Sage Journals: Discover world-class research

Abstract

Purpose

We aimed to establish nomograms to predict the survival in patients aged ≥45 years with lung squamous cell carcinoma and brain metastasis.

Methods

We collected patients diagnosed as lung squamous cell carcinoma with brain metastasis aged ≥45 years between 2010 and 2019 from the Surveillance, Epidemiology, and End Results database. Prognostic factors were determined by the univariate and multivariate Cox regression analysis, and then the nomogram was constructed to predict cancer-specific survival and overall survival. Nomograms were evaluated by decision curve analysis, the area under the receiver operating characteristic curve, calibration plot, concordance index, and risk group stratification.

Results

In total, 2437 patients were included, with 1706 and 731 in the cohorts of training and validation, respectively. The age, N stage, T stage, liver metastasis, chemotherapy, bone metastasis, along with radiotherapy were significant in predicting the survival, and adopted for the establishment of nomograms. In the training and validation sets, the concordance index were .713(95%CI:0.699–.728) & .700(95%CI:0.677–.722) in predicting cancer-specific survival and .715(95%CI:0.701–.729) & .712(95%CI:0.690–.735) in predicting overall survival, respectively. Besides, the area under the receiver operating characteristic curve for predicting cancer-specific survival and overall survival in the training set were all >.7 at 1-, 2-, and 3- years. Calibration plots proved the survival predicted by nomograms were consistent with the actual values. decision curve analysis revealed better clinical validity of the nomogram in predicting cancer-specific survival and overall survival at 1-year than TNM staging. Patients were stratified into the high-/low-risk groups according to the optimal cutoff value of 100.21 for cancer-specific survival and 91.98 for overall survival. A web-based probability calculator was constructed finally.

Conclusion

Two nomograms were developed for the prognostic prediction of lung squamous cell carcinoma patients with brain metastasis aged ≥45 years, providing guidance for decision-making in clinical practice.

Keywords

lung squamous cell carcinoma brain metastasis cancer-specific survival overall survival nomograms surveillance,epidemiology and end results database

Introduction

As one of the most common malignancy, the non-small cell lung cancer (NSCLC) accounts for nearly 85% of patients with lung cancer (LC), which is the leading cause of deaths in the world.¹ Brain metastasis (BM) occur in about 10%–25% of NSCLC patients at the first visit and about 50% of patients during treatment.² The proportion of lung squamous cell carcinoma (LUSC) in all the NSCLCs is about 25%–30%,³ especially more likely to occur in middle-aged and older patients. In contrast to non-squamous cell carcinoma, incidence of BM has been found to be much lower in LUSC.⁴ Nevertheless, there have been few studies on BM of LUSC.

LUSC patients with BM usually present with a poor prognosis regarding of its insidious onset as well as the lack of targeted therapy. Approximately 70% of patients are already at the intermediate/advanced stage at diagnosis, losing opportunities for complete surgical resection.⁵ Targetable driver mutations, including epidermal growth factor receptor (EGFR), anaplastic lymphoma kinase (ALK), or ROS proto-oncogene 1 (ROS1) alterations, are exclusively found in the histological samples of adenocarcinoma but not in LUSC.⁶ Considering the limitations of treatment methods, the survival of LUSC patients with BM is as short as 3-6 months in median,⁷ bringing a heavy burden on their families and the whole society.

One study has reported on the prognostic influences of metastatic NSCLC in the elderly.⁸ There is also a study reporting on prognostic influences in LC younger than 45 years of age.⁹ In the study, LC patients younger than 45 years of age were defined as having early onset LC. These patients accounted for approximately 5% of all LC patients.¹⁰ Hereditary cancer factors are considered to be the main cause of early onset LC.¹¹ The World Health Organization (WHO) classifies age as follows: younger people under the age of 44, middle-aged people between the ages of 45 and 59, and older people over the age of 60. LUSC is most often seen in middle-aged and older patients. To date, no studies have reported on the prognostic factors influencing the prognosis of middle-aged and elderly patients with LUSC with BM.

Early diagnosis together with timely therapy is potential for the improvement of the survival of LUSC patients with BM. Since the WHO classification of thoracic tumors in 2021 didn’t incorporate histopathologic criteria for the prognosis of LUSC,¹² TNM staging has become the most frequently used tool for clinical staging and prognostic assessment of LUSC. Unfortunately, the prognostic factors such as the sex, age, histologic grade, or treatments in these patients were neglected in TNM staging.¹³ Thus, efficient tools are urgently needed to provide a reference for the diagnosis as well as prognostic assessment of BM in LUSC cases.

Currently, nomograms have been recognized as a unique and reliable method for the prognostic prediction of various tumor in the past decade, including gastric cancer, breast cancer, and testicular cancer,^14-16 but few in LUSC patients combined with BM. NSCLC can be classified into LUSC, lung adenocarcinoma (LUAD), adenosquamous carcinoma (ADSQC), and large-cell lung carcinoma (LCLC) according to histopathological classification, and the prognosis of patients with different pathological types varies significantly. Patients with LC with BM have a poor prognosis, and untreated patients have a short survival. Therefore, more attention needs to be paid to this problem and the factors affecting prognosis need to be studied. Prognostic factors influencing LC with BM have been reported in studies.¹⁷ Other scholars have reported the prognostic factors of NSCLC with BM and small cell lung cancer (SCLC) with BM.^18,19 In addition, other scholars have reported the prognostic influencing factors of metastatic LUSC and metastatic LUAD, respectively.^20,21 To date, no prognostic model for LUSC with BM has been reported in the literature. Thus, this objective of this study was to develop appropriate nomograms to evaluate the prognosis of LUSC patients combined with BM using the Surveillance, Epidemiology and End Results (SEER) database, facilitating the decision-making in clinical practice. We present the following article in accordance with the RECORD reporting checklist.

Patients and Methods

Data Source

Data used in this study were obtained from the Surveillance, Epidemiology, and End Results (SEER) database, covering approximately 30% of the whole U.S. population. Besides, SEER*Stat (version 8.4.0.1) was used (http://seer.cancer.gov/SEERSTAT/) to access the database. The study was carried out complying with the Declaration of Helsinki (as revised in 2013). Local ethics approval or statements were not required because the clinical data used in this study were obtained from the public-access SEER database and thus, the requirement for informed consent was waived.

Patient Selection

Patients with a pathological diagnosis of LUSC (histologic type code: 8070-8074, 8083) between 2010 and 2019 from the SEER database were included. The exclusion criteria included (1) patients with non-brain metastasis (n = 94 561); (2) patients less than 45 years old (n = 56); (3) patients with non-pathological diagnosis (n = 623); (4) patients with non-first primary cancer (n = 834); (5) patients with unknown Race (n = 8); (6) patients with unknown Marital (n = 147); (7) patients without left or right laterality (n = 129); (8) patients with T0\Tx cancer staging (n = 298); (9) patients with Nx cancer staging (n = 90); (10) patients with unknown bone metastasis (n = 36); (11) patients with unknown liver metastasis (n = 27); (12) patients with unknown lung metastasis (n = 23); (13) patients with unknown surgery (n = 2); (14) patients with the survival <1 month (n = 311). The flowchart for inclusion and exclusion is presented in Figure 1.

Figure 1.

Flowchart for inclusion and exclusion of LUSC patients with BM aged ≥45 years.

Data Collection

The demographic information of the patients (such as the age), the clinicopathological information of the tumor (such as the primary location) as well as results of follow-up (such as the survival months) were collected from the SEER database.

The included patients were divided into 3 groups (45–60 years old, 61–70 years old, and ≥71 years old) using the optimal cutoff value of the age calculated by X-tile (version 3.6.1, Yale University School of Medicine, US). Figure S1 shows estimation of the cutoff value for the age determined by X-tile software. Besides, to avoid the error caused by incongruous stage system, we completed the transformation of TNM staging and evaluated the patient data according to the 7th edition of TNM.

Endpoints included cancer-specific survival (CSS) and overall survival (OS). CSS referred to the duration between the diagnosis and a specific cancer‐related death; while OS referred to the duration between the diagnosis and the all-cause death/the final follow‐up.

Statistical Analysis

All statistical analyses were carried out via SPSS (26.0, IBM Corporation, NY, USA) together with R (4.1.0, R Foundation for statistical computing, Vienna, Austria). Patients were allocated into either the training (70%) or validation (30%) cohort. Categorical data were compared between cohorts using the chi-square test. Besides, the prognostic factors for the training set were determined through univariate & multivariate Cox proportional regression, and then a nomogram was constructed for prognostic prediction. The concordance index (C-index) together with the area under the receiver operating characteristic curve (AUROC) was adopted to quantify the discriminative power of nomograms. In addition, calibration plots (1000 bootstrap replicates) were used to calibrate the nomograms. Decision curve analysis (DCA) was conducted for the evaluation of their clinical validity. Total points of individuals were then computed using the nomograms. Finally, all patients were stratified into groups with high/low risks using the optimal cutoff value of receiver operating characteristic (ROC) curves, and their survival were compared via Log-rank test as well as Kaplan–Meier (K-M) curves. A bilateral P value less than .05 was indicative of statistical significance.

Results

Analysis of Demographic as Well as Clinicopathological Characteristics

In total, there were 2437 patients with LUSC and BM between 2010 and 2019 included, and then randomly allocated to the training cohort (N = 1706) along with the validation cohort (N = 731). There were 1562 (64.10%) males and 875 (35.90%) females. Besides, their age ranged within 45–96 years, including 602 (24.70%), 973 (39.93%), and 862 (35.37%) in the groups of 45-60 years, 61-70 years, and ≥71 years, respectively, suggesting the predominance of LUSC with BM in elderly patients. The majority of patients (80.10%) were white. More than half of patients (52.03%) were married, the 27.99% were separated, divorced or widowed (SDW), and 19.98% were unmarried.

There were 57.90% of patients with LUSC in the right lung and 42.10% in the left lung. The lesions occurred mainly in the upper lobe, accounting for 54.37% of all. Patients at Grade III/IV, Grade I/II, and unknown Grade accounted for 13.46%, 29.63%, and 56.91%, respectively. According to the 7th edition of TNM staging of American Joint Committee on Cancer (AJCC), the primary tumors were mainly in T4, T3, T2, and T1 stages in 865 (35.49%), 628 (25.77%), 760 (31.19%), and 184 (7.55%) patients, respectively. In addition, the patients tended to have later T stages (25.77% in T3 and 35.49% in T4) and later N stages (50.47% in N2 and 17.32% in N3). Bone metastasis accounted for 27.37%, lung metastasis accounted for 21.99% and liver metastasis accounted for 17.85%. In terms of the treatments, radiotherapy, and chemotherapy were the predominant ones, accounting for 77.10% and 52.56%, respectively, only 4.51% of patients underwent surgery.

All the characteristics between cohorts were found to be not statistically significant (all P > .05) (Table 1).

Table 1.

Clinicopathological Characteristics of Patients Aged 45 Years or Older With LUSC With BM.

Characteristics	All N (%)	Training cohort N (%)	Validation cohort N (%)	p
Total	2437	1706	731
Sex
Male	1562(64.10)	1095(64.19)	467(63.89)	.924
Female	875(35.90)	611(35.81)	264(36.11)
Age
45-60	602(24.70)	423(24.79)	179(24.49)	.802
61-70	973(39.93)	674(39.51)	299(40.90)
≥71	862(35.37)	609(35.70)	253(34.61)
Marital
Unmarried	487(19.98)	329(19.28)	158(21.61)	.394
Married	1268(52.03)	892(52.29)	376(51.44)
SDW	682(27.99)	485(28.43)	197(26.95)
Race
White	1952(80.10)	1379(80.83)	573(78.39)	.38
Black	339(13.91)	228(13.36)	111(15.18)
Other	146(5.99)	99(5.80)	47(6.43)
Primary location
Lower lobe	702(28.81)	484(28.37)	218(29.82)	.551
Middle lobe	105(4.31)	70(4.10)	35(4.79)
Upper lobe	1325(54.37)	943(55.28)	382(52.26)
Other	305(12.52)	209(12.25)	96(13.13)
Grade
I/II	328(13.46)	217(12.72)	111(15.18)	.216
III/IV	722(29.63)	516(30.25)	206(28.18)
Unknown	1387(56.91)	973(57.03)	414(56.63)
Laterality
Left	1026(42.10)	722(42.32)	304(41.59)	.771
Right	1411(57.90)	984(57.68)	427(58.41)
T Stage
T1	184(7.55)	120(7.03)	64(8.76)	.489
T2	760(31.19)	533(31.24)	227(31.05)
T3	628(25.77)	447(26.20)	181(24.76)
T4	865(35.49)	606(35.52)	259(35.43)
N stage
N0	525(21.54)	368(21.57)	157(21.48)	.921
N1	260(10.67)	180(10.55)	80(10.94)
N2	1230(50.47)	857(50.23)	373(51.03)
N3	422(17.32)	301(17.64)	121(16.55)
Bone metastasis
No	1770(72.63)	1245(72.98)	525(71.82)	.59
Yes	667(27.37)	461(27.02)	206(28.18)
Liver metastasis
No	2002(82.15)	1407(82.47)	595(81.40)	.562
Yes	435(17.85)	299(17.53)	136(18.60)
Lung metastasis
No	1901(78.01)	1333(78.14)	568(77.70)	.854
Yes	536(21.99)	373(21.86)	163(22.30)
Radiation
No/Unknown	558(22.90)	392(22.98)	166(22.71)	.926
Yes	1879(77.10)	1314(77.02)	565(77.29)
Chemotherapy
No/Unknown	1156(47.44)	802(47.01)	354(48.43)	.55
Yes	1281(52.56)	904(52.99)	377(51.57)
Surgery
No	2327(95.49)	1637(95.96)	690(94.39)	.11
Yes	110(4.51)	69(4.04)	41(5.61)

SDW: separated, divorced, or widowed.

Univariate as well as Multivariate Cox Regression for Predictive factors of Prognosis in the Training Cohort

According to the findings in univariate Cox regression, the following 10 factors were significantly related to the CSS as well as the OS among the training set, including the age, T stage, bone metastasis, marital status, N stage, lung metastasis, chemotherapy, liver metastasis, radiotherapy, as well as surgery (Table 2). Subsequently, multivariate Cox regression reveled the following 7 factors, including the age, N stage, T stage, liver metastasis, chemotherapy, bone metastasis, as well as radiotherapy, to be significant predictive factors of the prognosis (Table 3).

Table 2.

Univariate and Multivariate Analyses of Risk Factors for CSS in Training Cohort.

Indicators	Univariate			Multivariate
Indicators	HR	95% CI	p	HR	95% CI	p
Sex
Male	REF
Female	.927	.836–1.028	.152
Age
45-60	REF			REF
61-70	1.25	1.100–1.421	.001	1.209	1.061–1.378	.004
≥71	1.495	1.311–1.704	<.001	1.243	1.084–1.426	.002
Marital
Unmarried	REF			REF
Married	.799	.700–.912	.001	.895	.782–1.025	.109
SDW	1.026	.887–1.187	.727	1.034	.891–1.2	.662
Race
White	REF
Black	1.144	.989–1.323	.069
Other	.825	.661–1.030	.089
Primary location
Lower lobe	REF
Middle lobe	.937	.722–1.216	.624
Upper lobe	1.014	.904–1.138	.813
Other	1.149	.971–1.359	.105
Grade
I/II	REF
III/IV	1.112	.947–1.306	.197
Unknown	.996	.856–1.158	.959
Laterality
Left	REF
Right	1.078	.975–1.192	.145
T Stage
T1	REF			REF
T2	1.223	.991–1.511	.061	1.218	.985–1.506	.069
T3	1.558	1.257–1.931	<.001	1.489	1.197–1.852	<.001
T4	1.512	1.226–1.865	<.001	1.442	1.163–1.789	.001
N stage
N0	REF			REF
N1	1.066	.884–1.285	.504	1.169	.968–1.412	.104
N2	1.201	1.056–1.365	.005	1.273	1.114–1.454	<.001
N3	1.05	.895–1.233	.549	1.235	1.043–1.462	.014
Bone metastasis
No	REF			REF
Yes	1.391	1.244–1.556	<.001	1.269	1.128–1.427	<.001
Liver metastasis
No	REF			REF
Yes	1.465	1.288–1.665	<.001	1.421	1.239–1.628	<.001
Lung metastasis
No	REF			REF
Yes	1.232	1.093–1.388	.001	.985	.867–1.12	.822
Radiation
No/Unknown	REF			REF
Yes	.625	.556-.703	<.001	.797	.706-.899	<.001
Chemotherapy
No/Unknown	REF			REF
Yes	.378	.342-.419	<.001	.392	.351-.437	<.001
Surgery
No	REF			REF
Yes	.648	.497–.845	.001	.799	.609–1.049	.106

REF: reference; HR, hazard ratio; SDW: separated, divorced, or widowed.

Table 3.

Univariate and Multivariate Analyses of Risk Factors for OS in Training Cohort.

Indicator	Univariate			Multivariate
Indicator	HR	95% CI	p	HR	95% CI	p
Sex
Male	REF
Female	.935	.841–1.039	.212
Age
45-60	REF			REF
61-70	1.254	1.100–1.429	.001	1.215	1.064–1.389	.004
≥71	1.485	1.298–1.698	<.001	1.24	1.078–1.428	.003
Marital
Unmarried	REF			REF
Married	.825	.720–.946	.006	.92	.800–1.059	.245
SDW	1.056	.909–1.227	.477	1.062	.911–1.239	.44
Race
White	REF
Black	1.121	.965–1.302	.135
Other	.801	.637–1.008	.058
Primary location
Lower lobe	REF
Middle lobe	.919	.702–1.202	.538
Upper lobe	1.01	.898–1.136	.865
Other	1.132	.953–1.345	.158
Grade
I/II	REF
III/IV	1.103	.935–1.300	.245
Unknown	1.003	.86–1.171	.967
Laterality
Left	REF
Right	1.074	.969–1.190	.175
T Stage
T1	REF			REF
T2	1.266	1.018–1.574	.034	1.259	1.011–1.567	.04
T3	1.579	1.265–1.971	<.001	1.504	1.201–1.884	<.001
T4	1.535	1.236–1.907	<.001	1.462	1.171–1.826	.001
N stage
N0	REF			REF
N1	1.111	.918–1.344	.282	1.214	1.001–1.471	.049
N2	1.237	1.084–1.412	.002	1.301	1.134–1.492	<.001
N3	1.06	.898–1.251	.492	1.233	1.036–1.468	.018
Bone metastasis
No	REF			REF
Yes	1.414	1.261–1.585	<.001	1.288	1.143–1.453	<.001
Liver metastasis
No	REF			REF
Yes	1.497	1.314–1.707	<.001	1.446	1.258–1.661	<.001
Lung metastasis
No	REF			REF
Yes	1.223	1.082–1.382	.001	.977	.856–1.114	.724
Radiation
No/Unknown	REF			REF
Yes	.638	.566–.720	<.001	.808	.713–.914	.001
Chemotherapy
No/Unknown	REF			REF
Yes	.388	.349–.430	<.001	.399	.357–.446	<.001
Surgery
No	REF			REF
Yes	.618	.469–.816	.001	.765	.576–1.016	.065

REF: reference; HR, hazard ratio; SDW: separated, divorced or widowed.

Construction and Calibration of Nomograms

We constructed nomograms for the prediction of the CSS (Figure 2(A)) as well as OS (Figure 2(B)) at 1-/2-/3-year among LUSC patients with BM using the 7 significant prognostic factors found in multivariate Cox regression models.

Figure 2.

The nomograms for predicting 1-, 2-, 3-year CSS and OS in the training set of LUSC patients with BM aged ≥45 years. (A) The nomogram in predicting CSS. (B) The nomogram in predicting OS.

We further calibrated the nomograms using calibration plots, and the predicted probabilities for CSS (Figure 3(A) and (B)) and OS (Figure 3(C) and (D)) by nomograms were found to be consistent with the actual observed values in the training and validation sets, suggesting a good accuracy of the nomograms.

Figure 3.

Calibration plots of the nomograms in predicting 1-, 2-, 3-year CSS and OS in LUSC patients with BM aged ≥45 years. (A) Calibration plots of the nomograms in predicting 1-, 2-, 3-year CSS in the training set. (B) Calibration plots of the nomograms in predicting 1-, 2-, and 3-year CSS in the validation set. (C) Calibration plots of the nomograms in predicting 1-, 2-, 3-year OS in the training set. (D) Calibration plots of the nomograms in predicting 1-, 2-, 3-year OS in the validation set. The horizontal axis is the predicted value in the nomogram, and the vertical axis is the observed value.

Validation of Nomograms Using C-Index, Area Under The Receiver Operating Characteristic Curve and Decision Curve Analysis

Internal validation was used to test discriminative power of nomograms. For the training and validation sets, the C-index was .713(.699–.728) and .700(.677–.722) in predicting CSS, and .715(.701–.729) and .712(.690–.735) in predicting OS, respectively.

Besides, the AUROC at one-, two-, and three-years in predicting CSS was found to be .759, .717, and .737 for the training set as well as .725, .651, and .647 for the validation set, respectively (Figure 4(A) and (B)); whereas the AUROC for the prediction of the OS at one-, two-, and three-years was .761, .722, and .741 for training set as well as .737, .669, and .687 for the validation set, respectively (Figure 4(C) and (D)).

Figure 4.

The AUROC for predicting 1-, 2-, and 3-year CSS and OS in LUSC patients with BM aged ≥45 years. (A) The AUROC at 1-, 2-, and 3-year for the prediction of CSS in the training set was .759, .717, and .737, respectively. (B) The AUROC at 1-, 2-, and 3-year for the prediction of CSS in the validation set was .725, .651, and .647, respectively. (C) The AUROC at 1-, 2-, and 3-year for the prediction of OS in the training set was .761, .722, and .741, respectively. (D) The AUROC at 1-, 2-, and 3-year for the prediction of OS in the validation set was .737, .669, and .687, respectively.

DCAs for the nomograms as well as TNM staging systems are shown in Figure 5. Results suggested that the nomograms for the prediction of CSS was more beneficial at one-year in either the training or validation set in contrast to TNM staging; In comparison, the nomogram for the prediction of CSS at 2-, 3-year in either the training or validation set showed no apparent advantages over TNM staging (Figure 5(A) and (B)). Meanwhile, the nomogram for the prediction of OS was more beneficial at one-year in either the training or validation set in contrast to TNM staging; in contrast, the nomogram for the prediction of OS at two- and three-year in either the training or validation set showed no apparent advantages over TNM staging (Figure 5(C) and (D)).

Figure 5.

DCA of the nomograms for predicting CSS and OS in LUSC patients with BM aged ≥45 years. (A) In the training set, the nomogram in predicting CSS at 1 year was more beneficial compared to TNM staging, while the nomogram for CSS at 2, 3-year had no apparent advantages over TNM staging. (B) In the validation set, the nomogram in predicting CSS at the 1-year was more beneficial compared to TNM staging, while the nomogram for CSS at 2, 3-year had no apparent advantages over TNM staging system. (C) In the training set, the nomogram in predicting OS at 1-year was more beneficial compared to TNM staging, while the nomogram in predicting OS at 2, 3-year had no apparent advantages over TNM staging. (D) In the validation set, the nomogram in predicting OS at 1-year was more beneficial compared to TNM staging, while the nomogram for OS at 2, 3-year had no apparent advantages over TNM staging system.

Based on the above findings, it could be inferred that the nomograms had a favorable discriminative capability.

Risk Stratification of Patients Based on Scores of Nomograms

Total score of each patient was computed using the nomograms. Besides, the optimal cutoff value was determined by the receiver operating characteristic (ROC) curve. Based on the optimal cutoff values (100.21 for the prediction of CSS; 91.98 for the prediction of OS), patients were then further stratified into the groups with high or low risks. In addition, K-M curve revealed that compared to the group with low risks, the rate of CSS or OS were lower in the group with high risks in both cohorts (Figure 6) with statistical significance.

Figure 6.

The Kaplan-Meier curves of LUSC patients with BM aged ≥45 years in the low-risk and high-risk groups. The K-M curve showed that the CSS rate of the patients in the high-risk group was significantly lower than that in the low-risk group both in the training set (A) and validation set (B). The K-M curve showed that the OS rate of the patients in the high-risk group was significantly lower than that in the low-risk group both in the training set (C) and validation set (D).

Construction of Web-Based Probability Calculator

To make our nomograms more convenient for the application in clinical practice, we constructed a web-based calculator for the probability calculation of OS and CSS (https://fengyang.shinyapps.io/DynNomapp_OS/ and https://fengyang.shinyapps.io/DynNomapp_CSS/) for LUSC patients with BM aged ≥45 years. The survival probability of patients could be directly displayed once the variables of individual patients were inputed into the calculator.

Discussion

There have been several clinical models established for LUSC patients.^20,22 In the present study, two nomograms were developed and specifically predicted the OS and CSS in LUSC patients with BM aged ≥45 years based on SEER database for the initial time. The age, N stage, T stage, liver metastasis, chemotherapy, bone metastasis, as well as radiotherapy were identified as predictive factors for the prognosis and included in the nomograms. Subsequently, patients were further stratified into groups with high-/low-risks according to the scores of nomograms. The two nomograms may have a potential for the convenient prediction of OS & CSS of individual patients, providing significant implications for decision-making in clinical practice.

The role of age in the prognosis of LUSC patients has been controversial. Hu et al²³ found the age not to be a predictor of the survival among NSCLC patients combined with pericardial effusion using the SEER database. However, Gu et al has demonstrated that increasing age was correlated with a poor lung cancer-specific survival (LCSS) for LUSC patients,²⁴ consistent with the findings in our study. Therefore, more studies would be needed to further evaluate the impact of age on the survival of LUSC patients. Previous literature has reported that marital status has been shown to significantly affect the prognosis of patients with LUAD,²⁵ as divorced patients are less likely to enjoy a relatively harmonious family environment. In our study, marital status was included as one of the basic demographic information. Our study found through statistical analysis that marriage was not an independent factor affecting prognosis, and therefore, marriage was not included in the final prediction model. Further future studies are needed to assess the role of marital status on prognosis.

TNM staging is an internationally recognized prognostic factor for LC. The 8th edition of AJCC staging system emphasized the importance of primary tumor size in the treatment and prognosis of LC.²⁶ The diameter of the primary tumor has been proved to be related with the number of clonal cells, which is indicative of differentiation of cells, the malignancy of tumors, as well as the possibility of distant metastases.²⁷ Several studies have revealed a significant association of tumor size with the prognosis of NSCLC patients after surgery, radiotherapy, and chemotherapy,^28-30 which was similar to the findings in our study.

Researches have proved bone/liver metastases to be associated with the prognosis of LC patients,³¹ which is consistent with the results of our study. Bone metastasis has been found in 30%–40% of NSCLC patients, among which approximately 6% developed within one year after the diagnosis of LC, characterized by the pain and pathological fractures; Patients with bone metastases showed 4.4 months shorter OS compared to those without bone metastases.³² The 1-year survival rate in patients without bone metastasis was about 37%, which was more than 3 times of that in those with bone metastasis.³³ Compared to brain and bone metastases, liver metastases presented the worst prognosis among single-organ metastases among the patients with LC, with the OS of only 4 months in median.³¹ An analysis based on the US oncology database between 2010 and 2015 found that the 1-year survival rate for all patients with liver metastases was 15.1%, which was obviously lower than 24.0% for patients without liver metastases.³⁴

With the advent of genomics, targeted therapies have brought significant benefits for LC patients. However, chemotherapy has remained the preferred treatment for patients with advanced LUSC in clinical practice due to the low rate of genetic mutations in them. Some researchers found that chemotherapy significantly prolonged their survival, including elderly patients.²² Radiotherapy is an essential treatment for patients with stage IV LUSC, mainly including local and systemic radiotherapy. An analysis based on SEER database between 2010 and 2015 found that radiotherapy was significantly associated with OS rates in LUSC.³⁵ Consistent with the above findings, our study also demonstrated the significant roles of chemotherapy and radiotherapy in LUSC patients.

In this study, there were still several limitations. First, selection bias would be inevitable since it was retrospective study and composed of only patients having complete data. Second, SEER database included no information of some specific prognostic factors, such as smoking history, family cancer history, genetic mutations, as well as the assessment of physical status. Third, the information on targeted therapies and immunotherapy were unavailable in the SEER database. Due to the lack of relevant information on tyrosine kinase inhibitors (TKIs) and immune checkpoint inhibitors treatment in the SEER database, the two factors were not included in the study, which is one of the limitations of our study. In the future, it is necessary to include TKIs and immune checkpoint inhibitors in clinical studies. Besides, there was no information on systemic therapies, especially the type of operations, dose for radiation, together with the choice of chemotherapeutic agents. Fourth, the SEER database only contained the data of the single-organ metastases at diagnosis but not during follow-up. Finally, only internal validation was performed in this study. More studies would be necessary for external validation in different populations to evaluate the predictive value of our nomograms in the future.

Conclusion

Two nomograms were successfully established for the prediction of the CSS/OS among LUSC patients combined with BM aged ≥45 years based on the significant prognostic factors, including the age, N stage, T stage, liver metastasis, chemotherapy, bone metastasis, as well as radiotherapy. The two nomograms presented with a good accuracy and a discriminative power, potentially making the clinical decision-making more convenient in the real world.

Supplemental Material

Supplemental Material - Development and Validation of Prognostic Nomograms for Lung Squamous Cell Carcinoma With Brain Metastasis in Patients Aged 45 Years or Older: A Population-Based Study

Supplemental Material for Development and Validation of Prognostic Nomograms for Lung Squamous Cell Carcinoma With Brain Metastasis in Patients Aged 45 Years or Older: A Population-Based Study by Feng Yang, Lianjun Gao, Qimin Wang, and Wei Gao in Cancer Control

Footnotes

Acknowledgments

We thank the SEER database for providing clinical information, and we also thank CureEdit for English language editing.

Declaration of Conflicting Interests

All authors have completed the ICMJE uniform disclosure form. The authors have no conflicts of interest to declare.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical Approval

The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The SEER database is an open-access cancer database that only contains de-identified patient data. Therefore, this study was exempted from the approval by the Institutional Review Board of China Rehabilitation Research Center.

ORCID iD

Feng Yang

Supplemental Material

Supplemental material for this article is available online.

Appendix

References

Siegel

Miller

Fuchs

Jemal

. Cancer statistics, 2021. CA Cancer J Clin. 2021;71(1):7-33. doi:10.3322/caac.21654.

Ulahannan

Khalifa

Faivre-Finn

Lee

. Emerging treatment paradigms for brain metastasis in non-small-cell lung cancer: an overview of the current landscape and challenges ahead. Ann Oncol. 2017;28(12):2923-2931. doi:10.1093/annonc/mdx481

Youlden

Cramb

Baade

. The International epidemiology of lung cancer: geographical distribution and secular trends. J Thorac Oncol: Official Publication of the International Association for the Study of Lung Cancer. 2008;3(8):819-831. doi:10.1097/JTO.0b013e31818020eb

Zhang

Jin

Gao

Guo

Chen

. The development and validation of a nomogram for predicting brain metastases in lung squamous cell carcinoma patients: an analysis of the surveillance, epidemiology, and end results (SEER) database. J Thorac Dis. 2021;13(1):270-281. doi:10.21037/jtd-20-3494

Zhou

Yang

, et al. Epidermal growth factor receptor (EGFR) mutations in non-small cell lung cancer (NSCLC) of Yunnan in southwestern China. Oncotarget. 2017;8(9):15023-15033. doi:10.18632/oncotarget.14706

Hsu

Tsai

, et al. Real-world treatment pattern and prognostic factors of stage IV lung squamous cell carcinoma patients. Kaohsiung J Med Sci. 2022;38(10):1001-1011. doi:10.1002/kjm2.12599

Park

Decker

Wilson

. Prophylactic cranial irradiation for patients with locally advanced non-small-cell lung cancer at high risk for brain metastases. Clin Lung Cancer. 2015;16(4):292-297. doi:10.1016/j.cllc.2014.11.005

Sun

Liu

Yang

Ren

Dai

Wang

. Construction and validation of prognostic nomograms for elderly patients with metastatic non-small cell lung cancer. The clinical respiratory journal. 2022;16(5):380-393. doi:10.1111/crj.13491

Dai

Wang

Liu

, et al. Development and validation of prognostic nomogram for lung cancer patients below the age of 45 years. Bosn J Basic Med Sci. 2021;21(3):352-363. doi:10.17305/bjbms.2020.5079

10.

Graziano

Comin

Crisci

, et al. Functional polymorphisms of the microsomal epoxide hydrolase gene: a reappraisal on a early-onset lung cancer patients series. Lung cancer (Amsterdam, Netherlands). 2009;63(2):187-193. doi:10.1016/j.lungcan.2008.05.004.

11.

Timofeeva

Kropp

Sauter

, et al. Genetic polymorphisms of MPO, GSTT1, GSTM1, GSTP1, EPHX1 and NQO1 as risk factors of early-onset lung cancer. Int J Cancer. 2010;127(7):1547-1561. doi:10.1002/ijc.25175

12.

WHO Classification of Tumors Editorial Board . Thoracic Tumors. 5th ed. Lyon, France: International Agency for Research on Cancer; 2021.

13.

Amin

Greene

Edge

, et al. The eighth edition AJCC cancer staging manual: Continuing to build a bridge from a population-based to a more "personalized" approach to cancer staging. CA Cancer J Clin. 2017;67(2):93-99. doi:10.3322/caac.21388

14.

Zhang

. Development and validation of prognostic nomogram for young patients with gastric cancer. Ann Transl Med. 2019;7(22):641. doi:10.21037/atm.2019.10.77

15.

Mao

Kong

Chen

. Development and validation of prognostic nomogram for germ cell testicular cancer patients. Aging. 2020;12(21):22095-22111. doi:10.18632/aging.104063

16.

Pan

Yang

Chen

Tong

. Nomogram for predicting the overall survival of patients with inflammatory breast cancer: A SEER-based study. Breast. 2019;47:56-61. doi:10.1016/j.breast.2019.05.015

17.

Hao

. Risk and prognostic factors of brain metastasis in lung cancer patients: A surveillance, epidemiology, and end results population-based cohort study. Eur J Cancer Prev : The Official Journal of the European Cancer Prevention Organisation (ECP). 2023;32:498-511. doi:10.1097/cej.0000000000000790

18.

Huang

Tong

Fan

Zhao

. Construction of a nomogram to predict the prognosis of non-small-cell lung cancer with brain metastases. Medicine. 2020;99(31):e21339. doi:10.1097/md.0000000000021339

19.

Ding

Sheng

, et al. Development and validation of a nomogram for the prediction of brain metastases in small cell lung cancer. The clinical respiratory journal. 2023;17(5):456-467. doi:10.1111/crj.13615

20.

Liu

Sun

Xiong

Zhang

Liu

. Construction and validation of prognosis nomogram for metastatic lung squamous cell carcinoma: A population-based study. Technol Cancer Res Treat. 2022;21:15330338221132035. doi:10.1177/15330338221132035

21.

Zhao

Gao

Tan

Xue

Gao

. Specific organ metastases and prognosis in lung adenocarcinoma. Thoracic cancer. 2023;14(8):736-745. doi:10.1111/1759-7714.14801

22.

Chen

Gao

Tang

You

Dong

. A prognostic model for elderly patients with squamous non-small cell lung cancer: A population-based study. J Transl Med. 2020;18(1):436. doi:10.1186/s12967-020-02606-3

23.

Zeng

. Prognostic factors and nomogram for cancer-specific death in non small cell lung cancer with malignant pericardial effusion. PLoS One. 2019;14(5):e0217007. doi:10.1371/journal.pone.0217007

24.

Lee

, et al. Integrated evaluation of clinical, pathological and radiological prognostic factors in squamous cell carcinoma of the lung. PLoS One. 2019;14(10):e0223298. doi:10.1371/journal.pone.0223298

25.

Zhu

Chen

Zhang

. Relationship between marital status and survival in patients with lung adenocarcinoma: A SEER-based study. Medicine. 2022;101(1):e28492. doi:10.1097/md.0000000000028492

26.

Rami-Porta

Asamura

Travis

Rusch

. Lung cancer - major changes in the American joint committee on cancer eighth edition cancer staging manual. CA Cancer J Clin. 2017;67(2):138-155. doi:10.3322/caac.21390

27.

Shan

Fan

Guo

Han

Wang

. Relationship between tumor size and metastatic site in patients with stage IV non-small cell lung cancer: a large SEER-based study. PeerJ. 2019;7:e7822. doi:10.7717/peerj.7822

28.

Käsmann

Niyazi

Blanck

, et al. Predictive and prognostic value of tumor volume and its changes during radical radiotherapy of stage III non-small cell lung cancer: A systematic review. Strahlenther Onkol : Organ der Deutschen Rontgengesellschaft [et al]. 2018;194(2):79-90. Prädiktiver und prognostischer Wert des Tumorvolumens und seiner Veränderungen während radikaler Strahlentherapie beim nicht-kleinzelligen Bronchialkarzinom im Stadium III: Ein systematischer Review. doi:10.1007/s00066-017-1221-y

29.

Yang

Shah

, et al. Long-term outcomes of surgical resection for stage IV non-small-cell lung cancer: A national analysis. Lung cancer (Amsterdam, Netherlands). 2018;115:75-83. doi:10.1016/j.lungcan.2017.11.021

30.

Ball

Mitchell

Giroux

Rami-Porta

. Effect of tumor size on prognosis in patients treated with radical radiotherapy or chemoradiotherapy for non-small cell lung cancer. An analysis of the staging project database of the International association for the study of lung cancer. J Thorac Oncol : Official Publication of the International Association for the Study of Lung Cancer. 2013;8(3):315-321. doi:10.1097/JTO.0b013e31827dc74d

31.

Riihimäki

Hemminki

Fallah

, et al. Metastatic sites and survival in lung cancer. Lung cancer (Amsterdam, Netherlands). 2014;86(1):78-84. doi:10.1016/j.lungcan.2014.07.020

32.

Kuchuk

Sabri

Hutton

Clemons

Wheatley-Price

. The incidence and clinical impact of bone metastases in non-small cell lung cancer. Lung cancer (Amsterdam, Netherlands). 2015;89(2):197-202. doi:10.1016/j.lungcan.2015.04.007

33.

Cetin

Christiansen

Jacobsen

Nørgaard

Sørensen

. Bone metastasis, skeletal-related events, and mortality in lung cancer patients: A Danish population-based cohort study. Lung cancer (Amsterdam, Netherlands). 2014;86(2):247-254. doi:10.1016/j.lungcan.2014.08.022

34.

Horn

Stoltzfus

Lehrer

, et al. Epidemiology of liver metastases. Cancer epidemiology. 2020;67:101760. doi:10.1016/j.canep.2020.101760

35.

Yang

Shen

Shao

Wang

Shen

. Based on the development and verification of a risk stratification nomogram: Predicting the risk of lung cancer-specific mortality in stage IIIA-N2 unresectable large cell lung neuroendocrine cancer compared with lung squamous cell cancer and lung adenocarcinoma. Front Oncol. 2022;12:825598. doi:10.3389/fonc.2022.825598

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.25 MB