Bioelectrical impedance-derived body composition and systemic therapy outcomes in non-small cell lung cancer

Introduction

Lung cancer remains a leading cause of cancer morbidity and mortality worldwide, and non-small cell lung cancer (NSCLC) accounts for most lung cancer diagnoses (1). During the past decade, systemic treatment has changed substantially with the introduction of immune checkpoint inhibitors (ICIs) and molecularly targeted therapy, including pembrolizumab for programmed death-ligand 1 (PD-L1)-positive NSCLC and tyrosine kinase inhibitors (TKIs) for actionable oncogenic alterations (2-4). Nevertheless, survival remains heterogeneous even among patients with similar tumor stage and molecular profiles, indicating that host-related factors may contribute to treatment response and disease progression (5).

Obesity and metabolic status have therefore attracted increasing attention in lung cancer research. Body mass index (BMI) is widely used because it is simple and standardized, and international definitions classify BMI below 18.5 kg/m² as underweight, 18.5–24.9 kg/m² as normal weight, and 25 kg/m² or higher as overweight (6). Obesity is a major public health problem, with increasing weight gain and projected obesity burden reported in population-level studies (7,8). However, the relationship between BMI and lung cancer is not linear. Epidemiologic studies and meta-analyses have reported inverse associations between BMI and lung cancer risk in some settings, and clinical studies have described better outcomes among overweight or obese patients with NSCLC, including those receiving definitive treatment, first-line pembrolizumab, or epidermal growth factor receptor tyrosine kinase inhibitor (EGFR-TKI) therapy (9-16).

These observations are often described as an obesity paradox, but BMI cannot explain the underlying biology. BMI does not distinguish adipose tissue from lean tissue, does not quantify regional fat distribution, and may misclassify patients with low muscle mass, visceral obesity, or sarcopenic obesity (17,18). Mechanistically, adiposity may influence tumor progression and treatment response through chronic inflammation, altered adipokine signaling, myeloid-derived suppressor cells, T-cell dysfunction, and programmed cell death protein 1 (PD-1)-related immune modulation (19-23). Importantly, central obesity and body shape appear to carry biological information beyond total body size, and visceral obesity has been implicated in lung cancer progression and the tumor immune microenvironment (14,18,24).

Body composition assessment offers a more granular approach to host phenotyping. Low skeletal muscle mass has been associated with inferior outcomes after NSCLC surgery, whereas fat mass, fat-free mass, and regional body composition patterns have shown prognostic relevance in population and disease-specific cohorts (25-27). Sex may further modify these associations, because adiposity-related cancer risk and adipose tissue biology differ between men and women (28). These data support the need to evaluate muscle and adipose compartments separately rather than using BMI as the sole host phenotype. A recent systematic review of ICI-treated NSCLC similarly found that sarcopenia, cachexia, and pretreatment weight loss were more consistently associated with clinical outcomes than BMI alone (29).

Several tools are available for body composition assessment, including computed tomography (CT), dual-energy X-ray absorptiometry (DXA), and bioelectrical impedance analysis (BIA) (30-32). CT and DXA can provide quantitative estimates of tissue compartments but are less convenient for repeated monitoring in routine oncology practice. BIA is noninvasive, low-cost, repeatable, and feasible in outpatient or bedside settings; prior studies have supported its use for estimating muscle and fat compartments, although standardization and validation against CT remain important (30,33-35). On this basis, we conducted an exploratory retrospective study to characterize BIA-derived muscle and adiposity parameters in NSCLC and to evaluate their associations with progression-free survival (PFS) and objective response rate (ORR) in patients receiving immunotherapy or EGFR-TKI therapy. We present this article in accordance with the STROBE reporting checklist (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-0718/rc).

Methods

Study design and patients

This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of The First Affiliated Hospital of Guangzhou Medical University (No. ES-2024-110-02). The requirement for written informed consent was waived because this retrospective study used de-identified clinical data, involved no additional intervention, and did not affect patient management. This single-center, retrospective observational study enrolled consecutive adults with histologically or cytologically confirmed NSCLC who underwent baseline multifrequency BIA before initiation of antitumor treatment at The First Affiliated Hospital of Guangzhou Medical University between March 2023 and March 2025. Clinical stage was assigned according to the ninth edition of the American Joint Committee on Cancer/Union for International Cancer Control (AJCC/UICC) tumor-node-metastasis (TNM) classification.

For cross-sectional stage-based comparisons, clinical stage was prespecified as a binary variable (stage I vs. stages II–IV). Stage I was considered to represent early-stage disease, whereas stages II–IV were considered to reflect greater disease burden and greater relevance to systemic therapy. The study flow is shown in Figure 1.

Figure 1 Study flow diagram. EGFR, epidermal growth factor receptor; NSCLC, non-small cell lung cancer; TKI, tyrosine kinase inhibitor.

Treatment cohorts

Outcome analyses were conducted in two prespecified stage II–IV systemic-therapy cohorts: patients receiving first-line ICI monotherapy or ICI plus chemotherapy, and patients with EGFR-mutant NSCLC receiving first-line EGFR-TKI monotherapy or EGFR-TKI plus chemotherapy.

Body-composition assessment

Body composition was assessed using a multifrequency BIA device (BC care860), which estimates fat mass, lean body mass, skeletal muscle mass, visceral fat area, and fluid compartments. BIA was selected because it is noninvasive and feasible for repeated assessment in routine clinical practice. Body-composition measurements were obtained at baseline before treatment initiation. For the longitudinal ICI analysis, eligible follow-up BIA measurements were those obtained 84–105 days after treatment initiation, generally corresponding to assessment after four 3-week treatment cycles and allowing up to one additional cycle for delayed reassessment. Serial BIA was not protocol-mandated and was performed as part of routine clinical practice. Patients were included if paired baseline and eligible follow-up measurements were available and radiographic disease progression had not occurred before follow-up BIA. For outcome analyses, BMI was evaluated both continuously and dichotomized as <25 vs. ≥25 kg/m², as specified. Fat and muscle mass indices were calculated by normalizing the corresponding mass by height squared.

Outcomes

The primary endpoint was PFS. For patients receiving systemic therapy, PFS was defined as the time from treatment initiation to radiographic disease progression or death from any cause, whichever occurred first. The secondary endpoint was ORR, defined as the proportion of response-evaluable patients who achieved a complete or partial response according to Response Evaluation Criteria in Solid Tumors (RECIST), version 1.1. Disease progression and response were assessed by radiologists or treating clinicians.

Statistical analysis

Continuous variables were reported as means ± standard deviations (SD) or median (range), while categorical variables were expressed as frequencies and percentages. Normality was assessed using the Shapiro-Wilk test. Between-group differences in body-composition variables by clinical stage and EGFR mutation status were evaluated using analysis of covariance (ANCOVA). Prior to the ANCOVA, the model assumptions were tested for normality of residuals and homoscedasticity using the Shapiro-Wilk test and Levene’s test, respectively. Both assumptions were met.

Sex-specific differences in body composition were considered in the outcome analyses of the immunotherapy cohort. Because few women had evaluable outcome data, both the PFS and ORR analyses were restricted to the same 94 male patients. Optimal cutoffs for body-composition variables were derived from PFS in these 94 men using the surv_cutpoint function in the R package survminer and were applied unchanged to the ORR analysis; no ORR-specific cutoff optimization was performed. PFS was analyzed using Kaplan–Meier methods and Cox proportional hazards models, whereas ORR was evaluated using logistic regression. Multivariable Cox and logistic regression models in the ICI cohort were adjusted for age, smoking history, clinical stage, and histologic subtype.

Paired changes in all 49 BIA-derived body-composition parameters were assessed using paired-sample t-tests, with P values adjusted using the Benjamini-Hochberg false discovery rate (FDR) procedure. Relative change was calculated as [(follow-up value − baseline value)/baseline value] × 100%. Variables included in the longitudinal survival analysis were categorized as a >10% decrease vs. no >10% decrease (reference), with the reference group including decreases of ≤10%, no change, and increases. Associations with PFS were evaluated using univariable Cox proportional hazards models; trunk fat mass was additionally evaluated in a multivariable model adjusted for age, sex, smoking history, clinical stage, and histologic subtype.

Because the EGFR-TKI cohort was small and included both men and women, body-composition parameters were analyzed as continuous variables after standardization within sex rather than using outcome-derived cutoffs. Separate Cox proportional hazards models were fitted for each parameter and adjusted for sex, with hazard ratios (HRs) expressed per 1 sex-specific SD increase.

HRs or odds ratios (ORs), as appropriate, with 95% confidence intervals (CIs) were calculated. A P value of <0.05 was considered statistically significant. Statistical analyses were performed using R software version 4.5.0 and SPSS version 26.0.

Results

Patient characteristics

A total of 49 body-composition parameters were assessed, including indices of fat mass, muscle mass, body fat percentage, waist-to-hip ratio, and other proportional measures. Total fat and muscle mass were normalized by height squared. Overall, 369 patients were included: 126 patients (34.1%) with stage I resectable NSCLC and 243 patients (65.9%) with stage II–IV unresectable NSCLC. The median age was 55.5 years in the stage I group and 64 years in the stage II–IV group. Males accounted for 37.3% of patients with stage I disease and 71.6% of those with stage II–IV disease. Adenocarcinoma was the most common histologic subtype, accounting for 95.2% of stage I tumors and 59.3% of stage II–IV tumors. Baseline characteristics are summarized in Table 1. Driver-alteration categories were mutually exclusive for tabulation, and no co-occurring driver alterations were identified in the analyzed cohort. Among patients with stage II–IV disease, baseline characteristics were further tabulated by systemic treatment cohort to contextualize the subsequent immunotherapy and EGFR-TKI analyses (Table 2).

Body composition by clinical stage and EGFR mutation status

Body-composition parameters were compared between patients with stage I disease and those with stage II–IV disease using ANCOVA adjusted for age and sex. Compared with patients with stage I disease, patients with stage II–IV disease had lower body fat mass, body fat percentage, visceral fat area, trunk fat mass, body fat mass index, visceral fat index (VFI), and trunk fat mass index. However, none of these differences remained statistically significant after Benjamini-Hochberg FDR correction. No significant between-group differences were observed for water, protein, muscle mass, skeletal muscle mass, fat-free mass, or extracellular water-related parameters (Table S1).

Within each clinical-stage subgroup, body-composition parameters were also compared between patients with EGFR-mutant and EGFR-wild-type tumors after adjustment for age and sex. In both the stage I and stage II–IV subgroups, no body-composition parameter differed significantly according to EGFR mutation status, either before or after FDR correction (Tables S2,S3).

Correlation of body-composition parameters

Pairwise correlation analyses were performed to assess potential collinearity among BIA-derived body-composition variables. Muscle-related parameters and adiposity-related parameters showed strong within-domain correlations, whereas correlations between the muscle and adiposity domains were weak to moderate. Similar correlation patterns were observed across sex and clinical-stage strata; therefore, representative variables from each body-composition domain were selected for subsequent analyses (Figure S1).

Body composition and PFS in the immunotherapy cohort

The immunotherapy outcome cohort included 125 patients with stage II–IV NSCLC who received first-line ICI-based therapy. Because of sex-specific differences in body composition and the limited number of women with evaluable follow-up, the primary PFS analysis was restricted to men. A total of 94 men were included in the primary PFS analysis. The median follow-up duration was 9 months (range, 6–26 months), and 39 PFS events occurred (41.4%), comprising 33 radiographic progressions and 6 deaths without prior documented progression.

To evaluate the prognostic impact of body composition, variables were dichotomized using PFS-derived optimal cutoffs determined exclusively among the 94 men included in the primary ICI PFS analysis (Table S4). In univariable Cox regression, higher skeletal muscle index (SMI; HR =0.49, 95% CI: 0.26–0.94; P=0.03) and higher soft lean mass index (SLMI; HR =0.48, 95% CI: 0.24–0.96; P=0.04) were associated with longer PFS. Conversely, higher visceral fat area (VFA; HR =2.39, 95% CI: 1.26–4.52; P=0.008) and higher VFI (HR =2.09, 95% CI: 1.11–3.93; P=0.02) were associated with a higher risk of a PFS event.

Variables associated with PFS in univariable analysis (P<0.05) were screened for multicollinearity using VIFs, and variables with a VIF greater than 4 were excluded. In multivariable Cox regression adjusted for age, smoking history, clinical stage, and histologic subtype, higher SMI was associated with longer PFS (HR =0.49, 95% CI: 0.25–0.97; P=0.04), whereas higher VFI remained independently associated with shorter PFS (HR =2.11, 95% CI: 1.09–4.07; P=0.03). Age, BMI, histologic subtype, and smoking status were not significantly associated with PFS (all P>0.05; Table S5).

Kaplan-Meier analysis showed that patients with higher SMI (P=0.02) and higher SLMI (P=0.02) had longer PFS than those with lower values. Patients with lower VFA (P=0.005) and lower VFI (P=0.02) also had longer PFS than patients with higher visceral fat levels (Figure 2).

Figure 2 Kaplan-Meier curves for progression-free survival according to selected body-composition parameters among male patients with stage II–IV NSCLC receiving immunotherapy (n=94). (A) SMI; (B) SLMI; (C) VFA; (D) VFI. NSCLC, non-small cell lung cancer; PFS, progression-free survival; SLMI, soft lean mass index; SMI, skeletal muscle index; VFA, visceral fat area; VFI, visceral fat index.

Based on their biological complementarity and their independent associations with PFS in multivariable Cox models, SMI and VFI were combined to define four body-composition phenotypes. Patients were further stratified into four groups according to SMI and VFI: low SMI/high VFI, low SMI/low VFI, high SMI/high VFI, and high SMI/low VFI. The low SMI/high VFI group had the shortest PFS, whereas the high SMI/low VFI group had the longest PFS, with an apparent risk gradient across the four groups (Figure 3).

Figure 3 Kaplan-Meier curves for PFS according to combined skeletal muscle index and visceral fat index groups among male patients with stage II–IV NSCLC receiving immunotherapy (n=94). Patients were stratified into four groups according to SMI and VFI status: low SMI/high VFI, low SMI/low VFI, high SMI/high VFI, and high SMI/low VFI. NSCLC, non-small cell lung cancer; PFS, progression-free survival; SMI, skeletal muscle index; VFI, visceral fat index.

Using the low SMI/high VFI group as the reference, univariable Cox analysis showed that the low SMI/low VFI group (HR =0.25, 95% CI: 0.10–0.65; P=0.004), high SMI/high VFI group (HR =0.24, 95% CI: 0.08–0.66; P=0.006), and high SMI/low VFI group (HR =0.18, 95% CI: 0.07–0.48; P=0.001) were associated with longer PFS (Table S6). After adjustment for age, smoking history, clinical stage, and histologic subtype, the high SMI/low VFI group retained the most favorable prognosis (HR =0.13, 95% CI: 0.04–0.38; P<0.001), corresponding to an 87% lower risk of a PFS event than the reference group. High SMI also remained protective among patients with elevated visceral fat (high SMI/high VFI vs. low SMI/high VFI: HR =0.29, 95% CI: 0.08–0.99; P=0.05). Among the 39 PFS events, 33 (84.6%) were radiographic progressions and 6 (15.4%) were deaths without prior documented progression. In the low SMI/high VFI group, 11 PFS events occurred, comprising 8 radiographic progressions and 3 deaths without prior documented progression. The corresponding progression/death-without-prior-progression event counts were 10/0 in the low SMI/low VFI group, 6/1 in the high SMI/high VFI group, and 9/2 in the high SMI/low VFI group. Among the 32 patients with high VFI, 24 (75.0%) had a BMI <25 kg/m²; this proportion was 17 of 18 (94.4%) in the low SMI/high VFI group.

Body composition and ORR

ORR was evaluated in the same 94 male patients included in the primary ICI PFS analysis. Among these patients, the ORR was 43.6%. The PFS-derived body-composition cutoffs were applied unchanged to the ORR analysis.

In univariable logistic regression, higher SMI was associated with higher ORR (OR, 2.41; 95% CI: 1.03–5.65; P=0.04), whereas VFI was not significantly associated with ORR (OR, 0.46; 95% CI: 0.19–1.12; P=0.09). After adjustment for age, smoking history, clinical stage, and histologic subtype, higher SMI remained associated with higher ORR (adjusted OR, 3.04; 95% CI: 1.20–7.72; P=0.02; Table S7).

When patients were stratified by the four SMI/VFI groups, all three groups had higher ORR than the low SMI/high VFI reference group in univariable analyses (Table S8). These associations persisted after adjustment for age, smoking history, clinical stage, and histologic subtype, with adjusted ORs of 6.48 (95% CI: 1.33–31.54; P=0.02) for low SMI/low VFI, 6.80 (95% CI: 1.31–35.23; P=0.02) for high SMI/high VFI, and 15.90 (95% CI: 2.95–85.63; P=0.001) for high SMI/low VFI.

Dynamic changes in body composition during immunotherapy

Among the 125 patients in the ICI outcome cohort, 73 underwent repeat BIA, whereas 52 did not. Among the 73 patients with repeat BIA measurements, 32 were excluded because the follow-up assessment fell outside the specified reassessment window, and two were excluded because radiographic disease progression occurred before reassessment. The remaining 39 patients (31.2%) had eligible paired measurements and comprised the longitudinal cohort. All 49 BIA-derived body-composition parameters were evaluated longitudinally and are reported in Table S9. After Benjamini-Hochberg FDR correction, significant increases were observed in body fat mass, BMI, percent body fat, waist-to-hip ratio, trunk fat mass, fat mass index, and trunk fat mass index (all FDR-adjusted P≤0.021). The proportions of upper-limb and trunk muscle mass relative to total muscle mass also increased significantly (FDR-adjusted P=0.02 and 0.02, respectively). Visceral fat area and VFI showed nominal increases but did not remain significant after FDR correction (both FDR-adjusted P=0.10). No significant changes were observed in absolute lean- or muscle-mass measures, including protein content, soft lean mass, fat-free mass, skeletal muscle mass, or regional muscle mass, or in fluid-related parameters. In the exploratory analysis, a >10% decrease in trunk fat mass vs. no >10% decrease was associated with shorter PFS in univariable analysis (HR, 2.06; 95% CI: 1.15–3.70; P=0.02) and after adjustment for age, sex, smoking history, clinical stage, and histologic subtype (adjusted HR, 4.56; 95% CI: 1.39–14.97; P=0.01). No other evaluated body-composition change was significantly associated with PFS (Table S10).

Body composition in patients receiving EGFR-TKI-based therapy

A total of 43 patients with stage II–IV NSCLC receiving EGFR-TKI-based therapy were included, comprising 16 men (37.2%) and 27 women (62.8%). The median follow-up duration was 9 months (range, 6–24 months), during which 13 PFS events occurred. Given the limited sample size and mixed-sex composition of this cohort, body-composition parameters were analyzed as continuous variables after standardization within sex rather than dichotomized using outcome-derived cutoffs. No body-composition parameter was significantly associated with PFS in the sex-adjusted Cox models. For SMI, the sex-adjusted HR per 1 sex-specific SD increase was 0.96 (95% CI: 0.50–1.82; P=0.89), and for VFI, it was 1.06 (95% CI: 0.59–1.87; P=0.85) (Table S11).

Discussion

In this single-center retrospective cohort of patients with NSCLC who underwent baseline BIA before antitumor treatment, body-composition parameters showed clinically relevant associations with outcomes, particularly in patients with stage II–IV disease receiving first-line ICI-based therapy. Higher SMI was associated with longer PFS and higher ORR, whereas higher VFI was associated with shorter PFS. The combined SMI/VFI phenotype further separated patients into distinct risk groups, with the low SMI/high VFI group having the poorest outcomes and the high SMI/low VFI group having the most favorable outcomes. These findings support the potential prognostic value of host body composition in ICI-treated NSCLC.

The stage- and EGFR-based cross-sectional analyses provide context for interpreting the outcome findings. Although several adiposity-related measures were nominally lower in stage II–IV disease than in stage I disease, these differences did not remain significant after FDR correction. In addition, no body-composition parameter differed significantly by EGFR mutation status within either stage subgroup. Thus, the associations observed in the systemic-therapy cohorts were unlikely to be explained solely by broad baseline differences across clinical stage or EGFR-defined subgroups.

The association between higher SMI and improved outcomes is biologically plausible. SMI reflects height-normalized skeletal muscle reserve and may capture nutritional status, systemic inflammation, functional reserve, and treatment tolerance more effectively than BMI. Cancer cachexia is characterized by ongoing skeletal muscle loss, with or without fat loss, and is associated with functional decline and adverse outcomes (36,37). Low muscle mass has also been associated with poorer survival and treatment-related toxicity across solid tumors (38-40). In patients with NSCLC receiving ICIs, prior studies and meta-analyses have reported associations between low muscle mass or sarcopenia and inferior survival (29,41-44). More recent evidence reinforces this relationship: in a recent cohort of advanced NSCLC, CT-defined sarcopenia was associated with inferior PFS and OS among patients receiving first-line chemoimmunotherapy (45), and a prospective study using BIA found shorter PFS and OS among patients with sarcopenia or pre-sarcopenia before first-line ICI therapy (46). Our findings are consistent with these data and suggest that BIA-derived SMI may be a feasible marker of host fitness in routine practice.

Higher VFI remained associated with shorter PFS after adjustment, underscoring the importance of fat distribution rather than body size alone. BMI cannot distinguish muscle, subcutaneous fat, and visceral adipose tissue, which may explain the lack of association between BMI and PFS in this study. This distinction may be particularly relevant in Asian populations, in whom conventional BMI-based definitions can under-recognize excess visceral and ectopic adiposity at lower body weights (47). In our male ICI cohort, 24 of 32 patients (75.0%) with high VFI had a BMI <25 kg/m², including 17 of 18 patients (94.4%) with the low SMI/high VFI phenotype. Thus, elevated VFI in this cohort largely reflected disproportionate visceral adiposity that was not captured by BMI alone. Visceral adipose tissue is metabolically active and may contribute to chronic low-grade inflammation, insulin resistance, altered adipokine signaling, and immune dysregulation (48). However, the relationship between adiposity and ICI outcomes is not uniform. Higher BMI has been associated with improved outcomes in some ICI-treated NSCLC cohorts, whereas CT-based studies evaluating visceral adiposity have reported heterogeneous findings (44,49-51). Differences in patient population, treatment line, sex distribution, body-composition methods, and cutoff definitions may account for these discrepancies.

The combined SMI/VFI analysis may provide a more integrated assessment of body-composition risk than either measure alone. SMI and VFI represent distinct biological domains: skeletal muscle reserve and visceral adiposity burden. The particularly unfavorable low SMI/high VFI phenotype may therefore reflect the coexistence of reduced muscle reserve and disproportionate visceral adiposity rather than simply lower overall body mass. This pattern further supports discordance between BMI and visceral adiposity in this subgroup. Consistent with the broader concept that combined muscle depletion and excess adiposity may confer higher risk, a recent study of ICI-treated lung cancer reported poorer survival among patients with sarcopenic obesity than among those with sarcopenia or obesity alone, although its BMI-based definition differs from the SMI/VFI phenotype used here (52).

Because PFS was defined as radiographic progression or death from any cause, we also examined the composition of PFS events. Radiographic progression accounted for 33 of 39 PFS events overall and 8 of 11 events in the low SMI/high VFI group, indicating that the unfavorable PFS observed in this phenotype was not driven solely by death events. Nevertheless, because cause-specific mortality was not adjudicated, a contribution from competing non-oncologic mortality cannot be excluded. Accordingly, the association between high VFI and shorter PFS should not be attributed exclusively to tumor-immune mechanisms. Because the SMI and VFI cutoffs were derived and evaluated within the same cohort without internal resampling or external validation, the observed effect estimates may be subject to optimism bias. These data-derived thresholds should therefore be considered exploratory and require validation in independent cohorts using prespecified cutoffs.

Dynamic BIA assessment may provide information beyond baseline measurements. Although several adiposity-related measures increased on average, a subset of patients experienced marked tissue loss, and the exploratory analysis showed that a >10% decrease in trunk fat mass was associated with shorter PFS. This directional analysis was informed by longitudinal studies in ICI-treated NSCLC linking on-treatment muscle or adipose-tissue loss with adverse outcomes. A 10% relative decline in CT-derived psoas muscle area has been used to define substantial muscle loss (53), and a recent study further evaluated treatment-related changes in muscle and adipose tissue during nivolumab therapy (54). Trunk fat loss may be a marker of systemic catabolism, reduced nutritional intake, treatment-related toxicity, or disease-related deterioration rather than a causal effect of fat depletion itself. However, the 10% threshold has not been validated for BIA-derived trunk fat mass, serial BIA was not protocol-mandated, and only 39 patients had eligible paired measurements. The longitudinal analysis was therefore limited to a selected subgroup and may be subject to selection bias. These findings should therefore be considered exploratory and require prospective validation with standardized serial assessments.

In the EGFR-TKI cohort, no clear association between sex-standardized body-composition parameters and PFS was observed in sex-adjusted models. These findings should be interpreted cautiously given the small cohort (43 patients and 13 PFS events). EGFR mutation subtype and TKI generation or agent were not incorporated into the survival models and may contribute to residual confounding, while longitudinal BIA measurements were not systematically available. Moreover, the median follow-up of 9 months indicates limited outcome maturity. Larger cohorts with longer follow-up and serial body-composition assessment are needed to clarify the prognostic relevance of body composition in EGFR-TKI-treated NSCLC.

BIA offers a practical means of jointly characterizing skeletal muscle reserve and visceral adiposity (30,33,35), two complementary host domains that are not distinguished by BMI (17,18). In the male ICI cohort, SMI and VFI showed opposing associations with PFS, their combined phenotype further stratified outcomes, and serial BIA enabled exploratory assessment of on-treatment body-composition changes. These findings, together with emerging evidence on the prognostic relevance of body composition in ICI-treated NSCLC (29,44), support further evaluation of BIA as a pragmatic tool for host phenotyping and longitudinal body-composition assessment in NSCLC. The main limitations include the single-center retrospective design, limited sample size and event number, restriction of the primary ICI outcome analyses to men, potential BIA measurement variability related to hydration and measurement conditions, and residual treatment heterogeneity.

Conclusions

Among men with NSCLC receiving ICI-based therapy, BIA-derived SMI and VFI were associated with treatment outcomes, whereas BMI was not associated with PFS. Higher SMI was associated with longer PFS and higher ORR, whereas lower VFI was associated with longer PFS. In the exploratory longitudinal analysis, a >10% decrease in trunk fat mass was associated with shorter PFS. These findings provide preliminary evidence that BIA-based body-composition profiling may support host-related risk stratification and longitudinal monitoring during NSCLC systemic therapy.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-0718/rc

Data Sharing Statement: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-0718/dss

Peer Review File: Available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-0718/prf

Funding: This work was supported by the National Key Research and Development Program of China (Nos. 2024ZD0520004, 2024ZD0520000, and 2022YFC2505105), the National Natural Science Foundation of China (No. 82373121), the Science and Technology Planning Project of Guangzhou (No. SL2023A03J01303), and the Major Project of Guangzhou National Laboratory (Nos. GZNL2023A02007 and SRPG22-017).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tlcr.amegroups.com/article/view/10.21037/tlcr-2026-0718/coif). W.L. serves as an unpaid Associate Editor-in-Chief of Translational Lung Cancer Research from May 2026 to April 2027. All authors report that this study was supported by the National Key Research and Development Program of China (Nos. 2024ZD0520004, 2024ZD0520000, and 2022YFC2505105), the National Natural Science Foundation of China (No. 82373121), the Science and Technology Planning Project of Guangzhou (No. SL2023A03J01303), and the Major Project of Guangzhou National Laboratory (Nos. GZNL2023A02007 and SRPG22-017). The authors have no other conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of The First Affiliated Hospital of Guangzhou Medical University (No. ES-2024-110-02). The requirement for written informed consent was waived because this retrospective study used de-identified clinical data, involved no additional intervention, and did not affect patient management.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Bray F, Laversanne M, Sung H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2024;74:229-63. [Crossref] [PubMed]
2. Zer A, Ahn MJ, Barlesi F, et al. Early and locally advanced non-small-cell lung cancer: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up. Ann Oncol 2025;36:1245-62. [Crossref] [PubMed]
3. Reck M, Rodríguez-Abreu D, Robinson AG, et al. Pembrolizumab versus Chemotherapy for PD-L1-Positive Non-Small-Cell Lung Cancer. N Engl J Med 2016;375:1823-33. [Crossref] [PubMed]
4. Shea M, Costa DB, Rangachari D. Management of advanced non-small cell lung cancers with known mutations or rearrangements: latest evidence and treatment approaches. Ther Adv Respir Dis 2016;10:113-29. [Crossref] [PubMed]
5. Santo V, Brunetti L, Pecci F, et al. Host-related Determinants of Response to Immunotherapy in Non-small Cell Lung Cancer: The Interplay of Body Composition, Metabolism, Sex and Immune Regulation. Curr Oncol Rep 2025;27:1427-47. [Crossref] [PubMed]
6. World Health Organization. Body mass index (BMI). [cited 2026 May 30]. Available online: https://www.who.int/data/gho/data/themes/topics/topic-details/GHO/body-mass-index.
7. Alvarnas A, Alvarnas J. Obesity and cancer risk: a public health crisis. Am J Manag Care 2019;25:SP332-3. [PubMed]
8. Samuel M, Park RY, Eastwood SV, et al. Trends in weight gain recorded in English primary care before and during the Coronavirus-19 pandemic: An observational cohort study using the OpenSAFELY platform. PLoS Med 2024;21:e1004398. [Crossref] [PubMed]
9. Smith L, Brinton LA, Spitz MR, et al. Body mass index and risk of lung cancer among never, former, and current smokers. J Natl Cancer Inst 2012;104:778-89. [Crossref] [PubMed]
10. Rauscher GH, Mayne ST, Janerich DT. Relation between body mass index and lung cancer risk in men and women never and former smokers. Am J Epidemiol 2000;152:506-13. [Crossref] [PubMed]
11. Duan P, Hu C, Quan C, et al. Body mass index and risk of lung cancer: Systematic review and dose-response meta-analysis. Sci Rep 2015;5:16938. [Crossref] [PubMed]
12. Leung CC, Lam TH, Yew WW, et al. Lower lung cancer mortality in obesity. Int J Epidemiol 2011;40:174-82. [Crossref] [PubMed]
13. Lam VK, Bentzen SM, Mohindra P, et al. Obesity is associated with long-term improved survival in definitively treated locally advanced non-small cell lung cancer (NSCLC). Lung Cancer 2017;104:52-7. [Crossref] [PubMed]
14. Yu D, Zheng W, Johansson M, et al. Overall and Central Obesity and Risk of Lung Cancer: A Pooled Analysis. J Natl Cancer Inst 2018;110:831-42. [Crossref] [PubMed]
15. Cortellini A, Ricciuti B, Tiseo M, et al. Baseline BMI and BMI variation during first line pembrolizumab in NSCLC patients with a PD-L1 expression ≥ 50%: a multicenter study with external validation. J Immunother Cancer 2020;8:e001403. [Crossref] [PubMed]
16. Ding ZY, Chen YY, Hong Y, et al. Obesity has an impact on the efficacy of EGFR-TKI in NSCLC patients harbouring EGFR mutation: A real-world study. Ann Oncol 2019;30:ii49-50.
17. Khaddour K, Gomez-Perez SL, Jain N, et al. Obesity, Sarcopenia, and Outcomes in Non-Small Cell Lung Cancer Patients Treated With Immune Checkpoint Inhibitors and Tyrosine Kinase Inhibitors. Front Oncol 2020;10:576314. [Crossref] [PubMed]
18. Ardesch FH, Ruiter R, Mulder M, et al. The Obesity Paradox in Lung Cancer: Associations With Body Size Versus Body Shape. Front Oncol 2020;10:591110. [Crossref] [PubMed]
19. Park J, Morley TS, Kim M, et al. Obesity and cancer--mechanisms underlying tumour progression and recurrence. Nat Rev Endocrinol 2014;10:455-65. [Crossref] [PubMed]
20. Canter RJ, Le CT, Beerthuijzen JMT, et al. Obesity as an immune-modifying factor in cancer immunotherapy. J Leukoc Biol 2018;104:487-97. [Crossref] [PubMed]
21. Hotamisligil GS. Inflammation and metabolic disorders. Nature 2006;444:860-7. [Crossref] [PubMed]
22. Ostrand-Rosenberg S. Myeloid derived-suppressor cells: their role in cancer and obesity. Curr Opin Immunol 2018;51:68-75. [Crossref] [PubMed]
23. Wang Z, Aguilar EG, Luna JI, et al. Paradoxical effects of obesity on T cell function during tumor progression and PD-1 checkpoint blockade. Nat Med 2019;25:141-51. [Crossref] [PubMed]
24. Barbi J, Patnaik SK, Pabla S, et al. Visceral Obesity Promotes Lung Cancer Progression-Toward Resolution of the Obesity Paradox in Lung Cancer. J Thorac Oncol 2021;16:1333-48. [Crossref] [PubMed]
25. Lading T, Media AS, Tellefsen B, et al. Sarcopenia and survival in surgically treated non-small cell lung cancer: A systematic review and meta-analysis. Eur J Surg Oncol 2025;51:110538. [Crossref] [PubMed]
26. Sedlmeier AM, Baumeister SE, Weber A, et al. Relation of body fat mass and fat-free mass to total mortality: results from 7 prospective cohort studies. Am J Clin Nutr 2021;113:639-46. [Crossref] [PubMed]
27. Xu S, Wen S, Yang Y, et al. Association Between Body Composition Patterns, Cardiovascular Disease, and Risk of Neurodegenerative Disease in the UK Biobank. Neurology 2024;103:e209659. [Crossref] [PubMed]
28. Rask-Andersen M, Ivansson E, Höglund J, et al. Adiposity and sex-specific cancer risk. Cancer Cell 2023;41:1186-1197.e4. [Crossref] [PubMed]
29. Golban C, Susa SR, Varga NI, et al. The Impact of Body Composition on Outcomes in NSCLC Patients Treated with Immune Checkpoint Inhibitors: A Systematic Review. Cancers (Basel) 2025;17:2765. [Crossref] [PubMed]
30. Branco MG, Mateus C, Capelas ML, et al. Bioelectrical Impedance Analysis (BIA) for the Assessment of Body Composition in Oncology: A Scoping Review. Nutrients 2023;15:4792. [Crossref] [PubMed]
31. Albanese CV, Diessel E, Genant HK. Clinical applications of body composition measurements using DXA. J Clin Densitom 2003;6:75-85. [Crossref] [PubMed]
32. Zhang L, Li J, Yang Z, et al. The development of an attention mechanism enhanced deep learning model and its application for body composition assessment with L3 CT images. Sci Rep 2024;14:28953. [Crossref] [PubMed]
33. Lee JK, Park YS, Kim K, et al. Comparison of Bioelectrical Impedance Analysis and Computed Tomography on Body Composition Changes Including Visceral Fat After Bariatric Surgery in Asian Patients with Obesity. Obes Surg 2021;31:4243-50. [Crossref] [PubMed]
34. Looijaard WGPM, Stapel SN, Dekker IM, et al. Identifying critically ill patients with low muscle mass: Agreement between bioelectrical impedance analysis and computed tomography. Clin Nutr 2020;39:1809-17. [Crossref] [PubMed]
35. Aleixo GFP, Shachar SS, Nyrop KA, et al. Bioelectrical Impedance Analysis for the Assessment of Sarcopenia in Patients with Cancer: A Systematic Review. Oncologist 2020;25:170-82. [Crossref] [PubMed]
36. Fearon K, Strasser F, Anker SD, et al. Definition and classification of cancer cachexia: an international consensus. Lancet Oncol 2011;12:489-95. [Crossref] [PubMed]
37. Argilés JM, López-Soriano FJ, Stemmler B, et al. Cancer-associated cachexia - understanding the tumour macroenvironment and microenvironment to improve management. Nat Rev Clin Oncol 2023;20:250-64. [Crossref] [PubMed]
38. Rier HN, Jager A, Sleijfer S, et al. The Prevalence and Prognostic Value of Low Muscle Mass in Cancer Patients: A Review of the Literature. Oncologist 2016;21:1396-409. [Crossref] [PubMed]
39. Martin L, Birdsell L, Macdonald N, et al. Cancer cachexia in the age of obesity: skeletal muscle depletion is a powerful prognostic factor, independent of body mass index. J Clin Oncol 2013;31:1539-47. [Crossref] [PubMed]
40. Prado CM, Lieffers JR, McCargar LJ, et al. Prevalence and clinical implications of sarcopenic obesity in patients with solid tumours of the respiratory and gastrointestinal tracts: a population-based study. Lancet Oncol 2008;9:629-35. [Crossref] [PubMed]
41. Takada K, Yoneshima Y, Tanaka K, et al. Clinical impact of skeletal muscle area in patients with non-small cell lung cancer treated with anti-PD-1 inhibitors. J Cancer Res Clin Oncol 2020;146:1217-25. [Crossref] [PubMed]
42. Ren B, Shen J, Qian Y, et al. Sarcopenia as a Determinant of the Efficacy of Immune Checkpoint Inhibitors in Non-Small Cell Lung Cancer: A Meta-Analysis. Nutr Cancer 2023;75:685-95. [Crossref] [PubMed]
43. Takenaka Y, Oya R, Takemoto N, et al. Predictive impact of sarcopenia in solid cancers treated with immune checkpoint inhibitors: a meta-analysis. J Cachexia Sarcopenia Muscle 2021;12:1122-35. [Crossref] [PubMed]
44. Chaunzwa TL, Qian JM, Li Q, et al. Body Composition in Advanced Non-Small Cell Lung Cancer Treated With Immunotherapy. JAMA Oncol 2024;10:773-83. [Crossref] [PubMed]
45. Lee H, Kim CG, Han S, et al. Sarcopenia Predicts Outcome After Chemoimmunotherapy, Not Chemotherapy, in Advanced Lung Cancer: Single-Centre Retrospective Study. J Cachexia Sarcopenia Muscle 2026;17:e70243. [Crossref] [PubMed]
46. Masuda T, Watanabe S, Suzuki R, et al. Muscle loss and pentraxin-3 as predictors of immunotherapy outcomes in non-small cell lung cancer: a prospective study. Sci Rep 2026;16:20915. [Crossref] [PubMed]
47. Lim S, Ji L, Tham KW, et al. Clinical obesity in Asian people: bridging the gap between adiposity and disease. Nat Rev Endocrinol 2026;22:615-24. [Crossref] [PubMed]
48. Ouchi N, Parker JL, Lugus JJ, et al. Adipokines in inflammation and metabolic disease. Nat Rev Immunol 2011;11:85-97. [Crossref] [PubMed]
49. Kichenadasse G, Miners JO, Mangoni AA, et al. Association Between Body Mass Index and Overall Survival With Immune Checkpoint Inhibitor Therapy for Advanced Non-Small Cell Lung Cancer. JAMA Oncol 2020;6:512-8. [Crossref] [PubMed]
50. Park JE, Jo J, Youk J, et al. Prognostic utility of body composition parameters based on computed tomography analysis of advanced non-small cell lung cancer treated with immune checkpoint inhibitors. Insights Imaging 2023;14:182. [Crossref] [PubMed]
51. Minami S, Ihara S, Tanaka T, et al. Sarcopenia and Visceral Adiposity Did Not Affect Efficacy of Immune-Checkpoint Inhibitor Monotherapy for Pretreated Patients With Advanced Non-Small Cell Lung Cancer. World J Oncol 2020;11:9-22. [Crossref] [PubMed]
52. Yang T, Wang W, Qiao Y, et al. Association and predictive value of sarcopenic obesity for the prognosis of lung cancer patients receiving immune checkpoint inhibitors. Front Nutr 2026;13:1687912. [Crossref] [PubMed]
53. Nishioka N, Uchino J, Hirai S, et al. Association of Sarcopenia with and Efficacy of Anti-PD-1/PD-L1 Therapy in Non-Small-Cell Lung Cancer. J Clin Med 2019;8:450. [Crossref] [PubMed]
54. Kocaaslan E, Güren AK, Akagündüz F, et al. Prognostic Value of Treatment-Related Body Composition Changes in Metastatic NSCLC Receiving Nivolumab. Medicina (Kaunas) 2026;62:98. [Crossref] [PubMed]