5). The lack of a significant difference in the hazard ratio between N0 and N1 stages may be due to the relatively small numbers of patients with N0 or N1 disease, given that survival curves according to N category differed significantly (P < 0.001; log-rank test). However, HRs for OS between T categories were not significant. These results are supported by a few previous studies by IASLC on the applicability of the TNM staging system to SCLC. Shepherd et al. reported that both the sixth and seventh editions of the TNM staging system could more precisely predict survival rates in SCLC, and these authors indicated that stratification by stage (I–III) should be incorporated in clinical trials [20]. Ignatius et al. reported that compared with the sixth edition, the seventh edition of the TNM staging system resulted in better discrimination of survivals in SCLC patients [21]. One of the remarkable aspects of the current study is that after we adjusted for other known prognostic factors to evaluate the applicability of TNM staging system more practically, the correlation between TNM stage and SCLC prognosis remained (Table 5). Therefore, our data may have clinical significance and
B.W. Jhun et al. / Lung Cancer 81 (2013) 65–70
69
Table 5 Multivariate analysis for overall survival in prognostic models. Characteristic
Model 1 HR (95% CI)
Age, years 1 <65 ≥65 1.2 (1.0–1.5) ECOG PS 1 0, 1 ≥2 2.1 (1.5–2.9) LDH, IU/L 1 <430 ≥430 1.8 (1.4–2.3) Albumin, g/dL ≥3.8 1 <3.8 1.1 (0.9–1.5) T descriptor 1 T0 0.6 (0.3–1.1) T1 1.4 (0.9–2.3) T2 1.0 (0.7–1.3) T3 1.2 (0.6–2.1) T4 N descriptor NA N0 N1 N2 N3 M descriptor NA M0 M1a M1b TNM stage grouping I + II NA III IV TNM stage grouping NA I II III IV Two-stage NA Limited Extensive
Model 2
Model 3
Model 4
Model 5
Model 6
P*
HR (95% CI)
P*
HR (95% CI)
P*
HR (95% CI)
P*
HR (95% CI)
P*
HR (95% CI)
P*
0.176
1 1.4 (1.1–1.8)
<0.001
1 1.2 (0.9–1.6)
<0.001
1 1.2 (1.0–1.6)
0.098
1 1.3 (1.0–1.7)
0.089
1 1.2 (1.0–1.5)
0.195
<0.001
1 1.9 (1.3–2.6)
<0.001
1 1.9 (1.3–2.7)
<0.001
1 1.9 (1.3–2.6)
<0.001
1 1.8 (1.3–2.5)
<0.001
1 1.9 (1.3–2.6)
<0.001
<0.001
1 1.6 (1.3–2.1)
<0.001
1 1.7 (1.2–2.3)
<0.001
1 1.6 (1.2–2.1)
<0.001
1 1.6 (1.2–2.1)
<0.001
1 1.6 (1.2–2.1)
<0.001
0.395
1 1.1 (0.9–1.5)
0.324
1 1.1 (0.8–1.5)
0.638
1 1.1 (0.9–1.4)
0.458
1 1.1 (0.9–1.4)
0.470
1 1.1 (0.9–1.5)
<0.298
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
0.100 0.196 0.999 0.999 1 3.9 (0.9–16.3) 1.9 (1.0–3.3) 1.4 (1.0–2.0) NA
NA
NA
NA
0.084 0.030 0.039 1 1.9 (1.2–3.1) 1.4 (0.9–2.2) NA
NA
NA
0.002 0.102 1 2.8 (1.3–6.0) 2.2 (1.5–2.9) NA
NA
0.004 <0.001 1 3.6 (0.6–22.0) 1.8 (0.8–3.8) 2.2 (1.6–3.0) NA
NA 0.999 0.180 <0.001 1 2.6 (2.0–3.4)
<0.001
All models are adjusted for age, ECOG PS, LDH, and albumin; Model 1, UICC 7th T descriptor. Model 2, UICC 7th N descriptor. Model 3, UICC 7th M descriptor. Model 4, UICC 7th three TNM stage grouping (I + II, III, and IV). Model 5, UICC 7th four TNM stage grouping (I, II, III, and IV). Model 6, Two-stage; HR, hazard ratio; NA, not applicable; *P-value for HR, relative to preceding row.
may provide more precise information for evaluating prognosis in SCLC using the TNM staging system. Our data for the T descriptor did not show a significant correlation between T category and OS, unlike the findings of some previous studies [11,20,21,24]. However, there have also been studies with results similar to ours. For example, Lim et al. evaluated survival in 59 limited-stage SCLC patients who underwent surgical resection or adjuvant chemotherapy, and found no clear survival difference according to T category [25]. Angeletti et al. evaluated the influence of T and N stages on long-term survival in 49 SCLC patients who underwent surgical resection or adjuvant chemotherapy. They showed that survival was significantly influenced by N, but not T category [26]. The contradictory findings for the T category may be partly explained by the characteristics of the tumors, which usually present with bulky lymphadenopathy and without a lung tumor or conglomerated lesion [27,28], making the accurate assessment of T category difficult. Further studies on the applicability of the TNM staging system in T0 SCLC without distant metastasis are needed. The roles of PET/CT scan in the staging of SCLC are still controversial. Some authors insisted that PET/CT scan might enable more accurate clinical staging [29–31]. However, PET/CT scan could have a false positive result to define mediastinal and distant metastases. SUVmax can also be variably interpreted by several factors including patient glucose level and variability in scanners. However, in
our institution, the protocol for PET/CT scan was consistent during study period. In the current study, 34 (10.6%) patients were restaged from limited to extensive due to the detection of distant metastasis based on PET/CT scan and the managements were changed from concurrent chemoradiotherapy to chemotherapy. The most common sites of metastasis was bone (n = 14), followed by chest wall or contra-lateral lung (n = 7), distant lymph nodes including intra-abdominal or axillary lymph nodes (n = 5), liver (n = 4), adrenal gland (n = 2), breast (n = 1), and brain (n = 1). Of these, 4 patients were pathologically confirmed (axillary lymph node = 2, bone = 1, breast = 1). Thirty metastatic lesions were not pathologically confirmed due to high risk of complications at the physician’s discretion. Interestingly, differences in OS according to SUVmax were not statistically significant in the present study, unlike in previous studies showing the relevance of metabolic parameters measured by PET/CT to prognosis of SCLC [19]. However, one report suggested the unsuitability of SUVmax as a prognostic factor. Van der Leest et al. evaluated the clinical role of SUVmax in 75 SCLC patients [32]. OS and disease-free survival were significantly different according to SUVmax only in patients with stage IV disease and who underwent chemotherapy; when all patients were considered, OS did not differ significantly between those with high and low SUVmax values (P = 0.81). Furthermore, there was no difference in OS according to high and low SUVmax among all treated
70
B.W. Jhun et al. / Lung Cancer 81 (2013) 65–70
patients (P = 0.342) or among all non-treated patients (P = 0.291) [32]. The current study has some limitations. First, relatively few early-stage, especially stage I and II, patients were included, as most SCLC patients initially present with advanced disease because of the tumor’s aggressiveness. The more severe cases were admitted to our tertiary referral center. The correlation between TNM stage and OS in SCLC in our study might have been stronger if more early-stage disease cases had been included. Second, not all suspected malignant fluid was confirmed by cytology or biopsy in this study. Among 75 patients who had suspected malignant pleural or pericardial effusion (Table 1), 26 with distant metastases did not undergo cytopathological examination. Among the remaining 49 patients, only 10 were found to have malignant cells by cytopathological examination; the other 39 patients were accepted as having malignant pleural effusion on the basis of bloody exudates that could not be explained by other reasons or clinical judgment that most pleural effusions with lung cancer are due to tumor. Finally, there was no direct comparison of two-stage system and TNM staging system in this study. Therefore, it cannot be ascertained that the TNM staging system is superior to the two-stage system.
[9]
[10]
[11] [12]
[13] [14]
[15]
[16]
[17]
5. Conclusion [18]
The UICC 7th TNM staging system can contribute to a more precise prognosis in patients with SCLC. Further prospective validation studies are required to confirm the applicability of this TNM staging system to SCLC in clinical practice.
[19]
[20]
Conflict of interest statement All authors have no financial disclosures or personal relationships with other people or organisations that could inappropriately influence their work.
[21]
[22]
Acknowledgments [23]
This study was supported by the Samsung Biomedical Research Institute (CA92092). All authors have no financial disclosures or personal relationships with other people or organisations that could inappropriately influence their work. References [1] Govindan R, Page N, Morgensztern D, Read W, Tierney R, Vlahiotis A, et al. Changing epidemiology of small-cell lung cancer in the United States over the last 30 years: analysis of the surveillance, epidemiologic, and end results database. J Clin Oncol 2006;24:4539–44. [2] Janssen-Heijnen ML, Coebergh JW. The changing epidemiology of lung cancer in Europe. Lung Cancer 2003;41:245–58. [3] Elias AD. Small cell lung cancer: state-of-the-art therapy in 1996. Chest 1997;112:251S–8S. [4] Murray N, Turrisi 3rd AT. A review of first-line treatment for small-cell lung cancer. J Thorac Oncol 2006;1:270–8. [5] Hann CL, Rudin CM. Management of small-cell lung cancer: incremental changes but hope for the future. Oncology (Williston Park) 2008;22:1486–92. [6] Johnson BE, Bridges JD, Sobczeck M, Gray J, Linnoila RI, Gazdar AF, et al. Patients with limited-stage small-cell lung cancer treated with concurrent twice-daily chest radiotherapy and etoposide/cisplatin followed by cyclophosphamide, doxorubicin, and vincristine. J Clin Oncol 1996;14:806–13. [7] Turrisi 3rd AT, Kim K, Blum R, Sause WT, Livingston RB, Komaki R, et al. Twicedaily compared with once-daily thoracic radiotherapy in limited small-cell lung cancer treated concurrently with cisplatin and etoposide. N Engl J Med 1999;340:265–71. [8] Foster NR, Mandrekar SJ, Schild SE, Nelson GD, Rowland Jr KM, Deming RL, et al. Prognostic factors differ by tumor stage for small cell lung cancer: a pooled
[24]
[25]
[26]
[27]
[28] [29]
[30]
[31]
[32]
analysis of North Central Cancer Treatment Group trials. Cancer 2009;115: 2721–31. Albain KS, Crowley JJ, LeBlanc M, Livingston RB. Determinants of improved outcome in small-cell lung cancer: an analysis of the 2,580-patient Southwest Oncology Group data base. J Clin Oncol 1990;8:1563–74. Osterlind K, Hansen HH, Hansen M, Dombernowsky P. Mortality and morbidity in long-term surviving patients treated with chemotherapy with or without irradiation for small-cell lung cancer. J Clin Oncol 1986;4:1044–52. Kreisman H, Wolkove N, Quoix E. Small cell lung cancer presenting as a solitary pulmonary nodule. Chest 1992;101:225–31. Urban T, Chastang C, Vaylet F, Mathieu M, Leclerc P, Paillotin D, et al. Prognostic significance of supraclavicular lymph nodes in small cell lung cancer: a study from four consecutive clinical trials, including 1,370 patients. “Petites Cellules” Group. Chest 1998;114:1538–41. Argiris A, Murren JR. Staging and clinical prognostic factors for small-cell lung cancer. Cancer J 2001;7:437–47. Hoffman PC, Golomb HM, Bitran JD, DeMeester TR, Cohen L, Griem ML, et al. Small-cell carcinoma of the lung: a five-year experience with combined modality therapy. Cancer 1980;46:2550–6. Sagman U, Maki E, Evans WK, Warr D, Shepherd FA, Sculier JP, et al. Small-cell carcinoma of the lung: derivation of a prognostic staging system. J Clin Oncol 1991;9:1639–49. Micke P, Faldum A, Metz T, Beeh KM, Bittinger F, Hengstler JG, et al. Staging small cell lung cancer: Veterans Administration Lung Study *Group versus International Association for the Study of Lung Cancer—what limits limited disease? Lung Cancer 2002;37:271–6. Shepherd FA, Ginsberg RJ, Haddad R, Feld R, Sagman U, Evans WK, et al. Importance of clinical staging in limited small-cell lung cancer: a valuable system to separate prognostic subgroups. The University of Toronto Lung Oncology Group. J Clin Oncol 1993;11:1592–7. Shepherd FA, Ginsberg RJ, Feld R, Evans WK, Johansen E. Surgical treatment for limited small-cell lung cancer. The University of Toronto Lung Oncology Group experience. J Thorac Cardiovasc Surg 1991;101:385–93. Zhu D, Ma T, Niu Z, Zheng J, Han A, Zhao S, et al. Prognostic significance of metabolic parameters measured by (18)F-fluorodeoxyglucose positron emission tomography/computed tomography in patients with small cell lung cancer. Lung Cancer 2011;73:332–7. Shepherd FA, Crowley J, Van Houtte P, Postmus PE, Carney D, Chansky K, et al. The International Association for the Study of Lung Cancer lung cancer staging project: proposals regarding the clinical staging of small cell lung cancer in the forthcoming (seventh) edition of the tumor, node, metastasis classification for lung cancer. J Thorac Oncol 2007;2:1067–77. Ignatius Ou SH, Zell JA. The applicability of the proposed IASLC staging revisions to small cell lung cancer (SCLC) with comparison to the current UICC 6th TNM Edition. J Thorac Oncol 2009;4:300–10. Stahel RGR, Havemann K. Staging and prognostic factors in small cell lung cancer: a consensus report. Lung Cancer 1989;5:119–26. Groome PA, Bolejack V, Crowley JJ, Kennedy C, Krasnik M, Sobin LH, et al. The IASLC Lung Cancer Staging Project: validation of the proposals for revision of the T, N, and M descriptors and consequent stage groupings in the forthcoming (seventh) edition of the TNM classification of malignant tumours. J Thorac Oncol 2007;2:694–705. Uno T, Sumi M, Sawa Y, Teshima T, Hara R, Ikeda H, et al. Process of care and preliminary outcome in limited-stage small-cell lung cancer: results of the 1995–1997 patterns of care study in Japan. Int J Radiat Oncol Biol Phys 2003;55:626–32. Lim E, Belcher E, Yap YK, Nicholson AG, Goldstraw P. The role of surgery in the treatment of limited disease small cell lung cancer: time to reevaluate. J Thorac Oncol 2008;3:1267–71. Angeletti CA, Macchiarini P, Mussi A, Basolo F. Influence of T and N stages on long-term survival in resectable small cell lung cancer. Eur J Surg Oncol 1989;15:337–40. Forster BB, Muller NL, Miller RR, Nelems B, Evans KG. Neuroendocrine carcinomas of the lung: clinical, radiologic, and pathologic correlation. Radiology 1989;170:441–5. Pearlberg JL, Sandler MA, Lewis Jr JW, Beute GH, Alpern MB. Small-cell bronchogenic carcinoma: CT evaluation. AJR Am J Roentgenol 1988;150:265–8. Bradley JD, Dehdashti F, Mintun MA, Govindan R, Trinkaus K, Siegel BA. Positron emission tomography in limited-stage small-cell lung cancer: a prospective study. J Clin Oncol 2004;22:3248–54. Fischer BM, Mortensen J, Langer SW, Loft A, Berthelsen AK, Petersen BI, et al. A prospective study of PET/CT in initial staging of small-cell lung cancer: comparison with CT, bone scintigraphy and bone marrow analysis. Ann Oncol 2007;18:338–45. Azad A, Chionh F, Scott AM, Lee ST, Berlangieri SU, White S, et al. High impact of 18F-FDG-PET on management and prognostic stratification of newly diagnosed small cell lung cancer. Mol Imaging Biol 2010;12:443–51. van der Leest C, Smit EF, Baas J, Versteijlen RJ, van Walree N, Hoogsteden HC, et al. SUV(max) during 18FDG-PET scanning in small cell lung cancer: Similar information as in non-small cell lung cancer? Lung Cancer 2012;76:67–71.