|Home | About | Journals | Submit | Contact Us | Français|
Purpose: The aim of this study was to compare the clinicopathologic prognostic factors between patients who underwent lung resection for adenocarcinoma (AD) and those with squamous cell carcinoma (SQ).
Methods: A database of patients with lung AD or SQ who underwent surgery with curative intent in our department from January 2008 to December 2014 was reviewed. Associations between various clinicopathologic factors, postsurgical recurrence-free survival (RFS), and overall survival (OS) were analyzed to find significant prognostic factors.
Results: A total of 537 lung cancer patients (AD, 434; SQ, 103) were included in this study. Although RFS was similar in patients with AD and SQ, OS was significantly poorer in those with SQ. Multivariate analysis in patients with AD revealed that age (≥69 vs. <69), lymphatic invasion, and histologic pleural invasion (p0 vs. p1–3) were associated with RFS, while gender and pleural invasion were associated with OS. In SQ, however, smoking, clinical stage, and pulmonary metastasis were associated with RFS in the multivariate analysis.
Conclusion: Since significant postoperative prognostic factors are quite different between lung AD and SQ, these two histologic types should be differently analyzed in a clinical study.
Lung adenocarcinoma (AD) and squamous cell carcinoma (SQ) are the two major histologic types of non-small-cell lung cancer (NSCLC), and have been considered to be almost similar in both postsurgical prognosis and chemotherapeutic responses for a long period of time. However, recent multiple studies on the postsurgical outcomes of these two histologic types indicate that overall survival (OS) is significantly better in AD when compared with SQ.1,2) In addition, differences in certain chemotherapeutic responses have been reported; for instance, a chemotherapeutic regimen of cisplatin and gemcitabine was more effective for SQ, whereas, for AD, a regimen involving cisplatin and pemetrexed was found to be more effective.3) In moleculartargeted therapy, the epidermal growth factor receptor tyrosine kinase inhibitors and anaplastic lymphoma tyrosine kinase inhibitors are known to be effective for the treatment of AD carrying activating mutations, but these mutations are rarely seen in SQ.4–6) Epidemiologically, an increase in the prevalence of AD, particularly in the non-smoking population, and a decrease in SQ prevalence has been observed worldwide.7,8) Therefore, in the present study, we analyzed the prognostic factors associated with the recently increasing ADs when compared with the decreasing SQs in the lung.
This retrospective study was approved by the Institutional Review Board of the St. Marianna University School of Medicine. A clinicopathologic database of patients with AD or SQ of the lung who underwent surgery with curative intent in our department from January 2008 to December 2014 was reviewed. Records were obtained and the surgical methods used for treatment were classified into the following four types: sublobar resection (wedge resection or segmentectomy), lobectomy (including bilobectomy), pneumonectomy, and combined lung resection with adjacent organs. Bronchial or tracheal resections without lung resection, patients with multiple lung cancers and different histologic types, and those who received preoperative induction therapy were excluded. Disease stages were determined according to the international staging criteria for lung cancer, published by the American Joint Committee on Cancer/Union for International Cancer Control/International Association for the Study of Lung Cancer in 2009.9)
Categorical variables for contingency tables were tested using the chi-square test, whereas continuous variables in multiple groups were tested using the Kruskal–Wallis test. Estimated survival rates after surgery were calculated using the Kaplan–Meier method, and survival differences between patient groups were tested using the log-rank test. The cox proportional hazard model was used for multivariate analysis of postsurgical survival. P <0.05 was considered as significant.
A total of 537 patients (434 ADs and 103 SQs) comprising 311 males and 226 females with a mean age of 69 ± 9 years (range, 22–89 years) were included in this study. The mean postoperative follow-up period was 29 ± 21 months (range, 1–82 months). Table 1 demonstrates the differences in clinicopathologic factors between AD and SQ patients. Significant differences were found in sex distribution, smoking habit, tumor size, maximum standardized uptake value in positron emission tomography (PET/SUVmax), clinical stage, lymphatic invasion, and venous invasion.
Although postsurgical 5-year recurrence-free survival (RFS) was similar in patients with AD (78.9%) and SQ (74.5%; P = 0.2126), the 5-year OS was significantly better in those with AD when compared with SQ (88.9% vs. 68.4%; P = 0.0002), suggesting that deaths from diseases other than lung cancer were frequent in patients with SQ. Similar results were observed in AD (n = 268) and SQ (n = 41) patients with clinical stage IA tumors, wherein the 5-year RFS was 87.8% and 88.3% (P = 0.8500), and the 5-year OS was 92.3% and 60.5% (P = 0.0008) in the AD and SQ patients, respectively.
Various clinicopathologic factors were significantly associated with RFS and OS in patients with AD in the univariate analysis (Table 2). In the multivariate analysis including all the significant factors found in the univariate analysis, age (≥69 vs. <69) and lymphatic invasion were found to be associated with RFS, sex was found to be associated with OS, and histologic pleural invasion (p0 vs. p1–3) was observed to be associated with both RFS and OS (Table 3). In contrast, fewer factors were significantly associated with RFS and OS in patients with SQ in the univariate analysis (Table 4). Smoking, clinical stage, pulmonary metastasis, and pathologic stage were associated with RFS and pulmonary metastasis alone was significantly associated with OS. Finally, smoking, clinical stage, and pulmonary metastasis were associated with RFS in the multivariate analysis (Table 5). In this multivariate analysis, pathologic stage was excluded from the analysis because of multicollinearity between pathologic stage and clinical stage.
In the present study, remarkable clinicopathologic differences were observed between the AD and SQ patients. The majority (90%) of patients with SQ were males and are thought to possess sex-related risk factors for OS because it is well known that life expectancy is shorter in males than in females in most countries. The prognosis of NSCLC has been found to be significantly poor in males than in females in several Japanese nationwide surveillances regarding postsurgical OS in lung cancer, performed every 5 years by the Japanese Joint Committee for Lung Cancer Registration (JJCLCR).10) Thus, one of the reasons for the poor prognosis in SQ patients may be attributed to the higher percentage of males in this study. Furthermore, the development of SQ is closely associated with tobacco smoking. The present study demonstrated that 95% of patients with SQ were current or past smokers. The smoking habit is associated with several life-threatening diseases such as lung emphysema, pneumonia, ischemic heart diseases, brain infarction, and several malignant tumors including lung cancer,11) which may worsen the OS in patients with SQ when compared to those with AD. Moreover, larger tumor size, higher SUVmax at the primary lesion, and increased lymphatic and venous invasions in patients with SQ demonstrates the typical local invasive nature of SQ when compared with AD. The RFS did not differ between the AD and SQ patients, when limited to those with c-stage IA, which indicates that the biologic malignant grades of these two histologic types might be comparable as a whole. Although many significant prognostic factors in AD were found in the univariate analysis in both RFS and OS, fewer prognostic factors were found in SQ. Some of these findings are similar to those found in a JJCLCR report, wherein females with pathologic stage I/II AD had a significantly better prognosis than males, but there was no significant prognostic difference between non-AD male and female patients with pathologic stage I/II disease.10) In our present study, male sex and higher age were worse prognostic factors for RFS in AD patients in the univariate analysis, suggesting these factors might be associated with biologic malignant grade of the tumor. Similar results in NSCLC were reported from other institutions.12,13)
Various factors such as age, sex, smoking habit, ethnicity, tumor size, SUVmax, visceral pleural invasion, lymphatic invasion, and venous invasion have been reported to predict survival in surgical cases of NSCLC.14–25) However, only a few reports have analyzed AD and SQ separately. In the present study, the most powerful prognostic factors for RFS and OS in AD were histologic pleural invasion. In our previous studies, we had reported that SUVmax 3.0 in the primary lesion was useful to identify node-negative patients25) and to select patients suitable for limited lung resection in NSCLC.26) However, SUVmax was only marginally significant (P = 0.0516) for RFS in AD in multivariate analysis in the present study. Multiple studies have shown that SUVmax is a significant prognostic factor in NSCLC.24,27,28) However, one study revealed a statistically significant correlation between SUVmax and prognosis in patients with AD, but not SQ,29) and also reported that other clinicopathologic factors did not have a role in predicting OS in patients with SQ. In another study, venous invasion was significantly associated with RFS in patients with AD but not SQ.22) Interestingly, the results of the present study are in accordance with the findings of these two studies. In SQ, multivariate analysis revealed that smoking, clinical stage, and pulmonary metastasis were significant prognostic factors for RFS. However, these results should be cautiously interpreted because extreme imbalance existed between the compared groups in non-smoker and pulmonary metastasis due to small number of patients.
In conclusion, higher percentages of smokers and male patients with lung SQ were speculated to have shorter OS when compared to AD. Various significant prognostic factors were found in patients with AD in univariate analysis, whereas fewer factors were predictive in SQ. Since significant postoperative prognostic factors are quite different between lung AD and SQ, these two histologic types should be differently analyzed in a clinical study.
There are no conflicts of interest to be declared.