Detection of cancer-associated cachexia in lung cancer patients using whole-body [18F]FDG-PET/CT imaging: A multi-centre study

Abstract

Background: Cancer-associated cachexia (CAC) is a metabolic syndrome contributing to therapy resistance and mortality in lung cancer patients (LCP). CAC is typically defined using clinical non-imaging criteria. Given the metabolic underpinnings of CAC and the ability of [¹⁸F]fluoro-2-deoxy-D-glucose (FDG)-positron emission tomography (PET)/computer tomography (CT) to provide quantitative information on glucose turnover, we evaluate the usefulness of whole-body (WB) PET/CT imaging, as part of the standard diagnostic workup of LCP, to provide additional information on the onset or presence of CAC.

Methods: This multi-centre study included 345 LCP who underwent WB [¹⁸F]FDG-PET/CT imaging for initial clinical staging. A weight loss grading system (WLGS) adjusted to body mass index was used to classify LCP into 'No CAC' (WLGS-0/1 at baseline prior treatment and at first follow-up: N = 158, 51F/107M), 'Dev CAC' (WLGS-0/1 at baseline and WLGS-3/4 at follow-up: N = 90, 34F/56M), and 'CAC' (WLGS-3/4 at baseline: N = 97, 31F/66M). For each CAC category, mean standardized uptake values (SUV) normalized to aorta uptake (<SUV_aorta>) and CT-defined volumes were extracted for abdominal and visceral organs, muscles, and adipose-tissue using automated image segmentation of baseline [¹⁸F]FDG-PET/CT images. Imaging and non-imaging parameters from laboratory tests were compared statistically. A machine-learning (ML) model was then trained to classify LCP as 'No CAC', 'Dev CAC', and 'CAC' based on their imaging parameters. SHapley Additive exPlanations (SHAP) analysis was employed to identify the key factors contributing to CAC development for each patient.

Results: The three CAC categories displayed multi-organ differences in <SUV_aorta>. In all target organs, <SUV_aorta> was higher in the 'CAC' cohort compared with 'No CAC' (P < 0.01), except for liver and kidneys, where <SUV_aorta> in 'CAC' was reduced by 5%. The 'Dev CAC' cohort displayed a small but significant increase in <SUV_aorta> of pancreas (+4%), skeletal-muscle (+7%), subcutaneous adipose-tissue (+11%), and visceral adipose-tissue (+15%). In 'CAC' patients, a strong negative Spearman correlation (ρ = -0.8) was identified between <SUV_aorta> and volumes of adipose-tissue. The machine-learning model identified 'CAC' at baseline with 81% of accuracy, highlighting <SUV_aorta> of spleen, pancreas, liver, and adipose-tissue as most relevant features. The model performance was suboptimal (54%) when classifying 'Dev CAC' versus 'No CAC'.

Conclusions: WB [¹⁸F]FDG-PET/CT imaging reveals groupwise differences in the multi-organ metabolism of LCP with and without CAC, thus highlighting systemic metabolic aberrations symptomatic of cachectic patients. Based on a retrospective cohort, our ML model identified patients with CAC with good accuracy. However, its performance in patients developing CAC was suboptimal. A prospective, multi-centre study has been initiated to address the limitations of the present retrospective analysis.

Keywords: Cachexia; Lung cancer; Metabolism; PET/CT; [18F]Fluoro‐2‐deoxy‐D‐glucose.

Figure 1

Flow chart for inclusion and stratification of lung cancer patients. [¹⁸F]FDG‐PET/CT, [¹⁸F]fluoro‐2‐deoxy‐D‐glucose positron emission tomography/computer tomography; CAC, cancer‐associated cacWB, whole‐body; WLGS, weight loss grading system.

Figure 2

Mean SUV_aorta distributions in target organs for ‘No CAC’ (white), ‘Dev CAC’ (grey) and ‘CAC’ (black) cohorts. Significant differences are indicated with stars (*P < 0.05, **P < 0.01, ***P < 0.001).

Figure 3

Correlations among imaging parameters (volumes and aorta>) of the target regions in the ‘No CAC’ (upper), ‘Dev CAC’ (middle) and ‘CAC’ (lower) cohorts. In the chord plots, external nodes represent the imaging parameters, while the thickness of the internal curves indicates the strength of the correlation. (A) All significant correlations between the parameters and regions considered. (B, C) Connectivity profiles of subcutaneous adipose tissue <SUV_aorta> and visceral adipose tissue <SUV_aorta>, respectively. SUV, standardized uptake value.

Figure 4

CatBoost classifier ROC curve (A) and SHAP analysis (B) for the binary classification between ‘No CAC’ and ‘CAC’ cohorts. The position of the dots to the left or right in the SHAP plot (B) indicates their influence toward a ‘No CAC’ or ‘CAC’ classification, respectively. The colour of the dots indicates the absolute value of each feature: Blue for lower and pink for higher values. ASAT, aspartate aminotransferase; BMI, body mass index, SUV, standardized uptake value; Vol, volume.

Figure 5

CatBoost classifier ROC curve (A) and SHAP analysis (B) for the binary classification between ‘No CAC’ and ‘Dev CAC’ cohorts. The position of the dots to the left or right in the SHAP plot (B) indicates their influence toward a ‘No CAC’ or ‘Dev CAC’ classification, respectively. The colour of the dots indicates the absolute value of each feature: Blue for lower and pink for higher values. ASAT, aspartate aminotransferase; BMI, body mass index, SUV, standardized uptake value; Vol, volume.