The early detection of lung cancer and the European SOLACE project in particular were the focus of an ECR session attended by experts from five countries. SOLACE aims to promote the structured and widespread introduction of lung cancer screening using low-dose CT in Europe.
Joanna Chorostowska-Wynimko from Warsaw, Poland, gave a brief introduction to the project: Launched in 2023, it now involves 37 partners from 15 EU member states and is conducting 13 pilot studies in 11 countries. It has two main objectives: Firstly, to establish a knowledge center and develop guidelines for mapping and evaluating ongoing screening initiatives in Europe, and secondly, to implement pilot projects that address hard-to-reach target groups, such as socially disadvantaged people, ethnic minorities or people with occupational exposure to harmful substances. A further aim of SOLACE is to create a standardized guideline recognized by the EU, based on over 250 existing screening guidelines and the GRADE methodology for scientific evaluation. In the long term, the guidelines and screening programs developed as part of SOLACE should become official EU standards and be applicable throughout Europe.
Comparison with the HANSE study
In his presentation, Jens Vogel-Claussen from Hanover, Germany, drew a comparison with the HANSE study, which served as the basis for SOLACE. The study aimed to recruit over 5,000 high-risk individuals and compare two methods of risk assessment – the PLCO risk score and the NELSON criteria. ‘The results show that the PLCO score was more efficient and identified more lung cancer cases than the Nelson criteria, while at the same time the number of screenings required to detect a cancer case was lower,’ says Vogel-Claussen.
Another key element was the digital infrastructure, which enabled automated recruitment and appointment allocation. Participants could register online, by telephone, or on site, with a system automatically identifying those at high risk. The CT scans were supported by the use of artificial intelligence (AI), which helped to assess nodules, calculate volumes, and determine the lung RADS score. As part of SOLACE, this approach was extended to target women and socially disadvantaged groups. The combination of automated processes and target group-specific measures led to a significant increase in the participation of women and socially disadvantaged people.
Assessment of pulmonary nodules
The presentation by Mario Silva from Parma, Italy, focused on the precise evaluation of pulmonary nodules and risk assessment. ‘Persistent, partly solid nodules have the highest risk of malignant changes, while solid nodules occur most frequently,’ explained Silva. ‘In addition to the size and density of the nodules, morphological characteristics also play an important role in risk assessment.’ Another key criterion is the volume doubling time (VDT). ‘Nodules with a VDT of less than 400 days are considered suspicious,’ says Silva.
However, analyzing screening scans is also an everyday challenge for radiologists, as Anna Kerpel-Fronius from Budapest, Hungary, explained in her presentation. ‘The overwhelming majority of scans are unremarkable. In the NELSON trial, only 119 out of 7,500 scans were assessed as positive, while 1,400 were considered unclear, and the rest showed no abnormalities. Radiologists, therefore, spend the majority of their working time assessing healthy lungs, which requires an enormous amount of concentration,’ emphasized Kerpel-Fronius. She emphasized the use of computer-aided detection (CAD) to identify smaller nodules and their growth. ‘However, CAD reaches its limits with very large tumors or complex accompanying findings,’ the expert noted.
Increasing workload
A survey on the workload of radiologists and other specialists revealed that 66.4% of respondents perceived an increase in workload as a result of screening. Nevertheless, 80 % did not feel overwhelmed. ‘This could be because structured reporting during screening leads to more routine and thus increases diagnostic certainty,’ surmised Viktoria Palm from Heidelberg, Germany. ‘Lung cancer screening contributes to the increasing workload as it is performed in addition to regular diagnostic CT scans. It is particularly problematic that the time windows for screening CTs are often shorter than for oncological CTs, which increases the pressure on staff,’ said the expert.
Training is essential
Helmut Prosch from Vienna, Austria, emphasized that training is essential for all professionals involved in screening. ‘Lung cancer screening harbors potential risks such as false alarms, overdiagnosis, overtreatment, radiation exposure and high costs. To minimize these risks, specialist associations have developed guidelines to optimize screening,’ explained Prosch. Nevertheless, there are still challenges: Many doctors do not understand the screening data correctly or misinterpret guidelines, which leads to low participation in screening programs. Prosch: ‘A survey conducted in 21 countries in 2023 showed that only four national specialist societies have so far introduced mandatory training programs for lung cancer screening, namely the Czech Republic, France, Italy, and Portugal.’ The European ‘You Can Scan EU’ project offers web-based and practice-oriented training, Spain is planning a master’s degree in lung cancer screening, and the ‘European Lung Cancer Screening Certification Project’ of the European Society of Radiology (ESR) is establishing a standardized certification program for radiologists consisting of workshops, webinars and practical examinations. The aim is to bundle these training initiatives in the planned European Lung Cancer Screening Alliance.
In the concluding discussion round on the topic of AI in lung cancer screening, all speakers were in favor of a European quality assurance system for AI products to ensure that only powerful and tested algorithms are used in clinical practice.
