Evaluation of the prognostic value of the new 9th edition Tumor-Node-Metastases (TNM) staging system for epidermal growth factor receptor (EGFR)-mutated lung adenocarcinoma patients with bone metastases

Abstract

Background

There are some changes in the new 9th edition Tumor-Node-Metastases (TNM) staging system for lung cancer, including subdividing M1c into M1c1 and M1c2 stage. The aim of this study was to assess the prognostic performance of the updated classification system and try to provide some real-world application data among advanced lung adenocarcinoma patients with bone metastases.

Methods

Advanced lung adenocarcinoma patients in M1c stage with bone metastases who receiving first-line first-generation epidermal growth factor receptor tyrosine kinase inhibitor (EGFR-TKI) and T790M-guided osimertinib as the second-line therapy were retrospectively screened from December 2016 to December 2021. A total of 126 patients were enrolled in this study. 62 patients and 64 patients were subdivided into M1c1 and M1c2 groups according to the 9th edition of TNM staging system.The first-line real-world progression-free survival (1LrwPFS), the second-line real-world progression-free survival (2LrwPFS), post-progression survival (PPS) and real-world overall survival (rwOS) were analyzed.

Results

The overall median rwOS was 40.1 months (95% CI 35.996–44.204). 1LrwPFS was 13.9 months (95% CI 12.653–15.147) and 2LrwPFS was 14.5 months (95% CI 11.665–17.335) for all patients.Patients in M1c2 stage was inferior to M1c1 stage patients in rwOS (35.2 months vs. 42.9 months, HR = 0.512, P = 0.005). 2LrwPFS was moderately correlated with rwOS (r = 0.621, R² = 0.568, P = 0.000). Multivariate analysis showed performance status (PS) score ≥ 2 and TP53 alteration positive were independent prognostic factors of worse rwOS.

Conclusions

More refined stratification of M1c according to the 9th edition of TNM staging system is conducive to the judgment of prognosis and the implementation of precision medicine for patients.

Introduction

Epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKI) have become standard first-line treatment [1–3]. Compared to conventional chemotherapy, EGFR-TKIs including gefitinib, erlotinib, icotinib, afatinib, dacomitinib and osimertinib improve progression-free survival (PFS) and overall survival (OS) of advanced non-small cell lung cancer (NSCLC) patients with sensitive epidermal growth factor receptor (EGFR) mutations, of which the third-generation EGFR-TKI such as osimertinib is preferred [4–11]. There are multiple treatment options of EGFR-TKI as the first-line treatment in real-world setting [12, 13], and treatment decisions mainly depend on the characteristics of tumors, the stage of tumor and the feasibility based on patient-related factors, particularly with regards to cost-effectiveness in clinical practice. Pharmacoeconomic studies suggest that the first-line first-generation EGFR-TKI and T790M-guided osimertinib as the second-line treatment was more cost-effective than the first-line osimertinib treatment [14–17]. In addition, APPLE trial updated survival data and showed comparable OS between the first-line osimertinib versus gefitinib followed by osimertinib in European Lung Cancer Congress 2023 [18]. In real-world setting, the first-line first-generation EGFR-TKI followed by the sequential use of third-generation EGFR-TKI with acquired EGFR T790M mutation will continue to play an irreplaceable role and be taken.

30–50% of advanced lung adenocarcinoma patients have initial bone metastases at diagnosis of lung cancer [19–22]. Bone metastases at diagnosis of lung cancer was closely associated with worse prognosis and potentially coexisted with complex multi-organ metastases, compared with non-bone metastases [23–26]. Although the clinical application of EGFR-TKI extended the survival of advanced lung adenocarcinoma patients, the survival of patients with bone metastases is worrying. In addition, survival data of patients with bone metastases was relatively rare, compared with survival studies of patients with brain metastases, and we need to pay more attention to the prognosis of these patients in real-world practice.

Staging system for lung cancer is very important to predict prognosis and select accurate treatment. The World Conference on Lung Cancer (WCLC) 2023 announced details about the ninth edition of the classification of primary tumors, lymph nodes and metastases (TNM) for lung cancer, updating M1c grouping. Multiple extrathoracic metastases, which was M1c descriptor in the 8th edition of TNM staging system, are now divided into multiple extrathoracic metastases in a single organ system (M1c1) and multiple extrathoracic metastases in multiple organ system (M1c2) in the 9th edition of TNM staging system. A more detailed stratification might be more conducive to the implementation of precision medicine for patients.

The aim of this study was to assess the prognostic performance of the 9th edition of TNM staging system among advanced lung adenocarcinoma patients in M1c stage with bone metastases and we try to provide some real-world survival data about application of the updated classification system.

Materials and methods

Patients

Clinical records of patients with stage IV lung adenocarcinoma from December 2016 to December 2021 at Shanghai Jiao Tong University School of Medicine affiliated Shanghai Chest Hospital were screened. Multiple bone metastases at initial diagnosis of lung cancer in M1c stage was subdivided into multiple extrathoracic metastases in bone (M1c1 group) and multiple extrathoracic metastases in bone and simultaneous other extrathoracic organ (M1c2 group) according to the 9th TNM staging system. The inclusion criteria for patient selection were as follows: (1) Histological or cytological examinations showed lung adenocarcinoma; (2) Patients in M1c stage were included according to the 9th edition of the TNM staging system; (3) Sensitive EGFR mutations (19 deletion or 21 L858R mutation) were tested positive by amplification refractory mutation system (ARMS) or Next-generation sequencing (NGS); (4) Patients had bone metastases at the initial diagnosis of lung cancer; (5) Patients received first-line first-generation EGFR-TKI and EGFR T790M mutation guided second-line osimertinib. Exclusion criteria were: (1) EGFR mutation was negative or not available; (2) Without initial bone metastases at diagnosis of lung cancer; (3) Insensitive EGFR mutations (Any mutation else other than 19 deletion and 21 L858R mutation); (4) First-generation EGFR-TKI as first-line therapy and third-generation EGFR-TKI as the second-line treatment with acquired T790M mutation was not taken; (5) EGFR-TKI treatment was combined with other chemotherapy; (6) Incomplete follow-up information. The patient selection processes were shown in Fig. 1.

Fig. 1.

Flowchart of patient selection steps. Abbreviations EGFR, epidermal growth factor receptor; EGFR-TKI, epidermal growth factor receptor tyrosine kinase inhibitor; M1c1 group, multiple extrathoracic metastases in bone; M1c2 group, multiple extrathoracic metastases in bone and simultaneous other extrathoracic organ

Diagnosis, treatment and follow-up

The following examinations were used as clinical baseline assessment for all patients to determine the stage of lung cancer: laboratory blood tests, chest computed tomography (CT), ultrasonic examination of abdomen and supraclavicular lymph nodes, brain magnetic resonance imaging (MRI), single-photon emission computed tomography (SPECT) bone scan or positron emission tomography (PET)-CT. Bone metastases was detected by bone scan or positron emission tomography (PET)-CT, confirmed with computed tomography (CT) or magnetic resonance imaging (MRI). Patients underwent laboratory blood tests, chest CT, ultrasonic examination of abdomen and supraclavicular lymph nodes and metastatic organ MRI in the follow-up treatment when treatment response was assessed.

One of the first-generation EGFR-TKIs, gefitinib (250 mg once a day, AstraZeneca, UK), erlotinib (150 mg once a day, Roche, Switzerland), or icotinib (125 mg three times a day, Beta, China) was used as the first-line treatment. After disease progressed, patients underwent liquid and/or tissue biopsy for EGFR T790M detection. Patients with acquired EGFR T790M mutation received third-generation EGFR-TKI such as osimertinib (80 mg once a day, AstraZeneca, UK) as the second-line treatment until disease progression, unacceptable adverse events.

The treatment response was assessed one month after the initiation of EGFR-TKI therapy and subsequently after every 8 weeks. Tumor responses were classified as complete response (CR), partial response (PR), stable disease (SD) and progressive disease (PD), according to the response evaluation criteria in solid tumors (RECIST) version 1.1 [27].

Data collection

Patients’ medical records and follow-up data from initial diagnosis of lung cancer to the last follow-up visit were all collected. Follow-up was performed by hospital visits and telephone conversations. The first-line real-world progression-free survival (1LrwPFS) was defined as the time interval from the initiation of first-generation EGFR-TKI therapy to the date of clinician-defined progression based on RECIST version 1.1. The second-line real-world progression-free survival (2LrwPFS) was defined as the time interval from the initiation of third-generation EGFR-TKI therapy to the date of clinician-defined progression. The post-progression survival (PPS) was measured from the date of tumor progression after second-line osimertinib until death or censored at the last follow-up. The real-world overall survival (rwOS) was determined from the first-generation EGFR-TKI treatment initiation to the date of death or last follow-up visit. The last follow-up occurred on October 21, 2023.

Statistical analysis

Statistical analysis was performed with SPSS statistical software, version 24.0(IBM-SPSS, Inc., Chicago, IL, USA). Pearson χ², Fisher’s exact test or the chi-square test were used for comparison of clinical features. Kaplan-Meier survival analysis and log-rank test were used to compare 1LrwPFS, 2LrwPFS, PPS and rwOS. Linear regression and Spearman rank correlation analysis were conducted to evaluate the correlations of 1LrwPFS, 2LrwPFS and PPS with rwOS. Hazard ratio (HR) and the corresponding 95% confidence intervals (CIs) were calculated by the Cox proportional hazards model. Statistical significance was defined as two-sided P < 0.05.

Results

Patient characteristics

A total of 3015 patients were retrospectively screened at Shanghai Chest Hospital and 126 patients were enrolled in this study according to the inclusion criteria. 62 patients and 64 patients were grouped into M1c1 group and M1c2 group. In this study, the characteristics of the two groups were well balanced (Table 1).

Table 1.

Demographic data of patients and comparison of baseline characteristics between two groups

Characteristics	M1c1 group(n = 62)	M1c2 group(n = 64)	P value	Total
Median age(range)	60.5(35–79)	59.5(37–79)	0.163	60.0(35–79)
Gender
Female	35(56.5%)	41(64.1%)	0.383	76(60.3%)
Male	27(43.5%)	23(35.9%)		50(39.7%)
Age group
<60 year	30(48.3%)	33(51.6%)	0.722	63(50.0%)
≥ 60 year	32(51.7%)	31(48.4%)		63(50.0%)
Smoking history
Yes	26(41.9%)	23(35.9%)	0.490	49(38.9%)
No	36(58.1%)	41(64.1%)		77(61.1%)
Lung lesion location
peripheral type	55(88.7%)	57(89.1%)	0.950	112(88.9%)
central type	7 (11.3%)	7(10.9%)		14(11.1%)
T stage
T1-2	41(66.1%)	35(54.7%)	0.207	76(60.3%)
T3-4	21(33.9%)	29(45.3%)		50(39.7%)
N stage
N0-1	7(11.3%)	5(7.8%)	0.557	12(9.5%)
N2-3	55(88.7%)	59(92.2%)		114(90.5%)
PS
0–1	54(87.1%)	56(87.5%)	0.946	110(87.3%)
≥ 2	8(12.9%)	8(12.5%)		16(12.7%)
EGFR mutation status
19deletion	42(67.7%)	37(57.8%)	0.249	79(62.7%)
21L858R	20(32.3%)	27(42.2%)		47(37.3%)
TP53 alteration
Yes	22(35.5%)	16(25.0%)	0.245	38(30.2%)
No	40(64.5%)	48(75.0%)		88(69.8%)
LDH(U/L)
≤ 250	47(75.8%)	53(82.8%)	0.331	100(79.4%)
>250	15(24.2%)	11(17.2%)		26(20.6%)
AKP(U/L)
≤ 120	38(61.3%)	44(68.8%)	0.380	82(65.1%)
>120	24(38.7%)	20(31.2%)		44(34.9%)
Calcium (mmol/L)
2.1–2.5	53(85.5%)	55(86.0%)	0.910	108(85.7%)
>2.5	1(1.6%)	2(3.1%)		3(2.4%)
<2.1	3(4.8%)	2(3.1%)		5(4.0%)
Missing	5(8.1%)	5(7.8%)		10(7.9%)
Spinal bone metastases
Yes	47(75.8%)	50(78.1%)	0.757	97(77.0%)
No	15(24.2%)	14(21.9%)		29(23.0%)
Nature of bone metastases
Osteogenic bone metastases	3(4.8%)	4(6.3%)	0.942	7(5.6%)
Osteolytic bone metastases	54(87.1%)	55(85.9%)		109(86.5%)
Both	5(8.1%)	5(7.8%)		10(7.9%)
Bisphosphonate was used regularly for at least 9 months
Yes	23(37.1%)	29(45.3%)	0.349	52(41.3%)
No	39(62.9%)	35(54.7%)		74(58.7%)
Radiotherapy for bone metastases
Yes	15(24.2%)	24(37.5%)	0.106	39(31.0%)
No	47(75.8%)	40(62.5%)		87(69.0%)

Abbreviations EGFR, epidermal growth factor receptor; LDH, Lactate dehydrogenase; AKP, Alkaline phosphatase; EGFR-TKI, Epidermal growth factor receptor tyrosine kinase inhibitor; PS, performance status; M1c1 group, multiple extrathoracic metastases in bone; M1c2 group, multiple extrathoracic metastases in bone and simultaneous other extrathoracic organ

The metastatic organs of all patients, coexisting with initial bone metastases in this study, were lung (66/126,52.4%), brain (40/126,31.7%), pleura (30/126,23.8%), liver (16/126,12.7%), adrenal gland (7/126,5.6%), distant lymph nodes (7/126,5.6%), pericardium (6/126,4.8%) and nodules of subcutaneous metastasis (2/126,1.6%) (Fig. 2).

Fig. 2.

Overview of metastatic organs of all patients

Comparison of survival outcome of two groups

The median follow-up time for all patients was 44.5 months (95%CI 37.9–51.1 months). The overall median 1LrwPFS was 13.9 months (95% CI 12.653–15.147), 2LrwPFS was 14.5 months (95% CI 11.665–17.335) and rwOS was 40.1 months (95% CI 35.996–44.204). Compared with patients in M1c1 group, patients in M1c2 group had worse rwOS (35.2 months vs. 42.9 months, HR = 0.512,95% CI 0.320–0.821, P = 0.005) (Fig. 3A) and shorter 2LrwPFS (12.8months vs. 17.0 months, HR = 0.575,95% CI 0.374–0.881, P = 0.011) (Fig. 3B). There was no statistically significant difference in 1LrwPFS (12.7months vs. 14.0months, HR = 0.838,95% CI 0.587–1.198, P = 0.333) and PPS (10.6 months vs. 6.2months, HR = 0.731,95% CI 0.475–1.123, P = 0.152) between M1c1 group and M1c2 group.

Fig. 3.

Survival analysis and efficacy analysis. (A) Kaplan–Meier curve of rwOS; (B) Kaplan–Meier curve of 2LrwPFS; (C) The best response outcome in M1c1 group; (D) The best response outcome in M1c2 group; (E) Objective response rate (ORR) and disease control rate (DCR) in M1c1 group and M1c2 group. Abbreviations rwOS, real-world overall survival; 2LrwPFS, second-line real-world progression-free survival; HR, hazard ratio; CI, confidence interval; ORR, objective response rate; DCR, disease control; M1c1 group, multiple extrathoracic metastases in bone; M1c2 group, multiple extrathoracic metastases in bone and simultaneous other extrathoracic organ

No CR was observed in M1c1 group and M1c2 group during third-generation EGFR-TKI treatment. The disease control rate (DCR) was 93.5% (58/62) and objective response rate (ORR) was 46.8% (29/62) in M1c1 group. The M1c2 group achieved a similar DCR (96.9%,62/64) and ORR (48.4%,31/64) (Fig. 3E). Detailed data for the best response outcome between M1c1 group and M1c2 group was shown in Figs. 3C and D.

Multivariate regression analysis on rwOS and 2LrwPFS

Based on the results of the univariate analysis, multivariate analysis used the Cox model to examine the impact of important clinical characteristics on rwOS and 2LrwPFS. In multivariate analysis, patients in M1c2 group (P = 0.002), performance status (PS) score ≥ 2 (P = 0.000) and TP53 alteration positive (P = 0.003) were independent prognostic factors of worse rwOS (Table 2). And patients in M1c2 group (P = 0.014) and PS score (P = 0.000) were related to shorter 2LrwPFS (Table 3).

Table 2.

Univariate and Multivariate analyses of clinical characteristics on real-world overall survival outcomes between two groups

	Univariate analysis on rwOS		Multivariate analysis on rwOS
Factors	HR (95%CI)	P	HR (95%CI)	P
M1c1/M1c2group	0.512 (0.320–0.821)	0.005*	0.472 (0.292–0.762)	0.002*
Gender (Female/Male)	1.098 (0.684–1.764)	0.698
Age group (<60/≥60 )	0.660 (0.419–1.041)	0.074
Smoking history (No/Yes)	1.285 (0.798–2.069)	0.302
Lung lesion location (peripheral/ central type)	0.766 (0.402–1.458)	0.417
T stage (T1-2/ T3-4)	0.598 (0.378–0.944)	0.027*	0.803 (0.498–1.293)	0.366
N stage (N0-1/ N2-3)	0.482 (0.193–1.205)	0.118
PS (0–1/≥2)	0.006 (0.001–0.026)	0.000*	0.005 (0.001–0.023)	0.000*
EGFR mutation status (19deletion/21L858R)	0.753 (0.473–1.198)	0.231
TP53 alteration (No/Yes)	0.592 (0.359–0.976)	0.040*	0.457 (0.274–0.762)	0.003*
LDH (U/L)(≤ 250/>250)	0.521 (0.308–0.882)	0.015*	0.753 (0.422–1.341)	0.335
AKP (U/L)(≤ 120/>120)	0.952 (0.591–1.533)	0.841
Spinal bone metastasis (No/ Yes)	1.001 (0.589–1.701)	0.998
Bisphosphonate was used regularly for at least 9 months (No/ Yes)	0989 (0.618–1.580)	0.962
Radiotherapy for bone metastases (No/Yes)	1.458 (0.879–2.419)	0.144
Initial brain metastases (No/ Yes)	0.638 (0.403–1.011)	0.055
Initial liver metastases (No/ Yes)	0.634 (0.331–1.211)	0.167

*P < 0.05

Abbreviations rwOS, real-world overall survival; PS, performance status; EGFR, epidermal growth factor receptor; LDH, Lactate dehydrogenase; AKP, Alkaline phosphatase; HR, hazard ratio; CI, confidence interval; M1c1 group, multiple extrathoracic metastases in bone; M1c2 group, multiple extrathoracic metastases in bone and simultaneous other extrathoracic organ

Table 3.

Univariate and Multivariate analyses of clinical characteristics on second-line real-world progression-free survival between two groups

Factors	Univariate analysis on 2LrwPFS		Multivariate analysis on 2LrwPFS
	HR (95%CI)	P	HR (95%CI)	P
M1c1/M1c2group	0.575 (0.374–0.881)	0.011*	0.582 (0.378–0.896)	0.014*
Gender (Female/Male)	0.944 (0.619–1.439)	0.788
Age group (<60/≥60 )	0.892 (0.588–1.353)	0.590
Smoking history (No/Yes)	1.148(0.746–1.767)	0.531
Lung lesion location (peripheral/ central type)	0.918 (0.487–1.729)	0.791
T stage (T1-2/ T3-4)	0.736(0.482–1.123)	0.155
N stage (N0-1/ N2-3)	0.587 (0.271–1.276)	0.179
PS (0–1/≥2)	0.215 (0.115-0.400)	0.000*	0.220(0.117–0.412)	0.000*
EGFR mutation status (19deletion/21L858R)	0.614 (0.403–0.937)	0.024*	0.695 (0.453–1.066)	0.095
TP53 alteration (No/Yes)	0.720 (0.458–1.134)	0.156
LDH (U/L)(≤ 250/>250)	0.680 (0.409–1.130)	0.136
AKP (U/L)(≤ 120/>120)	0.924 (0.595–1.436)	0.726
Spinal bone metastases (No/ Yes)	0.863 (0.528–1.410)	0.556
Bisphosphonate was used regularly for at least 9 months (No/ Yes)	0.867 (0.567–1.325)	0.510
Radiotherapy for bone metastases (No/Yes)	1.178 (0.749–1.852)	0.479
Initial brain metastases (No/ Yes)	0.727 (0.476–1.111)	0.141
Initial liver metastases (No/ Yes)	0.829 (0.440–1.562)	0.563

*P < 0.05

Abbreviations 2LrwPFS, second-line real-world progression-free survival; PS, performance status; EGFR, epidermal growth factor receptor; LDH, Lactate dehydrogenase; AKP, Alkaline phosphatase; HR, hazard ratio; CI, confidence interval; M1c1 group, multiple extrathoracic metastases in bone; M1c2 group, multiple extrathoracic metastases in bone and simultaneous other extrathoracic organ

In addition, log-rank tests demonstrated that rwOS differed for patients according to TP53 alteration status and PS. 2LrwPFS was different for patients according to performance status. Patients with TP53 alteration negative had a median rwOS of 42.9months compared to 33.4 months for patients with TP53 alteration positive (log-rank test, P = 0.038, Fig. 4A). Patients with PS0-1 had a median rwOS of 41.8months compared to 18.1 months for patients with PS ≥ 2 (log-rank test, P = 0.000, Fig. 4B). Patients with PS0-1 had a median 2LrwPFS of 16.4months compared to 8.0 months for patients with PS ≥ 2 (log-rank test, P = 0.000, Fig. 4C).

Fig. 4.

Kaplan-Meier curve according to prognostic factors and correlations of 1LrwPFS, 2LrwPFS and PPS to rwOS. (A) Kaplan-Meier curve of rwOS according to TP53 alteration status; (B) Kaplan-Meier curve of rwOS according to PS; (C) Kaplan-Meier curve of 2LrwPFS according to PS; (D) Correlation between rwOS and 1LrwPFS; (E) Correlation between rwOS and 2LrwPFS; (F) Correlation between rwOS and PPS. r values represent Spearman’s rank correlation coefficients. R² values represent linear regression. Abbreviations rwOS, real-world overall survival; 1LrwPFS, first-line real-world progression-free survival;2LrwPFS, second-line real-world progression-free survival; PS, performance status; M1c1 group, multiple extrathoracic metastases in bone; M1c2 group, multiple bone metastases and simultaneous other extrathoracic metastases

Correlations of 1LrwPFS, 2LrwPFS and PPS to rwOS

In this study, we assessed the associations of 1LrwPFS, 2LrwPFS and PPS to rwOS using patient-level data. Linear regression and Spearman rank correlation analysis demonstrated that 2LrwPFS was moderately correlated with rwOS (r = 0.621, R² = 0.568, P = 0.000) (Fig. 4E), while 1LrwPFS (r = 0.421, R² = 0.210, P = 0.000) and PPS (r = 0.406, R² = 0.154, P = 0.000) was lowly correlated with rwOS (Figs. 4D, F). 2LrwPFS had a moderate impact on rwOS.

Subgroup analysis of patients in M1c1 and M1c2 group

Survival analysis of rwOS and 2LrwPFS showed the superiority of M1c1 group over M1c2 group. Subsequently, we performed subgroup analysis to determine the survival effect of different bone metastases patterns (M1c1 group and M1c2 group) in various subgroups. All subgroup showed that M1c2 group was inferior to M1c1 group in rwOS (Fig. 5A) and 2LrwPFS (Fig. 1B).

Fig. 5.

Forest plots showing subgroup analysis of EGFR mutant advanced lung adenocarcinoma patients with bone metastases. (A) subgroup analysis of rwOS; (B) survival analysis of 2LrwPFS. Abbreviations EGFR, epidermal growth factor receptor; LDH, Lactate dehydrogenase; AKP, Alkaline phosphatase; PS, performance status; HR, hazard ratio; CI, confidence interval; M1c1 group, multiple extrathoracic metastases in bone; M1c2 group, multiple bone metastases and simultaneous other extrathoracic metastases

Discussion

In this study, we assessed the prognostic performance of the 9th edition of the TNM staging system among M1c stage lung adenocarcinoma patients with bone metastases. And we provided some real-world application data of the updated classification system and real-world survival outcome for EGFR-mutant advanced lung adenocarcinoma patients with bone metastases in M1c stage. A more refined stratification of M1c1 and M1c2 was conducted in EGFR-mutant advanced lung adenocarcinoma patients with bone metastases who receiving the third-generation EGFR-TKI as the second-line treatment after obtaining the first-line first-generation EGFR-TKI resistance with acquired EGFR T790M mutation. Our research revealed that patients in M1c1 and M1c2 stage had different survival outcome. Patients in M1c2 stage was inferior to M1c1 stage patients in rwOS (35.2 months vs. 42.9 months, HR = 0.512, P = 0.005), which suggested that more refined stratification of M1c can be conducive to judge the prognosis of patients.

The overall median rwOS was 40.1 months (95% CI 35.996–44.204), 1LrwPFS was 13.9 months (95% CI 12.653–15.147) and 2LrwPFS was 14.5 months (95% CI 11.665–17.335) for all patients. Real-world data reported from previous researches showed similar survival in terms of rwOS ,1LrwPFS and 2LrwPFS for advanced lung adenocarcinoma patients receiving the first-line first-generation EGFR-TKI and EGFR T790M mutation guided second-line Osimertinib [28–35]. For EGFR T790M-negative patients, the treatment options after the first-line EGFR-TKI failure were limited, and platinum based chemotherapy remained the main treatment with overall survival of approximately 20.0 months [36, 37]. In addition, patients in M1c2 stage had shorter 2LrwPFS (12.8months vs. 17.0 months, HR = 0.575, P = 0.011), compared with M1c1 stage patients. All subgroup analysis revealed that patients in M1c2 stage was inferior to patients in M1c1 stage in rwOS and 2LrwPFS. Multivariate analysis showed PS score ≥ 2 and TP53 alteration positive were independent prognostic factors of worse rwOS, and PS score ≥ 2 was related to shorter 2LrwPFS, which were consistent with the report of former researches [34, 38–40]. This study showed that 2LrwPFS was moderately correlated with rwOS (r = 0.621, R² = 0.568, P = 0.000), which provided some potentially valuable clues and might be prudent to consider how to improve 2LrwPFS and rwOS.

However, the reasons of worse prognosis for patients in M1c2 stage are not yet fully revealed. The possible reasons we speculated were as follows. First of all, simultaneous bone metastases and multi-organ metastases indicated high tumor load, high proliferation, migration, and invasion capacities of tumor cells, which were linked with poor survival. Second, with the existence of genetic heterogeneity in tumor sites, there may be different gene mutation status between primary tumor and metastatic sites, possibly resulting in divergent responses to therapies [41, 42]. The gene mutation status of patients in real-world setting was usually determined by primary tumor biopsy, resulting in the loss of comprehensive genetic information to some extent. Previous research revealed that biopsies from both primary tumor and metastatic sites gained additional genetic assessment information that was known as clinically significant mutations or with experimental drug targets, which helped clinicians to make a decision on personalized treatment and improve survival outcome for patients [43]. Third, the tumor microenvironment (TME) of tumor lesions might be changed during treatment with EGFR-TKI, compared with the status of initial lesion including primary tumor and metastatic sites, resulting in drug resistance and a poorer survival [44–46]. Last, what’s more important was that drug resistance mechanisms to EGFR-TKI were potentially heterogeneous and multiple coexisting among primary tumor sites and metastatic sites, which might lead to different therapeutic response to subsequent anti-tumor therapy [47–50].

Undeniably, some limitations existed in this study. First, it was a single-center retrospective study which would result in some biases, such as selection bias. Second, this research was conducted within the context of routine medical practice in real-world setting. The evaluation of disease progression reflected physician’s judgments and influenced by individual difference among physicians, which might lead to some upward bias in progression-free survival in contrast to rigorous clinical trials. Third, the sample size of this study was small. Large-scale prospective multi-center studies will be necessary to validate and further expand the current findings.

Conclusions

In summary, more refined stratification of M1c according to the 9th edition of the TNM staging system has advantages to judge the prognosis of lung adenocarcinoma patients with bone metastases and help to explore and select more precise treatment. And future clinical studies in larger cohorts are expected to validate additional findings that subsequent third-generation EGFR-TKI treatment for disease progression after the first-line first-generation EGFR-TKI treatment had a significant impact on overall survival time, which might provide potentially valuable clues of considering how to improve PFS of the second-line treatment and thus OS.

Abbreviations

NSCLC

Non-small cell lung cancer

EGFR

Epidermal growth factor receptor

EGFR-TKI

Epidermal growth factor receptor tyrosine kinase inhibitor

M1c1 group

Multiple extrathoracic metastases in bone

M1c2 group

Multiple extrathoracic metastases in bone and simultaneous other extrathoracic organ

ARMS

Amplification refractory mutation system

NGS

Next-generation sequencing

CT

Computed tomography.

MRI

Magnetic resonance imaging

SPECT

Single-photon emission computed tomography

PET-CT

Positron emission tomography computed tomography

RECIST

Response Evaluation Criteria in Solid Tumors

CR

Complete response

PR

Partial response

SD

Stable disease

PD

Progressive disease

ORR

Objective response rate

DCR

Disease control

LDH

Lactate dehydrogenase

AKP

Alkaline phosphatase

PS

Performance status

1LrwPFS

First-line real-world progression-free survival

2LrwPFS

Second-line real-world progression-free survival

PPS

Post-progression survival

rwOS

Real-world overall survival

HR

Hazard ratio

CI

Confidence intervals

TME

Tumor microenvironment

WCLC

The World Conference on Lung Cancer

Author contributions

Conceptualization: J P, F H, L J; Methodology: J P, F H; Formal analysis and investigation: J P, F H, X M，Y N, M M; Writing - original draft preparation: J P, F H; Writing - review and editing: J P, L J; Funding acquisition: L J; Resources: Y N, M M, L J; Supervision: L J.Final approval of manuscript: All authors.

Funding

This research was funded by the National Natural Science Foundation of China, grant number 82150005 and the Chinese Society of Clinical Oncology Tumor Research Fund, grant number Y-2019AZZD-0038.

Data availability

Data is provided within the manuscript.The data of this study are available from the corresponding author and first author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was conducted in accordance with the Declaration of Helsinki, and approved by the Institutional Review Board of Shanghai Chest Hospital (protocol code IS22015 on 22 February 2022). Informed consent was obtained from all patients involved in the study before the collection of information.

Consent for publication

Informed consent was obtained from all patients involved in the study.

Competing interests

The authors declare no competing interests.