The prostate gland lies within the lower pelvis immediately superior to the musculofascial floor. It is the largest accessory gland of the male reproductive tract, weighing approximately 30 grams in men without prostatic pathology. A fibrous capsule surrounds the gland, which is continuous with the surrounding connective tissue stroma, but is not present at the apex of the prostate. The fascia surrounding the prostate gland is variably adherent to the capsule. Shaped like an inverted tapered cylinder, the prostate base is immediately caudad to the bladder and its tapered apex cephalad to the urogenital diaphragm. The apex contains muscle fibers continuous with the urogenital diaphragm. The prostate lies posterior to the symphysis pubis, connected via the puboprostatic ligaments, and anterior to the rectum, separated from it by the anterior portion of Denonvilliers fascia (rectovesical septum), a thin layer of connective tissue that separates the prostate and seminal vesicles from the anterior rectal wall. The prostatic urethra passes through the gland from the base of the bladder to the membranous urethra. The ejaculatory ducts course from the convergence of the ducti deferens and the seminal vesicles obliquely, anteriorly and caudally, through the posterior prostate to communicate to join with the prostatic utricle to open into the prostatic urethra.

The blood supply of the prostate gland is from the prostatic branches of the inferior vesicle, pudendal, and middle rectal arteries. Venous drainage is shared with the penis, urethra, and lateral pelvic organs via the dorsal venous plexus (Santorini’s plexus) to the internal iliac vein. The predominant course of lymphatic drainage is to the internal iliac nodes, but alternative drainage routes to the obturator, external iliac, sacral, vesicle, and rarely the periaortic lymph nodes are common. The prostate gland has dual autonomic innervation from the prostatic nerve plexus. Parasympathetic innervation is provided by the pelvic splanchnic nerve (S2-4) and sympathetic innervation is provided by the inferior hypogastric plexus (T10-L2). The neurovascular bundle (NVB) courses between fascial layers on the posterior lateral surfaces of the prostate gland and is intimately related to the dorsal surfaces of the seminal vesicles.

The prostate has been anatomically described in terms of both lobes and zones (McNeal 1981). There are four lobes of the prostate: anterior, posterior, median, and lateral. The anterior lobe is located anterior to the urethra and contains no glandular tissue, just fibromuscular stroma. The median lobe lies in the center of the gland, with the urethra anteriorly and the ejaculatory ducts posteriorly. The paired right and left lateral lobes are the largest lobes and contain the bulk of glandular tissue, being separated into halves by the prostatic urethra. The posterior lobe lies at the dorsal portion of the gland and can be palpated via a digital rectal examination.

The zonal anatomy of the prostate (Fig. 2) was first described by McNeal in 1981 and is used more commonly. This system is based on embryonic development patterns and includes four prostatic zones described relative to the urethra: the peripheral zone, the central zone, the transition zone, and the anterior fibromuscular stroma. The peripheral zone is a horseshoe-shaped structure extending posteriorly and laterally around the inner regions of the prostate gland and contains approximately 70 % of the prostatic glandular tissue. It is embryologically derived from a double row of developing prostatic ducts that extend posterolaterally from the distal urethra. The central zone contains approximately 25 % of the glandular tissue and is formed from developing ducts that grow cephalad, posteriorly and laterally into the mesenchyme surrounding the ejaculatory ducts. It is a cone-shaped volume extending from the bladder base to the verumontanum, encompassing the ejaculatory ducts. The transition zone lays anteromedial to the central zone and surrounds the mid-prostatic urethra. Benign prostatic hypertrophy occurs most commonly in the transition zone. The fibromuscular stroma is the anterior part of the prostate and contains little to no glandular tissue but does account for one-third of the total bulk of tissue within the prostatic capsule.

The seminal vesicles consist of two lobulated glands posterior to the bladder and prostate gland that converge medially. They are inferior and lateral to the ampulla of the ductus deferens and lie against the fundus of the bladder. The majority of these glands lie in the retroperitoneum and are enclosed by dense fascia, except for the most cephalad portion, which is intraperitoneal. Their cephalad surface abuts the posterior bladder and their caudad surface interfaces with the anterior rectum at the rectovesical (Denonvilliers’) fascia. The ureters lie medial to each seminal vesicle. The seminal vesicles converge with the vas deferens to form the ejaculatory ducts which enter the prostate gland. They produce alkaline secretions that are added to the semen. Their blood supply is from the inferior vesicle artery and vein and middle rectal artery and vein. Their main lymphatic drainage is to the internal iliac lymph nodes. Sympathetic innervation is via the superior lumbar and hypogastric nerve, which regulates rapid contraction of smooth muscle cells during ejaculation. Parasympathetic innervation is from the pelvic splanchnic nerve and from the inferior hypogastric plexus.

The male urethra courses from the base of the bladder to the tip of the penis and is subdivided into prostatic, membranous, and spongy portions. It provides a conduit for urine and semen to be excreted from the body. The prostatic portion enters the anterior portion of the prostate gland and travels with slight anterior concavity so that it courses posteriorly toward the mid gland before resuming an anterior position at the prostatic apex. The change from the posterior to anterior course happens at a sharp, approximately 35-degree angle, at the level of the verumontanum, a midline protrusion along the posterior urethra where the ejaculatory ducts enter the prostatic urethra. The prostatic urethra is about 3-cm long and is the widest and most compliant portion of the urethra. The membranous urethra is approximately 1-cm long and extends through the urogenital diaphragm from the apex of the prostate gland to the bulb of the corpus spongiosum. The external urethral is a complex anatomic and physiologic structure closely related to the urogenital diaphragm. The innervation for the external sphincter travels near the apex of the prostate. The external sphincter is a circular muscle under partial voluntary control (innervated by the somatic nervous system) in the urogenital diaphragm and regulates the passage of urine through the urogenital diaphragm. Once the urethra enters the penile bulb, it becomes the spongy (penile) urethra which extends to the navicular fossa at the tip of the penis. The spongy urethra is about 6-cm long and is contained in the corpus spongiosum.

The penis is subdivided into three portions: the root, the body, and the glans. The root is anchored to the os pubis, symphysis pubis, and ischium by the crura and suspensory ligaments. The body of the penis extends between the root and the glans, a dome-shaped extension of the corpus spongiosum. The penis contains three erectile bodies, each encased in a fascial cover: two paired corpora cavernosa and the corpus spongiosum. All corpora are further encased together by the fibrous Buck’s fascia. The spongy urethra courses within the corpus spongiosum after entering this structure at the penile bulb, the most proximal portion of the spongiosum. The penile skin rests upon Colles’ fascia, which is continuous with Scarpa’s fascia of the abdominal wall.

The arterial supply of the penis and urethra is from the internal pudendal arteries which branch into the deep penile artery, the bulbourethral artery, and the dorsal artery of the penis. The deep artery supplies the corpora cavernosa, while the others supply the remainder of these organs, including the urethra and corpus spongiosum. There are multiple routes of venous drainage of the penis. The superficial dorsal vein is external to Buck’s fascia, while the deep dorsal vein lies deep into this thick fascial layer between the paired dorsal arteries. The dorsal veins join the pudendal plexus and eventually the internal pudendal vein. The penile skin and superficial fascia’s main route of lymphatic drainage is to the superficial inguinal lymph nodes. The predominant lymphatic pathway for the glans penis is through the external iliac nodes. The urethra drains to the internal and common iliac lymph nodes.

The anatomy of the pelvic floor is challenging to visualize on CT or MRI, and hence definition of the penile bulb varies. This may contribute to inconsistent reports (5–15). The penile bulb appears as an oval-shaped, hyperintense midline structure on T2-weighted MR images; on axial CT imaging it is bounded by the crura, corpora spongiosum, and the levator ani muscle. At University of California-San Francisco, the bulb is defined as the most proximal portion of the penis sitting immediately caudal to the prostate.

Many excellent textbook references are available with detailed descriptions of the histology of the organs of the distal male genitourinary tract (McAninch 2008; Bostwick 1998). Several cellular subtypes are evident upon histologic examination of the prostate, including basal, secretory, neuroendocrine, urothelial, and ejaculatory duct cells (Fig. 3a, b). Under the thin fibrous capsule of the prostate are smooth muscle layers and layers of collagen which also surround the prostatic urethra, forming the involuntary sphincter. Deeper to this layer is the prostatic stroma, which encases the glandular compound tubuloacinar secretory units, numbering 30–50. These units are arranged concentrically into three groups based on their relationship to the urethra. The majority of these ducts open on the floor of the prostatic urethra near the verumontanum. The main group is farthest from the urethra and constitutes the largest in number. The mucosal and submucosal glands, along with their adjacent stroma, undergo hyperplastic change with age, leading to compression of the urethra and symptoms consistent with benign prostatic hyperplasia (BPH).

The glandular acini are highly redundant and are lined by two layers of cells: a luminal layer of tall columnar cells and a layer of cuboidal cells adherent to the underlying basement membrane. Between prostatic glands is fibromuscular stroma containing smooth muscle, collagen, and elastic fibers. In parenchymal portion, glands consist of a simple high cuboidal/low columnar epithelium.

Histologically, each seminal vesicle consists of a single coiled tubular structure that forms a large irregular lumen with many mucosal folds. The epithelium is composed of mixed columnar and pseudostratified columnar cells with mucosal crypts generated by infoldings of the mucosa. It is surrounded by two muscular layers that contract to create positive pressure to move secretions through the lumen and into the ejaculatory duct. The duct of each seminal vesicle combines with the ductus deferens to form the ejaculatory duct.

Deep into the superficial urethral mucosa is the submucosa which contains smooth muscle, elastic tissue, and other stromal elements. The superficial histology of the urethra is dependent on its anatomic subsection. In the prostatic portion, it is lined with transitional epithelium. In the spongy portion, it contains stratified columnar epithelium until the navicular fossa, where a transition to non-keratinizing stratified squamous epithelium is seen. The lamina propria of the spongy portion of the urethra merges with the surrounding corpus spongiosum.

In the penis, cavernous bodies are irregular and lined with simple squamous endothelium and contain venous blood. The cavernous spaces of the cavernosa are larger than those of the spongiosum and have thinner stromal trabeculae, allowing the cavernosa to become more turgid than the corpus spongiosa when the penis is tumescent. (Fig. 3c, d).

The penile bulb consists of corpus cavernosum and responds during sexual arousal becoming distended, and is the base of the penis when erect.

The prostate gland is arranged in three concentric groups: 1. Main, one-half glandular; 2. Submucosal, one-fourth fibrous tissue; 3. Mucosal, one-fourth involuntary muscle (Fig. 4).

The prostate gland is influenced by the male sex hormones, i.e., testosterone and adrenal androgens which are converted into dihydrotestosterone (DHT) which is 30 × more potent that testosterone. The DHT stimulates growth of the normal prostate glandular epithelium. The prostate gland secretes prostate acid phosphatase and prostate-specific antigen (PSA) which are incorporated into seminal fluid of which only a small fraction enters circulating blood (i.e. 4 Ng/ml). As the prostate enlarges and undergoes malignant transformation, PSA increases to >4 Ng/ml and serves as a biomarker for cancer increasing to 4.0–10.0 Ng/ml. In addition the prostate gland secretes prostatic acid phosphatase (PAP), an enzyme that regulates prostate cell growth and metabolism of the glandular epithelium. The fibronolysin in the secretions liquefies semen.

Seminal vesicle secretions consist largely of prostaglandins (accounts for the name of prostate gland) which during ejaculation discharges its secretions and assists in flushing out the semen through the urethra.

During its erectile stage due to filling of its vascular spaces, allows for penetration of the vagina and its ejaculate of semen via the urethra, initiates the fertilization of the ovum (Fig. 4).

Male sexual performance consists of specific sequence of events (Fig. 4):

Sexual arousal → libidinous desire → erection → ejaculation → orgasm → detumescence

The ducts of the prostate functional units empty into the prostatic urethra. The secretory elements (ducts) are surrounded by smooth muscle and connective tissue that separate the secretory units. Prostatic secretions are expelled into the urethra and join other components of the seminal fluid when these smooth muscle units contract. Prostatic secretions are alkaline (pH 6.5) and serous, containing acid phosphatase, carbohydrates, lipids, lipofuscin, hormones, citric acid, amylase, and fibrolysin, which act to liquefy the seminal fluid.

Late histologic changes in irradiated benign prostate ducts include variable ductal atrophic change, cytologic and nuclear atypia, basal cell hyperplasia, increased foreign body giant cell reaction to corpora amylacea, nuclear pleomorphism, nuclear vacuolation, hyperchromatic DNA, and presence of prominent nucleoli. Several metaplastic changes have been noted, including mucinous metaplasia, squamous metaplasia, and Paneth-like cell change. There is no significant difference in the histologic appearance of irradiated benign glands when the patients are treated with androgen deprivation therapy (Gaudin et al. 1999; Magi-Galluzzi et al. 2003). The surrounding stromal cells also undergo chronic inflammatory processes. The molecular and cellular mechanisms underlying this chronic inflammatory change are well described. Briefly, after the initial insult of radiation therapy, a cascade of inflammatory mediators is initiated including TNF-alpha, interleukin-1, and interleukin-6 (Haase 2004; Bentzen 2006). In the later phases of fibrosis, TGF-beta and PDGF, among others, simulate the proliferation of fibroblasts and the synthesis or extracellular matrix constituents and MMPs. Connective tissue growth factor (CTGF) further promotes fibrotic change (Vozenin-Brotons et al. 2003).

The seminal vesicles produce a viscous fluid containing high amount of fructose that nourishes the sperm. Other components include simple sugars, amino acids, ascorbic acids, and prostaglandins.

Irradiation of the prostate gland leads to interstitial fibrosis. Atrophy and calcifications can be seen on CT and MR images (Fig. 6a, b). Radiation changes are also seen in the seminal vesicles, which lose volume and demonstrate low T2-weighted signal intensity on MR images after radiation. The prostatic and membranous urethras are very sensitive to radiation. The irradiated urethral tissue often develops strictures.

As aforementioned in the pathophysiology section, a vexing issue is persistence of cancer cells, the same is true for follow-up assessment by serial MRIs in follow-up. Comparing pre- to post-RT MRIs might be helpful in this regard, as is MR spectroscopy. Figure 6c illustrates an example of the utility of MR spectroscopy in defining tumor in the prostate.

Although the rectum is covered in a separate chapter, a discussion of this organ is merited here, as the appropriate dose–volume rectal constraints for prostate cancer radiotherapy is the most published area of normal tissue tolerance in this malignancy. The reason for this interest in rectal tolerance is clear. Unlike bladder toxicity, late rectal toxicity has been found to be directly and consistently correlated with physical treatment parameters such as dose and volume. Rectal complications are the main late toxicities that limit dose escalation in prostate cancer. General trends can be drawn from the available clinical literature regarding these constraints for conventionally fractionated radiation therapy despite institutional differences in prescribed dose, extent of contouring, radiation techniques, and outcomes measurement. Moderate to severe late rectal effects are dose dependent, occurring in around 25 % of men treated with 78 Gy and 13 % in men treated with 70 Gy (Pollack et al. 2002; Peeters et al. 2006a, b; Zietman et al. 2005). Despite this dose dependency, several reports indicate that careful planning techniques and intensity modulation can mitigate this toxicity and allow dose escalation without excess late GI toxicity (Zelefsky et al. 2002; Beckendorf et al. 2004; Peeters et al. 2006a, b).

Numerous studies have been published with late rectal bleeding as an isolated clinical endpoint with consistent contouring of the rectum. Jackson et al. from MSKCC reported on 171 patients treated with 3D conformal radiation therapy to a dose of 70.2 or 75.6 Gy. They found a significant dose–volume correlation with RTOG > 2 late bleeding and recommended, for the techniques and doses used, V40 < 60 % and, for patients treated with 75.6 Gy, a V77 < 14 % (Jackson et al. 2001). An Italian intergroup published two papers using RTOG > 2 late bleeding endpoints in patients treated in the range of 70–78 Gy and ultimately suggested the following dose–volume constraints: V50 < 60 %, V60 < 45 %, and V70 < 25 % (Fiorino et al. 2002, 2003). The recommendation of V70 < 25 %, using the same endpoint, was independently confirmed by the group at William Beaumont Hospital using adaptive image-guided radiation with prescription doses between 70.2 and 79.2 Gy (Vargas et al. 2005). Peeters et al. found that a V65 > 30 % was highly predictive for late bleeding requiring transfusions or interventional laser coagulation in patients treated between 68 and 78 Gy (Peeters et al. 2006c). When SOMA-LENT Grade > 2 late bleeding criteria was used as the clinical endpoint, recommendations are similar: V50 < 55 %, V60 < 40 %, V70 < 25 %, and V75 < 5 % (Fiorino et al. 2008; Fellin et al. 2009). In addition to these straightforward dose–volume constraints, several groups have generated normal tissue complication probability models and nomograms for late rectal bleeding that are generally consistent with the above studies (Rancati et al. 2004; Söhn et al. 2007; Valdagni Rancati et al. 2008).

Dose–volume constraint recommendations are only slightly altered when all forms of late rectal toxicity are included in outcomes analysis. The MD Anderson group reported on rectal complications from their randomized dose escalation trial and found that the rates of rectal complications were similar in both arms (70 and 78 Gy) but found a significant increase in late rectal complications when V70 > 25 % (Storey et al. 2000a, b; Kuban et al. 2008). When a different subset of patients treated in the same general dose range was evaluated from the same institution, the investigators found that the percentage of the rectum exposed to a certain dose, rather than an absolute volume, was predictive of late rectal toxicities. In this paper, the following recommendations were made: V60 < 40 %, V70 < 25 %, V75.6 < 15 %, and V78 < 5 % (Huang et al. 2002). Analysis of the high dose arm (74 Gy) of RTOG 94-06 yielded a recommendation to keep the V65 < 50 % to keep late rectal toxicity Grade I or less (Michalski et al. 2004). A recent report from MSKCC analyzing 478 patients treated to 86.4 Gy found a 4 % rate of CTCAE 3.0 late rectal toxicity with V47 < 53 % and V75.6 < 30 % at a median follow-up of 53 months (Cahlon et al. 2008). Numerous studies have found results similar to the results seen in these trials (Fonteyne et al. 2007; Karlsdóttir et al. 2008).

Taking these individual publications into account, there is general agreement that multiple dose–volume constraints in the intermediate (30–50 Gy) and high range (>70 Gy) regions of a histogram are prudent to shape the dose–volume histogram appropriately and to minimize the probability of late rectal bleeding to rates <10 %. It is also clear that while the volume of the rectum receiving a high dose is more predictive of late rectal toxicity, the amount of rectum receiving lower doses also predicts rectal toxicity, particularly with non-IMRT techniques. The distribution of dose on the rectum is also important, as increasing doses to the posterior rectal wall and upper rectum are associated with increase rates of late rectal toxicity (Skwarchuk et al. 2000; Fiorino et al. 2002; Heemsbergen et al. 2005; Peeters et al. 2006d; Munbodh 2008). A summary of recommended dose/volume constraints for late rectal injury is provided in Table 1. See “Radiation Induced Rectal Toxicity” for additional discussion.

Stool frequency has also been correlated with treatment-related dose–volume parameters. Fonteyne et al. reported that the rate of mild diarrhea was correlated with the volume of the rectum receiving 40 Gy, and chronic rectal urgency was correlated with the volume of the rectum receiving 70 Gy (Fonteyne et al. 2007). Similarly, Peeters reported that both the V40 and the mean rectal dose were predictive of stool frequency (Peeters et al. 2006c). See “Biophysiopatholo​gy of the Microvasculature​ and Microcirculation​” for additional discussion.

The prostate brachytherapy literature has consistently demonstrated that very small volumes of the rectum can tolerate very high doses of radiation, which is an important consideration in this era of highly conformal dose escalation. Snyder et al. found that when the rectum was contoured as a hollow structure from 9 mm below the prostate apex to 9 mm above the top of the seminal vesicles, the rate of grade 2 or greater proctitis was dependent on the volume of rectum receiving the prescription dose of 160 Gy. The rates of proctitis per absolute volume of rectal wall receiving prescription dose were: 0 % if < 0.8 cc, 7–8 % between 0.8 and 1.8 cc, and 25 %, and 24–25 % if > than 1.8 cc (Snyder et al. 2001). D’Amico’s group also found that the absolute volume of rectum receiving 100 Gy was predictive of needing argon plasma coagulation for late proctitis after brachytherapy, with no patients needing intervention when the volume of rectum receiving 100 Gy was below 8 cc, whereas 20 % of men needed intervention with argon plasma coagulation when the volume of rectum receiving 100 Gy was greater that 8 cc (Albert et al. 2008). Han et al. found a similar outcome when endoscopically proven radiation proctitis was used as the endpoint and the rectum was contoured as a solid structure, with the volume of the rectum receiving at least 100 % of the prescription dose being 2.5 cc in patients with endoscopically-confirmed proctitis and 0.6 cc in those with no evidence of bleeding (Han and Wallner 2001). The maximum point dose to the rectum after brachytherapy is also predictive of RTOG > 2 bleeding, with a 0.4 % toxicity rate for a maximum point dose of 150 Gy, a 1.2 % rate for a maximum point dose of 200 Gy, and a 4.7 % rate for a maximum point dose of 300 Gy (Waterman and Dicker 2003). Therefore, it is clear that small volumes of the rectum can receive very high doses, leading investigators to embrace more precise brachytherapy strategies, and by extrapolation, external beam treatments in combination with increasingly effective real-time image guidance solutions in an effort to dose escalate the prostate safely while sparing more of the rectum from the very high dose regions. It is evident that the smaller the amount of rectum irradiated, the less likely that late rectal bleeding will occur.

Dose volume constraints and their effect on acute GI toxicities from pelvic radiation has been studied less. Nonetheless, these appear also to be highly dependent on dose–volume parameters, typically with mean rectal dose and the volume of rectum receiving greater than 60 Gy. Peeters examined GI effects in the first 6 weeks of treatment and found that the mean rectal dose as well absolute and the relative amount of rectum receiving 5, 15, and 30 Gy were predictive of acute GI toxicity (Peeters et al. 2005a). Vavassori and Cheng also reported DVH correlations with rates of acute toxicity, including the mean dose, and the minimal dose to 10, 20, and 50 % of the rectum (Vavassori et al. 2007; Cheng et al. 2008). Other authors have offered more specific recommendations. Nuyttens’ group found that a V75 < 11 cc and a mean dose < 38 Gy minimized acute rectal toxicity (Nuyttens et al. 2002). Another report looking at patients treated with 78 Gy found that a V65 Gy < 20 % was protective against any grade I or higher acute GI toxicity (Karlsdttir et al. 2004). Of note, most of the above studies looking at acute side effects contoured the rectum as a solid structure, including the luminal contents. A summary of recommended dose/volume constraints for acute rectal effects is provided in Table 2.