Clinical characteristics and prediction model of long-term survival of patients with stage III non-small cell lung cancer.

BACKGROUND: The patients with stage III non-small cell lung cancer own a poor prognosis. We aimed to study the clinical characteristics of the patients with stage III non-small cell lung cancer and more than 5 years overall survival and establish a prognosis prediction model.
METHODS: A total of 5792 patients were separated from the Surveillance, Epidemiology, and End Results database between 2011 and 2015. Cox regression was performed to select the predictors of overall survival. The validation of the nomogram was implemented by using the concordance index, calibration curves. Kaplan-Meier curves were used to illustrate and compare the overall survival of patients in different clinical characteristics.
RESULTS: Multivariate analyses indicated factors such as age, sex, site, histology, grade, stage T, and N, surgical treatment were associated with prognosis. In the nomogram, we could predict the probability of overall survival for patients. The concordance index of the novel nomogram was 0.751. The calibration curves also showed good consistency in the probability of 5-year overall survival. The Kaplan-Meier curves showed that overall survival in the different age, sex, race, site, histology, grade, stage T and N, surgical treatment was accurately differentiated with a significantly statistical difference.
CONCLUSIONS: Patients with a highly differentiated adenocarcinoma and early stages of T and N who are less than 70 years of age and have an opportunity for surgery to undergo surgery have a higher five-year survival rate in patients with stage III non-small cell lung cancer. 2021 Translational Cancer Research. All rights reserved.

Introduction

Lung cancer is the most common cancer and the leading cause of cancer death in the world, which has the highest morbidity and mortality of malignant tumors (1). The number of deaths from lung cancer is larger than the sum of the breast, colon, and prostate cancer (2). In the pathological type of lung cancer, non-small cell lung cancer accounts for 80–85% and 30% of non-small cell lung cancer patients have already arrived at stages III at diagnosis with an unsatisfactory prognosis (3,4). At present, the research on the prognosis of stage III non-small cell lung cancer patients is mainly focused on its treatment, such as surgical method, radiotherapy and chemotherapy, immunotherapy, and so on. However, the study that focuses on the clinical characteristics of the patients with stage III non-small cell lung cancer which may influence the prognosis is relatively less. The purpose of our study is to study the clinical characteristics of the patients with stage III non-small cell lung cancer patients with more than 5 years of overall survival and establish a prognosis prediction model (5-7). We present the following article in accordance with the TRIPOD reporting checklist (available at: http://dx.doi.org/10.21037/tcr-20-3173).

Methods

Study population

SEER 18 population-based cancer Registries (1975 to 2017 dataset) were chosen to select patients, which is maintained by the National Cancer Institute and covers nearly 28% of the population in the United States. The inclusion criteria in our study were that (I) patients with stage III non-small cell lung cancer were diagnosed between 2010 and 2015; (II) patients aged 18 years or older who were diagnosed; (III) the information of patient consisted of the age of diagnosis, race, sex, primary site, laterality, stage T, stage N, grade, histology, surgery, survival status and time. Individuals who had unclear information were subsequently excluded. We collected and retrospectively analyzed the demographic data of all eligible cases.

The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). This article does not contain any studies with human participants or animals performed by any of the authors. All procedures performed in study involving human data were extracted freely from the SEER Research Data available to the public online (https://seer.cancer.gov/data/access.html).

Statistical analysis

The analysis of differences in clinical characteristics between the different groups was performed by using the chi-square test. An analysis of Cox regression was implemented to assess the risk factors of overall survival. It was set up in that P<0.05 was considered statistically significant. R software was used to set up a nomogram based on the numerous possible prognostic factors which could be associated with the overall survival. The validation of the nomogram was executed, and the results were evaluated based on the concordance index and the calibration curve. The concordance index was used to denote the predictive accuracy and differentiation ability of each factor of the nomogram. Calibration curves were performed to check the calibration of the nomogram. Kaplan–Meier curves were used to illustrate and compare the overall survival of patients in different clinical characteristics. Data extraction was collected using SEER*Stat software version 8.3.6.1 and data analyses were all performed using R software (version 4.0.0).

Results

Demographics of the entire cohort of patients

Overall, 5,792 patients with stage III non-small cell lung cancer were included among which 2,963 (51.16%) patients were diagnosed at the age of ≥70 years. Males accounted for 53.94% and the majority of patients were white (79.70%) which is related to the fact that the United States is a white-dominated country. In terms of the primary site, the percentages of the upper lobe and lower lobe were 64.93% and 30.27% respectively. For the laterality, the right tumors (58.34%) were the most common and the pathology of adenocarcinoma (53.49%) had a higher proportion than the squamous-cell carcinoma. For the grade, the grade II and grade III had a higher proportion (38.42% and 53.45%). Based on the AJCC guidelines for the staging of tumor nodule metastasis (TNM), T2, T3, and T4 had a higher proportion than T1. For lymph nodes, N2 (63.17%) was the most common. Only 1,898 (32.77%) of those patients underwent surgery ().

Table 1

Clinicopathologic characteristics of the entire cohort of patients

Characteristics	Number	%
Age
<70	2,829	48.84
≥70	2,963	51.16
Sex
Female	2,668	46.06
Male	3,124	53.94
Race
Black	679	11.72
White	4,616	79.70
Other	497	8.58
Site
Upper lobe	3,761	64.93
Middle lobe	278	4.80
Lower lobe	1,753	30.27
Laterality
Left	2,413	41.66
Right	3,379	58.34
Histology
Squamous-cell carcinoma	2,694	46.51
Adenocarcinoma	3,098	53.49
Grade
Grade I; Well differentiated	420	7.25
Grade II; Moderately differentiated	2,225	38.42
Grade III; Poorly differentiated	3,096	53.45
Grade IV; Undifferentiated; anaplastic	51	0.88
Stage-T
T1	768	13.26
T2	1,724	29.77
T3	1,520	26.24
T4	1,780	30.73
Stage-N
N0	715	12.34
N1	663	11.45
N2	3,659	63.17
N3	755	13.04
Chemotherapy
No/unknow	2,427	41.90
Yes	3,365	58.10
Radiotherapy
No/unknow	3,624	62.57
Yes	2,168	37.43
Surgery
Yes	1,898	32.77
No	3,894	67.23

Cox regression analysis for estimating the risk factors of overall survival

The Cox proportional hazards model was performed to investigate how each variable could help predict overall survival. Multivariate analyses indicated that factors such as age, sex, site, histology, grade, stage T, and N, chemoradiotherapy, surgical treatment were associated with the prognosis of patients ().

Table 2

Cox regression analysis for estimating the risk factors of overall survival

Characteristics	HR	95% CI of HR	P value
Age
<70	Reference
≥70	1.382	1.300–1.450	<0.001
Sex
Female	Reference
Male	1.210	1.138–1.287	<0.001
Race
Black	Reference
White	1.043	0.949–1.174	0.381
Other	0.877	0.763–1.007	0.063
Site
Upper lobe	Reference
Middle lobe	1.000	0.861–1.162	0.998
Lower lobe	1.104	1.033–1.180	0.004
Laterality
Left	Reference
Right	0.982	0.923–1045	0.567
Histology
Squamous-cell carcinoma	Reference
Adenocarcinoma	0.870	0.8224–0.935	<0.001
Grade
Grade I; Well differentiated	Reference
Grade II; Moderately differentiated	1.171	1029–1.333	0.017
Grade III; Poorly differentiated	1.308	1.153–1.485	<0.001
Grade IV; Undifferentiated; anaplastic	1.275	0.907–1.793	0.162
Stage-T
T1	Reference
T2	1.187	1.071–1317	0.001
T3	1.289	1.154–1.436	<0.001
T4	1.356	1.209–1.521	<0.001
Stage-N
N0	Reference
N1	1.205	1.047–1.387	0.009
N2	1.279	1.139–1.436	<0.001
N3	1.352	1.180–1.550	<0.001
Chemotherapy
No/unknow	Reference
Yes	2.321	2.212–2.567	<0.001
Radiotherapy
No/unknow	Reference
Yes	2.203	2.102–2.473	<0.001
Surgery
Yes	Reference
No	2.193	2.032–2.367	<0.001

Building and validating the novel nomogram

The model that incorporated the above factors was developed and presented as the nomogram (). In the nomogram, each predictor was scored based on a point scale. By adding up the total scores projected in the scale at the bottom, we could then predict the probability of overall survival for patients with stage III non-small cell lung cancer. The concordance index of the novel nomogram was 0.751, indicating the good discrimination ability of the model. The calibration curves also showed good consistency in the probability of 5-year overall survival between the actual observation and the nomogram prediction ().

Figure 1

Nomogram to predict the overall survival of patients with stage III non-small cell lung cancer.

Figure 2

Calibration curve showing nomogram-predicted overall survival compared with the actual overall survival.

Kaplan-Meier curves of overall survival for patients

The Kaplan-Meier curves showed that overall survival in the different age, sex, race, histology, grade, stage T, and N, chemoradiotherapy, surgical treatment was accurately differentiated with a significantly statistical difference ().

Figure 3

Kaplan-Meier curves of overall survival for patients in different clinical characteristics. (A) age, (B) sex, (C) chemotherapy, (D) radiotherapy, (E) laterality, (F) histology, (G) grade, (H) stage T, (I) stage N, (J) surgical treatment

Demographics of the patients with different overall survival

The significant difference in race, site, laterality was not found between the two groups. There was a significantly lower proportion of patients with the age of ≥70 years in the group with the overall survival of more than 5 years (P<0.001). The demographic and clinicopathologic characteristics like male, squamous-cell carcinoma were in the same way. However, the group with the overall survival of more than 5 years demonstrated a significantly higher proportion of patients with surgery. The group with the overall survival of more than 5 years had a higher proportion of well-differentiated tumors (grade I + grade II) (50.42% vs. 45.12%) as same as the stage T1 and N0 ().

Table 3

Demographic and clinicopathologic characteristics of different overall survival

Characteristics	Overall survival				P value
	≤5 years (n=5195)	% (89.69%)	>5 years (n=597)	% (10.31%)
Age					<0.001
<70	2,465	47.45%	364	60.97%
≥70	2,730	52.55%	233	39.03%
Sex					<0.001
Female	2,344	45.12%	324	54.27%
Male	2,851	54.88%	273	45.73%
Race					0.448
Black	607	11.68%	72	12.06%
White	4,150	79.88%	466	78.06%
Other	438	8.43%	59	9.88%
Site					0.291
Upper lobe	3,358	64.64%	403	67.50%
Middle lobe	248	4.77%	30	5.03%
Lower lobe	1,589	30.59%	164	27.47%
Laterality					0.153
Left	2,148	41.35%	265	44.39%
Right	3,047	58.65%	332	55.61%
Histology					<0.001
Squamous-cell carcinoma	2,467	47.49%	227	38.02%
Adenocarcinoma	2,728	52.51%	370	61.98%
Grade					0.038
Grade I; Well differentiated	364	7.01%	56	9.38%
Grade II; Moderately differentiated	1,980	38.11%	245	41.04%
Grade III; Poorly differentiated	2,806	54.01%	290	48.58%
Grade IV; Undifferentiated; anaplastic	45	0.87%	6	1.01%
Stage-T					<0.001
T1	657	12.65%	111	18.59%
T2	1,564	30.11%	160	26.80%
T3	1,380	26.56%	140	23.45%
T4	1,594	30.68%	186	31.16%
Stage-N					<0.001
N0	613	11.80%	102	17.09%
N1	579	11.15%	84	14.07%
N2	3,295	63.43%	364	60.97%
N3	708	13.63%	47	7.87%
Chemotherapy					<0.001
No/unknow	3,110	59.87%	212	35.51%
Yes	2,085	40.13%	385	64.49%
Radiotherapy					<0.001
No/unknow	3,542	68.18%	200	36.52%
Yes	1,653	31.82%	379	63.48%
Surgery					<0.001
Yes	1,527	29.39%	371	62.14%
No	3,668	70.61%	226	37.86%

Discussion

Non-small cell lung cancer is one of the most common types of lung cancer originating from alveolar epithelium or bronchial mucosal epithelium. However, the exact mechanism remains unclear. Moreover, more than 30% of patients with non-small cell lung cancer are diagnosed at advanced stages with a limited survival rate. The 5-year survival rate of stage III non-small cell lung cancer is from the range of 13% to 36% and the survival time is reduced when distant metastases occur (8). It has been widely accepted that surgery, chemotherapy, and radiotherapy are significant for patients with non-small cell lung cancer at advanced stages. The study that focuses on the clinical characteristics of stage III non-small cell lung cancer patients which may influence the prognosis is relatively less. In the present study, therefore, we have incorporated a total of 5,792 patients diagnosed with stage III non-small cell lung cancer using the SEER database to address the above-mentioned issue.

According to the global health statistics, 50% of non-small cell lung cancer patients were over 65 years old when they were first diagnosed and more than 30% of non-small cell lung cancer patients were 70 years old. The function of various organs of the body will be significantly reduced when patients are over 70 years old. Patients aged over 70 have their unique characteristics of pathology, physiology, and metabolism. Therefore, it has been a consensus to define non-small cell lung cancer patients aged ≥70 as elderly non-small cell lung cancer patients (9). The study of Ping showed that the median survival time of patients with lung cancer was 22.17 months in the young group (<50) and 20.15 months in the old group (>70). The cumulative survival time of the young group was better than that of the old group, but there was no statistically significant (P>0.05) (10). Mauri et al. also suggested that the prognosis of young lung cancer patients was similar to that of old lung cancer patients (11). However, some studies had shown that younger patients with lung cancer have a better prognosis, which is consistent with our study (12,13).

The reported studies had demonstrated that women with non-small cell lung cancer had a better prognosis than men with longer survival (14-16). Foeglé and colleagues found that there was no gender difference by retrospectively analyzing the outcomes of 1,738 non-small cell lung cancer patients (17). However, in our study, the 5-year survival of female patients with adenocarcinoma was significantly higher than that of male patients with squamous cell carcinoma patients, which might mean that adenocarcinoma had a better prognosis than squamous cell carcinoma. The significant difference of 5-year survival in age, sex, race, site, histology, grade, stage T, and N, chemoradiotherapy, surgical treatment was also found in the presented study except race, site, and laterality.

Multimodality therapy is an appropriate treatment approach for stage III non-small cell lung cancer patients. But the optimal treatment strategies are still indistinct. When surgical resection can be accomplished after adjuvant therapy with low risk and a good chance of complete resection, the survival time may be optimal. Despite the lack of information on chemoradiotherapy, we were still able to see that surgery could improve the prognosis of patients with stage III non-small cell lung cancer and there was a higher rate of surgical treatment among those with a survival of more than 5 years compared with those with the survival of less than 5 years. The results of our study are consistent with that of some previous studies (18).

There still are some several limitations that deserve attention in our study. First, because the study is a retrospective study, it will have an unavoidable confounding bias. Second, the details of chemotherapy and radiotherapy are not obtained in the SEER database, which hindered further prognostic analyses. Third, the population of our study was patients in America, which can not represent the global population. Finally, because the external validation was not accomplished, before the proposed nomogram can be used in clinical practice, it needs to be replicated and verified prospectively.

Conclusions

In conclusion, age, sex, histology, grade, stage T, and N, chemoradiotherapy, surgical treatment are associated with a 5-year survival of a patient with stage III non-small cell lung cancer. We constructed and validated a novel nomogram with relatively good accuracy to help predict 5-year overall survival for a patient with stage III non-small cell lung cancer. Clinicians can predict individualized survival and give treatment recommendations by using the prognostic model.