- Click here for explanation of classification scheme.

To investigate the impact of dietary counseling or nutrition supplements on outcomes in cancer patients on nutrition, morbidity and quality of life (QoL) both during and three months after radiotherapy (RT) for colorectal cancer (CRC).

• Free-living colorectal patients stratified based on their stage of GI cancer (I/II and III/IV) and referred for RT treatment of 50.4 Gy administered in 28 fractions, regardless if the radiation therapy was primary, adjuvant to surgery, combined with chemotherapy or with palliative intent
• Absence of renal disease or diabetes mellitus
• All patients who gave written consent to participate in the study. 

 

• Renal disease
• Diabetes mellitus.

Recruitment

Between the periods of July 2000 and March 2003, all consecutive CRC ambulatory patients referred  for RT were considered eligible, regardless of whether the proposed RT was primary, adjuvant to surgery, combined with chemotherapy or with palliative intent. 

Design

• Prospective, randomized, controlled trial
• Patients stratified into two groups according to stage of cancer, stage I/II or stage III/IV
• Patients then randomly assigned in permutation blocks of three:
  • Group one (G1; N=37): Received individualized dietary counseling
  • Group two (G2; N=37): Was asked to add and  consume two cans of high-protein liquid supplements(200kcals per 20g protein per 200cc) in addition to their usual diet
  • Group three (G3; N=37): Served as control group and maintained their usual diet
• Nutritional intake (24-hour recall), nutritional status using Ottery's Subjective Global Assessment and QoL (European Organization for Research and Treatment of Cancer Quality of Life Questionnaire version 3.0) were evaluated at baseline, at the completion of radiation therapy and three months after radiotherapy.

Blinding Used

• Randomization sequence was not disclosed to study personnel
• Patients randomly assigned at enrollment received sealed envelopes with random assignments to group one, group two or group three
• Before the first appointment, randomization envelopes were opened by person blind to the study procedures.

Intervention

After randomization, patients were assigned to one of three groups: 

• Group one (G1; N=37) received individualized dietary counseling based on regular foods.The diet prescription identified the type, amount and frequency of feeding; specified the calorie and protein level to attain; and included any restrictions and limited or increased individual dietary components
• Group two (G2; N=37) were asked to consume two cans per day of high-protein liquid supplement in addition to their usual diet. Supplements used throughout the study were always the same commercial brand. Each 200ml can provided 20g protein and 200kcal. Compliance was monitored by using a supplement consumption record, which was kept daily by patients and verified by a caregiver or relative.
• Group 3 (G3; N=37), the control group, were asked to maintain their ad libitum intake.
• Randomly assigned patients had scheduled visits and identical contact time with the research dietitian. 

Statistical Analysis

• Statistical analysis was performed using SPSS 11.5 (SPSS, Inc., Chicago, IL) and EPI-Info 2000 (Centers for Disease Control, Atlanta, GA).
• All analyses were conducted on an intention-to-treat-basis, and therefore available data from all study patients were used. If any missing data were observed, the missing value(s) would be replaced by the average of the study group,which would have no effect on the estimators. Study groups had been assessed for comparability at time of entry into the study.
• Data related to incidence, prevalence or frequency (symptoms, cancer stages and nutritional status categories) were expressed as number and percentages
• Data related to age was expressed as mean ± standard deviation and range
• Data related to energy and protein intake were expressed as the median and range
• The Quality of Life scores were expressed as median values
• The continous variables were analyzed using the one-way analyses of variance or Wilcoxon rank sum test as appropriate
• The categorical variables and incidence, prevalence or frequency were evaluated by the chi-square test
• Univariate or multiple correlations were assessed by two-tailed non-parametric Spearman's tests
• Statistical significance was set for a P value less than 0.05
• A minimum sample size of 58 patients was calculated to detect a difference in body weight of 1.9kg, in nutritional intake 25% and in QoL scores of 20% (i.e., an effect size of 0.9) with a significance level of 0.01 between groups and a power of 0.85
• Statistical power was based on on the changes observed in weight, nutritional intake and QoL from a pilot study conducted in 46 patients with CRC.

Timing of Measurements

• At baseline, the following were measured:
  • Medical history
  • Demography
  • Concommitant medications
  • Nutritional Status with PG-SGA
  • Weight
  • Diet history
  • QoL with EORTC QLQ-C30
• At day seven, 14, 21, 28 and day 35 of radiotherapy, the following were measured:
  • Concommitant medications were assessed
  • Nutritional status with PG-SGA
  • Weight
  • 24-hour recall
  • RT-induced morbidity with EORTC/RTOG (except at day seven)
  • Acceptability and compliance of nutritional intervention
• At day 42, end date of radiation therapy, the following were measured:
  • Concommitant medications were assessed
  • Nutritional status with PG-SGA
  • Weight
  • 24-hour recall
  • RT-induced morbidity with EORTC/RTOG
  • Acceptability and compliance of nutritional intervention
  • QoL with EORTC QLQ-C30
• At three months following radiotherapy, the following were measured:
  • Concommitant medications were assessed
  • Nutritional status with PG-SGA
  • Weight
  • Diet history
  • 24-hour recall
  • RT-induced morbidity with EORTC/RTOG
  • QoL with EORTC QLQ-C30
  Compliance to high protein supplements was measured using a daily consumption record kept by patient and verified by caregiver or relative.

Dependent Variables

• Variable one: Nutritional status measured by:
  • Ottery's Patient Generated Subjective Global Assessment
  • Anthropometric data including height, weight and BMI
• Variable two: Quality of life using the EORTC Quality of Life Questionnaire version 3.0 (EORTC-QLQ C30)
• Variable three: RT induced morbidity with EORT/RTOG
• Variable four: Nutritional intake derived from diet history(Burke's diet history) and 24-hour-recall food questionnaire (two weekdays and one weekend day). DIETPLAN version 5 for Windows was used to analyze nutrient data.

Independent Variables

• Therapeutic diet
• High-protein polymeric formula.

Control Variables

37 patients advised to maintain their ad libitum intake.

• Initial N: 111 patients; 66 males, 45 females
  • Group I:  Dietary counseling, regular foods; N=37
  • Group II: Protein supplements; N=37
  • Group III: Ad libitum intake; N=37
• Attrition (final N): All patients recruited completed the study and none were lost to follow-up
• Age: 58±15 years (range 32 to 88 years)
• Ethnicity: Assume Portuguese, though this was not directly mentioned in the study (study was conducted at Portuguese hospital)
• Other relevant demographics: TNM Stage was 45 in Stage I/II and 66 in Stage III/IV
• Anthropometrics:
  • At baseline, PG-SGA identified 15 malnourished patients in G1, 14 in G2 and 13 in G3
  • At baseline, BMI identified five malnourished patients in G1, four in G2 and three in G3(malnourished = BMI less than 20)
• Location: Radiotherapy Department of the Santa Maria University Hospital in Lisbon, Portugal.

Nutritional Intake

• Median estimated requirements and median nutritional intake similar in all groups
• In comparison with baseline data, at the end of RT, net intake increased by 535kcal per day in group one and 296kcal per day in group two
• Energy intake decreased by 285kcal per day in group three
• At three-month follow-up, group one maintained calorie intake where as patients in group two and group three decreased calorie intake.

Nutritional Status

Table 1: Changes in Nutritional Status During RT and at Three Months Categorized According to PG-SGA and BMI (Data Expressed as Numbers of Patients)

	Group One Decline		Group One Maintained or Improved		Group Two Decline		Group Two Maintained or Improved		Group Three Decline		Group Three Maintained or Improved
Methods	End RT	Three Months	End RT	Three Months	End RT	Three Months	End RT	Three Months	End RT	Three Months	End RT	Three Months	P*	P
PG-SGA	3	10	34	27	19	24	18	13	34	36	3	1	<0.002	<0.001
BMI	1	2	36	35	3	6	34	31	5	8	32	29	NS	NS

P^* indicates significance of statistical differences between intervention groups, regarding nutritional decline both at the end of RT and at three months.

P indicates significance of statistical differences between intervention groups, regarding maintenance or improvement of nutritional status at the end of RT and at three months.

Table 2: RT-induced Morbidity Categorized According to Severity Grades

	Group One Grade One		Group One Grade Two		Group Two Grade One		Group Two Grade Two		Group Three Grade One		Group Three Grade Two
Symptoms	End RT	Three Months	End RT	Three Months	End RT	Three Months	End RT	Three Months	End RT	Three Months	End RT	Three Months	P1	P2	P3
Anorexia	20	6	13	1	19	5	14	3	17	12	17	10	<0.02	<0.01	<0.001
Nausea or Vomiting	27	0	7	0	23	7	10	3	18	9	16	6	<0.001	0.17	<0.0001
Diarrhea	32	0	2	0	2	9	9	3	18	15	17	13	<0.0001	<0.5	<0.0001

Data is expressed as numbers of patients

P¹ expresses the significance of statistical differences between intervention groups, regarding the reduction of grade one symptom incidence between the end of RT and three months.

P² expresses the significance of statistical differences between intervention groups, regarding the reduction of grade two symptom incidence between the end of RT and three months.

P³ expresses the significance of statistical differences between intervention groups, regarding the reduction of grades one and two symptom incidence between the end of RT and three months.

Quality of Life

• Quality of Life Function Scores:
  • Quality of life function scores improved significantly (P<0.002) in group one despite RT-induced symptoms (P<0.05)
  • Linear positive association with improvement of patients' nutritional status (P<0.05)
  • Physical, role and emotional scores improved (P<0.05) in group two and these were proportional to the increase in protein intake (P=0.04)
  • In group three, all function scores decreased in association with worsening nutritional intake (P<0.0001) as well as their decline in nutritional status (P<0.001)
• Quality of Life Symptoms, scales and single items:
  • In group one, pain increased in association with anorexia (P=0.05), nausea or vomiting (P=0.04) and diarrhea (P=0.03)
  • In group two, decline in fatigue and pain were associated with anorexia (P<0.001), nausea or vomiting (P≤0.04) and diarrhea (P<0.002). Also, severity of sleep disturbance increased (P=0.02).
  • In group three, increased fatigue was associated with poorer nutritional intake (P<0.003) and with decline in nutritional status (P<0.001). Pain increased in association with nausea or vomiting and diarrhea (P<0.001).
  • In group three, sleep disturbance and appetite became worse and were associated with nausea or vomiting and diarrhea (P<0.002).

Three-month Follow-up (In Comparison with the End of RT)

• Group One: 
  • Maintained or improved their overall quality of life (P<0.02). This was positively and proportionally associated with maintenance or improvement of nutritional status (P<0.02) and adequate nutritional intake (P<0.01)
  • Function scales were improved or maintained (P<0.04)
  • Symptom scales or single item scales were similar to baseline scores
• Group Two:
  • Maintained or worsened their overall quality of life (P<0.03)
  • Patients had worse physical, role, emotional and cognitive functions (P<0.05), which were associated with poor dietary intake (P<0.003) and worsening nutritional status (P<0.002) 
  • Even with the improvement of pain, nausea or vomiting and diarrhea (P<0.04), sleep disturbance and anorexia deteriorated (P<0.03). Remaining scores were unchanged in comparison with the end of RT and were worse than at baseline.
• Group Three:
  • Function scores further worsened both in comparison with the end of RT and at baseline (P<0.004)
  • Worsening of function scores was significantly associated with inadequate nutritional intake (P<0.001) and decreased nutritional status (P<0.002)
  • Symptom scale scores compared to those thath were rated poorly at the end of RT and significantly worse than at baseline (P<0.001)
  • Worst scores were associated with insufficient dietary intake (P<0.005). 

• Concurrent individualized dietary counseling based on regular foods diet is the most effective means of improving patient's nutritional intake, nutritional status and quality of life
• Radiotherapy induced toxicity as it relates to symptom incidence and severity was lowest in those patients who received dietary counseling and education specific to their individual needs
• Dietary counseling significantly improved all quality of life function scores
• This prospective randomized control trial is the first to demonstrate that concurrent individualized dietary counseling, based on regular foods, is the most effective means of improving patients' nutritional intake, status and quality of life, thereby lessening radiation therapy-induced morbidity
• The patients who received oral supplements, only three of six function scores improved during supplementation, and these were proportional to the increase in dietary intake; however, once the supplementation was discontinued, most functional scores deteriorated
• Patients not submitted to any nutritional intervention experienced, throughout the entire study, a significant deterioration in function scores and fatigue in direct relationship to the worsening of their nutritional intake and nutritional status
• The authors did note that the quality of life tool was a subjective multi-dimensional construct.
   

Not-for-profit

0 Foundation associated with industry:

• Wide age range (32 to 88 years)
• Small sample size 
• Unclear if patient characteristics were similar among groups, i.e., how many patients were also receiving combined chemotherapy or RT and how many patients were s/p surgery in each group
• Even though the randomized clinical trial is the most powerful type of design to test what factors cause an illness or lead to a certain outcome, it is important to note the subjective nature of the quality of life evaluation instrument utilized
• Further long-term studies are required to study QoL evaluation in the patients. Oral nutritional supplement fact sheet information is required to know the protein source and other nutrients in the supplements. It is not clear whether their protein status is improved after supplementation and not clear of their initial protein status. In addition, biochemical assessments are also required to study the effects of the treatments and time on variables.