New method helps in the diagnosis of lung cancer – 07/25/2023 – Health

Cachexia is a complex metabolic syndrome characterized by severe loss of weight and muscle mass and is especially harmful in the case of non-small cell lung cancer, affecting approximately half of patients. Its early detection is important to predict the prognosis and guide better treatment decisions. The condition can lead to muscle weakness and, consequently, make it difficult to carry out simple day-to-day activities, which makes it more difficult to deal with the side effects of the treatment – ​​and with the disease itself. In the case of patients with lung cancer, it is even more harmful, as it worsens respiratory function.

Sarah Santiloni Cury, a postdoctoral fellow at the Institute of Biosciences at the Universidade Estadual Paulista (Unesp) in Botucatu, is studying ways to diagnose cachexia early and is one of the authors of an article published in the Journal of Translational Medicine on a new method for predicting this syndrome.

His research was awarded by the European Organization of Molecular Biology and the Federation of European Biochemical Societies, in an event that focused on artificial intelligence and machine learning approaches in cancer research.

“Multiple screening tools are used to measure the loss of muscle mass, one of them is the use of computed tomography images in the region of the third lumbar vertebra (L3). However, the images of patients with lung cancer usually do not include the L3”, he warns.

As a way to circumvent this limitation, the study showed that analyzing the pectoral muscle area also serves as a prognostic parameter, as the measurements showed an association with the clinical outcome of the disease. In this way, it was possible to establish reference values ​​based on the pectoral muscle area, something that had not yet been determined. Based on muscle dimensions, clinical data and tumor microenvironment profile, a patient classification model was constructed.

The research group used machine learning to generate a muscle loss prediction model. Initially, measurements of the pectoral muscle area of ​​211 patients with lung cancer were obtained from computed tomography scans publicly available at The Cancer Imaging Archive (TCIA) repository.

Subsequently, cutoff points were established using algorithms (CART and Cutoff Finder) on clinical, survival, and pectoral muscle area data. “We evaluated the effectiveness of our model in a validation set, which included 36 patients treated at the Faculty of Medicine of Unesp in Botucatu”, highlights Robson Francisco Carvalho, from the Biosciences Institute of Unesp in Botucatu, Cury’s advisor and co-author of the article.

“This study represents an important advance in the understanding of cachexia in patients with lung cancer, as it identified new potential mediators and biomarkers of the syndrome, using imaging tests and molecular analyzes for early diagnosis. Based on these discoveries, it will be possible to develop research on therapeutic strategies and offer better follow-up to patients”, explains Carvalho, a researcher supported by FAPESP in studies on cachexia.

Previously, the group had already demonstrated the relationship between the presence of certain tumor biomarkers and the risk of developing cachexia, in work also supported by the foundation. “We found that, among the tumor types that commonly induce cachexia, those of lung cancer show an increase in the expression of specific factors that contribute to muscle loss. Some of these factors act on receptors on the cell surface of muscle tissue cells, contributing to their loss”, points out Carvalho.

RNA sequencing data from the tumors revealed 90 affected genes in patients with low musculature, which are potentially secreted and interact with muscle cell receptors. They also allowed the identification of cells in the tumor microenvironment responsible for the secretion of cachexia-inducing factors.

These analyzes revealed that patients with low musculature had, for example, high proportions of a specific type of CD8+ T lymphocytes. While these cells are often associated with intense anticancer activity, in this case the scientists noted that they may be associated with a worse prognosis. Other researchers had already shown that CD8+ T cells induce adipose tissue loss in cachexia associated with chronic infection, but this association had not yet been made for lung cancer.

Thus, in addition to establishing parameters for predicting cachexia, the work also made it possible to identify cells that may be related to the syndrome, which opens up the possibility for new therapies. The tumor microenvironment, however, is complex, the researchers point out, and further studies are needed.

The work is also signed by Diogo de Moraes, Jakeline Santos Oliveira, Paula Paccielli Freire, Patricia Pintor dos Reis, Miguel Luiz Batista Jr and Érica Nishida Hasimoto and involved the University of São Paulo (USP), the State University of Campinas (Unicamp) and the Boston University School of Medicine.

The article Low muscle mass in lung cancer is associated with an inflammatory and immunosuppressive tumor microenvironment can be read here.