Saikrishna Yendamuri

M.D

Roswell Park Comprehensive Cancer Center

Professor and Chair of Thoracic Surgery

NLST (Learn more about this study)

NLST-669

May 12, 2020

Visceral Obesity is associated with adverse lung cancer outcomes

New treatment strategies for lung cancer are urgently needed. We are currently in the midst of a worldwide obesity epidemic, and contrary to common perception, ~65% of lung cancer patients are either overweight or obese (BMI>25. Obesity has been linked to increased cancer incidence across thirteen organ sites, but not in lung cancer. While obesity promotes lung cancer growth in animal models, human obesity, defined by a high BMI, is associated with decreased incidence of lung cancer as well as better oncological outcomes after resection of early lung cancer. Our recent investigations seem to unravel this “Obesity Paradox”. Mounting evidence shows that the adverse health outcomes of obesity are driven by visceral and not subcutaneous obesity – a distinct fat distribution phenotype that is not discernable by BMI alone. In order to assess the effect of visceral obesity on lung cancer outcomes, we measured the visceral fat area at a fixed level of the abdomen using CT scans performed pre-operatively (N=375 stage I NSCLC patients). In order to adjust for body size, we normalized this measurement to the sum of the visceral fat area (VFA) and subcutaneous fat area (SFA) measured at the same level. This ratio is termed the visceral fat index (VFI = VFA / SFA+VFA). We observe that multivariable analyses show that high VFI patients have lower overall survival (OS) as well as recurrence free survival (RFS) when compared to low VFI patients. BMI did not show any such association with survival. These observations show that the obesity paradox in lung cancer may be due to the confounding effects of the use of BMI as a measure of obesity. In this project, we seek to validate our findings using imaging and the corresponding clinical data from the NLST.
Patients with diagnosis of lung cancer with a CT scan available within 6 months from diagnosis will be included. Other inclusion criteria include patients with stage I or II non-small cell lung cancer undergoing surgery without neoadjuvant therapy. In this aim, we will measure the VFI of these patients at the L1 level (which is covered in all chest CT scans) as a measure of visceral adiposity and assess the association of VFI with overall survival, disease-specific survival as well as progression free survival over the duration of follow up, while controlling for all relevant clinical covariates. Histology will be classified into adenocarcinoma, squamous cell carcinoma and others. Race with be classified into black, white and others. Smoking status will be classified into former and current. Grade will be classified into I/II (low), III/IV (high) and unknown. Surgery will be classified into wedge / lobectomy and pneumonectomy. The association between survival measures and VFI as a continuous variable will be first analyzed in univariate and multivariable analyses. The patients will also be grouped into tertiles and the top and bottom tertiles will be compared. Analyses will also be performed to exploratorily define VFI cutoffs to maximize separation of survival measures. Such cutoffs may help inform future studies.