AI Transforms Cancer Survivorship Care

Published Date : 20/10/2025 

A recent study presented at the EONS session at the ESMO Congress 2025 highlights the growing use of artificial intelligence (AI) in cancer survivorship care. The study, conducted by Akile Karaaslan Eser, PhD, RN, of the University of Health Sciences in Ankara, Turkey, reveals the potential of AI tools in addressing various survivorship challenges. 

A new study highlights the use of artificial intelligence (AI)-based approaches to cancer survivorship care, with many such tools currently remaining in early development. These findings were presented at an EONS session at the ESMO Congress 2025 by Akile Karaaslan Eser, PhD, RN, of the University of Health Sciences in Ankara, Turkey.

The study, registered with CRD420250651277, was based on a comprehensive literature search of records within PubMed, Scopus, and Web of Science databases for articles published between March 17, 2015, and February 17, 2025. The study followed Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and had 2 independent reviewers evaluate findings.

Literature searches involved several keywords, such as “artificial intelligence,” “machine learning,” “cancer survivors,” “survivorship,” “cancer care,” and other similar terms. Included studies could involve original research using a variety of designs, such as randomized controlled trials, cohort studies, or cross-sectional studies. They were required to have an intervention involving AI-based methods and with adult cancer survivors as the study population.

From a total of 1096 records identified from database searches, Karaaslan Eser and colleagues found 52 articles eligible for inclusion in the analysis. Study types were observational in 60% of cases, while 20% were randomized controlled trials/pilot studies, and 20% were focused on model development.

A range of cancer types were identified among survivors across studies, with breast cancer being the most common (approximately 40%). This was followed by colorectal cancer (approximately 20%), hematologic cancers (approximately 15%), and various other cancer types.

In her presentation, Karaaslan Eser noted that findings suggested there has been growing use of AI and machine learning in addressing survivorship challenges. For example, machine learning models have been developed for predicting symptoms (21 studies; 40%), such as fatigue, insomnia, cognitive decline, and psychological distress.

Psychosocial outcomes were also frequently evaluated in studies (14 studies; 27%), involving topics such as return to work, quality of life, depression, or fear of recurrence. Cardio-oncology and survival were addressed in 10 studies (19%), and digital health solutions, involving mobile apps, chatbots, or wearable sensors, were evaluated in 17 studies (33%).

Models identified within studies had an accuracy ranging from 72% to 93%, and 50% had area-under-the-curve values of greater than 0.80. User satisfaction with digital tools ranged from 68% to 92%.

Karaaslan Eser indicated AI demonstrates strong potential to offer personalized, proactive, and efficient survivorship care but most studies are in the pilot or feasibility stage. “Larger, multicenter, randomized, controlled studies are urgently needed for validation and integration into routine practice,” she said in her presentation. 

Frequently Asked Questions (FAQS): 

Q: What is the main focus of the study presented by Akile Karaaslan Eser?

A: The main focus of the study is the use of artificial intelligence (AI) and machine learning in cancer survivorship care, highlighting the potential of these technologies to address various survivorship challenges.

Q: How many articles were included in the analysis?

A: A total of 52 articles were included in the analysis after a comprehensive literature search.

Q: What are the most common cancer types among survivors in the studies?

A: The most common cancer types among survivors in the studies were breast cancer (approximately 40%), followed by colorectal cancer (approximately 20%), and hematologic cancers (approximately 15%).

Q: What is the accuracy range of the AI models identified in the studies?

A: The accuracy of the AI models identified in the studies ranged from 72% to 93%, with 50% of the models having area-under-the-curve values of greater than 0.80.

Q: What is the main recommendation for future research in this field?

A: The main recommendation is for larger, multicenter, randomized, controlled studies to validate and integrate AI and machine learning tools into routine practice. 

More Related Topics :