This study examined the effect of regular and long-term nutritional counseling on dietary intake, anthropometric measurements, response rate and overall survival rate and quality of life in patients with cancer of the breast, ovary or lung (small cell) undergoing chemotherapy with curative intent.

All patients had:

1. Histologically-verified small-cell lung cancer. Ovarian cancer or breast cancer.
2. Measurable or assessable disease
3. Life expectancy of longer than three months
4. Eastern Cooperative Oncology Group (ECOG) performance status of two or better
5. All patients agreed to enter a chemotherapeutic treatment protocol.

• Prior chemotherapy (except adjuvant chemotherapy in breast cancer)
• Current or planned hormonal therapy (including therapy with anticatrogens and progesterone derivatives
• Major surgery within the last month
• Presence of ascities or clinically-detectable edema, malabsorption or other diseases (unrelated to cancer) that would necessitate dietary intervention
• Metastasis to the central nervous system.

Recruitment

Not described.

Design

• Longitudinal study
• All patients in the study were stratified according to primary disease, extent of weight-loss in the preceding three months and Eastern Cooperative Oncology Group performance status
• A random number table and a system of sealed envelopes to receive either regular and frequent nutrition counseling from the start of the chemotherapy were utilized
• Over the five-month time-frame, the Intervention Group was counseled by the study dietitian, while the Control Group received nutritional support (education) at the discretion of the assigned physician.

Blinding Used

Blinding was not discussed.

Intervention

• A three-day food diary was utilized to evaluate the dietary intake of the Counseled and Control Groups
  • Prior to randomization, each patient was instructed by the study dietitian to weigh and measure foods, utilizing scales and measuring cups
  • Each patient received instructions to record their food intake for three consecutive days (including one weekend day) six times during the five-month period by the study dietitian.
• Both groups were encouraged to take one vitamin-mineral tablet per day, which met the recommended allowance of micronutrients. The same dietitian counseled all patients randomized to receive nutritional intervention before the start of chemotherapy and twice monthly during the five months of the study. Counseling by the dietitian were also available upon the request.
  • The goal was to meet or exceed the Nordic Recommended Allowance for protein and energy requirements. These were daily energy intakes in the range of 1.5 to 1.7 basal energy (BEE), calculated from the Harris Bendict equation and protein intake of 1.0g to 1.2g per kg body weight.
  • Based on the diaries and nutrition-related symptoms, patients were counseled on an individual basis, according to advocated recommendations
  • Counseled patients were offered supplementation with commercial liquid diets, protein supplements or maltodextrin, if indicated
  • The subjects were given their preference among several commercial diets, in a choice of different flavors, (Salvimulsin, Ercopharm, Kvistgird, Denmark, Standard Top Up, Novo Nordisk, Copenhagen, Denmark, Nutrition Standard, Nutricia, Bagsvard, Denmark).
• Patients who submitted less than four food diaries had the data excluded from the analysis. The Cox's test was used to test differences in survival distribution. Survival curves were plotted using the method of Kaplan and Meier.

Baseline Nutritional Data
(Preceding the onset of chemotherapy)

[Note: Values are mean ±SD.]

Statistical Analysis

• Categorical variables were analyzed by chi square test and quantitative variables by analysis of variance
• The ANOVA model was a two-factor design with repeat measures on one factor
• The researchers determined that approximately 50 patients in each group were required in order to detect differences in a weight of five kg, considered to be clinically significant, at a level of significance (two-sided) of 0.05 and a power of 0.8.

Timing of Measurements

• All of the therapies were administered in repeating cycles, approximately every four weeks. Tumor responses were evaluated when patients completed three cycles (approximately two months) of protocol chemotherapy and again after six cycles (approximately five months).
• The first food recording was before the onset of the first chemotherapeutic series and the following five recordings during the week before each of the subsequent scheduled series. For the Intervention Group, the diaries were analyzed upon receipt to determine appropriate changes in nutritional counseling. The diaries of the Control Group were analyzed at the end of the study.
• Dietary counseling was standardized and performed by a trained dietitian two times per month during the five-month time frame of the research project
• All patients underwent anthropometric and laboratory assessments six times prior to the beginning of chemotherapy and five additional times at delivery of dietary records, usually on days of chemotherapy
• Patients answered the quality of life questionnaire three times. These included:
  • Before onset of chemotherapy
  • Before the fourth and sixth chemotherapeutic cycle, usually on the first day of a diet registration period.

Dependent Variables

• Response to chemotherapy
• All patients were staged according to running treatment protocols using physical and radiologic examinations; ultrasonic, radioisotopic and computed tomographic scan and needle biopsy cytology. All patients with ovarian cancer had a staging exploratory laparotomy. Various chemotherapeutic regimens were used:

• In this study, treatment "Responders" were defined as patients with CR or PR, while "Nonresponders" were defined as patients with NC or PD
• The patients with no response, assessed after three cycles of chemotherapy, had their chemotherapy regimes changed, whereas responders continued with the protocol treatment for at least another three cycles or until there was evidence of disease progression
• The patients who died were considered non-responders
• The response evaluation criteria was defined as follows:

Dietary Intake

Dietary data was coded and analyzed for energy and protein intake, utilizing a computerized version of Danish Food Composition Tables (Dankost, version 1.3 National Food Agency, Saborg, Denmark).

Anthropometric Measurements

• Weight (light clothing), determined by weighing on a calibrated physician's scale. The patient's recalled weight, at least three months prior to the cancer diagnosis, was used to assess weight loss.
• The mean values of triplicate measurements at four sites [triceps, biceps, subscapular and suprailiac (skinfolds)] were used to to calculate body density, using the equations reported by Durnin and Womersley
• Triceps skinfold, measured to the nearest millimeter using a Harpenden skin caliper (John Bull, British Indicators Ltd., England)
• Arm muscle area, measured using a non-stretchable tape
• Fat-free mass (FFM), from the calculated body density, the fat-free mass was derived using the Sin's equation
• Fat mass
• Mid-arm muscle area (bone-free arm muscle area) was calculated from arm circumference and triceps skinfold measurements, according to the equation reported by Heymsfield et al.

Laboratory Values

• Plasma albumin
  • The Quality of Life Index utilized to assess the quality of life of the cancer patients was a modified version of the QL-index developed by Spitzer et al. The questionnaire was chosen because it has been shown to be reliable and valid and because scores have been shown to be associated with degree of weight loss in a cancer population similar to the one under investigation. The QL index covers the following items:
    • Involvement in own occupation
    • Activities of daily living
    • Perception of own health and outlook on life (one question each)
    • Patient's appetite and pleasure of eating. Each item was scored categorically (zero, one or two). Each patient also completed a 100-mm linear analog scale (LAS) in answer to the question, How good is your quality of life? The answers on the LAS ranged from zero (extremely poor) to 100 (excellent).
• Survival rate.

Independent Variables

• Age (years)
• Sex
• Stage cancer (local:extensive)
• Primary tumor site
• Percent weight loss in prior three months.

Control Variables

Dietary counseling by a trained dietitian vs. no dietary counseling by a trained dietitian.

Initial N

137.

Characteristics of 57 Patients Who Received Dietary Counseling and 48 Control Patients

 *Presence of distant metastases (i.e., tumor-made metastasis classification of M) in breast and ovarian cancer patients and disease outside one hemitharax in small-cell lung cancer patients).

Attrition (Final N)

• 105
• 32 patients were not used in the analysis due to:
  • Early death or deteriorating functional status (N=13)
  • Rejected further participation within two months (N=11)
  • Delivery of insufficient number of dietary records (N=5)
  • Stopped protocol chemotherapeutic treatment (N=3).

Ethnicity

Not identified.

Location

Copenhagen, Denmark.

Changes From Initial (Preceding the Onset of Therapy) Energy and Protein Intakes In the Counseled (N=57) and Control (N=48) Groups

[Note: Values are mean ±SD]
*P<0.05 between months
§P<0.05 between groups.

Changes From Initial (Preceding the Onset of Therapy) Anthropometric Values in Counseled (N=57) and Control (N=48) Groups

[Note: Values are mean ±SD.]
*P<0.05 between months

Overall tumor response rate was not significantly different in the Counseled and Control Groups.

Patient Responses in the Counseled (N=58) and Control (N=47) Groups After Three Months of Chemotherapy

*P=0.83 difference between groups
¶P=0.11 difference between groups

[Note: Values are mean ±SD.]
*P<0.05 between months

Other Findings

There were no significant differences in the overall survival rate in the Counseled Group, compared with the Control Group.

The cumulative proportion surviving were:

Analysis by Nutritional Status

• Subset of patients nutritionally depleted (greater than 5% weight loss) at baseline
• No benefit from MNT above control
• Similar intake to weight-stable patients
• Higher QOL tools in weight-stable patients.

Frequent and professional dietary counseling targeted to recommended intake levels resulted in increased food intake during the five months' time-frame of the study in patients undergoing chemotherapy. However, no effect on weight, response rate, survival duration or quality of life could be demonstrated.

• The researchers appeared to adhere to the protocols for their study design. The longitudinal design focus on a population or sub-group, some members of which are exposed to or experience certain events over time, while others will not. Data are collected at more than one point in time and the reference period is prospective. The researchers look to the future in describing and explaining the occurrence of the characteristics of interest. 
• The author did not note the possibility of "recall basis" in relationship to the patients' report of weight prior to the beginning of the research study. It was noted that the patient's recalled weight, at least three months prior to the cancer diagnosis, was used to assess weight loss.
• Even after attrition, the number of patients remained very close to the parameters established (57 Counseled and 48 Control) in the statistical analysis. The researchers determined that the chance of detecting differences in weight of five kilograms, considered to be clinically significant, required at a level of significance (two-sided) of 0.05 and a power of 0.8, would require approximately 50 patients in each group.