Radiation and Cancer Physics

SS 18 - Physics 4 - Imaging for Response Assessment I

140 - Objective CT-Based Imaging Biomarkers of Radiation-Induced Lung Damage

Tuesday, October 23
8:05 AM - 8:15 AM
Location: Lila Cockrell Theatre

Objective CT-Based Imaging Biomarkers of Radiation-Induced Lung Damage
C. Veiga1, D. Landau2,3, A. Devaraj4, T. Doel1, Y. Ngai5, D. Hawkes1, and J. McClelland1; 1Centre for Medical Image Computing, University College London, London, United Kingdom, 2Department of Oncology, Guy's & St. Thomas' NHS Trust, London, United Kingdom, 3Department of Oncology, University College London Hospital, London, United Kingdom, 4Department of Radiology, Royal Brompton Hospital, London, United Kingdom, 5Cancer Research UK & UCL Cancer Trials Centre, University College London, London, United Kingdom

Purpose/Objective(s): No objective criteria exist to measure radiation-induced lung damage (RILD). This leads to variability in toxicity reporting and poor understanding of its relationship with dose. We recently developed a set of twelve CT-based imaging biomarkers of RILD 12-months after RT, which correlate with observed parenchymal change, volume reduction and pleural change. In this work we present the objective values of these biomarkers and investigate their correlation with respiratory function tests.

Materials/Methods: The biomarkers were calculated on a sub-group (n=50) of a non-randomized phase I/II trial of isotoxic chemoradiation in stage II/III non-small cell lung cancer patients. Respiratory function tests were available for 41 subjects at baseline and 12-months.

Results: The table describes the range of changes seen at 12-months. The biomarkers exhibited significant inter-patient variability in degree of measured change. All patients had evidence of RILD. The combination of sub-types measured with higher severity was variable. Decline in breathing function was assessed as change in FEV1% and DLCO. Five imaging biomarkers were moderately correlated with decline in FEV1% (Pearson’s correlation coefficient, ρ≤-0.3, p≤0.10), particularly those relating to mediastinal shift. No biomarker correlated with DLCO. These findings need to be investigated in larger cohorts.

Conclusion: We demonstrate the use of novel CT-based imaging markers to quantify RILD in RT-treated patients. The variability in radiological and clinical findings indicate individual patterns of normal tissue lung damage in response to RT. These biomarkers have the potential to allow objective scoring of RILD. This would be useful in developing a better understanding of the biology and relationships of RILD as well as a common reporting system.
Imaging biomarker description [units] Measured values (percentile) Correlation with FEV1%
5% 25% 50% 75% 95%
Normal lung volume shrinkage* [%] 3 12 22 34 61 -0.03 (p=0.83)
High-intensity lung volume at 12-months (multiplicative factor between IL+ and CL) – i.e., measure of consolidation) [1×] 1.0 1.5 2.0 3.1 4.8 -0.08 (p=0.64)
Lung width reduction* [%] 0 7 11 18 24 -0.30 (p=0.06)
Lung height reduction* [%] -3 3 8 12 34 0.00 (p=0.99)
Diaphragmatic elevation [mm] -11 1 9 18 56 -0.30 (p=0.05)
Change in diaphragmatic curvature* [%] 1 6 8 17 31 -0.39 (p=0.02)
Increase in the surface of diaphragm tented [mm2] 0 0 0 102 333 -0.18 (p=0.26)
Main bronchus rotation in coronal view* [%] 0 3 8 13 29 -0.36 (p=0.02)
Mediastinal shift toward the IL [mm] -1 1 6 10 16 -0.29 (p=0.07)
Anterior junction line rotation toward the IL [°] -2 2 4 7 13 -0.51 (p<0.01)
Thickening the anterior junction line thickness (multiplicative factor between time-points) [1] 0.5 1.1 1.8 3.2 10.2 -0.25 (p=0.15)
Increase in percentage of chest wall surface covered with pleural reactions [%] 0 3 7 15 39 -0.05 (p=0.76)
*relative to contralateral lung (CL); +IL=ipsilateral lung; ×1=dimensionless

Author Disclosure: C. Veiga: None. D. Landau: None. A. Devaraj: Honorary; Imperial College London. Scientific advisory board; Grail. T. Doel: Honorary Contract; University College London Hospitals NHS Foundation Trust. Y. Ngai: None. D. Hawkes: Honorary Appointment; University College Hospitals. Stock; Vision RT, Ixico plc. Strasbourg IUE; Strasbourg IUE. iFIND and 3D-Heart projects; Kings College London. J. McClelland: None.

Catarina Veiga, PhD

University College London

University College London: Research Associate: Employee

Dr Catarina Veiga, Royal Academy of Engineering Research Fellow at University College London

I acquired BSc in Physics at University of Minho (Braga, Portugal), and an MSc in Medical Physics at University of Porto (Porto, Portugal). I have then obtained my PhD in 2016 in Medical Physics at University College London (London, UK). My project consisted of using CBCT to evaluate the need for treatment adaptation in head & neck and lung patients, both for IMRT and proton therapy. After my PhD I have worked at UCL as research associate developing image analysis tools for radiation-induced lung damage. I have recently been awarded a Royal Academy of Engineering Research Fellowship to develop novel methods to study and predict late effects of radiation in childhood cancer survivors. My research interests are in the use of imaging to improve radiotherapy delivery, namely for applications in proton therapy, treatment planning and adaptation, and minimising side-effects of curative radiation.


Send Email for Catarina Veiga


140 - Objective CT-Based Imaging Biomarkers of Radiation-Induced Lung Damage

Attendees who have favorited this

Please enter your access key

The asset you are trying to access is locked. Please enter your access key to unlock.

Send Email for Objective CT-Based Imaging Biomarkers of Radiation-Induced Lung Damage