Abstract
An adenosquamous carcinoma (ASC) of the lung is a relatively rare tumor. In this multi-institutional cohort study, we tested the hypothesis that an ASC exhibits more aggressive clinical behavior as compared to adenocarcinoma (AC) and squamous cell carcinoma (SC).
This retrospective cohort study used a prospective database produced by the Japan National Hospital Organization Study Group for Lung Cancer over a 7-year period (operations from 1997 to 2003, follow-up data until March 2010). During that period, 4668 cases underwent an operation for various types of primary malignant lung tumors. When a sample from a tumor comprised at least 20% each of SC and AC, the case was classified as ASC. Pathologic staging was done according to the seventh edition of the International Union against Cancer (UICC) Tumor Node Matastasis (TNM) classification of malignant tumors.
We identified 114 patients with ASC (2.4%), 2993 with AC (64.2%), and 1369 with SC (29.3%). Kaplan-Meier survival curves for all stage cases, p-stage IA, IB, and IIIA tumors indicated that ASC cases had the least favorable survival. The 5-year survival rates for all stage cases were 23.3% for ASC, 58.0% for AC (p < 0.0001), and 40.8% for SC (p < 0.0001). The 5-year survival rates for p-stage IA were 42.0% for ASC, 81.8% for AC (p = 0.0005), and 63.4% for SC not significant (NS), while those for p-stage IB were 19.3%, 65.3% (p = 0.0024), and 46.8% (NS), respectively, and those for p-stage IIIA were 17.8%, 24.8% (p = 0.0154), and 18.8% (NS), respectively. There was a tendency for greater survival differences between ASC and AC in earlier tumor stages. A step-wise multivariable model demonstrated that sex, age, performance status, histology, tumor size, p-stage, operative method, and neoadjuvant/adjuvant therapy were independent prognostic factors.
ASC of the lung is more aggressive than AC and SC. The decreased survival of patients with ASC as compared with either of those single histology tumors suggests the need for a clinical trial of adjuvant chemotherapy that includes early-stage patients.
INTRODUCTION
Adenosquamous carcinoma (ASC) of the lung is a relatively rare tumor, comprising from 0.3% to 5% of non-small-cell lung cancer cases [1–8]. ASC is a mixed histologic tumor, as defined by the World Health Organization (WHO), as it has components of both adenocarcinoma (AC) and squamous cell carcinoma (SC), with each comprising at least 10% of the tumor [9]. Previously, the Japan Lung Cancer Society recommended a diagnosis of ASC when each component of the tumor comprised at least 20% of the tumor texture [10]. However, that definition was changed in 2003 to agree with that of the WHO [11]. Although several studies have suggested that an ASC of the lung is more aggressive than AC and SC [2–8], a few large series have been reported in medical literature, and the characteristics and prognosis remain poorly defined. In the present multi-institutional cohort study, we tested the hypothesis that ASCs exhibit a clinical behavior distinct from ACs and SCs of the lung, by applying the former more rigorous histologic criteria of the Japan Lung Cancer Society.
MATERIALS AND METHODS
This retrospective cohort study used a prospective database produced by the Japan National Hospital Organization Study Group for Lung Cancer over a 7-year period (operations from 1997 to 2003, follow-up data until March 2010). Data collection and analyses were approved and the need for obtaining informed consent from each patient was waived by an institutional review board. There were 4668 patients who underwent an operation during that period for various types of primary non-small-cell lung cancer. The tumor histological type was determined according to the fifth edition of the general rules for clinical and pathologic records published by the Japan Lung Cancer Society [10]. According to these criteria, when a routine microscopic examination using hematoxylin and eosin staining shows that a sample from a tumor is comprised of at least 20% each of SC and AC, the condition is classified as adenosquamous. Pathologic staging was done according to the seventh edition of the TNM classification of malignant tumors [12].
Statistical analyses were performed using the StatView 5.0 software package (SAS Institute Inc. Cary, NC, USA). Comparisons of the characteristics of each cell type were made using the chi-square test for nominal variables and the Student's t-test for continuous variables. Survival rates were calculated using the Kaplan—Meier method for each carcinoma type and statistical significance was evaluated using a log-rank test for comparisons of overall differences in survival distributions. Cox-proportional hazards analysis was used to identify independent predictors of survival and prognosis. Univariate predictors were considered to be significant with a value of p < 0.05 and entered into a step- wise multivariable model. Statistical significance was assumed for a two-tailed p value less than 0.05.
RESULTS
We identified 114 patients with ASC (2.4%), 2993 with AC (64.2%), 1369 with SC (29.3%), and 192 (4.1%) with large cell carcinoma. We compared the patients with ASC to those with AC and SC. The clinical characteristics based on the three types are shown in Table 1. The mean age of patients with AC was younger than that of those with ASC and SC (65.2, 68.7, and 68.5 years old, respectively, p < 0.0001). AC patients were more likely to be women and SC patients were more likely to be men, as compared to ASC patients (p < 0.0001). In the distribution of performance status, significantly more patients with ASC were classified as grade 1 (29.8%) compared to AC patients (15.0%, p < 0.0001), while there was no difference in performance status between patients with ASC and SC. SC patients had more smoking history than ASC patients and AC patients had less (p < 0.0001). Mean tumor size (mm) was larger in ASC patients (35.0 ± 18.5) than AC patients (28.8 ± 16.6, p < 0.0001), while there was no difference in tumor size between patients with ASC and SC. Significantly, more patients with ASC were presented with T2a or T3 invasion than those with AC (51.7% vs 38.8% and 18.4% vs 9.4%, respectively; p = 0.0002), while there was no difference in the distribution of T factor between patients with ASC and SC. In ASC patients, the reasons for T2a consistent with tumor size in 41 (69.5%), pleural invasion in 17 (28.8%), and tumor location in one (1.7%), while in AC patients, those consistent with tumor size in 702 (60.2%), pleural invasion in 462 (39.6%), and tumor location in two (0.2%). The reason for T3 in ASC patients consistent with tumor size in nine (42.8%), pulmonary metastasis in the same lobe in one (4.8%), and chest wall invasion in 11 (52.4%), while those in AC patients consistent with tumor size in 61 (21.6%), pulmonary metastasis in the same lobe in 91 (32.3%), chest wall invasion in 126 (44.7%), and tumor location in four (1.4%). Significantly, more patients with ASC were classified as N2 (23.7%) than those with AC (15.3%, p = 0.0185), while there was no difference between patients with ASC and SC. Consequently, pathologic stage distribution revealed that significantly more patients with ASC were presented with stage IIIA (24.6%) than those with AC (15.6%, p = 0.0007), while there was no difference between patients with ASC and SC. Also, a pneumonectomy was performed significantly more frequently in patients with ASC (7.9%) than those with AC (2.6%, p = 0.002), while there was no difference between patients with ASC and SC. There were no differences in regard to the proportion of complete resections among the three types. Also, there were no differences in regard to the ratio of patients who received neoadjuvant therapy or adjuvant therapy among the three types.
Characteristics of patients with adenosquamous carcinoma, adenocarcinoma, and squamous cell carcinoma, and comparisons among adenosquamous carcinoma and single histologic carcinoma types
| ASC | AC | p value AC versus ASC | SC | p value SC versus ASC | |
|---|---|---|---|---|---|
| Total no. of cases | 114 | 2993 | 1369 | ||
| Age in years (range) | 68.7 (34–86) | 65.2 (19–93) | <0.0001 | 68.5 (27–91) | NS |
| Sex, male/female (ratio) | 88/26 (3.38:1) | 1591/1402 (1.13:1) | <0.0001 | 1254/115 (10.9:1) | <0.0001 |
| Performance status | |||||
| 0 | 75 (65.8) | 2511 (83.8) | <0.0001 | 952 (69.6) | |
| 1 | 34 (29.8) | 449 (15.0) | 377 (27.5) | ||
| 2 | 4 (3.5) | 27 (0.9) | 34 (2.5) | NS | |
| 3 | 1 (0.9) | 5 (0.2) | 4 (0.3) | ||
| 4 | 0 (0.0) | 1 (0.1) | 2 (0.1) | ||
| Smoking history | |||||
| Yes | 87 (76.3) | 1393 (46.5) | <0.0001 | 1175 (85.8) | 0.0061 |
| No | 27 (23.7) | 1600 (53.5) | 194 (14.2) | ||
| Tumor size (mm) | 35.0 ± 18.5 | 28.8 ± 16.6 | <0.0001 | 37.9 ± 20.5 | NS |
| T factor (%) | |||||
| T0 | 0 (0.0) | 2 (0.1) | 0.0002 | 0 (0.0) | |
| T1a | 14 (12.3) | 744 (25.0) | 154 (11.3) | ||
| T1b | 14 (12.3) | 627 (21.0) | 210 (15.3) | ||
| T2a | 59 (51.7) | 1166 (38.8) | 573 (41.8) | NS | |
| T2b | 5 (4.4) | 96 (3.2) | 120 (8.8) | ||
| T3 | 21 (18.4) | 282 (9.4) | 263 (19.2) | ||
| T4 | 1 (0.9) | 76 (2.5) | 49 (3.6) | ||
| N factor (%) | |||||
| N0 | 71 (62.2) | 2253 (75.2) | 0.0185 | 920 (67.2) | |
| N1 | 15 (13.2) | 259 (8.7) | 230 (16.8) | NS | |
| N2 | 27 (23.7) | 457 (15.3) | 204 (14.9) | ||
| N3 | 1 (0.9) | 24 (0.8) | 15 (1.1) | ||
| Pathologic stage (%) | |||||
| 0 | 0 (0.0) | 2 (0.1) | 0.0007 | 0 (0.0) | |
| IA | 23 (20.2) | 1165 (38.8) | 299 (21.9) | ||
| IB | 30 (26.3) | 809 (26.9) | 375 (27.4) | ||
| IIA | 16 (14.0) | 233 (7.8) | 206 (15.1) | NS | |
| IIB | 12 (10.5) | 177 (5.9) | 173 (12.6) | ||
| IIIA | 28 (24.6) | 466 (15.6) | 263 (19.2) | ||
| IIIB | 1 (0.9) | 38 (1.3) | 21 (1.5) | ||
| IV | 4 (3.5) | 103 (3.4) | 32 (2.3) | ||
| Op. method (%) | |||||
| Pneumonectomy | 9 (7.9) | 79 (2.6) | 0.002 | 137 (10.0) | |
| Lobectomy | 98 (86.0) | 2613 (87.3) | 1110 (81.1) | NS | |
| Segmental/partial | 7 (6.1) | 301 (10.1) | 122 (8.9) | ||
| Resection (%) | |||||
| Complete | 105 (92.1) | 2818 (94.1) | NS | 1260 (92.0) | NS |
| Incomplete | 9 (7.9) | 175 (5.9) | 109 (8.0) | ||
| Neo/adjuvant therapy | |||||
| Yes | 35 (30.7) | 691 (23.1) | NS | 375 (37.4) | NS |
| No | 79 (69.3) | 2302 (76.9) | 994 (72.6) |
| ASC | AC | p value AC versus ASC | SC | p value SC versus ASC | |
|---|---|---|---|---|---|
| Total no. of cases | 114 | 2993 | 1369 | ||
| Age in years (range) | 68.7 (34–86) | 65.2 (19–93) | <0.0001 | 68.5 (27–91) | NS |
| Sex, male/female (ratio) | 88/26 (3.38:1) | 1591/1402 (1.13:1) | <0.0001 | 1254/115 (10.9:1) | <0.0001 |
| Performance status | |||||
| 0 | 75 (65.8) | 2511 (83.8) | <0.0001 | 952 (69.6) | |
| 1 | 34 (29.8) | 449 (15.0) | 377 (27.5) | ||
| 2 | 4 (3.5) | 27 (0.9) | 34 (2.5) | NS | |
| 3 | 1 (0.9) | 5 (0.2) | 4 (0.3) | ||
| 4 | 0 (0.0) | 1 (0.1) | 2 (0.1) | ||
| Smoking history | |||||
| Yes | 87 (76.3) | 1393 (46.5) | <0.0001 | 1175 (85.8) | 0.0061 |
| No | 27 (23.7) | 1600 (53.5) | 194 (14.2) | ||
| Tumor size (mm) | 35.0 ± 18.5 | 28.8 ± 16.6 | <0.0001 | 37.9 ± 20.5 | NS |
| T factor (%) | |||||
| T0 | 0 (0.0) | 2 (0.1) | 0.0002 | 0 (0.0) | |
| T1a | 14 (12.3) | 744 (25.0) | 154 (11.3) | ||
| T1b | 14 (12.3) | 627 (21.0) | 210 (15.3) | ||
| T2a | 59 (51.7) | 1166 (38.8) | 573 (41.8) | NS | |
| T2b | 5 (4.4) | 96 (3.2) | 120 (8.8) | ||
| T3 | 21 (18.4) | 282 (9.4) | 263 (19.2) | ||
| T4 | 1 (0.9) | 76 (2.5) | 49 (3.6) | ||
| N factor (%) | |||||
| N0 | 71 (62.2) | 2253 (75.2) | 0.0185 | 920 (67.2) | |
| N1 | 15 (13.2) | 259 (8.7) | 230 (16.8) | NS | |
| N2 | 27 (23.7) | 457 (15.3) | 204 (14.9) | ||
| N3 | 1 (0.9) | 24 (0.8) | 15 (1.1) | ||
| Pathologic stage (%) | |||||
| 0 | 0 (0.0) | 2 (0.1) | 0.0007 | 0 (0.0) | |
| IA | 23 (20.2) | 1165 (38.8) | 299 (21.9) | ||
| IB | 30 (26.3) | 809 (26.9) | 375 (27.4) | ||
| IIA | 16 (14.0) | 233 (7.8) | 206 (15.1) | NS | |
| IIB | 12 (10.5) | 177 (5.9) | 173 (12.6) | ||
| IIIA | 28 (24.6) | 466 (15.6) | 263 (19.2) | ||
| IIIB | 1 (0.9) | 38 (1.3) | 21 (1.5) | ||
| IV | 4 (3.5) | 103 (3.4) | 32 (2.3) | ||
| Op. method (%) | |||||
| Pneumonectomy | 9 (7.9) | 79 (2.6) | 0.002 | 137 (10.0) | |
| Lobectomy | 98 (86.0) | 2613 (87.3) | 1110 (81.1) | NS | |
| Segmental/partial | 7 (6.1) | 301 (10.1) | 122 (8.9) | ||
| Resection (%) | |||||
| Complete | 105 (92.1) | 2818 (94.1) | NS | 1260 (92.0) | NS |
| Incomplete | 9 (7.9) | 175 (5.9) | 109 (8.0) | ||
| Neo/adjuvant therapy | |||||
| Yes | 35 (30.7) | 691 (23.1) | NS | 375 (37.4) | NS |
| No | 79 (69.3) | 2302 (76.9) | 994 (72.6) |
ASC: adenosquamous carcinoma; AC: adenocarcinoma; SC: squamous cell carcinoma; segmental/partial: segmental or partial resection; lob: lobectomy; pneumo: pneumonectomy; Neo/adjuvant: neoadjuvant or adjuvant therapy.
Characteristics of patients with adenosquamous carcinoma, adenocarcinoma, and squamous cell carcinoma, and comparisons among adenosquamous carcinoma and single histologic carcinoma types
| ASC | AC | p value AC versus ASC | SC | p value SC versus ASC | |
|---|---|---|---|---|---|
| Total no. of cases | 114 | 2993 | 1369 | ||
| Age in years (range) | 68.7 (34–86) | 65.2 (19–93) | <0.0001 | 68.5 (27–91) | NS |
| Sex, male/female (ratio) | 88/26 (3.38:1) | 1591/1402 (1.13:1) | <0.0001 | 1254/115 (10.9:1) | <0.0001 |
| Performance status | |||||
| 0 | 75 (65.8) | 2511 (83.8) | <0.0001 | 952 (69.6) | |
| 1 | 34 (29.8) | 449 (15.0) | 377 (27.5) | ||
| 2 | 4 (3.5) | 27 (0.9) | 34 (2.5) | NS | |
| 3 | 1 (0.9) | 5 (0.2) | 4 (0.3) | ||
| 4 | 0 (0.0) | 1 (0.1) | 2 (0.1) | ||
| Smoking history | |||||
| Yes | 87 (76.3) | 1393 (46.5) | <0.0001 | 1175 (85.8) | 0.0061 |
| No | 27 (23.7) | 1600 (53.5) | 194 (14.2) | ||
| Tumor size (mm) | 35.0 ± 18.5 | 28.8 ± 16.6 | <0.0001 | 37.9 ± 20.5 | NS |
| T factor (%) | |||||
| T0 | 0 (0.0) | 2 (0.1) | 0.0002 | 0 (0.0) | |
| T1a | 14 (12.3) | 744 (25.0) | 154 (11.3) | ||
| T1b | 14 (12.3) | 627 (21.0) | 210 (15.3) | ||
| T2a | 59 (51.7) | 1166 (38.8) | 573 (41.8) | NS | |
| T2b | 5 (4.4) | 96 (3.2) | 120 (8.8) | ||
| T3 | 21 (18.4) | 282 (9.4) | 263 (19.2) | ||
| T4 | 1 (0.9) | 76 (2.5) | 49 (3.6) | ||
| N factor (%) | |||||
| N0 | 71 (62.2) | 2253 (75.2) | 0.0185 | 920 (67.2) | |
| N1 | 15 (13.2) | 259 (8.7) | 230 (16.8) | NS | |
| N2 | 27 (23.7) | 457 (15.3) | 204 (14.9) | ||
| N3 | 1 (0.9) | 24 (0.8) | 15 (1.1) | ||
| Pathologic stage (%) | |||||
| 0 | 0 (0.0) | 2 (0.1) | 0.0007 | 0 (0.0) | |
| IA | 23 (20.2) | 1165 (38.8) | 299 (21.9) | ||
| IB | 30 (26.3) | 809 (26.9) | 375 (27.4) | ||
| IIA | 16 (14.0) | 233 (7.8) | 206 (15.1) | NS | |
| IIB | 12 (10.5) | 177 (5.9) | 173 (12.6) | ||
| IIIA | 28 (24.6) | 466 (15.6) | 263 (19.2) | ||
| IIIB | 1 (0.9) | 38 (1.3) | 21 (1.5) | ||
| IV | 4 (3.5) | 103 (3.4) | 32 (2.3) | ||
| Op. method (%) | |||||
| Pneumonectomy | 9 (7.9) | 79 (2.6) | 0.002 | 137 (10.0) | |
| Lobectomy | 98 (86.0) | 2613 (87.3) | 1110 (81.1) | NS | |
| Segmental/partial | 7 (6.1) | 301 (10.1) | 122 (8.9) | ||
| Resection (%) | |||||
| Complete | 105 (92.1) | 2818 (94.1) | NS | 1260 (92.0) | NS |
| Incomplete | 9 (7.9) | 175 (5.9) | 109 (8.0) | ||
| Neo/adjuvant therapy | |||||
| Yes | 35 (30.7) | 691 (23.1) | NS | 375 (37.4) | NS |
| No | 79 (69.3) | 2302 (76.9) | 994 (72.6) |
| ASC | AC | p value AC versus ASC | SC | p value SC versus ASC | |
|---|---|---|---|---|---|
| Total no. of cases | 114 | 2993 | 1369 | ||
| Age in years (range) | 68.7 (34–86) | 65.2 (19–93) | <0.0001 | 68.5 (27–91) | NS |
| Sex, male/female (ratio) | 88/26 (3.38:1) | 1591/1402 (1.13:1) | <0.0001 | 1254/115 (10.9:1) | <0.0001 |
| Performance status | |||||
| 0 | 75 (65.8) | 2511 (83.8) | <0.0001 | 952 (69.6) | |
| 1 | 34 (29.8) | 449 (15.0) | 377 (27.5) | ||
| 2 | 4 (3.5) | 27 (0.9) | 34 (2.5) | NS | |
| 3 | 1 (0.9) | 5 (0.2) | 4 (0.3) | ||
| 4 | 0 (0.0) | 1 (0.1) | 2 (0.1) | ||
| Smoking history | |||||
| Yes | 87 (76.3) | 1393 (46.5) | <0.0001 | 1175 (85.8) | 0.0061 |
| No | 27 (23.7) | 1600 (53.5) | 194 (14.2) | ||
| Tumor size (mm) | 35.0 ± 18.5 | 28.8 ± 16.6 | <0.0001 | 37.9 ± 20.5 | NS |
| T factor (%) | |||||
| T0 | 0 (0.0) | 2 (0.1) | 0.0002 | 0 (0.0) | |
| T1a | 14 (12.3) | 744 (25.0) | 154 (11.3) | ||
| T1b | 14 (12.3) | 627 (21.0) | 210 (15.3) | ||
| T2a | 59 (51.7) | 1166 (38.8) | 573 (41.8) | NS | |
| T2b | 5 (4.4) | 96 (3.2) | 120 (8.8) | ||
| T3 | 21 (18.4) | 282 (9.4) | 263 (19.2) | ||
| T4 | 1 (0.9) | 76 (2.5) | 49 (3.6) | ||
| N factor (%) | |||||
| N0 | 71 (62.2) | 2253 (75.2) | 0.0185 | 920 (67.2) | |
| N1 | 15 (13.2) | 259 (8.7) | 230 (16.8) | NS | |
| N2 | 27 (23.7) | 457 (15.3) | 204 (14.9) | ||
| N3 | 1 (0.9) | 24 (0.8) | 15 (1.1) | ||
| Pathologic stage (%) | |||||
| 0 | 0 (0.0) | 2 (0.1) | 0.0007 | 0 (0.0) | |
| IA | 23 (20.2) | 1165 (38.8) | 299 (21.9) | ||
| IB | 30 (26.3) | 809 (26.9) | 375 (27.4) | ||
| IIA | 16 (14.0) | 233 (7.8) | 206 (15.1) | NS | |
| IIB | 12 (10.5) | 177 (5.9) | 173 (12.6) | ||
| IIIA | 28 (24.6) | 466 (15.6) | 263 (19.2) | ||
| IIIB | 1 (0.9) | 38 (1.3) | 21 (1.5) | ||
| IV | 4 (3.5) | 103 (3.4) | 32 (2.3) | ||
| Op. method (%) | |||||
| Pneumonectomy | 9 (7.9) | 79 (2.6) | 0.002 | 137 (10.0) | |
| Lobectomy | 98 (86.0) | 2613 (87.3) | 1110 (81.1) | NS | |
| Segmental/partial | 7 (6.1) | 301 (10.1) | 122 (8.9) | ||
| Resection (%) | |||||
| Complete | 105 (92.1) | 2818 (94.1) | NS | 1260 (92.0) | NS |
| Incomplete | 9 (7.9) | 175 (5.9) | 109 (8.0) | ||
| Neo/adjuvant therapy | |||||
| Yes | 35 (30.7) | 691 (23.1) | NS | 375 (37.4) | NS |
| No | 79 (69.3) | 2302 (76.9) | 994 (72.6) |
ASC: adenosquamous carcinoma; AC: adenocarcinoma; SC: squamous cell carcinoma; segmental/partial: segmental or partial resection; lob: lobectomy; pneumo: pneumonectomy; Neo/adjuvant: neoadjuvant or adjuvant therapy.
Next, we analyzed cumulative overall survival rates according to tumor type based on pathologic stage. Including all stage cases, Kaplan—Meier survival curves indicated that ASC cases had the least favorable survival (Fig. 1), as the 5-year survival rates were 23.3% for ASC, 58.0% for AC, and 40.8% for SC, which showed statistical differences among the three types (p < 0.0001 for ASC to AC and SC to AC, p = 0.0009 for ASC to SC). We selected stage IA, IB, and IIIA for analyses, because the numbers of patients with ASC in other stages was too small for valid comparisons. Survival curves for p-stage IA tumors indicated that ASC cases had the least favorable survival (Fig. 2), as the 5-year survival rates were 42.0% for ASC, 81.8% for AC, and 63.4% for SC, which showed statistical differences between ASC and AC patients (p = 0.0005), and between AC and SC (p < 0.0001). Furthermore, the survival curves for p-stage IB cases also indicated that ASC had the least favorable survival (Fig. 3), as those 5-year survival rates were 19.3%, 65.3%, and 46.8%, respectively, with a significant difference observed between ASC and AC (p = 0.0024), and between AC and SC (p < 0.0001). Finally, the survival curves for p-stage IIIA tumors also indicated that ASC had the least favorable survival (Fig. 4), with 5-year survival rates of 17.8%, 24.8%, and 18.8%, respectively, and a significant difference between ASC and AC (p = 0.0154).
Survival curves following resection for patients with all stage cases of adenocarcinoma (AC), squamous cell carcinoma (SC) and adenosquamous carcinoma (ASC).
Survival curves following resection for patients with pathologic stage IA adenocarcinoma (AC), squamous cell carcinoma (SC) and adenosquamous carcinoma (ASC).
Survival curves following resection for patients with pathologic stage IB adenocarcinoma (AC), squamous cell carcinoma (SC) and adenosquamous carcinoma (ASC).
Survival curves following resection for patients with pathologic stage IIIA adenocarcinoma (AC), squamous cell carcinoma (SC) and adenosquamous carcinoma (ASC).
A step-wise multivariable model demonstrated that sex, age, performance status, histology, tumor size, p-stage, operative method, and neoadjuvant/adjuvant therapy were independent prognostic factors (Table 2).
Multivariate analysis of prognostic factors for survival
| Variable | HR | CI | p Value |
|---|---|---|---|
| Sex (female vs male) | 0.593 | 0.514–0.684 | <0.0001 |
| Age (<64 vs >65) | 0.629 | 0.555–0.713 | <0.0001 |
| PS (0, 1 vs 2–4) | 0.444 | 0.326–0.606 | <0.0001 |
| Histology (AC vs SC vs ASC) | |||
| Adenocarcinoma | 0.481 | 0.361–0.641 | <0.0001 |
| Squamous cell carcinoma | 0.515 | 0.386–0.688 | <0.0001 |
| Tumor size (≤30 vs ≥31 mm) | 0.603 | 0.531–0.683 | <0.0001 |
| p-Stage (0, I vs II vs III vs IV) | |||
| p-Stage 0, IA, IB | 0.225 | 0.175–0.288 | <0.0001 |
| p-Stage IIA, IIB | 0.464 | 0.359–0.599 | <0.0001 |
| p-Stage IIIA, IIIB | 0.893 | 0.702–1.136 | 0.3572 |
| Op. method (seg/part vs lob vs peumo) | |||
| Segmental/partial | 1.075 | 0.829–1.393 | 0.5865 |
| Lobectomy | 0.750 | 0.613–0.917 | 0.0051 |
| Neo/adjuvant (no vs yes) | 0.859 | 0.756–0.976 | 0.0195 |
| Variable | HR | CI | p Value |
|---|---|---|---|
| Sex (female vs male) | 0.593 | 0.514–0.684 | <0.0001 |
| Age (<64 vs >65) | 0.629 | 0.555–0.713 | <0.0001 |
| PS (0, 1 vs 2–4) | 0.444 | 0.326–0.606 | <0.0001 |
| Histology (AC vs SC vs ASC) | |||
| Adenocarcinoma | 0.481 | 0.361–0.641 | <0.0001 |
| Squamous cell carcinoma | 0.515 | 0.386–0.688 | <0.0001 |
| Tumor size (≤30 vs ≥31 mm) | 0.603 | 0.531–0.683 | <0.0001 |
| p-Stage (0, I vs II vs III vs IV) | |||
| p-Stage 0, IA, IB | 0.225 | 0.175–0.288 | <0.0001 |
| p-Stage IIA, IIB | 0.464 | 0.359–0.599 | <0.0001 |
| p-Stage IIIA, IIIB | 0.893 | 0.702–1.136 | 0.3572 |
| Op. method (seg/part vs lob vs peumo) | |||
| Segmental/partial | 1.075 | 0.829–1.393 | 0.5865 |
| Lobectomy | 0.750 | 0.613–0.917 | 0.0051 |
| Neo/adjuvant (no vs yes) | 0.859 | 0.756–0.976 | 0.0195 |
HR: hazard ratio; CI: confidence interval; PS: performance status; seg/part: segmental or partial resection; lob: lobectomy; pneumo: pneumonectomy; Neo/adjuvant: neoadjuvant or adjuvant therapy.
Multivariate analysis of prognostic factors for survival
| Variable | HR | CI | p Value |
|---|---|---|---|
| Sex (female vs male) | 0.593 | 0.514–0.684 | <0.0001 |
| Age (<64 vs >65) | 0.629 | 0.555–0.713 | <0.0001 |
| PS (0, 1 vs 2–4) | 0.444 | 0.326–0.606 | <0.0001 |
| Histology (AC vs SC vs ASC) | |||
| Adenocarcinoma | 0.481 | 0.361–0.641 | <0.0001 |
| Squamous cell carcinoma | 0.515 | 0.386–0.688 | <0.0001 |
| Tumor size (≤30 vs ≥31 mm) | 0.603 | 0.531–0.683 | <0.0001 |
| p-Stage (0, I vs II vs III vs IV) | |||
| p-Stage 0, IA, IB | 0.225 | 0.175–0.288 | <0.0001 |
| p-Stage IIA, IIB | 0.464 | 0.359–0.599 | <0.0001 |
| p-Stage IIIA, IIIB | 0.893 | 0.702–1.136 | 0.3572 |
| Op. method (seg/part vs lob vs peumo) | |||
| Segmental/partial | 1.075 | 0.829–1.393 | 0.5865 |
| Lobectomy | 0.750 | 0.613–0.917 | 0.0051 |
| Neo/adjuvant (no vs yes) | 0.859 | 0.756–0.976 | 0.0195 |
| Variable | HR | CI | p Value |
|---|---|---|---|
| Sex (female vs male) | 0.593 | 0.514–0.684 | <0.0001 |
| Age (<64 vs >65) | 0.629 | 0.555–0.713 | <0.0001 |
| PS (0, 1 vs 2–4) | 0.444 | 0.326–0.606 | <0.0001 |
| Histology (AC vs SC vs ASC) | |||
| Adenocarcinoma | 0.481 | 0.361–0.641 | <0.0001 |
| Squamous cell carcinoma | 0.515 | 0.386–0.688 | <0.0001 |
| Tumor size (≤30 vs ≥31 mm) | 0.603 | 0.531–0.683 | <0.0001 |
| p-Stage (0, I vs II vs III vs IV) | |||
| p-Stage 0, IA, IB | 0.225 | 0.175–0.288 | <0.0001 |
| p-Stage IIA, IIB | 0.464 | 0.359–0.599 | <0.0001 |
| p-Stage IIIA, IIIB | 0.893 | 0.702–1.136 | 0.3572 |
| Op. method (seg/part vs lob vs peumo) | |||
| Segmental/partial | 1.075 | 0.829–1.393 | 0.5865 |
| Lobectomy | 0.750 | 0.613–0.917 | 0.0051 |
| Neo/adjuvant (no vs yes) | 0.859 | 0.756–0.976 | 0.0195 |
HR: hazard ratio; CI: confidence interval; PS: performance status; seg/part: segmental or partial resection; lob: lobectomy; pneumo: pneumonectomy; Neo/adjuvant: neoadjuvant or adjuvant therapy.
DISCUSSION
ASC of the lung is a relatively uncommon subtype of lung cancer and prior reports have suggested that it represents 0.3–5% of lung cancer cases [1–8]. We found ASC in 2.4% of 4668 patients with non-small-cell lung cancer who underwent surgery, using the rigorous histologic criteria formally presented by the Japan Lung Cancer Society. Although a few studies have reported nonsignificant differences in regard to postoperative survival between ASC and other histologic types [1,13], many more have indicated that ASC has a more aggressive behavior and worse prognosis than other histologic types of non-small-cell lung cancer [2–8]. However, a few large series have been presented in medical literature, thus the characteristics and prognosis of this entity remain poorly defined.
In regard to the more aggressive behavior and worse prognosis of ASC, Takamori and colleagues evaluated 2160 patients who underwent resection for primary lung cancer, and found 56 patients (2.6%) with ASC. The survival curves indicated that the outcome of ASC was worse than that of AC and SC, particularly in stage I and II cases [3]. Shimizu and colleagues [4] examined 1284 patients who underwent resection for primary lung cancer, including 44 cases (3.4%) of ASC. The 5-year survival rate for the ASC cases was 18.5%, which was significantly worse than that for AC (39.2%) and SC (38.7%) cases. Nakagawa and colleagues reviewed outcomes in 30 cases of resection for ASC. The cumulative survival rate for patients with ASC and pathologic stages IA—IIB in their study population was similar to that of patients with stage IIIA AC or SC [6]. Gawrychowski and colleagues [7] examined data for 96 patients with ASC, and found that the cumulative postoperative survival rate for patients with ASC at 5 years was 25.4% and after 10 years was 19.2%, as compared with 42.5% and 39.1%, respectively, for a contemporaneous cohort of patients with AC. In addition, their analysis of survival according to pathologic stage revealed that stage IB patients treated surgically for ASC showed significantly worse outcomes than patients who underwent surgery for AC, as the 5-year survival rate was 31.8% for ASC in comparison with 56.3% for AC. On the other hand, survival rates did not differ significantly between ASC and AC patients classified as stage IA, IIA, or IIB. Cooke and colleagues [8] examined a national database of patients surgically treated in the United States and identified 872 diagnosed with ASC. Their study cohort was limited to patients who underwent a lobectomy for early-stage and node-negative disease, and those with ASC represented 4.1% of the 21361 patients examined. Overall survival was significantly reduced for both ASC and SC cases, as compared with AC. The 5-year survival rate for patients with stage I tumors was 62.0% for ASC, 69.2% for SC, and 73.2% for AC (p < 0.0001). Although cases of ASC stage II showed a trend toward worse survival, there were no statistical differences among the three groups.
Similar to these studies, our analysis of survival found that ASC tumors resulted in a worse prognosis than AC and SC (Figs. 1–4). Including all stage cases, survival curves indicated that ASC cases had significantly the least favorable survival compared to other cell types. According to the tumor type based on pathologic stage, ASC cases in stage IA, IB, and IIIA revealed significantly worse survival compared to AC, while in comparison to SC, ASC showed a trend toward worse survival, though it was not significant. We found a tendency for larger differences in survival between ASC and single histology AC cases in earlier stages.
The reason why ASC tumors have more aggressive behavior than AC and SC tumors remains unresolved. As for the histogenesis of ASC, there are many possibilities, including AC with squamous metaplasia, collision tumor, and bipotential undifferentiated cell origin [3]. Niho and colleagues [14] analyzed the clonality of ASC and reported that squamous cell and AC components showed identical monoclonal patterns, which suggested that both originated from the same cell type. Kanazawa and colleagues [15] also suggested monoclonal transition from SC to AC in ASC. These findings support the hypothesis that ASC originates from a monoclonal expansion of a single mutant progenitor cell clone, which is different from AC and SC. Several studies have reported distinctive characteristics of ASC found in clinicopathologic examinations. Ruffini and colleagues [5] found that among ASC cases, high cell grading, advanced stage, and intratu-moral perineural invasion were significantly more evident than in the single histology population. Cakir and colleagues [16] also found that advanced stage, vascular invasion, and parietal pleural involvement was significantly more evident than in the single histology group. Furthermore, Bastide and colleagues performed comparative transcriptome analysis and suggested that ASC tumors are more complex than simple mixes of AC and SC components. They proposed that neuroendocrine differentiation and extracellular signalregulated kinase (ERK) proliferation pathways may be preferentially deregulated in ASC as compared to AC and SC, respectively, which could explain the high clinical aggressiveness of ASC [17].
We could not analyze details of the pathologic characteristics of ASC, such as cell grading, vascular invasion, or ratio of the component of AC and SC. Several authors mentioned that these factors are supposed to affect on survival, so further study is needed.
In conclusion, ASC of the lung is more aggressive than AC and SC. The decreased survival of patients with ASC as compared with the single histology AC and SC suggests the need for a clinical trial of adjuvant chemotherapy, including early-stage patients.
Conflict of interest: none declared.
