We read with great interest the recently published article by Ferrara et al. [1] in the Journal of Cachexia, Sarcopenia and Muscle. Based on whole-body [¹⁸F]FDG-PET/CT imaging, the study demonstrated the groupwise differences in the multi-organ metabolism of lung cancer patients (LCP) with and without cancer-associated cachexia (CAC), thus highlighting systemic metabolic aberrations symptomatic of cachectic patients and identifying LCP with CAC accurately by machine-learning model. We commend the authors for their valuable contributions and offer several suggestions which could further enhance the interpretation of these findings and provide valuable direction for future research.

Firstly, some detailed covariates related to comorbidities and lifestyle factors may not have been fully considered in the baseline analysis. Cachexia is a multifactorial syndrome influenced by cancer itself as well as comorbidities and lifestyle factors, including smoking, alcohol use and dietary pattern. Including the detailed covariates and performing subgroup analyses would provide more granular insights into how these variables interact with cachexia. Considering these factors can potentially reveal which subgroups may benefit the most from this analysis and further enhance the robustness of the study results.

Secondly, the assessment of CAC may be biassed. The study utilises the Weight Loss Grading System (WLGS) to assess cachexia, which classifies patients based on BMI and weight loss over 6 months [2]. There are concerns regarding its reliance on self-reported weight loss data and the lack of standardised weight measurements, which introduces the risk of recall bias. In future studies, incorporating more standardised weight measurement techniques, as well as functional assessments like grip strength or gait speed, could reduce bias and enhance the reliability of cachexia evaluation. Additionally, surgical resection is a common treatment in LCP, removing lesions in the lungs also causes varying degrees of weight loss. While the study has excluded WLGS 2 patients to minimise non–cachexia-related weight loss, they do not provide enough details regarding the type of treatment the study participants received during the study period. To address this, it would be beneficial to specify the types of treatments the patients underwent and assess their impact on weight loss separately from cachexia.

Thirdly, the imaging approach used in the study primarily focuses on the metabolic activity of solid organs, but it overlooks the imaging features from hollow organs and key tumour-related factors that could provide additional insights into cachexia. The tumour microenvironment, including both intratumoral and peritumoral regions, plays a critical role in systemic inflammation and metabolic dysregulation [3, 4]. Moreover, the hollow organs like the gastrointestinal tract are directly involved in digestion and nutrient absorption, both of which are crucial to energy homeostasis. It is surprising that the authors did not extract or analyse these imaging characteristics in their assessment. Future research should incorporate a more comprehensive extraction of imaging features to better understand the multifaceted mechanisms of CAC.

In conclusion, we greatly appreciate the valuable insights this article provides on the association between CAC in LCPs and whole-body [¹⁸F]FDG-PET/CT imaging. We look forward to future research that further explores and addresses these limitations, inspiring further discussion and more comprehensive evaluations in CAC research.

我们怀着极大的兴趣阅读了 Ferrara 等人 [1] 最近在《恶病质、肌肉减少症和肌肉杂志》上发表的文章。基于全身 [¹⁸F]FDG-PET/CT 成像，该研究证明了肺癌患者 （LCP） 伴和不伴癌症相关恶病质 （CAC） 多器官代谢的群体差异，从而突出了恶病质患者症状的全身代谢异常，并通过机器学习模型准确识别 LCP 与 CAC。我们赞扬作者的宝贵贡献，并提出一些建议，以进一步加强对这些发现的解释，并为未来的研究提供有价值的方向。

首先，基线分析可能没有充分考虑一些与合并症和生活方式因素相关的详细协变量。恶病质是一种多因素综合征，受癌症本身以及合并症和生活方式因素的影响，包括吸烟、饮酒和饮食模式。包括详细的协变量并执行子组分析将提供有关这些变量如何与恶病质相互作用的更精细见解。考虑这些因素可能会揭示哪些亚组可能从该分析中受益最大，并进一步增强研究结果的稳健性。

其次，CAC 的评估可能存在偏倚。该研究利用体重减轻分级系统 （WLGS） 来评估恶病质，该系统根据 BMI 和 6 个月内的体重减轻对患者进行分类 [2]。人们担心它依赖于自我报告的体重减轻数据，并且缺乏标准化的体重测量，这带来了召回偏倚的风险。在未来的研究中，结合更标准化的体重测量技术，以及握力或步态速度等功能评估，可以减少偏倚并提高恶病质评估的可靠性。此外，手术切除是 LCP 的常见治疗方法，切除肺部病灶也会导致不同程度的体重减轻。虽然该研究排除了 WLGS 2 患者以尽量减少非恶病质相关的体重减轻，但他们没有提供有关研究参与者在研究期间接受的治疗类型的足够详细信息。为了解决这个问题，指定患者接受的治疗类型并独立于恶病质评估它们对体重减轻的影响将是有益的。

第三，研究中使用的成像方法主要关注实体器官的代谢活动，但它忽略了空腔器官的成像特征和关键的肿瘤相关因素，这些因素可以为恶病质提供额外的见解。肿瘤微环境，包括瘤内和瘤周区域，在全身炎症和代谢失调中起关键作用 [3， 4]。此外，胃肠道等中空器官直接参与消化和营养吸收，这两者都对能量稳态至关重要。令人惊讶的是，作者在评估中没有提取或分析这些影像学特征。未来的研究应包括更全面的成像特征提取，以更好地了解 CAC 的多方面机制。

总之，我们非常感谢本文就 LCP 中的 CAC 与全身 [¹⁸F]FDG-PET/CT 成像之间关系提供的宝贵见解。我们期待未来的研究进一步探索和解决这些局限性，激发 CAC 研究的进一步讨论和更全面的评估。