Interfractional Variation in the Set-up of Pelvic Bony Anatomy and Soft Tissue, and Their Implications in Proton Therapy of the Prostate

Reviewer: Christine Hill-Kayser, MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: October 14, 2009

Presenter: A. Trofimov
Presenter's Affiliation: Massachusetts General Hospital, Boston, MA
Type of Session: Scientific


  • As proton therapy becomes more widely available, more and more patients with prostate cancers receive definitive radiation using protons.
  • Multiple dosimetric studies have demonstrated potential for improved dosimetry using protons versus photons, with potential sparing of vital structures such as the rectum and bladder.
  • Improved sparing of organs at risk may allow for dose escalation, and higher boost doses delivered with proton radiation have been demonstrated to improve outcomes (Zietman, 2005).
  • Despite the growing use of proton radiation in treatment of prostate cancers, concerns remain regarding ability to deliver adequate and safe dose to a region situated amongst many other soft tissues, which is thus mobile and may exhibit positional variation from fraction to fraction.
  • The study described here was undertaken to address these concerns through evaluation of several aspects of immobilization and treatment:
    • First, the daily dose distribution due to daily variation in patient anatomy and position.
    • Second, the effectiveness of the current set-up approaches employed at Massachusetts General Hospital (MGH).

Materials and Methods

  • Series of CT data from 10 patients treated for prostate cancer at Morristown Memorial Hospital were evaluated.
    • All patients were treated with intensity modulated radiation therapy (IMRT) to a total of 76 Gray (Gy).
    • All patients underwent treatment planning CT scan, as well as between 23 and 43 in-room scans done on various days throughout the radiation course. All patients underwent in-room CT scans on days 1-5 of the treatment course, and on approximately 33% of ensuing days.
    • Patients were immobilized supine with the legs straight. A grooved cushion was used to immobilize the knees and lower legs, and the feet were placed in toe straps.
    • Patient body mass index (BMI) ranged from 22.8 – 53 (with BMI>30 being classified as obesity).
  • Variations between the treatment-planning CT and daily CT scans were recorded for the following dimensions:
    • Hip rotation angles
    • Thickness of lateral subcutaneous adipose tissue
    • Lateral physical depth to distal surface of prostate.
  • Treatment planning for all 10 patients was then undertaken using the standard MGH approach.
    • The prostate and seminal vesicles were contoured, and a 5mm expansion applied for determination of planning target volume (PTV).
    • A 10 mm compensator expansion was applied.
    • A 3.5% expansion of the distal and proximal spread out Bragg peak edges was applied.
  • Daily variations in the volume treated to the prescription dose, as well as target coverage, were then evaluated.


  • For all 10 patients, significant variation was observed in lateral tissue thickness. Variation was at least 5 mm in all 10 patients, and at least 10 mm in 3 patients. The standard deviation was 1.7 – 3.6 mm for individual patients.
  • Hip rotation angle was also found to vary significantly from day to day for each patient, with standard deviation 1.4 – 4.8 degrees. Maximum variability exceeded 15 degrees at some point during the treatment course for 6 of 10 patients. Changes in hip rotation were noted to affect the soft tissue distance from skin to femur.
  • The standard deviation of the 98% isodose line was 2 – 3.5 mm. Shifts in the position of the 98% isodose line were correlated with variations in lateral tissue thickness and hip rotation.

Author's Conclusions

  • The authors conclude that hip rotation and soft tissue motion may cause substantial change in shape of the treated volume.
  • Compensator expansion was noted to prevent loss of dose to the target in most cases.
  • They note that in-room set-up verification is essential for maintaining target coverage. They note that improved immobilization techniques may allow reduction of PTV and collimator expansions, but that uncertainties in proton penetration will still exist; they thus note that refinements in proton range verification may have the largest impact on margin reduction.

Clinical/Scientific Implications

  • The authors have performed an interesting and well-done study, demonstrating significant motion of pelvic bony structure of the course of prostate cancer treatment.
  • Their findings have significant implications in the treatment of patients with proton radiation for prostate cancer. The very quality that may allow dosimetric improvement with proton therapy – improved conformality – may also increase risk of inadvertent target misses due to inter- and intrafractional variability in organ location.
  • For this reason, we must be particularly careful as we advance treatment options for prostate cancer and increasingly offer proton therapy to patients. Although theoretical studies demonstrate dosimetric benefit with use of protons for treatment of prostate cancer, proton versus photon radiation for prostate cancer have never been directly compared to one another. Although such a comparison may soon be underway, adequate data regarding long-term outcomes will take many years to accrue.
  • Patients are currently being treated with proton radiotherapy in a milieu in which long-term data may not be available. Studies such as the one presented here are valuable within the literature as they allow improved awareness of potential complications of proton radiation for prostate cancer even as further clinical data is awaited.

Frequently Asked Questions