A viscerally driven cachexia syndrome in patients with advanced colorectal cancer: contributions of organ and tumor mass to whole-body energy demands

Abstract

Background: Cancer cachexia-associated weight loss is poorly understood; energetically demanding tissues (eg, organ and tumor mass) and resting energy expenditure (REE) are reported to increase with advanced cancer.

Objective: The objective was to quantify the potential contribution of increasing masses of energetically demanding tissues to REE with colorectal cancer cachexia progression.

Design: A longitudinal computed tomography (CT) image review was performed to quantify organ size (liver, including metastases, and spleen) and peripheral tissues (skeletal muscle and adipose tissue) during colorectal cancer cachexia progression (n = 34). Body composition was prospectively evaluated by CT and dual-energy X-ray absorptiometry, and REE was determined by indirect calorimetry in advanced colorectal cancer patients (n = 18).

Results: Eleven months from death, the liver (2.3 +/- 0.7 kg) and spleen (0.32 +/- 0.2 kg) were larger than reference values. One month from death, liver weight increased to 3.0 +/- 1.5 kg (P = 0.010), spleen showed a trend to increase (P = 0.077), and concurrent losses of muscle (4.2 kg) and fat (3.5 kg) (P < 0.05) were observed. The estimated percentage of fat-free mass (FFM) occupied by the liver increased from 4.5% to 7.0% (P < 0.001). The most rapid loss of peripheral tissues and liver and metastases gain occurred within 3 mo of death. A positive linear relation existed between liver mass and measured whole-body REE (r(2) = 0.35, P = 0.010); because liver accounted for a larger percentage of FFM, measured REE . kg FFM(-1) . d(-1) increased (r(2) = 0.35, P = 0.010).

Conclusions: Increases in mass and in the proportion of high metabolic rate tissues, including liver and tumor, represented a cumulative incremental REE of approximately 17,700 kcal during the last 3 mo of life and may contribute substantially to cachexia-associated weight loss.

FIGURE 1

Representative patients showing change in (A) liver and spleen and (B) skeletal muscle and adipose tissue. Transverse computed tomography images from (A) a 54-y-old woman at the 12th thoracic vertebrae (i) 10.0 and (ii) 1.5 mo from death [the liver with metastases (LM) increased by 1.3 kg to 3.1 kg, and spleen (SP) increased by 0.30 kg to 0.50 kg], and (B) a 52-y-old man at the 3rd lumbar vertebrae (i) 7.7 and (ii) 0.8 mo from death [estimated whole-body muscle (SM) decreased by 8.8 kg to 17.0 kg, and whole-body adipose tissue (AT) decreased by 2.0 kg to 16.7 kg].

FIGURE 2

Time course rates of gain or loss for liver (including metastases), muscle, and adipose tissue from the retrospective colorectal cancer patient cohort (n = 34). Scan intervals were categorized relative to the time of death and divided into 5 categories. Mean rates of change were determined for each tissue at each time point. Best-fit regression lines were used to determine the overall rate of change relation over time. The rate of change in liver followed a polynomial relation: liver %change/100 d = 0.0017(time to death, in d)² − 0.7316(time to death, in d) + 75.56 (r² = 0.90) (▴ and dotted curve). The loss of skeletal muscle was logarithmic: skeletal muscle %change/100 d = 8.8303ln(time to death, in d) − 50.746 (r² = 0.99) (♦ and solid curve). Loss of adipose tissue was also logarithmic: adipose tissue %change/100 d = 32.029ln(time to death, in d) − 172.92 (r² = 0.95) (▪ and dashed curve).

FIGURE 3

Relation between measured resting energy expenditure (REE) and liver mass in the prospective colorectal cancer patient cohort (n = 18). Liver mass (including metastases) was determined by computed tomography image analysis. REE was determined by indirect calorimetry, and fat-free mass (FFM) was determined by dual-energy X-ray absorptiometry. Simple linear regression was used to assess relations. A: REE (kcal/d) = 343.52(liver mass, in kg) + 841.49 (r² = 0.35, P = 0.010). B: REE (kcal · kg FFM⁻¹ · d⁻¹) = 3.0011(%FFM occupied by the liver) + 20.513 (r² = 0.35, P = 0.010) (dashed line). The solid line indicates a similar relation occurring in the new cancer cachexia simulation incorporating the measured liver masses from the retrospective cohort (Table 1).

FIGURE 4

¹⁸F-Deoxyglucose positron emission tomography scan of a patient with extensive liver metastases.

FIGURE 5

A, B: Simulation of resting energy expenditure (REE) over 62 wk based on measured liver and spleen masses from the retrospective colorectal cancer patient cohort (n = 34). All calculations are based on an assumed liver-specific metabolic rate of 200 kcal · kg⁻¹ · d⁻¹ and a spleen-specific metabolic rate of 80 kcal · kg⁻¹ · d⁻¹. The original cachexia simulation (constant: 1.8 kg liver and 250 g spleen) (14) (dashed curve) can be contrasted with the cachexia simulation specifying the organ masses match the data at different time points from the retrospective cohort computed tomography images (Table 1) (solid curve). The healthy reference simulation in energy balance (constant: 1.8 kg liver and 250 g spleen) (dotted curve) and the reduced energy intake simulation (dashed dotted curve) are also shown (13).