Short summary

This project aims to investigate, how artificial intelligence can be applied in clinical settings, and based on risk-estimates, used to predict patients in high risk of lung cancer, patients in high risk of experiencing relapse of lung cancer after treatment, in order to offer qualified personalized treatment.

Rationale

Lungcancer remains the second most common type of cancer in Denmark. Every year 4500 people are diagnosed with lung cancer, which has the highest mortality of all types of cancer.

This is due to the fact that the disease has often spred at the time of diagnosis, and once there is not only local disease, treatment is limited to palliative and lifeprolonging treatment. If the disease has not spread at diagnosis, potentially curable treatment can be offered. 5-year survival remains though under 50%. This is due to micro-metastasis, not visible on scans at the time of diagnosis or after operation or chemo- and radiotheraphy. Therefore it is of outpost importance to identify characteristics and riskfactors, specific to patients at an early stage of disease, in order to improve patient survival and prognosis.

Data on patients who have undergone diagnostics due to a suspicion of lung cancer from the Region of Southern Denmark, within the last 10 years will be included. Risk estimates will be calculated, using mathematical algorithms, in order to predict certain outcome.

The project is divided into five sub studies, representing the patients "journey" throughout the different sectors, from private practise, examination in medical diagnostic centres, to oncological treatment at the Department of Oncology.

Project 1: A descriptive analysis of characteristics among patients with lung cancer, compared with a sample without malignancy, who have undergone diagnostics due to a suspicion of cancer. 

Project 2: A characterization of patients with vague unspecific symptoms of cancer, in order to create risk estimators that can help the private practitioner refer patients to further diagnostics at en earlier stage.

Project 3: A characterization of patients with a benign diagnosis, in order to avoid overtreatment and excess diagnostics.

Project 4: A characterization of patients undergoing potentially curative chemo-radiation therapy, who experience relapse of disease, in order to create risk estimators and personalize treatment, accordingly.

Project 5: A characterization of patients undergoing surgery and subsequently chemotherapy, in order to prevent relapse. The aim is again to create individual risk estimators in order to create individual treatment strategies.

The results will be validated on a share of the patient population, and subsequently tested prospectively. With this project we aim to create risk estimators supporting the clinician both in early detection of lung cancer, as well as in generating treatment plans based on individual health information.

Description of the cohort

Study 1: All patients diagnosed with lung cancer in the Region of Southern Denmark within the last 10 years.

Study 2: Patients refered to diagnostic centers, under a broard suspicion of malignancy

Study 3: Patients referred to diagnostic testings for lung cancer, who end up with a benign diagnosis.

Study 4: Patients referred to combination treatment with chemotherapy and radiation theraphy.

Study 5: Patients referred to chemotheraphy after curative surgery.

All five studies include patients in the Region of Southern Denmark, referred in decade jan 2009- jan 2019.

Data and biological material

A variety of registered data including symptoms, diagnosis, laboratory test results, scan results, demographics, prescription informatics.

Collaborating researchers and departments

Department of Oncology, Vejle University Hospital

• MD, associate professor, Torben Frøstrup Hansen, PhD
• MD, associate professor, Lars Henrik Jensen, PhD

Department of internal medicine, Vejle University Hospital

• MD, professor Ole Hilberg, DMSc

Department of Biochemistry and immunology, Vejle University Hospital

• MD, Claus Lohman Brasen, PhD

Publications associated with the project

We intent to publish peer-reviewed articles in relevant international journals.