Neha Vapiwala, MD
- Severe late rectal and/or bladder toxicity following prostate radiation therapy (RT) is relatively rare nowadays, but lower-grade, long-term toxicities can negatively impact patient quality of life.
- Until a genetically based tool becomes available for prostate cancer treatment, educated predictions of a patient’s suitability for RT are typically made through thorough review of the history, physical exam, and imaging studies as well as dosimetric evaluations.
- Short-term toxicities during RT are generally predictable and highly manageable in today’s high-tech environment, with weekly status checks designed to catch and address nuisance symptoms like urinary frequency and urgency.
One of the most challenging aspects of prostate cancer management in the nonmetastatic setting is also one of its greatest strengths: the myriad choices patients and providers face. In order to help guide patients with nonmetastatic prostate cancer along this arduous journey, one must first establish the patients’ treatment-related goals, personal biases and preferences, and baseline performance/physical status, with respect to genitourinary (GU), gastrointestinal (GI), and sexual function. Although a growing number of focal therapy approaches are emerging, the most commonly accepted and well-established options for clinically localized prostate cancer remains the trifecta of radical prostatectomy, radiation therapy (RT), and active surveillance (AS). Within both the surgical and radiation realms, however, additional layers of options emerge related to technique, approach (i.e., robotic vs. open surgery and brachytherapy vs. external beam RT [EBRT]), and adjunctive therapies, to name a few. Even AS requires an ongoing patient-provider conversation and evolving assessment of both the appropriate frequency of data collection and trigger for active intervention. Regardless of the intricacies and of the ultimate choice, the vast majority of men with nonmetastatic prostate cancer have favorable cancer-specific survival rates. A recent high-profile publication of a landmark randomized clinical trial highlighted the comparable and overall excellent oncologic outcomes seen with radical prostatectomy, EBRT, and AS for PSA-detected, clinically localized disease.1 As such, the implications of the patient’s chosen disease management approach on acute and chronic toxicities, and ultimately on quality of life, can dominate the decision-making process. Beyond an individual patient level, the societal impact of cancer survivors with chronic, treatment-related morbidity can be staggering. Therefore, a multidisciplinary approach to counseling patients, with input from urologic, radiation, and medical oncology colleagues, is the gold standard for balanced, comprehensive care.
Understanding the Toxicities
As part of any nuanced risk-benefit discussion, clear comprehension of treatment-related toxicities is critical. EBRT is a well-established definitive treatment for prostate cancer in a variety of settings beyond organ-confined disease, including locally advanced and postoperative cases.2,3 Marked advances over the past 15 to 20 years in imaging capabilities, as well as software and hardware updates for treatment planning and delivery, have enabled increasing adoption of dose escalation to the target (i.e., prostate gland and seminal vesicles).4-6 Multiple randomized studies in patients with clinically localized prostate cancer have demonstrated improved disease control with higher RT doses, albeit no overall survival benefit to date.7-11 In some instances, these gains are achieved at the expense of greater rectal and/or bladder toxicity rates. Fortunately, severe late rectal and/or bladder toxicity following prostate RT is relatively rare nowadays, but lower-grade, long-term toxicities can negatively impact patient quality of life (i.e., symptom-related “bother”) and thus tend to be more frequently observed and reported.
The etiology of and predictive factors for RT toxicity are topics of great interest and even greater complexity. Increasingly sophisticated radiation techniques such as intensity-modulated RT (IMRT) with photons or proton beam RT (PBRT) can offer improved target conformality, precision, and normal-tissue sparing, but there is a dearth of level 1 data establishing toxicity benefits.12-15 Furthermore, interpretation of the existing published literature is complicated by notable variation in the RT planning and delivery protocols, total doses and dose per fraction, and late-toxicity scoring criteria that were used.16,17
Relatively straightforward are the characteristics of a patient’s personal history, which are generally believed to significantly compound existing toxicity risks. Prior pelvic RT, bladder or rectal surgery, urethral stricture, significant transurethral resection of the prostate defect, bleeding diatheses, bowel disorders (in particular ulcerative colitis, but including diverticular disease), and certain autoimmune conditions are among these relative or absolute contraindications to prostate RT. Although these factors have not consistently been reported to increase toxicity, anticoagulant use is among the better-established risk factors for post-RT bleeding.18
Less straightforward are anatomical considerations that may predispose to toxicity through direct or indirect effects on the radiation dosimetry (defined as the calculated absorbed dose in tissue for a given exposure to ionizing radiation). For example, large-volume prostate glands (> 80-90 cc), “droopy” seminal vesicles that wrap around portions of the rectum, significant spinal scoliosis, and metal penile or pelvic prostheses can all pose challenges to reproducible and reliable daily patient positioning and/or accurate radiation dose calculations. Endorectal balloons and hydrogel rectal spacers are potential tools to improve internal organ immobilization/reproducibility and optimize spatial geometry, respectively, for protection of normal tissues during high-dose EBRT for prostate cancer.
Individual genetic variations are also complicated to predict, mainly in the form of single nucleotide polymorphisms within genes, which confer differing degrees of normal tissue “vulnerability” to radiation. Although there is a clear link between extreme radiation sensitivity and homozygous mutations in the ATM gene,19 this elegant paradigm does not apply to the vast majority of patients for whom response to radiation is likely a complex polygenic trait. The field of radiogenomics specifically emerged in part to elucidate genomic markers that are predictive for development of radiation toxicity. The Radiogenomics Consortium is a large-scale international collaborative effort to compile resources for identification of relevant single nucleotide polymorphisms and ultimately design a predictive tool for identifying those at greatest risk for adverse effects from radiation.
At present, until such a genetically based tool becomes available for prostate cancer treatment, educated predictions of a patient’s “suitability” for RT are typically made through thorough review of the history, a physical exam, and imaging studies. These considerations should factor into the risk-benefit assessment given the many options patients with localized prostate cancer have to weigh during what can be an overwhelming process. Once a patient with prostate cancer is deemed a reasonable candidate for definitive RT, a treatment-planning simulation is performed to develop a customized radiation plan, with detailed modeling of radiation dose distribution to the target and normal nearby structures. It is these dosimetric parameters that perhaps receive the greatest emphasis in that they are modifiable factors that could minimize potential toxicity from prostate radiation. Short-term toxicities during RT are generally predictable and highly manageable in today’s high-tech environment, with weekly status checks designed to catch and address nuisance symptoms such as urinary frequency and urgency. The vast majority of these acute complaints resolve several weeks to months following RT completion. But, as already noted, the greatest concern is ultimately late radiation-related toxicity given the quality-of-life detriment with chronic treatment-related complications, and that is our focus here.
Examining the Data
A meta-analysis by Ohri et al included 20 studies reporting late toxicity after definitive EBRT for localized prostate cancer to investigate the impact of radiation doses, techniques, and fields on moderate (grade ≥ 2) and severe (grade ≥ 3) GI and GU toxicity.20 The authors restricted the analysis to reports incorporating the RTOG/EORTC Late Radiation Morbidity Scoring Schema, but included a combination of retrospective and prospective phase I-III studies for nearly 12,000 patients. The studies comprised a wide range of follow-up periods, doses (64-80 Gy), and techniques, from the antiquated (two-dimensional [2D] RT in nine series) and less sophisticated (three-dimensional [3D] conformal RT in the majority) to more advanced (IMRT and PBRT) approaches. The median rates of late grade 2 or greater GI and GU toxicity were 15% and 17%, respectively, but the ranges spanned the gamut from 5% up to 41%. Grade 3 or greater adverse events were less common, at 2% and 3% for GI and GU, respectively. Univariate analyses uncovered no statistically significant associations between the reported toxicity rates and factors such as RT fields (i.e., inclusion of pelvic nodes for microscopic disease vs. prostate-region only), RT dose, RT technique, androgen deprivation therapy use, or time points of toxicity reporting. However, the rates in this meta-analysis with more modern RT techniques are reassuring compared to historical data on late grade 3 or greater toxicities from landmark RTOG prostate cancer studies using only 2D- and 3D-conformal RT, which noted 5% to 8% for GU toxicity and 1% to 4% for GI toxicity.21
Furthermore, analysis of the subset of randomized dose-escalation trials demonstrated that dose and technique can matter. A 10 to 14 Gy increase in RT dose (i.e., 64-70 Gy to 74-80 Gy) could roughly double the rate of severe late GI toxicity, whereas IMRT and PBRT were associated with decreased GI toxicity rates relative to 3D RT. Among the subset of “high RT dose” studies (73.8 Gy or higher), multivariate analysis found that, compared with 3D RT, IMRT was associated with a statistically significant decline in grade 2 or greater late GI toxicity, whereas PBRT use was associated with decreased grade 3 or greater late GI toxicity and a trend toward decreased grade 3 or greater late GU toxicity. Ohri et al did note that although relatively recent techniques have inherently shorter follow-up, there is no evidence in the meta-analysis that morbidity rates increased with longer follow-up or increasing time-to-toxicity reporting.20 In fact, the latter was associated with a statistically significant decrease in the incidence of both moderate and severe GI events, suggesting a real but transient nature of some adverse effects.
Prior reports have not been able to clearly delineate the impact of RT dose and RT technique on late morbidity. An Agency for Healthcare Research and Quality study noted no difference in toxicities with increasing RT dose, and the Institute for Clinical and Economic Review rates IMRT as “unproven with potential.”16 One thing is key—patient-level data, although cumbersome, are critical to the process, as large population-based datasets are only now capturing trends toward increasing dose and technique sophistication and do not adequately provide insights into consistently graded provider- or patient-reported toxicities. Even with patient-level data, the subjectivity of toxicity scoring and potential misattributions of cause can always muddy the waters.
Also difficult to capture and measure are the incremental benefits afforded by continued improvements in patient immobilization, onboard visualization prior to each radiation treatment, and the overall field of tumor imaging, which permits more accurate staging and localization to better target the RT. But each of these assets will continue to define our field and contribute to a more favorable therapeutic ratio—that elusive but critical balance between maximal tumor control and minimal normal tissue complications. These technologically driven efforts, combined with incorporation of biologically rooted biomarkers to predict RT toxicity profiles, will help radiation oncologists continue to pave the way toward increasingly personalized medicine and RT.
About the Author: Dr. Vapiwala is associate professor and vice chair of education in the Department of Radiation Oncology as well as assistant dean of students in the Perelman School of Medicine at the University of Pennsylvania.