IMRT sparing of normal tissues in locoregional treatment of breast cancer

A radiation oncologist and/or a clinical research assistant assessed the acute toxicity every week for the duration of treatment and 2 weeks post treatment, using the CTCAE v3.0.scale. Follow-ups were scheduled at 3 months post radiation and then every 6 months thereafter.

Smoking history, cardiac and lung comorbidities were collected. Eleven patients had a smoking history with tobacco consumption from 3 to 45 pack-years, 7 of them having a tobacco consumption at least of 10 pack-years. The prior cardiac history was significant for: 1 patient with valve surgery, 1 patient with bypass surgery, 1 patient with treated arrhythmia and high blood pressure (HBP), plus 5 patients with treated HBP. The respiratory history was significant for: 1 patient with treated chronic obstructive pulmonary disease (the heaviest smoker), and 4 patients with treated asthma.

Cardiac function and radiation effects were evaluated using persantine myoview perfusion and left ventricular function test performed at baseline, 3 and 12-month follow-up. Images for tomographic myocardial perfusion were acquired following persantine stress (50 mg IV) and at rest and by using IV Technetium-99 m tetrofosmin (99mTc) (334 MBq at rest and 1070 MBq at peak stress). Both the myocardial perfusion plus the left ventricular function at rest and post stress were evaluated. The myoview tests were initially done with 99mTc, but following the radioactive metabolite crisis and shortage of 99mTc, follow-up myoview tests were done with Thallium-201 chloride (Tl-201) administered intravenously at peak vasodilation (130 MBq). Cardiac function interpretation combined rating ECG, left ventricular function and myocardial perfusion as normal or not. A global score of zero (if no abnormalities at all) or 1 (if at least 1 abnormality reported) was attributed to each patient. When the global score was 1, tests were reviewed in detail to precisely define the abnormalities reported.

The evaluation of pulmonary toxicity was mainly clinical but a CT scan of the chest was scheduled at the 6-month follow-up to evaluate radiologic fibrosis.

This clinical trial was approved by the ethics committee of the Ottawa Hospital Research Institute, patients signed informed consent and the study was registered in the clinicaltrials.gov database with the following identifier: NCT00508352.

All patients completed the 5-week course of radiation without interruption. Skin erythema was modest and mainly grade 1 during the 3rd and 4th week (22/30 and 24/30 patients, respectively), and grade 2 during the 5th week (22/30 patients). Only 4/30 patients experienced grade 3 (any moist desquamation area) erythema, mostly recorded 2 weeks post radiation. Only 5 patients demonstrated grade 1 or 2 dyspnea but 3 of them already had symptoms pre-irradiation at baseline assessment. One patient had developed acute respiratory distress syndrome pre-irradiation during delivery of adjuvant taxane-based chemotherapy and was managed in the intensive care unit. A second patient developed bronchiolitis obliterans post taxane-based chemotherapy that merited an evaluation by a respirologist. Neither of these women developed any acute respiratory symptoms during radiation treatment. Eleven patients showed grade 1 temporary dysphagia from esophagitis.

The median follow-up of the whole group, including the deceased patients, is 58 (24–76) months. The evolution of the major moderately-late side effects is summarized in Table 2. Skin reactions, breast/chest-wall pain, respiratory and sensori-motor symptoms have been infrequent, and mostly grade 1. They were sometimes already present at baseline assessment. Six patients underwent breast reconstruction post mastectomy and none developed any significant complications, including no arm lymphedema recorded from physical examination or from arm measurements done by a nurse. One patient was diagnosed with acute leukemia 24 months post radiation treatment and is currently in complete remission post treatment. Her adjuvant chemotherapy regimen was FEC100-Docetaxel. One patient was diagnosed with a small cell lung cancer on the radiated side 30 months post radiation treatment. This patient had other lung cancer risk factors. The primary tumour was localized in an area that was covered by 40 to 50 Gy. It is possible that this small cell lung cancer was radiation induced. However, this causality is not specific to the IMRT technique as a tangential radiation beams technique would have covered the same area with the 40 to 50 Gy isodoses.

Women were scheduled for baseline, 3 and 12-month follow-up scintigraphy but some women declined the initial tests and some declined any follow-up tests, mostly because of the discomfort caused by the persantine injection used to simulate stress. The number of tests done and abnormalities reported are summarized in Table 3. Abnormalities have been recorded for 6 women, 4 radiated on the right side and 2 on the left. For 5 patients, the perfusion abnormalities are not related to the radiation treatment. Three were present at baseline and 2 at follow-up were related to imaging artefacts according to cardiology assessment. Only 1 patient’s abnormalities were possibly related to radiation. This patient had several cardiac comorbidities (arrhythmia, HBP, smoking habits) and her cardiac radiation doses were the following: heart mean dose:7.8Gy, V25:4%; left ventricle median dose:4.4Gy [2.2-28.9], V30:0%; LAD artery median dose:26Gy [3.7-43.5]. Following a gastrointestinal surgical procedure, this patient died from congestive heart failure post atrial flutter.

As summarized in Table 2, no significant late clinical pulmonary toxicity has been recorded. A maximum number of 3 patients experienced a grade 1 dyspnea or cough from 3 and up to the 12 month follow-up. The 2 patients who needed respirologist evaluation were all seen during or at the end of chemotherapy. At 30 months, all 26 remaining patients had no respiratory symptoms. A 6-month follow-up CT chest has been done for 25 out of 30 patients. For 18 patients, the CT chest was completely normal and for the other 7, the CT chest showed a minimal anterior fibrosis (grade 1 fibrosis).

No loco-regional recurrence has been diagnosed. Five patients have been diagnosed with distant recurrent disease. One patient developed a contralateral breast cancer with axillary nodal involvement at 11 months post completion of radiation treatment. This was in the context of a BRCA1 gene mutation. The same patient later developed metastatic disease. Four patients have been diagnosed with bone metastasis at least after 6 months of follow-up. Six deaths have been recorded: 4 patients from progression of metastatic breast cancer, 1 patient from a lung cancer and one due to a cardiac event post surgical procedure. The estimated Kaplan Meier survival at 5 years is 78% (CI 95%: 70-97%).

Clinically significant symptomatic radiation pneumonitis occurs in approximately 1–5% of patients irradiated for breast cancers [25]. One of the major concerns (explaining the low rate of accrual) with this pilot trial was the possibility of an increased risk of severe subacute respiratory symptoms related to the IMRT rotational beam technique. This technique results in an increase in the volume of lung receiving a low dose of radiation (<10Gy). Severe respiratory toxicity has been reported for mesothelioma patients treated by post operative IMRT when the mean lung dose and V20 and V5 were increased [26]. However in our group of patients, acute or moderately late lung effects were not seen. Most of the respiratory symptoms recorded were already present at baseline and some were partly related to the taxane-based regimen. These results may fit with some published data suggesting that reducing the volume of lung receiving high dose of radiation (≥40Gy) and reducing the mean lung dose decrease the risks of radiation pneumonitis [27]. The CT chest at the 6-month follow-up done for 25 (83%) of the patients was reported normal for 18 (72%) of them. The others showed minimal anterior fibrosis. Thus, we may postulate that locoregional IMRT was very well tolerated regarding pulmonary side effects. It could be noted that the average DVH values recorded for lungs in our clinical study are in the range of values recommended for radiation therapy post pneumonectomy for mesothelioma in the last Quantitative Analysis of Normal Tissue Effects in the Clinic (QUANTEC) review [28].

Cardiac perfusion defects of up to 30% have been seen on follow-up scintigraphy perfusion tests for irradiated left breast cancer patients [29‐31]. Myoview tests have been used as a surrogate marker of possible cardiac injury from radiation treatment. Fifteen out of eighteen left sided breast cancer patients had baseline, 3 and 12 month follow-up myoview tests and none of them detected a significant perfusion abnormality. The only patient who developed cardiac problems was diagnosed with an abnormal test at 30 months post radiation treatment. It is unclear if these abnormalities and subsequent cardiac problems were related to previous breast cancer treatments, but we cannot completely exclude radiation treatment causality for this patient. A recent study reported a percent increase of major coronary event by 7.4% for every 1 Gy of mean cardiac dose [10]. For a mean cardiac dose of 8 Gy (the mean cardiac dose recorded in our study for women radiated on the left or right side) the estimated percent increase of major coronary event is 40%. It is unclear if this estimated number from the Darby et al. paper can be used for locoregional breast radiation with IMRT rotational beam technique. Their study reported only the mean cardiac dose, but not the volume of heart receiving above 25 Gy and/or close to the prescription dose. Radiation by a combination of tangent beams treats a larger volume of heart with a dose equal or superior to 25 Gy, compare to IMRT rotational beam technique. On the other hand, IMRT rotational beam technique will radiate a larger volume of heart to a dose less than or equal to 10 Gy, and as a result the mean cardiac dose will be higher compared to the tangential beam radiation technique. We think that when evaluating a locoregional breast radiation planning, the mean cardiac dose is not the only factor to consider and that the V25 and/or V30, V40 are also important metrics. From the perfusion myoview tests results, we also note that a mean cardiac dose around 6–8 Gy seems to be safe for breast cancer women, if we can achieve a V25 below 5% and a V30 below 2.5%. We do not believe that the mean cardiac dose is the only metric to meaningfully compare 3D planning using tangent beams and the IMRT rotational beam technique.

As stated above and due to the shortage of radionuclides, 99mTc was replaced by Tl-201. One may argue that this may have altered the results of the myoview tests. However, the ROBUST study compared the performances of 3 radionuclide myocardial perfusion tracers for clinical practice in a large population of 2560 patients. That study concluded that all tracers performed well in clinical terms, with high sensitivity and specificity for angiographic stenosis and no differences in accuracy between the tracers [32]. Therefore, we may reasonably suggest that the negativity of myoview tests is likely due to the sparing of cardiac structure including the coronary arteries from a high dose of radiation.