Early Skeletal Muscle Loss in Non-Small Cell Lung Cancer Patients Receiving Chemoradiation and Relationship to Survival.

PURPOSE: Sarcopenia is associated with reduced survival in cancer. Currently, data on sarcopenia at presentation and muscle loss throughout treatment are unknown in patients receiving chemoradiation therapy (CRT) for non-small cell lung cancer (NSCLC). This study evaluated skeletal muscle changes in NSCLC patients receiving CRT and relationship with survival. METHODS: Secondary analysis of 41 patients with NSCLC treated with CRT assessed for skeletal muscle area and muscle density by computed tomography pre-treatment and 3 months post-treatment. Images at week 4 of treatment were available for 32 (78%) patients. Linear mixed models were applied to determine changes in skeletal muscle over time and related to overall survival using Kaplan-Meier plots. RESULTS: Muscle area and muscle density decreased significantly by week 4 of CRT (- 6.6 cm2, 95% CI - 9.7 to - 3.1, p < 0.001; - 1.3 HU, 95% CI - 1.9 to - 0.64, p < 0.001, respectively), with minimal change between week 4 of CRT and 3 months post-CRT follow-up (- 0.2 cm2, 95% CI - 3.6-3.1, p = 0.91; - 0.27, 95% CI - 0.91-0.36, p = 0.36, respectively). Sarcopenia was present in 25 (61%) and sarcopenic obesity in 6 (14%) of patients prior to CRT, but not associated with poorer survival. Median survival was shorter in patients with low muscle density prior to treatment although not statistically significant (25 months + 8.3 vs 53 months + 13.0, log-rank p = 0.17). CONCLUSION: Significant loss of muscle area and muscle density occurs in NSCLC patients early during CRT. A high proportion of patients are sarcopenic prior to CRT; however, this was not significantly associated with poorer survival.

A Deep Learning Model to Automate Skeletal Muscle Area Measurement on Computed Tomography Images.

BACKGROUND: Muscle wasting (Sarcopenia) is associated with poor outcomes in cancer patients. Early identification of sarcopenia can facilitate nutritional and exercise intervention. Cross-sectional skeletal muscle (SM) area at the third lumbar vertebra (L3) slice of a computed tomography (CT) image is increasingly used to assess body composition and calculate SM index (SMI), a validated surrogate marker for sarcopenia in cancer. Manual segmentation of SM requires multiple steps, which limits use in routine clinical practice. This project aims to develop an automatic method to segment L3 muscle in CT scans. METHODS: Attenuation correction CTs from full body PET-CT scans from patients enrolled in two prospective trials were used. The training set consisted of 66 non-small cell lung cancer (NSCLC) patients who underwent curative intent radiotherapy. An additional 42 NSCLC patients prescribed curative intent chemo-radiotherapy from a second trial were used for testing. Each patient had multiple CT scans taken at different time points prior to and post- treatment (147 CTs in the training and validation set and 116 CTs in the independent testing set). Skeletal muscle at L3 vertebra was manually segmented by two observers, according to the Alberta protocol to serve as ground truth labels. This included 40 images segmented by both observers to measure inter-observer variation. An ensemble of 2.5D fully convolutional neural networks (U-Nets) was used to perform the segmentation. The final layer of U-Net produced the binary classification of the pixels into muscle and non-muscle area. The model performance was calculated using Dice score and absolute percentage error (APE) in skeletal muscle area between manual and automated contours. RESULTS: We trained five 2.5D U-Nets using 5-fold cross validation and used them to predict the contours in the testing set. The model achieved a mean Dice score of 0.92 and an APE of 3.1% on the independent testing set. This was similar to inter-observer variation of 0.96 and 2.9% for mean Dice and APE respectively. We further quantified the performance of sarcopenia classification using computer generated skeletal muscle area. To meet a clinical diagnosis of sarcopenia based on Alberta protocol the model achieved a sensitivity of 84% and a specificity of 95%. CONCLUSIONS: This work demonstrates an automated method for accurate and reproducible segmentation of skeletal muscle area at L3. This is an efficient tool for large scale or routine computation of skeletal muscle area in cancer patients which may have applications on low quality CTs acquired as part of PET/CT studies for staging and surveillance of patients with cancer.