Role of Machine Learning-Based CT Body Composition in Risk Prediction and Prognostication: Current State and Future Directions

CT body composition analysis has been shown to play an important role in predicting health and has the potential to improve patient outcomes if implemented clinically. Recent advances in artificial intelligence and machine learning have led to high speed and accuracy for extracting body composition metrics from CT scans. These may inform preoperative interventions and guide treatment planning. This review aims to discuss the clinical applications of CT body composition in clinical practice, as it moves towards widespread clinical implementation.

Relation between skeletal muscle volume and prognosis in rectal cancer patients undergoing neoadjuvant therapy

The prognostic role of body composition indexes, and specifically sarcopenia, has recently been explored in different cancer types. However, conflicting results have been reported. Heterogeneity in cancer type, cancer stage or oncological treatments, as well as different methodology and definition of sarcopenia, could be accounted for different conclusions retrieved from literature. When focusing on colorectal cancer, it clearly appears that colon and rectal cancers are often treated as a single entity though they have different behaviors and treatments. Particularly, patients with advanced rectal cancer represent a peculiar group of patients that according to current guidelines are treated with neoadjuvant chemotherapy and radiotherapy followed by radical surgery. This review was restricted to a homogeneous group of patients with advanced lower rectal cancer and the aim of exploring whether there is a correlation between skeletal muscle depletion and prognosis. Literature was searched for articles related to patients with advanced rectal cancer undergoing neoadjuvant chemo-radiotherapy (NCRT) followed by radical surgery, in whom muscle mass and/or change in muscle mass during neoadjuvant treatment were measured. Eight full-text articles were selected and included in the present review. The main findings of our review were: (1) The majority of the studies defined sarcopenia as muscle mass alone over muscle strength or physical performance; (2) There was a great deal of heterogeneity in the definition and measures of sarcopenia, in the definition of cut-off values, and in the method to measure change in muscle mass; (3) There was not full agreement on the association between sarcopenia at baseline and/or after chemo-radiotherapy and prognosis, and only few studies found a significance in the multivariate analysis; and (4) It seems that a loss in skeletal muscle mass during NCRT is associated with the worst outcomes in terms of disease-free survival. In conclusion, analysis of muscle mass might provide prognostic information on patients with rectal cancer, however more robust evidence is needed to define the role of muscle depletion and/or muscle change during neoadjuvant treatments, related to this specific group of patients. If a prognostic role would be confirmed by future studies, the role of preoperative intervention aimed at modifying muscle mass could be explored in order to improve outcomes.

Older cancer patients receiving radiotherapy: a systematic review for the role of sarcopenia in treatment outcomes

BACKGROUND

Previous studies have evaluated the prognostic effects of sarcopenia in cancer patients receiving various treatments, including chemotherapy and surgery, but few studies have focused on radiotherapy (RT).

AIMS

We aimed to investigate the prevalence of sarcopenia and the relationship between sarcopenia and outcomes in older cancer patients who underwent RT without chemotherapy.

METHODS

A systematic review of the literature was conducted in Pubmed/Medline and Cochrane databases in September 2021. We used the search terms and medical subject heading terms "sarcopenia," "low muscle mass (LMM)," "low muscle strength," "LMM and low muscle strength," "LMM and low muscle strength and low physical performance," and "RT." Outcomes were overall survival (OS), progression-free survival, non-cancer death, cancer death, disease-specific survival, local failure-free survival, distant failure-free survival, and RT-related toxicities.

RESULTS

Among 460 studies, 8 studies were eligible for inclusion. The prevalence of sarcopenia was between 42.8% and 72%. Sarcopenia was not associated with OS or OS at 3 years in seven studies in which it was defined as the presence of LMM, while it was related in one study, in which it was defined as the concomitant presence of LMM and muscle strength/function.

DISCUSSION

There was heterogeneity between the studies because there was diversity in their inclusion criteria, definition and assessment methods used for detection of sarcopenia, considered cutoffs for low muscle mass and strength, cross-sectional locations on imaging to assess muscle mass and included covariates. The discrepancy in the results of the studies may also result from the variations in diagnoses, sample sizes, and treatment modalities. The low number of included studies and a small number of patients in each study limited generalizability.

CONCLUSIONS

Sarcopenia may be a prognostic factor, especially in OS when low muscle strength/function is integrated into its definition. We suggest that clinicians focus on muscle strength/function while considering sarcopenia and its association with cancer and RT-related outcomes.

Clinical impact of sarcopenia assessment in patients with hepatocellular carcinoma undergoing treatments

Changes in body composition are associated with poor outcomes in cancer patients including hepatocellular carcinoma (HCC). Sarcopenia, defined as the loss of skeletal muscle mass, quality and function, has been associated with a higher rate of complications and recurrences in patients with cirrhosis and HCC. The assessment of patient general status before HCC treatment, including the presence of sarcopenia, is a key-point for achieving therapy tolerability and to avoid short- and long-term complications leading to poor patients' survival. Thus, we aimed to review the current literature evaluating the role of sarcopenia assessment related to HCC treatments and to critically provide the clinicians with the most recent and valuable evidence. As a result, sarcopenia can be predictive of poor outcomes in patients undergoing liver resection, transplantation and systemic therapies, offering the chance to clinicians to improve the muscular status of these patients, especially those with high-grade sarcopenia at high risk of mortality. Further studies are needed to clarify the predictive value of sarcopenia in other HCC treatment settings and to evaluate its role as an additional staging tool for identifying the most appropriate treatment. Besides, interventional studies aiming at increasing the skeletal muscle mass for reducing complications and increasing the survival in patients with HCC are needed.

Association between specific presurgical anthropometric indexes and morbidity in patients undergoing rectal cancer resection

OBJECTIVES

Baseline body composition has been associated with dismal outcomes in patients undergoing a variety of major abdominal operations. Whether specific anthropometric indexes can predict morbidity after rectal resection has been poorly investigated. The aims of this study were to assess whether there is a relationship between body mass index and the different computed tomography-assessed body composition indexes, and whether the analysis of different body compartments could be predictive of short-term outcomes in patients undergoing curative surgery for rectal cancer.

METHODS

Computed tomography-derived measures of skeletal muscle and adipose tissue areas of patients undergoing surgery for rectal cancer between January 2009 and December 2016 were used to calculate population-specific thresholds of sarcopenia, subcutaneous adiposity, visceral adiposity, visceral obesity, sarcopenic obesity, and myosteatosis. Association between the aforementioned body composition features were related with overall complication, infection, and anastomotic leak.

RESULTS

During the study period, 311 patients received surgery and 173 were eligible for an accessible preoperative computed tomography imaging. After surgery, 59 (34.1%) patients experienced a complication, 29 an infection, and 10 an anastomotic failure. The overall morbidity rate was observed more frequently in patients with sarcopenia than in those without sarcopenia (39% versus 17.5%; P = 0.002) and infections (41.4% versus 21.5% respectively; P = 0.024). The presence of myosteatosis also was associated with a higher incidence of overall morbidity (33.9% versus 20.2% in patients without myoteatosis; P = 0.048). Anastomotic failure occurred in 6 of 10 patients with visceral obesity and in 24 of 112 (21.4%) patients without this condition (P = 0.007).

CONCLUSIONS

Some anthropometric indexes are accurate predictors of specific types of morbidity. These findings may allow a more accurate preoperative risk stratification.

Effects of Sarcopenia on Prognosis After Resection of Gallbladder Cancer

BACKGROUND

This study aimed to determine the prognostic significance of radiographic sarcopenia (RS) in patients with gallbladder cancer (GBC).

METHODS

From March 2001 to December 2013, 158 GBC patients who underwent curative intent surgery were included. The presence of RS was determined by skeletal muscle mass index using abdominal computed tomography.

RESULTS

The 1-, 3-, and 5-year overall survival (OS) rates were 63.6%, 41.9%, and 36.4%, respectively, for patients with RS (n = 88), and 84.3%, 62.6%, and 54.3%, respectively, for those without RS (n = 70) (P = 0.006). Multivariate analysis showed that RS (hazard rate [HR] 1.704, P = 0.024) was a significant prognostic factor for patient survival, as well as disease stage (IV: HR 7.181, P < 0.001), radicality (HR 2.830, P = 0.001), adjuvant therapy (HR 0.537, P = 0.017), and intraoperative blood loss ≥ 1 L (HR 1.851, P = 0.023).

CONCLUSIONS

This study showed a significant association between RS and OS in GBC patients. Because RS is the only significant prognostic factor that can be evaluated preoperatively, its assessment would be helpful to provide early preventive therapy allowing the maintenance of muscle mass and patient-tailored treatment based on their physiologic reserves (e.g., skeletal muscle mass).

The Obesity Paradox in Cancer: How Important Is Muscle?

Although higher body mass index (BMI) increases the incidence of many cancers, BMI can also exhibit a null or U-shaped relationship with survival among patients with existing disease; this association of higher BMI with improved survival is termed the obesity paradox. This review discusses possible explanations for the obesity paradox, the prevalence and consequences of low muscle mass in cancer patients, and future research directions. It is unlikely that methodological biases, such as reverse causality or confounding, fully explain the obesity paradox. Rather, up to a point, higher BMI may truly be associated with longer survival in cancer patients. This is due, in part, to the limitations of BMI, which scales weight to height without delineating adipose tissue distribution or distinguishing between adipose and muscle tissue. Thus, cancer patients with higher BMIs often have higher levels of protective muscle. We assert that more precise measures of body composition are required to clarify the relationship of body size to cancer outcomes, inform clinical decision-making, and help tailor lifestyle interventions.

Optimal perioperative care in peri-hilar cholangiocarcinoma resection

Surgical resection remains the only proven curative treatment for peri-hilar cholangiocarcinoma. Despite recent advances in liver surgery techniques and perioperative care, resection for peri-hilar cholangiocarcinoma remains associated with significant morbidity and mortality. Considerable variation in the perioperative management of these patients exists. Optimal perioperative management has the potential to deliver improved outcomes. This article seeks to summarize the evidence underpinning best practice in the perioperative care of patients undergoing resection of peri-hilar cholangiocarcinoma. The authors also seek to identify areas where research efforts and future clinical trials should be targeted.

Excess adiposity and survival in patients with colorectal cancer: a systematic review

Excess adiposity is an established risk factor for incident colorectal cancer (CRC) but whether this association extrapolates to poorer survival is unclear. We undertook a systematic review to examine relationships between measures of adiposity and survival in patients with CRC. For distinction, we included pre-diagnosis exposure and CRC-related mortality. We performed dose-response meta-analyses and assessed study quality using eight domains of bias. Six study categories were identified based on (i) timing of adiposity measurement relative to survival analysis time zero and (ii) clinical setting. Several types of adiposity measurements were reported; body mass index (BMI) was the commonest. For pre-diagnosis cohorts, baseline BMI negatively impacted on CRC-related mortality in men only (risk estimate per 5 kg m(-2)  = 1.19, 95% confidence intervals: 1.14-1.25). The other groups were pre-diagnosis BMI but diagnosis as time zero; population-based cohorts; treatment cohorts; observational analyses within adjuvant chemotherapy trials; patients with metastatic CRC - each had several biases (e.g. treatment selection, reverse causality) and sources of confounding (e.g. chemotherapy 'capping'). Overall, there was insufficient evidence for a strong link between adiposity and survival. These findings demonstrate an important principle: an established link between an exposure (here, adiposity) and increased cancer incidence does not necessarily extrapolate into an inferior post-treatment outcome.