Analysis of Interfraction and Intrafraction Variation During Tangential Breast Irradiation with an Electronic Portal Imaging Device
Source: Int J Radiat Oncol Biol Phys. 2005 Jun 1;62(2):373-8.
Authors: Smith RP, Bloch P, Harris EE, McDonough J, Sarkar A, Kassaee A, Avery S, Solin LJ.
During radiotherapy, a major goal is to deliver radiation to the target tissue while minimizing exposure to normal tissues. In tangential breast radiotherapy, the breast is the target and the lung and heart (during left-side irradiation) are the organs at risk. Maximizing the therapeutic ratio is a guiding principle during radiation portal design, however, what is actually delivered to the patient can deviate significantly from what is planned. Patient motion during treatment (intrafraction variation) and daily set-up errors (interfraction variation) during breast irradiation are evaluated in the study using EPID (electronic portal imaging device).
- Design: Patients were prospectively evaluated with 1,709 electronic portal images collected during their tangential breast irradiation
- Patients: 8 patients with early stage invasive breast cancer or DCIS treated with breast conservation therapy; 6 with left-sided and 2 with right-sided breast cancer.
- Patients were immobilized with an alpha cradle cast and instructed to maintain light, easy breathing
- During each treatment field, 14-18 electronic portal images were acquired at a rate of 53 frames/minute. Images were collected during during different treatment days (112 total days, although it is not clear how many days each patient was imaged)
- On the first image of each set, the lung, heart, and breast were manually contoured; an image analysis algorithm was used on subsequent images to identify these organs
- Central Lung Distance (CLD), the distance between the deep field edge and the interior chest wall at the central axis, was measured.
- In the 4 patients who were simulated on CT, corresponding heart and lung volumes were calculated based on the 2D EPID images
- The intrafraction variation (patient/organ motion) was minimal with variations in CLD (at most 0.25 cm). Range of lung area and heart area were also minimal.
- The interfraction variation (set up error) was LARGE compared to the intrafraction variation, with CLD's ranging from 0.59 cm in the most reproducible patient and the worst patient varying the CLD 2.94 cm (most were in the 1-2 cm range)
- Lung and heart areas (and volumes where possible) followed a similar trend where interfraction varying was much larger than intrafraction variation
- The authors point to the lack of a survival benefit to post-lumpectomy radiation as an indication that normal tissue toxicity (predominantly cardiovascular events) erases the cancer-specific survival benefit. More precise localization of the target and organs at risk could translate into a survival benefit to post-lumpectomy radiation in addition to the local control benefit.
- Sophisticated methods are being developed to avoid normal tissue toxicity, namely IMRT and ABC (active breathing control). Given the data presented here, the best IMRT plan can be rendered no better than conventional treatment if daily set up error is not addressed. There appears to be little benefit to be gained from ABC since intrafraction variation due to breathing motion is only 2.5 mm.
This paper uses a relatively new technology, electronic portal imaging, to assess how accurate radiation delivery is during tangential breast irradiation. Variations from daily set up errors were several-fold larger than variations from breathing. Similar studies in prostate radiotherapy have lead to incorporation of daily position verification techniques, such as ultrasound or fiducial markers. Here we see that similar problems plague breast radiotherapy, which is especially important in left-sided breast cancers where the heart is an organ at risk. Daily verification and adjustment of tangent fields using electronic portal imaging would likely make a bigger impact on sparing heart and lung than IMRT or ABC.