2 Clarke Drive
Cranbury, NJ 08512
© 2022 MJH Life Sciences™ and OncLive - Clinical Oncology News, Cancer Expert Insights. All rights reserved.
Developing and validating a predictive model for patient-reported health-related quality of life remains an elusive target, and one that has great potential to further improve satisfaction with outcome following prostate cancer treatment in a great number of prostate cancer survivors.
Daniel A. Hamstra, MD
Dean A. Shumway MD
Steven Eric Finkelstein, MD
Felix Feng, MD
Why is this article contemporary?
The options and alternatives for managing prostate cancer with radiation continue to expand. Herein, Heath et al provide a contemporary discussion of factors influencing patient reported rectal toxicity following external beam radiation therapy for prostate cancer.
In our current health climate, posttreatment health-related quality of life (HRQOL) is emerging as a main concern of patients. Measuring patient-reported HRQOL using validated survey instruments suggests these measures are more sensitive compared to physician-reported toxicity guidelines as counterparts.
Patient-reported HRQOL following radiation therapy remains strongly dependent on the technique, the total radiation dose delivered, and dosevolume relationships with different portions of the rectum. Imageguided radiation therapy and pelvic tissue spacers are 2 recent technical advancements that have facilitated limitation of untoward rectal toxicity.
Counseling is key with patient education focused that rectal HRQOL is dependent on individualized baseline function and treatment option selected. To minimize toxicity with respect to radiation induced rectal side effects, consideration should be given to dose received not only by the whole rectum but particularly by the inferior rectum.
Heath et al suggests that rectal doses both greater than and less than 70 Gy may influence bowel QOL with stronger associations with higher doses. Indeed, a number of DVH constraints have been suggested by single institutions; with the exception of the rectal V70<25%, these largely await validation in multi-institutional studies.
Thus, to increase contemporary patient satisfaction following prostate cancer treatment future multiinstitutionally validated predictive guidelines should be advanced that permit for better communication of expectation for rectal HRQOL changes following treatment.
The high incidence and relatively low mortality rate associated with prostate cancer, in combination with a lack of clearly demonstrated survival benefit among numerous treatment options, has resulted in post-treatment health-related quality of life (HRQOL) emerging as a primary concern of patients. Measuring patient-reported HRQOL using validated survey instruments has shown that these measures are more sensitive and reliable compared to physician-reported toxicity guidelines, which have been traditionally utilized. Dose-volume histogram (DVH) guidelines for limiting rectal toxicity and improving patient-reported HRQOL have been established in numerous single-institution studies yet await multi-institutional validation. Limiting dose to the inferior rectum in particular has been shown to limit rectal toxicity and improve patient-reported rectal HRQOL in a subset of patients undergoing intensity-modulated radiation therapy (IMRT). Image-guided radiation therapy (IGRT) and pelvic spacers are 2 recent advancements that have also helped to limit rectal toxicity. Developing and validating a predictive model for patient-reported HRQOL remains an elusive target, and one that has great potential to further improve satisfaction with outcome following prostate cancer treatment in a great number of prostate cancer survivors.
Radiation-Related Rectal Toxicity Development of Radiation Therapy Techniques That Limit Rectal Toxicity
Approximately one in 7 men living in the United States will be diagnosed with prostate cancer in their lifetime. However, it is estimated that as few as one in 38 men will die from this disease due in part to the natural history of the disease, as well as improved screening measures and treatment options.1 The 3 most common treatments for prostate cancer—radical prostatectomy (RP), brachytherapy (BT), and external beam radiation therapy (EBRT)—have all been shown to offer similar survival benefit and each has its own uniquely associated HRQOL implications following treatment.2 Due to the fact that most prostate cancer survivors do not die as a direct result of their disease, and over 94% of patients will live more than 15 years following diagnosis, HRQOL is often a primary concern for patients and physicians when determining a treatment plan.1,3 Offering patients concise predictive information on post-treatment HRQOL can be difficult due to the rapidly evolving technology used to treat prostate cancer in combination with patients and physicians often emphasizing different HRQOL outcomes.4 Due to the sensitivity of the rectum to radiation, efforts to reduce the dose to this area in the pelvis are often associated with improved patient-reported HRQOL as well as physician-scored toxicity.As the rectum has been shown to be the most radiation sensitive tissue in the pelvic cavity, and doses applied to this area is associated with an array of disease progression, survival, and HRQOL measures, careful planning of treatment dose to the rectum is essential.5-9 Significant morbidity was previously associated with prostate radiation delivered using two-dimensional radiation, in which the treatment field was defined based on skeletal anatomy from radiographs, and the rectum was not defined as an avoidance structure. In patients treated to a total dose of 64 to 72 Gy to the prostate (also including pelvic lymph nodes to 45 Gy with a randomization to inclusion of paraaortic lymph nodes) on RTOG 7506 and 7706 and using physician-reported toxicity measures, diarrhea was routinely observed during treatment, persistence of grade 1-2 diarrhea was noted in approximately 45% in the first month following completion of radiation. Approximately one-third of patients developed grade 1-2 proctitis in the same time period.10
With long-term follow-up of a minimum of 7 years, late grade 3 or greater gastrointestinal toxicity occurred in 3.3% of patients, including 0.4% with diarrhea, 1.6% with proctitis, and 1.7% with rectal bleeding or ulcer.11 This rate of toxicity is considered unacceptable by modern standards.
With development of 3-dimensional conformal radiation (3D-CRT), the dose delivered to the prostate was escalated to 78 Gy on RTOG 9406.12 Compared with the historical experience on RTOG 7506 and 7706, there was a significantly lower incidence of Grade 3 or worse late effects with 3D-CRT, despite the administration of a higher dose of radiation. These results were replicated in several institutional series.13-15 It is notable that these trials did not specifically use IMRT or IGRT, which are expected to result in an even lower rate of grade 2 gastrointestinal toxicity.5 On the most recent randomized controlled trial of high dose vs standard dose radiation, RTOG 0126, the protocol suggested the importance of daily online imaging or real-time ultrasound localization, but only required weekly orthogonal verification films. In patients randomized to the high-dose arm who received 79.2 Gy, there were no significant differences between 3D-CRT and IMRT for any patient-reported toxicity at any time point, nor when compared to baseline function, including the bowel subscore. However, the questionnaire that was used, the Functional Alterations due to Changes in Elimination (FACE) instrument, did not include questions about rectal bleeding, which had previously been described as the most common physician-reported gastrointestinal toxicity on the same trial.16 Analysis of physician- reported toxicity showed a significant reduction in grade 2+ acute collective gastrointestinal/genitourinary toxicity with IMRT and a trend toward reduced grade 2+ gastrointestinal toxicity with IMRT. Taken together, the analyses of patient-reported and physician-reported toxicity suggest a trend for a clinically meaningful reduction in late grade 2 or greater gastrointestinal toxicity with IMRT, although it is noteworthy that very few patients experienced significant toxicity in either arm.16
Patient-reported Health-Related Quality of Life Outcomes Substantial differences have been shown between patient- reported HRQOL and physician-scored toxicity, with physicians often underestimating the impact of toxicity on HRQOL.4 Due to the emphasis on rectal bleeding in physician-scored toxicity, efforts have been made to reduce dose to the mid-rectum. However, patients and their partners often prioritize fecal incontinence and urgency over rectal bleeding due to the great impact this can have on their daily life following treatment.17,18
Physician- scored toxicity also does not adequately capture patient bowel irritation and bother, measures that have been identified as important to patients and their partners.19 In order to more accurately monitor rectal HRQOL following treatment, many studies have started to focus more on patient-reported measures to assess HRQOL. One of the most popular and validated methods for collecting patient-reported HRQOL data after treatment for prostate cancer is the Expanded Prostate Cancer Index (EPIC) survey, which assesses HRQOL across five domains (urinary-irritative, urinary obstructive, rectal, sexual, and hormonal) and uses a 0 to 100 scale to assess outcomes. Measuring long-term HRQOL using patient- reported surveys, such as EPIC, can be instructive in that it captures the patient’s perspective on how symptoms change over time and how much these symptoms impact the patient's life (often expressed in terms of function and bother scales). In one group of patients treated at the University of Michigan for prostate cancer, HRQOL was measured using the EPIC instrument after EBRT, BT, and radical prostatectomy. EBRT was delivered using 3D-CRT, with 1.8 to 2.0 Gy daily fractions to a total of 55 to 80 Gy, and BT patients received a 160 Gy dose of implanted low-dose rate iodine-125. Of note the EBRT treated cohort used neither IMRT nor IGRT and as such utilized PTV margins likely now considered unusually large by contemporary standards. Similarly, BT treated patients utilized neither peripheral loading nor MRI-based planning which many now consider part of the standard of care. Baseline HRQOL was not measured for this cohort, so comparisons are difficult. Nevertheless, trends in HRQOL over time can be evaluated. In follow-up, BT patients had the lowest rectal HRQOL score 2.6 years after treatment (73.8), which improved at 6.2 years. EBRT patients had higher rectal HRQOL at 2.6 years (87.7), with little change at 6.2 years (87.9). Patients in the BT group reported increased bother with overall bowel problems at the first survey interval (17% at 2.6 years, 10% at 6.2 years), but at 6.2 years they reported slightly better outcomes in this category compared to EBRT patients (8% at 2.6 years, 13% at 6.2 years). Unpublished data following up on this cohort at 15.7 years post treatment indicates that rectal HRQOL was unchanged in the EBRT group from 6.2 to 15.7 years post treatment with a modest decline in rectal HRQOL for the BT group over this time frame although the reported QOL at 15.7 years after BT was on average similar to that observed after 2.6 years.20
Treatment Factors Associated with Rectal ToxicityDevelopment of Dose-Volume Histogram Guidelines
Although baseline quality of life was not measured in this cohort and radiation treatment therapies have been since improved upon, this cohort offers an opportunity to counsel patients on the changing nature of rectal quality of life years following different treatment options.19-21One issue with the previous study was that pretreatment HRQOL was not collected making it difficult to attribute how much of the ultimate HRQOL was related to treatment or baseline function. In contrast, the PROSTQA study followed 1201 patients and 625 spouses from multiple academic centers before and after treatment at 2, 6, 12, and 24 months. EBRT was utilized in 292 patients with 83% of patients completing the 2-year assessment. Median age was 69 (range 45 to 84) with 82% NCCN low-intermediate risk. Androgen deprivation therapy (ADT) was used in 31%, pelvic RT in 13%, and IMRT in 85%, with RT dose 75.6-79.2Gy.
BT patients were treated with transperineal low-dose rate isotopes using predominantly 125I and with a lower portion 103Pd. The PROSTQA study also utilized the EPIC-26 short form instrument to measure patientreported HRQOL changes.22
BT and EBRT patients both experienced a drop in rectal HRQOL 2 months following treatment, with many of these changes lasting for more than one year (Figure 1). Bother associated with overall bowel problem was reported by EBRT patients (16% at 2 months, 9% at 6 months, 9% at 12 months, 11% at 24 months.) at rates comparable with BT patients (15% at 2 months, 12% at 6 months, 9% at 12 months, 8% at 24 vmonths), as well as moderate-to-big problems with overall bowel function at 24-months (10% BT, 11% EBRT).
Analysis of this study has shown that in the EBRT cohort a rectal V70≥25% is correlated with an average 9.3 point decline in overall bowel score, meeting the criteria for a minimally important clinical difference set at 4.0 to 6.0 points.23 Additionally, those with V70≥25% reported a 3.9- fold increase in incidence of moderate-to-severe incontinence, a 3.6-fold increase in incidence of moderate-to-severe rectal bleeding, and 2.9-fold increase in incidence of moderate-to-severe bowel urgency compared with treated with V70<25%. Another factor associated with reduced rectal HRQOL following treatment was aspirin use, showing a 4.7-point decline in overall bowel HRQOL and a 2.8-fold increase in moderate to severe bloody stools.
For example, in those with V70<25% there was a 16% increase in the prevalence of small, moderate, or large problems with bowel urgency at 2-years (with the majority of this being a small problem) while in those with V70>25% there was a 38% increase in small, moderate, or large problems with bowel urgency. Similarly, small, moderate, or large problems with rectal bleeding was increased by 5% in those not taking aspirin during therapy which was 16% in those on aspirin. Interestingly, IMRT use did not independently correlate with rectal HRQOL as compared to 3D-conformal therapy in this study, albeit there was a small and potentially selected group treated with 3D-conformal therapy, and IMRT has shown to improve these measures in other reports.18
Dose to the Inferior Rectum
The PROSTQA study suggested that dose-volume parameters of the rectum might be associated with the probability of declines in rectal HRQOL. This result was confirmed in a single institutional study of 372 patients treated with dose-escalated IMRT between 2001 and 2010. This study builds on the previously established V70≤ 25% guideline and recommends dosing to the rectum at V70<10%, V65<20%, V40<40% when targeting the prostate and seminal vesicles, and V70<20%, V65<40%, V40<80% when including pelvic lymph nodes. Very few patients reported problems with overall bowel problem after treatment with these constraints (3% at 2 months, 3% at 6 months, 0% at 12 months, 3% at 18 months, 2% at 24 months), and zero patients reported problems with fecal incontinence or urgency after 24 months. Freedom from grade 2 GI toxicity was reported in 94% of patients after 2 years and in 92% of patients after 4 years24. Other groups have reported similar correlations between rectal dose volume constraints and late bowel HRQOL.14,25,26Most studies of rectal HRQOL have used a standard definition of the rectum between iliac crests (inferiorly) and sigmoid flexure (superiorly). However, analysis of the Dutch randomized dose escalation study has suggested that in regard to rectal toxicity not all portions of the rectum (superior to inferior) carry the same risk of affecting rectal toxicity.27 In one single institutional study 90 patients were treated with dose-escalated 3D-CRT, with common use of IMRT and image guidance, to a median dose of 77.7 Gy in 1.8 to 2.0 Gy fractions and patientreported rectal HRQOL was measured using the EPIC instrument. DVH’s were analyzed for whole, superior, middle, and lower rectum for 2 years following treatment.
Modern Technical Advances That Limit Rectal Toxicity Daily Image Guidance
In this analysis, the dose to the inferior rectum was significantly associated with incontinence, rectal bleeding, urgency, and overall bowel problems (Figure 2). Increasing mean dose to the whole rectum was also associated with a decreased bowel summary score. In contrast, there was little association between bowel HRQOL and dose received by the middle and superior rectum. Only rectal bleeding had a strong association with the middle rectum (the portion most closely adjacent to the prostate). As incontinence and urgency have been noted as the most important problems associated with reduced patient-reported HRQOL, dose to the inferior rectum potentially has a large impact on patient satisfaction with outcome.17Prostate motion between fractions is strongly related to changes in rectal volume28 and rectal distension is an important predictor of biochemical failure.29 There can also be movement of the prostate during treatment delivery due to rectal peristalsis.30 This inter- and intra-fraction variability in the position of the prostate moves independently from skeletal landmarks, such that pelvic bony anatomy is an inadequate surrogate for prostate alignment unless large margins are used.31,32 However, any volume expansion to account for this setup uncertainty increases the volume of rectal wall exposed to high doses. Treatment planning margins can be cut in half with use of intraprostatic fiducial markers and an on-line localization scheme, from 5 to 7 mm with localization based on bony anatomy to 3mm.32 Because any reduction in the radius of a sphere results in a dramatic decrease in its volume (since volume is calculated by the radius raised to the third power: 4/3 pr3), this reduction in treatment margins results in significantly less dose to the adjacent bladder and rectum. There is evidence to suggest that smaller margins used with fiducial markers result in clinically significant reduction in toxicity, with reduced physician-reported grade 2 rectal toxicity.33
Other methods of image-guidance for daily localization of the prostate include Calypso markers or cone-beam CT (CBCT). Calypso markers provide real-time localization and monitoring of the target position, with similar accuracy compared to x-ray localization of gold fiducial markers.30 In a study of 35 patients who underwent continuous, real-time tracking for a full course of radiotherapy, displacements 3mm for cumulative duration of 30 seconds were observed in 41% of sessions.30 The ability to monitor the position of the prostate throughout treatment may be particularly beneficial for hypofractionated treatments with longer treatment times and higher dose per fraction28, but is perhaps less meaningful for patients with a shortened treatment time, such as those treated with volumetric modulated arc delivery.
Pelvic Cavity Spacers
Daily kilovoltage cone-beam computed tomography provides an equivalent means of setup accuracy compared to fiducial markers34,35, although there is evidence supporting greater reproducibility and accuracy with alignment to fiducial markers.36 In summary, daily image guidance with on-line correction is an imperative component for margin reduction, dose-escalation, and reduced treatment toxicity.As physicians work to reduce dose to the rectum in order to improve patient-reported rectal HRQOL, a number of techniques have been utilized to increase the distance between the posterior prostate capsule and the anterior rectal wall. Endorectal balloons and hyaluronic acid gel injections have been shown to decrease dose to the rectum by up to 50% while achieving an average separation of between 10 and 15mm, which has the potential to impact rectal HRQOL when used to meet rectal strict DVH constraints. A recent randomized phase III trial utilized fiducial marker based IGRT with IMRT delivering 79.2 Gy in daily fractions of 1.8 Gy. Strict rectal dose-volume histogram (DVH) criteria were utilized with centralized plan review while patients were randomized to placement of a bio-absorbable spacer between the rectum and prostate or to control.37-41 Although placement of the SpaceOAR provided a small reduction in the likelihood of rectal toxicity, the difference was not statistically significant. The lack of a significant difference was attributed to uniform use of IGRT, smaller planning target volume margins (5 to 10 mm), IMRT with pre-determined dosimetric constraints, and careful attention to pre-treatment radiation treatment plan quality assurance with careful attention to meeting treatment goals. Another multi-institutional study examining perirectal hydrogel spacer application reported that with an average treatment group perirectal space of 12.6 +/- 3.9mm, compared to 1.6 +/- 2.0mm in the control group, a reduction in V70 from 12.4% to 3.3% was achieved. There were small differences in the median decline in bowel score. However, the proportion of patients with a clinically relevant decline in bowel QOL (as defined as 10+ points on the EPIC scale) was substantially lower in those treated with the rectal spacer than in the control group (11.6% vs 21.4%). This is suggestive that although all patients may not need a rectal spacer it is possible that there is a subgroup in whom it may improve rectal QOL. At present, it is unclear if this subgroup is best selected based upon baseline QOL, the ability to achieve rectal planning goals, or other health related parameters.Patient-reported HRQOL following radiation therapy is dependent on the radiotherapy technique, the total radiation dose delivered, and dose-volume relationships with different portions of the rectum. Patients should be counseled that their rectal HRQOL is dependent on their baseline function and their treatment option selected. In order to minimize rectal side effects related to radiation, particular attention should be given to dose received not only by the whole rectum but in particular by the inferior rectum. In addition, it does appear that rectal doses both greater than and less than 70 Gy may influence bowel QOL with stronger associations with higher doses. While a number of DVH constraints have been put forth by individual institutions, at this point, with the exception of the more commonly accepted rectal V70<25% level, these standards largely await validation in multi-institutional studies. In order to increase patient satisfaction with outcome following prostate cancer treatment, multi-institutionally validated predictive guidelines should be developed that allow for better communication of expectation for rectal HRQOL changes following treatment.
Gerard E. Heath, BS, and Dean A. Shumway, MD, are with the Department of Radiation Oncology, University of Michigan. Daniel A. Hamstra, MD, is with Texas Center for Proton Therapy, Irving TX. Corresponding author: Gerard Heath, BS, 36340 Barkley St., Livonia MI, 48154, Cell: (734) 664-7134, E-mail: firstname.lastname@example.org
This review article highlights technological advancements that seek to improve rectal quality of life in patients treated with radiation therapy for prostate cancer.
Daniel A. Hamstra, MD is employed by Texas Oncology. He has received grants from Novartis, is a paid consultant to Augmenix, and is on the advisory board of Medivation, Varian, GenomeDX, Myriad, and Janssen.