Choice of diet type
Analysis of diet history questionnaires revealed that 22 of the 29 patients in the KD group already had experience with consuming a LCD or KD over 6–24 months. In contrast, patients in the LCD or SD group started from a Western diet. The majority of patients selected LCD.
Energy and protein uptake
Since cancer patients need a considerable amount of energy and an increased daily intake of protein [16] we analyzed both components on the basis of the food diaries maintained by the patients in the outpatient phase. All three diets reached the goal intake of 25–30 kcal/kg/d for ambulatory patients Fig. 2A). However, the KD supported the patients with a mean of 32.5 ± 1.5 kcal/kg/d, which was significantly higher than the energy supply obtained with LCD (24.3 ± 0.7 kcal/kg/d; p < 0.0001) as well as with SD (26.4 ± 1.2 kcal/kg/d; p = 0.0005). Both KD (1.33 ± 0.07 g/kg/d) and LCD (1.2 ± 0.03 g/kg/d) supplied sufficient protein, while SD (0.98 ± 0.04) failed to reach the goal of 1.0 − 1.5 g/kg/d protein (Fig. 2B). Patients in the KD group reached the intended ketogenic ratio (KR; grams of fat divided by grams of carbohydrates plus protein) of 1.6:1 (1.65 ± 0.08) and exhibited stable ketosis according to the daily urine measurements (not shown). As expected, LCD patients had a higher KR than SD patients (Fig. 2C).
Physical performance
Because the majority of the participants in the KD group already were on a high fat diet at study entry, this group had a lower respiratory quotient (RQ; 0.79 ± 0.01) than LCD (0.85 ± 0.01) and SD (0.88 ± 0.02) at T0. During the intervention, the RQ in the KD group further decreased to 0.76 ± 0.01, almost reaching the 0.7 value of pure fat oxidation. The RQ in the LCD group remained stable (0.85 ± 0.01), while that of the SD group increased (0.9 ± 0.02), in line with the carbohydrate rich diet (Fig. 3A).
All three groups improved their physical performance during the intervention reflecting effects of the multimodal therapy. However, despite being the group with the highest percentage of advanced diseases, patients in the KD group performed best in the treadmill test at T0 with higher maximal oxygen uptake (VO2max/kg; 24.7 ± 1.2; Fig. 3B), and maximal workload (142 ± 5.8 Watt; Fig. 3C) as well as longer time to exhaustion (8.5 ± 0.4 min; Fig. 3D). At T20, the KD remained the fittest group. All three diet groups increased their VO2max/kg from T0 to T20 which was significant in both the LC (BHcv = 0.02) and SD groups (BHcv = 0.03). Improvements in workload and time to exhaustion missed significance. In all three groups, lactate (3 min after exhaustion) in peripheral blood was slightly higher (4.1–4.4 mmol/l) at T20 compared to T0 (3.7–3.8 mmol/l) without any significant inter- or intra-group differences (not shown).
Body composition
There was no significant intergroup difference in body mass index (BMI) at T0 but all three interventions induced a significant reduction until T20 (Fig. 3E). BMI decreased from 23.9 ± 0.7 kg/m2 to 22.9 ± 0.7 kg/m2 (BHcv = 0.05) in the KD group, from 27.3 ± 0.5 kg/m2 to 26.1 ± 0.5 kg/m2 (BHcv = 0.017) in the LCD group and from 26.8 ± 0.8 kg/m2 to 26.0 ± 0.9 kg/m2 (BHcv = 0.033) in the SD group.
The phase angel was very similar in all diet groups (KD/LCD/SD) at T0 (5.69 ± 0.12 /5.56 ± 0.05/5.68 ± 0.10) and remained stable until T20 (5.68 ± 0.14/5.57 ± 0.05/5.63 ± 0.10; Fig. 3F).
Neither at T0 nor at T20 was there a significant difference in bone density between the groups (Fig. 3G). Bone density slightly decreased in the KD (T0: 1.11 ± 0.02 g/cm2; T20: 1.09 ± 0.02 g/cm2) and LCD group (T0: 1.08 ± 0.01 g/cm2; T20: 1.07 ± 0.01 g/cm2) and remained stable at the lowest level in the SD group (T0: 1.06 ± 0.02 g/cm2; T20: 1.06 ± 0.02 g/cm2).
All three diet groups improved their muscle/fat mass ratio based on a significant reduction of total body fat (16.7% in KD; 10.5% in LCD and 5% in SD) and only a slight reduction in muscle mass (2.5% in KD; 1.4% in LCD and 1.8% in SD) during the intervention (Fig. 3H). The KD group exhibited the best ratio at T0 (2.07 ± 0.15) and further improved to 2.43 ± 0.14 at T20, which was significantly higher (BHcv = 0.025) compared to LCD and SD at T20.
Patients in the KD group had a lower amount of visceral fat mass at T0 (10.0 ± 0.9 kg) and further reduced it to 7.9 ± 0.8 kg at T20. While LCD patients had lost around 1.6 kg of visceral fat (from 14.3 ± 0.5 to 12.6 ± 0.6 kg, BHcv = 0.017), the reduction was 0.6 kg in the SD group (12.7 ± 0.9 to 12.1 ± 1.0; Fig. 3I). The percentage of visceral fat of total body fat improved in the KD (from 44.2 ± 1.3 to 42.37 ± 1.3) and LCD groups (from 48.9 ± 1.0 to 48.1 ± 1.2) while it remained stable in the SD group (46.8 ± 1.1 to 46.9 ± 1.3; Fig. 3J).
Quality of life (QoL)
QoL assessed with the EORTC-QLQC30 and BR23 questionnaires improved in all three diet groups. At T0, the KD group had a significantly higher global quality of life and lower fatigue than the other two groups, while all other parameters were without a significant difference (Table 2).
Table 2
Quality of life (EORTC-QLQ C30, version 3 + BR23)
| | KD | | | LCD | | | SD | | T0 Intergroup | T20 Intergroup |
| T0 | T20 | BHcv | T0 | T20 | BHcv | T0 | T20 | BHcv | BHcv | BHcv |
Glob. health/QoL | 61.2 ± 3.4 | 71.3 ± 4.5 | 0.006** | 47.2 2.1 | 64.7 2.4 | 0.006** | 55.1 ± 3.1 | 64.2 ± 3.7 | 0.042* | 0.016* | 0.055 |
Physical funct. | 77.9 ± 3.0 | 88.7 ± 4.3 | 0.060 | 66.7 ± 2.3 | 79.2 ± 1.9 | 0.006** | 70.3 ± 3.0 | 80.0 ± 2.9 | 0.024* | 0.063 | 0.003** |
Emotional funct. | 56.6 ± 5.0 | 74.2 ± 6.3 | 0.036* | 41.9 ± 3.2 | 62.0 ± 3.1 | 0.006** | 46.5 ± 5.1 | 58.0 ± 5.2 | 0.065 | 0.070 | 0.047* |
Cognitive funct. | 66.7 ± 5.6 | 70.1 ± 6.4 | 0.167 | 57.4 ± 3.3 | 64.7 ± 3.2 | 0.095 | 52.2 ± 6.2 | 61.8 ± 4.6 | 0.071 | 0.078 | 0.117 |
Social funct. | 56.9 ± 4.9 | 70.0 ± 6.6 | 0.077 | 57.2 ± 3.3 | 68.5 ± 3.4 | 0.030* | 57.0 ± 5.8 | 75.0 ± 4.6 | 0.024* | 0.18 | 0.148 |
Role funct. | 57.5 ± 5.1 | 70.8 ± 7.3 | 0.107 | 50 ± 3.2 | 63.6 ± 3.5 | 0.012* | 56.5 ± 5.8 | 61.8 ± 5.3 | 0.119 | 0.125 | 0.109 |
Fatigue | 42.9 ± 5.3 | 20.6 ± 5.7 | 0.060 | 62.6 ± 2.8 | 40.9 ± 3.2 | 0.006** | 57.0 ± 4.7 | 46.3 ± 4.3 | 0.083 | 0.023* | 0.008** |
Pain | 41.4 ± 6.0 | 29.2 ± 7.4 | 0.107 | 50.2 ± 3.7 | 40.2 ± 3.6 | 0.054 | 49.5 ± 6.2 | 31.9 ± 4.6 | 0.036* | 0.125 | 0.086 |
Dyspnea | 46.0 ± 5.6 | 30.0 ± 7.6 | 0.101 | 49.6 ± 3.4 | 34.2 ± 3.7 | 0.018* | 43.0 ± 6.6 | 33.3 ± 5.7 | 0.077 | 0.133 | 0.156 |
Insomnia | 58.6 ± 6.5 | 28.3 ± 7.3 | 0.048* | 63.6 ± 3.1 | 56.9 ± 3.8 | 0.119 | 51.6 ± 6.5 | 47.2 ± 8.9 | 0.155 | 0.094 | 0.039* |
Nausea/Vomiting | 8.6 ± 3.8 | 5.0 ± 2.7 | 0.077 | 7.2 ± 1.5 | 4.5 ± 1.2 | 0.131 | 14.0 ± 3.9 | 7.6 ± 3.1 | 0.083 | 0.102 | 0.133 |
Appetite loss | 12.6 ± 4.2 | 8.3 ± 4.1 | 0.161 | 17.0 ± 2.6 | 7.6 ± 2.0 | 0.060 | 16.1 ± 4.9 | 13.9 ± 4.5 | 0.137 | 0.141 | 0.102 |
Constipation | 14.9 ± 4.8 | 10.0 ± 4.3 | 0.113 | 14.4 ± 2.8 | 9.8 ± 2.5 | 0.125 | 16.1 ± 4.9 | 12.5 ± 5.2 | 0.179 | 0.172 | 0.164 |
Diarrhea | 12.6 ± 5.1 | 11.7 ± 4.4 | 0.173 | 15.5 ± 2.9 | 12.0 ± 2.9 | 0.143 | 12.9 ± 3.7 | 8.3 ± 3.6 | 0.149 | 0.148 | 0.156 |
Benjamini Hochberg corrected p-value (BHcv): BHcv < 0.05*; BHcv < 0.01** |
The improvement in fatigue was not significant in the KD group upon intervention, but still superior to the other groups; furthermore, global QoL improved significantly (BHcv = 0.006) and remained the highest among the diet groups. The same could be observed for emotional functioning (BHcv = 0.036) and insomnia (BHcv = 0.048); the latter improvement resulted in a significant advantage of the KD compared to the LCD and SD groups (BHcv = 0.039). Participants in the LCD group significantly improved in nine of the 14 parameters of QoL, and SD participants in four of those nine aspects, which were global QoL (BHcv = 0.042), pain (BHcv = 0.036), physical (BHcv = 0.024) and social (BHcv = 0.024) functioning. Symptoms of dyspepsia (nausea/vomiting, appetite loss, constipation and diarrhea) improved on all three dietary regimens resulting in an equal outcome at T20.
Blood parameters
Results of blood chemistry are shown in Table 3.
Table 3
blood parameters at start and end of intervention
| | | KD | LCD | SD | Between-group differences |
Parameter and dimension | | Normal range | T0 | T20 | Within-group difference (BHcv) | T0 | T20 | Within-group difference (BHcv) | T0 | T20 | Within-group difference (BHcv) | T0 BHcv | T20 BHcv |
TG | (mg/dl) | < 150 | 78.1 ± 5.1 | 82.0 ± 6.6 | 0.130 | 128.3 ± 8.0 | 106.5 ± 7.3 | 0.005** | 105.3 ± 8.9 | 106.0 ± 10.9 | 0.146 | 0.008** | 0.156 |
Chol | .(mg/dl) | < 200 | 234.6 ± 8.5 | 239.1 ± 11.4 | 0.135 | 233.5 ± 4.8 | 221.6 ± 4.8 | 0.005** | 215.2 ± 6.1 | 203.6 ± 5.5 | 0.047* | 0.125 | 0.047* |
HDL | (mg/dl) | > 65 | 79.9 ± 3.7 | 78.6 ± 4.2 | 0.078 | 64.5 ± 1.5 | 68.2 ± 1.6 | 0.026* | 65.2 ± 2.7 | 64.6 ± 6.2 | 0.057 | 0.016* | 0.055 |
LDL | (mg/dl) | < 130 | 141.8 ± 7.1 | 146.9 ± 8.7 | 0.010* | 152.0 ± .9 | 141.2 ± 3.7 | 0.005** | 137.2 ± 5.6 | 125.4 ± 4.8 | 0.052 | 0.109 | 0.047* |
LDL/HDL | (ratio) | < 2,5 | 1.9 ± 0.1 | 2.0 ± 0.2 | 0.177 | 2.5 ± 0.1 | 2.1 ± 2.3 | 0.005** | 2.2 ± 0.1 | 2.0 ± 0.1 | 0.089 | 0.023* | 0.102 |
TG/HDL | (ratio) | < 1.25 | 1.1 ± 0.1 | 1.1 ± 0.1 | 0.156 | 2.2 ± 0.2 | 1.7 ± 0.1 | 0.005** | 1.9 ± 0.2 | 1.8 ± 0.2 | 0.151 | 0.008** | 0.039* |
BG | (mg/dl) | 74–106 | 86.5 ± 1.8 | 85.3 ± 1.8 | 0.167 | 91.2 ± 1.4 | 88.7 ± 1.1 | 0.016* | 86.0 ± 1.4 | 85.5 ± 1.8 | 0.125 | 0.070 | 0.148 |
Crea | (mg/dl) | 0.7–1.1 | 0.83 ± 0.02 | 0.79 ± 0.02 | 0.052 | 0.78 ± 0.01 | 0.73 ± 0.01 | 0.010* | 0.79 ± 0.03 | 0.80 0.03 | 0.109 | 0.094 | 0.031* |
GFR | (ml/min) | 60–150 | 82.0 ± 2.9 | 85.6 ± 25 | 0.042* | 88.6 ± 1.6 | 92.9 ± 1.4 | 0.010* | 84.6 ± 3.8 | 83.9 ± 3.5 | 0.172 | 0.117 | 0.031* |
Uric acid | (mg/dl) | 2.6-6.0 | 4.8 ± 0.2 | 4.8 ± 0.2 | 0.083 | 5.3 ± 0.1 | 5.0 ± 0.1 | 0.031* | 5.3 ± 0.2 | 5.2 ± 0.3 | 0.156 | 0.094 | 0.188 |
BUN | (mg/dl) | 17–43 | 30.5 ± 1.2 | 33.1 ± 1.1 | 0.021* | 29.0 ± 0.7 | 32.5 ± 0.9 | 0.005** | 28.8 ± 1.3 | 29.1 ± 1.4 | 0.063 | 0.172 | 0.164 |
AP | (U/l) | 30–120 | 66.6 ± 4.7 | 64.2 ± 4.4 | 0.130 | 68.8 ± 2.4 | 67.0 ± 2.9 | 0.036* | 65.9 ± 4.8 | 65.3 ± 5.0 | 0.068 | 0.180 | 0.195 |
AST | (U/l) | < 31 | 28.8 ± 4.8 | 25.7 ± 4.0 | 0.089 | 30.1 ± 2.1 | 28.2 ± 1.9 | 0.094 | 24.2 ± 1.8 | 23.9 ± 1.8 | 0.167 | 0.086 | 0.133 |
CRP | (mg/l) | < 5.0 | 1.6 ± 0.4 | 0.83 ± 0.2 | 0.078 | 2.2 ± 0.23 | 1.9 0.2 | 0.052 | 2.3 ± 0.5 | 1.7 ± 0.4 | 0.112 | 0.195 | 0.133 |
TSH | (mU/l) | 0.3–3.5 | 1.5 ± 0.2 | 1.6 ± 0.2 | 0.141 | 1.7 ± 0.1 | 1.6 ± 0.1 | 0.073 | 1.0 ± 0.1 | 1.2 ± 0.2 | 0.104 | 0.063 | 0.078 |
Protein | (g/dl) | 6.6–8.3 | 6.7 ± 0.09 | 6.7 ± 0.09 | 0.182 | 6.8 ± 0.04 | 6.8 ± 0.04 | 0.182 | 6.9 ± 0.08 | 6.7 ± 0.09 | 0.042* | 0.148 | 0.24 |
Hemoglobin | (g/dl) | 12-15.5 | 13.1 ± 0.2 | 13.4 ± 0.2 | 0.167 | 13.2 ± 0.1 | 13.4 ± 0.1 | 0.125 | 13.2 ± 0.2 | 13.0 ± 0.2 | 0.250 | 0.250 | 0.125 |
Lekocyte | (103/µl) | 3.9–11 | 4.68 ± 0.25 | 5.17 ± 0.29 | 0.083 | 5.41 ± 0.16 | 5.82 ± 0.20 | 0.042* | 5.27 ± 0.31 | 5.56 ± 0.35 | 0.208 | 0.063 | 0.188 |
Insulin | (µU/ml) | 2–25 | 12.8 ± 2.0 | 12.3 ± 1.6 | 0.177 | 16.3 ± 0.9 | 14.0 ± 0.8 | 0.031* | 16.9 ± 2.3 | 16.6 ± 1.9 | 0.172 | 0.188 | 0.086 |
HOMA-IR | index* | 1.9/2.9 | 2.73 ± 0.4 | 2.79 ± 0.4 | 0.151 | 3.55 ± 0.3 | 3.14 ± 0.3 | 0.016* | 3.68 ± 0.5 | 3.58 ± 0.4 | 0.182 | 0.086 | 0.156 |
TG: triglycerides; Chol: total cholesterol; HDL: high density cholesterol; LDL: low density cholesterol; BG: blood glucose; Crea: creatinine; GFR: glomerular filtration rate; BUN: bound urea nitrogen; AP: alkaline phosphatase; AST: aspartate transaminase; CRP: C-reactive protein; TSH: thyoid-stimulating hormone; HOMA-IR: Homeostatic Model Assessment of Insulin Resistance *index: 1.9–2.9 = early insulin resistance; >2.9: significant insulin resistance |
Blood parameters for the three intervention groups measured at T0 and T20. Parameter values are given as mean ± standard error of the mean (SEM). The other columns give p-values derived from various tests: (i) within-group differences between T0 and T20 were evaluated using the Mann-Whitney-U test; (ii) between-group differences at T0 and T20 were evaluated using the Kruskal-Wallis test; p-values were adjusted following the Benjamini-Hochberg procedure with a false discovery rate of 25% resulting in Benjamini-Hochberg critical values (BHcv). *BHcv ≤ 0.05; **BHcv ≤ 0.01 |
At T0, the KD group had the lowest triglycerides (TG; 78.1 ± 5.1 mg/dl) and highest HDL concentrations (79.9 ± 3.7 mg/dl) compared to LCD (128.3 ± 8.0/64.5 ± 1.5) and SD (105.3 ± 8.9/65.2 ± 2.7), while LDL was similar between groups, resulting in significantly higher HDL/LDL (BHcv = 0.023) and TG/HDL (BHcv = 0.008) ratios. TG and HDL remained relatively stable until T20 in the KD and SD group, but there was a significant decrease of TG (BHcv = 0.005) and increase of HDL (BHcv = 0.026) in the LCD group. Of note, total cholesterol was highest in the KD group and remained high, while in the LCD and SD groups a significant reduction of cholesterol was observed, corresponding to a decrease in LDL with only a slight increase (LCD) or decrease (SD) in HDL. However, the final TG/HDL ratio in either the LCD (1.7 ± 0.1) or SD (1.8 ± 0.2) group did not reach the recommended ratio of < 1.25 that was only achieved in the KD group (1.1 ± 0.1). Due to the distinct changes in LDL and HDL, the LDL/HDL ratio was nearly the same in all three groups at T20 (2.0-2.1).
Throughout the intervention, no significant change occurred for blood glucose and insulin in the KD and SD groups. Here, blood glucose concentration was comparable in KD and SD, but insulin remained lowest in KD and highest in SD, resulting in the lowest HOMA-IR (2.8 ± 0.4) in the KD group and highest (3.6 ± 0.4) in the SD group at T20. There was a significant decrease of blood glucose in the LCD group (from 90.2 mg/dl to 88.7; BHcv = 0.016) which was accompanied by a significant decrease in insulin (from 16.3 to 14.0 µU/ml; BHcv = 0.031) and reduction of HOMA-IR to 3.14 ± 0.3 (BHcv = 0.016).
Concerning markers of kidney function, all thee intervention groups started from comparable ranges. While there was no influence of diet type on creatinine, glomerular filtration rate (GFR), uric acid and bound urea nitrogen (BUN) in the SD group, creatinine decreased (BHcv = 0.01) and GFR significantly increased in the KD (BHcv = 0.04) and LCD (BHcv = 0.01) group. Uric acid remained stable in the KD and SD groups and decreased in the LCD group (BHcv = 0.03); BUN increased on the KD (BHcv = 0.02) and LCD (BHcv = 0.005); however all parameters stayed within the normal range.
The liver parameters alkaline phosphatase (AP) and aspartate transaminase (AST) did not differ between groups at T0 and remained stable throughout the intervention with the exception of a significant decrease of AP in the LCD group (BHcv = 0.036).
At T0 and T20, c-reactive protein (CRP) was lowest in the KD group without significant differences within or between groups. Serum protein remained stable in the KD and LCD group but decreased in the SD group (BHcv = 0.04).
Hemoglobin remained relatively stable upon intervention, slightly raised on the KD (from 13.1 ± 0.2 g/dl to 13.4 ± 0.2) and LCD (from 13.2 ± 0.1 to 13.4 ± 0.1), and decreased on the SD (from 13.2 ± 0.2 to 13.0 ± 0.2). In all three groups, there was a positive development in leukocyte count that was significant in the LCD (from 5.41 ± 0.16 to 5.82 ± 0.2; BHcv = 0.042) and without significance in the KD (from 4.68 ± 0.25 × 103/µl to 5.17 ± 0.16; p = 0.04) and SD (from 5.27 ± 0.31 to 5.56 ± 0.35) group.