A total of 415 posters were presented at the European Lung Cancer Congress (ELCC) at the end of March, one of them from Medaffcon. The poster showcased key findings from a recent study about non-small cell lung cancer (NSCLC).
Medaffcon’s poster was presented by Ville Koistinen, Chief Pulmonologist from Kotka. Scientific Advisor Riikka Mattila represented Medaffcon at the congress.
Despite advancements in treatment, non-small cell lung cancer (NSCLC) remains one of the leading causes of cancer-related mortality worldwide. Understanding the factors that influence survival outcomes is crucial for optimizing treatment strategies and improving patient prognoses.
Medaffcon used high-quality real-world clinical data to develop a robust machine learning-based survival prediction model. Their findings demonstrate that the model achieves excellent predictive performance at specific follow-up time points as well as identifies key factors contributing to the prognoses of NSCLC patients treated with pharmacological lung cancer therapies.
The study included 1,085 adult NSCLC patients treated at HUS, Helsinki University Hospital, between 2018 and 2023. Survival time was calculated as the number of months from the initiation of first-line treatment to the death or the end of the study. Smoking status was extracted from the patient records using a natural language processing approach. A Random Survival Forest model was trained on the data following hyperparameter tuning, and its performance was evaluated using the area under the curve (AUC) at six specific follow-up time points. Variable importance with 95% confidence intervals was extracted from the model to rank predictive features and illustrate how the importance varies over time.
Overall survival was computed by fitting Kaplan Meir curves for the overall treated cohort, stratified by the covariates included in the prediction model.
According to Mattila, the poster especially caught the attention of Finnish pulmonologists, as it focused on their own patient population.
“In overall, the congress made it clear that lung cancer is leading the way in personalized medicine. Genetic profiling and biomarker use are more prevalent in lung cancer than in any other cancer type. New therapies are constantly being developed. The treatment landscape is evolving rapidly—especially for non-small cell lung cancer,” Mattila says.
According to Mattila, it is important to stay up to date on how new developments in biomarkers and treatments are reflected in real-world patient care. Medaffcon helps ensure that the understanding of these matters remains current.
The majority of ELCC (European Lung Cancer Congress) attendees are clinicians, including medical oncologists, radiation oncologists, thoracic surgeons, and respiratory physicians. Some industry representatives also attend.
This year, ELCC was held in Paris, France, with participants from over 90 countries—4,190 in total. Next year, ELCC will take place in Copenhagen, Denmark.
Medaffcon, founded in 2009, is a Nordic research and consulting company specializing in Real-World Evidence, Medical Affairs, and Market Access. With offices in Stockholm, Sweden, and Espoo, Finland, we provide expert services across the Nordic region. Our services combine strong medical and health economic expertise with modern data science.
The company employs some 30 experts. Since 2017, Medaffcon has been a subsidiary of Tamro Oyj and is part of the PHOENIX group, which is a leading provider of healthcare services in Europe.
The data team keeps Medaffcon's research projects on track and ensures that the research findings are scientifically sound. At the heart of the team’s work is the processing and analysis of patient data, particularly in Real-World Evidence (RWE) studies.
Johan Rehnberg started working as a Scientific Advisor at Medaffcon’s Swedish office in August 2024. He is a dynamic researcher who values opportunities to learn new things and develop his skills – opportunities that Medaffcon provides.
The algorithm was originally developed to extract smoking status from patient texts with purpose to analyze the effects of smoking on postoperative complications. Today, it is also being utilized in lung cancer research.
Scientific Advisor
PhD
Riikka joined Medaffcon in February 2021. She has a wide-ranging experience from different therapy areas from atherosclerosis and birth asphyxia to neurodegeneration. She has more than 15 years of experience in research, three of which in Max Planck Institute in Germany. Her PhD thesis from 2011 focused on cholesterol metabolism.
Riikka’s strenghts include broad know-how and interest in diverse therapy areas, as well as enthusiasm and experience in both written and verbal scientific communication. At Medaffcon she enjoys varied projects and effectiveness of research.
Real world evidence fascinates Riikka because there is so much data, and more accumulating all the time, and most of this data is unused. There is potential for findings to support clinicians and the pharmaceutical industry, as well as alleviate the lives of patients. She is also happy that decisions in health care are increasingly evidence based.