Comparison of Surveillance with Low-Dose and Contrast-Enhanced Chest Computed Tomography in Patients Disease-Free for 2 Years after Curative Resection for Lung Cancer

Bubse Na; Ji Hyeon Park; Kwon Joong Na; Samina Park; Chang Hyun Kang; Young Tae Kim; In Kyu Park

doi:10.4143/crt.2025.256

Articles

Page Path: HOME > Cancer Res Treat > Volume 58(2); 2026 > Article

Original Article Lung and Thoracic cancer Comparison of Surveillance with Low-Dose and Contrast-Enhanced Chest Computed Tomography in Patients Disease-Free for 2 Years after Curative Resection for Lung Cancer: Bubse Na¹, Ji Hyeon Park¹, Kwon Joong Na^1,2, Samina Park¹, Chang Hyun Kang¹, Young Tae Kim^1,2, In Kyu Park¹; Cancer Research and Treatment : Official Journal of Korean Cancer Association 2026;58(2):454-464.
DOI: https://doi.org/10.4143/crt.2025.256
Published online: June 5, 2025

¹Department of Thoracic and Cardiovascular Surgery, Seoul National University Hospital, Seoul National University College of Medicine, Seoul, Korea

²Cancer Research Institute, Seoul National University College of Medicine, Seoul, Korea

Correspondence: In Kyu Park, Department of Thoracic and Cardiovascular Surgery, Seoul National University Hospital, Seoul National University College of Medicine, 101 Daehak-ro, Jongno-gu, Seoul 03080, Korea
Tel: 82-2-2072-2342 E-mail: ikpark@snu.ac.kr

• Received: March 5, 2025 • Accepted: June 4, 2025

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

1,545 Views
60 Download

prev next

Full Article

Download PDF

Abstract

Purpose
Low-dose chest computed tomography (LDCT) is recommended for surveillance 2–3 years after curative resection of non-small cell lung cancer (NSCLC); however, supporting clinical evidence is limited. This study compared LDCT with contrast-enhanced chest computed tomography (CECT) in terms of recurrence detection and overall survival (OS) in patients 2 years after curative resection of NSCLC.
Materials and Methods
Among patients who underwent curative resection for NSCLC between January 2011 and December 2017 and survived for 2 years without recurrence, 2,083 patients were included. Comparisons between the LDCT and CECT groups were performed in both the entire cohort and propensity score-matched cohort. The primary outcome was the difference in overall survival. Secondary outcomes included time-to-recurrence, recurrence-free survival, and post-recurrence survival in each group.
Results
In the propensity score-matched population, the 5-year OS (96.0% for LDCT, 98.0% for CECT, p=0.097) and recurrence-free survival (RFS) (95.4% for LDCT, 96.0% for CECT, p=0.761) did not differ. The OS and RFS did not differ in subgroup analyses stratified by pathologic stage and histologic type. In the competing risk analysis, the overall 5-year cumulative incidence of recurrence did not differ between the two groups (4.56% for LDCT, 3.93% for CECT, p=0.765). When stratified by pathologic stage and histologic type, there was no significant difference in the cumulative incidence of recurrence. The distribution of recurrence sites did not differ between groups.
Conclusion
Similar OS and RFS were observed in LDCT and CECT surveillance in patients who achieved a 2-year disease-free status after curative resection for NSCLC.
Key words: Non-small-cell lung carcinoma, Practice guidelines as topic, Postoperative period, X-Ray computed tomography

Introduction

Postoperative surveillance after curative resection of non–small cell lung cancer (NSCLC) aims to detect recurrence and second primary lung cancers (SPLCs) to provide appropriate treatments that can prolong long-term survival. However, excessive surveillance can lead to wastage of resources, unnecessary secondary work-ups, overtreatment, and psychological burden on patients. It is reasonable to conduct minimal surveillance if long-term survival is not compromised. Thus, it is crucial to strike an appropriate balance between intensive and less intensive surveillance, as supported by evidence. For instance, in colon cancer, some studies suggest that high-frequency intensive follow-up is beneficial for survival and post-recurrence treatment [1,2]. However, no prospective or large-scale studies have been conducted on this subject in lung cancer.

Chest computed tomography (CT) is a fundamental imaging modality for the postoperative surveillance of lung cancer, as thoracic sites contribute to over half of the initial recurrences [3]. Furthermore, chest CT scan encompasses upper abdominal organs such as liver and adrenal glands, which are common sites for lung cancer metastasis.

Guidelines generally recommend intensive surveillance with contrast-enhanced chest CT (CECT) during the first 2 years after curative resection. However, there are discrepancies among the surveillance guidelines after the first 2 years. Recommendations vary according to the duration of intensive surveillance (either 2 or 3 years) and the use of CT protocols (CECT, non-contrast-enhanced CT [NECT], or low-dose CT [LDCT]). The European Society for Medical Oncology recommends CECT with 6-month intervals for the first 2 years and annual chest CTs (unspecified type) thereafter [4]. The American Society of Clinical Oncology suggests CECT for the first 2 years, followed by LDCT after that period [5]. The National Comprehensive Cancer Network (NCCN) guidelines recommend chest CT with or without contrast enhancement every 6 months for the first 2-3 years, followed by annual LDCT for patients with stage I-II NSCLC after primary treatment. For patients with stage III cancer, the NCCN advises chest CT with or without contrast enhancement every 3-6 months for the first 3 years, then every 6 months for 2 more years [6].

As the NCCN guidelines are frequently revised, some changes have been made to CT protocols and the duration of intensive surveillance. In 2015, LDCT was introduced as a standard surveillance modality after the first 2 years after curative resection [7]. In 2018, the duration of intensive surveillance was extended from 2 to 2-3 years [8]. However, these changes were largely based on an expert consensus, as studies on the optimal follow-up duration and CT protocols, particularly the use of contrast, are scarce.

Therefore, we conducted this study to compare the effects of LDCT and CECT surveillance on the long-term survival of patients who achieved a 2-year disease-free status after curative resection for NSCLC.

Materials and Methods

1. Patients

We retrospectively analyzed 3,566 patients who underwent curative surgery for NSCLC between January 2011 and December 2017. The study population comprised patients who had undergone surgery at least 5 years previously to have a sufficient follow-up duration for the evaluation of recurrence. Exclusion criteria were follow-up loss, recurrence or SPLC within 24±2 months, salvage surgery, incomplete resection, stage IV cancer, previous lung cancer history, concomitant cancer in other organs, chronic kidney disease or contrast allergy precluding CECT, and inappropriate type of CT. Finally, 2,083 patients who were recurrence-free for 24 months after curative surgery for NSCLC were included (Fig. 1). Demographic, radiologic, and clinicopathologic data, including age, sex, smoking history, Eastern Cooperative Oncology Group performance status (ECOG PS), histologic type, pathologic stage, extent of pulmonary resection, comorbidities, pattern of recurrence, treatment of recurrence, and survival, were retrieved from electronic medical records. Pathologic staging was reviewed and updated to align with the 8th edition of the TNM staging system [9]. Based on smoking history, patients were classified as nonsmokers, ex-smokers, and current smokers. Resection extent was grouped into sublobar resection (wedge resection+segmentectomy), lobectomy (lobectomy+bilobectomy), and pneumonectomy. The histological type was classified as adenocarcinoma, squamous cell carcinoma, and others. Detailed surveillance data, such as the date of radiological examinations and outpatient clinic visits, were also collected. The last follow-up date for survival and recurrence was May 31, 2023.

2. Surveillance strategy

Patients were followed routinely in the outpatient clinic after surgery at intervals of 3-6 months for the initial 2 years and intervals of 6-12 months during subsequent years. The scanned area of the chest CT scan extended from the lower poles of the thyroid gland to the upper abdomen and covered both adrenal glands. If patients complained of symptoms suggesting recurrence, closer follow-up and additional radiologic examinations such as ¹⁸F-fluoro-2-deoxyglucose positron emission tomography/computed tomography (PET/CT), magnetic resonance imaging (MRI), or bone scans were conducted. The choice between LDCT and CECT during surveillance was made at the discretion of the surgeon. Regular surveillance was performed until the postoperative 5th year, and further surveillance was performed at the discretion of each surgeon.

3. Diagnosis of recurrence

Patients with newly developed lesions or lesions of increasing size were suspected of having recurrence. Those with swollen lymph nodes compared to the size in previous CT scans or those with newly visible nodes on the CT scan were also suspected of recurrence. Any patients who were suspected of recurrence or SPLC were clinically judged by the Martini-Melamed criteria, and these cases were further discussed by multidisciplinary lung cancer experts if pathologic differentiation was not available. Patients with suspected lesions underwent PET/CT to determine whether the suspected lesion had an increased metabolic response that indicated the presence of a tumor, or they underwent biopsy to confirm the precise diagnosis. The sites of recurrence were classified as local, regional, or distant. Recurrence at the resection margin (stapler line, bronchial stump) was considered as local recurrence. Recurrences in the mediastinal or hilar lymph nodes, ipsilateral lung, or pleura were considered as regional recurrences. Recurrences in the supraclavicular lymph nodes, contralateral lung or pleura, bone, brain, non-regional thoracic lymph nodes, soft tissue, or abdominal organs were considered as distant recurrences. The type of modality in which recurrence was first noted was recorded.

4. Outcome measurement

The primary outcome measure was the overall survival (OS) rate. The secondary outcomes were recurrence-free survival (RFS), time-to-recurrence (TTR), and post-recurrence survival (PRS). The duration of OS was measured from the date of surgery to the date of death or last follow-up. The date of recurrence was defined as the date on which the disease was suspected. The RFS durations were measured from the date of surgery to the date of recurrence or death. The TTR was measured from the date of surgery to the date of recurrence. The PRS duration was measured from the date of recurrence to the date of death or last follow-up.

5. Statistical analysis

Patients were divided into two groups based on the type of chest CT used. The LDCT group comprised patients who underwent LDCT only rather than CECT from postoperative 24±2 months to appearance of lesion suspected as recurrence or end of surveillance. The CECT group comprised patients who underwent CECT at least once from postoperative 24±2 months until appearance of lesion suspected as recurrence was noted or end of surveillance. Patients with recurrences identified by radiological examinations other than chest CT, such as abdominopelvic CT, brain MRI, brain CT, or liver sonography, were categorized into the LDCT or CECT group according to the same classification method used for recurrences identified by chest CT, as previously mentioned.

Continuous variables were expressed as mean±standard deviation or median with interquartile range (IQR). Student t tests were used to compare continuous variables. Categorical variables were presented as numbers and proportions and compared using Pearson chi-squared test, Fisher exact test, and linear-by-linear association.

The Kaplan-Meier method and log-rank test were used to estimate and compare OS, RFS, and PRS between the groups.

Cox proportional hazards models were used for univariable and multivariable analyses of risk factors for OS and PRS in all pre-matched patients. Variables with a p-value less than 0.1 in the univariable analysis were included in the multivariable analysis with backward elimination of variables based on the likelihood ratio test.

As SPLC was expected to occur in a non-negligible number of cases, a competing risk analysis using the Fine-Gray sub-distribution hazard model was conducted, considering SPLC as a competing risk. Predictors of recurrence in all pre-matched patients were analyzed using Fine-Gray competing risk regression. Statistical significance was set at p < 0.05.

6. Propensity score matching analysis

In the propensity score matching analysis, age, sex, year of operation, histologic type, extent of resection, complete resection, and pathologic stage were used as covariates, and the propensity scores were matched one-to-one. We used the nearest-neighbor method without replacement, with a 0.15 caliper width.

All statistical analyses were performed using the R Software ver. 4.2.0 (R Foundation for Statistical Computing) and IBM SPSS ver. 26.0 (IBM Corp.).

Results

1. Entire population

The total number of patients was 2,083, with a median follow-up of 73.4 (IQR, 62.1 to 95.2) months. The mean patient age was 62.5±10.0 years, and there were 1,082 (51.9%) men and 1,001 (48.1%) women. There were 1,589 (76.3%) patients with pathologic stage I NSCLC; 283 (13.6%), stage II NSCLC; and 211 (10.1%), stage III NSCLC. The demographic characteristics of the patients in each group are summarized in Table 1.

The LDCT group comprised 466 patients (22.4%) with a median follow-up of 69.6 months (IQR, 61.3 to 89.8). The CECT group comprised 1,617 patients (77.6%) with a median follow-up of 75.3 months (IQR, 62.8 to 97.5).

Compared to the CECT group, the LDCT group had more patients with adenocarcinomas (78.8% vs. 75.2%, p < 0.001) and pathologic stage I NSCLC (89.7% vs. 72.4%, p < 0.001). Smoking history, year of surgery, and underlying comorbidities were not significantly different between the two groups (Table 1).

A total of 130 patients experienced recurrence, 83 (63.8%) of which were confirmed by biopsy. Recurrence in 31 (23.8%) patients was clinically diagnosed using PET/CT, without biopsy. Sixteen patients (12.3%) did not undergo a recurrence-confirming biopsy or PET/CT, but recurrence was confirmed using chest CT (n=13), brain MRI (n=1), and bone scans (n=2). There were 49 cases of SPLC detected during follow-up.

Among the 130 patients with recurrence, 90 (69.2%) had regional recurrence; 84 (64.6%), distant; and, five (3.8%), local recurrence. One hundred and seven patients (82.3%) had all their recurrence sites covered by thoracic CT. Fifty-eight patients (44.6%) had multiple recurrence sites in the area covered by thoracic CT. There was no significant difference in the proportion of recurrence sites between the LDCT and CECT groups (Table 2).

A total of 118 deaths (66 cancer-related, 33 noncancer-related, and 19 of unknown causes) occurred. Four deaths occurred at 120 months postoperatively.

No difference was found in the OS (5-year OS rate: 96.0% for LDCT vs. 97.6% for CECT, p=0.198) and RFS (5-year RFS rate: 95.4% for LDCT vs. 94.7% for CECT; p=0.272) between the two groups (S1 Fig.).

Univariable and multivariable analyses of risk factor for OS showed that age (hazard ratio [HR], 1.048; 95% confidence interval [CI], 1.026 to 1.071; p < 0.001), male sex (HR, 2.282; 95% CI, 1.498 to 3.475; p < 0.001), pathologic stage II versus I (HR, 2.670; 95% CI, 1.690 to 4.217; p < 0.001), pathologic stage III versus I (HR, 4.180; 95% CI, 2.639 to 6.622; p < 0.001), and previous cancer history (HR, 2.359; 95% CI, 1.529 to 3.640; p < 0.001) were significant poor risk factors. The type of surveillance CT was not associated with OS (p=0.199) (S2 Table).

In the competing risk regression analysis for recurrence, lobectomy versus sublobar resection (sub-distribution hazard ratio [sHR], 2.894; 95% CI, 1.230 to 6.810; p=0.015), pathologic stage II versus I (sHR, 5.057; 95% CI, 3.115 to 8.210; p < 0.001), pathologic stage III versus I (sHR, 6.940; 95% CI, 4.155 to 11.593; p < 0.001), and adenocarcinoma versus squamous cell carcinoma (sHR, 2.227; 95% CI, 1.271 to 3.906; p=0.005) were significant risk factors (S3 Table).

Five patients in the LDCT group (21.7%) and 10 patients in the CECT group (9.3%) did not receive treatment. Treatment for recurrence significantly improved PRS compared with no treatment in multivariable analysis (HR, 0.209; 95% CI, 0.104 to 0.419; p < 0.001). The type of surveillance CT was not associated with PRS in the multivariable analysis (S4 Table).

2. Propensity score matching analysis

After propensity score matching, a well-matched cohort of patients was obtained (S5 Table). Each group comprised 466 patients with variables that were not significantly different from each other (Table 1). Median follow-up durations of each group were 69.6 months (IQR, 61.3 to 89.8) in the LDCT group and 76.2 months (IQR, 62.7 to 86.0) in the CECT group. There were 23 and 22 cases of recurrence in the LDCT and CECT groups, respectively.

The OS did not differ between the LDCT and CECT groups (5-year OS rate, 96.0% for LDCT vs. 98.0% for CECT; p=0.097) (Fig. 2). In the subgroup analysis stratified by pathologic stage and histologic type, there was no difference in OS between the LDCT and CECT groups.

Additionally, RFS showed similar results. The 5-year RFS rates did not differ significantly between the LDCT (95.4%) and CECT (96.0%) groups (p=0.761). In the subgroup analysis stratified by pathologic stage and histologic type, there was no difference in RFS between the LDCT and CECT groups (Fig. 3).

The median TTR was 67.7 months (IQR, 60.8 to 87.3) in the LDCT group and 73.8 months (IQR, 61.8 to 94.0) in the CECT group.

In the competing risk analysis, the overall 5-year cumulative incidence of recurrence was 4.25% (95% CI, 3.58 to 4.91). The annual risks of recurrence in 3rd, 4th, and 5th postoperative years were 1.43%, 0.93%, and 0.26%, respectively. The LDCT and CECT groups showed no difference in the 5-year cumulative incidence of recurrence (4.56% [95% CI, 3.59 to 5.53] for LDCT, 3.93% [95% CI, 3.02 to 4.84] for CECT; p=0.765). No difference in the cumulative incidence of recurrence between the groups was found in the analysis restricted to pathologic stage I or II/III and adenocarcinoma or squamous cell carcinoma (Table 3, S6 Fig.).

The PRS was not significantly different between the LDCT and CECT groups (3-year post-recurrence OS rate: 44.8% for LDCT vs. 69.8% for CECT; p=0.074) (S7 Fig.). Five patients in the LDCT group (21.7%) and three patients in the CECT group (13.6%) did not receive treatment. Recurrence treatment (HR, 0.338; 95% CI, 0.122 to 0.935; p=0.037), regional recurrence (HR, 5.697; 95% CI, 1.280 to 25.354; p=0.022), and male sex (HR, 4.196; 95% CI, 1.390 to 12.665; p=0.011) were statistically significant factors for PRS (S8 Table).

Further, SPLC was documented in 22 patients in the matched population: 11 each in the LDCT and CECT groups. The overall 5-year cumulative incidence of SPLC was 1.86% (95% CI, 1.41 to 2.31), with an incidence rate of 0.37 (95% CI, 0.23 to 0.56) per 100 person-years. No difference was found in the cumulative incidence of SPLC between the groups (LDCT group: 1.76% [95% CI, 1.14 to 2.38] vs. CECT group: 1.95% [95% CI, 1.31 to 2.60]; p=0.903).

Discussion

The OS, RFS, and TTR were not significantly different between the LDCT and CECT groups of patients with lung cancer who were disease-free in the 2nd year after curative resection. PRS was attributed to the treatment of the recurrence and not the type of surveillance CT.

LDCT offers several advantages over CECT. First, LDCT is free of contrast-related complications. Hypersensitivity reactions to contrast agents can result in rashes, itching, nausea, and severe anaphylactic shock [10]. Additionally, patients with limited kidney function can undergo LDCT without the risk of contrast-induced nephropathy [11], avoiding the need for prevention strategies [12]. Contrast-associated acute kidney injury has been reported in 0%-21% of cases [13]. Furthermore, more than 30% of patients experience psychological stress related to the use of an intravenous contrast medium [14]. Second, LDCT significantly decreases radiation exposure compared with CECT. LDCT generally requires only 10% of the radiation dose used for CECT [15]. Cumulative radiation exposure increases the risk of developing new cancers [16], making LDCT advantageous for patients with a long life expectancy. With the increasing number of lung cancer survivors, surveillance is becoming a significant healthcare concern and expense [17]. In this context, reduced radiation exposure from LDCT can provide both medical and financial benefits [18,19]. Third, LDCT is more economical than CECT. A study from the United States reported that $199 could be saved in each case by opting for LDCT ($3,979) over CECT ($4,178) [20].

Although the benefits of LDCT are clear, the critical question is whether LDCT can detect recurrences as effectively as CECT. Delays or failures in the detection of recurrence can impair long-term survival. However, relevant research in this field is lacking.

This study demonstrated that the efficacy of recurrence detection with LDCT was comparable to that of CECT and that LDCT did not impair long-term prognosis. In the matched cohort, the 5-year OS rates in the LDCT and CECT groups were 96.0% and 98.0%, respectively. The 5-year RFS rate did not differ between the two groups (95.4% vs. 96.0%, p=0.761). Furthermore, the TTR (67.7 vs. 73.8 months) and cumulative incidence of recurrence (4.56% vs. 3.93%, p=0.765) did not differ between the two groups. Although our data may appear to show low recurrence rates, it actually reflects 3-year recurrence between the second and fifth postoperative years. Moreover, as over 90% of patients were stage I after matching, the recurrence rates appear to be consistent with those reported in previous studies [21].

PRS would represent the impact of treatment delay caused by diagnosis delay. Our study found no differences in PRS between the LDCT and CECT groups. The PRS for the CECT and LDCT groups showed no significant difference (HR, 0.477; p=0.081). Furthermore, CT type was not identified as a prognostic factor for PRS in multivariable analysis.

The results of our study can be explained in several ways. First, recurrences often appeared at multiple sites simultaneously. Of the 130 patients who experienced recurrence, 58 (44.6%) had metastases at multiple sites. In such cases, the likelihood of missing all metastatic lesions is very low, and detection of only one suspicious lesion may lead to further investigation. Second, CECT may be beneficial for the detection of intrapulmonary or hilar node recurrence and recurrence in hypervascular organs such as the liver. However, the incidence was 11.5% in hilar node metastases and only 2.3% of liver metastases in our cohort. The weight of these sites is reduced by the presence of multiple metastases to multiple organs. Studies comparing CT for detecting intrathoracic lymph node recurrence usually focused on each nodal station rather than on each patient [22]. A study comparing CECT and NECT reported that the number of enlarged lymph nodes detected by CECT was 11% higher than that detected by NECT (p=0.044) [23]. However, no significant differences were observed at the patient level. Another study showed that CECT was beneficial for hilar lymph nodes but not for mediastinal lymph nodes [24]. Although CECT offers radiological advantages, it remains unclear whether it offers a clinical advantage in postoperative surveillance. Third, the presence of metastases not detectable on chest CT may have partially influenced our study results. The incidence of metastases outside of the chest CT scans was only 17.7%. The mode of chest CT becomes irrelevant in this situation, and the use of LDCT may not impair the outcomes.

Only one previous study, involving 416 patients with stage I NSCLC, has examined the correlation between the type of surveillance CT and survival outcomes [25]. In the matched analysis of that study, there were no statistically significant differences in RFS or OS between the LDCT and CECT groups, similar to the findings of our study.

The effectiveness of LDCT in detecting recurrence and survival outcomes was observed across different pathologic stages and histologic types. This might be more obvious in patients with stage I cancer and adenocarcinoma, as supported by the large study population. Although the number of patients was relatively small, the results were statistically consistent for stages II and III. There has been no study that investigated postoperative surveillance in stage II/III. Thus, LDCT might be an option for the surveillance of stage II/III NSCLC after the second postoperative year. However, as the 5-year cumulative recurrence rate was 23.4% in stage II and III, a careful approach that considers each patient’s situation is warranted.

As the present study deals with long-term surveillance, SPLC was expected to occur in a non-negligible number of patients compared to the number of recurrences. Hence, SPLC is considered a competing risk factor for recurrence. Our data showed that the overall 5-year cumulative incidence of SPLC was 1.86%, which is consistent with the 1%-2% incidence reported in previous studies [26].

Our study had some limitations. First, this was a retrospective study. We used propensity score matching to reduce confounding factors that might influence the allocation of surveillance methods; however, there was a slight, although statistically insignificant, disparity in the number of recurrence events between the groups after matching. Second, this study primarily involved patients with stage I, as those with higher stages likely dropped out of the study cohort owing to more recurrences within the first two postoperative years. However, the number of patients with stage II/III NSCLC was larger than that reported in other similar studies [27] and this is the only study to compare LDCT and CECT in stage II/III NSCLC. Third, CT intervals and the number of CT examinations were not considered. Some studies have suggested that the surveillance intervals are not associated with patient survival [28]. Fourth, not all recurrences were confirmed using biopsy. Nevertheless, a biopsy rate of 64% may be acceptable in real-world practice, and multidisciplinary tumor board discussions, along with the use of PET/CT, can ensure reliable recurrence diagnoses in patients who did not undergo biopsy.

In conclusion, the LDCT and CECT groups showed no significant differences in OS, RFS, and TTR. This finding was consistent regardless of pathologic stage or histologic type. LDCT is a feasible option for the surveillance of patients with NSCLC who remain recurrence-free during the first two postoperative years.

Electronic Supplementary Material

Supplementary materials are available at Cancer Research and Treatment website (https://www.e-crt.org).

crt-2025-256_S1_Fig.pdf

crt-2025-256_S2_Table.pdf

crt-2025-256_S3_Table.pdf

crt-2025-256_S4_Table.pdf

crt-2025-256_S5_Table.pdf

crt-2025-256_S6_Fig.pdf

crt-2025-256_S7_Fig.pdf

crt-2025-256_S8_Table.pdf

NOTES

Ethical Statement

This study was approved by the Institutional Review Board (No. 2111-108-1272), and the need for patient consent was waived.

Author Contributions

Conceived and designed the analysis: Na B, Park IK.

Collected the data: Na B.

Contributed data or analysis tools: Na B, Park JH, Na KJ, Park S, Kang CH, Kim YT, Park IK.

Performed the analysis: Na B, Park IK.

Wrote the paper: Na B, Park IK.

Conflict of Interest

Conflict of interest relevant to this article was not reported. C.H. Kang is a proctor of Intuitive Surgical Korea Inc. and was paid for consultations, lectures, presentations, speaker bureaus, manuscript writing, or educational events by Intuitive Surgical Korea Inc. within the past 36 months. K.J.N. is a co-founder and stockholder of Portrai, Inc. (Republic of Korea), which is unrelated to this article. Y.T. Kim was paid for consultations, lectures, presentations, speaker bureaus, manuscript writing, and educational events from Johnson & Johnson and AstraZeneca within the past 36 months.

Funding

This research was supported by the Department of Thoracic and Cardiovascular Surgery Research Fund at Seoul National University Hospital.

Fig. 1.

Consort diagram of the study population. CECT, contrast-enhanced chest computed tomography; LDCT, low-dose non-contrast-enhanced chest computed tomography; NSCLC, non–small cell lung cancer.

Fig. 2.

Kaplan-Meier survival curves for overall survival (OS) in the matched population. (A) All patients. (B) Stage I. (C) Stage II and III. (D) Adenocarcinoma. (E) Squamous cell carcinoma. None of the subgroups showed significant differences in OS. The x-axis represents the time between curative resection and death. The survival probability was 100% until 26 months because the study population included patients who survived for 26 months without recurrence. CECT, contrast-enhanced chest computed tomography; LDCT, low-dose non-contrast-enhanced chest computed tomography.

Fig. 3.

Kaplan-Meier survival curves for recurrence-free survival (RFS) in the matched population. (A) All patients. (B) Stage I. (C) Stage II and III. (D) Adenocarcinoma. (E) Squamous cell carcinoma. None of the subgroups showed significant differences in RFS. The x-axis represents the time from curative resection to recurrence or death. The survival probability was 100% until 26 months because the study population included patients who survived for 26 months without recurrence. CECT, contrast-enhanced chest computed tomography; LDCT, low-dose non-contrast-enhanced chest computed tomography;.

Table 1.

Demographic data of the patients

Variable	Total (n=2,083)	Before matching			After matching
Variable	Total (n=2,083)	LDCT group (n=466)	CECT group (n=1,617)	p-value	LDCT group (n=466)	CECT group (n=466)	p-value
Age (yr)	62.5±10.0	61.7±11.2	62.7±9.7	0.065	61.7±11.2	61.4±9.6	0.725
Male	1,082 (51.9)	228 (48.9)	854 (52.8)	0.139	228 (48.9)	210 (45.1)	0.237
Smoking history
Non-smoker	1,115 (53.5)	264 (56.7)	851 (52.6)	0.190	264 (56.7)	287 (61.6)	0.252
Ex-smoker	642 (30.8)	140 (30.0)	502 (31.0)		140 (30.0)	119 (25.5)
Current smoker	326 (15.7)	62 (13.3)	264 (16.3)		62 (13.3)	60 (12.9)
ECOG PS
0	1,843 (88.5)	427 (91.6)	1,416 (87.6)	0.013	427 (91.6)	432 (92.7)	0.640
1	229 (11.0)	38 (8.2)	191 (11.8)		38 (8.2)	33 (7.1)
2	11 (0.5)	1 (0.2)	10 (0.6)		1 (0.2)	1 (0.2)
Resection extent
Sublobar resection	378 (18.1)	99 (21.2)	279 (17.3)	0.001	99 (21.2)	82 (17.6)	0.199
Lobectomy	1,661 (79.7)	366 (78.5)	1,295 (80.1)		366 (78.5)	383 (82.2)
Pneumonectomy	44 (2.1)	1 (0.2)	43 (2.7)		1 (0.2)	1 (0.2)
Year of operation
2011	221 (10.6)	39 (8.4)	182 (11.3)	0.562	39 (8.4)	44 (9.4)	0.967
2012	285 (13.7)	77 (16.5)	208 (12.9)		77 (16.5)	72 (15.5)
2013	243 (11.7)	75 (16.1)	168 (10.4)		75 (16.1)	74 (15.9)
2014	261 (12.5)	37 (7.9)	224 (13.9)		37 (7.9)	32 (6.9)
2015	330 (15.8)	78 (16.7)	252 (15.6)		78 (16.7)	84 (18.0)
2016	375 (18.0)	76 (16.3)	299 (18.5)		76 (16.3)	71 (15.2)
2017	368 (17.7)	84 (18.0)	284 (17.6)		84 (18.0)	89 (19.1)
Histologic type
ADC	1,583 (76.0)	367 (78.8)	1,216 (75.2)	< 0.001	367 (78.8)	384 (82.4)	0.369
SqCC	321 (15.4)	43 (9.2)	278 (17.2)		43 (9.2)	35 (7.5)
Others	179 (8.6)	56 (12.0)	123 (7.6)		56 (12.0)	47 (10.1)
Pathologic stage
I	1,589 (76.3)	418 (89.7)	1,171 (72.4)	< 0.001	418 (89.7)	422 (90.6)	> 0.99
II	283 (13.6)	37 (7.9)	246 (15.2)		37 (7.9)	29 (6.2)
III	211 (10.1)	11 (2.4)	200 (12.4)		11 (2.4)	15 (3.2)
DM	284 (13.6)	62 (13.3)	222 (13.7)	0.814	62 (13.3)	51 (10.9)	0.270
Underlying lung disease	246 (11.8)	58 (12.4)	188 (11.6)	0.629	58 (12.4)	45 (9.7)	0.174
Previous cancer history	360 (17.3)	83 (17.8)	277 (17.1)	0.732	83 (17.8)	94 (20.2)	0.358

Values are presented as mean±SD or number (%). ADC, adenocarcinoma; CECT, contrast-enhanced chest computed tomography; DM, diabetes mellitus; ECOG PS, Eastern Cooperative Oncology Group performance status; LDCT, low-dose non-contrast-enhanced chest computed tomography; SD, standard deviation; SqCC, squamous cell carcinoma.

Table 2.

Site of recurrence in pre-matched patients and matched patients

Site of recurrence	Overall cohort (n=130)	Patients with recurrence
		Before matching			After matching
		LDCT group (n=23)	CECT group (n=107)	p-value	LDCT group (n=23)	CECT group (n=22)	p-value
Local	5 (3.8)	0	5 (4.7)	0.585	0	0	-
Bronchial stump	4 (3.1)	0	4 (3.7)	> 0.99	0	0	-
Stapler line	1 (0.8)	0	1 (0.9)	> 0.99	0	0	-
Regional	90 (69.2)	17 (73.9)	73 (68.2)	0.592	17 (73.9)	16 (72.7)	0.928
Mediastinal node	28 (21.5)	7 (30.4)	21 (19.6)	0.270	7 (30.8)	4 (18.2)	0.339
Hilar node	15 (11.5)	5 (21.7)	10 (9.3)	0.142	5 (21.7)	3 (13.6)	0.699
Ipsilateral lung	52 (40.0)	8 (34.8)	44 (41.1)	0.573	8 (34.8)	9 (40.9)	0.672
Ipsilateral pleura or pleural effusion	27 (20.8)	4 (17.4)	23 (21.5)	0.783	4 (17.4)	5 (22.7)	0.722
Distant	84 (64.6)	16 (69.6)	68 (63.6)	0.584	16 (69.6)	15 (68.2)	0.920
SCN	14 (10.8)	2 (8.7)	12 (11.2)	> 0.99	2 (8.7)	1 (4.5)	> 0.99
Contralateral lung	40 (30.8)	9 (39.1)	31 (29.0)	0.338	9 (39.1)	7 (31.8)	0.608
Contralateral pleura or pleural effusion	1 (0.8)	0	1 (0.9)	-	0	0	-
Bone	16 (12.3)	2 (8.7)	14 (13.1)	0.736	2 (8.7)	2 (9.1)	> 0.99
Brain	15 (11.5)	3 (13.0)	12 (11.2)	0.729	3 (13.0)	4 (18.2)	0.699
Non-regional thoracic LN	6 (4.6)	1 (4.3)	5 (4.7)	> 0.99	1 (4.3)	0	> 0.99
Adrenal gland	4 (3.1)	1 (4.3)	3 (2.8)	0.546	1 (4.3)	0	> 0.99
Liver	3 (2.3)	0	3 (2.8)	> 0.99	0	0
Kidney	1 (0.8)	1 (4.3)	0	0.177	1 (4.3)	0	> 0.99
Soft tissue	4 (3.1)	0	4 (3.7)	> 0.99	0	2 (9.1)	0.233
Other extrathoracic organ	1 (0.8)	0	1 (0.9)	> 0.99	0	0
Recurrence all covered by chest CT	107 (82.3)	19 (82.6)	88 (82.2)	> 0.99	19 (82.6)	17 (77.3)	0.722
Multiple site recurrence in CT coverage	58 (44.6)	11 (47.8)	47 (43.9)	0.733	11 (47.8)	11 (50.0)	0.884

Values are presented as number (%). CECT, contrast-enhanced chest CT; CT, computed tomography; LDCT, low-dose non-contrast-enhanced chest CT; LN, lymph node; SCN, supraclavicular lymph node.

Table 3.

Incidence of recurrence stratified by pathologic stage and histologic type in matched patients

	No. of patients with recurrence	Incidence rate per 100 person-years (95% CI)	5-Year cumulative incidence^a) of recurrence (95% CI)	Gray’s p-value
Overall cohort	45	0.76 (0.56-1.02)	4.25 (3.58-4.91)	0.765
LDCT (n=466)	23	0.81 (0.51-1.21)	4.56 (3.59-5.53)
CECT (n=466)	22	0.72 (0.45-1.09)	3.93 (3.02-4.84)
Pathologic stage I	23	0.43 (0.27-0.64)	2.18 (1.67-2.69)	0.729
LDCT (n=418)	12	0.46 (0.24-0.81)	2.41 (1.66-3.17)
CECT (n=422)	11	0.39 (0.20-0.71)	1.95 (1.27-2.64)
Pathologic stage II, III	22	4.21 (2.64-6.37)	23.42 (18.9-27.93)	0.990
LDCT (n=48)	11	4.23 (2.11-7.57)	23.65 (17.32-29.99)
CECT (n=44)	11	4.19 (2.09-7.50)	22.98 (16.51-29.45)
ADC	39	0.84 (0.60-1.14)	4.44 (3.69-5.20)	0.894
LDCT (n=367)	19	0.86 (0.52-1.35)	4.67 (3.56-5.78)
CECT (n=384)	20	0.81 (0.50-1.26)	4.21 (3.18-5.24)
SqCC	6	1.28 (0.47-2.78)	8.01 (4.84-11.17)	0.521
LDCT (n=43)	4	1.62 (0.44-4.15)	9.70 (5.02-14.38)
CECT (n=35)	2	0.90 (0.11-3.24)	5.99 (1.81-10.17)

ADC, adenocarcinoma; CECT, contrast-enhanced chest computed tomography; CI, confidence interval; LDCT, low-dose non-contrast-enhanced chest computed tomography; SqCC, squamous cell carcinoma.

^a) As the study population is composed of patients who were recurrence-free during postoperative 2 years, actual duration of cumulative incidence is 3 years.

REFERENCES

1. Jeon J, Lee DB, Shin SJ, Han DH, Chang JS, Han YD, et al. Effect of high-versus low-frequency of abdominopelvic computed tomography follow-up testing on overall survival in patients with stage II Or III colon cancer. Clin Colorectal Cancer. 2023;22:307–17. Article PubMed
2. Pietra N, Sarli L, Costi R, Ouchemi C, Grattarola M, Peracchia A. Role of follow-up in management of local recurrences of colorectal cancer: a prospective, randomized study. Dis Colon Rectum. 1998;41:1127–33. PubMed
3. Koike T, Goto T, Kitahara A, Sato S, Saitoh M, Hashimoto T, et al. Characteristics and timing of recurrence during postoperative surveillance after curative resection for lung adenocarcinoma. Surg Today. 2017;47:1469–75. Article PubMed PDF
4. Postmus PE, Kerr KM, Oudkerk M, Senan S, Waller DA, Vansteenkiste J, et al. Early and locally advanced non-small-cell lung cancer (NSCLC): ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2017;28:iv1–21. Article PubMed
5. Schneider BJ, Ismaila N, Aerts J, Chiles C, Daly ME, Detterbeck FC, et al. Lung cancer surveillance after definitive curative-intent therapy: ASCO guideline. J Clin Oncol. 2020;38:753–66. PubMed
6. NCCN Clinical Practice Guidelines in Oncology (NCCN guidelines). Non-small cell lung cancer. National Comprehensive Cancer Network; 2024.
7. NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines). Non-small cell lung cancer. National Comprehensive Cancer Network; 2015.
8. NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines). Non-small cell lung cancer. National Comprehensive Cancer Network; 2018.
9. International Association for the Study of Lung Cancer. Staging manual in thoracic oncology. 2nd ed. Editorial Rx Press; 2016.
10. Fukushima Y, Taketomi-Takahashi A, Suto T, Hirasawa H, Tsushima Y. Clinical features and risk factors of iodinated contrast media (ICM)-induced anaphylaxis. Eur J Radiol. 2023;164:110880.Article PubMed
11. Chaudhari H, Mahendrakar S, Baskin SE, Reddi AS. Contrast-induced acute kidney injury: evidence in support of its existence and a review of its pathogenesis and management. Int J Nephrol Renovasc Dis. 2022;15:253–66. Article PubMed PMC PDF
12. Subramaniam RM, Suarez-Cuervo C, Wilson RF, Turban S, Zhang A, Sherrod C, et al. Effectiveness of prevention strategies for contrast-induced nephropathy: a systematic review and meta-analysis. Ann Intern Med. 2016;164:406–16. Article PubMed
13. Shah R, Le FK, Labroo A, Khan MR. Contrast-associated acute kidney injury. Quant Imaging Med Surg. 2020;10:891–4. Article PubMed PMC
14. Lange S, Medrzycka-Dabrowska W, Malecka-Dubiela A. Patient experience during contrast-enhanced computed tomography examination: anxiety, feelings, and safety. Safety. 2023;9:69.Article
15. Diederich S, Lenzen H, Windmann R, Puskas Z, Yelbuz TM, Henneken S, et al. Pulmonary nodules: experimental and clinical studies at low-dose CT. Radiology. 1999;213:289–98. Article PubMed
16. Albert JM. Radiation risk from CT: implications for cancer screening. AJR Am J Roentgenol. 2013;201:W81–7. Article PubMed
17. Rajapakse P. An update on survivorship issues in lung cancer patients. World J Oncol. 2021;12:45–9. Article PubMed PMC
18. Hendrick RE, Smith RA. Benefit-to-radiation-risk of low-dose computed tomography lung cancer screening. Cancer. 2024;130:216–23. Article PubMed
19. McCunney RJ, Li J. Radiation risks in lung cancer screening programs. Chest. 2014;145:618–24. Article
20. Priester JI, Simister SK, Sario M, Choi J, Pina D, Theriault R, et al. Scrutinizing the use of contrasted chest CTs in extremity sarcoma staging and surveillance. J Surg Oncol. 2024;129:523–30. Article PubMed
21. Rajaram R, Huang Q, Li RZ, Chandran U, Zhang Y, Amos TB, et al. Recurrence-free survival in patients with surgically resected non-small cell lung cancer: a systematic literature review and meta-analysis. Chest. 2024;165:1260–70. Article PubMed
22. Paolini M, Wirth K, Tufman A, Reiser M, Huber RM, Mueller-Lisse UG. Thoracic lymph node delineation at dose-reduced (1 mSv) dose-modulated contrast enhanced MDCT: a retrospective pilot study. Radiol Med. 2016;121:644–51. Article PubMed PDF
23. Cascade PN, Gross BH, Kazerooni EA, Quint LE, Francis IR, Strawderman M, et al. Variability in the detection of enlarged mediastinal lymph nodes in staging lung cancer: a comparison of contrast-enhanced and unenhanced CT. AJR Am J Roentgenol. 1998;170:927–31. Article PubMed
24. Takahashi M, Nitta N, Takazakura R, Nagatani Y, Ushio N, Murata K. Detection of mediastinal and hilar lymph nodes by 16-row MDCT: is contrast material needed? Eur J Radiol. 2008;66:287–91. Article PubMed
25. Nakada T, Takahashi Y, Sakakura N, Masago K, Iwata H, Ohtsuka T, et al. Postoperative surveillance using low-dose computed tomography for non-small-cell lung cancer. Eur J Cardiothorac Surg. 2022;63:ezac549Article PubMed PDF
26. Johnson BE. Second lung cancers in patients after treatment for an initial lung cancer. J Natl Cancer Inst. 1998;90:1335–45. Article PubMed
27. Benamore R, Shepherd FA, Leighl N, Pintilie M, Patel M, Feld R, et al. Does intensive follow-up alter outcome in patients with advanced lung cancer? J Thorac Oncol. 2007;2:273–81. Article PubMed
28. McMurry TL, Stukenborg GJ, Kessler LG, Colditz GA, Wong ML, Francescatti AB, et al. More frequent surveillance following lung cancer resection is not associated with improved survival: a nationally representative cohort study. Ann Surg. 2018;268:632–9. PubMed PMC

Figure & Data

REFERENCES

Citations

Citations to this article as recorded by

PubReader
ePub Link

Cite

CITE: export Copy Download Format; Close

Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:

RIS — For EndNote, ProCite, RefWorks, and most other reference management software
BibTeX — For JabRef, BibDesk, and other BibTeX-specific software

Include:

Citation for the content below
Citation and abstract for the content below

Comparison of Surveillance with Low-Dose and Contrast-Enhanced Chest Computed Tomography in Patients Disease-Free for 2 Years after Curative Resection for Lung Cancer

Cancer Res Treat. 2026;58(2):454-464. Published online June 5, 2025

DOI: https://doi.org/10.4143/crt.2025.256

XML Download

Abstract
Introduction
Materials and Methods
Results
Discussion
Electronic Supplementary Material
NOTES
REFERENCES

Comparison of Surveillance with Low-Dose and Contrast-Enhanced Chest Computed Tomography in Patients Disease-Free for 2 Years after Curative Resection for Lung Cancer

Fig. 1. Consort diagram of the study population. CECT, contrast-enhanced chest computed tomography; LDCT, low-dose non-contrast-enhanced chest computed tomography; NSCLC, non–small cell lung cancer.

Fig. 2. Kaplan-Meier survival curves for overall survival (OS) in the matched population. (A) All patients. (B) Stage I. (C) Stage II and III. (D) Adenocarcinoma. (E) Squamous cell carcinoma. None of the subgroups showed significant differences in OS. The x-axis represents the time between curative resection and death. The survival probability was 100% until 26 months because the study population included patients who survived for 26 months without recurrence. CECT, contrast-enhanced chest computed tomography; LDCT, low-dose non-contrast-enhanced chest computed tomography.

Fig. 3. Kaplan-Meier survival curves for recurrence-free survival (RFS) in the matched population. (A) All patients. (B) Stage I. (C) Stage II and III. (D) Adenocarcinoma. (E) Squamous cell carcinoma. None of the subgroups showed significant differences in RFS. The x-axis represents the time from curative resection to recurrence or death. The survival probability was 100% until 26 months because the study population included patients who survived for 26 months without recurrence. CECT, contrast-enhanced chest computed tomography; LDCT, low-dose non-contrast-enhanced chest computed tomography;.

Fig. 1.

Fig. 2.

Fig. 3.

Comparison of Surveillance with Low-Dose and Contrast-Enhanced Chest Computed Tomography in Patients Disease-Free for 2 Years after Curative Resection for Lung Cancer

Variable	Total (n=2,083)	Before matching			After matching
Variable	Total (n=2,083)	LDCT group (n=466)	CECT group (n=1,617)	p-value	LDCT group (n=466)	CECT group (n=466)	p-value
Age (yr)	62.5±10.0	61.7±11.2	62.7±9.7	0.065	61.7±11.2	61.4±9.6	0.725
Male	1,082 (51.9)	228 (48.9)	854 (52.8)	0.139	228 (48.9)	210 (45.1)	0.237
Smoking history
Non-smoker	1,115 (53.5)	264 (56.7)	851 (52.6)	0.190	264 (56.7)	287 (61.6)	0.252
Ex-smoker	642 (30.8)	140 (30.0)	502 (31.0)		140 (30.0)	119 (25.5)
Current smoker	326 (15.7)	62 (13.3)	264 (16.3)		62 (13.3)	60 (12.9)
ECOG PS
0	1,843 (88.5)	427 (91.6)	1,416 (87.6)	0.013	427 (91.6)	432 (92.7)	0.640
1	229 (11.0)	38 (8.2)	191 (11.8)		38 (8.2)	33 (7.1)
2	11 (0.5)	1 (0.2)	10 (0.6)		1 (0.2)	1 (0.2)
Resection extent
Sublobar resection	378 (18.1)	99 (21.2)	279 (17.3)	0.001	99 (21.2)	82 (17.6)	0.199
Lobectomy	1,661 (79.7)	366 (78.5)	1,295 (80.1)		366 (78.5)	383 (82.2)
Pneumonectomy	44 (2.1)	1 (0.2)	43 (2.7)		1 (0.2)	1 (0.2)
Year of operation
2011	221 (10.6)	39 (8.4)	182 (11.3)	0.562	39 (8.4)	44 (9.4)	0.967
2012	285 (13.7)	77 (16.5)	208 (12.9)		77 (16.5)	72 (15.5)
2013	243 (11.7)	75 (16.1)	168 (10.4)		75 (16.1)	74 (15.9)
2014	261 (12.5)	37 (7.9)	224 (13.9)		37 (7.9)	32 (6.9)
2015	330 (15.8)	78 (16.7)	252 (15.6)		78 (16.7)	84 (18.0)
2016	375 (18.0)	76 (16.3)	299 (18.5)		76 (16.3)	71 (15.2)
2017	368 (17.7)	84 (18.0)	284 (17.6)		84 (18.0)	89 (19.1)
Histologic type
ADC	1,583 (76.0)	367 (78.8)	1,216 (75.2)	< 0.001	367 (78.8)	384 (82.4)	0.369
SqCC	321 (15.4)	43 (9.2)	278 (17.2)		43 (9.2)	35 (7.5)
Others	179 (8.6)	56 (12.0)	123 (7.6)		56 (12.0)	47 (10.1)
Pathologic stage
I	1,589 (76.3)	418 (89.7)	1,171 (72.4)	< 0.001	418 (89.7)	422 (90.6)	> 0.99
II	283 (13.6)	37 (7.9)	246 (15.2)		37 (7.9)	29 (6.2)
III	211 (10.1)	11 (2.4)	200 (12.4)		11 (2.4)	15 (3.2)
DM	284 (13.6)	62 (13.3)	222 (13.7)	0.814	62 (13.3)	51 (10.9)	0.270
Underlying lung disease	246 (11.8)	58 (12.4)	188 (11.6)	0.629	58 (12.4)	45 (9.7)	0.174
Previous cancer history	360 (17.3)	83 (17.8)	277 (17.1)	0.732	83 (17.8)	94 (20.2)	0.358

Site of recurrence	Overall cohort (n=130)	Patients with recurrence
		Before matching			After matching
		LDCT group (n=23)	CECT group (n=107)	p-value	LDCT group (n=23)	CECT group (n=22)	p-value
Local	5 (3.8)	0	5 (4.7)	0.585	0	0	-
Bronchial stump	4 (3.1)	0	4 (3.7)	> 0.99	0	0	-
Stapler line	1 (0.8)	0	1 (0.9)	> 0.99	0	0	-
Regional	90 (69.2)	17 (73.9)	73 (68.2)	0.592	17 (73.9)	16 (72.7)	0.928
Mediastinal node	28 (21.5)	7 (30.4)	21 (19.6)	0.270	7 (30.8)	4 (18.2)	0.339
Hilar node	15 (11.5)	5 (21.7)	10 (9.3)	0.142	5 (21.7)	3 (13.6)	0.699
Ipsilateral lung	52 (40.0)	8 (34.8)	44 (41.1)	0.573	8 (34.8)	9 (40.9)	0.672
Ipsilateral pleura or pleural effusion	27 (20.8)	4 (17.4)	23 (21.5)	0.783	4 (17.4)	5 (22.7)	0.722
Distant	84 (64.6)	16 (69.6)	68 (63.6)	0.584	16 (69.6)	15 (68.2)	0.920
SCN	14 (10.8)	2 (8.7)	12 (11.2)	> 0.99	2 (8.7)	1 (4.5)	> 0.99
Contralateral lung	40 (30.8)	9 (39.1)	31 (29.0)	0.338	9 (39.1)	7 (31.8)	0.608
Contralateral pleura or pleural effusion	1 (0.8)	0	1 (0.9)	-	0	0	-
Bone	16 (12.3)	2 (8.7)	14 (13.1)	0.736	2 (8.7)	2 (9.1)	> 0.99
Brain	15 (11.5)	3 (13.0)	12 (11.2)	0.729	3 (13.0)	4 (18.2)	0.699
Non-regional thoracic LN	6 (4.6)	1 (4.3)	5 (4.7)	> 0.99	1 (4.3)	0	> 0.99
Adrenal gland	4 (3.1)	1 (4.3)	3 (2.8)	0.546	1 (4.3)	0	> 0.99
Liver	3 (2.3)	0	3 (2.8)	> 0.99	0	0
Kidney	1 (0.8)	1 (4.3)	0	0.177	1 (4.3)	0	> 0.99
Soft tissue	4 (3.1)	0	4 (3.7)	> 0.99	0	2 (9.1)	0.233
Other extrathoracic organ	1 (0.8)	0	1 (0.9)	> 0.99	0	0
Recurrence all covered by chest CT	107 (82.3)	19 (82.6)	88 (82.2)	> 0.99	19 (82.6)	17 (77.3)	0.722
Multiple site recurrence in CT coverage	58 (44.6)	11 (47.8)	47 (43.9)	0.733	11 (47.8)	11 (50.0)	0.884

	No. of patients with recurrence	Incidence rate per 100 person-years (95% CI)	5-Year cumulative incidence^a) of recurrence (95% CI)	Gray’s p-value
Overall cohort	45	0.76 (0.56-1.02)	4.25 (3.58-4.91)	0.765
LDCT (n=466)	23	0.81 (0.51-1.21)	4.56 (3.59-5.53)
CECT (n=466)	22	0.72 (0.45-1.09)	3.93 (3.02-4.84)
Pathologic stage I	23	0.43 (0.27-0.64)	2.18 (1.67-2.69)	0.729
LDCT (n=418)	12	0.46 (0.24-0.81)	2.41 (1.66-3.17)
CECT (n=422)	11	0.39 (0.20-0.71)	1.95 (1.27-2.64)
Pathologic stage II, III	22	4.21 (2.64-6.37)	23.42 (18.9-27.93)	0.990
LDCT (n=48)	11	4.23 (2.11-7.57)	23.65 (17.32-29.99)
CECT (n=44)	11	4.19 (2.09-7.50)	22.98 (16.51-29.45)
ADC	39	0.84 (0.60-1.14)	4.44 (3.69-5.20)	0.894
LDCT (n=367)	19	0.86 (0.52-1.35)	4.67 (3.56-5.78)
CECT (n=384)	20	0.81 (0.50-1.26)	4.21 (3.18-5.24)
SqCC	6	1.28 (0.47-2.78)	8.01 (4.84-11.17)	0.521
LDCT (n=43)	4	1.62 (0.44-4.15)	9.70 (5.02-14.38)
CECT (n=35)	2	0.90 (0.11-3.24)	5.99 (1.81-10.17)

Table 1. Demographic data of the patients

Table 2. Site of recurrence in pre-matched patients and matched patients

Values are presented as number (%). CECT, contrast-enhanced chest CT; CT, computed tomography; LDCT, low-dose non-contrast-enhanced chest CT; LN, lymph node; SCN, supraclavicular lymph node.

Table 3. Incidence of recurrence stratified by pathologic stage and histologic type in matched patients

ADC, adenocarcinoma; CECT, contrast-enhanced chest computed tomography; CI, confidence interval; LDCT, low-dose non-contrast-enhanced chest computed tomography; SqCC, squamous cell carcinoma.

As the study population is composed of patients who were recurrence-free during postoperative 2 years, actual duration of cumulative incidence is 3 years.