Impact of Primary Tumor Diameter and SUV_max on Pathological Lymph Node Involvement in Non-small Cell Lung Cancer

Non-small-cell lung cancer (NSCLC) accounts for 85% of all lung cancer cases. Approximately 40% of cases are metastatic upon diagnosis [1, 2]. Curative treatment modalities (surgery, chemoradiotherapy) are chosen based on the stage of the remaining non-metastatic patients.

The precise staging of NSCLC, particularly preoperative identification of lymph node (LN) metastasis, is critical for treatment planning and prognosis determination.

LN metastasis is a significant prognostic factor in patients with NSCLC because it is associated with a worse prognosis. This is because LN metastases often indicate that the cancer has spread beyond the lung to other parts of the body.

Patients with NSCLC who have larger tumors have an increased risk of LN metastasis. This is because larger tumors are more likely to invade the lymphatic system, which is a network of vessels containing white blood cells and other immune cells throughout the body.

LN metastasis can be determined by invasive or non-invasive methods. Non-invasive imaging methods, such as computed tomography (CT) and ¹⁸F-fluorodeoxyglucose positron emission tomography-CT (¹⁸F-FDG PET/CT), are widely used to identify involvement of the ipsilateral (N2) or contralateral mediastinal LNs (N3). Mediastinoscopy and endobronchial ultrasound-guided transbronchial needle aspiration are two invasive staging procedures that are frequently used after non-invasive staging results from CT and ¹⁸F-FDG PET/CT scans. ¹⁸F-FDG PET/CT using ¹⁸F-FDG, a glucose analog shown to be useful in detecting malignancy, is an extremely important imaging method in the staging of NSCLC. Here, the maximum standardized uptake value (SUV_max) of the primary tumor and primary tumor diameter (PTD) can be used to predict the risk of nodal metastasis, and the threshold SUV_max value is between 2.5 and 4.0 [3].

Although invasive methods remain the gold standard for mediastinal nodal staging, they come with disadvantages, such as the risk of complications, delay in treatment, and high cost. Hence, the correct use of non-invasive mediastinal nodal staging will help in selecting patients who will benefit most from invasive staging and preventing unnecessary invasive procedures. However, in patients clinically staged as N0 (no nodal involvement) on preoperative ¹⁸F-FDG PET/CT, the incidence of histologically proven occult nodal metastasis after surgery has been found to range between 8% and 58% in various stages [4, 5].

The relationship between PTD and LN metastases in clinical staging and the proper selection of treatment methods has been investigated in many previous studies. Some studies have revealed a close relationship between these two factors, suggesting that PTD could be used to predict positive LN metastases [6, 7]. However, others have concluded that there is no such specific correlation between these two factors [8, 9]. It is believed that combining the PTD and SUV_max values may provide a more sensitive method for detecting pathological LN metastasis.

In this study, we aimed to determine the role of the PTD and SUV_max in preoperative ¹⁸F-FDG PET/CT in predicting regional LN involvement, lymphatic invasion (LI), vascular invasion (VI), and pleural invasion (PI) in patients with NSCLC who were operated on without neoadjuvant therapy.

The computer registration system of our hospital was reviewed, and the archive file records of 70 patients who were diagnosed with NSCLC at the medical oncology clinic between October 2014 and February 2023 and who underwent surgery (segmentectomy, lobectomy, bilobectomy, pneumonectomy) after ¹⁸F-FDG PET/CT without receiving neoadjuvant therapy were retrospectively examined. The effects of PTD and SUV_max on postoperative LN, lymphatic, vascular, and PI in patient groups below and above the determined threshold value on preoperative ¹⁸F-FDG PET/CT were compared. Since an optimal cut-off value for specificity and sensitivity was not obtained with the receiver operating characteristic (ROC) curve for both the PTD and SUV_max of the primary tumor, patients were grouped based on the median values for the two parameters.

The demographic and clinical characteristics of the patients included in the study are presented in Table 1. Parameters such as PTD and SUV_max in preoperative ¹⁸F-FDG PET/CT and histopathological diagnosis of the tumor in postoperative pathology reports, LN involvement, LI, VI, and PI were recorded. The effects of PTD and SUV_max on pathological LN involvement, LI, VI, and PI were compared between patient groups below and above the determined threshold value. The median follow-up period was 30.4 (0.17-112.7).

This study was planned and conducted in accordance with Good Clinical Practices and the Declaration of Helsinki, and it was approved by the Ethics Committee of Aydın Adnan Menderes University Hospital (approval no: E-53043469-050.04.04-346879, date: 11.05.2023).

Statistical Analysis

Data were summarized using descriptive statistics, such as mean, standard deviation, median, minimum, maximum, frequency, and ratio. The distribution of variables was assessed using the Kolmogorov-Smirnov test. Independent quantitative data were analyzed using independent sample t-tests and Mann-Whitney U tests. The chi-square test was employed for qualitative independent data, with the Fisher’s test used when conditions for the chi-square test were not met. ROC curve analysis was performed to maximize the sensitivity, specificity, and accuracy for assessing the cut-off value for PTD and SUV_max. A p value of less than 0.05 was considered statistically significant. Statistical Package for the Social Sciences 22.0 software was used for the analyses.

For the PTD and SUV_max values of the primary tumor, since a cut-off value could not be obtained with the ROC curve with optimal specificity and sensitivity, patients were divided into 2 groups based on the median values.

There was no significant difference in age between patients with a PTD of ≥32 mm and <32 mm (p=0.073). In the group with ≥32 mm, the male gender was found to be more predominant (p=0.040) (Table 2).

The number of patients who smoked (p=0.164), the smoking duration (p=0.506), Eastern Cooperative Oncology Group (ECOG) status (p=0.329), tumor localization (p=0.203), and histological subtype (p=0.150) did not significantly differ between the groups. The number of patients who received adjuvant chemotherapy was significantly higher in the group with a PTD of ≥32 mm (p=0.000). There was no significant difference in the number of patients who received adjuvant radiotherapy (p=0.495, Table 2).

Age was not significantly different (p=0.073) in the groups of patients with SUV_max values ≥12.55 and <12.55 for the primary tumor. Regarding gender distribution, male sex was more predominant (p=0.040) in the group with SUV_max ≥12.55 (Table 3).

No significant differences were observed between the groups in the number of patients who smoked (p=0.643), duration of smoking (p=0.948), ECOG status (p=0.626), tumor localization (p=0.203), or histological subtype (p=0.055). In the group with a primary tumor SUV_max value ≥12.55, the number of patients who received adjuvant chemotherapy was significantly higher (p=0.002). Adjuvant radiotherapy was not significantly different between the number of patients who received it (p=1.000) (Table 3).

Pathological LN involvement (p=0.322), VI (p=0.122), LI (p=0.122), and PI (p=1.000) were not significantly different between the two groups according to PTD (Table 4).

Regarding the SUV_max values of the primary tumor, no significant difference was found between the two groups in terms of the pathological LN involvement (p=0.621), VI (p=0.122), LI (p=0.122) and PI (p=1.000) (Table 5).

A low positive correlation was observed between the PTD and SUV_max values (p=0.000, r=0.447) (Figure 1).

Our study showed that the preoperative PTD and SUV_max on ¹⁸F-FDG PET/CT were insufficient to predict pathological LN involvement in patients with NSCLC.

In clinical stage 1A patients, the prevalence of LN metastasis is low, and because of high false-positive rates, the role of ¹⁸F-FDG PET/CT is limited [10, 11]. Additionally, a study has shown that the utility of ¹⁸F-FDG PET/CT in predicting LN metastasis in subsolid lesions significantly decreases, with a sensitivity of 11.1%, specificity of 86.1%, and accuracy of 81.9% for LN staging in patients with subsolid adenocarcinomas (AC) with a PTD ≤3 cm [12].

This study showed that the incidence of LN metastasis in lung cancer increased with tumor size. Seok et al. [13] found that when patients were split into six groups based on tumor diameter: ≤0.5, 0.6-1, 1.1-1.5, 16-2.0, 2.1-2.5, and 2.6-3 cm, the rates of LN metastasis were 0%, 0%, 7%, 14%, 27%, and 31%. However, our study has shown similar results in terms of PTD among patients with NSCLC.

¹⁸F-FDG PET/CT is a valuable tool in oncology practice because it can be used to measure increased glucose metabolism in tumor cells. This is because tumor cells typically have a higher metabolic rate than normal cells, and they take up more glucose to fuel their growth. A SUV_max >2.5 is generally used as a cut-off value for malignancy. This indicates that lesions or tissues with an SUV_max >2.5 are more likely to be cancerous than those with an SUV_max 2.5. However, some benign conditions can also cause high ¹⁸F-FDG uptake, and there are some cancers that may have an SUV_max below 2.5.

Tuberculosis, obstructive pneumonia, anthracosis, or immune reactions resulting from granulomatous inflammation can lead to false-positive results. Low-grade malignancies can also result in false negatives [14, 15]. However, it is important to note that using a higher SUV_max cut-off value can also increase the false positive rate. This means that more patients will be identified as having metastasis even if they do not actually have metastasis. This can lead to unnecessary additional testing and procedures.

Some authors have claimed that using a higher SUV_max than the traditional 2.5 can increase the accuracy of metastasis presence. This is because a high SUV_max is more likely to indicate malignant tissue. However, it is important to note that no single SUV_max value is universally accurate for all cancer types. The optimal cut-off value varies depending on the specific type of cancer, tumor location, and other factors [16].

The SUV_max value was also a good indicator of tumor aggressiveness [17]. However, a study conducted in patients with resected stage 1 NSCLC found that the cellular components in the tumor varied widely and therefore did not correlate with 18F-FGD activity [18].

Tumor size and the presence of necrosis are other factors affecting a tumor’s SUV_max. Previous studies have shown a positive correlation between PTD and SUV_max [19, 20]. Similarly, in our study, a positive correlation was found between the PTD and SUV_max. However, there was no significant difference in pathological LN involvement between the two groups based on the SUV_max of the main tumor.

It is known that the likelihood of necrosis increases with tumor size, which is associated with a lower SUV_max. However, the relationship between SUV_max and necrosis was not evaluated in our study.

Tumor histologic type has also been identified to be associated with LN metastasis in patients with NSCLC. It is known that squamous cell cancers (SCC) have shorter doubling times and grow faster than AC, and therefore have higher glucose metabolism [21]. Moreover, glucose transporter-1 expression is higher in SCCs. Accordingly, previous studies have shown that SUV_max values are higher in SCC [22-24]. Kim et al. [25] found that the SUV_max was 10.8±4.4 for SCC and 8.8±3.2 for AC.

However, contrary to this, our study showed no significant difference in PTDs and SUV_max values according to histological subtype.

In studies involving patients with early-stage NSCLC, a high primary tumor SUV_max was found to be a strong indicator of lymphovascular invasion (LVI) and LN metastasis [23, 26]. In a study conducted by Koksal et al. [27], it was found that the SUV_max of the tumor did not correlate with the nodal (N) stage, and the median SUV_max did not show a significant difference between groups with and without LN metastasis. Similarly, in our study, the tumor SUV_max was insufficient to predict pathological LN metastasis. Additionally, no significant relationship was observed between SUV_max and LVI or PI.

Study Limitations

The important limitations of our study are its non-randomized and retrospective nature. In addition, the small number of patients might have affected the study outcome. Therefore, we believe that prospective studies with large cohorts are needed to evaluate the utility of preoperative ¹⁸F-FDG PET/CT in predicting LN involvement.

Ultimately, the decision to use a higher SUV_max cut-off value should be made on a case-by-case basis in consultation with a medical oncologist.