Abstract
Aim: To investigate the correlation between frequency of action level of interfractional rectal displacement requiring repeated precaution in patients with prostate cancer and late toxicity from image-guided intensity-modulated radiation therapy (IG-IMRT) using helical tomotherapy. Patients and Methods: We examined 264 patients who underwent IG-IMRT during 2007-2011. Megavoltage computed tomographic (MVCT) images were acquired before radiation therapy and was examined with soft-tissue matching by comparing treatment planning images within 9,345 fractions. Displacement of the anterior rectal region larger than 5 mm, requiring repeated precaution, was defined as the level of rectal displacement requiring action (ARD). Results: ARD was identified in 815 (7.7%) out of 9,345 fractions and at least once in 82% (216/264) of patients. The highest incidence of ARD (11%) was found during the initial week of treatment (first five and next five fractions), after which the incidence decreased to 6% (p<0.0001). Patients with lean body (lower body mass index (BMI) tended to have a higher incidence of ARD. We identified 16 (6%) cases of gastrointestinal toxicity and 12 (4.5%) genitourinary toxicities as a late adverse reaction (3 months or later after IG-IMRT). There was no correlation between ARD and late toxicity. Prostate-specific antigen (PSA) control was also similar (p=0.12) between those with ARD (96% at 5 year) and those without ARD (88%). Conclusion: ARD occurred predominantly in lean patients, during the initial week of treatment and became less likely over time. ARD was not correlated to late toxicity and PSA control, therefore, IG-IMRT technique was able to adequately control error due to interfractional prostate and rectal motion.
Radiotherapy has an established role in local prostate cancer treatment. With the advance of accurate radiation techniques, intensity-modulated radiation therapy (IMRT) and image-guided radiation therapy (IGRT), higher irradiation levels can be delivered to target areas without increasing the dose to organs at risk by reducing the set-up margin and planned target volume (PTV) (1-3).
Helical tomotherapy is a form of IMRT with the ability to acquire megavoltage computed tomographic (MVCT) images of a patient in the treatment position before therapy. This precise positioning using CT images allows not only for correct bone position (bone matching) used in conventional radiation therapy (2), but also for visualization and identification of the position of organs, such as the prostate, rectum and bladder (i.e. soft-tissue matching). We identified interfractional rectal displacement using repeated MVCT before treatment and corrected patient positioning using couch adjustment, a process known as image-guided IMRT (IG-IMRT) (4-6). However, we occasionally encounter considerable rectal displacement, that is sometimes impossible to correct using couch adjustment. We define this degree of displacement as a level of rectal displacement that requires further action (ARD) (6). In such cases, the patient is instructed to get up from the couch and drink water to fill the bladder and to void the rectum; a rectal enema is prescribed if required.
Patient characteristics.
Because rectal displacement can not only cause treatment failure but also toxicity, the aim of this study was to analyze the correlation between ARD and outcomes [prostate-specific antigen (PSA) control and late toxicity of radiotherapy, especially for rectal toxicity].
Patients and Methods
We investigated 264 male patients with prostate cancer who received IG–IMRT using helical tomotherapy (HI-ART TomoTherapy Inc., Madison, WI, USA) between 2009 and 2011. The patients were aged between 48 and 86 years (median=72 years). The patient characteristics are presented in Table I. Details of the treatment have been described elsewhere (4-6). In brief, vacuum cushions (Blue Bag, Medical Intelligence, Schwabmûenchen, Germany) were used to immobilize the patients in the supine position. Kilovoltage CT planning images were acquired for each patient (2-mm slice thickness), including a minimum of 5 cm above and below the level of the PTV (Aquilion 64; Toshiba Medical Systems Inc., Tokyo, Japan). The prostate gland and proximal seminal vesicles were contoured as the clinical target volume (CTV) with the aid of fused magnetic resonance imaging. In the initial 2.2 Gy/fraction schedule (2007-2009), the CTV–PTV expansion margin was 5 mm in all directions, not avoiding the rectum. The primary PTV was defined by margins of 3 mm posteriorly and 5 mm in all other dimensions around the prostate gland and seminal vesicles (4). The prescribed dose was 72.6 Gy in 33 fractions (2007-2009), or 72 Gy in 36 fractions (2009 on) for patients in the low-risk group (Damico's: classification: stage, T1c; Gleason score, <7; and PSA, <10 ng/ml) and 74 Gy in 37 fractions or 74.8 Gy in 34 fractions (2009 on) for patients in other risk groups, set as D95 (i.e., 95% of the PTV received the prescribed dose). The bladder and rectum (contoured from the anal verge to rectosigmoidal junction) were defined as organs at risk and the major constraints during inverse planning were that no more than 35% of the rectal volume and no more than 50% of the bladder volume would receive 40 Gy of radiation. Patients were routinely instructed to empty the rectum, but not the bladder, 1 h before treatment. Patients took magnesium oxide (1 g/day) and dimethicone (80 mg 3 times a day, a total of 240 mg/day) 7 days prior to planning CT until the completion of treatment. MVCT images (3.5 MV) were acquired through PTV before treatment delivery, with a minimum slice thickness of 4 mm and a field of view of 35 cm. The first MVCT images were taken and autofused with the kilovoltage CT treatment planning images, and the superior–inferior, anterior–posterior and right–left shifts were then calculated using automatic image fusion for bone matching. The fused images were manually inspected for prostate soft-tissue matching and verified and corrected by two clinicians (rotational corrections were not implemented at the time of this study). Patients were then shifted into the calculated position by adjusting the couch. We defined ARD if most of the anterior rectal region moved more than 5 mm and could not be corrected using couch adjustment (6). In these cases, the patient was asked to get up from the couch and to void the rectum. If necessary, a rectal enema was used to dislodge large-volume rectal stool or rectal gas. A second set of MVCT images were acquired to verify that the prostate shift had been corrected. If ARD persisted, further correction was implemented. The total time between image acquisition and treatment delivery was typically less than 10 min. However, if ARD occurred, the time delay could be in the range of hours.
The number of fractions (%) requiring action to correct rectal displacement (ARD) according to time of treatment.
The clinical characteristics of patients are shown in Table I. Prostate specific antigen failure was defined using the Phoenix definition (nadir +2 ng/ml) (7) or start of salvage hormonal therapy. Common Terminology Criteria for Adverse Events version 3.0 was applied to late toxicity analysis (8). Late toxicity [gastrointestinal (GI), genitourinary (GU)] was defined as any toxicity experienced 3 months after the completion of radiotherapy. All patients provided informed written consent to their participation in this study. This study was conducted in accordance with the Declaration of Helsinki and Institutional review Board approval (ERB-C-926).
Statistical analysis. All statistical analyses were performed using Stat-view 5.0 statistical software (SAS Institute, Inc., Cary, NC, USA). The percentage values were analyzed using the chi-square test and means were compared using the Student's t-test. All analyses used the conventional p<0.05 level of significance.
Prostate-specific antigen (PSA) control rate according to the need for action to correct rectal displacement (ARD) during image-guided radiotherapy (IMRT). a: PSA control rates at 5 years in patients requiring ARD (bold line) and those not (thin line). b: PSA control rates at 5 years between those requiring highly frequent (≥5 fractions) ARD (thick line) and those with less frequent (<5 fractions) ARD (thin line).
Comparison of characteristics between patients requiring and those not requiring action to correct rectal displacement (ARD).
Results
We treated 264 patients with 72 Gy/36 fractions (27 patients) or 74 Gy/37 fractions (229 patients). In total, 9,345 fractions and 10,279 MVCT images were analyzed for deviations of the prostate using soft-tissue matching. ARD occurred in 815 fractions (7.7%) out of 9,345 fractions (median of three ARD fractions per patient; range=1-21 ARD fractions per patient). During the treatment course with 36-37 fractionations, 54 patients experienced one ARD, 33 two ARD, 46 three ARD, 30 four ARD and 60 five or more ARD during the treatment course; a total of 82% (216/264) of patients experienced ARD at least once. The highest incidence of ARD of 11% (221/1970) was found during the initial week of treatment (first five fractions), after which the incidence decreased to 6% (700/12148) (p<0.0001) (Table II). The predisposing factors for ARD are presented in Table III. Univariate analysis identified a low body mass index (BMI) as a statistically significant predisposing factor for ARD (Table III). The frequency of ARD was found to have decreased with time (Table IV). In 2007, 96% (26/27) patients experienced ARD, and 79% (52/66) in 2008, 89% (47/53) in 2009, 82% (42/52) in 2010, and 73% (49/67) in 2011 (p=0.03).
We identified 16 (6%) late GI toxicity grade 2 or more and 12 (4.5%) late Gu toxicities of grade 2 or more (Table V). There was no correlation between ARD and late toxicity [GI: 6% (3/48) patients with ARD vs. 6% (13/216) without, p=0.99); GU: 2% (1/48) patients with ARD vs. 4% (11/216) without, p=0.7)]. PSA control was also similar between those with ARD (96% at 5 years) and those without (88%, p=0.12) (Figure 1).
We examined the background characteristics of 60 patients with particularly highly frequent occurrence of ARD (≥5 fractions). Highly frequent ARD (≥5 fractions) was also found to have decreased with time (Table IV): 26% (7/27) of patients experienced highly frequent ARD in 2007, 18% (12/66) in 2008, 42% (22/53) in 2009, 24% (12/52) in 2010, and 10% (7/67) in 2011 (p=0.002). There was no correlation between highly frequent ARD and late grade 2 or more toxicity [≥5 ARD vs. 0-4 ARD: GI: 10% (6/60) vs. 5% of patients (10/204, p=0.25); GU: 2% (1/60) vs. 5% (11/204), p=0.3] (Table V). PSA control was also similar between those with highly frequent ARD (90% at 5 years) and those with less frequent ARD (86% at 5 years) (p=0.36) (Figure 1).
Discussion
IGRT has the potential to reduce unnecessary irradiation to organs at risk, and facilitates precise target location using soft-tissue matching, which is important in radiotherapy of prostate cancer because the position of the prostate gland varies with the filling and emptying of the bladder and rectum (9-11). Prostate matching was not possible in the bone-matching era as a result of the lack of soft-tissue contrast in portal images (1, 2, 4-6, 9-13). Soft-tissue matching IG-IMRT enables the use of smaller PTV margins compared to bone matching IMRT (4-6, 9-13). However, MVCT sometimes reveals irregular rectal morphological changes (e.g. partial rectal expansion), which is sometimes impossible to correct using couch adjustment (6).
Thus we examined the ARD phenomenon and its influence on toxicity and PSA control. Although we identified lower BMI and larger rectal volume as significant predisposing factors for ARD in a previous study (6), the latter was not confirmed by the present study. However, having a lean body (low BMI) was confirmed as a risk factor for ARD. A large rectal volume identified on planning CT images was sometimes correlated to rectal bleeding or PSA failure (14-18), which could partly be explained by unintended movement causing prostate and rectal displacement during radiotherapy. Stasi et al. also pointed out the importance of rectal volume control before and during radiotherapy (18). In the present study, we did not find any correlation of ARD with rectal toxicity and PSA control. We, therefore, believe our repeated precautions corrected rectal displacement and indirectly confirm the role of IG-radiotherapy, which may reduce unnecessary irradiation to the rectum. In other words, this finding indirectly confirms the safety of IG-IMRT in reducing rectal dose even when interfractional error occurs. The error due to intrafractional rectal motion may be too small to affect clinical outcome (PSA control).
The frequency [n (%)] of action required to correct rectal displacement (ARD) according to treatment year.
It is plausible that ARD occurs more frequently during the first week of treatment and then decreases because patients become used to the process of radiotherapy over time. Patients are tense and anxious during the first week but as they become accustomed to radiotherapy, the muscles tend to become more relaxed. It also true that there is a learning curve of medical staff because we recognized that ARD frequency has decreased with increasing years of experience. This information is helpful because rectal displacement can not only cause treatment failure or toxicity but also be a bothersome phenomenon for both medical staff and patients.
There are several limitations to the present study. Firstly, our data were derived from a single-institutional retrospective experience, which should be confirmed in prospective multi-institutional trials with a larger number of patients. Next, we did not assess the real-time rectal movement (intrafractional motion) cited by several methods (19-22) during irradiation in this study. At initial installment of tomotherapy, we performed analysis for intrafractional error in 10 patients (740 MVCT images) by comparison of pre- and post-treatment MVCT. The intrafractional motion was 0.03 (−1.3 to 1.4) ± 0.39 mm in the right-left dimension, 0.08 (−1.8 to 0.28) ± 0.73 mm in the superior-inferior dimension, and 0.52 (−1.8 to 1.8) ± 0.63 mm in the anterior-posterior dimension. Wust et al. also reported that daily MVCT-based IGRT without markers might be a valid alternative (21). In conclusion, ARD occurred predominantly in lean patients, during the initial week of treatment and became less likely over time. ARD was not correlated to late toxicity and PSA control. The IG-IMRT technique adequately controls interfractional error.
Late toxicity according to the frequency of action required to correct rectal displacement (ARD).
Footnotes
↵Conflicts of Interest
The Authors state no conflicts of interest exist in regard to this study.
- Received August 14, 2017.
- Revision received September 8, 2017.
- Accepted September 14, 2017.
- Copyright© 2017, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.