Multivariate analysis for the estimation of target localization errors in fiducial marker-based radiotherapy

Area of Science:

• Medical Physics
• Radiation Oncology
• Image-Guided Therapy

Background:

• Real-time tumor tracking irradiation (RT-TKI) in lung cancer requires accurate target localization.
• Fiducial markers are used to estimate tumor position, but their accuracy can be affected by target-marker distance and respiratory motion.
• Understanding the factors influencing target localization error (TLE) is crucial for optimizing radiotherapy delivery.

Purpose of the Study:

• To evaluate the target localization error (TLE) based on target-marker distance (|TMD|), average respiratory motion (|aRM|), and the number of fiducial markers (n).
• To determine the relationship between these factors and TLE in lung cancer patients undergoing real-time tumor tracking irradiation.

Main Methods:

• 17 lung cancer patients received four fractions of RT-TKI with 4-5 fiducial markers implanted near the tumor.
• Target localization error (TLE) was calculated using multiple regression analysis (MRA) based on |TMD|, |aRM|, and n.
• Data were analyzed for TLEs meeting specific |TMD| and |aRM| criteria (≤58.7 mm and ≥10 mm, respectively).

Main Results:

• Target-marker distance (|TMD|) was greater for n=1 than n=3, while average respiratory motion (|aRM|) remained consistent across different numbers of markers.
• Multiple regression analysis indicated that |aRM| in the left-right direction was a significant contributor to TLE, though its overall impact on 3D TLE was limited.
• The calculated TLE increased with higher |TMD| and |aRM| and decreased with an increased number of markers (n), with median TLEs ranging from 1.6 to 2.0 mm.

Conclusions:

• While increased target-marker distance and respiratory motion elevate target localization error, the absolute difference in TLE across varying numbers of fiducial markers is small (≤0.6 mm).
• Fiducial marker-based radiotherapy for lung cancer is deemed acceptable provided the target-marker distance is maintained within 58.7 mm for an average respiratory motion of 10 mm or more.