Evaluation of the Clinical Relevance of Body Composition Parameters in Patients With Cancer Metastatic to the Liver Treated With Hepatic Arterial Infusion Chemotherapy

Abstract

The association between body composition parameters and toxicity associated with hepatic arterial infusion (HAI) chemotherapy regimens has not been analyzed. We analyzed data from patients with advanced cancer and liver metastases treated on a clinical trial of HAI oxaliplatin combination regimen. Patient characteristics, response and toxicities were analyzed in relevance with body composition data from CT images. Forty-eight of 57 patients (mean age 57 years; 60% women) had available CT scans. The most common diagnosis was colorectal cancer (22/48, 46%); 30/48 patients (63%) had body mass index (BMI) ≥ 25 kg/m2. Twenty of 48 patients were sarcopenic (42%). Grade 3–4 adverse events did not differ among patients with and without sarcopeniaor according to BMI. The median survival (95% confidence interval [CI]) was 167 (128–206) days for sarcopenic and 280 (214–346) days for non-sarcopenic patients (p=0.271). Among patients treated at the maximum tolerated dose, the median survival was 103 days for sarcopenic and 312 days for non-sarcopenic patients (p=0.173). Sarcopenia was present in 30% (6/20) of patients with reduction in tumor size post-treatment, and in 52% (14/27) of patients with increased tumor size (p=0.171). In conclusion, body composition was not associated with toxicities. Sarcopenia might be associated with shorter survival.

The role of computed tomography in evaluating body composition and the influence of reduced muscle mass on clinical outcome in abdominal malignancy: a systematic review

“…All studies recruited the sample in an appropriate manner, with the use of a cancer centre and an electronic database for identification of eligible patients and data collection. However, five studies recruited relatively small sample sizes 40,42,46,48,49 in comparison with the other studies, although one of these gathered data for 7 years, which was the longest recruitment period. 40 The largest sample size was 1474 and included all newly referred patients.…”

“…[45][46][47][48][49] The included studies involved 2584 participants, and all further characteristics are detailed in Table 1. Of the 10 trials, 5 were conducted in Canada, 40,41,45-47 1 in the United States, 42 1 in the United Kingdom, 43 1 in France, 49 1 in The Netherlands 44 and 1 split across Norway and Canada. 48 All trials were based at a single oncology centre, except for one that included patients from two countries 48 and one that conducted a multicentre study over nine sites.…”

“…Six studies followed up participants from diagnosis until death, and the primary outcome was overall survival time in relation to CT scan analysis. 40,42,44,45,47,48 The primary outcomes of the four remaining studies included the following: occurrence of postoperative infections in relation to CT scan; 41 the capacity of CT to provide specific detail on different tissues; 46 level of inflammatory markers, including C-reactive protein and albumin; 43 and level of treatment toxicity. 49 In all studies, secondary outcomes included recognition of malnutrition by the identification of reduced muscle mass using CT scans and identification of those underweight using BMI.…”

“…46 This was owing to the significant relationship of a single CT image to whole-body fat-free mass (r = 0.94, P o 0.001). 46 Eight studies recorded clinical outcome in relation to body composition assessed by CT. 40,[42][43][44][45][46][47][48] One study recorded postoperative infection, rehabilitation and length of stay and demonstrated that an increase in each of these outcomes was significantly associated with the identification of sarcopenia. 41 One study demonstrated that an elevated systemic inflammatory response was significantly associated with a low muscle index (P = 0.01).…”

“…40 The remaining eight studies were vague about training and blinding of analysts, and five of the studies failed to specify whether any intra-or inter-observer reliability tests had been carried out. 41,42,44,48,49 Analysis of CT scans The Slice-O-Matic software was used for the analysis of CT scans in all studies, except for the studies by Richards et al 43 and Vledder et al 44 One study used a governmental free-ware programme Records identified as reviews (n = 5).…”