Multicomponent Interventions for Adults With Cancer Cachexia: A Systematic Review

Abstract

Background: Cancer cachexia has substantial impacts on people's quality of life. There is no current gold standard treatment, but the complex pathophysiology of cachexia suggests that a multitargeted and individualised treatment approach is needed. We aimed to evaluate the extent to which multicomponent interventions have targeted the key features of cachexia and been tailored to individuals, and differential effects on quality of life.

Methods: We conducted a systematic review of multicomponent interventions for adults with cancer cachexia. We searched four databases, two clinical trial registers and MedRxiv on 20 June 2024. Intervention components were classified by intervention category (nutritional, exercise/physical activity, pharmacological and psychosocial), cachexia feature(s) targeted (reduced energy intake, altered metabolism, involuntary weight loss and decline in physical function) and level of tailoring. Within-arm standardised mean changes in quality of life over time, as well as standardised mean differences between study arms, were calculated.

Results: Sixty-two multicomponent interventions were included, of which two combined components from all four intervention categories, and nine targeted all four key features of cachexia. Eighteen multicomponent interventions were fully tailored and 30 were partly tailored to individuals. Within-arm standardised mean changes in quality of life were calculated for thirteen studies; all had a high risk of bias or raised concerns. In eleven studies, quality of life scores improved following the intervention, whereas in two studies they declined. Standardised mean differences between study arms were calculated for four studies; in three, the intervention arm showed a greater improvement in quality of life scores than the usual care arm. Amongst these data, there was no indication that the number of cachexia features targeted, or the extent of tailoring, was associated with a greater improvement in quality of life scores; however, the heterogeneity prevented us from concluding on our hypothesis.

Conclusions: This review mapped out in detail the combinations of intervention categories used, the key features of cachexia targeted, and the extent of tailoring across multicomponent interventions for adults with cancer cachexia. Only a small proportion of the multicomponent interventions targeted all four key features of cachexia, but most were either partly or fully tailored to individuals. Despite sixty-two multicomponent interventions being investigated, only four studies compared these to usual care and reported quality of life outcomes. High risk of bias, low sample sizes and variable outcome data remain challenges to the interpretability of results in this field.

Keywords: cachexia; cancer; multimodal; quality of life; treatment; weight loss.

FIGURE 1

PRISMA flow diagram illustrating the identification, screening and inclusion of records in the review. Abbreviations: CENTRAL, Cochrane; WHO ICTRP, World Health Organisation International Clinical Trial Registry Platform.

FIGURE 2

Risk of bias assessments for randomised trials, assessed using the RoB 2 tool. D1 randomisation process, D2 deviations from intended interventions, D3 missing outcome data, D4 measurement of the outcome and D5 selection of the reported result. Abbreviations: MA, megestrol acetate; ONS, oral nutritional supplements; PA, physical activity.

FIGURE 3

Risk of bias assessments for nonrandomised studies of interventions, assessed using the ROBINS‐I tool. D1 confounding, D2 selection of participants in the study, D3 classification of interventions, D4 deviations from intended interventions, D5 missing outcome data, D6 measurement of the outcome and D7 selection of the reported result.

FIGURE 4

UpSet plot illustrating combinations of intervention categories used by the multicomponent interventions. The bars on the left show how many multicomponent interventions involved used each intervention category (e.g., 48 multicomponent interventions used at least one nutritional component). On the right, the matrix represents each of the possible combinations of intervention categories (from left to right: all four categories, different combinations of three categories, different combinations of two categories and each single category). The bars above the matrix show how many multicomponent interventions are employed in each of the possible combinations. Abbreviation: PA, physical activity.

FIGURE 5

Sunburst diagrams representing the types of components that targeted (a) energy intake, (b) metabolism, (c) weight/muscle loss and (d) physical function amongst the 50 multicomponent interventions included in the review. Abbreviations: FA, fatty acid; MA, megestrol acetate; MPA, medroxyprogesterone; NIS, nutrition impact symptoms; ONS, oral nutritional supplements; PA, physical activity.

FIGURE 6

UpSet plot illustrating combinations of key features of cachexia targeted by multicomponent interventions. The bars on the left show how many multicomponent interventions targeted each key cachexia. On the right, the matrix represents each of the possible combinations of key features of cachexia targeted (from left to right: all four features, different combinations of three features, different combinations of two features and each single feature). The bars above the matrix show how many multicomponent interventions targeted each of the possible combinations of features.

FIGURE 7

Forest plot representing standardised mean within‐arm changes in quality of life over time, for the intervention arms of 13 studies where data were available, subgrouped by (a) the number of key features of cachexia targeted by the intervention and (b) the extent to which the intervention was tailored. Larger squares indicate larger sample size. Abbreviation: SMC, standardised mean change.

FIGURE 8

Forest plots representing standardised difference in means between study arms, for the four studies where data were available, subgrouped by (a) the number of key features of cachexia targeted by the intervention and (b) the extent to which the intervention was tailored. Larger squares indicate larger sample size. Abbreviations: CI, confidence interval; SMD, standardised mean difference.