Protective effect of obesity on survival in cancers treated with immunotherapy vanishes when controlling for type of cancer, weight loss and reduced skeletal muscle. (January 2023)

• Record Type: Journal Article
• Authors: Antoun, Sami; Lanoy, Emilie; Ammari, Samy; Farhane, Siham; Martin, Lisa; Robert, Caroline; Planchard, David; Routier, Emilie; Voisin, Anne Laure; Messayke, Sabine; Champiat, Stephane; Michot, Jean Marie; Laghouati, Salim; Lambotte, Olivier; Marabelle, Aurélien; Baracos, Vickie
• Is Part Of: European journal of cancer. Volume 178(2023)
• Page Start: 49
• Page End: 59
• Publication Date: 2023-01
• Subjects: Obesity paradox -- Immune checkpoint inhibitors -- Weight loss -- Reduced skeletal muscle -- Melanoma -- Non-small cell lung cancer; Cancer -- Periodicals; Neoplasms -- Periodicals; Cancer -- Périodiques; Cancer; Tumors; Electronic journals; Periodicals; Electronic journals; 616.994
• DOI: 10.1016/j.ejca.2022.10.013
• Languages: English
• ISSNs: 0959-8049

Abstract

Abstract: Introduction: Association of high body mass index (BMI) with longer survival has been reported in patients on immune checkpoint inhibitors (ICIs), but results are inconsistent. This 'obesity paradox' is potentially confounded by the effects of BMI change over time and of skeletal muscle depletion. Methods: We conducted a secondary analysis of a prospective cohort, including consecutive patients receiving ICI treatment for melanoma (n = 411) and non-small cell lung cancer (NSCLC) (n = 389) in routine care. Results: In the univariable analysis of the entire population, overweight/obesity (BMI ≥ 25 kg/m 2 ) was associated with longer survival (p < 0.01); however, this effect was limited to NSCLC (p < 0.01) and was absent in melanoma. Weight loss (WL) and reduced skeletal muscle mass were observed in patients within all BMI categories. WL was associated with shorter survival in multivariable analysis in both tumour sites (p < 0.01), and for NSCLC, BMI lost significance when WL was included (p = 0.13). In models further adjusted for CT-defined skeletal muscle mass, WL retained significance for both tumour types (p < 0.01), and reduced skeletal muscle only for NSCLC (p = 0.02) was associated with shorter survival. WL retained significance when biomarkers (lactate dehydrogenase enzyme, albumin and derived neutrophil to lymphocyte ratio) were added to the multivariable model. Conclusions: The so-called 'obesity paradox', counterintuitive association between high BMI and longer survival, vanished when controlling for confounders, such as type of cancer, and manifestations of depletion (WL and reduced skeletal muscle mass). Highlights: Association of high BMI with survival is eliminated when adjusted for confounders. Association of high BMI with survival disappears when adjusted for weight loss. Weight loss in NSCLC and melanoma is the predominant prognostic factor. In patients with NSCLC, reduced skeletal muscle is an independent prognostic factor.