Abstract Rationale Unintentional weight loss and malnutrition are common among cancer patients. Malnutrition has been associated with impaired health-related quality of life, less well-tolerated chemotherapy regimens and shorter life duration. In Belgium there is a lack of epidemiological data on malnutrition in oncology patients at advanced stages of the disease. Methods Malnutrition assessment data was collected through a prospective, observational study in 328 patients who started a neoadjuvant anticancer therapy regimen or who started 1st, 2nd or 3rd line anticancer therapy for a metastatic cancer via 3 visits according to regular clinical practice (baseline visit (BV) maximum 4 weeks before start therapy, 1st Follow up visit (FUV1) ± 6 weeks after start therapy, FUV2 ± 4 months after start therapy). Malnutrition screening was evaluated using the Nutritional Risk Screening score 2002 (NRS-2002)and the diagnosis of malnutrition by the GLIM criteria. In addition, SARC-F questionnaire and Fearon criteria were used respectively to screen for sarcopenia and cachexia. Results Prevalence of malnutrition risk at BV was high: 54.5% of the patients had a NRS ≥ 3 (NRS 2002) and increased during the study period (FUV1: 73.2%, FUV2: 70.1%). Prevalence of malnutrition based on physician subjective assessment (PSA) remained stable over the study period but was much lower compared to NRS results (14.0%—16.5%). At BV, only 10% of the patients got a nutrition plan and 43.9% received ≤ 70% of nutritional needs, percentage increased during FU period (FUV1: 68.4%, FUV2: 67.6%). Prevalence of sarcopenia and cachexia were respectively 12.4% and 38.1% at BV and without significant variation during the study period, but higher than assessed by PSA (11.6% and 6.7% respectively). Figures were also higher compared to PSA. There were modifications in cancer treatment at FUV1 (25.2%) and at FUV2 (50.8%). The main reasons for these modifications at FUV1 were adverse events and tolerability. Patient reported daily questionnaires of food ...