EBRT is commonly utilized to treat men with clinically localized or locally advanced prostate cancer. EBRT alone may be applied to any clinical stage—T1 to T4—but is also often used with adjunctive anti-androgen therapy for selected higher-risk patients. We will address issues in regard to radiation dose, volume, technique below, and here will focus on the combination of ADT with RT and the role of RT after RP.

For high-risk prostate cancer, there is level I evidence from randomized trials supporting an overall survival advantage with the combination of ADT and RT. The question of ADT added to RT was first evaluated in a series of randomized trials which revealed that for patients with locally advanced (T3–T4) or high-grade prostate cancer (Gleason 8–10) ADT when added to conventional dose EBRT increased biochemical control and decreased risks of metastasis, death from prostate cancer, and all cause mortality (38,39,40,41,42). Therefore, for patients with the highest risk disease the use of ADT for between 28 and 36 months along with conventional dose EBRT appears better than either EBRT alone or EBRT delivered with shorter duration ADT (4–6 months) (40,42). For example, on RTOG 92-02 the addition of 24 months adjuvant ADT in addition to 4 months neo-adjuvant ADT resulted in a an absolute 16% increase in biochemical control at 10 years (p < 0.0001) along with improvements in local control (10% improvement, p < 0.0001), metastasis (8% improvement, p < 0.0001), and disease-specific survival (5% improvement, p = 0.004). Overall survival, however, was not increased for all patients, but was significantly improved for those with Gleason 8 to 10 disease (32% vs. 45%, p = 0.006). More recently it was also demonstrated that for node-negative men with high-risk prostate cancer the addition of conventional dose RT dramatically increased disease outcome when combined with long-term ADT as compared to long-term ADT delivered alone (43). There was an absolute risk reduction of 50% for biochemical failure at 10 years (75% vs. 26%, p < 0·0001) with the addition of RT. More importantly, the cumulative incidence at 10 years for prostate cancer-specific mortality was 24% in the ADT alone group and 12% in the ADT plus RT group (p < 0.0001). While the cumulative incidence for overall mortality was 39% in the endocrine alone group and 30% in the endocrine plus RT group (p = 0.004).

The role of ADT in intermediate risk disease has been addressed in two recent randomized trials which would suggest a benefit to shorter course anti-androgen therapy (4–6 months) when treated with conventional dose RT of < = 70 Gy (44,45). The role for ADT in patients treated with higher dose RT (particularly if intermediate risk) is an area of active evaluation at this time. Finally, there does not appear to be meaningful benefit with the addition of short-term ADT in either cancer-specific or overall survival in men with clinical low-risk disease by NCCN risk criteria even when treated with RT dose < 70 Gy (45).

EBRT is also commonly utilized as either adjuvant or salvage therapy in the post-prostatectomy setting for patients with clinical high-risk features or with a rising PSA following RP. For adjuvant therapy there is now level I evidence from randomized trials supporting a decline in biochemical failure, metastasis, and prostate cancer-specific death
with an improvement in overall survival with the use of adjuvant RT (ART) (46,47). The South West Oncology Group (SWOG) 8794 trial randomized men (stage pT3N0M0 disease or positive surgical margins) to observation or ART arms where the RT dose ranged from 60 to 64 Gy, with treatment portals including the prostatic fossa and paraprostatic tissues. With >12 years of follow-up the use of ART was associated with a 50% relative reduction in the risk of PSA recurrence in the ART group (p < 0.001). The use of ART yielded an absolute 10% improvement in metastasis-free survival at 10 years (71% vs. 61%, p = 0.016) and an 8% improvement in overall survival (66% vs. 74%, p = 0.023). Interestingly, the magnitude of benefit was similar for those with or without detectable PSA postoperatively although the utility of super-sensitive PSA remains to be determined (18). Two additional randomized trials support similar benefit in biochemical control of prostate cancer using ART although follow-up in these studies is insufficient to address more clinically meaningful end-points (48,49).

The role of salvage RT for patients with a rising PSA after RP has not been evaluated in randomized trials (although certainly a significant portion of patients on both the SWOG and EORTC trials had detectable PSA at the time of irradiation and as such would be consistent with salvage therapy). In general approximately 30% to 40% of patients given salvage RT will have long-term disease control (50,51). Preoperative, pathologic, and other clinical features have been utilized to construct a nomogram predicting the likelihood of PSA control (51) which was subsequently validated in a large multi-institutional data set (50). In addition, others have presented data supporting an improvement in prostate cancer-specific survival with the use of salvage RT particularly in those with the most rapidly rising PSA (52). No studies have directly compared ART (delivered 10–12 w after surgery) due to the presence of high-risk pathologic features even with undetectable PSA to salvage RT delivered at the time of rising PSA. However, a number of retrospective reviews have suggested an improved rate of biochemical control using ART as compared to salvage RT (53,54). In addition, the role of ADT concurrent with either ART or salvage RT has not been clearly established with some favoring its use (55) with others finding no benefit to concurrent ADT with RT in the salvage setting (52). Nevertheless, in Great Britain it has been assumed that salvage RT is the standard of care and a study has been undertaken test if ART is superior to salvage RT, and at the same time to evaluate the role of ADT concurrent with RT using a four arm dual-randomized trial (the Radiotherapy and Androgen Deprivation in Combination After Local Surgery (RADICALS) trial) (56).

Interstitial brachytherapy is a popular alternative to EBRT for delivery of RT to the intact prostate. Potential advantages of brachytherapy over EBRT include the convenience of undergoing a 1-day outpatient procedure, the lack of organ motion that can confound EBRT planning and delivery, and the use of low-energy isotopes with rapid dose fall-off outside the prostate, potentially minimizing treatment-related morbidity. Most commonly, LDR permanent interstitial radioisotopic implants are performed using I-125, Pd-103, or more recently, Cs-131 seeds which are placed transperineally via needles inserted through a template. HDR brachytherapy can also be used by similar needle placement of catheters that are used for remote afterloading of a high activity Ir-192 source. Multi-institutional and long-term data suggest that brachytherapy performs comparably to surgery or EBRT for low-risk patients (57,58,59). Brachytherapy may also be combined with a conventional course of EBRT to 45 to 50.4 Gy, particularly for patients with adverse risk features (60).

Since the incidence of urethral stricture can be high after transurethral prostatic resection (TURP), implant is usually not recommended for stage T1a and T1b tumors. Other relative contraindications to brachytherapy include large prostate size, a prominent median lobe, or significant preexisting obstructive urinary symptoms, as these may be associated with technical difficulty in performing an adequate implant or increased risks of complications. Patients with locally extensive tumors—clinical stages T3 and T4—are generally not suitable candidates for isotope therapy except possibly as a boost technique. Brachytherapy is also starting to have a role in salvage treatment after localized recurrence for patients treated with definitive RT although the results from this form of treatment at this time are very limited (61,62,63,64,65).

Cryotherapy, or freezing the prostate by circulating liquid nitrogen through interstitial catheters, is also utilized as curative treatment for clinically localized disease. Early results reported high complication rates although with newer techniques the complication rates have declined. Nevertheless, sexual dysfunction remains with near 100% incidence (35). A Canadian randomized trial did compare short-term ADT followed by either EBRT (68–73.5 Gy) or cryotherapy in men with localized prostate cancer where most (80%) had NCCN high-risk disease (66). Sexual dysfunction was significantly greater in those treated with cryotherapy (67). There were no significant differences in clinical outcome between treatment arms although a number of treatment features were suboptimal for the RT group including: low RT dose (most ≤70 Gy) (68,69), no pelvic RT (70), and only a short course of ADT (40,42). Each of these factors has been demonstrated alone to increase the rate of biochemical control for men treated with EBRT for prostate cancer. In addition, early repeat cryotherapy was not considered a treatment failure so a modest number of patients were retreated after biopsy-proven residual disease in the cryotherapy arm. Nevertheless, cryotherapy may be an option for patients with biopsy-proven recurrence after previous RT (71).

Noncurative options for prostate cancer include watchful waiting and active surveillance (both covered above) as well as androgen-deprivation hormonal therapy. Types of hormonal therapies include: orchiectomy, luteinizing hormone-releasing hormone (LHRH) agonists (e.g., leuprolide, goserelin) or antagonists (e.g., abarelix, degarelix), nonsteroidal anti-androgens (e.g., flutamide, bicalutamide) and estrogens. These hormonal agents can be used as debulking and anti-metastatic measures; however, these treatments are not curative unless combined with effective local therapy (43). They are, however, usually the first-line and most effective therapies available for disseminated disease. New systemic innovations are on the horizon with the recent reporting of the first cytotoxic chemotherapy agents that demonstrated improved survival in men with metastatic, hormone-refractory prostate cancer (72,73). In addition, newer hormonal agents are being developed which have resulted in promising PSA-based responses even in men with what was considered hormone refractory prostate cancer (74,75,76). The roles of these and other newer agents both alone and when combined with RT remain to be seen.

Once the process of defining the patient positioning and localization have been determined the process of treatment planning can commence. Most centers currently use forward planned, CT-based, 3D-CRT techniques, or inverse-planned IMRT. Although commonly utilized previously the added benefit of rectal, bladder, and/or urethral contrast when using CT-based planning is unclear. If the decision is made to treat the pelvic nodes, the classical superior bony landmarks in the AP view include the bottom of the L5 vertebral body to encompass the common, internal and external iliac nodes, midway through the sacroiliac joints to encompass the distal common, internal and external iliac lymph nodes or the bottom of the sacroiliac joints to encompass the internal and external iliac lymph nodes. The inferior margin is often set 1 cm below the ischial tuberosities inferiorly, and 1.5 to 2 cm to either side of the lateral bony pelvis laterally. In the lateral projection the anterior border is placed at the most anterior point of the pubis symphysis. The posterior border generally falls along the posterior ischium, although the field is usually shaped to bisect the rectum when filled with 20 to 30 mL barium GI contrast, administered via rectal tube. Retrograde urethrography may be performed using a small-gauge catheter to administer 5 to 10 mL of 30% Hypaque into the urethra with immediate penile clamping. The narrow point, or beak, in the contrast column denotes the level where the urethra passes through the urogenital diaphragm (109,110). The most inferior aspect of the prostate gland is typically located 1 cm above the urethral apex, and the inferior field margin is placed 1 cm below the apex or beak, allowing a 2-cm inferior margin on the prostate.

In most practices, CT planning has now largely supplanted the setting of fields using 2D fluoroscopic techniques. CT images facilitate the identification and localization of both target and normal tissue structures (87). Lymph node regions follow the path of the common iliac artery and its bifurcation into the internal and external iliac arteries, and can usually be easily traced even without intravenous contrast administration. Compared with the estimated general localization provided by bony landmarks in 2D planning, the size, shape, and location of the prostate and particularly the seminal vesicles can be visualized and defined more precisely with CT planning. The seminal vesicles drape around the rectum in some patients, and standard blocking in 2D fields set using rectal contrast may not provide adequate coverage of these structures. Identification of the prostate apex on axial CT is sometime difficult and may be facilitated with MRI or with sagittal and coronal reconstruction of CT data sets (111). Normal tissue structures including the rectum, bladder, loops of small bowel, penile bulb, and even important anatomic anomalies such as pelvic or horseshoe kidneys can be readily identified and localized with CT planning.

Fluoroscopic or CT simulation for treatment planning may be performed with the patient either supine or prone, with or without customized immobilization as discussed above. External marks are placed on the patient or immobilization device, often chosen to designate a preliminary isocenter position. These locations may be delineated with radiopaque markers taped to the patient in the x, y, and z axes that can be referenced to the image set, facilitating the transposition of the isocenter from the treatment plan to the patient at the time of treatment. For CT-based planning, CT images are obtained at 1 to 5 mm intervals through the levels to be encompassed within the treatment fields. Target and normal structures can then be defined and expanded to create the relevant PTVs on axial CT images within specialized treatment planning software (Fig. 26.3).

The issues regarding immobilization technique, patient positioning, and image guidance were discussed above. Common choices for immobilization include the Alpha Cradle™, thermoplastic cast material similar to Aquaplast, and “vacuum lock” molds. The CT scan must be done with the patient in the planned treatment position in the immobilization device, with attention toward patient comfort and maximization of daily reproducibility. Alternatively, for ease of position adjustment, some practitioners utilize supine position without an immobilization device if daily prostate localization techniques are employed (see above). An advantage of the prone position is to shift the seminal vesicles forward, away from the
rectum, potentially improving rectal sparing (112). On the other hand, prone positioning may be associated with more internal organ motion, including respiratory motion particularly in men with any degree of obesity (100). Roach et al. advocated nonuniform expansions around the prostate because of the variation in prostate movement depending on direction, the status of rectum and bladder filling as well as radiation tolerance of these two structures (113). The recommendations included anterior and inferior margins of 2 cm, lateral and superior margins of 1.5 cm, a posterior margin of 0.5 to 0.75 cm, and that CT planning be performed with the rectum empty and the bladder full. If bladder fullness is inconsistent from treatment to treatment, superior and inferior margins are more generous and accommodating, and this also helps to keep small bowel out of the field. If the rectum is empty when fields are planned, a rectum that is fuller during treatment only pushes the prostate further into the field, and 2 cm anterior margins should still provide adequate coverage of the CTV. However, with the use of image-guided therapy smaller margins may be appropriate (6).

Traditionally, a four-field technique is generally used to treat pelvic lymph nodes: anterior, posterior, right, and left lateral portals. Blocks can be drawn manually on simulation films or BEV digitally reconstructed radiographs (DRRs) from CT planning software. Construction of these blocked fields can be facilitated by reviewing the pelvic CT scan and entering lymph node volumes that track along the course of the iliac vasculature (Fig. 26.4). Depending on the external contour, which can be done manually or automatically by CT scan, small angle wedges on the lateral fields may help to produce a more symmetrical dose distribution. In addition, IMRT is being utilized with greater frequency to treat pelvic lymph nodes in which case prostate, seminal vesicles, pelvic lymph nodes, and OAR must all be defined for appropriate treatment planning and optimization. Examples of recommended OAR tolerances are provided (Table 26.5) (87).

Two-dimensional planning to treat the prostate and seminal vesicles has classically used three fields (anterior, right, and left lateral fields), four fields (anterior, posterior, right, and left lateral fields), or even two bilateral 120° arc fields. With 2D planning, target volumes tend to be more generous and blocking more conservative to allow for the inaccuracies of manual transposition of PTVs to the simulation plain films. For CT-based planning, CT images are entered into a treatment planning software package. The lymph node regions, prostate, and seminal vesicles are outlined by the physician, and can be expanded by adding appropriate margins to establish the CTVs and PTVs. As noted previously, margins may vary between institutions and individual physicians and by the use of IGRT. Normal structures are also outlined including bladder, rectum, penile bulb, femoral heads, loops of small bowel or any other structures to which the physician may wish to limit dose. Treatment planning software packages allow manipulation of virtual 3D reconstructions of these volumes, allowing a variety of views for the planner, including “beam’s eye view” (BEV) perspectives. These features are designed to allow the user to manipulate beam orientation (gantry, table, and collimator angles) and shaping (blocking or multileaf collimation) to encompass targets and avoid normal structures. For forward-planned 3D conformal treatment planning, a combination of beams is devised complete with beam modifiers such as wedges or compensators. High-energy beams such as 10 to 18 MV are customary for pelvic lymph node and prostate irradiation, and the beams should certainly be 6 MV or greater. The software allows rapid calculation of the dose that can be displayed on the overlying images in the form of isodose lines or surfaces (Fig 26.5). Additional tools including dose-volume histograms (DVHs) helps the clinician to compare treatment plans to determine the optimal plan for coverage of target volumes and avoidance of normal tissue structures (Fig. 26.6).

Forward planned 3D CRT beam arrangements for prostate cancer treatment may include the standard axial anterior, posterior, and lateral beams common to 2D prostate plans, but can also easily make use of oblique and nonaxial beam angles. Rotation of the collimator is commonly used to place the orientation of the multi-leaf collimator appropriately so as to maximize normal tissue blocking and target coverage. Segmental fields, physical wedges and dynamic wedge techniques can be used to improve target coverage and dose homogeneity. In addition to the classic
four-field box technique, 3D CRT beam configurations include a six-field conformal technique (114) as well as a four-field variation using lateral and anterior obliques that eliminate the posterior-oblique beams. Marsh et al. have found that the four-field technique, using laterals and anterior–inferior obliques, often improves the rectal DVH compared with axial plans (115) (Fig 26.7). This technique requires supine patient positioning to achieve table clearance for the anterior–inferior beams.

The application of IMRT technology to prostate cancer EBRT planning has enabled further improvements in normal tissue DVHs, mainly due to the ability of this technology to achieve convex dose distributions, reducing rectal dose (Fig. 26.8). The clinical significance of this advance toward reducing rectal toxicity remains undetermined although some groups have suggested that with the combination of IGRT and IMRT high dose EBRT can be delivered without increases in toxicity as compared to that observed in historical control groups treated to lower doses with less sophisticated techniques (5). For instance Zelefsky et al. noted an actuarial rate of grade 2 or greater rectal toxicity at 10 years of 13% in patients using 3D CRT and 5% using IMRT (p < 0.0001) (116) Of note IMRT patients received higher doses of RT than 3D conformal patients, but IGRT was also used with much greater frequency in the IMRT group as well. The inverse planning process starts with the contouring of target and normal tissue volumes on an axial CT image set—just as for 3D CRT planning. However, instead of inputting a candidate beam arrangement into the planning software and calculating a dose distribution for that plan, the IMRT software package uses an optimization algorithm to find a beam “solution” that maximizes a set of “cost functions.” Typically,

the process begins with a certain number of axial beam angles for the algorithm to utilize. Each beam is subdivided into square “beamlets” that represent the incremental movement of multileaf collimator leafs across the field in 1 cm or less increments. The algorithm “modulates” the fluence through each of these beamlets in an iterative process, evaluating and comparing each candidate plan to identify a better one. Each plan is evaluated for dose coverage of target volumes, as well as dose avoidance for normal tissue structures. Target coverage and normal tissue structure sparing is prioritized based on clinical judgment, and a cost for achieving or failing to achieve each objective is assigned. The algorithm will iterate until it identifies a plan that best achieves the stated objectives. Once the optimization algorithm produces a plan, a formal dose calculation is performed, and a quality assurance procedure is followed to ensure that the delivered dose will be within a desired level of accuracy. For the prostate, where the target shape and volume are relatively consistent from patient to patient, a template can be established to make the treatment planning process more efficient. Examples of typical dose constraints for rectum, bladder, femoral heads, and both large and small bowel are given in Table 26.5. The sparing of the penile bulb is controversial as different authors have come to differing conclusions about the utility of sparing the penile bulb in preserving sexual function (117). However, it is clear that if a penile bulb dose constraint is observed that it should not be at the cost of decreasing dose to the PTV. An example of an IMRT plan as well as dose volume histograms for both an IMRT plan and a 3D-CRT plan are given in Figure 26.9.

The sharp dose gradients that can be created using highly conformal techniques such as IMRT create a significant risk of missing the target due to inter- and intra-fraction setup uncertainty and organ motion. Therefore, highly conformal therapy should be combined with IGRT to ensure the accuracy of treatment and to minimize marginal misses due to setup error or changes in prostate position.

As advances in treatment techniques have allowed more conformal treatment of the target volumes with decreased dose to normal tissues, improved outcomes have been documented with increased doses to the prostate. However, the optimal treatment doses for various risk categories of prostate cancer remain unknown. With traditional 2D treatment techniques, doses of 45 to 50 Gy to the pelvis, 55 to 65 Gy to the seminal vesicles and 65 to 72 Gy to the prostate were typical. With the advent of 3D CRT, IMRT, and other highly conformal techniques, doses have been escalated without significant increases in acute side effects or late complications (118,119,120). In a landmark trial, the MD Anderson Cancer Center compared 70 to 78 Gy using 3D CRT techniques and found that patients receiving the higher dose had significant improved biochemical failure-free survival, particularly those patients with a pretreatment PSA above 10 ng/mL or with higher-risk clinical features (Table 26.6) (121). Toxicity; however,
was increased as a function of the volume of rectum treated to higher doses. Another significant randomized trial of dose escalation used a combination of external beam photon radiotherapy with a proton beam boost. Zietman et al. have reported that for mostly low-intermediate risk prostate cancer patients receiving 79.2 Gy had a 15% absolute reduction in the risk of biochemical recurrence at 10-years as compared to those receiving 70.2 Gy (32.4% vs. 16.7%) (2). Given the lack of biologic difference between protons and photons there is no reason to presume that these results are unique to the use of a proton boost. Using this technique, although there was a slight increase in mild acute rectal toxicity, there was no significant difference in clinically substantial toxicity rates or patient reported quality of life (122). These randomized trials as well as others (123,124), combined with many retrospective studies, give strong evidence that dose escalation above 70 Gy may significantly improve prostate cancer biochemical cure rates. However, at present, dose escalated radiation has not resulted in clear differences in rates of metastasis, prostate cancer-specific death, or overall survival. Encouragingly, when dose-escalation is performed using conformal techniques toxicity rates seem to be acceptable (4,33,124). The potential interaction between dose-escalation and hormonal therapy in terms of disease control and toxicity has not yet been thoroughly explored although both the Dutch and British dose escalation studies included patients with short-term ADT; therefore, it is clear that dose escalation may still provide a benefit even in the setting of ADT use (123,124).