Abstract
Background/Aim: To evaluate if topical support therapy during static-intensity modulated radiotherapy (sIMRT) course is able to equal the characteristic minimum risk for radiation proctitis of Image-guided volumetric modulated arc therapy (IG-VMAT) treatment among localized prostate cancer patients. Patients and Methods: Rectal toxicity data of the above patients were retrospectively collected throughout three different clinical periods at our Radiotherapy Deparment: from October 2011 to December 2012, prostate cancer patients were treated with sIMRT and in advance supported by means of daily topical corticosteroids; from January 2013 to November 2016, topical corticosteroids were replaced by daily hyaluronic acid enemas; from December 2016 to May 2018 eligible patients were treated with newly introduced IG-VMAT supported by only on-demand topical corticosteroids. Results: Among 359 eligible patients, IG-VMAT was proven generally more effective than sIMRT supported by topical medications in terms of proctitis reduction, although without clinical and practical relevance. Conclusion: Topical medications might have a role in radiation proctitis prevention.
- Static intensity modulated radiotherapy
- prostate cancer
- radiation proctitis
- radiation induced adverse events
- image guided volumetric modulated arc radiotherapy
- topical support therapy
- hyaluronic acid enema
- beclomethasone dipropionate
- hypofractionated radiotherapy
Radiation induced proctitis (RIP) in prostate cancer patients submitted to curative radiation therapy (RT) is one of the most feared side effects, which negatively affects the quality of life (1). Such disease occurs with a rate of over 10% even using modern external beam radiotherapy techniques (2). The spectrum of symptoms is so wide that at least 5 distinct assessment scales, both objective (3-5) and subjective (6), have been proposed to classify the RIP severity. Whatever the symptoms experienced, a single pathogenesis is recognized, which identifies the onset of this process with a complex evolution of the damage in rectal mucosa cells (7). Radiation oncologists generally adopt two approaches in the attempt to decrease RIP occurrence: a) the use of topical medications or drugs that can be taken orally; b) the reduction in the amount of the rectum or its subvolumes (i.e. anterior or entire rectal wall) exposed to radiation by adopting more conformal radiotherapy techniques or using medical devices such as rectal spacer or gold fiducials.
The aim of the present article is to report the evolution of the rectal toxicity rate along a real-life daily clinical practice in a single Institution’s Radiotherapy Department that, throughout a decade, has experienced the use of new molecules for supportive therapy in prostate cancer patients at risk for radiation proctitis and the implementation of a more modern radiotherapy technique.
Patients and Methods
We retrospectively analyzed the recorded rectal toxicity data of prostate cancer patients curatively treated in our center from October 2011 to May 2018. During that time our clinical practice changed progressively, both in terms of the kind of rectal supportive therapy and the technical approach. In fact we are able to distinguish three different historical periods in our radiotherapy department: i) patients irradiated from October 2011 to December 2012 were submitted to multiple static Intensity modulated radiotherapy (IMRT) technique and in advance supported by means of daily topical corticosteroids: namely rectal beclomethasone dipropionate suppositories (group A); ii) patients treated from January 2013 to November 2016 received RT using the same technique, but topical corticosteroids were replaced by the use of daily rectal hyaluronic acid (HA) enriched with a pool of amino acids (glycine, L-proline, L-leucine, L-lysine hydrochloride) enema (group B); iii) patients treated from December 2016 to May 2018 using Volumetric Modulated Arc Therapy (VMAT), with an image-guided set-up for a more accurate and rectal sparing dose delivery (IGRT-VMAT) without the use of systematic therapy, which was administered only on-demand at the time of RIP occurrence (generally topical corticosteroids) (group C). Both the group A and B were invited to daily rectal self-administration at night before sleeping, starting from the first day of radiotherapy, for its entire duration and up to two weeks after the end. Patients eligible for this study had to meet the following eligibility criteria: localized prostate cancer, biopsy-proven diagnosis, no anorectal symptoms at baseline, signed informed patient consent, total radiation dose ≥60 Gy, normo- or hypofractionated schedules (range=2-2.7 Gy/day), follow-up period of at least 24 months, no inflammatory bowel disease, no previous pelvic radiotherapy, no anorectal diseases (i.e. hemorrhoids, fissures, fistula, cancer, etc.). Two requirements were extremely important for CT simulation and daily treatment of all patients: empty rectum and full bladder. To achieve this, patients were advised to follow a high fiber diet and to drink 500 ml of water 30 min before set-up positioning and daily treatment. These tricks allow to reduce the amount of bowel loops and rectum exposed to radiation dose. CT simulation was performed in the supine position with 3 mm thickness slices from bi-iliac line to a transverse plan passing through the proximal and middle thirds of femurs. A dedicated immobilization device was used to loosen the tense muscles of back or legs and to ensure patient comfort. IMRT treatments were made with 6 MV energy photons and multiple static fields (seven fields technique); set-up verification was conducted by daily matching of bony landmarks between Digital Reconstructed Radiographs (DRR) and orthogonal 2D megavoltage (MV) electronic portal imaging device (EPID) images (anteroposterior and lateral). IGRT-VMAT treatments were made with the same energy photons and two coplanar arcs (one clockwise from 210° to 150° and the other one anti-clockwise from 150° to 210°); in this case set-up verification was conducted by daily matching of prostate (if present) or prostate bed (if adjuvant) and other nearby soft tissues/pelvic structures between planning CT scan and an online high-quality kV-cone beam CT (kvCBCT) for a three-dimensional evaluation of target position. In static intensity modulated radiotherapy treatments the clinical target volume (CTV) included prostate±seminal vesicles or prostate bed and was enlarged to generate the planning target volume (PTV) through an asymmetrical margin of 10 mm in every direction, except posteriorly where the presence of anterior rectal wall limits its expansion to 5 mm; for the more accurate IGRT-VMAT treatments these margins were 30% narrower: 7 mm in the cranio-caudal, latero-lateral and anterior direction, 3 mm in the posterior one. For the contouring of the organs at risk we relied on the Radiation Therapy Oncology Group (RTOG) atlas. The dose prescription was made in accordance with International Commission on Radiation Units & Measurements (ICRU) recommendations and the Quantitative Analyzes of Normal Tissues Effects in the Clinic (QUANTEC) dose constraints, such as the following: V50 <50%, V60 <35%, V65 <25%, V70 <20%, and V75 <15% for conventional fractionation, V32 ≤50%, V50 ≤25.8% and v60 ≤10% for hypofractionation. Acute (during or within 3 months after RT) and chronic (from 3-6 months after the end of treatment) rectal toxicities were assessed according to the RTOG criteria and the RTOG/European Organization for Research and Treatment of Cancer (EORTC) criteria and recorded weekly during radiation and every two months after the end. For each level of toxicity, the highest grade experienced by the patient is reported. All procedures performed in this study were in accordance with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants included in the study.
Statistical analysis. The endpoint of the study was evaluation of acute and chronic rectal toxicities rates in the three patient groups. Associations between variables were analyzed; chi-squared test was used for estimating group differences between the three arms. Fisher’s exact test (two-tailed) was used for analysis of the differences in terms of proctitis frequency or severity between IMRT (medications group) and IGRT-VMAT cohorts. Statistical analysis was performed using MedCalc for Windows, version 19.4 (MedCalc Software, Ostend, Belgium). p<0.05 was considered statistically significant.
Results
In total 359 patients met the eligibility criteria: 93 comprise group A, 175 group B and 91 group C. The median age of the whole population was 72 years (range=47-86 years). Two hundred and thirteen patients were treated definitively, and 146 in the adjuvant setting. Two hundred and seventy patients were treated with a normofractionated schedule (2 Gy/day), 89 with a hypofractionated schedule (2,7 Gy/day). Patient characteristics are summarized in Table I. Concerning acute rectal toxicity, 190 grade G0 cases, 148 G1, and 21 G2 were recorded, respectively. Regarding chronic rectal toxicity, a total of 304 grade G0 cases, 43 G1, and 12 G2 was registered. No grade G3 or more was reported in any of the three groups. HA enema reduced the severity of acute and chronic rectal toxicity when compared to bechlometasone dipropionate suppositories, but not its global incidence, as evidenced by a significantly lower grade G2 toxicity in the group B (p<0.05) in the face of a comparable rate of grade ≥G1 toxicity rate with respect to group A. Putting together group A and group B into the medications group and comparing this with group C we were able to state the following: 1) No acute toxicity rate was significantly higher in the IGRT group compared to the medications group (Figure 1); 2) in the medications group both acute G1 and G2 toxicities were significantly more frequent than in the IGRT group (Figure 2); 3) the IGRT group confirmed the advantage over the medications group even if considering exclusively the patients treated with hypofractionation of the radiation dose (Figure 3); 4) the IGRT treatment for prostate bed involved a significantly lower risk of acute (p<0.001), but not chronic (p=0.44), toxicity ≥G1 when compared to multiple static fields IMRT and medications (Tables II and III); 5) overall, chronic rectal toxicity was not considerably different between the medications group and the IGRT group, also in the subgroup analysis, that is evaluating hypofractionated treatment or prostate bed targeting as discriminating factors (Figure 4). Considering the whole cohort, the hypofractionation of the radiation dose did not show a higher frequency of rectal toxicity (G0 vs. >G0). Considering only clinically relevant toxicities (>G1), no significat differences were found between 3D-IGRT and medications group. Also, significant differences were not found at the subgroup analysis according to fractionation (conventional versus hypofractionated) or radiation therapy site (prostate versus prostate bed) (Figures 5, 6, 7, 8, 9 and 10). Within the hypofractionated subgroup (89) none of patients treated with 3D-IGRT complained for acute >G1 proctitis and only one experienced a chronic G2 proctitis in comparison with 4 and 1 cases, respectively, in the medications counterpart (p>0.05). All patients completed the planned schedule with a good tolerance to radiotherapy treatment, with the exception of one patient who voluntarily decided to interrupt the treatment for grade G2 genitourinary toxicity when 54 Gy out of the programmed 60 Gy dose was reached. Compliance with self-administration of medications was good in all patients. Only one case out of 12 grade G2 chronic toxicity needed topical formalin instillation for persistent intermittent rectal bleeding, risk for anemization and resistance to other conservative medications; in the remaining 11 patients chronic G2 toxicity regressed spontaneously or by means of topical corticosteroids. No difference in acute and chronic RIP was observed depending on patients’ age.
Discussion
Our study demonstrated that IMRT for prostate (or prostate bed if post-operative) target, both with static fields and with volumetric modulated arcs, offers an acceptable rate of acute and chronic rectal toxicity: a median dose of 80 Gy for the group treated with conventional fractionation and 70.2 Gy for the hypofractionated group produced no episode of radiation proctitis ≥G3 and few cases of G2. Thus, the escalated dose approach by IMRT delivery did not appear to affect rectal tolerance. Overall, the 3D-IGRT-VMAT resulted in a lower frequency of acute but not chronic rectal toxicity (although in the latter case there was still a trend toward statistical significance: p=0.06) compared to static 2D-EPID-IMRT. This observation applied to the entire sample and did not seem to be influenced by the type of radiotherapy dose fractionation or by the target for treatment (prostate vs. prostate bed). It was evident that the administration of medications simultaneously with the delivery of an IMRT plan verified by 2D-EPID failed to fill the gap in terms of rectal toxicity reduction with the 3D-IGRT-VMAT: a more accurate set-up verification would seem to be more effective in preventing radiation proctitis compared to a topical therapy based on hyaluronic acid or corticosteroids. However, if we want to discriminate for the clinical relevance of rectal toxicity (≤G1 vs. >G1), i.e. purely considering episodes of acute and chronic proctitis requiring medications or analgesics (>G1), these were not significantly more frequent among patients underwent 2D-EPID-IMRT compared to those treated with 3D-IGRT-VMAT. Such a finding did not appear to be affected by the type of dose fractionation or by the therapeutic aim (radical vs. postoperative). This would confirm the adequacy and the still good performance of static IMRT assisted by a verification system through 2D-EPID and by topical supportive therapy. Similar results are therefore not sufficient to discourage radiation oncologists who do not have a verification system using Cone Beam CT, even more so that the late rectal toxicity rate and severity grade did not differ significantly between the two radiotherapy techniques used in the present study. The already low toxicity profile of 3D-IGRT-VMAT could be further improved by the topical administration of medications, even different from those used in this study. Such a hypothesis could be addressed in a specific clinical research.
Our results are significantly better than those found in the cohort treated with External Beam-IGRT by Mohammed et al. (8); these authors reported acute and late ≥Grade 2 gastrointestinal (GI) toxicities in 49% and 20% of patients, respectively, treated with an unverified daily 3D-CRT technique and without any information on supportive care for proctitis prevention. In the work of Gill et al. (9), the acute ≥ Grade 2 proctitis rate did not differ significantly between IGRT and non-IGRT groups (6% vs. 15%, p=0.08), as in our case. By keeping with radiation dose constraints and in agreement with those suggested by Pederson et al. (10), the late ≥ Grade 2 GI toxicities rate in our experience was even lower than expected (3.3% versus 9%), probably due to a systematic use of supportive therapy during treatment or, alternatively, to a daily 3D-IGRT. This latter option, as demonstrated by Wortel et al. (11-13), further reduced proctitis rate, improving what was already proven for IMRT in respect to 3D-CRT (14, 15). This was due to a more accurate set-up verification through direct visualization of soft tissues by means of kV Cone Beam CT, whose potential in terms of safety for dose delivery can be further increased with implanted fiducial markers for prostate positioning (16, 17), surpassing and supplanting bony landmark based matching (18). The online set-up verification through kV CBCT immediately prior to daily dose delivery allows to minimize the risk of late ≥ Grade 2 proctitis, even with radiotherapy techniques typically less conformal and homogeneous than IMRT. For example, Hama et al. reached a rate of grade 2 late GI adverse events equal to only 3.8% at 5 years by adopting a simplified 3D-CRT that involved the use of two AP (anterior posterior)/PA (posterior anterior) and two lateral fields coupled with the live CBCT-guided positioning of the edge of the lead blocks on the boundary line between the prostate and the anterior rectal wall to maximize rectum sparing (19). The effectiveness and the need for image-guided dose delivery had already been hypothesized from preliminary experiences such as that of Bohrer et al. (20) that, by means of a daily stereotactic ultrasound positioning for prostatic target, reported a significantly lower risk of proctitis than that of non-IGRT patients against the background of a no longer used not-escalated dose. The extreme reliability of the IGRT for prostate cancer allows the delineation of tighter margins for PTV and consequently the reduction of harmful radiation exposure of nearby healthy tissues, including the rectum (12). Such an assumption was valid also for 2D-EPID-IGRT through the visualization of metallic stents inside the prostate on the orthogonal anterior-posterior / lateral Xrays pairs for set-up verification (21). The set-up verification system for prostate cancer patients through orthogonal portal images matched by bony landmarks with digital reconstructed radiographs does not allow to accurately visualize the amount of rectal gas, unless it is very large, unlike high-quality Kv CBCT imaging. Nonetheless, the real burden of changes of gas content on proctitis risk has been questioned by Sasaki et al. (22, 23). Actually, these authors showed that rectal gas had a minimum effect on dose distribution in the rectum: the lower gas-related dose scattering led to a decrease of rectal V75Gy on the boundary between dorsal target side and ventral OAR side, simultaneously to a clinically irrelevant increase of rectal volume exposed to moderate dose (≤60 Gy). In addition, the presence of gas within the rectum caused a reduction in the minimum dose, V100% and Dmean for CTV: the decline of dose scattering reduced the energy deposit where the dose was already lower (i.e. posterior edge of target) and, in parallel, facilitated the photon trasmission in high-dose regions (i.e. more pronounced hotspots). These authors concluded that the presence of gas did not alter the rectal tolerance and radiotherapy treatment may be safely delivered. Notably, the RT techniques less affected by gas content were static IMRT and full arc VMAT, similar to those used in our work. It is unclear whether these dosimetric perturbations are prognostically decisive in terms of local control and then of the related oncologic outcomes. These uncertainties should be addressed to a specific trial. According to the above results, we hypothesize that EPID-guided bony landmarks matching for set-up verification may be enough for a safe dose delivery at the level of the peri-target rectal area. On the other hand, an excessive gas volume could result in a significant prostate displacement and therefore radiotherapy target missing. In order to avoid this, the value of kvCBCT is irreplaceable. In our experience we used a reduced posterior margin for PTV among patients treated with daily 3D-IGRT-VMAT: this factor might have contributed to its better tolerance with respect to static 2D-EPID-IMRT. Conversely, the better CTV delineation related to the use of Magnetic Resonance imaging has not been shown to lead to a significant reduction in the risk of proctitis (24). For such reason, we deemed sufficient the target delineation on CT-scan images. Different quantitative and qualitative dosimetric approaches have been vetted in order to identify which one is the best predictor of radiation proctitis in recent years. Wortel et al. suggested that spatial information about dose distribution obtainable from Dose Surface Maps could be more reliable than classical Dose Volume Histogram (DVH). Moreover, these authors reported that IG-IMRT patients with acute grade ≥2 proctitis had a worse spatial distribution of moderate dose levels (≈25-50 Gy) (25). Such a finding was not confirmed by the work of Ishii et al. (26) who demonstrated no more frequent proctitis rate among patients with larger rectal volume exposed to low- and medium-dose levels. We have strictly followed the QUANTEC suggestions regarding dose limits for organs at risk. Paleny et al. (27) argued that those relative dose volume limits were effectively representative of risk prediction of proctitis, excluding the need for absolute dose volume parameters for RT in prostate cancer patients. Nonetheless, Ozkan et al. (28) questioned the robustness of QUANTEC constraints and suggested that new more reliable dosimetric goals should be proposed to limit acute rectal toxicity more effectively in prostate cancer patients during IMRT treatment. We believe that this last proposition deserves to be brought forward to better suit the technological potential of recent high-performing radiotherapy techniques (29). Regarding the rectal toxicity rate in the subgroup of patients treated at the prostate bed level, our results are in agreement with those published by Borghetti et al. (30); we also recorded a significantly lower risk of rectal disorders among patients treated with IGRT-VMAT. Among our series of prostate cancer patients supported by medications during 2D-EPID-IMRT treatment we reported an acceptable clinically relevant proctitis rate. Such a finding could decrease the need to resort to invasive procedures (i.e. rectal spacer or retractor) to limit the injurious radiation exposure of the anterior rectal wall (31, 32). These surgical procedures, even though extremely promising to decrease rectal toxicity, are still burdened by a not insignificant risk of iatrogenic major adverse events (33, 34). The medications used in our experience had a very low risk profile and, in terms of healthcare costs, are far less expensive than the above invasive solutions (35). By still focusing on the cost-effectiveness of IGRT and the good rectal tolerance related to the support of simple and effective topical medications to a less advanced IMRT treatment (36, 37) among our localized prostate cancer patients, we are able to state that our results are consistent with those obtained by Tøndel et al. (38). In fact, these authors reported no clinical benefit of daily online CBCT IGRT plus tighter margins for prostate target (7 mm) over weekly offline portal imaging plus larger margins to PTV delineation (15 mm), so questioning the need for and the costs of IGRT (39). Nevertheless, it must be emphasized that a lower overall risk of rectal disorder, that is by preventing also acute G1 events during treatment, can improve patient compliance with the planned RT schedule. However, from this perspective, we reported no RT breaks due to rectal discomfort. In our series daily hyaluronic acid enema was not inferior to beclomethasone suppositories; thus, the first medication could have a further advantage, apart from a practical nil pharmacologic risk, over the second one, that is the possibility of administration even in those cases in which the use of corticosteroids is contraindicated (i.e. infections, perforations, obstructions, local abscesses). Current literature is deficient in terms of level 1 evidence regarding the best supportive care for proctitis, thus allowing broad scope for the physician’s discretion based on his own experience (40). Our results confirm the feasibility of aggressive RT in older patients; in fact, we have not observed an excessive rate of RIP in any of the three groups despite a median age of 72 years (41). Actually, it is desirable that the findings of this study will be confirmed by a prospective trial. In fact, in this way, our experience is marred by some obvious critical issues, such as the presence of a noncontemporaneous control bias (due to the consecutiveness of the data collection) and the lack of an instrumental evaluation (i.e. rectosigmoidoscopy) for a cross-check of the subjective assessment reported by the patient. Furthermore, the particular route of administration in patients treated with topical therapy is traditionally disliked and although in this series an excellent obligingness to medical prescription has been reported, the request for administration of the drug at home (to reduce patient discomfort), not under direct medical supervision, could have led to an overestimation of the actual compliance and, consequently, an underestimation of the therapeutic power of the two medicaments here investigated. By trusting in what patients reported to us, our work confirmed the advantage of 3D-IGRT-VMAT over multiple static IMRT coupled with a set-up verification based on 2D-EPID in terms of reduction of the severity of acute radio-induced proctitis among prostate cancer patients and that this gap cannot be bridged by the addition of topical support therapy based on corticosteroids or HA. Clearly an accurate set-up verification is the best method of prevention. Relative Dose Volume planning aims by interpreting traditional DVH were confirmed as reliable in our work. To date the peculiarity of this study is that, to our knowledge, it is the first to compare 2D-EPID-IMRT systematically supported by topical medications with 3D-IGRT-VMAT for prevention of radiation proctitis among localized prostate cancer patients.
Conclusion
The IMRT, both with static and volumetric fields, allows a very low clinically relevant proctitis rate among localized prostate cancer patients. The IGRT would seem to further reduce the overall risk of proctitis and this could imply a better adherence of the patient to the therapeutic protocol. Topical supportive therapy with hyaluronic acid enema or beclomethasone suppositories might have a role in preventing RIP through an immediate restorative action on rectal mucosa strained by radiation. Further studies, of the prospective trial type, are needed to understand if and when IGRT is really necessary.
Footnotes
Authors’ Contributions
GF conceived and supervised the study. VU collected the data. VV performed the statistical analysis. GF wrote the manuscript with support from LM. GRB, VAM and SII performed the calculation of treatment plans. GF, GRB, IRC, AT, VV, SP, SL and AC contributed to the interpretation of the results and literature research.
Conflicts of Interest
The Authors report no conflicts of interest in relation to this study.
- Received February 13, 2021.
- Revision received February 22, 2021.
- Accepted February 23, 2021.
- Copyright © 2021 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.