Evaluation of a protocol to detect malnutrition and provide nutritional care for cancer patients undergoing chemotherapy

Cancer is highly prevalent and a major cause of morbidity and mortality worldwide. With improved survival due to early diagnosis and more effective treatments, oncological care is shifting toward a multidisciplinary model that addresses overall patient well-being, including nutritional status. Weight loss and malnutrition are common in cancer and are associated with worse outcomes and quality of life. Guidelines (e.g., ESPEN) emphasize early detection and prompt nutritional action to prevent or mitigate cachexia, which can become refractory. Prior studies suggest that early nutritional intervention reduces catabolism, improves clinical outcomes, and may enhance survival, particularly in high-risk cancers (e.g., oesophageal, gastric). In 2016, the authors’ hospital designed and implemented a nutritional care model aiming for early detection of nutritional risk, periodic assessment and monitoring, and timely intervention before refractory cachexia, to improve weight maintenance or gain during chemotherapy.

Design: Prospective study implementing a nutritional care model for adult outpatients (≥18 years) with de novo diagnosed solid tumours (any stage) starting chemotherapy. Patients with cultural/cognitive barriers to understanding study aims were excluded. Consecutive inclusion was used. Ethics approval obtained; written informed consent provided by all participants. Risk stratification and workflow: Two groups by tumour-related nutritional risk: - Group 1 (high risk): Head and neck; upper digestive tract (oesophagus, stomach); pancreas; bile ducts. Direct referral to nutrition consultation, usually within 3 days of Tumour Committee review or during diagnostic admission. Initiation of an Individualised Nutritional Care Programme (assessment, intervention, vigilant follow-up) per ESPEN guidelines, ranging from counseling to specialized nutrition support based on malnutrition severity. - Group 2 (low risk): All other solid tumours (e.g., colon, lung, breast, gynaecological). Underwent Nutriscore screening at first oncologist or outpatient pharmacy visit before chemotherapy. Patients with Nutriscore ≥5 underwent PG-SGA classification (A: well-nourished; B: moderate risk/malnutrition; C: severe malnutrition). Those classed B/C were referred to the nutrition clinic and started on the Individualised Nutritional Care Programme. Those with Nutriscore <5 received periodic re-evaluation with weight checks at each chemotherapy cycle and during perioperative periods; nutritional recommendations were available to all on request. Operational steps (Table 1): 1) Tumour Committee classifies tumour type; Step 2 assigns high-risk tumours to Group 1; Step 3 assigns others to Group 2. - Group 1: Rapid nutritional approach; individualized plan and follow-up. - Group 2: Nutriscore screening at treatment start. If Nutriscore ≥5: PG-SGA assessment; well-nourished patients get weight control each cycle; PG-SGA B/C referred to nutrition clinic. If Nutriscore <5: reassessment each cycle. Monitoring and visit cadence: Individualized monitoring plan with periodic reviews (most commonly Day 1 of each chemotherapy cycle) to assess adherence, tolerance, and efficacy. Average of five nutrition consultations per patient; mean follow-up 6 months. Screening/instruments: Nutriscore (0–9 points; considers 3-month involuntary weight loss, decreased appetite, tumour location, oncology treatment); nutritional risk defined as Nutriscore ≥5. Nutriscore validated in Spanish population against PG-SGA and MST (sensitivity 97.3%, specificity 95.9%). PG-SGA used for malnutrition classification (A/B/C). Weight definitions: - Normal weight (NW): Patient-reported weight during last 3 months. - Initial weight (IW): Weight at first application of the nutritional care model; expressed as % weight loss relative to NW. - Chemotherapy start weight (CSW): Weight at start of chemotherapy. - Chemotherapy end weight (CEW): Weight at end of chemotherapy. Effectiveness endpoint: % weight gained or maintained at CEW compared with IW. Clinically relevant weight loss (WL) defined as ≥2%. Timing definition: Early application defined as nutrition care initiated before chemotherapy or within 7 days after starting. Statistical analysis: Descriptive statistics for quantitative (central tendency, dispersion, position) and qualitative (frequency distribution) variables. Wilcoxon rank test and Mann-Whitney U for paired and independent quantitative comparisons; chi-square test for categorical comparisons; p<0.05 considered significant. Sample size: assuming ~300 eligible annually and 80% weight gain/maintenance post-protocol, for 95% confidence and 3% precision, n=209 required; inflated by 10% to 230 to allow for losses. A multivariate logistic regression modeled end-of-treatment outcome (WL>2% vs. weight gain/maintenance) with tumour type, age, sex, and treatment intention as covariates (p<0.05 significance). Ethics: Approved by the Hospital Costa del Sol Medical Ethics Committee. Consent to participate obtained; consent for publication not applicable.

- Cohort: 295 initially enrolled; 51 (17%) did not complete chemotherapy and 10 were lost to follow-up, leaving 234 for analysis. - Need for individualized nutrition care: 84/234 (36%) required individualized nutritional intervention and vigilant follow-up. - Of these 84: 27 (32.1%) were Group 1 (high-risk tumours); 12 had Nutriscore ≥5 at treatment start (the 27 high-risk also had Nutriscore ≥5); 45 were initially not at risk but developed WL during chemotherapy and were captured through periodic re-evaluations. - Baseline characteristics (all patients vs. those needing care): Mean age 59±11 vs. 60±11 years; female 41.9% vs. 60.7%; BMI 26.6±4.8 vs. 24.5±4.0; initial WL (NW–CSW) 3.9%±7.3 vs. 10.1%±7.4; Nutriscore ≥5: 16.7% overall vs. 46.4% among those needing care. - Nutritional support: 68% of those in individualized care required enteral nutrition (most common formulas: hypercaloric/high-protein with/without fiber and diet formulas). - Timing: Median 9 days (IQR 9) from diagnosis to first nutrition consultation; median 26 days (IQR 54) from first nutrition consultation/program initiation to start of chemotherapy. For the subgroup initially not at risk but later referred (n=45), nutrition consultation occurred on average 43 days (IQR 85) after chemotherapy start. Outcomes: - Overall effectiveness: Mean %WL at inclusion in the programme (relative to NW) was −3.6%±8.2. By end of chemotherapy, mean change was 0%±7.3; difference significant (p<0.001). Overall, 71.0% achieved weight gain or maintenance by the end of chemotherapy (vs. initial weight). - Group 1 and Group 2 with Nutriscore ≥5 and PG-SGA B/C (n=39): Median %WL 9.2% (IQR 10.8) vs. NW at first nutrition visit; after individualized care, median %WL vs. IW at chemotherapy start was 0% (IQR 1.6). By end of chemotherapy, 58% gained or maintained weight (vs. IW). - Group 2 initially not at risk, later with Nutriscore ≥5 and PG-SGA B/C (n=45): Median %WL at start of chemotherapy 6.1% (IQR 11.6) vs. NW; during treatment WL increased to 12.2% (IQR 11.4) prompting referral. Despite later initiation, WL slowed substantially; by end, WL reduced to 0.68% relative to usual weight at protocol start; 54% gained or maintained weight; differences significant (p<0.05). - By tumour site (median %WL start vs. median %WG/WM end; p-value): - Head-neck (n=5): −16.3 (IQR 18.1) to 1.7 (15.6); p=0.080. - Oesophagus-stomach (n=13): −7.8 (11.5) to −2.9 (10.6); p=0.101. - Pancreas-bile ducts (n=9): −7.9 (14.5) to −5.6 (12.5); p=0.678. - Colorectal (n=37): −6.1 (9.7) to 0.0 (6.7); p=0.002. - Gynaecological (n=32): −5.2 (2.2) to 1.7 (3.9); p=0.003. - Lung (n=45): −0.6 (15.5) to 0.0 (7.3); p=0.044. - Other (n=9): −3.8 (7) to 0.0 (16.8); p=0.028. - Breast (n=72): 0.0 (5.8) to 0.0 (4.8); p=0.226. - Bladder (n=12): −2.0 (5.8) to 1.0 (4.1); p=0.308. - Logistic regression: No significant associations between outcome (WL>2% vs. gain/maintenance) and tumour type, age, sex, or treatment intention. Interpretation: More than one-third of patients initiating chemotherapy were candidates for early nutritional intervention. Early identification and individualized care led to high rates of weight maintenance or gain, with significant improvements overall and in several tumour groups.

The study demonstrates that a structured, early nutritional care model incorporating screening (Nutriscore), assessment (PG-SGA), tailored intervention, and vigilant follow-up effectively identifies cancer outpatients at risk of malnutrition and favorably influences weight trajectories during chemotherapy. Addressing the research aim, 36% of patients beginning chemotherapy required individualized nutritional care, and 71% of the entire cohort achieved weight maintenance or gain by the end of treatment. Early referral of high-risk groups and ongoing monitoring in lower-risk groups captured additional patients who developed weight loss during therapy, underscoring the necessity of continuous surveillance, including the role of outpatient pharmacy encounters at each chemotherapy cycle. Findings align with ESPEN recommendations and prior literature indicating that early nutritional intervention mitigates cachexia-related catabolism and improves clinical outcomes. While high-risk gastrointestinal sites (e.g., oesophageal-gastric, pancreatic) did not show significant weight gains, many achieved stabilization, suggesting that earlier or more intensive interventions may be needed for these cancers. The lack of significant predictors in multivariable analysis implies broad applicability of the protocol across demographics and treatment intents. Overall, the model operationalizes multidisciplinary, proactive nutritional support, contributing to improved tolerance and potentially better quality of life during chemotherapy.

This outpatient-focused, systematized, and individualized nutritional care model enables early detection and management of malnutrition risk across tumour locations. By initiating nutritional assessment and intervention before or early in chemotherapy and maintaining vigilant follow-up each cycle, 71% of patients achieved weight maintenance or gain, indicating effective mitigation of treatment-associated nutritional decline. The protocol has been incorporated into routine practice at the authors’ institution. Future research should include controlled comparisons, evaluate intensified strategies (e.g., home parenteral nutrition) for high-risk gastrointestinal cancers, and assess impacts on clinical endpoints such as treatment tolerance, hospitalizations, and survival.

- No randomized or concurrent control group due to ethical constraints, limiting causal inference; comparisons are primarily to historical data. - Heterogeneous tumour population complicates comparison with literature that often focuses on single tumour types, and may dilute effects in specific subgroups. - Some delays in initiating nutritional intervention for patients who developed risk during treatment (e.g., average 43 days post-chemotherapy start for later referrals) could affect outcomes. - Reliance on patient-reported normal weight may introduce recall bias.