Weight loss in cancer patients: A plea for a better awareness of the issue

Luigi Mariani, Salvatore Lo Vullo, Federico Bozzetti

Research output: Contribution to journalArticlepeer-review

Abstract

Purpose: Although weight loss is of both prognostic and predictive relevance in oncologic patients, its assessment is often neglected. Aims of the present investigation were to define the prevalence and severity of weight loss in adult outpatients with a variety of solid tumors, and determine the association patterns with patient-, cancer-, and therapy-related factors. Methods: Among an outpatient series of 1,556 cancer patients, weight loss information was obtained for 1,540 patients. Weight loss was analyzed by means of multiple regression models, logistic models, and nomograms, according to age, gender, site of primary, UICC stage, Eastern Cooperative Oncology Group (ECOG) performance status, therapy, and symptoms type and degree. Results: Weight loss, relative to usual body weight, was 7.1% on average in the whole series, and clinically significant (≥10%) in 589 patients (38%). Factors most strongly associated with WL were site of primary, ECOG performance status, anorexia syndrome, and fatigue. These, together with oncologic therapy, were important factors for predicting significant weight loss. Conclusions: Weight loss turned out to be frequent and clinically significant. We believe that this sign should deserve major attention by the oncologists to pursue the benefits that early nutritional support prospectively yields in terms of quality of life and clinical outcome improvement.

Original languageEnglish
Pages (from-to)301-309
Number of pages9
JournalSupportive Care in Cancer
Volume20
Issue number2
DOIs
Publication statusPublished - Feb 2012

Keywords

  • Cancer cachexia
  • Neoplasm
  • Nomogram
  • Nutritional support
  • Weight loss

ASJC Scopus subject areas

  • Oncology

Fingerprint Dive into the research topics of 'Weight loss in cancer patients: A plea for a better awareness of the issue'. Together they form a unique fingerprint.

Cite this