1. HOME
  2. News
  3. AI analysis of PET/CT images can predict side effects of immunotherapy in lung cancer

AI analysis of PET/CT images can predict side effects of immunotherapy in lung cancer

Nov 11 2024

Dr. Watanabe and his teams from Niigata University have revealed that PET/CT image analysis using artificial intelligence (AI) can predict the occurrence of interstitial lung disease, known as a serious side effect of immunotherapy in lung cancer.

Immunotherapy has dramatically improved the treatment outcomes of primary lung cancer; however, it sometimes causes a serious side effect called interstitial lung disease. Interstitial lung disease is characterized by scarring (fibrosis) of the lungs and may be life-threatening owing to respiratory failure. Unfortunately, it is difficult to predict the occurrence of interstitial lung disease induced by immunotherapy. Accordingly, effective methods for predicting the risk of developing interstitial lung disease after immunotherapy are required.

This retrospective study investigated 165 patients with primary lung cancer who received immunotherapy at Niigata University Medical and Dental Hospital. As it is suggested that interstitial lung disease arises when inflammatory cells activated through immunotherapy damage healthy lung as well as cancer cells, the researchers hypothesized that patients with severe inflammation in healthy lungs prior to immunotherapy are more likely to develop interstitial lung disease after the treatment. Dr. Watanabe and his teams focused on PET/CT scan, a nuclear imaging test that is able to detect inflammation in the whole body. The researchers quantified the degree of inflammation in noncancerous lungs, namely lung regions without cancer, using AI analysis of PET/CT images. The study demonstrated that the risk of developing interstitial lung disease after immunotherapy is approximately 6.5 times higher in patients with high inflammation in the noncancerous lung than in those with low inflammation.

Dr. Yamazaki says "PET/CT is generally performed to detect cancer metastasis, but it would potentially be useful for estimating the risks of side effects associated with cancer treatment. The results of our study may not only help to predict the occurrence of interstitial lung disease after immunotherapy, but also to elucidate the mechanism of this serious side effect. We should conduct a multicenter prospective study for further investigation."


Example of AI analysis of PET/CT images
Using AI, the lung region without cancer on PET/CT images (green-colored region, called noncancerous lung) is automatically extracted. Then, the degree of inflammation within the extracted lung region is calculated. The red arrow denotes lung cancer, and the white arrows indicate lymph node metastasis.

Publication Details

Journal: Academic Radiology
Title: 18F-FDG-PET/CT Uptake by Noncancerous Lung as a Predictor of Interstitial Lung Disease Induced by Immune Checkpoint Inhibitors
Authors: Motohiko Yamazaki, Satoshi Watanabe MD, PhD, Masaki Tominaga, Takuya Yagi, Yukari Goto, Naohiro Yanagimura, Masashi Arita, Aya Ohtsubo, Tomohiro Tanaka, Koichiro Nozaki, Yu Saida, Rie Kondo, Toshiaki Kikuchi, Hiroyuki Ishikawa
Doi: 10.1016/j.acra.2024.08.043

News release

The article was released in EurekAlert, the online publication of the American Association for the Advancement of Science.

More News