The principle of radical radiotherapy is to maximize tumor control while minimizing radiation damage to surrounding normal tissues. With the advancement of radiotherapy techniques, especially the application of intensity-modulated radiotherapy (IMRT) and image-guided radiotherapy (IGRT) in recent years, the tumor control rate and patient quality of life have been significantly improved [1,2]. However, toxic side effects associated with radiotherapy are still present, which may be related to the significantly higher doses of radiotherapy [3]. Late rectal reactions caused by prostate radiotherapy are a great concern. Rectal injury associated with radiotherapy includes frequent bowel movements, urgency, diarrhea, and fecal incontinence caused by rectal irritation, and blood in the stool and abdominal pain are sometimes present. These symptoms either mildly or severely affect the quality of life of patients [4]. Several large-scale retrospective studies have analyzed the influencing factors of acute and late rectal injury to provide some reference for the selection of treatment options for patients; the relationship between rectal injury and rectal dose-volume has also been analyzed and several predictive models of rectal injury have been established [5-16]. It is now believed that appropriate dose-volume restriction of the rectum, depending on the patient’s specific situation, is the most effective way to reduce rectal injury [17-21]. If IGRT can be used daily, we can appropriately reduce the extent of CTV to PTV outreach, thus reducing the extent of the high-dose area of the rectum without compromising the treated tumor [22-24]. The placement of separators, such as hyaluronic acid and polyethylene glycol, between the prostate and the anterior rectal wall allows the anterior rectal wall to be set back, reducing the dose to the anterior wall [25-28], and although this method has limited usefulness for conventional separation irradiation, it can significantly reduce the incidence of rectal injury for large split radiotherapy with a single dose of more than 5 Gy [29]. Other methods such as endorectal hydrocolloid (to reduce prostate motion) [30] and drugs (to protect normal tissue) [31-33]. In this review, we will describe 1. the symptoms, causes, and incidence of rectal injury caused by radiotherapy for prostate cancer; 2. how dose-volume limitation is applied to the rectum; 3. changes in rectal location, volume, and actual received dose during image-guided intensity-modulated radiotherapy and how they are measured; 4. predictive models of rectal injury used internationally; and 5. methods to mitigate rectal injury after radiotherapy for prostate cancer, including technical, physical and biological methods. 1. rectal injury caused by radiotherapy For many years late rectal bleeding (late means at least 6 months after the end of radiotherapy) has been the only endpoint for determining rectal injury after radiotherapy, because it is a relatively objective observation. However, acute and late gastrointestinal reactions can have many manifestations: blood in the stool, rectal stricture, decreased rectal compliance, and decreased rectal volume, where the decreased rectal volume can in turn cause fecal incontinence and frequent rectal peristalsis. Damage to the muscles of the rectum and anal canal area can lead to fecal incontinence and anal stricture. The pathophysiologic mechanisms of rectal injury are complex. Factors contributing to rectal damage include: rectal mucosal injury, rectal volume changes, sensory function, rectal distensibility, function of the anal sphincter, changes in pelvic floor structures due to altered anal canal pressure, nerve damage, and fecal traits [34]. Knowing these possible factors, all presented rectal injuries should be able to find their corresponding causes and lesion sites [45], and by analyzing which sites may cause late radiotherapy reactions, some methods to avoid and mitigate the reactions should be taken in advance. The radiation damage grading criteria of RTOG and EORTC and CTCAE 4.0 scoring criteria are commonly used for grading adverse reactions, and the latter is more detailed in describing specific toxic symptoms. Summarizing data from previous studies, moderate to severe gastrointestinal reactions ranged from 2% to 20% [36-41]. There are several reasons for such large differences: different prescription doses (70-90 Gy), different radiotherapy techniques (conventional radiotherapy, 3D-CRT, IMRT, IGRT), different ranges of clinical target areas (prostate, prostate + seminal vesicles, whole pelvis), and whether endocrine therapy is combined. To determine the damage to the rectum, the first step is to define the minimum follow-up years. Rectal injury caused by radiotherapy is a late response and is generally considered to require a minimum follow-up of 30-36 months for a true assessment. There is a maximum state of severity for any toxic side effect, i.e., there is a peak. In the case of rectal injury, we define as a peak: a moderately severe event (e.g., rectal bleeding) that requires medical intervention. At the same time, the rectal response may be accompanied by some chronic, progressive damage (e.g., fecal incontinence). Such symptoms are characterized by a longitudinal progression over time and are used to determine the average degree of injury over a longer period of time [42-44], so this longitudinal degree of injury can be used as a measure of severe, persistent symptoms. To summarize, to describe radiotherapy-related rectal injury, it is important to first be clear which symptoms are rectal injury, and then to know the causes of these symptoms, the requirements for follow-up time, the grading of toxicities, and the definition and selection of peak and mean toxicities. The above definitions are described separately in the above article. 2. Rectal dose-volume limitation The dose-volume limitation of the organ at risk is a very important topic. In the past, different studies have used different methods to define the rectal contour, including the rectal volume outlined on CT (including the rectal wall and rectal contents air or feces) and the internal and external contours of the rectal wall outlined manually. Different DVH maps were obtained with these two methods. Overall, similar rectal dose-volume constraints were obtained for all studies. In two reviews in this area [45,46], the dose-volume limit was controlled to 40-75 Gy in order to keep the late rectal response below 15% for grade ≥2 and 10% for grade ≥3. The specific dose varied slightly among reports, probably V40 Gy <60-80%, V60 Gy <35-45%, V70 Gy <15-25% , V75 Gy <5-10 % [45-50]. It is important to note that the clinical studies that proposed the above dose limits mainly used late rectal bleeding as an evaluation criterion for injury. Other reactions to injury, such as late fecal incontinence and frequent stools, have been reported by investigators only after 2006. Although the incidence of these reactions is very low (5%), they significantly affect the daily life of prostate cancer patients as chronic symptoms. Therefore dose limitation is again proposed: V40Gy < 80%, or mean dose < 45-50gy [51,52]. Meeting this requirement keeps the incidence of injury below 1-2%. Generally these dose limits are easily achieved with the use of intensity modulation techniques, while for 3D conformal radiotherapy, the ability to meet such dose limits depends on the anatomical structure and the prescribed dose. Also regarding the dose-volume limitation of the pelvic floor muscles, in 2012 Smeenk.R.J [35] et al. reported the relationship between dose-volume limitation of the pelvic floor muscles and fecal incontinence and fecal urgency. To reduce the occurrence of fecal urgency and incontinence, the study proposed several average dose limits: 30 Gy for the internal anal sphincter, 10 Gy for the external anal sphincter, 50 Gy for the puborectalis muscle, and 40 Gy for the anal levator muscle. 3. Changes in rectal position, volume, and actual received dose during image-guided intensity-modulated radiotherapy and measurement methods Several papers have reported on changes in the volume and dose of the rectum during radiotherapy. In 2006, MD-Anderson reported 10 patients with prostate cancer [53], each of whom underwent daily IGRT-IMRT with a total of 390 repCTs, together with 10 pCTs, in which the same physician outlined the target area of 400 CTs and "ported" the planning system to repCT to calculate In 2010, the Fox Chase Center [54] reported 20 patients who underwent IGRT with 139 CTs, requiring V65 ≤ 17% and V40 ≤ 35%, and counted the actual doses received in the rectum, with 27% and 28% of patients not meeting these requirements, respectively. Neither of these two studies involved the occurrence of adverse effects, but their methods of measuring rectal volume and dose changes during radiotherapy are worth studying. In 2010 J.A. Hatton et al [55] compared the DVH maps of prostate, rectum, and bladder before and during radiotherapy in 12 patients with prostate cancer, where the rectum was compared with the planned volume of 40 Gy, 60 Gy, and 70 Gy during radiotherapy, and the results were statistically significant in 9 patients. In 2013 Maria Thor et al [56] compared the relationship between rectal bladder volume, dose changes and late toxicities during image guided intensity modulated radiotherapy in 38 patients with locally progressive prostate cancer, which is the first study in recent years on the relationship between dose volume changes and clinical adverse effects during radiotherapy. 38 patients did not undergo special rectal bladder preparation before radiotherapy, and the prostate The three CTVs were defined as CTV67.5 (prostate + affected seminal vesicles), CTV60 (prostate + seminal vesicles), and CTV50 (CTV60 + pelvic lymph nodes). The risk organs were outlined in the following areas: rectum (recto-B junction to the lower edge of the anal canal), bladder (bladder neck to bladder floor). CT (repCT) is performed twice a week during the course of radiotherapy, and the initial treatment plan is then "transposed" into the repCT, and the actual doses received by the prostate, rectum and bladder during treatment are calculated. The Eclipse v.10.0 system was used. The study counted changes in rectal and bladder volumes before and after treatment and found that, in general, rectal volumes were smaller for ≥2 degrees GI reactions compared to minor adverse reactions, and in particular, statistically significant values were obtained when comparing the actual rectal volumes where acute GI occurred (≥2 degrees vs. 0-1 degrees: 75 vs. 23.5px3). This result can be explained in this way: because the more severe rectal reactions are usually manifested by diarrhea, frequent stools, or even pus and blood stools, the reduction of rectal volume caused by these changes is a normal physiological phenomenon. Moreover, when the rectal content decreases during the actual treatment, the rectal volume decreases, and a larger proportion of the rectal wall will enter the high-dose area directly, so more serious adverse reactions will occur. Therefore, to disregard the effect of rectal contents, the study also analyzed rectal DWH maps (doseCwall histograms) and found that rectal dose volume limitation in DWH maps correlated more closely with adverse reactions than in DVH maps. The study also compared the bioequivalent uniform dose (gEUD) of rectum and bladder during planning and treatment and found that the dose of the more heavily reactive organs was high, where the relationship between rectal volume and the degree of acute GI reactions for equivalent doses (EQD2) above 76 Gy was compared with rectal DWH maps and the results were statistically significant. Due to the small number of cases in this study, some results are open to debate. However, the idea and method of this study are worthy of our study, such as we can compare the relationship between the percentage of volume in the rectal high-dose area and the degree of occurrence of adverse reactions during the actual treatment. 4. Predictive models Several large-scale prospective clinical studies have found that late rectal injury is significantly associated with clinical features and dosimetric distribution, and the incidence of toxic side effects can be predicted. At the same time, there are several diseases that can increase the gastrointestinal response in radiotherapy patients, such as history of abdominal surgery before radiotherapy, history of cardiovascular disease, history of diabetes mellitus, anticoagulant use, hemorrhoids, etc. Some studies have also emphasized the continuous effect between late response and acute phase response, suggesting that acute phase gastrointestinal response is an independent predictor of late response. Although we have some understanding of the dosimetric and clinical factors affecting rectal injury, there are still some patients who present with responses that we did not expect, suggesting that certain genetic factors of radiosensitivity may also be present in these patients. There is some evidence on genetic susceptibility to advanced rectal bleeding: Burri [57] and Damaraju [58] et al. suggested that susceptibility to late side effects after radiotherapy for prostate cancer may be associated with single nucleotide polymorphisms (SNPs) in SOD2, XRCC1, XRCC3 or in XRCC3, LIG4, MLH1, CYP2D6, ERCC2 presence was associated. In addition, it has been demonstrated that decreasing the expression of microRNAs in AKR1B1, BAZB1, LSM7, NUDT1, PSMB4, SEC22L1,UBB can increase the late toxic effects. Increasing the expression of microRNAs in DDX17, DRAP1, RAD23 and SRF predicts resistance to late rectal bleeding [59]. However, these studies are only from individual centers and the results are not yet identical from study to study. In addition to these models for biomarker studies, several statistical models on dosimetric and clinical information have been developed internationally to predict individual toxicities related to radiotherapy. If the model predicts a high probability of rectal toxicities in that patient, the treatment regimen may be optimized and modified accordingly. If external radiotherapy is not the best option, other treatment options, such as brachytherapy or surgery, may be selected by this model. The first type of model is the NTCP model (normal tissue complication probability) [69]. This model is a mathematical, biophysical model that predicts the likelihood of normal tissue adverse reactions in a certain percentage of patients through certain calculations. This model is based on the dose distribution of the DVH map of the target organ, and calculations are performed to finally obtain the degree of risk of toxic side effects. When the dose and irradiation volume increase, the NTCP increases, depicting a dose-NTCP relationship plotted as an S-shaped curve. Although the dose-volume relationship is different in different organs, this relationship is basically defined as a mixture of two extreme cases: tandem organs consider the maximum dose, parallel organs describe the median dose, and mixed organs have to consider both median and maximum dose. Recently, NTCP models have been proposed to predict clinical risk factors.DeFraene et al [14] developed a model (512 patients, 36 months follow-up data) to predict grade 3 rectal bleeding response (including abdominal surgery and cardiovascular disease), grade 3 late fecal incontinence (including abdominal surgery and diabetes) and grade 3 late increased stool frequency (including high stool frequency at baseline level). There is also a survey [15] that published NTCP models for 669 patients after 36-month follow-up, grade 2-3 advanced rectal bleeding (including abdominal surgery and acute phase injury), severe chronic fecal incontinence and mean fecal incontinence (including colonic lesions and abdominal surgery).NTCP models can be used to optimize treatment planning and individualize treatment. The second type of model is somewhat different in that it is a multivariate model that can predict all relevant variables (dose, clinical factors, genetic and molecular factors) to obtain an individualized assessment of the risk of toxicities. Using models that predict the possibility of very high and unacceptable toxicities in certain patients prior to radiotherapy, radiotherapists will limit the normal organ dose-volume distribution relatively more strictly to reduce adverse effects. These models have now been used to predict late rectal bleeding and late fecal incontinence and present the results of the analysis in clear and understandable graphs [61-64]. However, there is no model that is widely accepted and used. Because there are differences in treatment providers and treatment presence, models cannot be applied to all individuals. To date, no model has been able to include all potential differences (planned and actual dose differences), and none of them takes into account the spatial distribution of dose due to the influence of organ-tissue radiosensitivity heterogeneity. Despite its many shortcomings, predictive models have after all helped us to some extent to reduce the occurrence of rectal injury due to radiotherapy in those patients who are radiosensitive. In terms of individualized treatment, we can identify those who are at high risk and observe this group with close follow-up, and we can prevent drug interventions. 5. Ways to mitigate rectal injury 5.1 Technical advances in radiotherapy Several studies have shown that local control of prostate cancer can be improved by increasing the prescribed dose. However this advantage is offset by the ensuing gastrointestinal and urological adverse effects, especially in the era of 3DCRT [65-69]. With the improvement of planning systems and multi-leaf grating techniques, intensity-modulated radiotherapy has matured. Several studies, particularly the Memorial SloanCKettering Cancer Center high-dose radiotherapy study, showed that high-dose IMRT did not increase toxicities.Zelefsky et al [18] reported that 1571 patients received 3DCRT or IMRT radiotherapy at doses of 66-81 Gy, all IMRT patients received a dose of 81 Gy, with a median follow-up of 10 years. The incidence of ≥ grade 2 GI responses was significantly lower than in the 3DCRT group (5% vs 13%, P < 0.0001). Cahion et al [19] reported that 478 patients received 86.4 Gy IMRT with a median follow-up time of 4.4 years and grade 2 and grade 3 responses of 3% and 1%, respectively. These data suggest that the prescribed dose can be increased to 78-80 Gy under intensity-modulated radiotherapy, with the incidence of periodic toxicities comparable to that of conformal radiotherapy at 70 Gy. Compared with conformal radiotherapy, intensity-modulated radiotherapy can reduce the PTV boundary. Therefore, it is difficult to say whether the reduction in toxicities is due to the optimization of the dose distribution of intensity-modulated radiotherapy or to the reduction of the radiation field. The ability to more precisely target organ locations through daily image guidance allows for a further reduction in toxicities, which is certain. Several methods are available: intraprostatic implantation of metal markers [70], daily pelvic CT scans [71], and intraprostatic implantation of magnetic sensing devices [72]. The aim of all these methods is to allow more accurate localization of radiotherapy. With IGRT, the PTV extent of the prostate is reduced, thus protecting a portion of the anterior rectal wall from high-dose irradiation. A study by Zelefsky et al [73] at New York Memorial Hospital demonstrated that daily IGRT reduced advanced urological and rectal injury and that the reason was analyzed precisely because of the reduction of the CTV outreach. Currently at this oncology center, the posterior prostate outgrowth continues to be reduced from the original 6 mm. 5.2 Endorectal balloon In patients undergoing IGRT-IMRT, daily insertion of an endorectal balloon (latex or silicone balloon filled with air or water) into the rectum reduces the movement of the organ during each irradiation [30]. Several studies [74-78] compared prostate motility with and without the use of balloons. It can be seen that the use of a water balloon has a clear advantage in limiting the boundaries of CTV and PTV. However, because of the different methods and techniques used by different institutions, no definite conclusions can be drawn at this time. Therefore, we need to do further work to verify and correct the protocol of balloon placement. It was found that the placement of the balloon resulted in a significant reduction in the dose to the posterior wall of the rectum with medium to high doses of radiation therapy. The larger the balloon, the lower the dose to the posterior wall. This is because the balloon filling the rectum pushes the anterior wall of the rectum toward the anterior high dose area while keeping the posterior wall away from the high dose area. When the CTV contains a seminal vesicle, we would expect the balloon volume to be larger. Currently, there are few reports on the incidence of rectal injury after balloon use. Only one study directly compared 67.5 Gy 3DCRT radiotherapy [79] with and without the use of an endorectal balloon (n=24). Few late rectal injuries and mucosal injuries occurred in the group with balloons compared to the group without balloons. Due to the small sample size, there was no statistical significance. Regarding the endorectal balloon, a large prospective clinical study is needed to confirm whether it can truly translate this dosimetric advantage into an advantage in mitigating toxic effects. 5.3 Tissue fillers The intensity of radiation decreases as it passes through a certain distance. When we use appropriate objects to replace normal organs close to the target area, we can reduce the volume of normal organs located in the high dose area and thus mitigate toxic side effects. The anatomy between the rectum and prostate is well suited for the insertion of tissue fillers. the Denonvilliers crypt, also known as the rectal prostatic crypt, is located behind the prostate and separates the prostate and bladder from the rectum. The components of this crypt are dense collagen, smooth muscle and elastic fibers. Analysis of radical prostatectomy samples showed that tumor progression may invade this crypt, but not beyond. It is possible to separate the Denonvilliers' crypt from the rectal mesentery, thus increasing the distance from the rectum to the prostate. Several clinical studies have attempted to fill this anatomical gap with different methods, thus reducing the irradiated dose to the rectum [80-86]. Methods include 1) injection of synthetic polyethylene glycol hydrogel into the prostate-rectum gap and 2) insertion of a self-expanding and degradable balloon (which is filled with saline) in the body. All of these methods were phase I-II pilot studies in small group populations with four main objectives: to assess the safety of the procedure, to evaluate the stability and degradation time of the filler during radiotherapy, to evaluate the prostate-rectal spacing, and to measure the dose distribution to the target and risk organs after increasing the gap. All studies concluded consistently that the use of tissue fillers to separate the anterior rectal wall from the prostate was effective in reducing the median-maximum dose (50-75 Gy) received by the rectum. However, there is no clear answer as to whether a lower dose will necessarily result in a reduction in clinical toxicities. This is mainly because these studies are small sample data and the follow-up time is still short (median follow-up less than 12 months). In fact, with the current intensity-modulated techniques (high-dose, conventional or moderately separated radiotherapy) and dose-volume limitations, the incidence of advanced rectal injury is already low. Therefore, it is uncertain whether the use of these fillers will further reduce the late rectal response. However, tissue fillers may be well suited for large-segment radiotherapy (more than 5 Gy per session), which is often referred to as stereotactic radiotherapy (SBRT).SBRT is a promising technique because it allows for a high dose rate dose distribution that approximates brachytherapy, while being cost-effective, without the risks associated with puncture, and shortening hospital days. However, SBRT increases the rectal response in the acute phase. This is where the use of tissue fillers becomes very important. 5.4 Drugs and other compounds Prevention and mitigation of acute phase rectal injury is important because early damage to the intestinal mucosa or vascular endothelium is strongly associated with the late occurrence of mucosal injury. Some studies have investigated the possible principles of action of some drugs for the prevention of acute rectal injury. Butyrate and short-chain fatty acids have been shown to prevent and moderate symptoms in the acute phase. These drugs promote mucosal repair, increase arterial vascular resistance, and increase mucosal blood flow and oxygen uptake, thus allowing the mucosa to be repaired as soon as possible [87,88]. In addition, there were applications of soy isoflavones and epinephrine retention enemas to prevent rectal reactions in the acute phase, but all had statistically insignificant results due to the small number of participants. In conclusion, these studies give us a hint that we can manage rectal injury caused by radiotherapy for prostate cancer with certain drugs. These drugs may reduce the clinical risk or the risk of rectal injury in patients. However, these results are only preliminary and need to be validated in a larger population. In conclusion, the reduction of rectal injury during radiotherapy for prostate cancer is a topic of great interest in the field of radiotherapy. Over the past 15 years, we have obtained a large amount of data on dose-volume effects, developed various models for predicting toxicities, applied IGRT-IMRT, and worked to give patients the most optimal dose distribution of treatment. At the same time, there are many physical and pharmacological methods to prevent the toxic side effects caused by radiotherapy, but these methods need to be confirmed in larger clinical trials.