ABSTRACT
PURPOSE
By correcting the effect of tumor size on metabolic activity, the maximum standardized uptake value-to-tumor size (SUVmax:tumor size) ratio on fluorodeoxyglucose F18 positron emission tomography (18F-FDG PET)/computed tomography (CT) scans can be a prognostic parameter of non-small cell lung cancer (NSCLC). The current study evaluates the prognostic value of SUVmax:tumor size ratio on pretreatment 18F-FDG PET/CT scans in patients with NSCLC. Furthermore, the SUVmax:tumor size ratio is compared with other established PET parameters.
METHODS
This study included 108 patients with NSCLC who underwent pretreatment 18F-FDG PET/CT scans and curative lung surgery. The associations between the SUVmax:tumor size ratio and other conventional PET parameters were investigated. The recurrence-free survival according to the SUVmax:tumor size ratio was also analyzed. In addition, the SUVmax:tumor size ratio was compared according to postoperative pathologic findings.
RESULTS
In total, 72 (66.7%) of the 108 participants presented with adenocarcinoma (ADC). Nineteen (17.6%) patients experienced recurrence during a median follow-up period of 32.3 months. The median SUVmax:tumor size ratio was 2.37 (1.23 for ADCs and 3.90 for other histologic types). The SUVmax:tumor size ratio was associated with SUVmax and mean SUV, as well as metabolic tumor volume and total lesion glycolysis. Patients with an SUVmax:tumor size ratio higher than the median had a worse recurrence outcome than those with an SUVmax:tumor size ratio lower than the median. Participants with ADC who presented with lymphovascular invasion had a higher SUVmax:tumor size ratio than those without. The presence of lymph node metastasis and advanced histologic grade were associated with a high SUVmax:tumor size ratio in patients with ADC.
CONCLUSION
The SUVmax:tumor size ratio on pretreatment 18F-FDG PET/CT scans was associated with aggressive tumor behavior and poor outcome in NSCLCs, particularly ADC.
CLINICAL SIGNIFICANCE
The SUVmax:tumor size ratio on pretreatment 18F-FDG PET/CT scans has a prognostic value in patients with NSCLCs, especially ADC.
Main points
• Fluorodeoxyglucose F18 positron emission tomography (18F-FDG PET)/computed tomography (CT) is a widely used imaging tool for non-small cell lung cancer (NSCLC) management.
• The maximum standardized uptake value (SUVmax)-to-tumor size ratio on 18F-FDG PET/CT scans was introduced to assess the metabolic activity of NSCLC by correcting the effect of tumor size.
• The SUVmax:tumor size ratio on 18F-FDG PET/CT scans was associated with aggressive tumor behavior and poor outcome in NSCLCs, particularly adenocarcinoma.
Despite advancement in prevention, screening, and management in recent decades, non-small cell lung cancer (NSCLC) is among the leading causes of cancer-related mortality worldwide.1 Fluorodeoxyglucose F18 (18F-FDG) positron emission tomography/computed tomography (PET/CT) is a widely used imaging tool for NSCLC management. The role of 18F-FDG PET/CT scanning in determining the status of solitary pulmonary nodules (malignant or not), staging of lung cancer, planning of radiation therapy, and evaluating treatment response is well established.2
The 18F-FDG PET/CT technique is used for prognostic prediction in NSCLC. Studies have shown that 18F-FDG uptake is related to prognosis. Among these studies, one revealed the presence of histological invasion in early-stage adenocarcinoma (ADC).3 In addition, in a meta-analysis of surgically resected NSCLC, a high standardized uptake value (SUV) and other metabolic parameters [e.g., metabolic tumor volume (MTV) and total lesion glycolysis (TLG)] were poor prognosis factors of disease-free survival and overall survival.4 However, the prognostic significance of the maximum SUV (SUVmax), a representative parameter of 18F-FDG uptake in tumors, has not as yet been completely elucidated.5 The inconsistent results can be attributed to various confounding factors affecting the SUVmax.
The SUVmax-to-tumor size ratio (SUVmax:tumor size) on 18F-FDG PET/CT scans was introduced to assess the metabolic activity of NSCLC by correcting the effect of tumor size, an established prognostic factor, on the SUV. Studies have shown that it has a prognostic value in patients with NSCLC.6, 7 Moreover, this indicator does not correspond to the fact that thresholds are not easy to apply when measuring metabolic parameters in the early stages of NSCLC because of the generally low SUV values. However, studies supporting the use of this indicator in clinical settings are still lacking.
The current study investigates the prognostic value of the SUVmax:tumor size ratio on pretreatment 18F-FDG PET/CT scans in patients with NSCLC. Furthermore, the SUVmax:tumor size ratio is compared with other established PET parameters.
Methods
Participants
This study recruited 131 consecutive patients who underwent an 18F-FDG PET/CT scan between March 2020 and December 2021 and lung surgery within 30 days after examination at our institution. Among the patients, 13 with benign postoperative pathologic results, four with small cell lung cancer, three who received neoadjuvant treatment, two with distant metastasis at the time of surgery, and one with myxofibrosarcoma were excluded from the analysis. Finally, 108 patients were included in the study (Figure 1).
The Kangbuk Samsung Hospital Institutional Review Board (decision no: KBSMC 2024-02-040, date: 03/05/2024) approved this study. The need for written informed consent from the participants was waived.
Assessment of the medical records of the patients
The medical records of the patients were assessed for clinical and tumor characteristics, including pathologic findings after surgery, type of treatment, clinical follow-up results, and recurrence diagnosis. Cancer stage was based on the American Joint Committee on Cancer Staging Manual, 8th edition. Lung cancer recurrence was diagnosed via pathologic examination or imaging results assessed by the attending physician. Recurrence-free survival (RFS) was defined as the date of recurrent lesion detection or last follow-up date from the date of surgery
Fluorodeoxyglucose F18 positron emission tomography/computed tomography scan acquisition
The patients fasted for at least 6 h, and each patient’s blood glucose level was <200 mg/dL at the time of 18F-FDG injection. PET/CT scan images were obtained approximately 60 min after the intravenous administration of 18F-FDG at a dose of 2.96 MBq/kg. The Discovery MI system (GE Healthcare, Milwaukee, WI, USA) was used for 18F-FDG PET/CT scan examination. Following the CT scan (120 kVp; 10–80 mA; section thickness: 3.75 mm), emission PET was performed from the thigh to the skull base (1.5 min per bed). The PET images were reconstructed using the Bayesian penalized likelihood reconstruction algorithm with attenuation-corrected
images.
Image analyses
The PET/CT scan images were reviewed on a dedicated workstation (AW; GE Healthcare, Chicago, IL, USA) by a nuclear medicine physician who can recognize lesions that would be surgically resected. A volume of interest was drawn over lung cancer lesions with an 18F-FDG uptake greater than that of surrounding background activity. The semi-automatic method was used to delineate the boundaries of the tumor with the SUV-based contouring software (volume viewer software on GE AW 4.7). The SUV threshold was set to 2.5. Data on the SUVmax, mean SUV (SUVmean), and MTV were recorded. TLG was defined as the product of MTV and SUVmean. If the SUVmax of a lung cancer lesion was <2.5, only SUVmax was measured, and other volumetric PET parameters were omitted. The longest diameter of the lung cancer lesion on 18F-FDG PET/CT scan images was also measured. The SUVmax:tumor size ratio was calculated by dividing the SUVmax of the lung cancer lesion by the longest diameter of the lesion.
Statistical analysis
The clinical characteristics and PET/CT scan parameters of the participants were compared according to histologic findings using the Mann–Whitney U test, chi-square test, or Fisher’s exact test. The association
between the SUVmax:tumor size ratio and other PET parameters was evaluated using Kendall’s tau-b correlation coefficient. Survival analysis was performed using the Kaplan–Meier method with subgroup survival estimates compared using the log-rank test. The Cox proportional hazards regression model was used to analyze the prognostic impact of variables. Finally, the SUVmax:tumor size ratio was compared according to postoperative pathologic findings. In the analysis using SUVmax, cases in which patients had lung cancer lesions without a discernable 18F-FDG uptake were treated as zero.
Jamovi version 2.3.28 was used to perform statistical analysis. A P value of <0.05 was considered statistically significant.
Results
Characteristics of the participants
Table 1 shows the clinical characteristics of all patients. Their median age was 66 years (range: 22–83) and 62 (57.4%) were men. In total, 72 (66.7%) of the 108 participants were diagnosed with ADC, whereas 36 patients presented with histologic findings other than ADC. The findings included squamous cell carcinoma [n = 21 (58.3%)], pleomorphic carcinoma [n = 9 (25.0%)], large-cell carcinoma [n = 4 (11.1%)], adenosquamous carcinoma [n = 1 (2.8%)], and atypical carcinoid [n = 1 (2.8%)]. Patients diagnosed with ADC had a lower median age than those with other histologic types. Furthermore, the proportion of male patients diagnosed with ADC was lower than that of male patients with other histologic types. Overall, 58 (53.7%) of the 108 patients were diagnosed with stage IA or lower lung cancer, and most of them were diagnosed with ADC. In total, 94 (87.0%) of the 108 patients underwent lobectomy, whereas 24 (22.2%) received adjuvant treatment. In total, 19 (17.6%) patients developed recurrence during a median follow-up period of 32.3 months.
Fluorodeoxyglucose F18 positron emission tomography/computed tomography scan parameters
The 18F-FDG uptake in 101 (93.5%) of the 108 patients with lung cancer was discernable. The median SUVmax was 5.71 and the median tumor size was 2.35 cm. The median SUVmax:tumor size ratio was 2.37 (range: 0–25.4). More specifically, the median SUVmax:tumor size ratios were 1.23 for ADCs and 3.90 for other histologic types. Binomial logistic regression analysis of the preoperative SUVmax:tumor size ratio for histologic type other than ADC returned an odds ratio of 2.13 [95% confidence interval (CI): 1.56–2.91]. The area under curve was 0.872, the sensitivity was 50.0%, and the specificity was 86.1 when the cut-off value was 0.5. There were 31 (28.7%) patients with an SUVmax of <2.5, and all these patients were diagnosed with ADC. The volumetric PET parameters of 76 patients were measured. However, in one patient, the volumetric PET parameters could not be measured due to image data issues. Table 2 shows the volumetric PET parameters. Each PET parameter distribution was biased, with a significant frequency in the near-zero interval (Figure 2).
The SUVmax:tumor size ratio was found to be correlated with SUVmax and SUVmean (Kendall’s tau-b = 0.678 and 0.495, respectively; P < 0.001) and MTV and TLG (Kendall’s tau-b = 0.191 and 0.243; P = 0.015 and 0.002, respectively). However, the correlation strength of MTV and TLG was not as strong as that of SUVmax or SUVmean. Similar results were obtained for the ADC subgroup. Nevertheless, the SUVmax:tumor size ratio was not significantly associated with MTV or TLG in other patients with NSCLC.
Maximum standardized uptake value-to-tumor size ratio and recurrence-free survival analysis
Patients with an SUVmax:tumor size ratio higher than the median (2.37) had a worse recurrence outcome than those with an SUVmax:tumor size ratio lower than the median (Figure 3). In patients with ADC, an SUVmax:tumor size ratio of >1.23 was consistently associated with worse outcomes (Figure 4). Based on univariate analysis, the hazard ratio of a high SUVmax:tumor size ratio for recurrence was 6.01 (1.75–20.65, P = 0.004) in all participants and 6.22 (0.75–51.68, P = 0.091) in patients with ADC (n = 72). A high SUVmax:tumor size ratio was associated with poor prognosis based on the multivariate analysis using clinical variables, albeit that it did not reach statistical significance (Table 3).
Association between the maximum standardized uptake value-to-tumor size ratio and aggressive tumor behavior based on pathologic findings following surgery
The preoperative 18F-FDG PET/CT scan parameters according to postoperative pathologic results were compared in patients with ADC and those with other histologic types. In the ADC group, patients with lymphovascular invasion had a higher SUVmax:tumor size ratio than those without (3.53 vs. 0.98, P < 0.001). Patients with lymph node metastasis had a higher SUVmax:tumor size ratio than those without (3.70 vs. 1.04, P = 0.006). An advanced histological grade was associated with a high SUVmax:tumor size ratio in ADC (Figure 5). In patients with ADC, the odds ratios of the SUVmax:tumor size ratio were 1.57 (1.14–2.30, P = 0.010) for lymphovascular invasion, 1.42 (0.99–2.09, P = 0.050) for lymph node metastasis, and 1.92 (1.33–2.98, P = 0.001) for histologic grade 3 lesions.
Unlike the ADC group, the other histologic type group showed no significant differences in SUVmax:tumor size ratio according to the lymphovascular invasion (3.80 vs. 4.50, P = 0.346), the presence of lymph node metastasis (3.77 vs. 5.18, P = 0.220) and histologic grade 3/4 lesions (4.15 vs. 4.82, P = 0.387).
Discussion
The clinical implications for SUVmax:tumor size ratio on pretreatment 18F-FDG PET/CT in surgically resected NSCLC were evaluated. The SUVmax:tumor size ratio was found to be associated with conventional metabolic PET parameters in patients with ADC. An SUVmax:tumor size ratio higher than the median was associated with worse recurrence outcomes. Patients with ADC who presented with lymphovascular invasion, lymph node metastasis, or histologic grade 3 lesions based on the pathologic results had a higher SUVmax:tumor size ratio than those without. However, the results did not significantly differ in the non-ADC group.
In addition to SUVmax, the volumetric PET parameters, including MTV and TLG, for the prognostic stratification of NSCLC have been evaluated.8 However, there is no clear standard method for segmentation to measure the parameters, and the various methods affect the measurements.9 An SUV threshold of 2.5 is widely used to delineate the tumor, and this value was also adopted in this study. However, other methods can be utilized. A specific percentage of SUVmax as a threshold or a gradient method without threshold are also frequently used to define tumors. However, all these methods have a common weakness in tumors with a generally low 18F-FDG uptake, such as ADC.10 In the present study, the SUVmax of 31 (43.1%) of 72 patients with ADCs was <2.5. Hence, volumetric PET parameters were not measured in these patients. In contrast, the SUVmax:tumor size ratio could be measured in most cases and was correlated with the volumetric parameters in ADCs.
In this study, an SUVmax:tumor size ratio higher than the median was associated with a poor RFS. This result is similar to that of previous studies on SUVmax:tumor size ratio.6, 7 Furthermore, in patients with ADC, a higher SUVmax:tumor size ratio was consistently related to worse outcomes. This study did not reach the level of prognostic prediction modeling via multivariate analysis. However, this outcome can be achieved in further studies with a larger number of participants and an extended follow-up period.
A higher SUVmax:tumor size ratio was associated with unfavorable postoperative pathologic findings in ADCs. This is consistent with the explanation than an increase in 18F-FDG uptake is related to the aggressive characteristics of tumor cells.11 Tumor cell density can be another factor affecting the SUVmax:tumor size ratio and is associated with tumor grade in lung ADC.12 However, in the present study, no significant association was found in the remaining NSCLC group except ADC. The lack of association could not be explained based on the small number of participants alone. It seems reasonable to distinguish ADC from the other histologic types when evaluating the prognostic value of 18F-FDG uptake in NSCLC.
Interestingly, in a meta-analysis on the prognostic value of 18F-FDG PET/CT scanning for surgically resected NSCLC, the results presented by classifying the values into lower and higher than an SUVmax of 6.0 had somewhat different patterns.4 The adjusted hazard ratio of the SUVmax for disease-free survival was 4.63 (2.53–8.48) in the sub-threshold group and was higher than 1.68 (95% CI: 1.07–2.63) in the above threshold group. Therefore, the association between the tumor 18F-FDG uptake and prognosis may weaken following disease progress. Furthermore, acidosis can reduce 18F-FDG uptake by inhibiting aerobic glycolysis with cancer progression.13, 14
The current study has several limitations. First, this retrospective study focused on patients with NSCLC who underwent surgical resection, meaning selection bias could have affected the results. The potential value of predicting histologic tumor type by preoperative SUVmax:tumor size ratio could be additionally evaluated using a larger number of participants with various histologic tumor types in the future. In addition, various factors might have influenced the clinical judgement of physicians regarding treatments, such as the type of treatment modality and adjuvant treatment following surgery. Among them, 18F-FDG PET/CT scan findings might have had an influence to an uncontrolled extent. Second, the follow-up duration of this study was limited. Thus, overall survival could not be evaluated. Controlled studies with a larger number of patients and extended follow-up periods should be conducted to validate the prognostic value of the SUVmax:tumor size ratio in NSCLC. Compared with other complicated parameters, the SUVmax:tumor size ratio can be advantageous for multicenter research as it involves less variation based on the institution.
In conclusion, the SUVmax-to-tumor size ratio on 18F-FDG PET/CT scanning was associated with aggressive tumor behavior and poor outcomes in NSCLCs, particularly ADC.
Conflict of interest disclosure
The authors declared no conflicts of interest.