There are typographical errors within the article and these have been corrected below:
-
In Table 1, the VATI value for weight stable patients is incorrect and is currently written as a repeat of the value above [55.52 (48.40–64.22)]:

The correct value for weight stable VATI is, in fact, 66.12 (24.02–87.17). The corresponding P value remains correct.
-
In para 1 of p. 1145, we say: ‘Comparing VAT between the groups, we found 2,200 probesets regulated (10‐fold up to 2.3‐fold down) between CC and CWS, and 1253 probesets regulated (nine‐fold up to 2.3‐fold down) between CC and control’.
The sentence should read: ‘Comparing VAT between the groups, we found 2200 probesets regulated (FDR < 0.05; 10‐fold up to 2.3‐fold down) between CC and control, and 1253 probesets regulated (FDR < 0.05; 9‐fold up to 2.3‐fold down) between CWS and control’. The corresponding data in the Abstract and Table 2 remain correct.
- In Figure 6, the patient groups are labelled incorrectly. Specifically, the ‘CC’ (cancer cachexia) group has been labelled ‘CWL’ (meaning cancer weight‐losing, our previous term for this group of patients). ‘CWL’ has not been used as a term in the rest of the manuscript, and it would be more consistent and helpful to the reader to continue to label the group ‘CC’. There is also a missing P value. We include a new version of Figure 6 below with the correct labels. The corresponding text in the manuscript remains correct.

The methodological technique of principal components analysis has been misspelt throughout the manuscript as principle components analysis. Specifically, this occurs on three occasions: the 1st paragraph of p. 1144; the last paragraph of p. 1144; and the legend to Figure 1 on p. 1144.
