Transitioning from conventional radiotherapy to intensity-modulated radiotherapy for localized prostate cancer: changing focus from rectal bleeding to detailed quality of life analysis

With the advent of modern radiation techniques, we have been able to deliver a higher prescribed radiotherapy dose for localized prostate cancer without severe adverse reactions. We reviewed and analyzed the change of toxicity profiles of external beam radiation therapy (EBRT) from the literature. Late rectal bleeding is the main adverse effect, and an incidence of >20% of Grade ≥2 adverse events was reported for 2D conventional radiotherapy of up to 70 Gy. 3D conformal radiation therapy (3D-CRT) was found to reduce the incidence to ∼10%. Furthermore, intensity-modulated radiation therapy (IMRT) reduced it further to a few percentage points. However, simultaneously, urological toxicities were enhanced by dose escalation using highly precise external radiotherapy. We should pay more attention to detailed quality of life (QOL) analysis, not only with respect to rectal bleeding but also other specific symptoms (such as urinary incontinence and impotence), for two reasons: (i) because of the increasing number of patients aged >80 years, and (ii) because of improved survival with elevated doses of radiotherapy and/or hormonal therapy; age is an important prognostic factor not only for prostate-specific antigen (PSA) control but also for adverse reactions. Those factors shift the main focus of treatment purpose from survival and avoidance of PSA failure to maintaining good QOL, particularly in older patients. In conclusion, the focus of toxicity analysis after radiotherapy for prostate cancer patients is changing from rectal bleeding to total elaborate quality of life assessment.


INTRODUCTION
Prostate cancer is one of the most prevalent solid tumors diagnosed in men in the USA and developed countries. Recent research in numerous randomized controlled trials demonstrated that increasing the prescribed dose in the treatment of localized prostate cancer improves biochemical control in several risk categories: low-, intermediate-and high-risk prostate cancer patients, at least for certain subgroups of patients, as summarized in two recent meta-analyses [1,2] (Table 1). should be considered, if not from their prostate cancer, then from one of the many competing causes of death. Therefore, it is important to determine what could most likely cause their demise. In high-risk patients who are relatively younger (<70 years old at diagnosis), dose escalation leads to a much higher likelihood of dying of a cause other than cancer. Perhaps equally notable, patients who are aged >70 years during treatment never die of prostate cancer when the dose is escalated to 78 Gy or with hormonal treatment [4]. These accomplishments in outcome must be weighed against the complication rate. Fortunately, technology and parameters for dose restriction to normal tissues have provided measures to ensure that the therapeutic index remains high. In this document, we attempted to review the change in toxicity profiles from 2D radiation to the era of image-guided radiotherapy in the face of a dramatic increase in the number of older patients. We analyzed the changing trends in adverse effects of external beam radiotherapy (EBRT). Although there are many good outcomes of brachytherapy (BT) for localized prostate cancer, to keep the analysis simple we did not include BT. The PubMed database was searched for relevant articles published after 1990. We included only studies published in English assessing adverse effects in patients following curative EBRT that had large sample sizes (more than 100 patients) and/or important findings.

LITERATURE REVIEW
From conventional (2D) radiotherapy to 3D conformal radiotherapy Standard 2D planning techniques used until the 1990s with limited total doses of up to 70 Gy were expected to cause toxicity. In the 1990s, 3D planning techniques were developed, and 3D conformal radiation therapy (3D-CRT) was combined with computer software to integrate CT images of the patient's internal anatomy. These approaches allowed physicians to work with a high-dose irradiated volume. The role of dose escalation has been estimated in several randomized controlled trials, and the results indicate that a higher dose improves PSA control with elevated toxicity, mainly in the form of rectal bleeding [1,2,[5][6][7][8][9][10][11][12][13][14][15][16] (Table 1). Most of the evidence of late radiation toxicity comes from those 3D-CRT dose escalation studies. Dearnaley et al. conducted a randomized controlled trial to compare the toxicity of 2D with 3D-CRT with a standard dose of 64 Gy in daily 2-Gy fractions and concluded that conformal techniques significantly lower the risk of late radiation-induced proctitis after radiotherapy for prostate cancer [5]. In the 225 men treated, significantly fewer men developed radiation-induced proctitis and bleeding in the conformal group than in the conventional group (37% vs 56% ≥ Radiation Therapy Oncology Group (RTOG) Grade 1, P = 0.004; 5% vs 15% RTOG ≥ Grade 2, P = 0.01). There were no differences between the groups with respect to Michalsky [15]  bladder function after treatment (53% vs 59% ≥ Grade 1, P = 0.34; 20% vs 23% ≥ Grade 2, P = 0.61). After a median follow-up period of 3.6 years, there was no significant difference between the groups in local tumor control. Koper et al. reported that conformal radiotherapy at a dose level of 66 Gy does not significantly decrease the incidence of gastrointestinal (GI) rectal (10% vs 7%), anal and genitourinary (GU) bladder toxicity compared with conventional radiotherapy in a Phase 3 trial [6]. There is a significant relationship between acute and late toxicity and the anal volume exposed to 90% of the tumor dose. GI and GU symptoms at the start have a major impact on late toxicity.
Yoshioka et al. compared late toxicity for 2D-with 3D-CRT using uniform radiotherapy of 70 Gy in 35 fractions, employing the classical four-field technique with gantry angles of 0°, 90°, 180°and 270°in 362 patients at five institutions with a median follow-up of 4.5 years (range, 1.0-11.6) [7]. The 5-year overall and cause-specific survival rates were 93% and 96%, respectively. The mean ± SD of portal field size in the right-left, superior-inferior and anterior-posterior directions was 10.8 ± 1.1, 10.2 ± 1.0 and 8.8 ± 0.9 cm for a 2D simulation and 8.4 ± 1.2, 8.2 ± 1.0 and 7.7 ± 1.0 cm for a 3D simulation (P < 0.001), respectively. No Grade 4 or 5 late toxicity was observed. The actuarial 5-year Grade 2-3 GU and GI late toxicity rates were 6% and 14% respectively, whereas the corresponding late rectal bleeding rate was 23% for a 2D simulation and 7% for a 3D simulation (P < 0.001). The use of a CT simulation and the resultant reduction in portal field size were significantly associated with reduced late GI toxicity, and particularly with less rectal bleeding.
Consequently, several dose escalation studies have been conducted (Table 1) [8][9][10][11][12][13][14][15][16]. Viani et al. performed a meta-analysis of seven randomized controlled trials with a total patient population of 2812 [1]. Pooled results from these studies showed a significant reduction in the incidence of biochemical failure in patients with prostate cancer treated with high-dose radiotherapy (P < 0.0001). On the other hand, there was no difference in the mortality rate (P = 0.38) or in specific prostate cancer mortality rates (P = 0.45) between the groups receiving high-dose radiotherapy and conventionaldose radiotherapy. Nevertheless, there were more cases of late Grade >2 GI toxicity after high-dose radiotherapy than after conventional dose radiotherapy. In the subgroup analysis, patients classified as being at a low (P = 0.007), intermediate (P < 0.0001), and high risk (P < 0.0001) of biochemical failure all showed a benefit from high-dose radiation therapy.
Zelefsky et al. compared outcomes between 830 3D-CRT and 741 IMRT treatments and concluded that serious late toxicity is unusual, despite the delivery of high radiation doses from 66-81 Gy with a median follow-up of 10 years [17]. Higher doses were associated with increased GI and GU Grade 2 toxicity, but the risk of proctitis was significantly reduced with IMRT. Acute symptoms were a precursor of late toxicity in these patients. After 10 years, the actuarial likelihood of the development of ≥ Grade 2 GI toxicity was 9%. The use of IMRT significantly reduced the risk of GI toxicity compared with patients treated with conventional 3D-CRT (from 13% to 5%; P < 0.001). Among patients who experienced acute GI symptoms, the 10-year incidence of late toxicity was 42%, compared with 9% in those who did not experience acute symptoms (P < 0.0001). The 10-year incidence of late Grade ≥ 2 GU toxicity was 15%. Patients treated with 81 Gy IMRT had a 20% incidence of GU symptoms 10 years later, compared with 12% in patients treated with lower doses (P = 0.01). From the same institute, Spratt et al. reported results from a large cohort of 1002 patients treated with high-dose radiation of 86.4 Gy with a median follow-up period of 5.5 years (range, 1-14 years) [18]. A total of 587 patients (59%) were treated with neoadjuvant and concurrent androgen deprivation therapy (ADT). For low-, intermediate-and high-risk groups, 7-year biochemical relapse-free survival outcomes were 98.8%, 85.6% and 67.9%, respectively (P < 0.001). The incidence of actuarial 7-year Grade ≥2 late GI and GU toxicity was 4.4% and 21.1%, respectively. Late Grade 3 GI and GU toxicity was experienced by seven patients (0.7%) and 22 patients (2.2%), respectively.
Vora et al. reported an improved PSA control rate as a result of high-dose IMRT compared with conventional-dose 3D-CRT without elevated toxicity. A total of 416 patients with a minimum follow-up of 3 years (median 5 years) were included [18]. Of these, 271 patients received 3D-CRT with a median dose of 68.4 Gy (range, 66-71 Gy). Next, 145 patients received IMRT with a median dose of 75.6 Gy (range, 70.2-77.4 Gy). The 5-year biochemical control rate was 74.4% and 84.6% with 3D-RT and IMRT, respectively (P = 0.0326). The high-dose IMRT group experienced greater acute GU toxicity (P = 0.094) than the 3D-CRT group, but the difference was not statistically significant. There were no differences in acute GI (P = 0.83), chronic GU (P = 0.33), and chronic GI (P = 0.24) toxicity between the two groups.
Sheets et al. reported that the use of IMRT vs CRT increased from 0.15% in 2000 to 95.9% in 2008 [21]. In propensity score-adjusted analysis (P = 12 976), men who received IMRT vs CRT were less likely to receive a diagnosis of GI morbidity (absolute risk, 13 [22]. Of 763 patients randomized to the 79.2 Gy arm, 748 were eligible and evaluable: 491 and 257 were treated with 3D-CRT and IMRT, respectively. For both bladder and rectum, the volumes receiving 65, 70 and 75 Gy were significantly lower with IMRT (for all P < 0.0001). For Grade ≥2 acute GI/GU toxicity, both univariate and multivariate analysis showed a statistically significant decrease in Grade ≥2 acute collective GI/GU toxicity for IMRT. There were no significant differences between 3D-CRT and IMRT in acute or late Grade ≥2 or Grade ≥3 GU toxicity. In multivariate analysis, IMRT showed a 26% reduction in Grade ≥2 late GI toxicity (P = 0.099). Acute Grade ≥2 toxicity was associated with late Grade ≥3 toxicity (P = 0.005). RT modality was not significant, whereas white race (P = .001) and rectal V70 ≥15% were associated with G2+ rectal toxicity (P = 0.034). Thus, IMRT is associated with a significant reduction in acute Grade ≥2 GI/GU toxicity. There is a trend for a clinically meaningful reduction in late Grade ≥2 GI toxicity with IMRT. The occurrence of acute GI toxicity and large (>15%) volumes of rectum >70 Gy are associated with late rectal toxicity. Ariskus et al. assessed long-term tumor control and toxicity outcomes after high-dose IMRT in 170 patients who received 81 Gy with a median follow-up period of 99 months [23]. The 10-year PSA control rates were 81% for the low-risk group, 78% for the intermediate-risk group, and 62% for the high-risk group. The 10-year cause-specific mortality rates were 0%, 3% and 14%, respectively. The 10-year likelihood of developing Grade 2 and 3 late GU toxicity was 11% and 5%, respectively; and the 10-year likelihood of developing Grade 2 and 3 late GI toxicity was 2% and 1%, respectively.
To our knowledge, only one manuscript dealt with the constraints of IMRT, but the data were not significant in multivariate analysis. Pederson et al. reported that a 4-year absence of maximal Grade ≥2 late toxicity is observed in 81% and 91% of patients in terms of GU and GI symptoms respectively, with a median follow-up period of 41 months after 76 Gy of IMRT [25]. In multivariate analysis, wholepelvis IMRT was associated with Grade ≥2 GU toxicity, and age was associated with Grade ≥ 2 GI toxicity. The absence of Grade ≥ 2 GI toxicity after 4 years was observed in 100% of men with rectal V70 ≤ 10%, V65 ≤ 20% and V40 ≤ 40%; 92% of men with rectal V70 ≤ 20%, V65 ≤ 40% and V40 ≤ 80%; and 85% of men exceeding these criteria (P = 0.13). These criteria were more strongly associated with GI toxicity in men aged ≥70 years (P = 0.07). At present, no confirmed constraints exist in IMRT, and further studies are required.

From IMRT to image-guided radiation therapy
Image-guided radiation therapy (IGRT) is the process of frequent 2D and 3D imaging, in the course of a radiation treatment, intended to direct radiation therapy using imaging coordinates of the actual radiation treatment plan. This approach allows physicians to deliver accurate radiation therapy with a reduction in the set-up margin (Table 3) [26]. This technique is associated with an improvement in biochemical tumor control among high-risk patients and a lower rate of late urinary toxicity compared with a similar dose of IMRT. This group of patients was retrospectively compared with a similar cohort of 190 patients without fiducial markers (non-IGRT). The 3-year likelihood of Grade ≥2 urinary toxicity for IGRT and non-IGRT cohort was 10.4% and 20.0%, respectively (P = 0.02). Multivariate analysis identifying predictors of Grade ≥2 late urinary toxicity demonstrated that in addition to the baseline International Prostate Symptom Score (IPSS), IGRT was associated with significantly less late urinary toxicity compared with the non-IGRT group. The incidence of Grade ≥2 rectal toxicity was low in both treatment groups (1.0% and 1.6%, respectively; P = 0.81). No differences in PSA relapse-free survival outcomes were observed in lowand intermediate-risk patients when either treated with IGRT or not treated with IGRT. Nonetheless, in high-risk patients, a significant improvement (97% vs 77.5%, P = 0.05) was observed 3 years after treatment with IGRT compared with non-IGRT.
Vora et al. reported [28] long-term disease control and chronic toxicity in 302 patients. Chronic toxicity was measured at the peak in symptoms and at the last visit. The median radiation dose delivered was 75.6 Gy (range, 70.2-77.4), and 35.4% of the patients received ADT. The patients were followed up until death or for 6-138 months (median, 91) for those alive at last evaluation. At last follow-up, only 0% and 0.7% of patients had persistent Grade ≥ 3 GI and GU toxicity, respectively.
Tomita et al. reported helical tomotherapy (HT) results for 241 patients with a median follow-up time of 35 months [29]. Late Grade 2-3 rectal toxicity was observed in 18 patients (7.4%). Age, the maximum dose for the rectum, V70 and V60 of the ≥ Grade 2 toxicity group were significantly higher than in the ≤ Grade 1 toxicity group (P = 0.000 93, 0.048, 0.0030 and 0.0021, respectively). None of the factors was significant
Potential mechanisms involved in the development of incontinence could be the reduced absorption capacity of the rectal mucosa, which may be expected to have a large volume effect as well as neurovascular damage impairing the musculature surrounding the rectum. Several recent studies produced evidence of dose-volume relations for late rectal incontinence [36][37][38]. It was demonstrated recently that a DVH constraint of rectum V40 < 65% or V40 < 80% (or a mean rectal dose of < 45-50 Gy) reduces the risk of late incontinence [6, 18, 20, 36-38, 58, 59, 61-63, 66-71]. Although late incontinence is quite a rare side-effect in modern radiotherapy, the application of this constraint has the potential to reduce the risk to <2%. In addition, several authors found a link to acute adverse reactions of Grade 2 and 3, which correlates strongly with the mean dose; these data suggest that the reduction of the dose bath delivered to the whole rectum may have an impact on the risk of acute toxicity [37,38,74]. Detailed analysis of the subarea DVH could provide further insights into the incontinence risks [33,38,63,73]. Heemsbergen et al. reported a subarea difference: for bleeding and a mucus loss, the strongest correlation was found for the dose delivered to the upper 70-80% of the anorectal region (P < 0.01) [73]. For soiling and fecal incontinence, they found the strongest association with the dose delivered to the lower 40-50% of the anorectal region. For example, the anal canal was contoured by taking the caudal 3 cm of the anorectal portion [38]; 53 Gy delivered to the anal surface was found to be an important constraint [75]. Al-Abany et al. also reported dose constraints: a dose V35 < 60% or V40 < 40% of the anal sphincter region volume for fecal leakage [76]. A recent study proposed more detailed dose constraints: 30 Gy delivered to the internal anal surface, 10 Gy to the external anal surface, 50 Gy to the puborectalis muscle, and 40 Gy to the levator ani muscles [68].

Genitourinary adverse reactions
Mild acute irritative urinary symptoms have been reported in several studies, whereas total urinary incontinence and other severe late urinary symptoms (i.e. urethral stricture) are rare.
In the case of the bladder, there is a clear dose effect when the whole organ is irradiated (i.e. for cystitis) [78]. On the other hand, in the case of prostate irradiation, the cranial portion of the bladder is generally spared, whereas the bladder neck and urethra are irradiated near the prescribed dose [80]. The lack of knowledge about the dose-volume modeling of bladder toxicity probably reflects the difficulties with accurate assessment of the amount of bladder wall that receives a certain dose. This is because large variations are observed in the bladder shape during treatment because of variable filling. Serial behavior was reported recently for late mild to severe toxicity [54], whereas serial-parallel behavior was reported for chronic moderate or severe urinary toxicity [80]. Both studies indicated that the fraction of bladder receiving >78-81 Gy is most predictive of late GU toxicity [17,54,80].

DISCUSSION
There are many modalities in radiation therapy, which cause a range of incidences of late GI toxicity. Kim et al. analyzed 28 088 patients using the SEER data. The most common GI toxicity is GI bleeding or ulceration. GI toxicity rates are 9.3 per 1000 person-years after 3D-CRT, 8.9 per 1000 personyears after IMRT, 20.1 per 1000 person-years after proton therapy, and 2.1 per 1000 person-years for patients receiving conservative management. Radiation therapy is the most significant factor associated with an increased risk of GI toxicity (HR, 4.74; 95% CI, 3.97-5.66). Even after 5 years, the radiation group continues to experience significantly higher rates of new GI toxicity than the conservative management group (HR, 3.01; 95% CI, 2.06-4.39) [91].
The RTOG or CTCAE scoring system has been widely used for assessment of toxicity but not enough to meet the requirements, according to a recent radiotherapy outcome survey. This is because in these scoring systems, compliance-related symptoms (such as stool frequency) and proctitis-related symptoms (such as rectal bleeding) are combined into one overall score. This feature may result in a loss of information and may obscure the relation between dose-volume parameters and complications [43]. Accordingly, several trials added a patient self-assessment questionnaire to obtain detailed information on morbidity. In addition, longitudinal assessment may add more useful information than peak score analysis can [43,63,68]. Gulliforde et al. found that endpoint-stool frequencystatistically significant dose-volume constraints are only derived by a longitudinal definition of toxicity in the outcome analysis of the MRC RT01 trial [63]. By the same token, an apparent association exists between acute side-effects experienced during the course of radiotherapy and the development of late toxicity. Heemsbergen et al. noted such an association between acute and late GI toxicity and postulated that late effects are a direct consequence of the initial tissue injury, which is reflected in acute symptoms resulting from inflammation of normal tissue [77]. According to their report, the presence of diarrhea during treatment is associated with a higher risk of late Grade ≥2 toxicity in late proctitis. They found that acute toxicity during treatment often manifests as tenesmus and internal hemorrhoid inflammation, which are associated with a higher likelihood of late proctitis. In addition, acute urinary symptoms that manifest during radiotherapy are linked to an increased risk of late Grade 2 urinary adverse events. Kim et al. [92] reported the long-lasting nature of GU toxicity: Grade 2-4 GU toxicity attributable to radiation therapy persists 10 years after treatment and thereafter based on comparison of 60 134 patients who received radiation therapy with 25 904 who underwent observation.
High-dose irradiation and/or hormonal therapy result in excellent outcomes, not only in PSA control, but also in overall survival. Nguyen et al. reported good 5-and 10-year actuarial overall survival rates (no ADT plus 75.6 Gy, 87.3% and 72.0% respectively; and ADT plus 75.6 Gy, 92.3% and 72% respectively; P = 0.0035) [4]. We also obtained similar results: 70 Gy plus ADT achieve 91-93% of overall survival after 5 years [7,93]. Therefore, we should pay attention to adverse effects and quality of life (QOL) rather than disease control because almost 90% of the patients after EBRT live longer than 5 (or 10) years.
Multiple health-related QOL studies have been conducted using the IPSS, IIEF, and the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire for Prostate Cancer 25 items (QLQ-PR25) etc. Such comparison between radical prostatectomy, EBRT, BT, and combined approaches uncovers a link between observed toxicity and QOL. For example, Sanda et al. prospectively measured outcomes reported by 1201 patients and 625 spouses or partners at multiple centers before and after radical prostatectomy, BT or EBRT [94]. Adjuvant ADT is associated with worse outcomes across multiple QOL domains among patients receiving BT or radiotherapy. Patients in the BT group report long-lasting urinary irritation, bowel and sexual symptoms, and transient problems with vitality or hormonal function. Adverse effects of prostatectomy on sexual function are mitigated by nerve-sparing procedures. After prostatectomy, urinary incontinence is frequent, but urinary irritation and obstruction are improved, particularly in patients with a large prostate. No treatment-related deaths occurred in that study; serious adverse events were rare. Their results suggest that treatment-related symptoms are exacerbated by obesity, large prostate size, high PSA score and older age. Black patients report a lower degree of satisfaction with the overall treatment outcomes. Changes in QOL are significantly associated with the degree of outcome satisfaction among patients and their spouses or partners. However, there are several problems with the use of QOL questionnaires. For example, the IPSS is considered a major QOL questionnaire in the treatment of prostate cancer, but IPSS was constructed mainly for prostate hypertrophy symptoms. Thus, this questionnaire cannot evaluate adverse effects after prostatectomy (the IPSS of most patients improves after prostatectomy). Therefore, when it comes to comparison of different treatment methods, accurate QOL evaluation is a challenge.
The impact of age on prostate cancer outcomes was found not only in PSA control and survival but also in QOL in less aggressive prostate cancers in older men [95], independent of other clinical features. When adjusted for other covariates, age >70 years still correlates with decreased OS (HR, 1.56 [95% CI] 1.43-1.70 P < 0.0001) and with a decreased incidence of metastasis (HR, 0.72 [95% CI, 0.63-0.83], P < 0.0001) and prostate cancer-specific death (HR, 0.78 [95% CI, 0.66-0.92], P < 0.0001). Although the biological underpinnings of this finding remain unknown, stratification by age in future trials is warranted. Several reports show that adverse reactions occur more frequently in older patients [32,33,77]. In this context, major data provided by a clinical trial (i.e. a large randomized controlled trial) were based on the data from patients younger than 80 years of age.
There are several limitations to our study. First, we did not analyze BT (although there are plenty of data in the literature) because we focused on the changes in adverse effects as a result of the advancement of EBRT from 2D to IMRT and IGRT. Second, as a result of this we did not analyze particle therapy because of the limited use of this therapy (both proton and carbon ion) in patients with prostate cancer except for clinical studies. Finally, hypofractionated radiotherapy was also excluded from this analysis, even though there is a hypothesis that hypofractionation has a radiobiological advantage in prostate carcinoma because of the low α/β ratio. This topic-the influence of fractionation-is beyond the scope of this study and will be explored in future studies.
In conclusion, the focus of toxicity analysis following radiotherapy for prostate cancer patients is changing from rectal bleeding to total elaborate QOL assessment.