Assessment of the correlation between visceral adiposity and liver fat in metastatic colorectal cancer patients

Abstract

Purpose

The study aims to evaluate the correlation between visceral adiposity, fatty liver, and survival in patients with metastatic colorectal cancer (mCRC).

Methods

The study included 131 adult patients with treatment-naive mCRC. The visceral and liver fat content was measured using baseline computed tomography (CT) images. The analysis used the 50th percentile (131.80 HU) visceral adiposity value as a cutoff. The visceral and liver fat content association with patient characteristics and outcomes was assessed.

Results

In the overall cohort, neither visceral adiposity (median OS 37.8 vs 36.7 months, HR = 0.83; p = 0.428) nor liver steatosis (median OS 46.0 vs 33.9 months, HR = 0.81; p = 0.370) showed significant association with OS. However, in patients with BMI ≥ 25 kg/m², liver steatosis was associated with significantly shorter survival (median OS 35.2 vs 59.5 months; adjusted HR = 0.49; p = 0.040). Visceral adiposity remained non-significant across BMI subgroups.

Conclusion

We observed a possible association between liver steatosis and OS in patients with mCRC in the high BMI subgroup. Prospective studies are essential to validate these findings and the role of liver steatosis in the prognostic assessment of mCRC patients.

Introduction

Obesity is linked to several types of cancer, including pancreatic, gastric, colorectal, and hepatic cancers [1, 2]. Obesity-associated cancers constitute about 27% of the global cancer burden [3, 4]. Colorectal cancer (CRC) is the third most commonly diagnosed cancer and a leading cause of cancer death. The liver is the most common site for CRC metastasis due to portal blood flow from the bowel to the liver [5, 6].

Recent research findings have demonstrated that the risk of CRC may increase by over 40% for both overweight and obese individuals [7]. As research into the energy mechanisms within tumor cells has expanded, studies increasingly highlight the potential significance of body fat density, particularly visceral and liver fat, in tumor cell metabolism and microenvironment [8–10]. So, visceral and liver fat are important factors in shaping how colorectal cancers grow and spread. Research shows that metastatic tumor cells often change their energy metabolism, increasing fatty acid synthesis and tweaking fatty acid oxidation [11]. These changes are crucial for the survival and growth of cancer cells and how they affect the surrounding tumor environment, possibly boosting their ability to metastasize and influencing patient outcomes [12]. However, the specific influence of visceral adipose tissue (VAT) and fatty liver on CRC metastasis has remained a subject of inquiry. It is postulated that VAT components play a pivotal role in fostering an environment conducive to developing colorectal tumors [13]. This is achieved through altering cellular metabolism and the modulation of communication within the tumor microenvironment via various mediators, including free fatty acids, cytokines, and hormones [14].

Previous studies have shown the association between fatty liver, visceral adiposity, early mortality, long-term survival outcomes, postoperative complications, and chemotherapy toxicity in localized colorectal cancer [15]. Data in patients with metastatic CRC are contradictory. A previous meta-analysis showed that CRC patients with a higher BMI had reduced mortality compared with normal-weight CRC patients [16]. In contrast, another study showed an increased risk of all-cause mortality and cancer-specific mortality in obese patients [17]. Higher BMI may have paradoxical effects on survival: it could offer nutritional protection compared to cachexia, while the inflammatory environment associated with obesity might indicate a poorer prognosis. Given the preclinical evidence implicating the role of adiposity in metastasis, we sought to evaluate the status of liver steatosis and visceral adiposity, including the prevalence, associated factors, and prognostic role in terms of overall survival in patients with metastatic colorectal cancer.

Materials and methods

Data collection

The retrospective cohort study encompassed adults (≥ 18 years) diagnosed with stage IV mCRC treated at the Hacettepe University Oncology Hospital (Ankara, Turkiye) between 2018 and 2023. Newly diagnosed, previously untreated mCRC patients who had baseline abdominal CT imaging before treatment and had a follow-up of at least 1 year were included. There were no specific exclusion criteria. The baseline characteristics, including age at diagnosis, gender, mutation status (KRAS, BRAF, NRAS), mismatch repair (MMR) status (pMMR/dMMR), primary tumor side (left, right), sites, and number of metastases (liver, other), were extracted from an anonymous database. Liver steatosis and visceral fat ratio were central to our study hypothesis as established markers of metabolic health and systemic inflammation. BMI was included to account for overall obesity, while diabetes, hypertension, and coronary artery disease were considered due to their relevance in metabolic dysfunction and cancer outcomes. RAS and BRAF mutation status, given its prognostic impact in mCRC, was included to control for molecular heterogeneity. Lastly, metastasis site and number were incorporated as critical clinical predictors of disease burden and prognosis. All data were extracted for statistical analysis in a database with the agreement of the local Health Sciences Research Ethics Committee.

Anthropometric measurements

Weight and height measurements were taken at the baseline examination upon CRC diagnosis, and BMI was calculated as kg/m².

Measurement of visceral and liver fat

Fat segmentation was evaluated through CT scans of the abdominal region obtained before treatment while the patient was in a dorsal decubitus position. Measurements were performed at the middle third lumbar vertebral corpus (L3) level after transferring the patient images from the PACS system to Syngo.via software (Siemens Healthcare, Forchheim, Germany) (Fig. 1). The program outputs the measured values in cm³ and standardizes the z-axis of the voxels in the section to 1 cm by adjusting the measured values according to a section thickness of 1 cm (for example, multiplying the value measured at a section thickness of 5 mm by 2). The new values correspond to the area values provided in cm² in the literature, and these adjusted values were used for area measurement results.

Fig. 1.

Axial CT images demonstrating the sequential stages of visceral fat quantification, including the raw image (left), processed segmentation (top right), and isolated visceral fat (bottom right)

After selecting the level (mid-corpus of L3) and thickness (1 cm) of the CT image in axial view, the volume of interest (VOI) for visceral adipose tissue measurement was automatically defined by the Syngo.via software. The visceral fat compartment within the VOI was measured by quantifying pixels within a density range of − 200 to − 40 Hounsfield units (HU) using Syngo.via client software [18]. Visceral organ parenchyma within the VOI was automatically excluded during segmentation due to its higher density (> − 40 HU). Bowel lumens, which presented densities in the range of − 200 to − 40 HU, were automatically classified as fat content by the software in the color-coded CT image and were excluded manually.

Fatty liver has been qualitatively and quantitatively assessed on non-contrast and contrast-enhanced CT images. On non-contrast CT, normal hepatic parenchyma typically exhibits greater attenuation than the spleen. Steatosis of the liver reduces its attenuation, making it less than that of the spleen. Furthermore, large intrahepatic vascular structures appear hypodense due to the lower attenuation of blood compared to healthy hepatic tissue. In patients with advanced hepatosteatosis, intrahepatic vascular structures may appear more hyperdense than in normal liver parenchyma. A hepatic attenuation value below 40 HU or a liver-to-spleen attenuation difference of ≥ 10 HU is diagnostic of hepatic steatosis on non-contrast CT images. The non-contrast CT scans of the patients were evaluated according to these two methods mentioned above, and the presence or absence of hepatosteatosis was classified [19, 20]. Measurement of density in the liver parenchyma was performed by manually placing a region of interest (ROI) at segments IV and VIII on non-contrast CT and obtaining the mean density value. The diagnosis of fatty liver on contrast-enhanced CT images was based on visual assessment of the density difference between the liver and spleen, requiring a density difference of ≥ 20 HU between the two [21]. We assessed all quartiles of VAT measurements as cut-offs to test the association with survival, and the median value was the best prognosticator for survival compared to other quartiles. Thus, we used the 50th percentile (131.80) value as a cutoff.

Statistical analysis

Statistical analyses were conducted using the SPSS (IBM SPSS Statistics 25) program. To determine whether the numeric variables had a normal distribution was evaluated by the histogram curve, Q-Q Plot graphics, and Kolmogorov–Smirnov test. To define the OS time, the period from diagnosis to the last follow-up or death was used, and the period between diagnosis and the time of disease progression/death was depicted as progression-free survival (PFS) time. Kaplan–Meier analysis was employed to conduct survival analysis, and survival times were compared between prognostic subgroups using the log-rank test. Cox regression analyses were used for multivariate analysis, and hazard ratios were calculated using a 95% CI.

Results

Baseline characteristics

This study involved 131 treatment-naive patients diagnosed with mCRC between January 2018 and December 2022. The median age of the patients was 61 years (Inter Quartile Range (IQR): 51–68). As shown in Table 1, the majority of tumors were located in the colon (n = 102, 77.9%), while 29 patients (22.1%) had rectal tumors. In terms of laterality, most tumors originated from the left colon (80.9%). The most common molecular status was pan-RAS and BRAF wt (58.0%). Most patients had a single metastasis site (73.3%) and liver metastasis (79.4%). Patients were mostly treated with irinotecan-based doublets (59%) followed by oxaliplatin-based doublets (41%), and anti-VEGF or anti-EGFR treatment was used per RAS/BRAF mutation status. Of the 131 patients included, 49 (37.4%) underwent primary tumor resection, while 44 patients (33.6%) had metastasectomy during the treatment course. These procedures were performed either to manage complications (such as obstruction or bleeding) or as part of a curative-intent strategy in selected patients showing a favorable response to systemic therapy.

Table 1.

Baseline characteristics of the patients

Age, median, ❨IQR❩		61.12 (51.43–67.93)
Gender, n (%)	Female	61 (46.6)
	Male	70 (53.4)
Primary tumor side, n (%)	Left	106 (80.9)
	Right	25 (19.1)
Location, n (%)	Colon	102 (77.9)
	Rectum	29 (22.1)
Metastasis site, n (%)	Liver	104 (79.4)
	Other	27 (20.6)
Metastasis site number, n (%)	One	96 (73.3)
	Two or more	35 (26.7)
Molecular status, n (%)	K-RASm	40 (30.5)
	N-RASm	2 (1.5)
	BRAF V600Em	13 (9.9)
	RAS and BRAF wt	76 (58.0)
MSI status, n (%)	pMMR	126 (96.2)
	dMMR	5 (3.8)
BMI, n (%)	< 25 kg/m2	60 (45.8)
	≥ 25 kg/m2	71 (54.2)
Visceral fat	> 50th perc	65 (49.4)
	≤ 50th perc	66 (50.4)
Fatty liver	Positive	60 (54.8)
	Negative	71 (45.2)
Diabetes mellitus, n (%)	Yes	72 (55.0)
	No	59 (45.0)
Hypertension, n (%)	Yes	41 (33.9)
	No	80 (66.1)
Coronary artery disease, n (%)	Yes	15 (12.4)
	No	106 (87.6)

MSI: Microsatellite  Instability, BMI: Body Mass Index,  IQR :Interquartile Range 

During the median follow-up period of 57.4 months, 78 patients (59.5%) died. The median OS of all patients was 37.1 months. Patients with RAS or BRAF mutation had shorter survival compared to the wild type (HR = 2.07, 95% CI 1.32–3.24, p = 0.001), and patients with a higher number of metastases (multiple vs single, HR = 2.14, 95% CI 1.33–3.44, p = 0.002) were associated with lower survival (Table 2.).

Table 2.

Univariate analysis of overall survival for all patients, normal and high BMI patients

		All patients		BMI ≥ 25		BMI < 25
		HR (95% CI)	P value	HR (95% CI)	P value	HR (95% CI)	P value
Liver steatosis	Absence	0.81 (0.52–1.27)	0.370	0.56 (0.30–1.04)	0.066	1.11 (0.55–2.23)	0.760
	Presence (ref)
Visceral fat ratio 50.perc	≤ 131.80	0.83 (0.53–1.30)	0.428	0.61(0.30–1.23)	0.166	0.98 (0.47–2.04)	0.962
	> 131.80 (ref)
RAS and BRAF status	wt (ref)	2.07 (1.32–3.24)	0.001	1.92 (1.04–3.53)	0.035	2.13 (1.10–4.15)	0.025
	Mutant
Metastasis site number	1 (ref)	2.14 (1.33–3.44)	0.002	1.84 (0.98–3.43)	0.054	3.11 (1.40–6.87)	0.005
	> 1
Location	Rectum	0.81 (0.47–1.40)	0.819	0.85 (0.42–1.71)	0.857	0.81 (0.34–1.96)	0.818
	Colon (ref)
Metastasectomy	Presence	0.34 (0.20–0.59)	< 0.001	0.42 (0.22–0.80)	0.009	0.21 (0.06–0.70)	0.011
	Absence (ref)
BMI	≥ 25 (ref)	0.90 (0.57–1.14)	0.665	–-	–-	–-	–-
	< 25
Metastasis site	Liver	0.62 (0.37–1.04)	0.071	0.48 (0.24–0.93)	0.031	0.90 (0.39–2.08)	0.810
	Extrahepatic (ref)
DM	Absence	1.15 (0.73–1.79)	0.534	1.25 (0.67–2.32)	0.472	1.04 (0.53–2.01)	0.915
	Presence (ref)
Hypertension	Absence	1.01 (0.63–1.60)	0.977	1.62 (0.87–3.02)	0.128	0.49 (0.23–1.04)	0.064
	Presence (ref)
CAD	Absence	0.83 (0.35–1.73)	0.619	0.92 (0.32–2.60)	0.871	0.73 (0.26–2.07)	0.556
	Presence (ref)

BMI: Body Mass Index, DM: Diabetes Mellitus, CAD: Coronary Artery Disease, HR: Hazard Ratio, CI : Confidence Interval, ref : reference

There were no associations between fatty liver and age (p = 0.675), gender (p = 0.380), BMI categories (p = 0.159), diabetes (p = 0.487), sites of metastases (p = 0.284), MSI status (p = 1.000), molecular status (p = 0.992), and primary tumor side (p = 0.373). The visceral fat ratio in patients was significantly higher with high BMI (p < 0.001) and in those with diabetes (p = 0.045). Detailed data on complications related to oncological or surgical treatment were not collected; however, we reassessed patients’ information and found no fatal complications other than those related to disease progression.

VAT, fatty liver, and OS

When all patients were evaluated, it was seen that neither liver steatosis (median OS, 46.0 vs 33.9 months; HR = 0.81; 95% CI, 0.52–1.27; p = 0.370) nor the amount of visceral fat tissue (median OS, 37.8 vs 36.7 months; HR = 0.83; 95% CI, 0.53–1.30; p = 0.428) were associated with survival (Table 2., Fig. 2A, B). The presence of RAS and BRAF mutation (HR 2.07 (1.32–3.24), p = 0.001) and > 1 metastatic sites (HR 2.14 (1.33–3.44), p = 0.002) were significantly associated with lower OS (Table 2.).

Fig. 2.

Kaplan–Meier curves for overall survival based on fat distribution. A Comparison of OS between patients with the presence or absence of liver fattiness (p = 0.370, log-rank test). B Comparison of OS between patients with visceral fat ratio at or below the 50th percentile (≤ 131.80) and those above the 50th percentile (> 131.80) (p = 0.427, log-rank test)

Subgroup analysis for OS was performed in patients with normal and high BMI. Among patients with high BMI, univariate analysis for OS showed that the presence of liver steatosis, RAS and BRAF mutation, presence of extrahepatic metastases and number of metastatic sites, and presence of metastasectomy were associated with OS, while the presence of diabetes mellitus (HR = 1.25; 95% CI, 0.67–2.32; p = 0.534), hypertension (HR, 1.62; 95% CI, 0.87–3.02; p = 0.064), and coronary artery disease (HR, 0.92; 95% CI, 0.32–2.60; p = 0.619) were not associated with OS (Table 2.).

BMI body mass index, DM diabetes mellitus, CAD coronary artery disease, HR hazard ratio, CI confidence interval, perc percentile, ref reference.

Multivariate analysis showed that after adjustment for the number and site of metastases, and presence of metastasectomy and mutation status, those with liver steatosis had shorter survival than those without fatty liver (median OS, 59.5 vs 35.2 months; HR = 0.49; 95% CI, 0.25–0.96; p = 0.040) (Fig. 3, Table 3). However, among patients with a normal BMI, OS was similar between patients with and without fatty liver (HR 1.11, 95% CI 0.55–2.23, p = 0.760).

Fig. 3.

Kaplan–Meier analysis of overall survival stratified by liver steatosis in the BMI ≥ 25 subgroup (p = 0.040, log-rank test)

Table 3.

Multivariate Cox-regression analysis for high BMI patients

		HR (95% CI)	P value
Liver steatosis	Absence	0.49 (0.25–0.96)	0.040
	Presence (ref)
Visceral fat ratio 50th percentile	≤ 131.80	0.65 (0.32–1.32)	0.237
	> 131.80 (ref)
RAS and BRAF status	wt	1.13 (0.54–2.37)	0.733
	Mutant (ref)
Metastasis site number	1	0.40 (0.20–0.78)	0.008
	> 1 (ref)
Metastasis site	Liver	0.58 (0.29–1.17)	0.133
	Extrahepatic (ref)
Metastasectomy	Presence	0.44 (0.22–0.85)	0.015
	Absence (ref)

Visceral fat tissue was not associated with survival in the high BMI (HR = 0.61, 95% CI 0.30–1.23, p = 0.166) and normal BMI (HR = 0.98, 95% CI 0.47–2.04, p = 0.962) subgroups.

Discussion

In this study, we have shown that liver steatosis and visceral fat levels were not associated with survival in patients with mCRC; however, in patients with a high BMI (> 25), the presence of fatty liver was associated with shorter OS after adjustment for other clinical and molecular risk factors.

The liver is the main site for metastasis in CRC [22]. Despite this, only a handful of studies have delved into the relationship between preexisting liver diseases and the onset of CRC liver metastasis, as well as the subsequent prognosis of affected patients. A noteworthy meta-analysis involving ten studies with a total of 349 colorectal cancer patients found that individuals with chronic liver conditions—such as hepatitis, cirrhosis, and fatty liver disease—seem to have a lower chance of developing liver metastasis. This fascinating result suggests that liver diseases might play a role in the mechanisms that lead to the formation of liver metastases, a concept often referred to in the literature as the “seed–soil hypothesis.” On the other hand, some studies paint a more concerning picture. For instance, research by Wu et al. in 2018 identified a significant link between pre-existing non-alcoholic fatty liver disease (NAFLD) and worse outcomes for CRC patients. Their findings indicated that the presence of fatty liver—particularly NAFLD—plays a critical role in tumorigenesis and angiogenesis [23]. This effect is mediated by an uptick in proangiogenic inflammatory cytokines, notably interleukin-6 (IL-6) and tumor necrosis factor-alpha, which are known to promote the development and spread of tumors. Another study has demonstrated that increased liver fatness is linked to reduced survival rates. While the distribution of patients with and without liver steatosis was comparable across different BMI categories, those with a higher BMI exhibited shorter survival times [24]. Research indicates that the activation of the WNT-β catenin pathway—an essential pathway in the development of colorectal cancer—may be influenced by elevated leptin levels in obese patients [25, 26]. Our findings suggest that hepatic steatosis may be associated with poorer outcomes in mCRC patients with high BMI. While these results are hypothesis-generating, they suggest a possible prognostic value of metabolic status at diagnosis, which requires further validation in prospective cohorts.

Numerous epidemiological studies and extensive meta-analyses have consistently demonstrated a significant link between being overweight or obese and the rising occurrences of CRC [27]. The rate of mortality of CRC patients also significantly increases with obesity. Daniel et al. 2016 analyzed 634 CRC patients and reported a higher rate of recurrence in obese individuals after the diagnosis of CRC compared to those with normal body weight. Consequently, the mortality rate is almost fivefold higher in obese CRC patients. Moreover, the obese condition is associated with heightened post-operative morbidity and poor outcomes due to slow healing, resistance to anti-angiogenic therapy, and recurrence [28]. However, some studies contradict these results. For example, in a meta-analysis conducted by Dai et al. in 2007, it was shown that colorectal cancer mortality rates in patients with BMI > 30 were similar to those in patients with low BMI [29]. In our study of metastatic colorectal cancer, we examined the impact of visceral and liver fat on survival. Given the absence of a standard threshold for cancer prognosis, we selected the 50th percentile (median) cut-off for VAT to differentiate between high and low visceral adiposity in our cohort. There is no consensus on the optimal cutoff for VAT in the literature. We preferred to use dichotomized values rather than continuous values to quantify better and express the magnitude of the risk. This median-based approach is consistent with prior oncology studies and ensures balanced group sizes for robust statistical analysis [30, 31]. The presence of high visceral fat did not significantly predict survival outcomes across all patients. Furthermore, obesity (BMI ≥ 25) was not a univariate predictor of survival. However, in the subgroup with a BMI ≥ 25, the absence of fatty liver correlated with better survival outcomes (HR = 0.44; CI, 0.21–0.81; p = 0.014).

In this study, the typical localization of the primary tumor was on the left side, with 80.9% of the colon tumors [32, 33], and the majority of the patients had one metastasis site [34], mostly the liver. Similar to the literature, the results showed that patients with mutations in RAS or BRAF had significantly shorter survival times (HR = 2.07, p = 0.001) [35]. These molecular markers are well-recognized indicators of poor prognosis in CRC and suggest more aggressive tumor biology. Moreover, the number of metastatic sites > 1 was associated with poorer survival outcomes (HR = 2.14; p = 0.002), underscoring metastatic burden in determining patient prognosis.

We also found that patients who underwent metastasectomy lived significantly longer than those who did not, and this trend was particularly evident among individuals with a BMI over 25 (HR = 0.44, p = 0.015). The survival advantage associated with metastasectomy aligns with existing clinical experience, especially in cases where metastatic disease is limited and patients respond well to initial systemic therapy. Another crucial aspect to consider is how a patient’s metabolic health can impact both their eligibility for surgery and the eventual success of the procedure. For example, patients with fatty liver or obesity-related inflammation might have special biological conditions that affect their recovery or the long-term benefits of a metastasectomy. This highlights the importance of moving beyond just tumor size and genetic markers when planning future treatments; instead, we might also integrate a patient’s metabolic profile into our strategy.

Additionally, our study shows that the visceral fat ratio is higher in patients with high BMI (p < 0.001) and diabetes (p = 0.045). The study did not find that diabetes status per se was an independent predictor of OS after accounting for adiposity, perhaps because it was colocalized with visceral fat, liver fat, and BMI. Indeed, many of our patients with diabetes likely had fatty liver and visceral obesity as part of metabolic syndrome. Clinicians managing mCRC patients with such comorbidities should be mindful of the potential for worse outcomes and consider optimization of the patient’s metabolic health alongside cancer-specific treatment.

The results of this study highlight the importance of adiposity assessment in mCRC patients. Assessing liver steatosis may offer additional prognostic context in selected subgroups, pending further validation. Given the observed link between liver steatosis and patient survival, we need to investigate whether interventions—like changes in diet or medications—could improve outcomes for certain patient groups by targeting their metabolic health. Achieving ideal weight should be emphasized for these patients without compromising adequate nutrition. Another issue is escalating treatment in these patients (triplet vs doublet) might be an option to be tested in clinical trials. Using body composition data from a patient’s initial CT scan suggests that it could offer valuable additional insights into their prognosis. If validated, such information may contribute to more personalized treatment strategies in mCRC [36]. The addition of lifestyle and metabolic interventions with modifiable risk factors such as fatty liver and visceral obesity along with traditional treatments may maximize the results in cancer [37]. A study recently revealed that an intensive dietary regimen for a year in survivors with colorectal cancer causes weight loss and fat mass increase reduction when compared with standard treatment and indicates targeted nutritional intervention may inhibit obesity-related cancer growth following treatment [38]. Besides, research on obesity-targeted medical interventions like metformin revealed mortality reduction in diabetic CRC patients, and it modulates the metabolic pathways in the development of the cancer [39]. Recent studies include anti-inflammatory or immunomodulatory interventions like IL-23 inhibitors, which would be effective in obese CRC patients and are subject to further studies on the role of IL-23 in cancer development. Overall, these strategies exemplify a trend toward personalized treatment of mCRC that accounts for host metabolic phenotype: an approach where oncology, hepatology, and lifestyle medicine converge.

The study’s limitation is its single-center retrospective design with a small sample size, resulting in inherent bias of all retrospective studies such as confounding bias and limitations of post hoc analyses. All the confounding factors could not be adjusted due to missing data and the small sample size. Our analysis did not include detailed information on second- or third-line regimens, radiotherapy, or treatment-related toxicities. However, we did incorporate data on whether patients underwent metastasectomy, as this can have an important impact on prognosis. Patients with mCRC are mostly treated with oxaliplatin or irinotecan-based doublets in our institution. The sample size was not sufficient for further analysis based on treatment protocols, and the studies of the efficacy of other interventions might be biased as they were confounded by indication. Another limitation of this study is the lack of inter-rater reliability assessment for imaging analysis, particularly for liver fat quantification, which was measured density in the liver parenchyma. However, visceral fat was not measured manually. Instead, it was quantified using software that automatically detects and calculates pixel values within a specific density range in CT images, thus minimizing operator dependence. Despite these limitations, the study’s strengths include the use of objective imaging measures for adiposity, a homogeneous treatment-naïve cohort, and the analysis of both host and tumor factors together.

Exploring the connection between adiposity and cancer progression could provide new insights into therapeutic interventions. Understanding the role of metabolic and inflammatory pathways may allow the design of better treatment strategies for mCRC patients. Our study suggests a potential prognostic relevance of fatty liver in high-BMI mCRC patients, while the role of visceral fat appears less clear and requires further investigation. These results suggest the need for further research on integrating metabolic and oncologic parameters in the treatment of mCRC, with personalized approaches.

Abbreviations

BMI

Body mass index

CAD

Coronary artery disease

CRC

Colorectal cancer

CT

Computed tomography

CI

Confidence interval

dMMR

Deficient mismatch repair

HU

Hounsfield units

IQR

Inter quartile range

mCRC

Metastatic CRC

NAFLD

Non-alcoholic fatty liver disease

pMMR

Proficient mismatch repair

ROI

Region of interest

VOI

Volume of interest

Author contribution

Conceptualization, C.E.C., F.O., and O.D.; methodology, C.E.C.; software, C.E.C., E.C., and V.G.; validation, E.C. and O.D.; formal analysis, C.E.C., E.C.; investigation, C.E.C., E.C.; resources, C.E.C.; data curation, C.E.C., E.C.; writing—original draft preparation, C.E.C.; writing—review and editing, E.C., O.D., F.O., and M.R.O.; visualization, C.E.C., E.C., and V.G.; revision, O.D., M.R.O., F.O., T.K.S., and C.E.C; supervision, O.D. and F.O.; project administration, F.O. and O.D.. All authors have read and agreed to the published version of the manuscript.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

This study was approved by the Institutional Review Board at Hacettepe University (GO 23/558) and was exempt from informed consent due to its retrospective nature.

Institutional review board statement

All procedures performed in studies involving human participants were by the ethical standards of the institutional and national research committee, the 1964 Helsinki Declaration and its later amendments, or comparable ethical standards. The ethics committee approved the study of Hacettepe University under the reference number GO 23/558.

Informed consent

Not applicable.

Competing interests

The authors declare no competing interests.