Introduction
Lung cancer is one of the most common cause of cancer incidence and mortality worldwide [1]. The treatment and prognosis for the non-small cell lung cancer (NSCLC) are poor and identifying the prognostic factors for these patients is challenging and in clinical interest [2]. Commonly used tumor-nodes-distant metastasis (TNM) classification still remains the primary and mostly independent prognostic factor for overall survival (OS) in NSCLC patients [3]. Standard imaging techniques in staging and assessing therapy response in NSCLC includes computed tomography (CT), rentgenography (RTG), endoscopic examination, endobronchial ultrasonography (EBUS) or esophageal ultrasonography (EUS) and, more recently, positron emission tomography/computed tomography with the most commonly used radiopharmaceutical — 18F-fluoro-2-deoxy-D-glucose positron emission tomography/computed tomography ([18F]FDG PET/CT) which has been a helpful imaging modality over the past decade in diagnosis and assessing therapy response in NSCLC [2, 3]. Overall survival is highly dependent on the stage of the disease and, according to some authors, preoperative [18F]FDG uptake in primary tumor is associated with OS and time to recurrence (TTR) [2, 4]. Moreover, an [18F]FDG PET-derived parameters provide additional information to the TNM stage, especially metabolic parameters of the tumor expressed with metabolic tumor volume (MTV) and total lesion glycolysis (TLG) vary with TNM stage and thus can be used as a biological description system for lung cancer [5].
The aim of the study was to assess [18F]FDG PET-derived parameters of the primary tumor on overall survival in patients with cN1-cN3 NSCLC.
Materials and methods
Patients characteristics
In the present study 59 consecutive patients (35 male, 24 female) with untreated NSCLC were included retrospectively. All patients gave their informed consent for the examination. Most of the patients (n = 44) received an curative treatment, while in 12 patients with N3 stage and 3 with T4 stage, because of the tumor involvement, palliative treatment was performed. A summary of patient and tumor characteristics is given in Table 1. Patients with the presence of distant metastases were excluded from the analysis. All patients had been fasting for at least 6 hours before the examination (average glucose level was 102.91 ± 23.41 mg/dL). OS was defined as time from primary diagnosis (taken from Greater Poland Cancer Registry) to death or the last info. Patients’ clinical stage was defined using the TNM 8th edition. Ethical approval as well as Bioethics Committee approval was waived, because of the retrospective nature of the study.
|
Characteristics |
Value |
|
Age (years) |
|
|
Mean ± SD Median |
68 ± 11 68 |
|
Sex |
|
|
Male |
35 (59.3) |
|
Female |
24 (40.7) |
|
Curative treatment |
|
|
Yes |
44 (75) |
|
No |
15 (25.4) |
|
Surgery |
|
|
Yes |
9 (15) |
|
No |
50 (84.7) |
|
T stage |
|
|
T1 |
8 (13.6) |
|
T2 |
13 (22) |
|
T3 |
17 (28.8) |
|
T4 |
21 (35.6) |
|
N stage |
|
|
N1 |
20 (33.9) |
|
N2 |
18 (30.5) |
|
N3 |
21 (35.6) |
|
M stage |
|
|
M0 |
59 (100) |
|
UICC stage |
|
|
II |
8 (13.6) |
|
III |
51 (86.4) |
[18F]FDG PET acquisition
All patients underwent a hybrid [18F]FDG PET/CT scan prior to therapy. [18F]FDG PET/CT scans (3D PET acquisition, 90 s per bed position) were performed on a Gemini TF PET/CT (Philips Healthcare, Best, The Netherlands). Data acquisition started 45 ± 25 min (50–70) after intravenous (i.v.) injection of [18F]FDG with mean activity of 364 ± 75 MBq. Scans were performed from the skull vertex to mid-thigh with scan time 1.30 min per table. CT scans (100–150 mAs, 120 kV, slice thickness of 5 mm) were performed before PET imaging without changing the patient’s position. Tomographic images were reconstructed using the BLOB-OS-TF reconstruction (3 iterations, 33 subsets) and CT based attenuation correction.
Image analysis
The metabolically active part of the primary tumor was delineated in the PET data by an automatic algorithm based on adaptive thresholding considering the local background [6, 7]. For the resulting regions of interest (ROI) the maximum and mean standardized uptake value (SUVmax and SUVmean), the metabolic active tumor volume (MTV) and the total lesion glycolysis (TLG = MTVxSUVmean) were computed. ROI definition and analysis was performed using the ROVER software, version 3.0.62 (ABX, Radeberg, Germany).
Statistical analysis
Survival analysis was performed with respect OS. The association of OS with clinical as well as quantitative PET parameters was analyzed using univariate Cox proportional hazard regression in which the PET parameters were included as metric parameters. PET parameters showing a significant effect in this analysis were further analyzed in univariate Cox regression using binarized PET parameters. The cutoff values were calculated by minimizing the p-value in univariate Cox regression as described in [8]. The probability of survival was computed and rendered as Kaplan-Meier curves. Independence of parameters was analyzed by multivariate Cox regression.
Statistical significance was assumed at a P-value of less than 0.05. Statistical analysis was performed with the R language and environment for statistical computing version 4.1.1 [9].
Results
Patient and tumor characteristics is presented in Table 1.
Univariate Cox analysis of clinical parameters and metric PET parameters revealed
MTV as a significant prognostic factor for OS (p = 0.024). TLG and curative treatment showed a trend for significance (p < 0.1) (Tab. 2). Other investigated parameters did not reach significance and were, therefore, not further analysed.
|
Parameter |
HR |
95% CI |
p-value |
|
Sex male |
1.44 |
0.79–2.63 |
0.23 |
|
Age > 68y |
0.87 |
0.48–1.57 |
0.64 |
|
Curative treatment |
0.57 |
0.3–1.1 |
0.096 |
|
Surgery |
0.67 |
0.3–1.5 |
0.33 |
|
T-stage > 3 |
1.62 |
0.88–2.97 |
0.12 |
|
N-stage > 1 |
1.41 |
0.75–2.65 |
0.29 |
|
UICC-stage > II |
1.38 |
0.58–3.28 |
0.46 |
|
MTV |
1.01 |
1–1.02 |
0.024 |
|
TLG |
1.001 |
1–1.002 |
0.068 |
|
SUVmax |
0.97 |
0.91–1.03 |
0.37 |
|
SUVmean |
0.96 |
0.87–1.07 |
0.44 |
After binarization also TLG was a significant factor for OS (HR = 2.06, 95% CI: 1.05–4.05, p = 0.035) in univariate analysis. However, this analysis revealed a notably larger HR for MTV (HR = 3.08, 95% CI: 1.46–4.05, p = 0.003) compared to TLG. Corresponding Kaplan-Meier curves are shown in Figure 1. In multivariate Cox regression (MTV and curative treatment) MTV remained a significant factor (HR = 1.01, 95% CI: 1.0–1.02, p = 0.047) indicating its independent prognostic value.
Discussion
Multiple authors in their works showed that MTV and TLG are independent prognostic factors for OS. Nappi et al. in their study conducted on a group of 103 NSCLC patients staged IIIB and IV showed that primary tumor SUVmax, up to 6.3, MTV up to 8.4 cm3 and TLG up to 259 is associated with worse progression free survival in NSCLC patients [2]. Moreover, they also noted that in patients with lymph node involvement primary tumor MTV showed significantly (p < 0.05) better outcome in those with lower values compared to those with higher primary MTV values. The cut-off value for discrimination of these two groups was 10.9 cm3 (94% sensitivity, 54% specificity) [2]. Our study in multivariate COX analysis discriminate 1 potential PET-derived parameters for OS in NSCLC patients with N1-N3 lymph node involvement. The MTV value obtained in our study varies from those presented by Nappi et al., which might be caused by the notably smaller group of patients in this study.
Wang et al. on a group of 92 nonsurgical NSCLC patients showed in univariate analysis that OS was associated with MTV (cut-off 10 ml) and TLG (100 g) [10]; however, in multivariate analysis OS was not essential to any of [18F]FDG PET-derived parameters, neither for primary tumor nor lymph nodes [16]. In our study we observed that only MTV remained a significant independent prognostic value in respect to OS in NSCLC patients.
Chardin et al. indicate that poor OS is associated with primary tumor MTV above 36.5 cm3 (p < 0.001) and TLG above 267 (p < 0.001) [11], while Kwon et al. noted that patients in stage I NSCLC and tumor size > 3 cm and SUVmax > 9 of primary tumor showed a poor 3-year survival rate [4]. 41% of patients with SUVmax value between 7.2 and 14.2 were still alive, while in SUVmax with less than 3.4 — 77% of patients showed 5-year survival rate [4]. This study indicate that only primary tumor MTV is a significant prognostic factor in stage cN1-cN3 NSCLC patients, while other clinical and PET-derived parameters did not show any association with OS in the analysed group.
In a recent published meta-analysis by Pellegrino et al. it was noted that volumetric PET-based parameters like MTV and TLG were relevant prognostic factors in NSCLC patients either in staging, after induction therapy or in the assessing response to applied therapy and are better in determination of OS and PFS than SUVmax [12]. They also noted that the higher SUVmax, MTV and TLG of the primary tumor caused the risk of recurrence or death increase in NSCLC patients. Im et al. in their study showed that TLG as well as MTV were strong predictors in early and advanced stages [13]. In another study Hyun et al. performed an analysis on a group of 161 patients with stage IIIA-N2 NSCLC and noticed that T stage is associated with OS and SUVmax value of the primary tumor with DFS [14]. Our analysis showed that no clinical parameters were significant for OS, while from PET-derived parameters only MTV of the primary tumor was significant in stage cN1–cN3 non-small-cell lung cancer.
Ma et al. on a group of 203 NSCLC patients showed that MTV is associated with OS in an early stage of disease, while no significant differences were noted in a late stage in Cox multivariate analysis [15]. Machtay et al. proceeded a large prospective and multi-centre study on a group of 250 stage III NSCLC patients and showed that pre-treatment SUVmax is not associated with survival rates in these patients [16]. Similarly, in a recent study, even on a notably lower group of patients, we concluded the same results as presented above: SUVmax did not show any association with respect to OS, while MTV was significant.
A limitation of this study is that it was conducted on a small group of patients and it was a retrospective study. Furthermore, all patients were diagnosed with NSCLC; however, we don’t have data about histology of NSCLC: squamous cell carcinoma or adenocarcinoma, which might also have an influence on the results obtained. Nevertheless, our study showed comparable results in terms of MTV of the primary tumor, which might be used to stratify patients in cN1–cN3 NSCLC.
Conclusion
The metabolic volume of the primary tumor is an independent prognostic factor in NSCLC patients with cN1-cN3 lymph node involvement and should be taken into account in assessing OS in these patients. Commonly used SUVmax of the primary tumor was not a predictor in the assessed group. Further studies on a larger and homogenous group of patients are needed to confirm obtained results.
Conflicts of interest
Authors declare no conflict of interest.
Funding
None declared.
Ethical approval
Ethical approval was not necessary for the preparation of this article because of the retrospective nature of the study.