Pulmonary sarcomatoid carcinoma: an analysis of a rare cancer from the Surveillance, Epidemiology, and End Results database

Pulmonary sarcomatoid carcinoma: an analysis of a rare cancer from the Surveillance,... Abstract OBJECTIVES Pulmonary sarcomatoid carcinoma (PSC) is a rare malignant neoplasm that accounts for a small percentage of non-small-cell lung carcinoma (NSCLC). At least 10% of PSCs has a spindle and/or giant cell component, which is often associated with a poor prognosis. We reviewed the Surveillance, Epidemiology, and End Results (SEER) database for the clinicopathological characteristics and surgical outcomes of PSCs. METHODS The SEER database (1973–2013) was queried for PSC. A comparison between PSC and other NSCLC patients was performed. Cox regression for overall survival (OS) and logistic regression for node-positive predictors were performed. A propensity-matched (1:2) analysis (including age, gender, grade and stage) among surgically treated cases was done to compare OS in PSC versus other NSCLCs. RESULTS A total of 955 899 NSCLC patients were identified; of these, 4987 patients had been diagnosed with PSC (0.52%). Men represented 60.9% of cases, with a median age of 68 years. The median size of the tumour was 5 cm and 3.5 cm in PSCs and NSCLCs, respectively (P < 0.001). PSC patients had significantly less Stage I, more high-grade tumours, advanced T stage, N+ disease and M1 disease (P < 0.001). In the PSC cohort, the most significant predictor of N+ disease on multivariate analysis was advanced T stage (P < 0.001). Predictors of OS in Stages I/II PSC on multivariate analysis were advanced age [P < 0.001, hazard ratio (HR) = 1.03], male gender (P = 0.024, HR = 1.25), carcinosarcoma (P = 0.002, HR = 1.76), grade (P = 0.033, HR = 1.81), T stage (P = 0.003, HR = 1.75), N status (P = 0.001, HR = 1.90) and surgical resection (P < 0.001, HR = 0.58). Among matched surgically resected cohorts, a poorer prognosis for OS was evident in PSCs in early stages (I/II) than in other NSCLCs (P = 0.009). CONCLUSIONS PSC patients present with more advanced stage and with worse survival outcomes than other NSCLC patients. While surgical resection conveys a survival advantage in PSC, this group represents a population at a high risk for relapse and should be evaluated for novel adjuvant therapies. Sarcomatoid carcinoma, Survival, Radiation, SEER database, Node-positive predictors, Propensity score match INTRODUCTION Pulmonary sarcomatoid carcinoma (PSC) is a rare malignant neoplasm, accounting for less than 1% of all lung cancers [1]. It may contain both carcinomatous and sarcomatous components and is often associated with a poor prognosis [2, 3]. The World Health Organization (WHO) classification system recognizes 5 types of PSCs: pure spindle cell, pure giant cell, carcinosarcoma, pulmonary blastoma and pleomorphic carcinoma [2]. Although a diagnosis of PSC can be suggested on small biopsies and cytology specimens, a definitive diagnosis should only be made on the resected specimens [2]. There are little data from retrospective studies about these rare tumours. Controversy exists with regard to the prognostic significance of PSCs [4, 5]. Many studies are small case series with a focus mostly on pathological descriptions [4–9], with little clinical outcomes data. We reviewed a large national database for clinicopathological characteristics and surgical outcomes for this tumour. METHODS Data sources The Surveillance, Epidemiology, and End Results (SEER) database of the National Cancer Institute (NCI) was queried retrospectively to identify PSCs among lung cancer patients based on inpatient status. The Public Use version of data collected from the SEER 18 registries between 1973 and 2013 was used for this study. We included only patients with single primary tumour (sequence number = 0 or 1), as survival in patients with multiple primary tumours could not be ascribed to a single anatomical cancer site. Study population and inclusion criteria According to the International Classification of Diseases for Oncology, 3rd edition codes (ICD-1-O Codes), pleomorphic carcinoma (8022/3), spindle cell carcinoma (8032/3), giant cell carcinoma (8031/3), carcinosarcoma (8980/3) and pulmonary blastoma (8972/3) are grouped under the term PSC [2, 10]. Among our cohort, giant and spindle cell carcinoma represented (72.9%), pleomorphic carcinoma (13.4%), carcinosarcoma (11.5%) and pulmonary blastoma (2.2%) of the entire cohort. Data of patients for whom information was obtained solely from the autopsy or death certificate and information that lacked survival time were excluded. Study variables Available data in the SEER files that included variables, such as gender, age, race, year of diagnosis, histological subgroup, grade, stage with its components T, N and M stage, tumour size, surgical resection, survival months and vital status, and whether the patient was alive or dead at the last follow-up were retrieved. Patients who died in less than a month were coded as having a survival time of zero in the SEER registries. We assigned a survival time of 1.5 months according to a previously established and standard epidemiological convention [11]. Survival and follow-up data Overall survival (OS) is identified as the time ranging from the date of diagnosis until the date of death from any cause or the date of the last follow-up. The interquartile range (IQR) of the follow-up period among our cohort was 10–91 months with a median of 41 months. The median (IQRs) follow-up period in giant and spindle cell carcinomas was 33 (9–90) months, pleomorphic carcinoma was 42 (9–74) months, carcinosarcoma was 53 (20.25–101) months and pulmonary blastoma was 80 (20–135) months. Statistical methods The primary outcome was to identify the OS in PSCs in the whole group and in different stages of PSCs and histology using the Kaplan–Meier survival curves with log-rank test. The secondary outcomes were to compare different clinicopathological parameters and survival data between PSCs and a contemporaneously treated group of non-small-cell lung carcinoma (NSCLC) cohorts. Subgroup analysis of our PSC cohort, which was represented by 2 historical periods of 20 years, early (1973–1993) vs late (1994–2013), was conducted. Continuous variables were presented as median and IQR and compared using the Mann–Whitney U-test. Categorical variables were reported as numbers (frequency percentages) and compared using the Pearson’s χ2 test. Cox regression analysis was used to identify the factors predicting survival in both univariate and multivariate analyses. Logistic regression analysis for predictors of node-positive disease was performed. A propensity score (PS)-matched analysis (including age as a continuous variable, gender, grade and stage) among surgically treated cases was done to compare OS in PSC patients versus other NSCLC patients. Matching was done 1 to 2 with the use of the nearest neighbour methods without replacement using 0.01 caliper width. The Kaplan–Meier survival curves were done and compared using log-rank test, where a P-value of <0.05 was considered statistically significant. Hazard ratios (HRs) and 95% confidence intervals (95% CIs) were calculated. Statistical analyses were conducted using IBM SPSS software (Released 2013; IBM SPSS Statistics for Windows, Version 22.0; IBM Corp., Armonk, NY, USA), PS-matching package version 3.03 and SPSS statistics R essentials. RESULTS Demographics and clinicopathological data A total of 955 899 NSCLC patients were identified; of these, 4987 patients had PSC (0.52%). The majority were men (60.9%), and the median age was 68 years. In all, 11.4% of the PSC patients versus 11% in other NSCLC patients were Blacks (P < 0.001). High-grade tumours represented 97.2% of PSCs vs 61.9% of other NSCLCs (P < 0.001). Stage I represented 20.5% of PSCs vs other NSCLCs (24.2%, P < 0.001, Table 1). More advanced T stages (T3–4) were prevalent in PSCs (54.8%) vs other NSCLCs (45.5%, P < 0.001). Median tumour size was 5 (3.3–7.5) cm and 3.5 (2.3–5.4) cm in PSCs and other NSCLCs, respectively (P < 0.001). Significantly more patients with PSC had node-positive disease (49.7% vs 45.3% in other NSCLCs; P < 0.001) as well as M1 disease (44.3% vs 42.2% in NSCLCs; P < 0.001). Table 1: Comparison between PSC and NSCLC cohorts   PSC (n = 4987)  NSCLC (n = 950 912)  P-value  Patient characteristics         Age (years), median (IQR)  68 (59–76)  70 (61–77)  <0.001   Female gender, n (%)  1948 (39.1)  406 279 (42.7)  <0.001   Stage, n (%)          I  779 (20.5)  172 231 (24.2)  <0.001    II  321 (8.4)  39 259 (5.5)    III  995 (26.1)  189 637 (26.7)    IV  1713 (45)  307 910 (43.4)   Grade 3–4, n (%)  2528 (97.2)  295 672 (61.9)  <0.001   Surgery, n (%)    Yes  2011 (40.3)  247 018 (26)  <0.001   Survival [median OS (5-year OS)], n (%)          All stages  6 months (15.5)      Matched surgical cohort (clinicopathological characteristics)   Age (years), median (IQR)  67 (59–74)  67 (59–74)  0.86   Female gender, n (%)  960 (44.9)  485 (45.4)  0.80   Stage, n (%)           I  842 (39.4)  418 (39.1)  0.81     II  389 (18.2)  194 (18.1)     III  611 (28.6)  296 (27.7)     IV  296 (13.8)  161 (15.1)   Grade 3–4, n (%)  2046 (95.7)  1023 (95.7)  1.00   Survival by stage [median OS, 5-year OS)], n (%)          All stages  18 months (30.5)  27 months (32.5)  0.004    Stage I/II  32 months (40.3)  47 months (43.9)  0.009    Stage III  12 months (23.2)  18 months (21.9)  0.38    Stage IV  5 months (8.5)  7 months (7.9)  0.30    PSC (n = 4987)  NSCLC (n = 950 912)  P-value  Patient characteristics         Age (years), median (IQR)  68 (59–76)  70 (61–77)  <0.001   Female gender, n (%)  1948 (39.1)  406 279 (42.7)  <0.001   Stage, n (%)          I  779 (20.5)  172 231 (24.2)  <0.001    II  321 (8.4)  39 259 (5.5)    III  995 (26.1)  189 637 (26.7)    IV  1713 (45)  307 910 (43.4)   Grade 3–4, n (%)  2528 (97.2)  295 672 (61.9)  <0.001   Surgery, n (%)    Yes  2011 (40.3)  247 018 (26)  <0.001   Survival [median OS (5-year OS)], n (%)          All stages  6 months (15.5)      Matched surgical cohort (clinicopathological characteristics)   Age (years), median (IQR)  67 (59–74)  67 (59–74)  0.86   Female gender, n (%)  960 (44.9)  485 (45.4)  0.80   Stage, n (%)           I  842 (39.4)  418 (39.1)  0.81     II  389 (18.2)  194 (18.1)     III  611 (28.6)  296 (27.7)     IV  296 (13.8)  161 (15.1)   Grade 3–4, n (%)  2046 (95.7)  1023 (95.7)  1.00   Survival by stage [median OS, 5-year OS)], n (%)          All stages  18 months (30.5)  27 months (32.5)  0.004    Stage I/II  32 months (40.3)  47 months (43.9)  0.009    Stage III  12 months (23.2)  18 months (21.9)  0.38    Stage IV  5 months (8.5)  7 months (7.9)  0.30  IQR: interquartile range; NSCLC: non-small-cell lung carcinoma; OS: overall survival; PSC: pulmonary sarcomatoid carcinoma. Among the PSCs, predictors of N +  disease on multivariate analysis were advanced T stage (P < 0.001; Table 2). N +  disease was evident in 52.3% of giant/spindle cell carcinoma, 49% in pleomorphic carcinoma, 38.5% of carcinosarcoma and 25.6% of pulmonary blastoma (P < 0.001). Surgery was done in 2011 patients (40.3%). Procedure was known in 1979 patients [local destruction (laser ablation, cautery or fulguration) in 1.2% of patients, sublobar resection in 14.1% of patients, lobe/bilobectomy in 1170 (59.1%) patients, pneumonectomy in 201 (10.2%) patients and resection/surgery NOS in 305 (15.4%) patients]. Table 2: Predictors of N+ among PSC cohort (n = 1671/4987) Independent variables  Univariate predictors   Multivariate predictors   OR (95% CI)  P-value  OR (95% CI)  P-value  Age (continuous variable)  0.996 (0.991–1.002)  0.17      Gender   Female (n = 1337)  Reference         Male (n = 1983)  1.08 (0.94–1.24)  0.26      Race           White (n = 2785)  Reference    Reference     Black (n = 370)  1.24 (0.99–1.55)  0.051  1.25 (0.89–1.75)  0.19   Others/unknown (n = 165)  0.99 (0.73–1.37)  0.99  0.94 (0.58–1.51)  0.79  Year of diagnosis (continuous variable)  0.99 (0.98–1.01)  0.43      Histology   Pleomorphic carcinoma (n = 494)  Reference    Reference     Giant and spindle cell carcinomas (n = 2510)  1.14 (0.94–1.39)  0.18  1.09 (0.83–1.45)  0.52   Pulmonary blastoma (n = 43)  0.36 (0.18–0.73)  0.004  0.66 (0.20–2.17)  0.49   Carcinosarcoma (n = 273)  0.65 (0.48–0.88)  0.005  0.92 (0.59–1.44)  0.73  Differentiation grade   G1–2 (n = 51)  Reference    Reference     G3–4 (n = 1846)  2.53 (1.36–4.70)  0.003  1.51 (0.74–3.06)  0.26  T stage           T0, 1 (n = 335)  Reference    Reference     T2 (n = 993)  1.41 (1.08–1.85)  0.012  1.17 (0.82–1.66)  0.38   T3, 4 (n = 1463)  3.46 (2.67–4.48)  <0.001  2.28 (1.62–3.21)  <0.001  M stage           M0 (n = 1972)  Reference    Reference     M1 (n = 1294)  3.37 (2.90–3.90)  <0.001  2.31 (1.83–2.92)  <0.001  Independent variables  Univariate predictors   Multivariate predictors   OR (95% CI)  P-value  OR (95% CI)  P-value  Age (continuous variable)  0.996 (0.991–1.002)  0.17      Gender   Female (n = 1337)  Reference         Male (n = 1983)  1.08 (0.94–1.24)  0.26      Race           White (n = 2785)  Reference    Reference     Black (n = 370)  1.24 (0.99–1.55)  0.051  1.25 (0.89–1.75)  0.19   Others/unknown (n = 165)  0.99 (0.73–1.37)  0.99  0.94 (0.58–1.51)  0.79  Year of diagnosis (continuous variable)  0.99 (0.98–1.01)  0.43      Histology   Pleomorphic carcinoma (n = 494)  Reference    Reference     Giant and spindle cell carcinomas (n = 2510)  1.14 (0.94–1.39)  0.18  1.09 (0.83–1.45)  0.52   Pulmonary blastoma (n = 43)  0.36 (0.18–0.73)  0.004  0.66 (0.20–2.17)  0.49   Carcinosarcoma (n = 273)  0.65 (0.48–0.88)  0.005  0.92 (0.59–1.44)  0.73  Differentiation grade   G1–2 (n = 51)  Reference    Reference     G3–4 (n = 1846)  2.53 (1.36–4.70)  0.003  1.51 (0.74–3.06)  0.26  T stage           T0, 1 (n = 335)  Reference    Reference     T2 (n = 993)  1.41 (1.08–1.85)  0.012  1.17 (0.82–1.66)  0.38   T3, 4 (n = 1463)  3.46 (2.67–4.48)  <0.001  2.28 (1.62–3.21)  <0.001  M stage           M0 (n = 1972)  Reference    Reference     M1 (n = 1294)  3.37 (2.90–3.90)  <0.001  2.31 (1.83–2.92)  <0.001  Variables with P-value ≤0.10 in univariate analysis were involved in MVA. CI: confidence interval; OR: odds ratio; MVA: multivariate analysis; PSC: pulmonary sarcomatoid carcinoma. Subgroup analysis of the PSC patients revealed that late time period (1994–2013) has 3865 (77.5%) patients with median age of 70 (60–77) vs 63 (54–71) years in the early group (P < 0.001). This was reflected in the percentage of patients who underwent surgery (39.3 vs 48.1%). Regarding histology, pleomorphic carcinoma predominates in late years (13.9 vs 11.8%) with similar predomination of pulmonary blastoma (2.6 vs 0.8%; P < 0.001). There was no statistically significant difference regarding race, grade, stage or OS (P > 0.05). Among 2011 (40.3%) patients who underwent surgery, subgroup analysis according to the two 20-year periods revealed that late time period (1994–2013) had 1505 (74.8%) patients with a median age of 67 (58–75) vs 60 (53–68) years in the early group (1973–1993; P < 0.001). Although the late time period group had older patients, they showed a better OS [median OS was 20 months (95% CI = 17.2–22.8) vs 13 months (95% CI = 10.6–15.4) for the early group, P = 0.013] that may be related to change in the histological subtypes with predominance of less aggressive types in the late time period. Pleomorphic carcinoma predominated in the late years (18.3 vs 9.5%) with similar predomination of pulmonary blastoma (5.8 vs 1.3%); however, carcinosarcoma decreased in the late years (15 vs 25.5%; P < 0.001). A higher number of nodes were sampled/dissected in the late years [median (IQR): 8 (4–13) vs 7 (4–11), P = 0.005] with lower use of radiation therapy in the late years (28.2 vs 38.1%, P < 0.001) and there was no statistically significant difference regarding race, grade or stage (P > 0.05). Survival data Among the whole PSC cohort, the median OS was 6 (95% CI 5.68–6.32) months with a 5-year OS of 15.5%. Predictors of OS in Stages I/II PSC on multivariate analysis were advanced age (P < 0.001, HR = 1.03), male gender (P = 0.024, HR = 1.25), carcinosarcoma (P = 0.002, HR = 1.76), grade (P = 0.033, HR = 1.81), T stage (P = 004, HR = 1.75), N status (P = 0.001, HR = 1.90) and surgical resection (P < 0.001, HR = 0.58; Table 3). The median OS in pleomorphic carcinoma was 35 (95% CI 23.49–46.51) months, giant/spindle cell carcinoma was 22 (95% CI 17.97–26.03) months, pulmonary blastoma was 132 (95% CI 40.20–223.80) months and carcinosarcoma was 24 (95% CI 20.02–27.99) months (P < 0.001; Fig. 1). The 5-year OS was shortest in carcinosarcoma (17%) and longest in pulmonary blastoma (58.3%) (P < 0.001; Fig. 1). Median OS for N0 was 11 (95% CI 9.69–12.31) months, N1 was 7 (95% CI 5.97–8.03) vs 4 (95% CI 3.56–4.44) months and 2 (95% CI 1.27–2.73) months for N2 and N3, respectively. Table 3: Predictors of OS among PSC cohort Stage I/II (n = 1100) Independent variables  Univariate predictors   Multivariate predictors   HR (95% CI)  P-value  HR (95% CI)  P-value  Age (continuous variable)  1.03 (1.027–1.040)  <0.001  1.03 (1.02–1.04)  <0.001  Gender   Female (n = 620)  Reference    Reference     Male (n = 480)  1.29 (1.12–1.49)  0.001  1.25 (1.03–1.52)  0.024  Race           White (n = 942)  Reference         Black (n = 105)  1.02 (0.81–1.28)  0.88       Others/unknown (n = 53)  0.99 (0.71–1.37)  0.94      Year of diagnosis (continuous variable)  0.99 (0.98–1.01)  0.52      Histology           Pleomorphic carcinoma (n = 201)  Reference    Reference     Giant and spindle cell carcinomas (n = 730)  1.15 (0.94–1.40)  0.17  1.14 (1.89–1.46)  0.32   Pulmonary blastoma (n = 32)  0.49 (0.30–0.80)  0.004  1.21 (0.56–2.61)  0.64   Carcinosarcoma (n = 137)  1.41 (1.09–1.82)  0.009  1.76 (1.23–2.53)  0.002  Differentiation grade   G1–2 (n = 34)  Reference    Reference     G3–4 (n = 669)  2.55 (1.55–4.21)  <0.001  1.81 (1.05–3.12)  0.033  T stage           T1 (n = 247)  Reference    Reference     T2 (n = 667)  1.51 (1.26–1.81)  <0.001  1.46 (1.13–1.89)  0.004   T3 (n = 142)  2.06 (1.57–2.70)  <0.001  1.75 (1.21–2.53)  0.003  N stage           N0 (n = 920)  Reference         N + (n = 129)  1.59 (1.29–1.95)  <0.001  1.90 (1.32–2.74)  0.001  Stage            I (n = 779)  Reference          II (n = 321)  1.70 (1.45–1.99)  <0.001      Surgery           No (n = 210)  Reference    Reference     Yes (n = 883)  0.38 (0.32–0.46)  <0.001  0.58 (0.43–0.76)  <0.001  Independent variables  Univariate predictors   Multivariate predictors   HR (95% CI)  P-value  HR (95% CI)  P-value  Age (continuous variable)  1.03 (1.027–1.040)  <0.001  1.03 (1.02–1.04)  <0.001  Gender   Female (n = 620)  Reference    Reference     Male (n = 480)  1.29 (1.12–1.49)  0.001  1.25 (1.03–1.52)  0.024  Race           White (n = 942)  Reference         Black (n = 105)  1.02 (0.81–1.28)  0.88       Others/unknown (n = 53)  0.99 (0.71–1.37)  0.94      Year of diagnosis (continuous variable)  0.99 (0.98–1.01)  0.52      Histology           Pleomorphic carcinoma (n = 201)  Reference    Reference     Giant and spindle cell carcinomas (n = 730)  1.15 (0.94–1.40)  0.17  1.14 (1.89–1.46)  0.32   Pulmonary blastoma (n = 32)  0.49 (0.30–0.80)  0.004  1.21 (0.56–2.61)  0.64   Carcinosarcoma (n = 137)  1.41 (1.09–1.82)  0.009  1.76 (1.23–2.53)  0.002  Differentiation grade   G1–2 (n = 34)  Reference    Reference     G3–4 (n = 669)  2.55 (1.55–4.21)  <0.001  1.81 (1.05–3.12)  0.033  T stage           T1 (n = 247)  Reference    Reference     T2 (n = 667)  1.51 (1.26–1.81)  <0.001  1.46 (1.13–1.89)  0.004   T3 (n = 142)  2.06 (1.57–2.70)  <0.001  1.75 (1.21–2.53)  0.003  N stage           N0 (n = 920)  Reference         N + (n = 129)  1.59 (1.29–1.95)  <0.001  1.90 (1.32–2.74)  0.001  Stage            I (n = 779)  Reference          II (n = 321)  1.70 (1.45–1.99)  <0.001      Surgery           No (n = 210)  Reference    Reference     Yes (n = 883)  0.38 (0.32–0.46)  <0.001  0.58 (0.43–0.76)  <0.001  Variables with P-value ≤0.10 in univariate analysis were involved in MVA. CI: confidence interval; HR: hazard ratio; MVA: multivariate analysis; OS: overall survival; PSC: pulmonary sarcomatoid carcinoma. Figure 1: View largeDownload slide The Kaplan–Meier overall survival for different histology in Stage I/II pulmonary sarcomatoid carcinomas; CS: carcinosarcoma; G/S: giant/spindle cell carcinoma; PB: pulmonary blastoma; PLC: pleomorphic carcinoma. Figure 1: View largeDownload slide The Kaplan–Meier overall survival for different histology in Stage I/II pulmonary sarcomatoid carcinomas; CS: carcinosarcoma; G/S: giant/spindle cell carcinoma; PB: pulmonary blastoma; PLC: pleomorphic carcinoma. An analysis of propensity-matched groups of surgically resected PSCs (n = 1069) and other NSCLCs (n = 2138) revealed worse OS among PSCs in all stages with a median and 5-year OS of 18 (95% CI 15.20–20.80) months and 30.5%, respectively, compared with 27 (95% CI 24.31–29.69) months and 32.5%, respectively, in the other NSCLCs (P = 0.009; Fig. 2, Table 1). Figure 2: View largeDownload slide The Kaplan–Meier overall survival in PSC versus NSCLC among the matched cohort. NSCLC: non-small-cell lung carcinoma; PSC: pulmonary sarcomatoid carcinoma. Figure 2: View largeDownload slide The Kaplan–Meier overall survival in PSC versus NSCLC among the matched cohort. NSCLC: non-small-cell lung carcinoma; PSC: pulmonary sarcomatoid carcinoma. Figure 3: View largeDownload slide The Kaplan–Meier overall survival for different stages in PSC versus NSCLC among the matched cohort. NSCLC: non-small-cell lung carcinoma; PSC: pulmonary sarcomatoid carcinoma. Figure 3: View largeDownload slide The Kaplan–Meier overall survival for different stages in PSC versus NSCLC among the matched cohort. NSCLC: non-small-cell lung carcinoma; PSC: pulmonary sarcomatoid carcinoma. In Stages I/II, the median and 5-year OS was 32 (95% CI 24.48–39.52) months and 40.3% compared with 47 (95% CI 41.61–52.39) months and 32.5%, respectively, in the other NSCLCs. In advanced stage (Stage IV), median OS was 5 (95% CI 3.78–6.22) months in PSCs vs 7 (95% CI = 5.77–8.23) months in NSCLCs (Fig. 3 Table 1). On analysing radiation therapy variable, we found a median and 5-year OS of 5 months (8%) for patients known to have received radiation therapy (n = 2060) vs 6 months (20.6%) for patients known to have not received it (P < 0.001). For patients who did not undergo surgery, 1396 (48.8%) patients had radiotherapy with a median and 5-year OS of 4 months (3.4%) compared with 2 months (4.8%) patients who did not receive radiotherapy (P < 0.001). Regarding sequencing of surgery and radiation therapy, we found that 576 (11.6%) patients known to have received adjuvant radiation therapy, 78 (1.6%) known to have received neoadjuvant radiation therapy and only 11 (0.4%) patients had both neoadjuvant and adjuvant radiation therapy. The median and 5-year OS for patients who had neoadjuvant or both neoadjuvant and adjuvant radiation therapy was 11 months (28.6%) vs 9 months (15.1%) for the adjuvant radiation therapy group (P = 0.018). No data regarding chemotherapy or targeted therapy were available in the public version of SEER database. To avoid heterogeneity among the long time frame of our study, we did a subgroup analysis from 2004 till the end of our study. We matched surgically resected PSC (n = 465) to a contemporaneously surgically resected NSCLC (n = 930) based on the same variables used before (age, gender, grade and stage) that revealed a median and 5-year OS of 24 months (37.9%) for PSC vs 42 months (42.7%) for other NSCLC (P = 0.001). A total of 656 (13.2%) patients were alive at the end of the study. The main causes of death among our cohort were death from lung cancer in 3574 (71.6%), heart disease in 171 (3.4%), chronic obstructive pulmonary disease in 84 (1.7%) and documented death from other causes of death in 61(1.2%) patients, whereas the remaining patients died from discrete causes, each represented <1% of the cohort. DISCUSSION This study has analysed a large cohort of PSC patients. We have leveraged the large SEER database to systematically study this rare cancer. Brambilla et al. [12] in 2001 reported that sarcomatoid histology is rare and found only in 0.1–0.4% of all patients with NSCLC. In this work, PSC accounted for 0.52% of NSCLCs. Of note, the gender distribution seen in this study (60%) is nearly similar to that reported by Martin et al. [3] (54%) but different from the proportion of men observed by others [7, 13], who reported that more than 90% of their cohort were men. In line with prior series, most of these tumours, when graded, were high grade [13, 14]. Prior series reported a short median survival between 8 and 19 months [3, 5, 8, 9, 13, 15–17] for patients with PSC, which is inferior to other NSCLCs. However, only some of the patients in these series underwent surgical resection. Other series reported no survival differences; however, their cohorts were small (<40 each) [4, 18]. Even in early-stage PSCs (Stage I/II), the OS in our study is only a median of 32 months. Similar to that reported by Martin et al. [3], this study reported poor survival outcome in a PS-matched analysis of PSCs versus other NSCLCs is related to histology and is independent of other tumour features. In the series by Nakajima et al. [4] in which the authors included 37 and 647 patients of PSCs and NSCLC, respectively, there was no significant survival differences. However, they did not perform multivariable analysis or stage-matched comparisons. Also, Pelosi et al. [5] reported similar cancer-related survival among Stage I PSC versus other NSCLCs. Thirty-one PSC cases were included in their analysis; however, Stage III represented only 13%. Similar to NSCLC patients, PSC patients with advanced age had worse survival [19]. Carcinosarcomas are clonal tumours [20, 21] developing through sarcomatoid change (rhabdomyosarcoma, chondrosarcoma and osteosarcoma) in a carcinoma (adenocarcinoma or squamous carcinoma) [20, 21] TP53 mutations are common in contrast to KRAS and EGFR. The presence of EGFR mutation is rare in carcinosarcoma; however, when present it is associated with better survival outcomes with the use of anti-EGFR treatment [22, 23]. Prior small case series have reported that PSC is resistant to conventional chemotherapy [24], emphasizing the need to identify targetable genomic alterations. In the SEER database study by Wright et al. [25], carcinosarcomas of the bladder were reported to be more aggressive with poorer survival and higher recurrence when compared with other subtypes of bladder sarcomatoid carcinoma. The authors suggested separation of bladder carcinosarcomas as a separate category from sarcomatoid carcinoma [25]. This raises the question of whether or not we should also consider pulmonary carcinosarcomas as a separate entity from PSC as it was associated with the poorest survival among the remaining PSC histological subtypes. The role of radiation therapy in PSC remains controversial and needs future randomized controlled trials as a prior series reported that irradiation causes sarcomatous or anaplastic changes in carcinoma [4, 26]. We found lower OS in patients who received radiation therapy, although only 1 month difference between both groups, but we are studying an already aggressive disease. Strengths and limitations The strength of this study is that it is the largest reported PSC series in the literature and utilizes a PS-matched analysis. The inherent limitation of the analysis is that it is retrospective, and our results must be tempered by the limitations of available information included in the SEER database. For example, little information exists with regard to the details of radiation and chemotherapy. CONCLUSIONS PSC patients present with more advanced stage disease with worse survival compared to other NSCLC patients. While surgical resection conveys a survival advantage in PSC, this group represents a high-risk population for relapse and should be evaluated for novel adjuvant therapies. ACKNOWLEDGEMENTS The authors wish to thank Mrs Sara E. Cain, Senior Medical Secretary, Department of Cardiothoracic Surgery, for her efforts in editing English language in some parts in the manuscript. Conflict of interest: none declared. REFERENCES 1 Yendamuri S, Caty L, Pine M, Adem S, Bogner P, Miller A. Outcomes of sarcomatoid carcinoma of the lung: a surveillance, epidemiology, and end results database analysis. Surgery  2012; 152: 397– 402. Google Scholar CrossRef Search ADS PubMed  2 Travis WD, Brambilla E, Nicholson AG, Yatabe Y, Austin JH, Beasley MB et al.   The 2015 World Health Organization classification of lung tumors: impact of genetic, clinical and radiologic advances since the 2004 classification. J Thorac Oncol  2015; 10: 1243– 60. Google Scholar CrossRef Search ADS PubMed  3 Martin LW, Correa AM, Ordonez NG, Roth JA, Swisher SG, Vaporciyan AA et al.   Sarcomatoid carcinoma of the lung: a predictor of poor prognosis. Ann Thorac Surg  2007; 84: 973– 80. Google Scholar CrossRef Search ADS PubMed  4 Nakajima M, Kasai T, Hashimoto H, Iwata Y, Manabe H. Sarcomatoid carcinoma of the lung. Cancer  1999; 86: 608– 16. Google Scholar CrossRef Search ADS PubMed  5 Pelosi G, Gasparini P, Cavazza A, Rossi G, Graziano P, Barbareschi M et al.   Multiparametric molecular characterization of pulmonary sarcomatoid carcinoma reveals a nonrandom amplification of anaplastic lymphoma kinase (ALK) gene. Lung Cancer  2012; 77: 507– 14. Google Scholar CrossRef Search ADS PubMed  6 Fishback NF, Travis WD, Moran CA, Guinee DG, McCarthy WF, Koss MN. Pleomorphic (spindle/giant cell) carcinoma of the lung. Cancer  1994; 73: 2936– 45. Google Scholar CrossRef Search ADS PubMed  7 Chang Y-L, Lee Y-C, Shih J-Y, Wu C-T. Pulmonary pleomorphic (spindle) cell carcinoma: peculiar clinicopathologic manifestations different from ordinary non-small cell carcinoma. Lung Cancer  2001; 34: 91– 7. Google Scholar CrossRef Search ADS PubMed  8 Nappi O, Glasner SD, Swanson PE, Wick MR. Biphasic and monophasic sarcomatoid carcinomas of the lung: a reappraisal of “carcinosarcomas” and “spindle-cell carcinomas”. Am J Clin Pathol  1994; 102: 331– 40. Google Scholar CrossRef Search ADS PubMed  9 Nishida K, Kobayashi Y, Ishikawa Y, Satoh Y, Okumura S, Nishimura H et al.   Sarcomatoid adenocarcinoma of the lung: clinicopathological, immunohistochemical and molecular analyses. Anticancer Res  2001; 22: 3477– 83. 10 Fritz A, Percy C, Jack A, Shanmugaratnam K, Sobin L, Parkin DM et al.   International Classification of Diseases for Oncology . Geneva: World Health Organization, 2000. 11 Koepsell TD, Weiss NS. Epidemiologic Methods: Studying the Occurrence of Illness . UK: Oxford University Press, 2014. Google Scholar CrossRef Search ADS   12 Brambilla E, Travis WD, Colby TV, Corrin B, Shimosato Y. The new World Health Organization classification of lung tumours. Eur Respir J  2001; 18: 1059– 68. Google Scholar CrossRef Search ADS PubMed  13 Rossi G, Cavazza A, Sturm N, Migaldi M, Facciolongo N, Longo L et al.   Pulmonary carcinomas with pleomorphic, sarcomatoid, or sarcomatous elements: a clinicopathologic and immunohistochemical study of 75 cases. Am J Surg Pathol  2003; 27: 311– 24. Google Scholar CrossRef Search ADS PubMed  14 Pelosi G, Sonzogni A, De Pas T, Galetta D, Veronesi G, Spaggiari L et al.   Review article: pulmonary sarcomatoid carcinomas: a practical overview. Int J Surg Pathol  2010; 18: 103– 20. Google Scholar CrossRef Search ADS PubMed  15 Raveglia F, Mezzetti M, Panigalli T, Furia S, Giuliani L, Conforti S et al.   Personal experience in surgical management of pulmonary pleomorphic carcinoma. Ann Thorac Surg  2004; 78: 1742– 7. Google Scholar CrossRef Search ADS PubMed  16 Ro JY, Chen JL, Lee JS, Sahin AA, Ordóñez NG, Ayala AG. Sarcomatoid carcinoma of the lung. Immunohistochemieal and ultrastructural studies of 14 cases. Cancer  1992; 69: 376– 86. Google Scholar CrossRef Search ADS PubMed  17 Hummel P, Cangiarella JF, Cohen J-M, Yang G, Waisman J, Chhieng DC. Transthoracic fine-needle aspiration biopsy of pulmonary spindle cell and mesenchymal lesions. Cancer Cytopathol  2001; 93: 187– 98. Google Scholar CrossRef Search ADS   18 Pelosi G, Fraggetta F, Nappi O, Pastorino U, Maisonneuve P, Pasini F et al.   Pleomorphic carcinomas of the lung show a selective distribution of gene products involved in cell differentiation, cell cycle control, tumor growth, and tumor cell motility: a clinicopathologic and immunohistochemical study of 31 cases. Am J Surg Pathol  2003; 27: 1203– 15. Google Scholar CrossRef Search ADS PubMed  19 Shokralla HA, Rahouma M. Prognostic clinico-pathological features of 99 cases advanced non-small cell lung cancer—Egyptian National Cancer Institute. Adv Lung Cancer  2016; 4: 29. Google Scholar CrossRef Search ADS   20 Pardo J, Aisa G, de Alava E, Sola JJ, Panizo A, Rodríguez-Spiteri N et al.   Primary mixed squamous carcinoma and osteosarcoma (carcinosarcomas) of the lung have a CGH mapping similar to primitive squamous carcinomas and osteosarcomas. Diagn Mol Pathol  2008; 17: 151– 8. Google Scholar CrossRef Search ADS PubMed  21 Dacic S, Finkelstein SD, Sasatomi E, Swalsky PA, Yousem SA. Molecular pathogenesis of pulmonary carcinosarcoma as determined by microdissection-based allelotyping. Am J Surg Pathol  2002; 26: 510– 6. Google Scholar CrossRef Search ADS PubMed  22 Vokes EE, Chu E. Anti-EGFR therapies: clinical experience in colorectal, lung, and head and neck cancers. Oncology (Williston Park)  2006; 20: 15– 25. Google Scholar PubMed  23 Stiles BM, Nasar A, Hussein MK, Ghaly GR, Ahmed MR, Port JL et al.   Routine molecular testing of resected early-stage lung adenocarcinoma with targeted next-generation sequencing demonstrates a high rate of actionable mutations. J Thorac Oncol  2016; 11: S44– 5. Google Scholar CrossRef Search ADS   24 Ouziane I, Boutayeb S, Mrabti H, Lalya I, Rimani M, Errihani H. Sarcomatoid carcinoma of the lung: a model of resistance of chemotherapy. N Am J Med Sci  2014; 6: 342– 5. Google Scholar CrossRef Search ADS PubMed  25 Wright JL, Black PC, Brown GA, Porter MP, Kamat AM, Dinney CP et al.   Differences in survival among patients with sarcomatoid carcinoma, carcinosarcoma and urothelial carcinoma of the bladder. J Urol  2007; 178: 2302– 7. Google Scholar CrossRef Search ADS PubMed  26 Steuer CE, Behera M, Liu Y, Fu C, Gillespie TW, Saba NF et al.   Pulmonary sarcomatoid carcinoma: an analysis of the National Cancer Data Base. Clin Lung Cancer  2017; 18: 286– 92. Google Scholar CrossRef Search ADS PubMed  APPENDIX. CONFERENCE DISCUSSION Dr E. Vallieres(Seattle, WA, USA): You are reporting on a 40-year period of data here. You are telling me that during these 40 years, the pathology definitions of these rare tumours have not evolved and have not changed? Because you and I both know they have, so how accurate is the information you have just presented? Dr B. Lee(New York, NY, USA): It is true, we see it even in non-small-cell lung cancer, how the definitions of terms are so different even this year, compared to 5 years ago. I do think that in the earlier stages, 40 years ago, the groupings were probably a little bit more put into one, or in fact back then perhaps they were not even called specific tumours that we use the terms for today. I think the terminology that we were able to cite in this article is definitely what we are more accustomed to in 2017, while term carcinosarcoma or pulmonary blastoma, these probably were not actually defined tumours 40 years ago. © The Author 2017. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png European Journal of Cardio-Thoracic Surgery Oxford University Press

Pulmonary sarcomatoid carcinoma: an analysis of a rare cancer from the Surveillance, Epidemiology, and End Results database

Loading next page...
 
/lp/ou_press/pulmonary-sarcomatoid-carcinoma-an-analysis-of-a-rare-cancer-from-the-9dq0ZhfhPO
Publisher
Oxford University Press
ISSN
1010-7940
eISSN
1873-734X
D.O.I.
10.1093/ejcts/ezx417
Publisher site
See Article on Publisher Site

Abstract

Abstract OBJECTIVES Pulmonary sarcomatoid carcinoma (PSC) is a rare malignant neoplasm that accounts for a small percentage of non-small-cell lung carcinoma (NSCLC). At least 10% of PSCs has a spindle and/or giant cell component, which is often associated with a poor prognosis. We reviewed the Surveillance, Epidemiology, and End Results (SEER) database for the clinicopathological characteristics and surgical outcomes of PSCs. METHODS The SEER database (1973–2013) was queried for PSC. A comparison between PSC and other NSCLC patients was performed. Cox regression for overall survival (OS) and logistic regression for node-positive predictors were performed. A propensity-matched (1:2) analysis (including age, gender, grade and stage) among surgically treated cases was done to compare OS in PSC versus other NSCLCs. RESULTS A total of 955 899 NSCLC patients were identified; of these, 4987 patients had been diagnosed with PSC (0.52%). Men represented 60.9% of cases, with a median age of 68 years. The median size of the tumour was 5 cm and 3.5 cm in PSCs and NSCLCs, respectively (P < 0.001). PSC patients had significantly less Stage I, more high-grade tumours, advanced T stage, N+ disease and M1 disease (P < 0.001). In the PSC cohort, the most significant predictor of N+ disease on multivariate analysis was advanced T stage (P < 0.001). Predictors of OS in Stages I/II PSC on multivariate analysis were advanced age [P < 0.001, hazard ratio (HR) = 1.03], male gender (P = 0.024, HR = 1.25), carcinosarcoma (P = 0.002, HR = 1.76), grade (P = 0.033, HR = 1.81), T stage (P = 0.003, HR = 1.75), N status (P = 0.001, HR = 1.90) and surgical resection (P < 0.001, HR = 0.58). Among matched surgically resected cohorts, a poorer prognosis for OS was evident in PSCs in early stages (I/II) than in other NSCLCs (P = 0.009). CONCLUSIONS PSC patients present with more advanced stage and with worse survival outcomes than other NSCLC patients. While surgical resection conveys a survival advantage in PSC, this group represents a population at a high risk for relapse and should be evaluated for novel adjuvant therapies. Sarcomatoid carcinoma, Survival, Radiation, SEER database, Node-positive predictors, Propensity score match INTRODUCTION Pulmonary sarcomatoid carcinoma (PSC) is a rare malignant neoplasm, accounting for less than 1% of all lung cancers [1]. It may contain both carcinomatous and sarcomatous components and is often associated with a poor prognosis [2, 3]. The World Health Organization (WHO) classification system recognizes 5 types of PSCs: pure spindle cell, pure giant cell, carcinosarcoma, pulmonary blastoma and pleomorphic carcinoma [2]. Although a diagnosis of PSC can be suggested on small biopsies and cytology specimens, a definitive diagnosis should only be made on the resected specimens [2]. There are little data from retrospective studies about these rare tumours. Controversy exists with regard to the prognostic significance of PSCs [4, 5]. Many studies are small case series with a focus mostly on pathological descriptions [4–9], with little clinical outcomes data. We reviewed a large national database for clinicopathological characteristics and surgical outcomes for this tumour. METHODS Data sources The Surveillance, Epidemiology, and End Results (SEER) database of the National Cancer Institute (NCI) was queried retrospectively to identify PSCs among lung cancer patients based on inpatient status. The Public Use version of data collected from the SEER 18 registries between 1973 and 2013 was used for this study. We included only patients with single primary tumour (sequence number = 0 or 1), as survival in patients with multiple primary tumours could not be ascribed to a single anatomical cancer site. Study population and inclusion criteria According to the International Classification of Diseases for Oncology, 3rd edition codes (ICD-1-O Codes), pleomorphic carcinoma (8022/3), spindle cell carcinoma (8032/3), giant cell carcinoma (8031/3), carcinosarcoma (8980/3) and pulmonary blastoma (8972/3) are grouped under the term PSC [2, 10]. Among our cohort, giant and spindle cell carcinoma represented (72.9%), pleomorphic carcinoma (13.4%), carcinosarcoma (11.5%) and pulmonary blastoma (2.2%) of the entire cohort. Data of patients for whom information was obtained solely from the autopsy or death certificate and information that lacked survival time were excluded. Study variables Available data in the SEER files that included variables, such as gender, age, race, year of diagnosis, histological subgroup, grade, stage with its components T, N and M stage, tumour size, surgical resection, survival months and vital status, and whether the patient was alive or dead at the last follow-up were retrieved. Patients who died in less than a month were coded as having a survival time of zero in the SEER registries. We assigned a survival time of 1.5 months according to a previously established and standard epidemiological convention [11]. Survival and follow-up data Overall survival (OS) is identified as the time ranging from the date of diagnosis until the date of death from any cause or the date of the last follow-up. The interquartile range (IQR) of the follow-up period among our cohort was 10–91 months with a median of 41 months. The median (IQRs) follow-up period in giant and spindle cell carcinomas was 33 (9–90) months, pleomorphic carcinoma was 42 (9–74) months, carcinosarcoma was 53 (20.25–101) months and pulmonary blastoma was 80 (20–135) months. Statistical methods The primary outcome was to identify the OS in PSCs in the whole group and in different stages of PSCs and histology using the Kaplan–Meier survival curves with log-rank test. The secondary outcomes were to compare different clinicopathological parameters and survival data between PSCs and a contemporaneously treated group of non-small-cell lung carcinoma (NSCLC) cohorts. Subgroup analysis of our PSC cohort, which was represented by 2 historical periods of 20 years, early (1973–1993) vs late (1994–2013), was conducted. Continuous variables were presented as median and IQR and compared using the Mann–Whitney U-test. Categorical variables were reported as numbers (frequency percentages) and compared using the Pearson’s χ2 test. Cox regression analysis was used to identify the factors predicting survival in both univariate and multivariate analyses. Logistic regression analysis for predictors of node-positive disease was performed. A propensity score (PS)-matched analysis (including age as a continuous variable, gender, grade and stage) among surgically treated cases was done to compare OS in PSC patients versus other NSCLC patients. Matching was done 1 to 2 with the use of the nearest neighbour methods without replacement using 0.01 caliper width. The Kaplan–Meier survival curves were done and compared using log-rank test, where a P-value of <0.05 was considered statistically significant. Hazard ratios (HRs) and 95% confidence intervals (95% CIs) were calculated. Statistical analyses were conducted using IBM SPSS software (Released 2013; IBM SPSS Statistics for Windows, Version 22.0; IBM Corp., Armonk, NY, USA), PS-matching package version 3.03 and SPSS statistics R essentials. RESULTS Demographics and clinicopathological data A total of 955 899 NSCLC patients were identified; of these, 4987 patients had PSC (0.52%). The majority were men (60.9%), and the median age was 68 years. In all, 11.4% of the PSC patients versus 11% in other NSCLC patients were Blacks (P < 0.001). High-grade tumours represented 97.2% of PSCs vs 61.9% of other NSCLCs (P < 0.001). Stage I represented 20.5% of PSCs vs other NSCLCs (24.2%, P < 0.001, Table 1). More advanced T stages (T3–4) were prevalent in PSCs (54.8%) vs other NSCLCs (45.5%, P < 0.001). Median tumour size was 5 (3.3–7.5) cm and 3.5 (2.3–5.4) cm in PSCs and other NSCLCs, respectively (P < 0.001). Significantly more patients with PSC had node-positive disease (49.7% vs 45.3% in other NSCLCs; P < 0.001) as well as M1 disease (44.3% vs 42.2% in NSCLCs; P < 0.001). Table 1: Comparison between PSC and NSCLC cohorts   PSC (n = 4987)  NSCLC (n = 950 912)  P-value  Patient characteristics         Age (years), median (IQR)  68 (59–76)  70 (61–77)  <0.001   Female gender, n (%)  1948 (39.1)  406 279 (42.7)  <0.001   Stage, n (%)          I  779 (20.5)  172 231 (24.2)  <0.001    II  321 (8.4)  39 259 (5.5)    III  995 (26.1)  189 637 (26.7)    IV  1713 (45)  307 910 (43.4)   Grade 3–4, n (%)  2528 (97.2)  295 672 (61.9)  <0.001   Surgery, n (%)    Yes  2011 (40.3)  247 018 (26)  <0.001   Survival [median OS (5-year OS)], n (%)          All stages  6 months (15.5)      Matched surgical cohort (clinicopathological characteristics)   Age (years), median (IQR)  67 (59–74)  67 (59–74)  0.86   Female gender, n (%)  960 (44.9)  485 (45.4)  0.80   Stage, n (%)           I  842 (39.4)  418 (39.1)  0.81     II  389 (18.2)  194 (18.1)     III  611 (28.6)  296 (27.7)     IV  296 (13.8)  161 (15.1)   Grade 3–4, n (%)  2046 (95.7)  1023 (95.7)  1.00   Survival by stage [median OS, 5-year OS)], n (%)          All stages  18 months (30.5)  27 months (32.5)  0.004    Stage I/II  32 months (40.3)  47 months (43.9)  0.009    Stage III  12 months (23.2)  18 months (21.9)  0.38    Stage IV  5 months (8.5)  7 months (7.9)  0.30    PSC (n = 4987)  NSCLC (n = 950 912)  P-value  Patient characteristics         Age (years), median (IQR)  68 (59–76)  70 (61–77)  <0.001   Female gender, n (%)  1948 (39.1)  406 279 (42.7)  <0.001   Stage, n (%)          I  779 (20.5)  172 231 (24.2)  <0.001    II  321 (8.4)  39 259 (5.5)    III  995 (26.1)  189 637 (26.7)    IV  1713 (45)  307 910 (43.4)   Grade 3–4, n (%)  2528 (97.2)  295 672 (61.9)  <0.001   Surgery, n (%)    Yes  2011 (40.3)  247 018 (26)  <0.001   Survival [median OS (5-year OS)], n (%)          All stages  6 months (15.5)      Matched surgical cohort (clinicopathological characteristics)   Age (years), median (IQR)  67 (59–74)  67 (59–74)  0.86   Female gender, n (%)  960 (44.9)  485 (45.4)  0.80   Stage, n (%)           I  842 (39.4)  418 (39.1)  0.81     II  389 (18.2)  194 (18.1)     III  611 (28.6)  296 (27.7)     IV  296 (13.8)  161 (15.1)   Grade 3–4, n (%)  2046 (95.7)  1023 (95.7)  1.00   Survival by stage [median OS, 5-year OS)], n (%)          All stages  18 months (30.5)  27 months (32.5)  0.004    Stage I/II  32 months (40.3)  47 months (43.9)  0.009    Stage III  12 months (23.2)  18 months (21.9)  0.38    Stage IV  5 months (8.5)  7 months (7.9)  0.30  IQR: interquartile range; NSCLC: non-small-cell lung carcinoma; OS: overall survival; PSC: pulmonary sarcomatoid carcinoma. Among the PSCs, predictors of N +  disease on multivariate analysis were advanced T stage (P < 0.001; Table 2). N +  disease was evident in 52.3% of giant/spindle cell carcinoma, 49% in pleomorphic carcinoma, 38.5% of carcinosarcoma and 25.6% of pulmonary blastoma (P < 0.001). Surgery was done in 2011 patients (40.3%). Procedure was known in 1979 patients [local destruction (laser ablation, cautery or fulguration) in 1.2% of patients, sublobar resection in 14.1% of patients, lobe/bilobectomy in 1170 (59.1%) patients, pneumonectomy in 201 (10.2%) patients and resection/surgery NOS in 305 (15.4%) patients]. Table 2: Predictors of N+ among PSC cohort (n = 1671/4987) Independent variables  Univariate predictors   Multivariate predictors   OR (95% CI)  P-value  OR (95% CI)  P-value  Age (continuous variable)  0.996 (0.991–1.002)  0.17      Gender   Female (n = 1337)  Reference         Male (n = 1983)  1.08 (0.94–1.24)  0.26      Race           White (n = 2785)  Reference    Reference     Black (n = 370)  1.24 (0.99–1.55)  0.051  1.25 (0.89–1.75)  0.19   Others/unknown (n = 165)  0.99 (0.73–1.37)  0.99  0.94 (0.58–1.51)  0.79  Year of diagnosis (continuous variable)  0.99 (0.98–1.01)  0.43      Histology   Pleomorphic carcinoma (n = 494)  Reference    Reference     Giant and spindle cell carcinomas (n = 2510)  1.14 (0.94–1.39)  0.18  1.09 (0.83–1.45)  0.52   Pulmonary blastoma (n = 43)  0.36 (0.18–0.73)  0.004  0.66 (0.20–2.17)  0.49   Carcinosarcoma (n = 273)  0.65 (0.48–0.88)  0.005  0.92 (0.59–1.44)  0.73  Differentiation grade   G1–2 (n = 51)  Reference    Reference     G3–4 (n = 1846)  2.53 (1.36–4.70)  0.003  1.51 (0.74–3.06)  0.26  T stage           T0, 1 (n = 335)  Reference    Reference     T2 (n = 993)  1.41 (1.08–1.85)  0.012  1.17 (0.82–1.66)  0.38   T3, 4 (n = 1463)  3.46 (2.67–4.48)  <0.001  2.28 (1.62–3.21)  <0.001  M stage           M0 (n = 1972)  Reference    Reference     M1 (n = 1294)  3.37 (2.90–3.90)  <0.001  2.31 (1.83–2.92)  <0.001  Independent variables  Univariate predictors   Multivariate predictors   OR (95% CI)  P-value  OR (95% CI)  P-value  Age (continuous variable)  0.996 (0.991–1.002)  0.17      Gender   Female (n = 1337)  Reference         Male (n = 1983)  1.08 (0.94–1.24)  0.26      Race           White (n = 2785)  Reference    Reference     Black (n = 370)  1.24 (0.99–1.55)  0.051  1.25 (0.89–1.75)  0.19   Others/unknown (n = 165)  0.99 (0.73–1.37)  0.99  0.94 (0.58–1.51)  0.79  Year of diagnosis (continuous variable)  0.99 (0.98–1.01)  0.43      Histology   Pleomorphic carcinoma (n = 494)  Reference    Reference     Giant and spindle cell carcinomas (n = 2510)  1.14 (0.94–1.39)  0.18  1.09 (0.83–1.45)  0.52   Pulmonary blastoma (n = 43)  0.36 (0.18–0.73)  0.004  0.66 (0.20–2.17)  0.49   Carcinosarcoma (n = 273)  0.65 (0.48–0.88)  0.005  0.92 (0.59–1.44)  0.73  Differentiation grade   G1–2 (n = 51)  Reference    Reference     G3–4 (n = 1846)  2.53 (1.36–4.70)  0.003  1.51 (0.74–3.06)  0.26  T stage           T0, 1 (n = 335)  Reference    Reference     T2 (n = 993)  1.41 (1.08–1.85)  0.012  1.17 (0.82–1.66)  0.38   T3, 4 (n = 1463)  3.46 (2.67–4.48)  <0.001  2.28 (1.62–3.21)  <0.001  M stage           M0 (n = 1972)  Reference    Reference     M1 (n = 1294)  3.37 (2.90–3.90)  <0.001  2.31 (1.83–2.92)  <0.001  Variables with P-value ≤0.10 in univariate analysis were involved in MVA. CI: confidence interval; OR: odds ratio; MVA: multivariate analysis; PSC: pulmonary sarcomatoid carcinoma. Subgroup analysis of the PSC patients revealed that late time period (1994–2013) has 3865 (77.5%) patients with median age of 70 (60–77) vs 63 (54–71) years in the early group (P < 0.001). This was reflected in the percentage of patients who underwent surgery (39.3 vs 48.1%). Regarding histology, pleomorphic carcinoma predominates in late years (13.9 vs 11.8%) with similar predomination of pulmonary blastoma (2.6 vs 0.8%; P < 0.001). There was no statistically significant difference regarding race, grade, stage or OS (P > 0.05). Among 2011 (40.3%) patients who underwent surgery, subgroup analysis according to the two 20-year periods revealed that late time period (1994–2013) had 1505 (74.8%) patients with a median age of 67 (58–75) vs 60 (53–68) years in the early group (1973–1993; P < 0.001). Although the late time period group had older patients, they showed a better OS [median OS was 20 months (95% CI = 17.2–22.8) vs 13 months (95% CI = 10.6–15.4) for the early group, P = 0.013] that may be related to change in the histological subtypes with predominance of less aggressive types in the late time period. Pleomorphic carcinoma predominated in the late years (18.3 vs 9.5%) with similar predomination of pulmonary blastoma (5.8 vs 1.3%); however, carcinosarcoma decreased in the late years (15 vs 25.5%; P < 0.001). A higher number of nodes were sampled/dissected in the late years [median (IQR): 8 (4–13) vs 7 (4–11), P = 0.005] with lower use of radiation therapy in the late years (28.2 vs 38.1%, P < 0.001) and there was no statistically significant difference regarding race, grade or stage (P > 0.05). Survival data Among the whole PSC cohort, the median OS was 6 (95% CI 5.68–6.32) months with a 5-year OS of 15.5%. Predictors of OS in Stages I/II PSC on multivariate analysis were advanced age (P < 0.001, HR = 1.03), male gender (P = 0.024, HR = 1.25), carcinosarcoma (P = 0.002, HR = 1.76), grade (P = 0.033, HR = 1.81), T stage (P = 004, HR = 1.75), N status (P = 0.001, HR = 1.90) and surgical resection (P < 0.001, HR = 0.58; Table 3). The median OS in pleomorphic carcinoma was 35 (95% CI 23.49–46.51) months, giant/spindle cell carcinoma was 22 (95% CI 17.97–26.03) months, pulmonary blastoma was 132 (95% CI 40.20–223.80) months and carcinosarcoma was 24 (95% CI 20.02–27.99) months (P < 0.001; Fig. 1). The 5-year OS was shortest in carcinosarcoma (17%) and longest in pulmonary blastoma (58.3%) (P < 0.001; Fig. 1). Median OS for N0 was 11 (95% CI 9.69–12.31) months, N1 was 7 (95% CI 5.97–8.03) vs 4 (95% CI 3.56–4.44) months and 2 (95% CI 1.27–2.73) months for N2 and N3, respectively. Table 3: Predictors of OS among PSC cohort Stage I/II (n = 1100) Independent variables  Univariate predictors   Multivariate predictors   HR (95% CI)  P-value  HR (95% CI)  P-value  Age (continuous variable)  1.03 (1.027–1.040)  <0.001  1.03 (1.02–1.04)  <0.001  Gender   Female (n = 620)  Reference    Reference     Male (n = 480)  1.29 (1.12–1.49)  0.001  1.25 (1.03–1.52)  0.024  Race           White (n = 942)  Reference         Black (n = 105)  1.02 (0.81–1.28)  0.88       Others/unknown (n = 53)  0.99 (0.71–1.37)  0.94      Year of diagnosis (continuous variable)  0.99 (0.98–1.01)  0.52      Histology           Pleomorphic carcinoma (n = 201)  Reference    Reference     Giant and spindle cell carcinomas (n = 730)  1.15 (0.94–1.40)  0.17  1.14 (1.89–1.46)  0.32   Pulmonary blastoma (n = 32)  0.49 (0.30–0.80)  0.004  1.21 (0.56–2.61)  0.64   Carcinosarcoma (n = 137)  1.41 (1.09–1.82)  0.009  1.76 (1.23–2.53)  0.002  Differentiation grade   G1–2 (n = 34)  Reference    Reference     G3–4 (n = 669)  2.55 (1.55–4.21)  <0.001  1.81 (1.05–3.12)  0.033  T stage           T1 (n = 247)  Reference    Reference     T2 (n = 667)  1.51 (1.26–1.81)  <0.001  1.46 (1.13–1.89)  0.004   T3 (n = 142)  2.06 (1.57–2.70)  <0.001  1.75 (1.21–2.53)  0.003  N stage           N0 (n = 920)  Reference         N + (n = 129)  1.59 (1.29–1.95)  <0.001  1.90 (1.32–2.74)  0.001  Stage            I (n = 779)  Reference          II (n = 321)  1.70 (1.45–1.99)  <0.001      Surgery           No (n = 210)  Reference    Reference     Yes (n = 883)  0.38 (0.32–0.46)  <0.001  0.58 (0.43–0.76)  <0.001  Independent variables  Univariate predictors   Multivariate predictors   HR (95% CI)  P-value  HR (95% CI)  P-value  Age (continuous variable)  1.03 (1.027–1.040)  <0.001  1.03 (1.02–1.04)  <0.001  Gender   Female (n = 620)  Reference    Reference     Male (n = 480)  1.29 (1.12–1.49)  0.001  1.25 (1.03–1.52)  0.024  Race           White (n = 942)  Reference         Black (n = 105)  1.02 (0.81–1.28)  0.88       Others/unknown (n = 53)  0.99 (0.71–1.37)  0.94      Year of diagnosis (continuous variable)  0.99 (0.98–1.01)  0.52      Histology           Pleomorphic carcinoma (n = 201)  Reference    Reference     Giant and spindle cell carcinomas (n = 730)  1.15 (0.94–1.40)  0.17  1.14 (1.89–1.46)  0.32   Pulmonary blastoma (n = 32)  0.49 (0.30–0.80)  0.004  1.21 (0.56–2.61)  0.64   Carcinosarcoma (n = 137)  1.41 (1.09–1.82)  0.009  1.76 (1.23–2.53)  0.002  Differentiation grade   G1–2 (n = 34)  Reference    Reference     G3–4 (n = 669)  2.55 (1.55–4.21)  <0.001  1.81 (1.05–3.12)  0.033  T stage           T1 (n = 247)  Reference    Reference     T2 (n = 667)  1.51 (1.26–1.81)  <0.001  1.46 (1.13–1.89)  0.004   T3 (n = 142)  2.06 (1.57–2.70)  <0.001  1.75 (1.21–2.53)  0.003  N stage           N0 (n = 920)  Reference         N + (n = 129)  1.59 (1.29–1.95)  <0.001  1.90 (1.32–2.74)  0.001  Stage            I (n = 779)  Reference          II (n = 321)  1.70 (1.45–1.99)  <0.001      Surgery           No (n = 210)  Reference    Reference     Yes (n = 883)  0.38 (0.32–0.46)  <0.001  0.58 (0.43–0.76)  <0.001  Variables with P-value ≤0.10 in univariate analysis were involved in MVA. CI: confidence interval; HR: hazard ratio; MVA: multivariate analysis; OS: overall survival; PSC: pulmonary sarcomatoid carcinoma. Figure 1: View largeDownload slide The Kaplan–Meier overall survival for different histology in Stage I/II pulmonary sarcomatoid carcinomas; CS: carcinosarcoma; G/S: giant/spindle cell carcinoma; PB: pulmonary blastoma; PLC: pleomorphic carcinoma. Figure 1: View largeDownload slide The Kaplan–Meier overall survival for different histology in Stage I/II pulmonary sarcomatoid carcinomas; CS: carcinosarcoma; G/S: giant/spindle cell carcinoma; PB: pulmonary blastoma; PLC: pleomorphic carcinoma. An analysis of propensity-matched groups of surgically resected PSCs (n = 1069) and other NSCLCs (n = 2138) revealed worse OS among PSCs in all stages with a median and 5-year OS of 18 (95% CI 15.20–20.80) months and 30.5%, respectively, compared with 27 (95% CI 24.31–29.69) months and 32.5%, respectively, in the other NSCLCs (P = 0.009; Fig. 2, Table 1). Figure 2: View largeDownload slide The Kaplan–Meier overall survival in PSC versus NSCLC among the matched cohort. NSCLC: non-small-cell lung carcinoma; PSC: pulmonary sarcomatoid carcinoma. Figure 2: View largeDownload slide The Kaplan–Meier overall survival in PSC versus NSCLC among the matched cohort. NSCLC: non-small-cell lung carcinoma; PSC: pulmonary sarcomatoid carcinoma. Figure 3: View largeDownload slide The Kaplan–Meier overall survival for different stages in PSC versus NSCLC among the matched cohort. NSCLC: non-small-cell lung carcinoma; PSC: pulmonary sarcomatoid carcinoma. Figure 3: View largeDownload slide The Kaplan–Meier overall survival for different stages in PSC versus NSCLC among the matched cohort. NSCLC: non-small-cell lung carcinoma; PSC: pulmonary sarcomatoid carcinoma. In Stages I/II, the median and 5-year OS was 32 (95% CI 24.48–39.52) months and 40.3% compared with 47 (95% CI 41.61–52.39) months and 32.5%, respectively, in the other NSCLCs. In advanced stage (Stage IV), median OS was 5 (95% CI 3.78–6.22) months in PSCs vs 7 (95% CI = 5.77–8.23) months in NSCLCs (Fig. 3 Table 1). On analysing radiation therapy variable, we found a median and 5-year OS of 5 months (8%) for patients known to have received radiation therapy (n = 2060) vs 6 months (20.6%) for patients known to have not received it (P < 0.001). For patients who did not undergo surgery, 1396 (48.8%) patients had radiotherapy with a median and 5-year OS of 4 months (3.4%) compared with 2 months (4.8%) patients who did not receive radiotherapy (P < 0.001). Regarding sequencing of surgery and radiation therapy, we found that 576 (11.6%) patients known to have received adjuvant radiation therapy, 78 (1.6%) known to have received neoadjuvant radiation therapy and only 11 (0.4%) patients had both neoadjuvant and adjuvant radiation therapy. The median and 5-year OS for patients who had neoadjuvant or both neoadjuvant and adjuvant radiation therapy was 11 months (28.6%) vs 9 months (15.1%) for the adjuvant radiation therapy group (P = 0.018). No data regarding chemotherapy or targeted therapy were available in the public version of SEER database. To avoid heterogeneity among the long time frame of our study, we did a subgroup analysis from 2004 till the end of our study. We matched surgically resected PSC (n = 465) to a contemporaneously surgically resected NSCLC (n = 930) based on the same variables used before (age, gender, grade and stage) that revealed a median and 5-year OS of 24 months (37.9%) for PSC vs 42 months (42.7%) for other NSCLC (P = 0.001). A total of 656 (13.2%) patients were alive at the end of the study. The main causes of death among our cohort were death from lung cancer in 3574 (71.6%), heart disease in 171 (3.4%), chronic obstructive pulmonary disease in 84 (1.7%) and documented death from other causes of death in 61(1.2%) patients, whereas the remaining patients died from discrete causes, each represented <1% of the cohort. DISCUSSION This study has analysed a large cohort of PSC patients. We have leveraged the large SEER database to systematically study this rare cancer. Brambilla et al. [12] in 2001 reported that sarcomatoid histology is rare and found only in 0.1–0.4% of all patients with NSCLC. In this work, PSC accounted for 0.52% of NSCLCs. Of note, the gender distribution seen in this study (60%) is nearly similar to that reported by Martin et al. [3] (54%) but different from the proportion of men observed by others [7, 13], who reported that more than 90% of their cohort were men. In line with prior series, most of these tumours, when graded, were high grade [13, 14]. Prior series reported a short median survival between 8 and 19 months [3, 5, 8, 9, 13, 15–17] for patients with PSC, which is inferior to other NSCLCs. However, only some of the patients in these series underwent surgical resection. Other series reported no survival differences; however, their cohorts were small (<40 each) [4, 18]. Even in early-stage PSCs (Stage I/II), the OS in our study is only a median of 32 months. Similar to that reported by Martin et al. [3], this study reported poor survival outcome in a PS-matched analysis of PSCs versus other NSCLCs is related to histology and is independent of other tumour features. In the series by Nakajima et al. [4] in which the authors included 37 and 647 patients of PSCs and NSCLC, respectively, there was no significant survival differences. However, they did not perform multivariable analysis or stage-matched comparisons. Also, Pelosi et al. [5] reported similar cancer-related survival among Stage I PSC versus other NSCLCs. Thirty-one PSC cases were included in their analysis; however, Stage III represented only 13%. Similar to NSCLC patients, PSC patients with advanced age had worse survival [19]. Carcinosarcomas are clonal tumours [20, 21] developing through sarcomatoid change (rhabdomyosarcoma, chondrosarcoma and osteosarcoma) in a carcinoma (adenocarcinoma or squamous carcinoma) [20, 21] TP53 mutations are common in contrast to KRAS and EGFR. The presence of EGFR mutation is rare in carcinosarcoma; however, when present it is associated with better survival outcomes with the use of anti-EGFR treatment [22, 23]. Prior small case series have reported that PSC is resistant to conventional chemotherapy [24], emphasizing the need to identify targetable genomic alterations. In the SEER database study by Wright et al. [25], carcinosarcomas of the bladder were reported to be more aggressive with poorer survival and higher recurrence when compared with other subtypes of bladder sarcomatoid carcinoma. The authors suggested separation of bladder carcinosarcomas as a separate category from sarcomatoid carcinoma [25]. This raises the question of whether or not we should also consider pulmonary carcinosarcomas as a separate entity from PSC as it was associated with the poorest survival among the remaining PSC histological subtypes. The role of radiation therapy in PSC remains controversial and needs future randomized controlled trials as a prior series reported that irradiation causes sarcomatous or anaplastic changes in carcinoma [4, 26]. We found lower OS in patients who received radiation therapy, although only 1 month difference between both groups, but we are studying an already aggressive disease. Strengths and limitations The strength of this study is that it is the largest reported PSC series in the literature and utilizes a PS-matched analysis. The inherent limitation of the analysis is that it is retrospective, and our results must be tempered by the limitations of available information included in the SEER database. For example, little information exists with regard to the details of radiation and chemotherapy. CONCLUSIONS PSC patients present with more advanced stage disease with worse survival compared to other NSCLC patients. While surgical resection conveys a survival advantage in PSC, this group represents a high-risk population for relapse and should be evaluated for novel adjuvant therapies. ACKNOWLEDGEMENTS The authors wish to thank Mrs Sara E. Cain, Senior Medical Secretary, Department of Cardiothoracic Surgery, for her efforts in editing English language in some parts in the manuscript. Conflict of interest: none declared. REFERENCES 1 Yendamuri S, Caty L, Pine M, Adem S, Bogner P, Miller A. Outcomes of sarcomatoid carcinoma of the lung: a surveillance, epidemiology, and end results database analysis. Surgery  2012; 152: 397– 402. Google Scholar CrossRef Search ADS PubMed  2 Travis WD, Brambilla E, Nicholson AG, Yatabe Y, Austin JH, Beasley MB et al.   The 2015 World Health Organization classification of lung tumors: impact of genetic, clinical and radiologic advances since the 2004 classification. J Thorac Oncol  2015; 10: 1243– 60. Google Scholar CrossRef Search ADS PubMed  3 Martin LW, Correa AM, Ordonez NG, Roth JA, Swisher SG, Vaporciyan AA et al.   Sarcomatoid carcinoma of the lung: a predictor of poor prognosis. Ann Thorac Surg  2007; 84: 973– 80. Google Scholar CrossRef Search ADS PubMed  4 Nakajima M, Kasai T, Hashimoto H, Iwata Y, Manabe H. Sarcomatoid carcinoma of the lung. Cancer  1999; 86: 608– 16. Google Scholar CrossRef Search ADS PubMed  5 Pelosi G, Gasparini P, Cavazza A, Rossi G, Graziano P, Barbareschi M et al.   Multiparametric molecular characterization of pulmonary sarcomatoid carcinoma reveals a nonrandom amplification of anaplastic lymphoma kinase (ALK) gene. Lung Cancer  2012; 77: 507– 14. Google Scholar CrossRef Search ADS PubMed  6 Fishback NF, Travis WD, Moran CA, Guinee DG, McCarthy WF, Koss MN. Pleomorphic (spindle/giant cell) carcinoma of the lung. Cancer  1994; 73: 2936– 45. Google Scholar CrossRef Search ADS PubMed  7 Chang Y-L, Lee Y-C, Shih J-Y, Wu C-T. Pulmonary pleomorphic (spindle) cell carcinoma: peculiar clinicopathologic manifestations different from ordinary non-small cell carcinoma. Lung Cancer  2001; 34: 91– 7. Google Scholar CrossRef Search ADS PubMed  8 Nappi O, Glasner SD, Swanson PE, Wick MR. Biphasic and monophasic sarcomatoid carcinomas of the lung: a reappraisal of “carcinosarcomas” and “spindle-cell carcinomas”. Am J Clin Pathol  1994; 102: 331– 40. Google Scholar CrossRef Search ADS PubMed  9 Nishida K, Kobayashi Y, Ishikawa Y, Satoh Y, Okumura S, Nishimura H et al.   Sarcomatoid adenocarcinoma of the lung: clinicopathological, immunohistochemical and molecular analyses. Anticancer Res  2001; 22: 3477– 83. 10 Fritz A, Percy C, Jack A, Shanmugaratnam K, Sobin L, Parkin DM et al.   International Classification of Diseases for Oncology . Geneva: World Health Organization, 2000. 11 Koepsell TD, Weiss NS. Epidemiologic Methods: Studying the Occurrence of Illness . UK: Oxford University Press, 2014. Google Scholar CrossRef Search ADS   12 Brambilla E, Travis WD, Colby TV, Corrin B, Shimosato Y. The new World Health Organization classification of lung tumours. Eur Respir J  2001; 18: 1059– 68. Google Scholar CrossRef Search ADS PubMed  13 Rossi G, Cavazza A, Sturm N, Migaldi M, Facciolongo N, Longo L et al.   Pulmonary carcinomas with pleomorphic, sarcomatoid, or sarcomatous elements: a clinicopathologic and immunohistochemical study of 75 cases. Am J Surg Pathol  2003; 27: 311– 24. Google Scholar CrossRef Search ADS PubMed  14 Pelosi G, Sonzogni A, De Pas T, Galetta D, Veronesi G, Spaggiari L et al.   Review article: pulmonary sarcomatoid carcinomas: a practical overview. Int J Surg Pathol  2010; 18: 103– 20. Google Scholar CrossRef Search ADS PubMed  15 Raveglia F, Mezzetti M, Panigalli T, Furia S, Giuliani L, Conforti S et al.   Personal experience in surgical management of pulmonary pleomorphic carcinoma. Ann Thorac Surg  2004; 78: 1742– 7. Google Scholar CrossRef Search ADS PubMed  16 Ro JY, Chen JL, Lee JS, Sahin AA, Ordóñez NG, Ayala AG. Sarcomatoid carcinoma of the lung. Immunohistochemieal and ultrastructural studies of 14 cases. Cancer  1992; 69: 376– 86. Google Scholar CrossRef Search ADS PubMed  17 Hummel P, Cangiarella JF, Cohen J-M, Yang G, Waisman J, Chhieng DC. Transthoracic fine-needle aspiration biopsy of pulmonary spindle cell and mesenchymal lesions. Cancer Cytopathol  2001; 93: 187– 98. Google Scholar CrossRef Search ADS   18 Pelosi G, Fraggetta F, Nappi O, Pastorino U, Maisonneuve P, Pasini F et al.   Pleomorphic carcinomas of the lung show a selective distribution of gene products involved in cell differentiation, cell cycle control, tumor growth, and tumor cell motility: a clinicopathologic and immunohistochemical study of 31 cases. Am J Surg Pathol  2003; 27: 1203– 15. Google Scholar CrossRef Search ADS PubMed  19 Shokralla HA, Rahouma M. Prognostic clinico-pathological features of 99 cases advanced non-small cell lung cancer—Egyptian National Cancer Institute. Adv Lung Cancer  2016; 4: 29. Google Scholar CrossRef Search ADS   20 Pardo J, Aisa G, de Alava E, Sola JJ, Panizo A, Rodríguez-Spiteri N et al.   Primary mixed squamous carcinoma and osteosarcoma (carcinosarcomas) of the lung have a CGH mapping similar to primitive squamous carcinomas and osteosarcomas. Diagn Mol Pathol  2008; 17: 151– 8. Google Scholar CrossRef Search ADS PubMed  21 Dacic S, Finkelstein SD, Sasatomi E, Swalsky PA, Yousem SA. Molecular pathogenesis of pulmonary carcinosarcoma as determined by microdissection-based allelotyping. Am J Surg Pathol  2002; 26: 510– 6. Google Scholar CrossRef Search ADS PubMed  22 Vokes EE, Chu E. Anti-EGFR therapies: clinical experience in colorectal, lung, and head and neck cancers. Oncology (Williston Park)  2006; 20: 15– 25. Google Scholar PubMed  23 Stiles BM, Nasar A, Hussein MK, Ghaly GR, Ahmed MR, Port JL et al.   Routine molecular testing of resected early-stage lung adenocarcinoma with targeted next-generation sequencing demonstrates a high rate of actionable mutations. J Thorac Oncol  2016; 11: S44– 5. Google Scholar CrossRef Search ADS   24 Ouziane I, Boutayeb S, Mrabti H, Lalya I, Rimani M, Errihani H. Sarcomatoid carcinoma of the lung: a model of resistance of chemotherapy. N Am J Med Sci  2014; 6: 342– 5. Google Scholar CrossRef Search ADS PubMed  25 Wright JL, Black PC, Brown GA, Porter MP, Kamat AM, Dinney CP et al.   Differences in survival among patients with sarcomatoid carcinoma, carcinosarcoma and urothelial carcinoma of the bladder. J Urol  2007; 178: 2302– 7. Google Scholar CrossRef Search ADS PubMed  26 Steuer CE, Behera M, Liu Y, Fu C, Gillespie TW, Saba NF et al.   Pulmonary sarcomatoid carcinoma: an analysis of the National Cancer Data Base. Clin Lung Cancer  2017; 18: 286– 92. Google Scholar CrossRef Search ADS PubMed  APPENDIX. CONFERENCE DISCUSSION Dr E. Vallieres(Seattle, WA, USA): You are reporting on a 40-year period of data here. You are telling me that during these 40 years, the pathology definitions of these rare tumours have not evolved and have not changed? Because you and I both know they have, so how accurate is the information you have just presented? Dr B. Lee(New York, NY, USA): It is true, we see it even in non-small-cell lung cancer, how the definitions of terms are so different even this year, compared to 5 years ago. I do think that in the earlier stages, 40 years ago, the groupings were probably a little bit more put into one, or in fact back then perhaps they were not even called specific tumours that we use the terms for today. I think the terminology that we were able to cite in this article is definitely what we are more accustomed to in 2017, while term carcinosarcoma or pulmonary blastoma, these probably were not actually defined tumours 40 years ago. © The Author 2017. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.

Journal

European Journal of Cardio-Thoracic SurgeryOxford University Press

Published: Apr 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

Monthly Plan

  • Read unlimited articles
  • Personalized recommendations
  • No expiration
  • Print 20 pages per month
  • 20% off on PDF purchases
  • Organize your research
  • Get updates on your journals and topic searches

$49/month

Start Free Trial

14-day Free Trial

Best Deal — 39% off

Annual Plan

  • All the features of the Professional Plan, but for 39% off!
  • Billed annually
  • No expiration
  • For the normal price of 10 articles elsewhere, you get one full year of unlimited access to articles.

$588

$360/year

billed annually
Start Free Trial

14-day Free Trial