Changes of Tumor Motion Patterns During a Course of Radiation Therapy for Lung Cancer

Reviewer: Neha Vapiwala, MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: October 22, 2003

Presenter: Kenneth M. Forster, PhD
Presenter's Affiliation: MD Anderson Cancer Center, Houston, TX
Type of Session: Scientific


    Patient and target immobilization with daily reproducibility are the cornerstones of accurate and precise radiation therapy delivery. The treatment of lung cancers has long posed the challenge of retaining this accuracy and precision in the face of natural respiratory motion. This movement has implications for both tumor coverage as well as normal tissue-sparing. This study was designed to evaluate whether NSCLC tumors move with patient respiration, and how this movement changes during an entire six-week course of definitive thoracic radiation.

Materials and Methods

  • 25 patients enrolled between October 2000 and September 2002
  • Only 9 patients received a repeat motion study at the completion of radiation, limiting analysis to this small group.
  • Eligible patients had biopsy-proven lung cancer with no prior surgery or RT to the area, minimal atelectasis, no pleural effusions, and tumor in one lobe only.
  • Study entrants were scheduled to undergo at least a 6-week course of definitive radiation therapy.
  • Patients were trained in spirometry-assisted breath hold technique, and ability to cooperate with this was required.
  • Before each motion study, patients underwent pulmonary function tests using a Sensor Medics Vmax22 computer-controlled occlusion spirometer. Vital capacity and displaced repiratory volume were recorded.
  • Tumor motion study performed while patient applied spirometry-assisted breath hold technique.
  • CT scanning of entire lung volume acquired at normal end inspiration (100% tidal volume) and end expiration (0% tidal volume).
  • Repeat scanning within one week of radiotherapy completion.
  • Gross tumor volume (GTV) contoured on each CT data set by the same physician and quality-controlled with review by a secoond radiation oncologist.
  • Registration of all study CT images with the free-breathing treatment planning CT, using vertebral body landmarks only.
  • Determination of tumor motion and vector displacement based on position (P)= P(ins) - P(exp)
  • Measurement of breathing patterns and lung volumes


  • All 9 analyzed patients demonstrated significant variation in breathing patterns from start to end radiotherapy course.
  • All patients had increased amplitude and frequency of tidal breathing at end of RT compared to beginning.
  • Vital capacity increase in 2 patients of >30% from start to end of RT.
  • Right-left motion change >3 mm in 5 patients (mean = 6mm, range 1-13mm)
  • Anterior-posterior motion change >3 mm in 5 patients (mean = 5.6mm, range 0-15mm)
  • Superior-inferior motion change >3 mm in 4 patients (mean = 3.4mm, range 0-6mm)
  • In 3 patients, the direction of motion actually reversed from start compared to end (eg: from post --> ant on inspiration before RT to ant--> post on inspiration after RT
  • End-expiration positions changed by >5 mm in ALL BUT ONE of the patients (mean=7.2mm, maximum change=12mm)

Author's Conclusions

  • Respiratory motion of NSCLC lesions occurs, and changes considerably throughout the course of radiation therapy.
  • These displacements can be as high as 1 cm or more, and can not be routinely predicted during the intial treatment planning phase.
  • Regular monitoring of tumor motion and location might be needed throughout the course of radiation in order to assure adequate tumor coverage with maximal normal-tissue sparing.

Clinical/Scientific Implications

    The results of this trial are both enlightening and concerning. It certainly gives radiation oncologists an appreciation for the limits we face in planning and delivering accurate treatment to NSCLC patients. The inablity to predict for the degree of tumor motion in a group of patients, combined with the large variation that is possible within any one patient as treatment proceeds, pose a very real challenge. The advent of innovative 4-D CT scanning techniques may help to improve this issue in the future. Implanted fiducial markers may also play a role in assessing the concordance between what is planned and what is actually delivered. The authors also point out the implications of their findings on the use of respiratory gating. They advocate the need for a multi-institutional study to help address this matter further, particulary if more "precise" techniques such as IMRT and proton therapy are to achieve their intended goals.

Oncolink's ASTRO Coverage made possible by an unrestricted Educational Grant from Siemens Medical Solutions.