Time between imaging and surgery is not a risk factor for upstaging of clinical stage IA non-small-cell lung cancer

Abstract

Open in new tabDownload slide

OBJECTIVES

The timing of preoperative imaging in patients with lung cancer is a debated topic, as there are limited data on cancer progression during the interval between clinical staging by imaging and pathological staging after resection. We quantified disease progression during this interval in patients with early stage non-small-cell lung cancer (NSCLC) to better understand if its length impacts upstaging.

METHODS

We retrospectively reviewed our institutional database to identify patients who underwent surgery for clinically staged T1N0M0 NSCLC from January 2015 through September 2022. Tumour upstaging between chest computed tomography (CT) and surgery were analysed as a function of time (<30, 30–59, ≥60 days) for different nodule subtypes. We analysed data across 3 timeframes using Pearson’s chi-squared and analysis of variance tests.

RESULTS

During the study period, 622 patients underwent surgery for clinically staged T1N0M0 NSCLC. CT-to-surgery interval was <30 days in 228 (36.7%), 30–59 days in 242 (38.9%) and ≥60 days in 152 (24.4%) with no differences in patient or nodule characteristics observed between these groups. T-stage increased in 346 patients (55.6%) between CT imaging and surgery. Among these patients, 126 (36.4%) had ground-glass nodules, 147 (42.5%) had part-solid nodules and 73 (21.1%) had solid nodules. CT-to-surgery interval length was not associated with upstaging of any nodule subtype (full-cohort, P = 0.903; ground-glass, P = 0.880; part-solid, P = 0.858; solid, P = 0.959).

CONCLUSIONS

This single-centre experience suggests no significant association between tumour upstaging and time from imaging to lung resection in patients with clinical stage IA NSCLC. Further studies are needed to better understand the risk factors for upstaging.

INTRODUCTION

Lung cancer is the second most common cancer in both men and women in the USA (excluding non-melanoma skin cancer) and the most lethal neoplasm [1]. The American Cancer Society estimates that there will be 238 340 new cases of lung cancer and 127 070 deaths from lung cancer in 2023 [1]. The implementation of lung cancer screening programmes and advances in high-resolution computed tomography (CT) have facilitated earlier detection of non-small-cell lung cancer (NSCLC) in recent years [2], making it increasingly important to understand how early stage tumours may progress to more advanced stages.

The 8th edition of the tumour, lymph node and metastasis (TNM) staging system for lung cancer implemented changes to the T descriptor, which accounts for tumour size, tumour invasion and location of any separate tumour with respect to the primary tumour. Now, each centimetre increase in the tumour’s diameter corresponds to a different stage, adding an important prognostic factor to the staging system [3–6]. The solid component of semi-solid lesions seen on chest CT scans is used to determine the clinical tumour stage [6]. International guidelines still recommend that measurements of both the solid component and the whole tumour, including any ground-glass component, are made and recorded [3]. However, the criteria for distinguishing lung nodules into solid, part-solid and ground-glass remain controversial and warrant complete consensus [7].

Solid tumours typically double in size every 100–400 days, while part-solid tumours typically double in size every 3–5 years [8]. Therefore, it is reasonable to recommend longer follow-up intervals for patients with part-solid nodules as compared with those recommended for patients with solid nodules [7]. The British Thoracic Society guidelines for lung cancer treatment recommend performing a chest CT scan within 4 weeks of surgical treatment for T3 and T4 lesions [9]. Similarly, the National Comprehensive Cancer Network (NCCN) Clinical Practice Guidelines in Oncology advise using CT or positron emission tomography (PET)–CT images taken within 59 days for staging when planning a surgical procedure [10]. However, these recommendations are based on limited evidence, and no uniform consensus supports a specific, acceptable interval [9, 10].

Our study aimed to evaluate whether longer duration between imaging and surgery increases the risk of upstaging of the tumour in patients diagnosed with clinical stage IA NSCLC. To this end, we measured changes in tumour size (T stage) between the last CT scan before surgery and the surgery itself in people with clinical stage IA NSCLC. We used the 8th edition of the TNM staging system to see how time affects T-upstaging in this group of people.

MATERIALS AND METHODS

Ethics statement: the prospectively maintained lung resection database at the Brigham and Women’s Hospital was reviewed. Approval for the review of hospital records was obtained from the Institutional Review Board on 23 January 2023 (protocol 2023P0000470), and the requirement for informed consent was waived.

We identified all patients in the database with pathologically proven, clinical stage IA NSCLC who underwent surgery between January 2015 and September 2022. Clinical stage IA disease was defined based on Union for International Cancer Control/American Joint Committee on Cancer/International Association for the Study of Lung Cancer (UICC/AJCC/IASLC) criteria, utilizing chest CT and/or PET–CT scans for evaluation [3–5]. Patients who received preoperative neoadjuvant chemotherapy or radiation, had incomplete data or had a CT scan-to-surgery interval exceeding 180 days were excluded from the analysis.

Clinical data analysed included patient age, sex, body mass index and forced expiratory volume in 1 s. The chest CT scan data included nodule characteristics (solid, part-solid or ground-glass), size (total nodule size and size of the solid component) and tumour location. Tumour size was acquired from the official radiology report. The largest dimension of the solid component for solid and part-solid nodules, as well as the largest dimension of ground-glass nodules, was used for clinical staging. Additionally, tumour size at pathological review, pathological nodal classification and pathological stage were documented.

The primary end-point was the rate of lung nodule upstaging as a function of the time between the last diagnostic imaging and the date of surgical intervention. A key secondary end-point was the impact of nodule type on tumour upstaging. For the comparative analyses, the patients were divided into 3 groups based on the time interval between the last pre-surgical imaging and the surgical procedure: group 1, <30 days; group 2, 30–59 days; and group 3, 60–180 days.

Categorical data among the 3 distinct timeframes were compared using Pearson’s chi-squared test, while continuous data were compared using analysis of variance tests or non-parametric alternatives for non-normal data. Further, we stratified the association between upstaging and time by nodule subtype and compared it with the chi-squared test. All hypothesis tests were 2-sided with P-values <0.05 indicating statistical significance. Statistical analysis was performed using SAS 9.4 software.

RESULTS

Clinical findings

A total of 622 patients diagnosed with clinical stage IA NSCLC underwent surgical resection between January 2015 and September 2022 and met inclusion criteria for the study. The median age of the patients was 69 years (interquartile range 63–74 years), 425 patients were female (68.3%) and the median forced expiratory volume in 1 s for the 554 patients with available lung function data was 89% of predicted (interquartile range 70–100% of predicted) (Table 1). Prior to surgery, 389 patients (62.5%) underwent combined chest CT and PET scans for clinical staging, while 233 patients (37.5%) were staged exclusively with chest CT scans. Part-solid nodules were observed in 293 patients (47.1%), 183 (29.4%) had solid nodules and 146 (23.5%) had ground-glass nodules. The distribution of clinical T stage among the patients was as follows: 22.8% Tis, 11.3% T1mi, 32.5% T1a, 25.7% T1b and 7.7% T1c.

Wedge resection was the most common surgical procedure and was performed in 299 patients (48.1%). Segmentectomy was performed in 178 patients (28.6%), and lobectomy was performed in 145 (23.3%). Adenocarcinoma was the most common histologic subtype and was diagnosed in 561 patients (90.2%).

Interval between preoperative computed tomography and surgical procedure

The median interval between the last chest CT scan and surgical resection was 38 days (interquartile range 21–58). Among the patients, 228 (36.7%) underwent surgery within 30 days of their last chest CT scan, 242 (38.9%) between 30 and 59 days after their last scan and 152 (24.4%) had surgery 60 or more days after their last scan. Forty-six patients (7.4%) received surgical intervention within 7 days of a CT scan, while 48 (7.7%) had surgical treatment more than 3 months after clinical staging using chest imaging (Fig. 1). The clinical and surgical characteristics of these 3 groups (<30-day, 30–59-day and >60-day interval) were similar (Table 2).

Tumour size upstaging and its association with time and nodule subtype

Out of 622 patients, 346 (55.6%) experienced upstaging due to an increase in pathologically measured tumour size as compared with the nodule size observed in the presurgery imaging study. However, in univariable analysis, no significant differences were found in the rate of tumour upstaging among the 3 patient groups stratified by the interval between imaging and surgery. Upstaging occurred in 56.6% of patients who underwent surgery within 30 days of imaging, 54.5% of patients with surgery with a 30- to 59-day interval and 55.9% of patients who underwent surgery ≥60 days after imaging (P = 0.903). Similarly, when time was modelled continuously with a cubic spine, there was no effect of the interval between imaging and surgery on upstaging (P = 0.89) (Supplementary Material, Fig. S1).

Patients with ground-glass nodules exhibited a significantly higher upstaging rate (n = 126, 86.3%) as compared with patients with part-solid nodules (n = 147, 50.2%) or solid nodules (n = 73, 39.9%) (P < 0.001). However, in univariate analysis, no significant differences in tumour upstaging were observed across different intervals between imaging and surgery within each of the nodule subtypes (Table 3). When upstaging, time and nodule type were examined in a multivariable model with time as a continuous variable, there was similarly no interaction between time and upstaging (nodule effect P < 0.001; time effect P = 0.88; time × nodule effect P = 0.64).

Of 346 patients with upstaging after pathological analysis, 50 were found to have T2a or higher NSCLC. Among these patients, 40 were reclassified as T2a, of whom 33 were reclassified due to the detection of pleural invasion and not due to tumour measurements >3 cm. Two patients were reclassified as T2b due to tumour size, and 8 were reclassified as T3 due to satellite nodules in the same lobe (Fig. 2).

Impact of tumour size upstaging on stage group classification

All patients included in the study were initially classified as having stage IA NSCLC according to the clinical TNM system. Most patients (n = 559, 89.8%) remained in the same stage classification after pathological evaluation, while 63 patients (10.2%) experienced upstaging between clinical TNM staging and pathological TNM staging of their NSCLC. Among the upstaged patients, 36 (6%) were reclassified as stage IB, 1 as stage IIA, 19 as stage IIB and 7 as stage IIIA (Fig. 3). Of the patients with upstaged pathological TNM tumour classifications, 18 (2.9%) were due to lymph node involvement, including 11 patients (1.8%) upstaged to N1 and 7 (1.1%) upstaged to N2.

DISCUSSION

Although the T-stage increased in over half our patients when clinical staging prior to surgery and pathological staging after resection were compared, the time from imaging to resection had no significant impact on the rate of tumour upstaging. Moreover, we observed a TNM upstaging rate of only 10.2%, which is in line with the 14–45.6% range reported in previous studies [11–13]. The majority of our patients experienced upstaging due to changes detected in tumour size rather than changes detected in lymph node involvement, which was limited by the extent of lymph node dissection in each patient. A substantial portion of patients in our cohort would not have required a change in the treatment approach employed, as they remained classified as stage IA.

Surgical resection remains the gold standard for patients with stage I NSCLC, and the recommended time for surgery is ∼8 weeks after diagnosis [10]. In our study cohort, the median time between the clinical staging and surgery was 38 days, which is closely aligned with previous research demonstrating a median time of 36.5 days between the detection of abnormal radiographs and lung cancer treatment [14]. Notably, real-world data have indicated that only 32.1% patients undergo surgery resection within 62 days after imaging [15]. While most of our patients underwent surgery within the recommended timeframe, 24.4% underwent surgery after 60 days. A previous study demonstrated that cardiac clearance, smoking cessation and tumour staging were common reasons for delays between diagnosis and surgical resection [16]. However, the reasons underlying the longer time interval for the 152 patients with resection ≥60 days after imaging in the cohort studied here warrant further investigation.

Pathological upstaging of stage I NSCLC is recognized to have a negative impact on post-surgery outcomes, leading to reduced disease-free and overall survival [15]. Nodule growth has been associated with an increased risk of tumour invasion and a significant reduction in overall survival in previous studies as well [15]. However, few studies have analysed the rate or impact of T-upstaging of early stage tumours after surgery. Our study revealed T-upstaging in 55.6% of the patients when we considered any increase in the nodule size within T1 subgroups. This rate is slightly higher than 28–46% range reported in early studies [15–18]. Our consideration of T1-stage subdivisions in this study likely accounts for this higher rate of T-upstaging. Notably, the upstaging rate was more pronounced for ground-glass nodules (86.3%). This can most likely be attributed to challenges in accurately measuring and characterizing ground-glass nodules on CT-imaging. For this study, nodule measurements came from the original radiology reports in the medical record, and variability typical in real-life clinical scenarios may have affected the accuracy of the measurements. Most of our cohort had adenocarcinomas, and this may have played a significant role in explaining the high incidence of ground-glass nodules in this study. However, no ground-glass nodules were upstaged beyond T2a. Thus, the patients retained stage IA status despite the high rate of T-upstaging when ground-glass morphology was present.

Importantly, we did not observe any significant differences in upstaging as a function of the time between final imaging and surgery. The potential impact of surgical delays on patients with lung cancer remains a topic of debate. Bott and colleagues demonstrated that a delay of more than 8 weeks was independently associated with pathological upstaging [11]. Heiden and colleagues reported that surgical delays exceeding 12 weeks were correlated with an increased risk of recurrence and worse survival [18]. In contrast, several others have observed findings similar to those observed in our study. McPherson and colleagues found no impact of the timing of surgery on the risk of tumour upstaging [15]. Similarly, Maiga and colleagues found no effect of time to surgery on increased tumour growth, with 42% of their cohort experiencing an interval longer than 60 days between lung cancer diagnosis and surgery [16]. Likewise, Quarterman and colleagues found that waiting time for surgery had no significant impact on tumour growth or pathological upstaging [19]. The absence of a clear relationship between time to surgery and T-upstaging suggests that other clinical and biological factors, rather than time to surgery alone, may contribute to the risk of tumour growth.

Limitations

Our study has several limitations inherent to its retrospective design. The analysis was conducted using data from a single centre, and we acknowledge the possibility of selection bias. These data may not apply to tumour subtypes other than adenocarcinoma, as other tumour types were not well represented. Wedge resection was performed in almost half of our cohort (48%), and therefore it is difficult to compare the rate of upstaging we observed due to pathological N stage with upstaging observed as a result of pathological N stage in studies comprised primarily of anatomic resections [20]. Although we suspect this indicates earlier-stage tumours in our patients, it was not possible to dissect the reasons for selection of a non-anatomic resection in this retrospective review. Furthermore, patients who underwent neoadjuvant treatment were excluded from this study to minimize its potentially confounding influence on tumour size. Future research that explores the relationship between neoadjuvant treatment, time to surgery and T-upstaging is necessary. In the future, we plan to increase follow-up time to analyse the impact of tumour upstaging on survival for our cohort.

CONCLUSION

Although 56% of patients had T-upstaging when pathological staging after resection was compared with clinical staging based on the last imaging study performed before surgery, TNM upstaging was only observed in 10% of patients with clinical stage IA NSCLC. Importantly, we did not observe any significant correlation between the duration from imaging to lung resection and tumour upstaging, indicating a multifaceted process in which various factors affect tumour development. Additionally, our findings suggest that there may be an increased risk of underestimating ground-glass nodules during the clinical staging process. Further investigations are required to gain a more comprehensive understanding of the factors influencing both staging and tumour progression and an understanding of how neoadjuvant chemotherapy and immunotherapy influence these factors. We plan to follow these patients for an additional 5 years to analyse the impact of delaying the time of surgery on the patients’ prognosis.

Presented at the 31st Annual Meeting of the European Society of Thoracic Surgeons, Milan, Italy, 6 June 2023.

SUPPLEMENTARY MATERIAL

Supplementary material is available at EJCTS online.

FUNDING

This study received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

Conflict of interest: none declared.

DATA AVAILABILITY

The data underlying this article will be shared on reasonable request to the corresponding author.

Author contributions

Rafael R. Barcelos: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Writing—original draft; Writing—review and editing. Evert Sugarbaker: Data curation; Investigation; Writing—review and editing. Kelvin F. Kennedy: Formal analysis; Investigation; Methodology; Writing—review and editing. Miles McAllister: Data curation; Writing—review and editing. Sangmin Kim: Data curation; Writing—review and editing. Julio Herrera-Zamora: Data curation; Writing—review and editing. Rachel Leo: Data curation; Writing—review and editing. Scott Swanson: Conceptualization; Writing—review and editing. Paula Ugalde Figueroa: Conceptualization; Formal analysis; Investigation; Methodology; Supervision; Validation; Writing—review and editing.

Reviewer information

European Journal of Cardio-Thoracic Surgery thanks Alain Jean Poncelet, Claudia Bardoni, Ilies Bouabdallah and the other anonymous reviewers for their contribution to the peer review process of this article.

REFERENCES

ABBREVIATIONS

ABBREVIATIONS
 
• AJCC

  American Joint Committee on Cancer

 
• CT

  Computed tomography

 
• IASLC

  International Association for the Study of Lung Cancer

 
• NCCN

  National Comprehensive Cancer Network

 
• NSCLC

  Non-small-cell lung cancer

 
• PET

  Positron emission tomography

 
• UICC

  Union for International Cancer Control