Early in-depth symptomatic investigations could have major impact on overall lung cancer mortality
Lung cancer remains the leading cause of cancer-related deaths globally, and this is often due to late detection, leading to the inability to receive curative treatment. Researchers from KI, Karolinska University Hospital and Umeå University have made strides in identifying pre-diagnostic symptoms and sensations of significance for the identification of primary lung cancer.
Hi there Adrian Levitsky, first author of an article recently published in Scientific Reports entitled “Early symptoms and sensations as predictors of lung cancer: a machine learning multivariate model”. As part of your signature postdoc work, what are the most important results of your study?
“We have discovered a set of seven background variables and 63 early symptoms and sensations that may help to identify patients needing further referral for lung cancer investigation. The most significant symptoms or sensations associated with lung cancer included a cough that varied over the day, back pain/aches/discomfort, early satiety, appetite loss, and having less strength.”
How can this new knowledge contribute to improving human health?
“This new knowledge of early risk symptoms and sensations may lead to earlier flagging to refer a patient for specialized diagnostic workup. This could potentially reduce the death toll from lung cancer, which remains the leading cause of cancer-related deaths globally. This death toll is in large part due to late detection, as it has proven difficult to identify signs of the disease early. Thus, particularly when the patient is diagnosed in the earliest stages of disease, the chances of curative treatment is more likely and survival is far better”
How did you conduct your study?
“This study was made possible through hundreds of patients who were referred to lung specialists at the Karolinska University Hospital. Before their specialist visit, they would fill out an interactive tablet e-questionnaire (The Patient EXperience of Bodily Changes for Lung Cancer Investigation, PEX-LC) developed by our research team; and which allowed them to express their unique symptomatic profiles. The e-questionnaire included a vast array of 285 possible early descriptors and was divided up into different modules that the patient would fill in only if relevant for them.”
Where do you go from here?
“With further research and precision in the set of symptoms and sensations we identified, including testing the e-questionnaire in a more general audience, a tool could later be developed to enable earlier referral and lung cancer diagnostic decision-making.”
The study is led by Associate Professor Lars E. Eriksson at the Department of Learning, Informatics, Management and Ethics and is a collaboration between researchers at Karolinska Institutet, Karolinska University Hospital and Umeå University. The study was funded by the Vårdal Foundation, the Swedish Research Council (Vetenskapsrådet), and the Strategic Research Area Health Care Science (SFO-V). Development of the PEX-LC e-questionnaire was made possible in collaboration with patient partners, healthcare professionals, and consultants in the industry.
Early symptoms and sensations as predictors of lung cancer: a machine learning multivariate model
Levitsky A, Pernemalm M, Bernhardson BM, Forshed J, Kölbeck K, Olin M, Henriksson R, Lehtiö J, Tishelman C, Eriksson LE
Scientific Reports, 11 November 2019, s41598-019-52915-x