Introduction

Lung cancer remains the second most common tumor in the world1. Since 2011, as cancer screening by low-dose computed tomography (LDCT) has been shown to reduce lung cancer mortality, LDCT has been extensively utilized, contributing to more and more small pulmonary nodules are detected. Nevertheless, the high false-positive rate of LDCT has sparked controversies regarding its usage, as it may induce patient anxiety, overdiagnosis, and radiation-induced lung cancer deaths2,3. Overdiagnosis and overtreatment are prevalent in ground-glass lung adenocarcinoma nodules4. Despite several guidelines have put forth management strategies of ground-glass nodules (GGNs) based on its growth law, there still lacks sufficient evidence and remains debated5,6,7. As persistent GGNs likely represent early lung adenocarcinoma or precancerous lesions, surgical resection is considered to be the most effective treatment approach. However, early nodule resection may lead to the deterioration of lung function and quality of life, which may hinder the resectability of the second more aggressive lung cancer4. In order to avoid additional risks caused by invasive procedure, conservative management is commonly recommended for slow-growing GGNs. And active surveillance (AS) based on LDCT is often employed to assess the malignancy risk of GGNs during follow-up8.

Previous study has shown that the consolidation tumor ratio (CTR) serves as a prognostic factor for GGNs, wherein a GGN with CTR ≤ 0.5, also referred to as predominantly ground-glass nodules, exhibits a more favorable prognosis compared to a GGN with CTR > 0.59. When both the tumor diameter ≤ 3 cm and CTR ≤ 0.5 are employed as radiological diagnostic criteria, the specificity for diagnosing low-invasive lung cancer (without lymph node involvement and vascular invasion) reaches 100%10. When lung cancer is suspected, concerning about cancer progressing, solid nodules and mixed ground glass nodules (mGGN) with CTR > 0.5 are usually recommended for surgical resection6. However, it is still controversial whether GGN with CTR ≤ 0.5 is active surgery or follow-up observation. Considering the risk of tumor progression and the psychological and economic burden brought by longstanding tumors and lung cancer screening, the feasibility of follow-up observation in incidentally detected GGNs with tumor diameters ≤ 3 cm and CTR ≤ 0.5 remains unknown. To our knowledge, only one study has compared the pathological outcomes of delayed resection after active surveillance and early resection, revealing no significant differences between the two choices of timing of surgery11. And to date, no previous study compared long-term outcomes of the two management strategies.

Therefore, the objective of this study is to compare the pathological results and long-term survival results of early surgery and delayed surgery after at least one year of follow-up for ground‐glass component predominant lung adenocarcinoma patients, and to explore the feasibility of follow-up monitoring for patients with lung adenocarcinomas exhibiting predominant ground-glass nodules.

Materials and methods

Patients

This retrospective study was approved by the Medical Ethics Committee of the First Medical Center of the Chinese PLA General Hospital and was conducted in accordance with the Declaration of Helsinki. (approval number: S2022-177-01). The requirement for informed consent was waived by the Medical Ethics Committee of the First Medical Center of the Chinese PLA General Hospital owing to the retrospective nature of this study. Patients who underwent surgical resection for lung adenocarcinoma from January 1, 2013 to August 31, 2017 were enrolled in the study.

The inclusion criteria were as follows: (1) The histopathological examination confirmed the diagnosis of minimal invasive adenocarcinoma (MIA) or invasive adenocarcinoma (IAC). (2) Thin-slice thoracic CT (section thickness 1–1.25 mm) within one month before operation. (3) Lesions was less than 3 cm and manifested as pGGNs or mGGNs with CTR ≤ 0.5 on CT at initial screening. (4) After a GGN was detected, surgical resection performed within three months or after follow-up longer than one year. Exclusion criteria included: (1) CT image quality was undiagnosable. (2) The pathological diagnosis was mucinous adenocarcinoma. (3) Patients had received anti-tumor chemotherapy prior to surgery. (4) Previous history of lung malignancy. (5) The preoperative follow-up was discontinuous. The specific recruitment process was illustrated in Fig. 1. According to the criteria, 279 patients were finally included, with 210 patients assigned to the early surgery group (ES group) and 69 patients assigned to the surgery after follow-up group (FS group).

Figure 1
figure 1

Flowchart of cases screening.

Full size image

Imaging protocol

The CT scans were performed using either Philips Brilliance 256i CT (Netherland) or GE Optima CT660 (USA). All patients were instructed to receive chest CT scans during an inspiratory breath-hold in the supine position. The scanning range was from lung apices to costo-phrenic angles. CT scanning parameters were as follows: voltage 120 kV, automatic tube current, pitch, 0.993, image matrix, 512 × 512; the reconstructed slice thickness, 1.00 mm or 1.25 mm. All images were analyzed in lung window setting (width, 1500 HU; level, − 600 HU).

Clinical characteristics and imaging features

Clinical characteristics were collected through electronic medical record system, including age, gender, smoking history, surgical procedure, and postoperative histopathological results. The cases were divided into those of minimally invasive adenocarcinoma (MIA) and invasive adenocarcinoma (IAC). According to the largest percentage of each histological component, the subtypes of invasive lung adenocarcinoma were divided into lepidic predominant, acinar predominant, papillary predominant, micropapillary predominant or solid predominant. Two experienced chest radiologists (> 10 years of experience)), who were blinded to the pathological results, independently reviewed the CT images. Inconsistent results were resolved by consultation. The CT imaging features include GGN size, types (pGGN or mGGN), the size of the solid component of nodules, lobar location and CTR. GGN size was defined as the largest diameter of the nodule in the sagittal, coronal, and axial planes. The size of the solid components measured using the longest diameter in three dimensions. And CTR was defined as the ratio of the maximum diameter of consolidation to the maximum diameter of the tumor.

Follow-up and study outcomes

The postoperative follow-up time was determined from the date of surgery, and terminated on August 31, 2022. The follow-up methods included outpatient, reexamination, and telephone follow-up. The recurrence and metastasis were defined based on imaging results and pathological examination. Overall survival (OS) was defined as the duration of time from the date of the surgery to the date of death due to any cause or the last follow-up. Disease-free survival (DFS) was defined as the time from the date of surgery to disease recurrence, metastasis, or death from any cause. The preoperative follow-up time was defined as the interval between the initial and final CT scans before surgery.

Statistical methods

Continuous variables were reported as either medians (with interquartile range [IQR]) or means (standard deviation) and compared using the t-test or Wilcoxon rank-sum test, depending on whether the continuous variable was normally distributed. Categorical variables were reported as frequencies and percentages and compared by χ2 test or Fisher’s exact. The propensity score matching (PSM) was calculated using a logistic regression model, including the following covariates: nodule size and CTR. Patients were matched with a caliper definition of 0.02 and a match ratio of 1:1. A two-sided α of less than 0.05 was considered statistically significant. The OS and DFS curves were estimated using the Kaplan–Meier method. All statistical analyses were done using the SPSS software, version 26.0 (IBM SPSS Statistics Version 26.0, IBM Corp. https://www.ibm.com/products/spss-statistics).

Results

The FS group had a mean preoperative follow-up time of 1146.3 ± 661.3 days, ranging from 367 to 3136 days. These patients underwent an average of 4.5 ± 2.4 CT examinations, ranging from 2 to 13. The ES group had an average time from the initial CT examination to surgery of 10.1 ± 8.1 days, ranging from 0 to 47 days. Demographic and clinical characteristics of patients were shown in Table 1. 210 (75.27%) patients were included in the ES group, the median (IQR) postoperative follow-up time was 81 (65–92) months. And 69 (24.73%) patients were in FS group with 63 (57–75) months median post-operative follow-up time. There were no significant differences in age, gender, smoking history of patients, the solid component size, CTR and lobar location of nodules in two groups. The GGN size in ES group were larger than those in FS group (17.8 ± 5.8 vs. 13.3 ± 4.4; P < 0.001). And lobectomy was more frequently performed in ES group (73.3% vs. 53.6%; P = 0.003). There was a difference (P = 0.030) in pathological subtype between the two groups and lepidic predominant adenocarcinoma was more common in FS group. After PSM, baseline characteristics of the two groups of patients were balanced (Table 2).

Table 1 Baseline characteristics of patients.
Full size table
Table 2 Baseline characteristics of patients after propensity score matching.
Full size table

The specific CT data comparison before and after follow-up in FS group is shown in Table 3. In group FS, 22 GGNs (31.9%) remained stable and 47 GGNs (68.1%) showed growth with solid component growth in 23 GGNs and glass-ground-opacity component growth in 24 GGNs, and the median growth time were 24 months. The solid component growth GGN had more aggressive pathological results (P = 0.033). This is described in more detail in Table 4.

Table 3 Comparison of CT data before and after follow-up in FS group.
Full size table
Table 4 Comparison of the characteristics between the solid components’ growth group and the solid components’ non-growth group.
Full size table

There were no significant differences in the recurrence rate (0.95% vs 1.4%) and death rate (0.5% vs 1.4%) between the two groups. The long-term survival outcomes in the two groups after PSM are illustrated in Fig. 2. The Kaplan–Meier (KM) curves indicate no significant differences in OS and RFS between the two groups. Among the death cases, one case died in ES group due to postoperative metastasis of lung cancer, and another case died in FS group due to gastrointestinal bleeding. Postoperative recurrence occurred in 3 patients. In ES group, there were 2 patients. One patient relapsed 18 months after operation and died 1 months later, while the other patient relapsed 41 months later, and was still alive by the time of follow-up. In FS group, 1 case recurred for 20 months and survived after 62 months of follow-up.

Figure 2
figure 2

Recurrence-free and overall survival curves between ES group and FS group after PSM.

Full size image

Discussion

In the present study, we compared the pathological subtypes and prognosis of lung adenocarcinoma between the ES group and the FS group, which was similar in both groups.

Currently, the management strategies of GGN were still controversial. Because lung adenocarcinoma has a long incubation period from obtaining the mutation of driving oncogene to the date of clinical diagnosis of lung cancer. They remain static for a long time before obtaining secondary mutation, and then evolve into clinical invasive lung cancer12. PET/CT was generally believed valuable for preoperative diagnosis and prognosis judgment of lung cancer. However, it was shown not effective in preoperative detection of lymph nodes or distant metastasis in patients with mGGN with solid part size of 3 cm or less13. CT-guided transthoracic needle biopsy is a very common method for diagnosing lung masses. However, a previous study revealed that the diagnosis accuracy only ranged from 35.2 to 80.0% of GGNs of less than 2 cm due to its rare solid components14. Owing to the limited tissue amount obtained by puncture, the pathological results of puncture of some nodules may be different from those of surgery15. Meanwhile, although puncture biopsy is a minimally invasive operation, complications are still possible after operation. Therefore, the management of GGNs is mainly based on regular follow-up and surgical intervention. American College of Chest Physicians considered GGNs are often malignant if they are growing or developing solid components and prompted further evaluation and/or consideration of resection6. The Fleischner Society is suggested that individualized regular follow-up plans should be made according to the size of nodules and solid components5. For indeterminate pulmonary nodules, an American survey study among 419 experienced pulmonologists suggested that pulmonologists would make different decisions according to the age, smoking history and nodule size16. There are also differences between different doctors. Some want to actively intervene, while others suggest conservative follow-up.

Whether the long-term follow-up of GGNs is safe, such as increase pathological invasiveness and reduce survival time, is still unknown. To the best of our knowledge, this retrospective cohort study is the first time to compare long term prognosis between the early surgery group and the delayed surgery after follow-up group. The follow-up time of this study is long, and there is no significant difference in DFS or OS between the early resection group and the follow-up group during the follow-up period. Previous clinical studies have proved that ground-glass nodular lung adenocarcinomas had a good prognosis. The Multicenter Italian Lung Detection (MILD) screening trial revealed the long-term prognosis of unresected mGGNs with a solid component less than 5 mm through conservative management. During the long-term follow-up of 9.3 ± 1.2 years, none of the 2303 subjects died of lung cancer caused by mGGNs4. Hiroyuki et al. analyzed the long-term prognosis of GGN lung adenocarcinoma with CTR ≤ 0.5, and found the 10-year overall survival rate and recurrence-free survival rate were 93.1% and 90.5% respectively8. This study also had a high survival rate in follow-up group and demonstrated that the prognosis of follow-up group is not worse than that of early operation group.

Researchers generally believe that the development of solid components in GGNs resulted in an increase of the probability of GGN malignancy. A study focusing on postoperative pathology of GGNs shows that pathological classification in growth group is more invasive than that in non-growth group11. In the present study, the pathology results remained consistent with the previous study. However, the prognosis of the two subgroups were no statistical difference. Although GGNs were inert, it still has the potential for malignant transformation. This may explain why GGNs in growth group were more invasive.

Histologic subtypes been shown to be closely correlated with prognosis of lung adenocarcinoma. Among the three patients with recurrence in this study, the focus of first patient contained solid components, and second patient contained micropapillary components, which recurred 41 months and 20 months after operation, respectively. According to the prognosis of patients, the International Association for the Study of Lung Cancer pathology panel had established a grading system for invasive pulmonary adenocarcinoma and placed any tumor with 20% or more of high-grade patterns (solid, micropapillary, or complex gland) into grade 3, which indicated poor prognosis17. Because the pathological diagnosis results in electronic medical record system did not contain detailed data, only the predominant subtypes were recorded in this study, and new grading system was not adopted. Nevertheless, in this study, the proportion of poor prognosis in cases with high-grade pattern’s subtype is obviously higher than that in the whole. Therefore, although the evidence is not sufficient, we still suggest that those patients wiht poorly differentiated nodules should be actively followed up after surgery, even though CTR is less than 0.5 (Supplementry material).

There remain some limitations in this study. Firstly, this study is a small-sample retrospective investigation. In the future, we plan to design a larger-scale prospective study to further validate our findings. Secondly, the lesions of patients in the early surgery group were generally considered to have more aggressive imaging signs, which may have brought a selection bias. Thirdly, due to mental strain of patients, they cannot tolerate long-term follow-up for GGN, even the nodule was stable. Therefore, in this study, all patients with a follow‐up period longer than one year were included in the follow-up group. Fourthly, some of the persistent ground glass nodules prove to be benign18. This part of the patients was not enrolled in the present study. Future real-world research may further explore the relationship between surgical resection and follow-up survival outcome of screen-detected GGN patients, not only limited to cancers. Finally, because of the indolent characteristics of ground-glass nodular lung adenocarcinoma, there were almost no tumor recurrence or death in both groups, which largely reduced the value of evidence. Therefore, future studies with larger sample size and longer follow-up period are warranted.

In conclusion, pathological subtypes of lung adenocarcinoma were similar between the early surgery group and the delayed surgery after follow-up group. And the outcomes in the two groups were close and excellent. One-year follow-up was proved to be a feasible and appropriate management method for ground‐glass component predominant lung adenocarcinoma patients.