The Utility of FDG-PET for assessment of nodal staging in subsolid clinical stage IA (T1N0) lung adenocarcinoma

D. Narasimhasanth. Kamtam¹, J. B.. Shrager¹, N. S. Lui², I. A. Elliott³, D. Z. Liou¹, H. Haiwei. Guo², L. M. Backhus⁴, M. F. Berry^2
1Stanford University School of Medicine, Stanford, California ²Stanford University, Stanford, California ³Stanford, Cupertino, California ⁴Stanford University, Dept. of Cardiothoracic Surgery, Stanford, California

Purpose: Positron-emission tomography (PET), along with computed tomography (CT), is considered the standard of care for clinical staging of lung cancer patients. This study assessed FDG-PET use in evaluating subsolid lung adenocarcinomas, where the utility is less clear as these typically more indolent cancers often exhibit low FDG avidity.

Methods: Patients with cT1N0 subsolid lung adenocarcinomas were selected from a retrospective review of 1101 patients who underwent surgical resection of lung adenocarcinoma from 1/2006 through 6/2022. Patients with clinical/pathological tumor size >30mm, ipsilateral hilar/central lymph nodes (LN)>1cm on CT, and purely solid tumors were excluded. Clinical LN status was stratified into negative or positive based on SUVmax (threshold≥2.5 or greater than mediastinal blood pool). Positives reported as non-specific/reactive were considered negative, whereas those reported as indeterminate/suspicious for malignancy were considered PET LN-positive. The sensitivity and specificity of PET for determining pathological LN status were assessed. Additionally, the optimal thresholds of %solid nodule component (%SNC) and tumor size with the highest degree of association with PET LN positivity were determined by identifying the maximum χ2 value among associations between the subgroups and PET LN positivity based on thresholds of progressively increasing %SNC (0.1 to 0.9) and tumor size (0.5cm to 2.5cm).

Results: FDG-PET was available in 499(93.3%) of 535 patients. The pathological LN-positive rate was 8.5% (46/535). There was no significant difference in the disease-free survival (DFS) between individuals who underwent PET scans and those who did not [113.84±1.23 vs. 115.84±3.1 months, p=0.51], but patients with PET scans were older with larger and more solid tumors. The specificity for PET in predicting pathological node positivity was 92.8%, but sensitivity was only 21.7%. Multivariate analysis demonstrated tumor size (OR 1.085;95%CI, 1.019-1.155, p=0.011), %SNC (OR 1.015;95%CI, 1.002-1.029, p< 0.001), and age (OR 1.055; 95%CI, 1.008-1.104, p=0.021) as the only independent predictors of LN positivity on PET (Table 1). Tumor size (AUROC=0.656,95%CI:0.57-0.74,p=0.003), age (AUROC=0.659,95%CI:0.56-0.76,p=0.003), and %SNC (AUROC=0.62, 95%CI:0.53-0.71,p=0.046) demonstrated moderate accuracy for predicting LN positivity on PET. Subgroup analysis based on progressively increasing %SNC and tumor size demonstrated %SNC≥0.6 [8.4%(26/309) vs. 2.7%(6/226),p=0.006] and tumor size≥19mm [10%(23/230) vs. 3.0%(9/305),p < 0.001] had the most significant association with PET LN positivity. Conversely, 167(33.4%) patients who underwent PET scanning had 218 incidentally detected hypermetabolic lesions unrelated to lung cancer, out of which only 8(4.7%) were proven to be pre-malignant/malignant [thyroid carcinoma (2), breast cancer (1), renal cell carcinoma (1), colon cancer/polyps (2), lymphoma (1), pancreatic neoplasm (1)] that required further therapy.

Conclusion: FDG-PET scan use for subsolid lung adenocarcinomas has high specificity but limited sensitivity for predicting pathological LN positivity and is not associated with increased DFS. FDG-PET scans for these cancers more often detect clinically unimportant findings rather than change lung cancer management, particularly for smaller and less-solid tumors.

Identify the source of the funding for this research project: None