Abstract
Background
The combined prognostic role of skeletal muscle index (SMI), subcutaneous fat index (SFI), and visceral fat index (VFI) in non-metastatic non-small cell lung cancer (NSCLC) remains unclear.
Methods
Consecutive non-metastatic NSCLC patients who underwent radical pulmonary resection at Yunnan Cancer Hospital from January 2013 to December 2018 were analyzed. Preoperative CT-derived SMI, SFI, and VFI at the third lumbar vertebra level (L3) were stratified into sex-specific high and low groups. A composite index was created based on the count of low-value measures among the three indices. Cox regression evaluated associations with overall survival (OS) and relapse-free survival (RFS).
Results
A total of 1661 patients (mean age 58.9 ± 9.5 years; 911 men [54.8%] and 750 women [45.2%]) were enrolled, with a median follow-up of 73.97 months (95% CI: 72.80–75.10). Low SMI (HR = 1.49, 95% CI: 1.16–1.92, p = 0.002), low SFI (HR = 1.54, 95% CI: 1.22–1.94, p < 0.001), and low VFI (HR = 1.69, 95% CI: 1.32–2.17, p < 0.001) were associated with poorer OS. The composite index demonstrated poorer OS with an increasing number of low indices (SMI, SFI, and VFI) (p for trend < 0.001): one-low (HR = 1.50, 95% CI: 1.16–1.94, p = 0.002), two-low (HR = 1.71, 95% CI: 1.29–2.28, p < 0.001), and all-low (HR = 2.99, 95% CI: 1.89–4.71, p < 0.001). Similar trend occurred for RFS (all p < 0.05).
Conclusion
Preoperative SMI, SFI, and VFI were independently associated with prognosis in patients with NSCLC, and a composite index integrating these measures may provide valuable complementary information for risk stratification.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12890-026-04235-w.
Keywords: Non-small cell lung cancer, Body composition, Computed tomography, Skeletal Muscle, Prognosis
Introduction
Among patients with non-small cell lung cancer (NSCLC) undergoing radical resection, approximately 30% to 55% experience disease recurrence or death [1]. Skeletal muscle index (SMI), subcutaneous fat index (SFI), and visceral fat index (VFI) are established body composition metrics with recognized prognostic relevance in cancer populations [2–4]. These indices have also been associated with differential immunotherapy responses and varied clinical outcomes [5, 6]. However, current prognostic studies examining these factors exhibit considerable heterogeneity, and their combined prognostic utility remains insufficiently explored.
Body composition analysis offers a more precise assessment of muscle and adipose tissue distribution than body mass index (BMI) [7–9]. Recent evidence emphasizes the impact of changes in muscle and fat compartments on cancer progression [10, 11]. Computed tomography (CT) has emerged as a reliable imaging modality for quantitative assessment of body composition [12–14]. While many studies have employed chest CT scans to evaluate body composition in relation to NSCLC outcomes [15–20], significant discrepancies exist between measurements obtained at the chest level and those derived from the third lumbar vertebra (L3) level [21].
Notably, L3-level measurements are considered more standardized and representative of whole-body composition, yet their prognostic significance in NSCLC has not been fully established [22]. Furthermore, recent findings indicate functional crosstalk between skeletal muscle and adipose tissue [23], underscoring the importance of evaluating their combined prognostic impact. Therefore, a comprehensive investigation of L3-based SMI, SFI, and VFI—and their integration into a composite index—may enhance risk stratification in patients with resectable NSCLC.
This study aims to assess the prognostic value of preoperative SMI, SFI, and VFI in patients with NSCLC undergoing radical resection, with a focus on their combined impact on survival outcomes.
Methods
Participants
A total of 3125 consecutive patients with NSCLC who underwent curative-intent surgical resection at Yunnan Cancer Hospital between January 2013 and December 2018 were retrospectively screened. Exclusion criteria included: (1) errors in clinical documentation (n = 17); (2) preoperative L3 CT images were unavailable due to hospital changing its imaging storage system, or because patients underwent emergency surgery or were referred from external hospitals and did not undergo preoperative CT scans at our institution (n = 1294); (3) unknown pathological stage (n = 121); (4) suboptimal CT image quality (n = 28); and (5) outlier or unmeasurable subcutaneous fat area (n = 4). After applying these criteria, a total of 1661 patients with complete clinical and imaging data were included in the final analysis (Figure 1).
Fig. 1.
Patient selection flowchart
Quantitative body composition assessment and normalization
A CT-based body composition analysis protocol was implemented, utilizing SliceOmatic software (version 6.0; Tomovision, Montreal, QC, Canada) for quantitative assessment of cross-sectional areas (cm²) of skeletal muscle and adipose tissue at the L3 level on non-contrast CT scans. The Hounsfield unit (HU) thresholds for tissue segmentation were defined as follows: (1) Skeletal muscle area (SMA): -29 to 150 HU, including the psoas major, erector spinae, quadratus lumborum, transversus abdominis, external and internal obliques, and rectus abdominis muscles; (2) Subcutaneous adipose tissue (SAT) : -190 to -30 HU, comprising superficial adipose deposits between the muscular layers and skin in both abdominal and dorsal regions; (3) Visceral adipose tissue (VAT) : -150 to -50 HU, encompassing intra-abdominal fat deposits delineated by the parietal peritoneum or transversalis fascia. The anatomical boundaries of muscle and adipose tissues were precisely delineated using standardized semi-automated segmentation techniques by board-certified radiologists specializing in abdominal imaging, who were blinded to patients’ clinical characteristics and outcomes. SMI, SFI, and VFI (unit: cm²/m²) were calculated by dividing the cross-sectional areas of SMA (cm²), SAT (cm²), and VAT (cm²) by height squared (m²) [2, 24].
Classification criteria for SMI, SFI, VFI and their composite index
In accordance with established methodologies [25, 26], optimal cutoff values for SMI, SFI, and VFI in relation to OS were determined using the surv_cutpoint function from the “survminer” package in R software, with gender-specific thresholds applied, enabling binary stratification of participants into high versus low value cohorts. The Composite index was generated by summing the number of indices (SMI, SFI, VFI) below their gender-specific cutoffs.
Outcome definition
The primary endpoint was defined as overall survival (OS), measured from the date of surgery to either death from any cause or the last follow-up. Relapse-free survival (RFS), defined as the time from surgery to locoregional recurrence, distant metastasis, or death from any cause, whichever occurs first.
Statistical analysis
Baseline characteristics were summarized as numbers (percentages). Categorical variables were compared between groups using χ² tests or Fisher’s exact tests. Kaplan-Meier curves were constructed, and between-group differences were assessed with log-rank tests. Multivariable Cox regression models evaluated the associations between body composition and clinical outcomes, reporting adjusted hazard ratio (HR) with 95% confidence intervals (CI). A sensitivity analysis was conducted following the exclusion of patients with follow-up durations shorter than 6 months. Subgroup analyses were performed stratified by gender, age, pathological stage, tumor location, histological type, adjuvant chemotherapy status, and surgical approach. The subgroup analysis results were visually presented using forest plots generated by the R package “forestploter”. A statistical significance threshold of α = 0.05 was applied for all analyses. All data cleaning and analytical procedures were performed using R statistical software (version 4.4.1).
Results
Patient characteristics
Among initially enrolled 3125 patients, 1661 ultimately met the predefined analytical criteria (Figure 1). Comparative baseline data between included and excluded populations are presented in Supplementary Table 1. The study population included 911 men (54.8%) and 750 women (45.2%) (Table 1), with an overall mean age of 58.9 ± 9.5 years (Supplementary Table 1). There were 432 deaths observed during follow-up. Median OS and RFS durations were 73.97 (interquartile range [IQR]: 46.13–90.77) and 70.50 (IQR: 28.63–87.73) months, respectively. Male patients had median SMI, SFI, and VFI measurements of 48.57, 29.74, and 32.26 cm²/m², respectively. The adopted cutoff values for these indices were 41.41, 14.14, and 9.18 cm²/m², correspondingly. Among female patients, the median values for SMI, SFI, and VFI were 39.00, 56.61, and 29.71 cm²/m², respectively. The corresponding cutoff values were established at 33.79, 52.32, and 16.98 cm²/m², respectively.
Table 1.
Baseline characteristics of body composition profiles in the study cohort
| Variablea | Totalb | Skeletal muscle indexb | p c | Subcutaneous fat indexb | p c | Visceral fat indexb | p c | |||
|---|---|---|---|---|---|---|---|---|---|---|
| LowHigh | LowHigh | LowHigh | ||||||||
| Number of patients | 1661 | 277 (16.7) | 1384 (83.3) | 485 (29.2) | 1176 (70.8) | 323 (19.4) | 1338 (80.6) | |||
| Age, years | < 0.001 | 0.752 | < 0.001 | |||||||
| < 60 | 865 (52.1) | 86 (31.0) | 779 (56.3) | 256 (52.8) | 609 (51.8) | 201 (62.2) | 664 (49.6) | |||
| ≥ 60 | 796 (47.9) | 191 (69.0) | 605 (43.7) | 229 (47.2) | 567 (48.2) | 122 (37.8) | 674 (50.4) | |||
| Sex | 0.326 | < 0.001 | < 0.001 | |||||||
| Male | 911 (54.8) | 144 (52.0) | 767 (55.4) | 162 (33.4) | 749 (63.7) | 137 (42.4) | 774 (57.8) | |||
| Female | 750 (45.2) | 133 (48.0) | 617 (44.6) | 323 (66.6) | 427 (36.3) | 186 (57.6) | 564 (42.2) | |||
| BMI, kg/m2 | < 0.001 | < 0.001 | < 0.001 | |||||||
| < 18.5 | 86 ( 5.2) | 47 (17.0) | 39 ( 2.8) | 70 (14.4) | 16 ( 1.4) | 63 (19.5) | 23 ( 1.7) | |||
| 18.5–24.9 | 1264 (76.1) | 223 (80.5) | 1041 (75.2) | 407 (83.9) | 857 (72.9) | 256 (79.3) | 1008 (75.3) | |||
| ≥ 25 | 311 (18.7) | 7 ( 2.5) | 304 (22.0) | 8 ( 1.6) | 303 (25.8) | 4 ( 1.2) | 307 (22.9) | |||
| Smoking | 0.729 | < 0.001 | 0.022 | |||||||
| Yes | 672 (40.5) | 109 (39.4) | 563 (40.7) | 126 (26.0) | 546 (46.4) | 111 (34.4) | 561 (41.9) | |||
| No | 970 (58.4) | 164 (59.2) | 806 (58.2) | 357 (73.6) | 613 (52.1) | 210 (65.0) | 760 (56.8) | |||
| Unknown | 19 ( 1.1) | 4 ( 1.4) | 15 ( 1.1) | 2 ( 0.4) | 17 ( 1.4) | 2 ( 0.6) | 17 ( 1.3) | |||
| Hypertension | 0.008 | < 0.001 | < 0.001 | |||||||
| Yes | 226 (13.6) | 36 (13.0) | 190 (13.7) | 40 ( 8.2) | 186 (15.8) | 15 ( 4.6) | 211 (15.8) | |||
| No | 1178 (70.9) | 181 (65.3) | 997 (72.0) | 362 (74.6) | 816 (69.4) | 260 (80.5) | 918 (68.6) | |||
| Unknown | 257 (15.5) | 60 (21.7) | 197 (14.2) | 83 (17.1) | 174 (14.8) | 48 (14.9) | 209 (15.6) | |||
| Diabetes mellitus | 0.012 | 0.021 | 0.001 | |||||||
| Yes | 81 ( 4.9) | 14 ( 5.1) | 67 ( 4.8) | 13 ( 2.7) | 68 ( 5.8) | 3 ( 0.9) | 78 ( 5.8) | |||
| No | 1323 (79.7) | 204 (73.6) | 1119 (80.9) | 390 (80.4) | 933 (79.3) | 272 (84.2) | 1051 (78.6) | |||
| Unknown | 257 (15.5) | 59 (21.3) | 198 (14.3) | 82 (16.9) | 175 (14.9) | 48 (14.9) | 209 (15.6) | |||
| COPD | 0.012 | 0.518 | 0.897 | |||||||
| Yes | 57 ( 3.4) | 7 ( 2.5) | 50 ( 3.6) | 16 ( 3.3) | 41 ( 3.5) | 12 ( 3.7) | 45 ( 3.4) | |||
| No | 1346 (81.0) | 211 (76.2) | 1135 (82.0) | 386 (79.6) | 960 (81.6) | 263 (81.4) | 1083 (80.9) | |||
| Unknown | 258 (15.5) | 59 (21.3) | 199 (14.4) | 83 (17.1) | 175 (14.9) | 48 (14.9) | 210 (15.7) | |||
| Histologic type | 0.626 | 0.003 | 0.372 | |||||||
| Adenocarcinoma | 1283 (77.2) | 210 (75.8) | 1073 (77.5) | 400 (82.5) | 883 (75.1) | 259 (80.2) | 1024 (76.5) | |||
| Squamous cell carcinoma | 300 (18.1) | 51 (18.4) | 249 (18.0) | 64 (13.2) | 236 (20.1) | 51 (15.8) | 249 (18.6) | |||
| Other | 78 ( 4.7) | 16 ( 5.8) | 62 ( 4.5) | 21 ( 4.3) | 57 ( 4.8) | 13 ( 4.0) | 65 ( 4.9) | |||
| Pathological stage | 0.044 | 0.495 | 0.085 | |||||||
| Ⅰ | 959 (57.7) | 145 (52.3) | 814 (58.8) | 287 (59.2) | 672 (57.1) | 173 (53.6) | 786 (58.7) | |||
| Ⅱ | 249 (15.0) | 54 (19.5) | 195 (14.1) | 65 (13.4) | 184 (15.6) | 46 (14.2) | 203 (15.2) | |||
| Ⅲ | 453 (27.3) | 78 (28.2) | 375 (27.1) | 133 (27.4) | 320 (27.2) | 104 (32.2) | 349 (26.1) | |||
| N stage | 0.893 | 0.424 | 0.669 | |||||||
| < N2 | 1310 (78.9) | 221 (79.8) | 1089 (78.7) | 386 (79.6) | 924 (78.6) | 258 (79.9) | 1052 (78.6) | |||
| ≥N2 | 297 (17.9) | 48 (17.3) | 249 (18.0) | 80 (16.5) | 217 (18.5) | 57 (17.6) | 240 (17.9) | |||
| Unknown | 54 ( 3.3) | 8 ( 2.9) | 46 ( 3.3) | 19 ( 3.9) | 35 ( 3.0) | 8 ( 2.5) | 46 ( 3.4) | |||
| Tumor location | 0.637 | 0.606 | 0.468 | |||||||
| Upper lobe | 840 (50.6) | 136 (49.1) | 704 (50.9) | 240 (49.5) | 600 (51.0) | 157 (48.6) | 683 (51.0) | |||
| Non–upper lobe | 821 (49.4) | 141 (50.9) | 680 (49.1) | 245 (50.5) | 576 (49.0) | 166 (51.4) | 655 (49.0) | |||
| Preoperative CEA, ng/mL | 0.397 | 0.031 | 0.008 | |||||||
| < 5 | 1112 (66.9) | 179 (64.6) | 933 (67.4) | 306 (63.1) | 806 (68.5) | 194 (60.1) | 918 (68.6) | |||
| ≥ 5 | 440 (26.5) | 75 (27.1) | 365 (26.4) | 150 (30.9) | 290 (24.7) | 107 (33.1) | 333 (24.9) | |||
| Unknown | 109 ( 6.6) | 23 ( 8.3) | 86 ( 6.2) | 29 ( 6.0) | 80 ( 6.8) | 22 ( 6.8) | 87 ( 6.5) | |||
| Chemotherapy | 0.151 | 0.030 | 0.529 | |||||||
| Yes | 800 (48.2) | 122 (44.0) | 678 (49.0) | 213 (43.9) | 587 (49.9) | 150 (46.4) | 650 (48.6) | |||
| No | 861 (51.8) | 155 (56.0) | 706 (51.0) | 272 (56.1) | 589 (50.1) | 173 (53.6) | 688 (51.4) | |||
| Radiotherapy | 0.908 | 0.279 | 0.865 | |||||||
| Yes | 107 ( 6.4) | 16 ( 5.8) | 91 ( 6.6) | 24 ( 4.9) | 83 ( 7.1) | 18 ( 5.6) | 89 ( 6.7) | |||
| No | 1542 (92.8) | 259 (93.5) | 1283 (92.7) | 458 (94.4) | 1084 (92.2) | 303 (93.8) | 1239 (92.6) | |||
| Unknown | 12 ( 0.7) | 2 ( 0.7) | 10 ( 0.7) | 3 ( 0.6) | 9 ( 0.8) | 2 ( 0.6) | 10 ( 0.7) | |||
aBMI Body mass index, CEA Carcinoembryonic Antigen, COPD Chronic obstructive pulmonary disease
bData are N(%)
cp, using chi-square test or exact Fisher test
As demonstrated in Table 1, among the baseline characteristics, 277 patients (16.7%) exhibited low SMI, which was significantly associated with advanced age, lower BMI, pathological stage, hypertension, diabetes mellitus, and chronic obstructive pulmonary disease (COPD). Low SFI was identified in 485 patients (29.2%) and was significantly associated with female sex, lower BMI, non-smoking history, adenocarcinoma histology, elevated preoperative carcinoembryonic antigen (CEA) levels, absence of chemotherapy, hypertension, and diabetes mellitus. Low VFI was identified in 323 patients (19.4%) and was significantly associated with younger age, female sex, lower BMI, non-smoking history, elevated CEA levels, hypertension, and diabetes mellitus (all p < 0.05).
Association between individual body composition and clinical outcome
As demonstrated in Figures 2A, B, and C, 5-year OS disadvantage for low vs. high SMI (68.8%, 95% CI [63.3–74.7%] vs. 79.4%, 95% CI [77.2–81.6%], p < 0.001); Poorer outcomes with low SFI (73.7% [69.8–77.8%] vs. 79.3% [77.0−81.8%], p = 0.010); Similarly, the low VFI group showed reduced 5-year OS relative to the high VFI group (68.7%, 95% CI [63.7–74.1%] vs. (79.8%, 95% CI [77.6–82.1%], p < 0.001). Consistent results were observed for SMI and VFI in relation to 5-year RFS (all p < 0.05) (Figure 3A and C). No significant difference in 5-year RFS was observed between the high and low SFI groups (p = 0.050) (Figure 3B).
Fig. 2.
Kaplan-Meier survival curves comparing overall survival among patient groups stratified by skeletal muscle index (SMI) (A), subcutaneous fat index (SFI) (B), visceral fat index (VFI) (C), and their composite index (number of low-value indices: SMI, SFI, and VFI) (D). Statistical significance was determined using log-rank tests
Fig. 3.
Kaplan-Meier survival curves comparing relapse-free survival among patient groups stratified by skeletal muscle index (SMI) (A), subcutaneous fat index (SFI) (B), visceral fat index (VFI) (C), and their composite index (number of low-value metrics: SMI, SFI, and VFI) (D). Statistical significance was determined using log-rank tests
The results of the multivariate analysis are presented in Table 2. Low SMI predicted worse OS (HR = 1.49, 95% CI: 1.16–1.92, p = 0.002); Low SFI impaired OS (HR = 1.54, 95% CI: 1.22–1.94, p < 0.001); Low VFI was associated with increased mortality risk (HR = 1.69, 95% CI: 1.32–2.17, p < 0.001). The associations between these three indices and RFS demonstrated similar statistical significance (all p < 0.05) (Table 2).
Table 2.
Multivariable-adjusted analyses of skeletal muscle index, subcutaneous fat index, visceral fat index, and composite index in relation to overall survival and relapse-free survival
| Variablea | N Patients |
Overall Survival | Relapse-free Survival | ||||
|---|---|---|---|---|---|---|---|
| N Events |
HR (95% CI)a | p b | N Events |
HR (95% CI)a | p b | ||
| SMI | |||||||
| High | 1384 | 332 | Ref | 461 | Ref | ||
| Low | 277 | 100 | 1.49 (1.16–1.92) | 0.002 | 129 | 1.35 (1.09–1.67) | 0.006 |
| SFI | |||||||
| High | 1176 | 285 | Ref | 401 | Ref | ||
| Low | 485 | 147 | 1.54 (1.22–1.94) | < 0.001 | 189 | 1.39 (1.13–1.70) | 0.002 |
| VFI | |||||||
| High | 1338 | 322 | Ref | 458 | Ref | ||
| Low | 323 | 110 | 1.69 (1.32–2.17) | < 0.001 | 132 | 1.30 (1.04–1.64) | 0.021 |
| Composite index* | |||||||
| All high | 979 | 222 | Ref | 320 | Ref | ||
| One low | 348 | 98 | 1.50 (1.16–1.94) | 0.002 | 131 | 1.36 (1.09–1.70) | 0.006 |
| Two low | 265 | 77 | 1.71 (1.29–2.28) | ≤ 0.001 | 98 | 1.38 (1.07–1.78) | 0.013 |
| All low | 69 | 35 | 2.99 (1.89–4.71) | < 0.001 | 41 | 2.13 (1.43–3.16) | < 0.001 |
| P for trend | < 0.001 | < 0.001 | |||||
aCI Confidence interval, HR Hazard ratio, Ref Reference, SMI Skeletal muscle index, SFI Subcutaneous fat index, VFI Visceral fat index
bMultivariate analysis was adjusted for Sex, Age, Smoking history, Hypertension, Diabetes mellitus, Chronic obstructive pulmonary disease, BMI, CEA, N stage, Chemotherapy, Radiotherapy, Pathological stage, Tumor location, Histologic type
*Composite index, number of low values in preoperative SMI, SFI, and VFI
Body composition composite index as prognostic biomarkers
The worst prognostic outcomes were observed in patients exhibiting low index of all three indices (SMI, SFI, and VFI; all p < 0.05) in the combined analysis (Figures 2D and 3D). The 5-year OS was 80% (95% CI: 78.3–83.4%) in the all-high group versus 55.4% (95% CI: 44.6–69.0%), 73.2% (95% CI: 67.9–78.9%), and 76.6% (95% CI: 72.2–81.3%) in groups with all-low, two-low, or one-low, respectively.
Multivariate analysis demonstrated progressively worse OS with an increasing number of low indices (SMI, SFI, or VFI; p for trend < 0.001): HR = 2.99 (95% CI: 1.89–4.71; p < 0.001) for the all-low group, HR = 1.71 (95% CI: 1.29–2.28; p < 0.001) for the two-low group, and HR = 1.50 (95% CI: 1.16–1.94; p = 0.002) for the one-low group, vs. all-high group, and a consistent pattern was also observed in RFS (all p < 0.05) (Table 2). The all-low group had the worst 3-year and 5-year survival outcomes across all pathological stages except for stage IIIB (Supplementary Fig. 1).
Subgroup and sensitivity analyses
Sensitivity analyses confirmed the robustness of the primary findings, with all associations remaining statistically significant (p < 0.05; Supplementary Table 2). Subgroup analyses showed that SMI, SFI, or VFI were consistently associated with shorter OS across most strata (Figure 4). However, this association was not statistically significant in patients younger than 60 years, those with pathological stage I/II disease, or those who underwent thoracoscopic surgery—although the directional trends remained consistent with the overall cohort. Furthermore, significant interactions were observed between body composition indices and both pathological stage (p = 0.018) and histological type (p = 0.038), suggesting effect modification (Figure 4).
Fig. 4.
Forest plots present the analysis results of body composition abnormality (defined by low skeletal muscle index [SMI], subcutaneous fat index [SFI], or visceral fat index [VFI]) subgroups stratified by clinicopathological variables
Further stratified analyses showed that a low composite index was associated with poor prognosis in both stages I/II and stage III disease, although the risk patterns differed by stage. In early-stage disease, low SMI and low VFI were associated with increased risk, with the poorest survival observed only when all three indices (SMI, SFI, and VFI) were concurrently reduced, indicative of severe systemic wasting. In advanced-stage disease, low SMI, low SFI, and low VFI were each independent risk factors, with the highest risk seen in patients with concomitant reductions in two indices, whereas further reduction across all three indices did not amplify risk (Supplementary Table 3).
Discussion
This study developed a novel composite body composition index by integrating SMI, SFI, and VFI based on low-value index counts. Associations were identified between composite indices and clinical prognosis in NSCLC patients. Multivariable analyses showed worsening OS and RFS with increasing numbers of below-threshold body composition indices (SMI, SFI, or VFI). Reduced SMI was independently associated with worse clinical outcomes, consistent with previous studies assessing SMI at alternative anatomical landmarks [16, 18, 27]. A recent systematic review further supports the prognostic relevance of SFI and VFI, demonstrating their associations with poor survival in various cancers [28]. Our findings confirm these associations in NSCLC, aligning with results from multiple independent cohorts [29–31]. Contrary to our results, Barbi et al. reported that high VFI predicted worse OS and RFS [32]. These differences may result from distinct cutoff selection methods (maximal selected rank statistics vs. tertile-based approach) and inherent tumor heterogeneity. Reduced muscle mass may adversely affect functional status and overall health in patients with NSCLC [33].
This study found that patients with concomitant low SMI, SFI, and VFI generally have poorer prognosis, suggesting that concurrent depletion of muscle and fat may constitute a high-risk body composition phenotype. Muscle wasting may reduce the secretion of immunoregulatory myokines, while abnormal fat loss or redistribution may disrupt adipokine signaling and metabolic regulation, thereby impairing immune and metabolic functions [34, 35]. This combined deficiency of muscle and fat may compromise the body’s capacity to adapt to inflammatory and energetic stress, reduce treatment tolerance, and increase the risk of cachexia, ultimately contributing to adverse outcomes [36, 37]. Although these mechanisms are supported by experimental and clinical evidence, their precise pathways require further investigation.
Another key finding of this study was the significant interaction between pathological stage, histologic type, and abnormalities in body composition (all p < 0.05). Although body composition was significantly associated with overall survival in the overall cohort, no statistically significant association was observed in the subgroup of patients with stages I/II disease (Figure 4). This difference may primarily be attributed to the clinical characteristic differences between patients at different disease stages. Advanced-stage patients are more likely to experience significant metabolic disturbances and systemic inflammation, which can exacerbate the impact of body composition deterioration on treatment tolerance and survival outcomes [38, 39]. By contrast, prognosis in early-stage disease is more strongly driven by tumor biology and the completeness of surgical resection, potentially attenuating the observable influence of body composition [40]. Notably, even in stages I/II patients, the direction of the hazard ratios was consistent with that of the overall cohort (Figure 4), and mortality risk remained significantly increased in those in the composite index all-low group (Supplementary Table 3), indicating that severe body composition impairment has prognostic relevance across all disease stages. Collectively, these findings suggest that the prognostic impact of body composition may be stage dependent and tends to become more apparent as the disease progresses.
This study has important clinical implications. According to current clinical guidelines, patients undergoing treatment surveillance are routinely required to undergo contrast-enhanced CT scans of the chest, abdomen, and pelvis every 6–12 weeks [41], which provides an objective basis for standardized, non-invasive quantitative assessment at the third lumbar vertebral level and enables longitudinal monitoring of SMI, SFI, and VFI. Beyond the prognostic value of individual parameters, a composite index derived from these measures may more effectively identify patients at risk of concurrent skeletal muscle and adipose tissue depletion. Recent evidence suggests that reductions in skeletal muscle mass are associated with attenuated responses to immune checkpoint inhibitors [42, 43], while the distribution and quality of adipose tissue may also influence the tumor immune microenvironment [31]. From a clinical perspective, declines in SMI may be improved through individualized nutritional support and structured exercise interventions [44], whereas adipose tissue abnormalities may require integrated management incorporating metabolic regulation and immunonutritional strategies [45, 46]. The composite index can serve as a practical tool for risk stratification, helping to identify high-risk patients early and guide supportive treatment. Although this study did not analyze immunotherapy prognosis, its potential value in immunotherapy evaluation warrants further exploration.
Notably, unlike prior studies that focused on individual body composition indices [47, 48], this study concurrently evaluated SMI, SFI, and VFI to assess their combined prognostic value. Additionally, the retrospective cohort design featured an extended follow-up period of up to 10 years, providing robust long-term outcome data.
It should be noted that 1,294 patients (41% of the initial cohort) were excluded due to missing CT images or having undergone imaging at external hospitals, which may introduce selection bias. However, the included and excluded patients were generally comparable in most key clinical characteristics, including body composition metrics, demographic features, tumor subtypes, and primary treatment modalities, with no significant differences. Although some variables, such as BMI, comorbidities, and pathological stage, showed statistically significant differences, the absolute magnitude of these differences was limited (mostly 5–8%) and were adjusted for as covariates in multivariable analyses (Supplementary Table 1). Therefore, the potential impact of this bias on the internal validity of the study is considered to be manageable.
Nevertheless, several limitations of this study should be acknowledged. As a single-center retrospective analysis, the sample exhibited limited heterogeneity, potentially affecting the generalizability of the findings to the broader population of non-small cell lung cancer patients. In addition, the absence of dynamic data on body composition, as well as information on patients’ physical performance and nutritional interventions, may result in residual confounding. Future large, multicenter prospective studies with longitudinal body composition assessments are needed to validate these findings and clarify the prognostic roles of SMI, SFI, and VFI in patients with NSCLC.
Conclusion
This study showed that preoperative SMI, SFI, and VFI were associated with survival outcomes in patients with NSCLC. A graded relationship was observed, whereby an increasing number of below-threshold body composition indices was associated with progressively worse OS and RFS. Given the lack of detailed physical fitness data, residual confounding cannot be excluded; nevertheless, the combined assessment of SMI, SFI, and VFI may provide complementary prognostic information and aid risk stratification in patients undergoing surgical resection.
Supplementary Information
Supplementary Material 1: Supplement Figure 1 Comparison of survival rates based on composite index (number of low-value indices: skeletal muscle index, subcutaneous fat index, and visceral fat index) across pathological stages. 3-Year Survival (A); 5-Year Survival (B).
Acknowledgements
The authors would like to express their sincere gratitude to all participants involved in data collection, statistical guidance, and study design for their invaluable contributions.
Abbreviations
- BMI
Body mass index
- CEA
Carcinoembryonic Antigen
- CI
Confidence interval
- CT
Computed tomography
- HR
Hazard ratio
- HU
Hounsfield unit
- IQR
Interquartile range
- L3
Third lumbar vertebra
- NSCLC
Non-small cell lung cancer
- OS
Overall survival
- RFS
Relapse-free survival
- SAT
Subcutaneous adipose tissue
- SFI
Subcutaneous fat index
- SMA
Skeletal muscle area
- SMI
Skeletal muscle index
- STROBE
Strengthening the Reporting of Observational Studies in Epidemiology
- VAT
Visceral adipose tissue
- VFI
Visceral fat index
Authors’ contributions
Fahui Chen, Zhihui Shi, and Hua Bai contributed to study conception, data analysis, and manuscript drafting. Hongtao Lei, Hongjiang Pu, Yang Yang, Yongmei Wu, Ying Zhao, Xin Ning, Guanghong Yan, Mengmei Liu, Yani Li, Sifan Duan, Hanqun Liu, Chang Shu, Lu Lu, and Wenjing Xia collected and organized the data. Xin Ning, Xuewen Zhang prepared Figs. 1, 2 and 3. Zhenhui Li and Dingyun You supervised the study and revised the manuscript. All authors reviewed and approved the final version.
Funding
This study has received funding by the grant of the Innovative Research Team of Yunnan Province (grant No. 202405AS350016), the First-Class Discipline Team of Kunming Medical University (grant No. 2024XKTDTS17), Medical high-end talents training program of Yunnan Health Commission (grant No. L-2024009), Yunnan Applied Basic Research Projects-Kunming Medical University Union Foundation (grant No. 202501AY070001-003), Biomedical Specialization of Yunnan Province (grant No. 202402AA310012), Academician (Expert) Workstation Program (grant No. 202505AF350006), and Key Laboratory of Integrated Care for Geriatric Chronic Diseases (Kunming Medical University, School of Nursing), Yunnan Provincial Education Department (grant No. 2024HTHLYB02), Yunnan Fundamental Research Projects (grant No. 202101AZ070001-103).
Data availability
We sincerely regret that the data used in this study cannot be made publicly available. The datasets involve sensitive patient information, and due to the retrospective nature of the study, it was not possible to obtain informed consent from all participants. Moreover, our research is subject to institutional confidentiality agreements, which limit the ability to share these data externally. We kindly ask for your understanding and are happy to provide further methodological details upon request. Regarding the original data of this study, please contact the corresponding author, Dingyun You, via email at youdingyun@kmmu.edu.cn.
Declarations
Ethics approval and consent to participate
The study protocol was approved by the Institutional Review Board of Kunming Medical University (approval no. KMMU2025MEC154). In accordance with the retrospective nature of the study and the full de-identification of patient data, the ethics committee granted a waiver of informed consent. This study was conducted in compliance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. All procedures involving human data were performed in accordance with the Declaration of Helsinki.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Fahui Chen, Zhihui Shi and Hua Bai are contributed equally, and shared the first authorship.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Zhenhui Li and Dingyun You are contributed equally, and shared the last authorship.
Contributor Information
Zhenhui Li, Email: lizhenhui@kmmu.edu.cn.
Dingyun You, Email: youdingyun@kmmu.edu.cn.
References
- 1.Uramoto H, Tanaka F. Recurrence after surgery in patients with NSCLC. Translational lung cancer Res. 2014;3:242. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.McGovern J, Dolan RD, Horgan PG, Laird BJ, McMillan DC. Computed tomography-defined low skeletal muscle index and density in cancer patients: observations from a systematic review. J Cachexia Sarcopenia Muscle. 2021;12:1408–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Maurits JS, Sedelaar JM, Mulders PF, Aben KK, Kiemeney LA, Vrieling A. Skeletal muscle radiodensity and visceral adipose tissue index are associated with survival in renal cell cancer–A multicenter population-based cohort study. Clin Nutr. 2022;41:131–43. [DOI] [PubMed] [Google Scholar]
- 4.Lee JH, Kang D, Ahn JS, Guallar E, Cho J, Lee HY. Obesity paradox in patients with non-small cell lung cancer undergoing immune checkpoint inhibitor therapy. J Cachexia Sarcopenia Muscle. 2023;14:2898–907. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Khan A, Welman CJ, Abed A, et al. Association of computed tomography measures of muscle and adipose tissue and progressive changes throughout treatment with clinical endpoints in patients with advanced lung cancer treated with immune checkpoint inhibitors. Cancers. 2023;15:1382. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Lopez P, Newton RU, Taaffe DR, et al. Associations of fat and muscle mass with overall survival in men with prostate cancer: a systematic review with meta-analysis. Prostate Cancer Prostatic Dis. 2022;25:615–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Caan BJ, Cespedes Feliciano EM, Kroenke CH. The Importance of Body Composition in Explaining the Overweight Paradox in Cancer-Counterpoint. Cancer Res. 2018;78:1906–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Strulov Shachar S, Williams GR. The Obesity Paradox in Cancer-Moving Beyond BMI. Cancer Epidemiol Biomarkers Prev. 2017;26:13–6. [DOI] [PubMed] [Google Scholar]
- 9.Cespedes Feliciano EM, Kroenke CH, Caan BJ. The Obesity Paradox in Cancer: How Important Is Muscle? Annu Rev Nutr. 2018;38:357–79. [DOI] [PubMed] [Google Scholar]
- 10.Jin Y, Ma X, Yang Z, Zhang N. Low L3 skeletal muscle index associated with the clinicopathological characteristics and prognosis of ovarian cancer: a meta-analysis. J cachexia sarcopenia muscle. 2023;14:697–705. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Brown KA, Scherer PE. Update on adipose tissue and cancer. Endocr Rev. 2023;44:961–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Yang M, Shen Y, Tan L, Li W. Prognostic Value of Sarcopenia in Lung Cancer: A Systematic Review and Meta-analysis. Chest. 2019;156:101–11. [DOI] [PubMed] [Google Scholar]
- 13.Mitsiopoulos N, Baumgartner RN, Heymsfield SB, Lyons W, Gallagher D, Ross R. Cadaver validation of skeletal muscle measurement by magnetic resonance imaging and computerized tomography. J Appl Physiol (1985). 1998;85:115–22. [DOI] [PubMed] [Google Scholar]
- 14.Rosenquist KJ, Pedley A, Massaro JM, et al. Visceral and subcutaneous fat quality and cardiometabolic risk. JACC Cardiovasc Imaging. 2013;6:762–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Gezer NS, Bandos AI, Beeche CA, Leader JK, Dhupar R, Pu J. CT-derived body composition associated with lung cancer recurrence after surgery. Lung Cancer. 2023;179:107189. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Vedire Y, Nitsche L, Tiadjeri M, et al. Skeletal muscle index is associated with long term outcomes after lobectomy for non-small cell lung cancer. BMC Cancer. 2023;23:778. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hata A, Muraguchi Y, Nakatsugawa M, et al. Automated chest CT three-dimensional quantification of body composition: adipose tissue and paravertebral muscle. Sci Rep. 2024;14:32117. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Mantz L, Mercaldo ND, Simon J, et al. Preoperative Chest CT Myosteatosis Indicates Worse Postoperative Survival in Stage 0-IIB Non-Small Cell Lung Cancer. Radiology. 2025;314:e240282. [DOI] [PubMed] [Google Scholar]
- 19.Fu L, Ding H, Mo L, et al. The association between body composition and overall survival in patients with advanced non-small cell lung cancer. Sci Rep. 2025;15:3109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Huang Y, Cun H, Mou Z, et al. Multiparameter body composition analysis on chest CT predicts clinical outcomes in resectable non-small cell lung cancer. Insights Imaging. 2025;16:32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Grønberg BH, Sjøblom B, Wentzel-Larsen T, et al. A comparison of CT based measures of skeletal muscle mass and density from the Th4 and L3 levels in patients with advanced non-small-cell lung cancer. Eur J Clin Nutr. 2019;73:1069–76. [DOI] [PubMed] [Google Scholar]
- 22.Liu X, Ji W, Zheng K, et al. The correlation between skeletal muscle index of the L3 vertebral body and malnutrition in patients with advanced lung cancer. BMC Cancer. 2021;21:1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Argilés JM, López-Soriano J, Almendro V, Busquets S, López‐Soriano FJ. Cross‐talk between skeletal muscle and adipose tissue: a link with obesity? Med Res Rev. 2005;25:49–65. [DOI] [PubMed] [Google Scholar]
- 24.Golder AM, Sin LKE, Alani F, et al. The relationship between the mode of presentation, CT-derived body composition, systemic inflammatory grade and survival in colon cancer. J Cachexia Sarcopenia Muscle. 2022;13:2863–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Bradley NA, Walter A, Dolan R, et al. Evaluation of the prognostic value of computed tomography-derived body composition in patients undergoing endovascular aneurysm repair. J Cachexia Sarcopenia Muscle. 2023;14:1836–47. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Zheng Y, Luo Q, He Y, et al. Associations of body composition and inflammatory parameters with survival in patients with resectable non-small cell lung cancer receiving neoadjuvant chemoimmunotherapy. Transl Lung Cancer Res. 2025;14:3009–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Recio-Boiles A, Galeas JN, Goldwasser B, et al. Enhancing evaluation of sarcopenia in patients with non-small cell lung cancer (NSCLC) by assessing skeletal muscle index (SMI) at the first lumbar (L1) level on routine chest computed tomography (CT). Support Care Cancer. 2018;26:2353–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Lou J, Guo Y, Li L, et al. Explanation of the obesity paradox of immunotherapy in cancer patients using CT-derived adipose composition parameters: A systematic review and meta-analysis. Int Immunopharmacol. 2025;144:113699. [DOI] [PubMed] [Google Scholar]
- 29.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 into Imaging. 2023;14:182. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Makrakis D, Rounis K, Tsigkas AP, et al. Effect of body tissue composition on the outcome of patients with metastatic non-small cell lung cancer treated with PD-1/PD-L1 inhibitors. PLoS ONE. 2023;18:e0277708. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Degens J, Dingemans AC, Willemsen ACH, et al. The prognostic value of weight and body composition changes in patients with non-small-cell lung cancer treated with nivolumab. J Cachexia Sarcopenia Muscle. 2021;12:657–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.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. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Bye A, Sjøblom B, Wentzel-Larsen T, et al. Muscle mass and association to quality of life in non-small cell lung cancer patients. J Cachexia Sarcopenia Muscle. 2017;8:759–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Czaja-Stolc S, Chatrenet A, Potrykus M et al. Adipokines and Myokines as Markers of Malnutrition and Sarcopenia in Patients Receiving Kidney Replacement Therapy: An Observational, Cross-Sectional Study. Nutrients. 2024;16:248016. [DOI] [PMC free article] [PubMed]
- 35.Park MJ, Choi KM. Interplay of skeletal muscle and adipose tissue: sarcopenic obesity. Metabolism. 2023;144:155577. [DOI] [PubMed] [Google Scholar]
- 36.Mariean CR, Tiucă OM, Mariean A, Cotoi OS. Cancer Cachexia: New Insights and Future Directions. Cancers (Basel). 2023;15:5590. [DOI] [PMC free article] [PubMed]
- 37.Zhu Y, Zhang Y, Li M, et al. Prognostic Value of Systemic Inflammation, Nutritional Status and Sarcopenia in Patients With Amyotrophic Lateral Sclerosis. J Cachexia Sarcopenia Muscle. 2024;15:2743–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.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. [DOI] [PMC free article] [PubMed]
- 39.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. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Yang X, Xing Z, Long S, et al. Comparative effectiveness for early-stage NSCLC without lymph node involvement based on prospective studies. Int J Surg. 2025;111:5535–48. [DOI] [PubMed] [Google Scholar]
- 41.Riely GJ, Wood DE, Ettinger DS, et al. Non-Small Cell Lung Cancer, Version 4.2024, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw. 2024;22:249–74. [DOI] [PubMed] [Google Scholar]
- 42.Wang J, Cao L, Xu S. Sarcopenia affects clinical efficacy of immune checkpoint inhibitors in non-small cell lung cancer patients: A systematic review and meta-analysis. Int Immunopharmacol. 2020;88:106907. [DOI] [PubMed] [Google Scholar]
- 43.Trestini I, Belluomini L, Dodi A, et al. Body composition derangements in lung cancer patients treated with first-line pembrolizumab: A multicentre observational study. J Cachexia Sarcopenia Muscle. 2024;15:2349–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Yang Y, Pan N, Luo J, Liu Y, Ossowski Z. Exercise and Nutrition for Sarcopenia: A Systematic Review and Meta-Analysis with Subgroup Analysis by Population Characteristics. Nutrients. 2025;17:2342. [DOI] [PMC free article] [PubMed]
- 45.Xu S, Lu F, Gao J, Yuan Y. Inflammation-mediated metabolic regulation in adipose tissue. Obes Rev. 2024;25:e13724. [DOI] [PubMed] [Google Scholar]
- 46.Choi W, Woo GH, Kwon TH, Jeon JH. Obesity-Driven Metabolic Disorders: The Interplay of Inflammation and Mitochondrial Dysfunction. Int J Mol Sci. 2025;26:9715. [DOI] [PMC free article] [PubMed]
- 47.Rounis K, Makrakis D, Tsigkas A, et al. 1299P Effect of fat tissue composition on the outcome of patients with metastatic non-small cell lung cancer treated with PD-1/PD-L1 inhibitors. Ann Oncol. 2021;32:S1006. [Google Scholar]
- 48.Tao J, Fang J, Chen L, et al. Increased adipose tissue is associated with improved overall survival, independent of skeletal muscle mass in non-small cell lung cancer. J Cachexia Sarcopenia Muscle. 2023;14:2591–601. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Material 1: Supplement Figure 1 Comparison of survival rates based on composite index (number of low-value indices: skeletal muscle index, subcutaneous fat index, and visceral fat index) across pathological stages. 3-Year Survival (A); 5-Year Survival (B).
Data Availability Statement
We sincerely regret that the data used in this study cannot be made publicly available. The datasets involve sensitive patient information, and due to the retrospective nature of the study, it was not possible to obtain informed consent from all participants. Moreover, our research is subject to institutional confidentiality agreements, which limit the ability to share these data externally. We kindly ask for your understanding and are happy to provide further methodological details upon request. Regarding the original data of this study, please contact the corresponding author, Dingyun You, via email at youdingyun@kmmu.edu.cn.