Imaging in Localized Prostate Cancer

Introduction

Prostate cancer affects one in seven men in the United States, and it is estimated that 12% of affected men will die from their disease. Prostate cancer represents a spectrum of disease ranging from indolent with a low risk of mortality to very aggressive with a high risk of metastases leading to death. For most patients, prostate cancer will not be the cause of death but may nonetheless be a source of great anxiety and may lead to treatment-related adverse events. The paradox of prostate cancer is that while it is mostly a slow-growing, nonlife-threatening diagnosis, it can also be an aggressive cancer in which metastases portend a median 5-year survival of only 28%. Motivated by the hope that early detection could lead to improved outcomes, population screening with prostate-specific antigen (PSA) was initiated beginning in the late 1980s. Now that over a quarter of a decade has elapsed, it is clear that PSA screening has also led to overdiagnosis and overtreatment of low-grade prostate cancer since PSA cannot discriminate between low- and high-risk cancers with reasonable specificity. Moreover, many men underwent workups for a variety of pathologies including prostatitis and benign prostatic hyperplasia (BPH), which cause false-positive increases in PSA. The results of two recently completed large trials of screening, including the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial, and the European Randomized Study of Screening for Prostate Cancer showed a high rate (17–50%) of overdiagnosis of low-grade prostate cancer based on routine PSA screening with no perceptible benefit to the patient. Up to the time of this report, many men with low-risk disease were being aggressively treated with surgery or radiation therapy for low-risk disease. While this treatment no doubt cured patients of their prostate cancer, it also left them with decades of decreased quality of life. As a result of these two studies and several other smaller studies that reached the same conclusion, the US Preventative Screening Task Force (USPSTF) in 2012 issued a grade of “D” for PSA screening using existing criteria. This has been widely interpreted to mean PSA screening is of no value to patients. However, there is still no suitable alternative and no better screening method currently available. In 2015, a 3221-patient study in Toronto, Canada looked at the impact of the new PSA screening guidelines set forth by the USPSTF. As a result of curtailment of PSA screening by family practitioners, there was a 42.8% decrease in per-month detection of clinically significant prostate cancer (Gleason >7) in the year following the release of the updated PSA screening recommendation. Thus, while reducing the use of PSA screening certainly reduces the overdiagnosis of low-grade disease, it also reduces the potentially life-saving diagnosis of higher-grade disease. It is therefore important that a new balance be struck between the extremes of screening everyone and screening no one.

There is some hope that imaging, particularly prostate MRI, might prove a useful adjunct to PSA screening. It is often lost in the discussion that PSA is only partly to blame for the current dilemma. Elevated PSA leads to a systematic but undirected prostate biopsy where it is quite likely that small incidental islands of low-grade disease could be discovered. Thus, this “random” biopsy is equally to blame for the unintended diagnosis of many low-grade tumor islets. Meanwhile, larger and more consequential lesions outside of the normal biopsy template are likely to be missed. Therefore, the addition of MRI could be helpful in detecting lesions within the prostate more likely to be clinically significant and directing biopsies into those lesions while avoiding biopsies of normal-appearing tissue, which is nonetheless likely to contain indolent islands of low-grade cancer. Over the past decade, there has been rapid improvement in image-guided biopsy technology. Most importantly, the acceptance and use of multiparametric prostate MRI in combination with transrectal ultrasound (TRUS) biopsy has increased clinicians’ ability to detect clinically significant disease while reducing the detection of low-grade disease, and is now directly impacting the clinical management of patients with prostate cancer. While it is still too early to definitively state that this approach will result in a more efficacious screening method for prostate cancer, it is nonetheless very promising and readily applied today whereas new biomarkers are still years away from the clinic. This chapter will discuss the current state of imaging in prostate cancer detection, including ultrasound, multiparametric MRI (mpMRI), and PET/CT.

Ultrasound

TRUS is a widely available, portable, readily repeatable, and relatively inexpensive imaging technique. Ultrasound enables real-time visualization of the prostate, thus allowing the clinician to determine the gland volume and the distinction between the peripheral zone (PZ) and transition zone (TZ). Prostate cancers usually appear as hypoechoic regions on TRUS; however, this pattern can be mimicked by a variety of other pathologies such as BPH and inflammatory process. Moreover, not every prostate cancer is hypoechoic, and lesions commonly appear isoechoic, which reduces the diagnostic utility of TRUS. ECE (extracapsular extension) detection by TRUS is also limited except in very extensive cases. Color or power Doppler modes in TRUS can improve the tumor detection rate, but this is dependent upon the extent of angiogenesis. In smaller or less aggressive cancers, this affect may be minimal and lead to lower sensitivity rates. Today TRUS is mainly used for the purpose of guiding biopsies during systematic, blind sampling of the prostate.

More recently, contrast-enhanced TRUS with microbubbles has been reported to improve the sensitivity for tumor detection. Microbubbles are 5–10 μm gas-filled bubbles that can be seen on ultrasound. However, microbubbles tend to act like blood pool agents because of their large size, and often, only the vessels themselves are visualized. A study on contrast-enhanced TRUS demonstrated recently the ability to differentiate prostate cancer from normal tissue. It reported a sensitivity of 100%, but a specificity of only 48% in patients having previous negative biopsies but rising PSA values. While the addition of contrast-enhancement ultrasound (CEUS) is a promising contribution to conventional TRUS biopsy, the value of CEUS is controversial, with experts achieving excellent results but several multicenter trials achieving mediocre results. As with all facets of ultrasound, the method is still highly operator-dependent and results will vary according to skill and experience. More research is warranted before this can be considered a routine option in clinical practice.

Magnetic resonance imaging

mpMRI is now considered to be the most powerful noninvasive diagnostic method for the detection of prostate cancer. While needle biopsy and histopathology remain the gold standard of diagnosis, mpMRI offers a high level of confidence in diagnosing clinically significant cancers. A study of 1003 men by the National Cancer Institute showed the significant improvement in prostate cancer detection with the use of mpMRI followed by targeted MRI-TRUS fusion-guided biopsy.

Prostate MRI can be acquired at 1.5T or 3T, with or without the use of an endorectal coil (ERC). There is an understanding that while the lower field strength of 1.5T is sufficient for evaluation, imaging at 3T is likely to be of higher quality. This is because 3T scanners exhibit a higher signal-to-noise ratio (SNR). However, multiple studies have shown minimal difference in outcomes between prostate MRIs conducted at 1.5T with an ERC versus 3T using only phased-array surface coils. The routine use of the ERC is still debated, particularly at 3T. Newer MRI units are capable of performing excellent prostate MRIs without the ERC, thus reducing cost, time, and patient discomfort. Nonetheless, several studies have determined that there is benefit to using ERC at 3T. Heijmink et al. found mpMRI with ERC to have better accuracy for tumor localization when compared to mpMRI with body coil. Turkbey et al. recently showed increased positive predictive value of 80% versus 64% for mpMRI with and without an ERC, respectively, in 20 patients who underwent radical prostatectomy. However, the clinical consequence of this slightly increased sensitivity with the ERC has not yet been determined. Thus, in general, better-quality scans can be obtained with ERC, and this translates to higher sensitivity and specificity; however, it is unclear whether the use of ERC is justified in terms of better patient outcomes.

Eventually, it is expected that routine use of ERC will no longer be used for the reasons just mentioned. With time and technological advancement in MR technology, mpMRI with ERC may be reserved for very specific purposes – such as in local staging after histopathological diagnosis and especially in the postprostatectomy follow-up after biochemical recurrence.