Insulin/IGF-1 Signaling Is Downregulated in Barrett's Esophagus Patients Undergoing a Moderate Calorie and Protein Restriction Program: A Randomized 2-Year Trial

Abstract

Obesity and associated insulin resistance (Ins-R) have been identified as important risk factors for esophageal adenocarcinoma development. Elevated calories and protein consumption are also associated with Ins-R and glucose intolerance. We investigated the effect of a 24-month moderate calorie and protein restriction program on overweight or obese patients affected by Barrett's esophagus (BE), as no similar dietary approach has been attempted to date in this disease context. Anthropometric parameters, levels of serum analytes related to obesity and Ins-R, and the esophageal insulin/IGF-1 signaling pathway were analyzed. This study is registered with ClinicalTrials.gov, number NCT03813381. Insulin, C-peptide, IGF-1, IGF-binding protein 3 (IGFBP3), adipokines, and esophageal expression of the main proteins involved in insulin/IGF-1 signal transduction were quantified using Luminex-XMAP® technology in 46 patients who followed the restriction program (IA) and in 54 controls (CA). Body mass index and waist circumference significantly decreased in 76.1% of IA and 35.2% of CA. IGF-1 levels were reduced in 71.7% of IA and 51.8% of CA. The simultaneous reduction of glycaemia, IGF-1, the IGF-1/IGFBP3 ratio, and the improvement in weight loss-dependent insulin sensitivity, were associated with the downregulation of the insulin/IGF-1 signal on BE tissue. The proposed intervention program was an effective approach to counteract obesity-associated cancer risk factors. The improvement in metabolic condition resulted in a downregulation of the ERK-mediated mitogenic signal in 43.5% of patients, probably affecting the molecular mechanism driving adenocarcinoma development in BE lesions.

Keywords: Barrett’s esophagus; calorie–protein restriction; esophageal adenocarcinoma prevention; insulin/IGF-1 signal transduction; obesity.

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure 1

The trial’s CONSORT flow diagram.

Figure 2

Waist circumference variation (ΔWC) and BMI (ΔBMI) in the control arm (CA) and intervention arm (IA) at the end of the 24 months of the intervention program. (a) ΔWC and (b) ΔBMI were calculated as the difference between the value measured at T24 and the baseline value. The dashed red lines (ΔWC = −4 cm and ΔBMI = −1.11 kg/m2) show the cutoffs between the two arms, which were mathematically corresponding to the 25th percentile of the whole data distribution. Each square represents a single patient from the CA and each triangle represents a single patient from the IA.

Figure 3

Total protein expression involved in insulin/IGF-1 signal transduction in esophageal tissue affected by Barrett’s esophagus. Protein expression was analyzed at baseline (T0, white boxes) and 24 months later (T24, black boxes) in both the control arm (CA) and intervention arm (IA). Data are expressed as the median (Q1; Q3) related to an arbitrary unit (A.U.), defined as described in the Materials and Methods Section. The Wilcoxon matched-pairs signed-rank test was used to compare measurements at T0 and T24 within the same group. The Mann–Whitney U test was performed to evaluate differences between the CA and IA at baseline. ** p < 0.01.

Figure 4

Main protein phosphorylation levels in insulin/IGF-1 signal transduction in esophageal tissue affected by Barrett’s esophagus. Protein expression was analyzed at baseline (T0) and 24 months later (T24) in both the control arm (CA) and the intervention arm (IA). Data are expressed as the median (Q1; Q3) of relative phosphorylation levels related to an arbitrary unit (A.U.) defined as described in the Materials and Methods Section. Each symbol represents a different data set: ● CA at T0, ∎ CA at T24, ▲ IA at T0, ▼IA at T24. The Wilcoxon matched-pairs signed-rank test was used to compare measurements at T0 and T24 within the same group. The Mann–Whitney U test was performed to evaluate differences in baseline measurements between the CA and IA. * p < 0.05.

Figure 5

Identification of different patient subpopulations based on the modulation of IRS1 expression after 24 months from the start of the intervention program. Patients were divided on the basis of the downregulation (lower) or upregulation (higher) of IRS1 expression at the end of the intervention program compared to the basal expression at the time of enrollment. Within the lower IRS1 subpopulation, patients who had improved glycaemic control and patients who had unchanged or worsened glycaemic control were analyzed separately. For each group of patients, data regarding anthropometric parameters, serum analytes, total expression, and the relative phosphorylation level of proteins extracted from Barrett’s lesion were reported. Data indicate the median variation (T24−T0) of the parameter considered, followed by a p-value (the Wilcoxon matched-pairs signed-rank test was used to compare measurements at T0 and T24 within the same group). A reduction in protein levels was indicated as the median of the variation expressed as a percentage (%). An increase in protein level was indicated as n-fold of the amount detected at baseline. Parameters not included in this diagram were not significantly modified. 1 The Fisher’s exact test was performed to assess differences between the proportions of patients belonging to the two arms within each subpopulation identified. The up/down arrow indicates an increase/decrease in the parameter’s measure at 24 months compared to baseline. The green arrow indicates a positive change (protective against cancer evolution) in the parameter considered. The red arrow indicates a negative change (procancer) in the parameter considered.