This study demonstrated a significant association between the ORR to ICIs and PMI. Notably, patients with a higher PMI had a higher incidence of irAEs, which are thought to be caused by ICI-induced immune cells. These findings indicate that PMI can serve not only as a prognostic indicator in patients with cancer, but also as a predictor of their response to ICI therapy.
PMI was correlated with the ORR to ICIs, with responders having a significantly higher mean PMI than non-responders. Grouping patients based on the cutoff value of the median PMI showed higher ORR and DCR in the high- PMI group than in the low- PMI group. Furthermore, the mean PFS and OS in the high PMI group were double those in the low- PMI group, although these differences were not statistically significant. The lack of statistical significance is attributable to the small sample size.
Recently, several parameters have been evaluated as biomarkers for predicting the response to ICI therapy. Of these, skeletal muscle mass is correlated with survival in patients with advanced NSCLC treated with nivolumab or a combination of ICIs and cytotoxic chemotherapy [5, 6]. Additionally, PMI is correlated with the response to PD-1 inhibitors in patients with NSCLC [7]. Similar associations have also been reported in patients receiving ICIs for gastric cancer and renal cell carcinoma [8–11]. Two independent meta-analyses have shown that sarcopenia has a negative effect on the overall response and survival in patients treated with ICIs for various types of cancer, consistent with the results of this study.
This study revealed a strong association between the PMI and incidence of irAEs. This contrasts with previous reports on conventional cytotoxic chemotherapy, in which patients with sarcopenia tend to have more adverse events and higher discontinuation rates [1, 17–20]. The difference in the effect of PMI on response to ICIs and cytotoxic chemotherapy can be attributed to fundamental differences in the mechanisms of these adverse events. Biologically, irAEs are inflammatory toxicities induced by overactivated T cells reinvigorated by ICIs [21]. In a mouse model of ICI-induced myocarditis, clonally expanded cytotoxic CD8+ T cells infiltrated the myocardium and were necessary for the development of myocarditis [22]. Similar findings have been reported for ICI-induced colitis. Single-cell RNA sequencing of colon biopsies from patients with ICI-induced colitis revealed an expansion of cytotoxic CD8+ T cells [23].
These findings suggest that the antitumor effects of ICIs and irAEs have a common underlying mechanism. In other words, the incidence of irAEs is a sign of sufficient activation of the host immune system by ICIs, and patients who experience irAEs can be deemed to be “responders” to ICIs. The presence of irAEs was strongly associated with both the clinical response of ICIs and survival in the present study. Several studies have had similar findings in NSCLC and in other types of cancer [24–27].
The associations between PMI and both clinical response and irAEs suggest that PMI is not only a prognostic indicator but can also be used as surrogate marker for patient immunocompetence, which predicts the clinical response to ICIs. Skeletal muscle cells secrete various cytokines called myokines, which communicate with other organs, such as the liver, pancreas, cardiovascular system, brain, and bones [28]. Recently, skeletal muscle has been shown to regulate both innate and adaptive immune responses via surface molecules and myokines, and loss of skeletal muscle mass and sarcopenia may cause immune senescence [3, 4]. Preclinical research has shown that skeletal muscle also interacts with the gut microbiota, which affect the tumor microenvironment via their metabolites [29–31]. A clinical study of patients with lung cancer has shown that sarcopenia is associated with changes in the gut microbiota and their metabolites [32].
These connections between skeletal muscle and the immune system may explain the greater potential utility of PMI for predicting the effectiveness of ICIs than for predicting response to conventional cytotoxic chemotherapy. However, a recent meta-analysis which found that sarcopenia had a negative effect on the clinical response to ICIs was inconclusive regarding the relationship between sarcopenia and irAEs [33, 34]. Considering that this lack of clarity is possibly due to the limited number of studies on irAEs, additional large studies are necessary to reach a conclusion.
The question of whether interventions to improve the sarcopenic status of patients with cancer can improve ICI response and overcome resistance to ICIs remains unanswered. Several interventions such as nutritional support, physical exercise, and anamorelin have been investigated, but none of these interventions have demonstrated clear clinical benefits on the survival of patients with cancer [1, 35]. An ongoing trial including patients with melanoma receiving ICI therapy, with the intervention group undergoing diet modulation and regular exercise, is assessing the effect of exercise on response to ICI therapy (NCT04866810). As this study was retrospective, we cannot infer causality. Therefore, prospective clinical trials with well-defined interventions to increase patients’ PMI are required to provide conclusive evidence on the relationship between PMI and response to ICI therapy.