Development and Validation of Nomograms to Predict Overall Survival and Cancer-Specific Survival for Non-Small Cell Lung Cancer with Chest Wall Invasion: A Population-Based Study

DOI: https://doi.org/10.21203/rs.3.rs-2510232/v1

Abstract

Background: Chest wall invasion is a relatively kind of infrequent direct tumor extension in non-small cell lung cancer (NSCLC) with a poor survival outcome. Risk factors that impact overall survival (OS) and cancer-specific survival (CSS) remain unclear. Therefore, we aimed to explore prognostic factors in NSCLC patients with chest wall invasion and construct predictive nomograms to predict both OS and CSS in NSCLC patients with chest wall invasion.

Methods: We extracted a total of 2091 patients diagnosed with primary NSCLC with chest wall invasion between 2010 and 2015 from the Surveillance, Epidemiology, and End Results (SEER) database. The total patients were divided into two groups randomly, the training cohort (1463 patients) and the validation cohort (628 patients). Univariate and multivariate Cox regression analyses were applied to distinguish the independent prognostic factors. Two prognostic nomograms for OS and CSS were established based on the training cohort and were evaluated in both cohorts. The concordance index (C-index), receiver operating characteristic curves (ROC), calibration curves, and decision curve analysis (DCA) curves were applied to assess the performance of these two nomograms.

Results: After multivariate Cox analysis, age, sex, histology, grade, N stage, M stage, surgery, and chemotherapy were identified as independent prognostic factors for OS,  meanwhile, age, histology, grade, N stage, M stage, surgery, and chemotherapy for CSS.

The C-index of the nomogram for OS in the training and validation cohorts was 0.711 and 0.716, respectively. The C-index of the nomogram for CSS in the training and validation cohorts was 0.721 and 0.726, respectively. The ROC curves, calibration curves, DCA curves, and K-M survival curves also exhibited good predictive performance in the training and validation cohorts of these two prognostic nomograms.

Conclusion: Two nomograms provide a useful and reliable tool to predict both OS and CSS in NSCLC patients with chest wall invasion. These nomograms can provide strong references to facilitate clinic decisions.

Introduction

According to the Global Cancer Statics 2020, malignant lung cancer is the leading cause of cancer mortality worldwide, with about 2.2 million new cases and 1.8 million new deaths in 2020[1]. Among lung cancer, non-small cell lung cancer (NSCLC) is the most common subset, accounting for 75 to 80% of the patients. Chest wall invasion by direct extension in lung cancer is a rare phenomenon, ranging from approximately 5–8%[24]. In addition, the anatomical structure of the chest wall consists of endo-thoracic fascia, parietal pleural, intercostal muscles, rib, external intercostal muscles, intercostal vessel, and intercostal nerves[5]. Chest wall invasion is considered an independent T descriptor in the AJCC 8th TNM staging system, meanwhile, tumors size less than 7 cm with chest wall involvement are defined as T3, and can be further categorized into IIB to IV stage combined with the presence of lymph node and distant metastasis[6]. As previous studies have reported, the most common symptom of patients with NSCLC invading the chest wall is chest pain which indicates a highly specific feature of chest wall involvement[7]. Besides, symptoms like dyspnoea and hemoptysis are also reported, especially in the case of large invading lesions.

These tumors invading the chest wall are considered to be potentially resectable. Coleman et al first reported the surgical management of primary lung cancer with chest wall involvement in 1947[8]. Since then, many retrospective studies have indicated that patients with lung cancer involving the chest wall can benefit from surgery, especially in cases without lymph node metastasis[9]. In a study of 1855 NSCLC patients with complete surgery, 104 cases (5.6%) had chest wall invasion and the 5-year survival rate in patients with parietal pleura invasion only was better than those with soft tissue invasion[10]. Besides, they concluded that in cases with chest wall invasion, lymph node involvement could lead to a worse survival rate, and no difference was described in lung adenocarcinoma and squamous cell carcinoma. In another subset of 100 patients with non-superior sulcus lung cancer invading the chest wall, Gregory et al demonstrated that the 5-year survival rate was only 45%, meanwhile higher pathological stage, without radiological response to induction therapy and cardiovascular comorbidity were correlated with poor OS[11]. Although comprehensive therapy based on surgery, chemotherapy, radiotherapy, and immunotherapy has developed rapidly, the 5-year survival rate still varies from 25–55%, with a worse survival outcome in patients who undergo incomplete surgical resection or occurred lymph node metastasis[12, 13]. What’s more, some reports described that the rate of local disease recurrence within the primary site or chest wall has ranged widely from 1–17%, too[1416]. There were a large number of studies focusing on the NSCLC patients invading the chest wall, however, most of them were limited to a small sample or based on patients with surgery. Incomplete resection and lymph node metastasis have been consistently confirmed as poor prognostic factors for NSCLC patients with chest wall involvement, whereas the survival impact of other factors is still unknown.

As a widely used novel statistic method in clinical investigation, nomograms can accurately predict distant metastasis, OS, and CSS of cancer patients. Most existing nomograms, however, are derived from NSCLC patients without chest wall invasion. Besides, nomograms for NSCLC patients with chest wall invasion have not been published until now. Developing more specific nomograms that directly aim at NSCLC patients with chest wall invasion, rather than general NSCLC patients, could be of greater clinical value.

Therefore, we aimed to explore the independent associated prognostic factors for OS and CSS in NSCLC patients with chest wall invasion from the SEER database. Meanwhile, we developed two novel nomograms to predict OS and CSS in NSCLC patients with chest wall invasion based on the demographic and clinicopathologic variables. As a result, this will facilitate individualized patient care as well as medical therapy.

Materials And Methods

Database

We applied the specialized database “Incidence–SEER Research Plus Data, 18 Registries, Nov 2020 Sub (2000–2018)” to extract clinical-pathological data of patients through the SEER*Stat software, version 8.4.0.

Patients Collection

The status of chest wall invasion has been recorded between 2004 and 2015 based on the term, CS extension (code 600 and 610), so the patients were recognized from the SEER database between 2010 and 2015. The inclusion criteria were as follows: (a) malignant tumor located in the main bronchus and lung (Site code: C340-C349); (b) patients diagnosed with primary NSCLC and unique; (c) diagnostic confirmation based on positive histology; (d) the status of chest wall extension was recorded; (e) T stage, N stage, and M stage on the basis of the 7th edition AJCC staging system was complete. Besides, clinical variables including age at diagnosis, sex, race, grade, tumor site, laterality, surgery, radiotherapy, and chemotherapy were contained. The exclusion criteria were as follows: (a) survived less than 1 month after diagnosis and unknown; (b) patients aged < 18 years; (c) unknown data on race, marital, grade, laterality, surgery, radiotherapy, and chemotherapy. Finally, a total of 2091 patients matched the criteria in the primary cohort.

OS was defined as the interval between the initial diagnosis and the occurrence of any cause of death. CSS was defined as the interval between the initial diagnosis and the occurrence of lung cancer-specific death. OS was the primary endpoint whereas CSS was the secondary endpoint. Age at diagnosis was divided into under 60 years old, 60–70 years old, and over 70 years old. Laterality was grouped into other (Left and bilateral) and right. Histology was grouped into adenocarcinoma, squamous cell carcinoma (SCC), and others. Tumor size was grouped into 1 (≤ 3 cm), 2 (>3 cm, ≤ 5 cm), 3 (>5 cm, ≤ 7 cm), and 4 (>7 cm). All the selected processes of the primary cohort were shown in Fig. 1.

Statistical Analysis

In this retrospective study, R software (version 4.0.3) was applied to perform all statistical analyses and P-value < 0.05 (two sides) was regarded as statistical significance. Furthermore, we transformed all continuous variables into categorical variables except survival time to simplify the analyses. Count and percentage were used to summarize categorical variables.

In the primary cohort with chest wall invasion, training and validation cohorts were extracted through R software with a ratio of 7:3 randomly, meanwhile, the distribution and difference between the two cohorts were examined by the Chi-square test or Fisher’s exact test. The univariate and multivariate Cox proportional hazards regression analyses were performed to determine independent OS- and CSS-related factors. Risk factors with P-value < 0.05 in the univariate Cox analysis were further analyzed in the multivariate Cox analysis. Two prognostic nomograms were established based on the independent prognostic factors in the training cohort by the “rms” package. Time-dependent receiver operating characteristic (ROC) curves were performed to predict 1-year, 3-year, and 5-year OS and CSS, and the corresponding area under the curve (AUC) was measured to show the discrimination as well as C-index. To determine the consistency between predicted and actual probability, calibration curves were generated. Decision curve analysis (DCA) curves were generated to assess the clinical benefits and improved performance of the nomograms. Besides, all NSCLC patients were divided into high-risk and low-risk groups based on the median risk score in both cohorts. Kaplan–Meier (K–M) survival analysis with the log-rank test was adopted to differentiate OS and CSS between the two groups in both cohorts.

Results

Baseline characteristics 

After identifying patients based on the inclusion and exclusion criteria, 2091 NSCLC patients with chest wall invasion were enrolled in our study. The median follow-up time was 13 months (1-107 months). At the end of follow-up, 1694 (81.0%) death of any cause, and 1433 (68.5%) cancer-specific death were observed. The 1-year OS rate, 3-year OS rate, and 5-year OS rate were 51.9%, 27.3%, and 21.3%, respectively. Among the cohort, 1220 (58.3%) patients were males and 1572 (75.1%) patients were aged 60 years. White (80.2%) was the majority of the population, while others counted for 19.8%. The most common T, N, and M stage were T3, N0, and M0 respectively. 1034 patients (49.5%) underwent surgery, 1059 patients (50.6%) underwent radiotherapy, and 1300 patients (69.9%) underwent chemotherapy. The histology of lung cancer was adenocarcinoma, squamous cell carcinoma, and others. Other variables were exhibited in Table 1. The training cohort (1463 patients) and validation cohorts (628 patients) were generated randomly. No significant difference could be found between the two subgroups in the primary cohort (Table 1). K-M curves showed that patients could benefit from surgery whether with lymph node metastasis (Figure 2A, C) and distant metastasis (Figure 2B, D) or not. 

Prognostic risk factors

As shown in Table 2, age, sex, marital, histology, grade, T stage, N stage, M stage, tumor size, surgery, and chemotherapy were distinguished as risk factors for OS after univariate Cox regression analysis. Then, multivariate Cox regression analysis further confirmed that age, sex, histology, grade, N stage, M stage, tumor size, surgery, and chemotherapy were the independent prognostic factors to predict OS in NSCLC patients with chest wall involvement.

As exhibited in Table 3, for CSS, we observed similar results in univariate Cox regression analysis except for sex. Age, histology, grade, N stage, M stage, tumor size, surgery, and chemotherapy were finally confirmed as independent risk factors to predict CSS in NSCLC patients with chest wall involvement.

Nomograms construction and validation

A prognostic nomogram for OS and another prognostic nomogram for CSS were established based on the independent risk factors that were confirmed by multivariate analysis (Figure 3A-B). In the OS nomogram, the C-index was 0.711 in the training cohort, showing a good discrimination ability of the nomogram as well as 0.716 in the validation cohort. While in the CSS nomogram, the C-index was 0.721 and 0.726 in the training and validation cohorts, respectively. In the OS nomogram, the AUCs of the training cohort for the 1-, 3-, and 5-year reached 0.773, 0.744, and 0.776, while 0.794, 0.781, and 0.790 in the validation cohort, respectively (Figure 4A-B). The K-M curves demonstrated that a significantly worse OS was observed among patients in the high-risk group (Figure 4C-D). 

In the CSS nomogram, the AUCs for the 1-, 3-, and 5-year reached 0.781, 0.747, and 0.776 in the training cohort, while 0.808, 0.779, and 0.791 in the validation cohort, respectively (Figure 5A-B). In addition, the patients in the high-risk group showed significantly worse CSS than the patients in the low-risk group by K-M analysis (Figure 5C-D). What’s more, in both nomograms, calibration curves exhibited outstanding consistency between predicted and actual survival at the 1-, 3-, and 5-year in the training and validation cohorts, respectively (Figure 6-7). In these two nomograms, DCA curves at the 1-, 3-, and 5-year exhibited that the nomogram had excellent predictive accuracy in both cohorts (Figure 8-9).

Discussion

We conducted this large-scale population-based retrospective study based on NSCLC patients with chest wall invasion to investigate prognostic predictors via the national cancer registry SEER database. Besides, independent prognostic associated factors for NSCLC patients with chest wall invasion for OS and CSS were eventually confirmed. Furthermore, to predict OS and CSS conveniently, we established two prognostic nomograms with reliable accuracy and discriminative ability which were validated by ROC, calibration, K-M, and DCA curves. These nomograms can serve as a practical tool for clinicians to identify patients with a high risk of poor survival and to determine the optimal clinical treatment for NSCLC patients initially diagnosed with chest wall invasion.

The chest wall is the most common site invaded by peripheral lung cancer, and its adverse survival impact is clear because the current 8th TNM staging system classified it as an independent T3 stage for NSCLC with the tumor size less than 7 cm[17, 18]. In the past, the tumor invading the chest wall has been considered a contraindication for surgery for a long time. Surgery has been gradually accepted by thoracic surgeons since Gronquist et al published the first encouraging survival results of patients with chest wall invasion after surgery in 1947[8]. To date, With the rapid development of surgical instruments and skills, more and more studies have demonstrated that surgery with R0 resection can improve survival in NSCLC patients with chest wall invasion. For example, in a study of 104 NSCLC patients with chest wall invasion, Giancarlo et al indicated that surgery could improve the survival outcome significantly during the last 3 decades and advocated the performance of radical en bloc resection for the treatment of NSCLC with chest wall invasion[19]. Another retrospective study consisting of 135 NSCLC patients, it does not worsen the quality of life and pulmonary function of patients who underwent pulmonary resection with chest wall removal compared with patients who underwent pulmonary resection only[20]. Therefore, surgery for NSCLC patients with chest wall invasion is feasible and en bloc resection is recommended as the primary choice, even if it requires removing the partial chest wall. In our study, surgery was identified as an independent positive prognostic factor for both OS and CSS as previously reported. Additionally, our results indicated that surgery had a significant survival benefit even in NSCLC patients with lymph node metastasis and distant metastasis. Nonetheless, it requires more prospective studies to evaluate the survival impact of surgery in patients with distant metastasis.

Chest wall involvement in NSCLC patients is often categorized as locally advanced stage in which adjuvant therapy is inevitable. Chemotherapy is recommended as first-line treatment in many cancers, especially in the advanced stage like small cell lung cancer, pleural mesothelioma, and ovarian cancer[2123]. In a large database study of 2326 eligible NSCLC patients with chest wall invasion (T3N0) constructed by Drake, they summarized that patients who were treated with adjuvant chemotherapy after en bloc resection (R0) had significantly better median survival than those without chemotherapy before and after propensity score matching[24]. Similarly, Gao et al found adjuvant chemotherapy could bring a survival benefit in NSCLC patients invading the chest wall without lymph node invasion[25]. In addition, patients with malignant pleural effusion could benefit from surgery and adjuvant chemotherapy[26] and sugemalimab combined with chemotherapy demonstrated a statistically significant and clinically meaningful progression-free survival improvement in patients with metastatic NSCLC[27]. As the same as the previous studies, our results revealed that NSCLC patients with chest wall invasion can benefit from chemotherapy. However, the benefits of radiotherapy in patients with chest wall invasion remain debatable. Magdeleinat et al revealed that radiotherapy could not improve survival even if a possibility of complete resection existed, but, they found that lymph node involvement was not a contradiction to surgery[28]. Another study demonstrated adjuvant radiation therapy had no significant benefits on local recurrence and overall survival as well[25]. Consistent with these studies, our study indicated that radiotherapy did not influence the OS and CSS in NSCLC patients with chest wall invasion. Whereas, in a prospective, multi-institutional phase II study (CJLSG0801), induction radiation plus chemotherapy followed by surgery was safe and effective with a high rate of pathologic response for patients with NSCLC invading the chest wall[29]. Largacha et al demonstrated that induction chemotherapy combined with high-dose radiation followed by surgical resection was correlated with the improvement of OS in patients with non-superior sulcus lung cancer with chest wall invasion[30]. What’s more, other studies also revealed that radiotherapy had a positive benefit on survival[25, 31]. In my opinion, those studies that considered radiotherapy useless were limited by small samples, current irradiation techniques, and incomplete resection. In the end, I would like to emphasize that in the present study, we can not distinguish neoadjuvant therapy and adjuvant therapy from the SEER database. Thus, more clinical studies are required to clarify the role of radiotherapy and chemotherapy in NSCLC patients with chest wall invasion.

Tumor size has already been a T descriptor in TNM staging system and its influence on direct tumor extension and survival is widely accepted. In a study conducted by Lee, tumor size (> 5 cm) and lymph node involvement were negatively related to OS[32]. To explore the impact of the depth of the chest wall invasion on survival, Wu et al suggested that compared with pT4, rib invasion has a poorer prognosis, and patients with parietal pleura invasion and tumor size ranging from 5.1cm to 7.0cm could be up-classified from pT3 to pT4[33]. A larger tumor size led to worse OS and CSS in our study as the previous conclusion. Chest wall invasion tends to involve mediastinal lymph nodes and distant organs which exhibit adverse impact on survival. Recently, Jones said that pathological nodal status and higher pathological stage were associated with poor OS in locally advanced NSCLC with chest wall invasion[11]. Similarly, Burkhart et al identified that node-positive and male were related to poor survival[34]. And beyond that, a deep learning model demonstrated that node-positive and distant metastasis had a negative survival benefit[35]. Our multivariate analysis concluded that tumor size, lymph node invasion, and distant metastasis were independent negative predictors for OS and CSS, which was in line with current opinions. As for histology, Burkhart et al indicated that no significant difference in impact on survival was observed between adenocarcinoma and squamous cell carcinoma[34], as well as Facciolo, reported[10]. Our analyses indicated that there was also no significant difference between adenocarcinoma and squamous cell carcinoma while others were a significant risk factor for OS and CSS. It requires more studies concentrating on histopathology to clarify its survival impact. Older age was associated with poor survival in many cancers[3640], which was consistent with our results. The reason why older age can worse survival may be that patients with older age have a poor physical condition, have a high risk of metastasis and are more likely to die from other diseases. In a pan-cancer analysis, the incidence and survival of cancers vary significantly by sex which indicates males generally have lower incidence and survival compared to females[41]. Other studies demonstrated that compared with men, women had better survival in metastatic pancreatic cancer[42], esophageal cancer[43], and large cell lung cancer[44]. Male was an independent risk factor for OS while not for CSS in our study which needs further studies to illustrate. Increasing tobacco use, increasing exposure to oncogenic agents, engaging in high-risk behaviors, and fewer health care services for males may attribute to the difference in survival between males and females.

Nomograms can predict prognosis efficiently and accurately in other tumors. Mo et al combined age, CEA, perineural invasion, circumferential resection margin status, grade, lymph nodes harvested, mismatch repair deficiency status, and T stage to predict survival in stage II colorectal cancer [45]. In germ cell testicular cancer, a nomogram for CSS was established based on age, race, AJCC stage, TM stage, SEER stage, and radiotherapy[46]. To date, there are many studies on the prognosis and nomograms of NSCLC, nonetheless, there is still no study concentrated on the prognosis in NSCLC patients with chest wall invasion based on the clinical characteristics, which leads to worse cancer prognosis. In the present study, two novel prognostic nomograms were established with good performance in accuracy, discrimination, and predictive benefits.

However, there are still some limitations in this study. First, prospective randomized controlled studies are required to confirm our results because of the selection bias of the retrospective study. Second, due to the lack of external validation in the present study, an inherent bias can not avoid. Third, we don’t include complete information on surgery such as R0 resection or not, chest wall reconstruction or not, and so on, which may affect survival outcomes. Finally, the depth of chest wall invasion is not described in detail which may confound the survival impact of chest wall invasion.

Conclusions

We comprehensively demonstrated that age, sex, histology, grade, N stage, and M stage, tumor size, surgery, and chemotherapy were the independent risk factors for OS, while age, histology, grade, N stage, and M stage, tumor size, surgery, and chemotherapy for CSS in NSCLC patients with chest wall invasion. Two prognostic nomograms were developed based on the above results with good performance. These two nomograms are clinically useful to assess the prognosis of NSCLC patients with chest wall invasion and could facilitate clinical decision-making.

Abbreviations

NSCLC

non-small cell lung cancer

OS

overall survival

CSS

cancer-specific survival

C-index

concordance index

ROC

receiver operating characteristic curves

DCA

decision curve analysis

SCC

squamous cell carcinoma

NOS

not otherwise specified

Declarations

Author contributions 

JY and HY conceptualized and designed the study. HY and GWZ generated the figures and tables. JY, HY and GWZ completed the statistical results. JY and HY wrote the initial manuscript. BTY scrutinized in every aspect of this study who critically reviewed the original article. JY, and HY contributed equally to this work.

Funding

None.

Acknowledgements

Not applicable.

Data availability

All data of the patients in our study are accessible in the Surveillance, Epidemiology, and End Results database (http://www.seer.cancer.gov/).

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

None.

References

  1. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, Bray F: Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 2021, 71(3):209-249.
  2. Riquet M, Arame A, Le Pimpec Barthes F: Non-small cell lung cancer invading the chest wall. Thoracic surgery clinics 2010, 20(4):519-527.
  3. Stoelben E, Ludwig C: Chest wall resection for lung cancer: indications and techniques. European journal of cardio-thoracic surgery : official journal of the European Association for Cardio-thoracic Surgery 2009, 35(3):450-456.
  4. Matsuoka H, Nishio W, Okada M, Sakamoto T, Yoshimura M, Tsubota N: Resection of chest wall invasion in patients with non-small cell lung cancer. European journal of cardio-thoracic surgery : official journal of the European Association for Cardio-thoracic Surgery 2004, 26(6):1200-1204.
  5. Birchard SJ, Schertel ER: Chapter 167 - Principles of Thoracic Surgery. In: Saunders Manual of Small Animal Practice (Third Edition). edn. Edited by Birchard SJ, Sherding RG. Saint Louis: W.B. Saunders; 2006: 1723-1731.
  6. Zhao M, Wu J, Deng J, Wang T, Haoran E, Gao J, Xu L, Wu C, Hou L, She Y et al: Proposal for Rib invasion as an independent T descriptor for non-small cell lung cancer: A propensity-score matching analysis. Lung Cancer 2021, 159:27-33.
  7. Filosso PL, Sandri A, Guerrera F, Solidoro P, Bora G, Lyberis P, Ruffini E, Oliaro A: Primary lung tumors invading the chest wall. J Thorac Dis 2016, 8(Suppl 11):S855-S862.
  8. Coleman FP: Primary Carcinoma of the Lung, with Invasion of the Ribs: Pneumonectomy and Simultaneous Block Resection of the Chest Wall. Annals of surgery 1947, 126(2):156-168.
  9. Wang L, Yan X, Zhao J, Chen C, Chen C, Chen J, Chen KN, Cao T, Chen MW, Duan H et al: Expert consensus on resection of chest wall tumors and chest wall reconstruction. Translational lung cancer research 2021, 10(11):4057-4083.
  10. Facciolo F, Cardillo G, Lopergolo M, Pallone G, Sera F, Martelli M: Chest wall invasion in non-small cell lung carcinoma: a rationale for en bloc resection. J Thorac Cardiovasc Surg 2001, 121(4):649-656.
  11. Jones GD, Caso R, No JS, Tan KS, Dycoco J, Bains MS, Rusch VW, Huang J, Isbell JM, Molena D et al: Prognostic factors following complete resection of non-superior sulcus lung cancer invading the chest wall. European journal of cardio-thoracic surgery : official journal of the European Association for Cardio-thoracic Surgery 2020, 58(1):78-85.
  12. Kawaguchi K, Yokoi K, Niwa H, Ohde Y, Mori S, Okumura S, Shiono S, Ito H, Yano M, Shigemitsu K et al: Trimodality therapy for lung cancer with chest wall invasion: initial results of a phase II study. Ann Thorac Surg 2014, 98(4):1184-1191.
  13. Tandberg DJ, Kelsey CR, D'Amico TA, Crawford J, Chino JP, Tong BC, Ready NE, Wright A: Patterns of Failure After Surgery for Non-Small-cell Lung Cancer Invading the Chest Wall. Clin Lung Cancer 2017, 18(4):e259-e265.
  14. Riquet M, Lang-Lazdunski L, Le PB, Dujon A, Souilamas R, Danel C, Manac'h D: Characteristics and prognosis of resected T3 non-small cell lung cancer. Ann Thorac Surg 2002, 73(1):253-258.
  15. Choi Y, Lee IJ, Lee CY, Cho JH, Choi WH, Yoon HI, Lee YH, Lee CG, Keum KC, Chung KY et al: Multi-institutional analysis of T3 subtypes and adjuvant radiotherapy effects in resected T3N0 non-small cell lung cancer patients. Radiation oncology journal 2015, 33(2):75-82.
  16. Gould PM, Bonner JA, Sawyer TE, Deschamps C, Lange CM, Li H: Patterns of failure and overall survival in patients with completely resected T3 N0 M0 non-small cell lung cancer. International journal of radiation oncology, biology, physics 1999, 45(1):91-95.
  17. Kawaguchi K, Miyaoka E, Asamura H, Nomori H, Okumura M, Fujii Y, Nakanishi Y, Eguchi K, Mori M, Sawabata N et al: Modern surgical results of lung cancer involving neighboring structures: a retrospective analysis of 531 pT3 cases in a Japanese Lung Cancer Registry Study. J Thorac Cardiovasc Surg 2012, 144(2):431-437.
  18. Detterbeck FC, Boffa DJ, Kim AW, Tanoue LT: The Eighth Edition Lung Cancer Stage Classification. Chest 2017, 151(1):193-203.
  19. Roviaro G, Varoli F, Grignani F, Vergani C, Pagano C, Maciocco M, Romanelli A: Non-small cell lung cancer with chest wall invasion: evolution of surgical treatment and prognosis in the last 3 decades. Chest 2003, 123(5):1341-1347.
  20. Liu M, Wampfler JA, Dai J, Gupta R, Xue Z, Stoddard SM, Cassivi SD, Jiang G, Yang P: Chest wall resection for non-small cell lung cancer: A case-matched study of postoperative pulmonary function and quality of life. Lung Cancer 2017, 106:37-41.
  21. Popat S, Baas P, Faivre-Finn C, Girard N, Nicholson AG, Nowak AK, Opitz I, Scherpereel A, Reck M: Malignant pleural mesothelioma: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up(☆). Annals of oncology : official journal of the European Society for Medical Oncology 2022, 33(2):129-142.
  22. Sun A, Durocher-Allen LD, Ellis PM, Ung YC, Goffin JR, Ramchandar K, Darling G: Guideline for the Initial Management of Small Cell Lung Cancer (Limited and Extensive Stage) and the Role of Thoracic Radiotherapy and First-line Chemotherapy. Clinical oncology (Royal College of Radiologists (Great Britain)) 2018, 30(10):658-666.
  23. Armstrong DK, Alvarez RD, Bakkum-Gamez JN, Barroilhet L, Behbakht K, Berchuck A, Chen LM, Cristea M, DeRosa M, Eisenhauer EL et al: Ovarian Cancer, Version 2.2020, NCCN Clinical Practice Guidelines in Oncology. Journal of the National Comprehensive Cancer Network : JNCCN 2021, 19(2):191-226.
  24. Drake JA, Sullivan JL, Weksler B: Adjuvant chemotherapy improves survival in patients with completely resected T3N0 non-small cell lung cancer invading the chest wall. J Thorac Cardiovasc Surg 2018, 155(4):1794-1802.
  25. Gao SJ, Corso CD, Blasberg JD, Detterbeck FC, Boffa DJ, Decker RH, Kim AW: Role of Adjuvant Therapy for Node-Negative Lung Cancer Invading the Chest Wall. Clin Lung Cancer 2017, 18(2):169-177.e164.
  26. Li X, Li M, Lv J, Liu J, Dong M, Xia C, Zhao H, Xu S, Wei S, Song Z et al: Survival Benefits for Pulmonary Adenocarcinoma With Malignant Pleural Effusion After Thoracoscopic Surgical Treatment: A Real-World Study. Frontiers in oncology 2022, 12:843220.
  27. Zhou C, Wang Z, Sun Y, Cao L, Ma Z, Wu R, Yu Y, Yao W, Chang J, Chen J et al: Sugemalimab versus placebo, in combination with platinum-based chemotherapy, as first-line treatment of metastatic non-small-cell lung cancer (GEMSTONE-302): interim and final analyses of a double-blind, randomised, phase 3 clinical trial. The Lancet Oncology 2022, 23(2):220-233.
  28. Magdeleinat P, Alifano M, Benbrahem C, Spaggiari L, Porrello C, Puyo P, Levasseur P, Regnard JF: Surgical treatment of lung cancer invading the chest wall: results and prognostic factors. Ann Thorac Surg 2001, 71(4):1094-1099.
  29. Kawaguchi K, Yokoi K, Niwa H, Ohde Y, Mori S, Okumura S, Shiono S, Ito H, Yano M, Shigemitsu K et al: A prospective, multi-institutional phase II study of induction chemoradiotherapy followed by surgery in patients with non-small cell lung cancer involving the chest wall (CJLSG0801). Lung Cancer 2017, 104:79-84.
  30. Muñoz-Largacha JA, Rao SR, Brinckerhoff LH, Daly BD, Fernando HC, Litle VR, Suzuki K: Induction chemoradiation is associated with improved survival in chest wall invasion lung cancer. Tumori 2019, 105(4):331-337.
  31. Kawaguchi K, Yokoi K, Niwa H, Ohde Y, Mori S, Okumura S, Shiono S, Ito H, Yano M, Shigemitsu K et al: Trimodality therapy for lung cancer with chest wall invasion: initial results of a phase II study. 2014, 98(4):1184-1191.
  32. Lee CY, Byun CS, Lee JG, Kim DJ, Cho BC, Chung KY, Park IK: The prognostic factors of resected non-small cell lung cancer with chest wall invasion. World J Surg Oncol 2012, 10:9.
  33. Wu LL, Li CW, Li K, Qiu LH, Xu SQ, Lin WK, Ma GW, Li ZX, Xie D: The Difference and Significance of Parietal Pleura Invasion and Rib Invasion in Pathological T Classification With Non-Small Cell Lung Cancer. Frontiers in oncology 2022, 12:878482.
  34. Burkhart HM, Allen MS, Nichols FC, 3rd, Deschamps C, Miller DL, Trastek VF, Pairolero PC: Results of en bloc resection for bronchogenic carcinoma with chest wall invasion. J Thorac Cardiovasc Surg 2002, 123(4):670-675.
  35. She Y, Jin Z, Wu J, Deng J, Zhang L, Su H, Jiang G, Liu H, Xie D, Cao N et al: Development and Validation of a Deep Learning Model for Non-Small Cell Lung Cancer Survival. JAMA network open 2020, 3(6):e205842.
  36. Lin YT, Hsu PK, Hsu HS, Huang CS, Wang LS, Huang BS, Hsu WH, Huang MH: En bloc resection for lung cancer with chest wall invasion. Journal of the Chinese Medical Association : JCMA 2006, 69(4):157-161.
  37. Yin Z, Wang Y, Wu Y, Zhang X, Wang F, Wang P, Tao Z, Yuan Z: Age distribution and age-related outcomes of olfactory neuroblastoma: a population-based analysis. Cancer management and research 2018, 10:1359-1364.
  38. Tang X, Zhou X, Li Y, Tian X, Wang Y, Huang M, Ren L, Zhou L, Ding Z, Zhu J et al: A Novel Nomogram and Risk Classification System Predicting the Cancer-Specific Survival of Patients with Initially Diagnosed Metastatic Esophageal Cancer: A SEER-Based Study. Annals of surgical oncology 2019, 26(2):321-328.
  39. Lin S, Mo H, Li Y, Guan X, Chen Y, Wang Z, Yuan P, Wang J, Luo Y, Fan Y et al: Development and validation of a nomogram for predicting survival of advanced breast cancer patients in China. Breast (Edinburgh, Scotland) 2020, 53:172-180.
  40. Bai Y, Wei C, Zhong Y, Zhang Y, Long J, Huang S, Xie F, Tian Y, Wang X, Zhao H: Development and Validation of a Prognostic Nomogram for Gastric Cancer Based on DNA Methylation-Driven Differentially Expressed Genes. International journal of biological sciences 2020, 16(7):1153-1165.
  41. Dong M, Cioffi G, Wang J, Waite KA, Ostrom QT, Kruchko C, Lathia JD, Rubin JB, Berens ME, Connor J et al: Sex Differences in Cancer Incidence and Survival: A Pan-Cancer Analysis. Cancer epidemiology, biomarkers & prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology 2020, 29(7):1389-1397.
  42. Pijnappel EN, Schuurman M, Wagner AD, de Vos-Geelen J, van der Geest LGM, de Groot JB, Koerkamp BG, de Hingh I, Homs MYV, Creemers GJ et al: Sex, Gender and Age Differences in Treatment Allocation and Survival of Patients With Metastatic Pancreatic Cancer: A Nationwide Study. Frontiers in oncology 2022, 12:839779.
  43. Qiu MJ, Yang SL, Wang MM, Li YN, Jiang X, Huang ZZ, Xiong ZF: Prognostic evaluation of esophageal cancer patients with stages I-III. Aging 2020, 12(14):14736-14753.
  44. Shi Y, Chen W, Li C, Qi S, Zhou X, Zhang Y, Li Y, Li G: Clinicopathological characteristics and prediction of cancer-specific survival in large cell lung cancer: a population-based study. J Thorac Dis 2020, 12(5):2261-2269.
  45. Mo S, Zhou Z, Li Y, Hu X, Ma X, Zhang L, Cai S, Peng J: Establishment and validation of a novel nomogram incorporating clinicopathological parameters into the TNM staging system to predict prognosis for stage II colorectal cancer. Cancer cell international 2020, 20:285.
  46. Mao W, Wu J, Kong Q, Li J, Xu B, Chen M: Development and validation of prognostic nomogram for germ cell testicular cancer patients. Aging 2020, 12(21):22095-22111.

Tables

Table 1 Baseline characteristics among NSCLC patients with chest wall invasion.


All
 (N=2091)

Training cohort
 (N=1463)

Validation cohort
 (N=628)

P

Age





<60

519 (24.8%)

361 (24.7%)

158 (25.2%)

0.892

>70

812 (38.8%)

573 (39.2%)

239 (38.1%)


60-70

760 (36.3%)

529 (36.2%)

231 (36.8%)


Sex





Female

871 (41.7%)

616 (42.1%)

255 (40.6%)

0.530

Male

1220 (58.3%)

847 (57.9%)

373 (59.4%)


Race





Black

250 (12.0%)

172 (11.8%)

78 (12.4%)

0.382

Other

163 (7.8%)

107 (7.3%)

56 (8.9%)


White

1678 (80.2%)

1184 (80.9%)

494 (78.7%)


Marital





Married

1123 (53.7%)

776 (53.0%)

347 (55.3%)

0.364

Unmarried

968 (46.3%)

687 (47.0%)

281 (44.7%)


Grade





I

90 (4.3%)

63 (4.3%)

27 (4.3%)

0.568

II

645 (30.8%)

464 (31.7%)

181 (28.8%)


III

1295 (61.9%)

892 (61.0%)

403 (64.2%)


IV

61 (2.9%)

44 (3.0%)

17 (2.7%)


Histology





Adenocarcinoma

936 (44.8%)

641 (43.8%)

295 (47.0%)

0.395

SCC

871 (41.7%)

622 (42.5%)

249 (39.6%)


Others

284 (13.6%)

200 (13.7%)

84 (13.4%)


Laterality





Other

909 (43.5%)

637 (43.5%)

272 (43.3%)

0.962

Right

1182 (56.5%)

826 (56.5%)

356 (56.7%)


Tumor site





Main bronchus

20 (1.0%)

13 (0.9%)

7 (1.1%)

0.385

Upper lobe

1451 (69.4%)

1018 (69.6%)

433 (68.9%)


Middle lobe

54 (2.6%)

37 (2.5%)

17 (2.7%)


Lower lobe

446 (21.3%)

320 (21.9%)

126 (20.1%)


Overlapping lesion

41 (2.0%)

28 (1.9%)

13 (2.1%)


NOS

79 (3.8%)

47 (3.2%)

32 (5.1%)


T stage





T3

1811 (86.6%)

1272 (86.9%)

539 (85.8%)

0.485

T4

280 (13.4%)

191 (13.1%)

89 (14.2%)


N stage





N0

971 (46.4%)

671 (45.9%)

300 (47.8%)

0.617

N1

281 (13.4%)

194 (13.3%)

87 (13.9%)


N2

659 (31.5%)

474 (32.4%)

185 (29.5%)


N3

180 (8.6%)

124 (8.5%)

56 (8.9%)


M stage





M0

1368 (65.4%)

964 (65.9%)

404 (64.3%)

0.514

M1

723 (34.6%)

499 (34.1%)

224 (35.7%)


Tumor size





1

334 (16.0%)

225 (15.4%)

109 (17.4%)

0.135

2

625 (29.9%)

458 (31.3%)

167 (26.6%)


3

567 (27.1%)

385 (26.3%)

182 (29.0%)


4

565 (27.0%)

395 (27.0%)

170 (27.1%)


Surgery





No

1057 (50.6%)

746 (51.0%)

311 (49.5%)

0.567

Yes

1034 (49.5%)

717 (49.0%)

317 (50.5%)


Chemotherapy





No

791 (37.8%)

561 (38.3%)

230 (36.6%)

0.461

Yes

1300 (62.2%)

902 (61.7%)

398 (63.4%)


Radiation





No

1032 (49.4%)

725 (49.6%)

307 (48.9%)

0.811

Yes

1059 (50.6%)

738 (50.4%)

321 (51.1%)


SCC: squamous cell carcinoma; Tumor size: 1: ≤ 3 cm, 2:>3 cm, ≤ 5 cm, 3:>5 cm, ≤ 7 cm, 4: >7 cm

 

Table 2 Univariate and multivariate Cox regression analyses of risk factors correlated with OS in NSCLC patients with chest wall invasion.


Univariate analysis

Multivariate analysis


HR

95%CI

P

HR

95%CI

P

Age, years







<60

Reference






60-70

1.152

0.987-1.344

0.073

1.150

0.982-1.347

<0.001

>70

1.612

1.389-1.871

<0.001

1.495

1.277-1.749

<0.001

Sex







Female

Reference






Male

1.170

1.042-1.315

0.008

1.273

1.128-1.437

<0.001

Race







Black

Reference






Other

0.833

0.632-1.097

0.192




White

1.044

0.873-1.248

0.637




Marital







Married

Reference






Unmarried

1.171

1.044-1.312

0.007

1.062

0.943-1.196

0.324

Grade







I

Reference






II

1.128

0.842-1.512

0.418

1.261

0.936-1.699

0.128

III

1.276

0.961-1.696

0.092

1.400

1.048-1.871

0.023

IV

1.660

1.087-2.536

0.019

1.765

1.137-2.741

0.012

Histology







Adenocarcinoma

Reference






SCC

1.120

0.989-1.267

0.073

1.114

0.977-1.272

0.108

Others

1.367

1.150-1.625

<0.001

1.253

1.045-1.501

0.015








Laterality







Other

Reference






Right

0.993

0.885-1.114

0.903




Tumor site







Main bronchus

Reference






Upper lobe

0.647

0.366-1.144

0.135




Middle lobe

0.618

0.313-1.221

0.166




Lower lobe

0.884

0.496-1.575

0.674




Overlapping lesion

0.539

0.263-1.103

0.091




Lung, NOS

1.205

0.636-2.281

0.567




T stage







T3

Reference






T4

1.364

1.157-1.608

<0.001

0.883

0.742-1.052

0.163

N stage







N0

Reference






N1

1.266

1.056-1.517

0.011

1.210

1.005-1.458

0.045

N2

1.782

1.563-2.032

<0.001

1.443

1.245-1.672

<0.001

N3

2.454

2.003-3.006

<0.001

1.564

1.253-1.954

<0.001

M stage







M0

Reference






M1

2.180

1.936-2.454

<0.001

1.763

1.528-2.034

<0.001

Tumor size







1

Reference






2

1.104

0.918-1.329

0.293

1.131

0.937-1.366

0.200

3

1.498

1.241-1.807

<0.001

1.357

1.116-1.649

0.002

4

1.735

1.440-2.089

<0.001

1.454

1.197-1.765

<0.001

Surgery







No

Reference






Yes

0.410

0.365-0.462

<0.001

0.549

0.475-0.636

<0.001

Chemotherapy







No

Reference






Yes

0.648

0.577-0.728

<0.001

0.523

0.462-0.593

<0.001

Radiotherapy







No

Reference






Yes

0.942

0.841-1.056

0.308




SCC: squamous cell carcinoma; Tumor size: 1: ≤ 3 cm, 2:>3 cm, ≤ 5 cm, 3:>5 cm, ≤ 7 cm, 4: >7 cm

 

Table 3 Univariate and multivariate Cox regression analyses of prognostic factors correlated with CSS in NSCLC patients with chest wall invasion.


Univariate analysis

Multivariate analysis


HR

95%CI

P

HR

95%CI

P

Age, years







<60

Reference






60-70

1.077

0.912-1.271

0.383

1.110

0.937-1.315

0.225

>70

1.494

1.274-1.753

<0.001

1.403

1.185-1.662

<0.001

Sex







Female

Reference






Male

1.103

0.973-1.251

0.126




Race







Black

Reference






Other

0.869

0.645-1.171

0.357




White

1.065

0.876-1.294

0.528




Marital







Married

Reference






Unmarried

1.205

1.064-1.364

0.003

1.023

0.901-1.16

0.729

Grade







I

Reference






II

1.033

0.755-1.413

0.838

1.166

0.847-1.604

0.347

III

1.233

0.911-1.669

0.175

1.352

0.993-1.841

0.056

IV

1.699

1.088-2.653

0.020

1.702

1.071-2.706

0.024

Histology







Adenocarcinoma

Reference






SCC

1.438

1.197-1.727

<0.001

1.114

0.977-1.272

0.108

Others

1.069

0.934-1.224

0.331

1.253

1.045-1.501

0.015








Laterality







Other

Reference






Right

0.987

0.871-1.118

0.839




Tumor site







Main bronchus

Reference






Upper lobe

0.621

0.342-1.128

0.118




Middle lobe

0.624

0.305-1.274

0.195




Lower lobe

0.815

0.445-1.493

0.508




Overlapping lesion

0.529

0.248-1.131

0.101




Lung, NOS

1.200

0.616-2.339

0.592




T stage







T3

Reference






T4

1.447

1.216-1.723

<0.001

0.881

0.733-1.06

0.180

N stage







N0

Reference






N1

1.385

1.138-1.687

0.001

1.305

1.066-1.597

0.010

N2

1.990

1.725-2.295

<0.001

1.524

1.299-1.790

<0.001

N3

2.766

2.23-3.431

<0.001

1.648

1.300-2.088

<0.001

M stage







M0

Reference






M1

2.406

2.118-2.732

<0.001

1.763

1.528-2.034

<0.001

Tumor size







1

Reference






2

1.122

0.914-1.377

0.271

1.132

0.919-1.395

0.245

3

1.503

1.221-1.850

<0.001

1.361

1.097-1.690

0.005

4

1.907

1.557-2.336

<0.001

1.580

1.279-1.952

<0.001

Surgery







No

Reference






Yes

0.375

0.329-0.426

<0.001

0.536

0.457-0.629

<0.001

Chemotherapy







No

Reference






Yes

0.668

0.588-0.757

<0.001

0.517

0.451-0.592

<0.001

Radiotherapy







No

Reference






Yes

0.979

0.864-1.108

0.731




SCC: squamous cell carcinoma; Tumor size: 1: ≤ 3 cm, 2:>3 cm, ≤ 5 cm, 3:>5 cm, ≤ 7 cm, 4: >7 cm