Purpose: Cancer-associated weight loss is associated with poor prognosis in advanced malignancy; however, its pretreatment prevalence and survival impact are inadequately described in large cohorts. Such data, stratified by tumor type and stage, may facilitate the optimal and timely allocation of complementary care, leading to improvements in patient survival and quality of life.
Methods: We performed a retrospective cohort study of 3,180 consecutively treated adult patients with lung or GI (including colorectal, liver, and pancreatic) cancer. Pretreatment cancer-associated weight loss was based on the international consensus definition of cachexia. Prevalence and survival impact of pretreatment cancer-associated weight loss were evaluated using the Kaplan-Meier method and compared using log-rank test.
Results: Cancer-associated weight loss was observed at the time of cancer diagnosis in 34.1% of patients. Pretreatment weight loss was documented in 17.6%, 25.8%, 36.6%, and 43.3% of stage I, II, III, and IV cancers, respectively. Wasting was common regardless of tumor type, with prevalence at diagnosis ranging from 27.3% in patients with colorectal cancer to 53.4% in patients with gastroesophageal cancer. Pretreatment weight loss was associated with reduced overall survival after adjusting for stage, size, grade, comorbidity, age, sex, and tobacco history (hazard ratio, 1.26; 95% CI, 1.13 to 1.39).
Conclusion: Pretreatment cancer-associated weight loss is common, even in early-stage disease, and is independently associated with reduced survival. Minimal weight loss represents a clinically distinct entity with an associated overall survival intermediate to that of no weight loss and overt wasting. Early diagnosis and treatment of cancer-associated wasting offers a novel therapeutic avenue for reducing cancer mortality.