Introduction

Incidence of CRC: Jordan vs. worldwide

Colorectal cancer (CRC) ranks as the third most common type of cancer worldwide, accounting for 10.0% of cases according to the latest GLOBCAN statistics1. Mortality rates are equal between male and female patients, with both genders accounting for 9.3% and 9.4% of recorded deaths, respectively.

According to the Global Burden of Disease (GBD), the incidence of CRC is higher in developed countries. However, it is steadily increasing in low- and middle-income countries, posing growing financial and health challenges in maintaining and optimizing quality health services2. The latest report from the Jordan Cancer Registry of 2019 revealed that CRC is the second most prevalent type of cancer in Jordan, accounting for 11.6% of cancer diagnoses, surpassed only by breast cancer (20.3%). Interestingly, CRC appears to affect males more than females, with 13.7% of male cancer cases being CRC compared to 9.7% for females.3. In terms of mortality, CRC stands as the second leading cause of cancer-related death in males at 11%, trailing behind lung cancer. Among females, CRC ranks third in terms of mortality at 10.4%, following breast cancer and leukemia2.

Risk factors and BMI

Several risk factors have been associated with a higher likelihood of developing CRC, including obesity, physical inactivity, smoking, unhealthy lifestyle, and genetic factors among others4,5,6,7.

Body mass index (BMI) is a commonly used measure of body fat calculated based on a person's weight and height (kg/m2)8. Although BMI is widely used to measure obesity9, it is not an accurate indicator of body fat as it does not account for fat distribution or the weight of bones and muscles10. Two types of fat have been linked with obesity; visceral fat and subcutaneous fat, both of which cannot be measured using BMI. Studies have shown that visceral fat, which is the metabolically active form of fat, contributes to the secretion of proinflammatory cytokines and adipokines (tumor necrosis factor and interleukin-6) which induce the high risk of CRC carcinogenesis11,12. On the other hand, subcutaneous fat is linked to favourable outcome in CRC patients13 and can be used as positive metabolic profile for glucose and lipid levels.

Unfortunately, BMI cannot distinguish between increased visceral fat, subcutaneous fat, or muscle mass14, making it an inadequate tool for assessing cancer risk. While BMI is commonly used in medical practice to assess patient obesity and overall risk, its limitations in accurately assessing body composition make it an imperfect tool for predicting cancer risk.

CRC and muscle mass

Several studies have demonstrated the importance of muscle mass in predicting survival rates and outcomes for CRC patients15,16,17. A recent comprehensive review15 showed that the frequency of sarcopenia (low muscle mass) in CRC patients ranges between 12 and 60%.

Factors associated with sarcopenia can be either patient-related, such as physical inactivity, malnutrition, and body composition, or cancer-related, including weight loss and muscle mass deterioration resulting from treatment18. Sarcopenia has also been used as a biomarker to predict chemotherapy tolerance and toxicity in CRC patients18 and several studies have used it to predict surgical complications, reduced survival, and poor quality of life in CRC patients19,20.

Patients with low muscle mass are best identified by computed tomography (CT), as it is considered the gold standard method to measure the mass and quality of muscles in addition to other body composition factors21.

In addition to low muscle mass, elevated levels of fat distribution, particularly abdominal visceral fat, represent significant risk factors in CRC. This is primarily attributed to visceral fat's capability to promote the abdominal tumorigenic environment, thereby increasing the likelihood risk of CRC development through various mechanisms, including enhanced cancer cell proliferation, angiogenesis, and the induction of a protumorigenic microenvironment12. Furthermore, visceral fat contributes to systemic chronic inflammation by releasing proinflammatory cytokines and tumour necrosis factor-alpha22.

Aim

The objective of this study was to explore how Visceral Adipose Tissue (VAT), Subcutaneous Adipose Tissue (SAT), and Muscle Mass (MM) correlate with overall survival (OS) in colorectal cancer (CRC) patients treated at King Hussein Cancer Center (KHCC), a leading comprehensive cancer center in Jordan. Utilizing CT scans, we measured muscle mass, visceral fat, and subcutaneous fat, and investigated its relationship with demographic variables (such as age and gender) and clinical indicators (including cancer stage, grade, and primary treatment) to patient survival.

Materials and methods

We screened a total of 2280 patients with CRC who visited KHCC between 2007 and 2018. Inclusion criteria required patients to have a confirmed diagnosis of CRC regardless of the stage, and an available CT scan at the lower edge of L3 vertebral level before treatment initiation. Patients who had CT scans after treatment interventions, types of cancer other than CRC, CT scans at different levels than L3, or low-quality CT scans were excluded from the study. Patients with missing BMI measures at diagnosis were also excluded. The slice thickness of the CT scan used for eligible patients was 3 mm, which is sufficient to provide higher resolution images with more details.

Image-J software (version 1.52a), which is a Java-based image processing program, developed by the U.S. National Institutes of Health (NIH) and available for free in the public domain (https://imagej.nih.gov/ij/), was used to measure the visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), and muscle mass (MM) in square centimeters (cm2) (Fig. 1). The instructions on how to use ImageJ were followed as mentioned in the instruction manual23. Three physicians were trained to measure VAT, SAT, and MM from the CT scans using the same ImageJ version and measuring technique. The threshold values applied to measure the CT scans were determined through visual inspection and experimentation with different thresholds until optimal separation of VAT, SAT, and MM was achieved.

Figure 1
figure 1

CT scans for CRC patients at the lower edge of L3 level, at different thresholds using ImageJ. (A) CT scan of CRC patient at lower edge of L3 level. (B) The area in red is the fat (visceral and subcutaneous) after setting the threshold between (10–90). The visceral fat was selected and measured at the indicated threshold. (C) Visceral fat was cut out, leaving subcutaneous fat alone to be measured at the same threshold (10–90). (D) The area in red is the muscle area after setting the threshold between (90–170). The area was selected and measured at the indicated threshold.

Full size image

Demographic, pathological, and clinical parameters were collected for the patients. Additionally, height and weight at diagnosis were collected to calculate BMI. To evaluate the agreement and reproducibility among the measurements obtained by the three physicians, 78 CT scans were independently measured three times by different physicians. Subsequently, the intraclass correlation coefficient (ICC) was calculated to evaluate the consistency among the measurements obtained by the three physicians.

Receiver Operating Characteristic (ROC) curves were generated to determine cutoff points for VAT, SAT, and MM based on gender. Survival rates were estimated using the Kaplan–Meier method and compared between groups using the Log-rank test. Multivariate Cox regression was used to assess prognostic factors. A significance criterion of P ≤ 0.05 was used in the analysis, and all analyses were performed using SAS version 9.4 (SAS Institute Inc, Cary, NC).

The data collected in this study adhered to HIPAA-compliant standards and was ethically approved by the Institutional Review Board (IRB) at King Hussein Cancer Center (study number: 17 KHCC 99). Informed consent was waived due to retrospective nature of study design.

Results

Agreement and reproducibility

The results of the two-way mixed effects, absolute agreement ICC for VAT, SAT, and MM were 0.88, 0.76, and 0.75, respectively, with a 95% confidence interval (CI) (Table 1). ICC values exceeding 0.70 indicate good reliability24. These findings suggest that the measurements obtained by the three physicians were consistent and reproducible.

Table 1 Intra-class Correlation Coefficient (ICC): two-way mixed effects for absolute agreement.
Full size table

Descriptive analysis

Of 2280 screened CRC patients, 408 patients diagnosed at KHCC between 2007 and 2018 met our inclusion criteria and were included in the analysis.

The mean age at diagnosis was 56.9 ± 13.2 (± SD) years, median follow-up was 93.3 months (range: 40.9–182), and 233 (57.1%) were male patients. Of these patients, 216 (52.9%) had colon cancer and 192 (47.1%) had rectum or rectosigmoid cancer. A total of 350 (85.8%) patients’ cancers were grade 1 or 2. Of the measured CT scans, 210 (51.5%) patients had high VAT, 188 (46.1%) had high SAT, and 228 (55.9%) had high MM. Table 2 shows the descriptive analysis of the demographic of CRC in the sample cohort.

Table 2 Descriptive analysis for demographic and clinical outcomes.
Full size table

Table 3 summarizes the cut-off points for the area under the curve (AUC) for MM, VAT and SAT, stratified according to gender. The ROC curves are for MM, VAT and SAT for male and females are represented in the supplementary materials (Figs. S1–S6).

Table 3 AUC cut off points for MM, VAT and SAT, stratified according to gender:
Full size table

Survival analysis

At the follow-up time in March 2022, 145 patients (35.4%) of the patients had died. The median survival of the entire cohort was 96.3 months (95% CI 63.1, 72.2) and the 5-year OS rate was 67.8% (Fig. 2).

Figure 2
figure 2

Kaplan–Meier plot demonstrating 5-year OS probability for CRC patients. The 5-year OS rate for CRC is 67.8% (95% CI 63.1–72.2). Of the entire cohort, 263 CRC patients were alive at follow-up time (64.5%).

Full size image

Patients who had surgery had a higher survival rate compared with those who did not have surgery (80.7% vs 33.6% P < 0.0001). Tumors of grades 1 and 2 were compared with tumors of grades 3 and 4 and showed a significantly better survival rate (71.0% vs. 46.7% P = 0.0015). Additionally, significant differences in survival rates between age groups and TNM stages were observed (P = 0.0014 and P < 0.0001 respectively) (Table 4).

Table 4 Univariate analysis of 5-year OS in CRC patients.
Full size table

High MM was associated with a better 5-year overall survival rate in CRC patients (71.8% vs. 62.7%, P = 0.0224) (Fig. 3). In contrast, SAT and VAT were not associated with OS when comparing high to low values (67.9% vs 67.6% P = 0.74), (67.9% vs 67.6%, P = 0.6351), respectively. Moreover, there was no association between BMI and cancer site groups and OS (P = 0.6085, P = 0.5062, and P = 0.5793 respectively) (Table 4).

Figure 3
figure 3

Kaplan–Meier plot demonstrating OS probability for CRC patients stratified by MM, high vs. low. The 5-year OS rate for CRC patients with high muscle mass was 71.8% (95% CI 65.8–77.5) and for patients with low muscle mass 62.7% (95% CI 55.5–69.6), p-value = 0.02.

Full size image

Using a multivariate Cox regression adjusting for age, stage, grade, surgery, VAT, SAT, and MM, no significant association was observed for VAT, SAT, or grade with a five-year OS rate (P = 0.474, 0.863, 0.101 respectively). However, there was an association between age, stage, surgery, and MM with a five-year OS rate (P = 0.016, < 0.0001, 0.0001, 0.040 respectively) (Table 5).

Table 5 Multivariate cox regression analysis for CRC patients.
Full size table

Discussion

In this study, we investigated the relationship between Visceral Adipose Tissue (VAT), Subcutaneous Adipose Tissue (SAT), and Muscle Mass (MM) measured using CT scans with OS in CRC patients treated at KHCC in Jordan. Our findings showed a significant association between high muscle mass and improved OS rates in CRC patients, aligning with previous research emphasizing the positive impact of increased MM on survival outcomes in cancer patients25,26,27. This analysis revealed that OS in CRC patients was influenced by multiple predictive factors, including surgical intervention, tumour grade, TNM stage, and age at diagnosis, all of which demonstrated significant associations with survival outcomes28,29.

The 5-years OS rate of 67.8% (95% CI 63.1–72.2) of our population was similar to other studies conducted in Brazil 63.5%30 and a population-based study in 9 European countries 71·1% (95% CI 70·7–71·4)31 which can be attributed to similarities in the high level of quality of care provided to CRC patients. A previously published study utilizing data from the Jordan’s cancer registry during the period of 2005–2010 reported a 5-years OS of 58.2% for CRC patients32. The notable increase in survival rates between their study and our findings suggests positive progress in the level of care and advancements in treatment modalities over time.

To accurately measure MM, VAT, and SAT, our study employed a CT scan-based method33 using ImageJ software. Three independent physicians measured the images to ensure data accuracy and reliability.

A study conducted in biliary duct cancer patients showed that high MM was associated with a high survival rate (HR 0.46, 95% CI 0.22–0.95, P = 0.037)25 which is similar to our results (HR 0.997, 95% CI 0.994–1, P = 0.0397). Another meta-analysis27 on rectal cancer patients with sarcopenia showed a poor survival rate (HR 2.10, 95% CI 1.33–3.32, P = 0.001) compared to patients with high MM, which is in line with our results that high muscle mass provides survival benefit for patients with CRC.

Moreover, a higher level of muscle mass is usually linked to a favourable inflammatory profile due to the release of myokines. These myokines play a key role in the anti-inflammatory effects of physical activity, helping to counteract metabolic disturbances and the effects of adipokines34,35. In addition to the pivotal role of physical exercise on increasing the muscle mass, it positively influences metabolic pathways and energy homeostasis by enhancing aerobic capacity, insulin sensitivity, glucose control, and oxidative capacity, among other adaptations35. All these factors contribute to the survival of CRC and MM relationship, which is usually altered during cancer35.Based on our knowledge and existing literature highlighting the adverse effects of VAT in the abdominal region, we explored the association between VAT and OS in CRC patients. Interestingly, our findings indicated no significant impact of high VAT (HR 1.001, 95% CI 1.003–0.4738, P = 0.4738) on OS, similar to results from a study on Korean CRC patients (HR 0.656; 95% CI 0.402–1.071; P = 0.092)36. However, this is contrary to other studies reporting a significant association between increased VAT (HR 2.61, 95% CI 1.155–5.924, P = 0.020) and poor survival outcomes37,38.

We also investigated the association between SAT and OS in CRC patients, and our findings did not indicate a significance on OS (HR 1.000, 95% CI 0.999–1.001, P = 0.8633). Our results are similar to the above-mentioned study37, (HR 1.18, 95% CI 0.614–2.036, P = 0.715) but contrary to the study on Korean CRC patients among other studies (HR 0.505, 95% CI 0.266–0.957; P = 0.036)36,38.

This discrepancy may be attributable to the relatively small sample size and the heterogeneity of the CRC patient population in our study. Genetic factors specific to the Middle Eastern population may also contribute to this variation5,6. These include IL-17 polymorphisms which play a crucial role in inflammation, and autoimmune diseases and are responsible for CRC growth and invasion5,39. Moreover, CRC driver mutations like RAS and BRAF mutations vary in their prevalence in Middle Eastern compared to Western countries7,40. However, this data was not available for the population of patients we studied and further research is needed to study these variable and their interplay with body composition and effect on CRC outcome.

Additionally, we rigorously controlled for potential confounding factors such as age at diagnosis, gender, tumour grade, and TNM stage, enhancing the validity and reliability of our findings. Our findings in the multivariate analysis revealed that only the age at diagnosis, TNM stage group, surgery, and MM were significant independent predictors of the patient’s survival. Our results are similar to findings in other Middle Eastern populations41,42 and analyses based on the SEER database43. Tumour grade did not emerge as a significant predictor of CRC survival in multivariate analysis, which is contrary to other published research in Western populations like the US and Canada using their National Cancer Database44 and Eastern populations such as Korean CRC patients45.

It is important to18 acknowledge the limitations of our study, primarily due to its retrospective design which resulted in missing data that could contribute important insights to the question of body composition and its association with CRC outcome. This includes the absence of lifestyle data encompassing physical activity and diet, both of which have the potential to impact body composition, fat, and muscle distribution, and could be associated with OS in CRC patients46,47, in addition to their role in cancer prevention48. Unfortunately, these data were not present in our electronic medical records at the time of diagnosis nor during the patient’s treatment course.

While the integration of artificial intelligence (AI) is rapidly growing in the radiology field -making CT scan analysis of body composition faster, more accurate, and potentially more informative- the need for human oversight remains important to make medical judgement49. Our study would have benefited from using AI to quantify the measurements of SAT, VAT and MM by improving the consistency and reducing human error in measurements. However, this would require training of the AI model by introducing massive amount of medical data using advanced computing power, which is not readily available in our healthcare setting.

Future research with larger sample sizes and more diverse patient cohorts is needed to better understand the association between body composition and CR outcome. Such studies would help validate our findings and elucidate the underlying mechanisms involved, thereby informing clinical practice to improve CRC outcomes and enhance the quality of life for patients. Additionally, investigations exploring genetic factors and genetic variations across different populations may help to answer questions specific to the Middle Eastern population and provide further insights into the relationship between body composition and survival outcomes in CRC patients.

Conclusion

This study provides evidence supporting the association between high muscle mass and improved OS rates in CRC patients. Additionally, TNM stage, surgical intervention, and age at diagnosis were identified as significant independent predictors of OS. Notably, VAT and SAT did not demonstrate a significant association with OS. These findings highlight the importance of assessing body composition, especially muscle mass as a valuable prognostic indicator in CRC patients.

Furthermore, integrating routine physical activity and promoting healthy lifestyle habits in the management of CRC patients hold potential benefits. Future studies can explore the impact of exercise and nutritional interventions on body composition and OS rates, contributing to the development of comprehensive treatment strategies.

In addition, nutritional counselling can be a crucial aspect of CRC patient care. Exploring the effects of tailored nutritional interventions, including increased protein intake, on body composition and muscle mass, OS rates, and quality of life, can provide valuable insights into optimizing patient outcomes.

By addressing these areas, we can emphasize the significance of physical activity and healthy lifestyle choices, and incorporate nutritional counselling into the comprehensive care of CRC patients. These factors can have the potential to advance the field and contribute to improved survival rates, enhanced quality of life, and better overall outcomes for individuals with CRC.