Correlation between patients' anatomical characteristics and interfractional internal prostate motion during intensity modulated radiation therapy for prostate cancer

Factors Affecting Prostate Displacement During Volumetric Modulated Arc Therapy in Prone Position After High-Dose-Rate Brachytherapy for Prostate Cancer

Purpose

In irradiating the prostate and pelvic lymph node regions, registration based on bony structures matches the pelvic lymph node regions but not necessarily the prostate position, and it is important to identify factors that influence prostate displacement. Therefore, we investigated factors influencing prostate displacement during volumetric modulated arc therapy after single-fraction high-dose-rate brachytherapy (HDR-BT) for prostate cancer and the trends in displacement for each fraction.

Methods and Materials

Seventy patients who underwent pelvic volumetric modulated arc therapy of 46 Gy in the prone position 15 days after 13 Gy HDR-BT were included. Prostate displacement relative to bony structures was calculated using cone beam computed tomography. Systematic error (SE) and random error (RE) were evaluated in the right-left (RL), craniocaudal (CC), and anteroposterior (AP) directions. The association with clinical and anatomic factors on the planning computed tomography or magnetic resonance imaging was analyzed. Prostate volume change (PVC) was defined as the volume change at 2 days after HDR-BT. Displacement trends were individually examined from the first to 23rd fractions.

Results

The mean SE in the RL, CC, and AP directions was -0.01 mm, -2.34 mm, and -0.47 mm, respectively. The root mean square of the RE in the RL, CC, and AP directions was 0.44 mm, 1.14 mm, and 1.10 mm, respectively. SE in the CC direction was independently associated with bladder volume (P = .021, t statistic = 2.352) and PVC (P < .001, t statistic = -8.526). SE in the AP direction was independently associated with bladder volume (P = .013, t statistic = -2.553), PVC (P < .001, t statistic = 5.477), and rectal mean area (P = .008, t statistic = 2.743). RE in the CC direction was independently associated with smoking (P = .035). RE in the AP direction was associated with PVC (P = .043). Gradual displacement caudally and posteriorly occurred during the irradiation period.

Conclusions

Anatomic characteristics of the bladder, rectum, and prostate predict SE. Smoking and PVC predict RE. In particular, whether PVC is ≥140% affects setting internal margins.

Does body mass index or subcutaneous adipose tissue thickness affect interfraction prostate motion in patients receiving radical prostate radiotherapy?

Aim

It is unclear whether body mass index (BMI) is a useful measurement for examining prostate motion. Patient’s subcutaneous adipose tissue thickness (SAT) and weight has been shown to correlate with prostate shifts in the left/right direction. We sought to analyse the relationship between BMI and interfraction prostate movement in order to determine planning target volume (PTV) margins based on patient BMI.

Materials and methods

In all, 38 prostate cancer patients with three implanted gold fiducial markers in their prostate were recruited. Height, mass and SAT were measured, and the extent of interfraction prostate movement in the left/right, superior/inferior and anterior/posterior directions was recorded during each daily fiducial marker-based image-guided radiotherapy treatment. Mean corrective shift in each direction for each patient, along with BMI values, were calculated.

Results

The median BMI value was 28·4 kg/m2 (range 21·4–44·7). Pearson’s product-moment correlation analysis showed no significant relationship between BMI, mass or SAT and the extent of prostate movement in any direction. Linear regression analysis also showed no relationship between any of the patient variables and the extent of prostate movement in any direction (BMI: R2=0·006 (ρ=0·65), 0·002 (ρ=0·80) and 0·001 (ρ=0·86); mass: R2=0·001 (ρ=0·87), 0·010 (ρ=0·54) and 0·000 (ρ=0·99); SAT: R2=0·012 (ρ=0·51), 0·013 (ρ=0·50) and 0·047 (ρ=0·19) for shifts in the X, Y and Z axis, respectively). Patients were grouped according to BMI, as BMI<30 (n=25, 65·8%) and BMI≥30 (n=13, 34·2%). A two-tailed t-test showed no significant difference between the mean prostate shifts for the two groups in any direction (ρ=0·320, 0·839 and 0·325 for shifts in the X, Y and Z axis, respectively).

Findings

BMI is not a useful parameter for determining individualised PTV margins. Gold fiducial marker insertion should be used as standard to improve treatment accuracy.