Management of prostate cancer

A 61-year-old man is referred for discussion of management options for prostate cancer in July 2015. He was diagnosed with prostate cancer in July 2013, and has been on active surveillance since that time.

He presented to his local urologist in May 2013 with a rising serum prostate-specific antigen (PSA) level—most recently 8.36 ng/dL (Table 1). A systematic transrectal ultrasound-guided biopsy in July 2013 demonstrated Gleason 3+3 adenocarcinoma of the prostate in 1 of 12 cores. An Oncotype DX^® (Genomic Health, Redwood City, CA) genomic test from the biopsy tissue demonstrated a Genomic Prostate Score™ (GPS) of 21, predicting a 9% likelihood of dominant Gleason pattern 4 or higher and a 7% risk of non–organ-confined disease. Overall, the GPS predicted very low risk, with an 86% likelihood of favorable pathology. On this basis, the patient elected active surveillance.

Table 1.

Patient Data

Date	PSA Level (ng/mL)	Oncotype DX® GPS	Biopsy/Treatment	Nuclear Medicine Bone Scan	MRI
12/5/06	1.7
11/27/07	2.2
12/30/08	2.6
12/10/09	4.4
11/29/11	6.51
12/5/12	6.29
5/22/13	8.36
7/25/13
8/21/13		GPS 21 (very low risk)
12/6/13	9.1, 5%
3/20/14	8.6
3/28/14			Gleason 3+3 in 1 of 12 biopsy cores
3/10/15	9.9
8/10/15					PI-RADS 5 in bilateral mid anterior, PI-RADS 3 in left lateral apex
9/21/15			MRI-targeted biopsy, Gleason 5+4 cancer in multiple cores
10/12/15				Normal
11/17/15			RARP; pT3a, N0, Gleason 5+4, positive margins
2/18/16	0.05
2/25/16	0.05
6/24/16			Completed 6600 cGy EBRT
10/1/16	<0.01

EBRT, external beam radiation therapy; GPS, Genomic Prostate Score™ (Genomic Health, Redwood City, CA); MRI, magnetic resonance imaging; PI-RADS, Prostate Imaging Reporting and Data System; PSA, prostate-specific antigen; RARP, robot-assisted radical prostatectomy.

His PSA level remained stable over the next year and was 8.6 ng/dL in March 2014. A repeat systematic biopsy in March 2014 demonstrated 1-mm Gleason 3+3 prostate cancer in 1 of 12 cores. The PSA began to rise, reaching a peak level of 9.9 in early 2015 (Table 1). Aside from the diagnosis of prostate cancer, the patient is healthy, with no competing risks for mortality.

He denies erectile dysfunction or bothersome lower urinary tract symptoms, and reports a Sexual Health Inventory for Men score of 25, without the use of phosphodiesterase type 5 inhibitors, and an International Prostate Symptom Score of 3. He exercises regularly and quit smoking 20 years prior. On examination, his prostate is smooth, symmetric, and without nodules or induration. He wishes to determine whether active surveillance remains a reasonable option for him.

Multiparametric prostate magnetic resonance imaging (MRI) on August 8, 2015, demonstrated a 26 × 13-mm anterior mid gland and apex Prostate Imaging Reporting and Data System (PI-RADS) 5 region of suspicion, with extension of tumor beyond the anterior fibromuscular stroma (Figure 1A-C). Additionally, a 7 × 5-mm left posterolateral apex PI-RADS 3 (intermediate suspicion) region is noted, corresponding with the site of previous cancer detection (Figure 1D-F). An MRI-targeted risk stratification biopsy, with MRI to ultrasound coregistration, performed on September 24, 2015, demonstrated Gleason 5+4 cancer within the anterior MRI-targeted regions, Gleason 4+5 cancer within the left apex MRI-targeted region, and Gleason 4+4 cancer within the systematic sample. A bone scan demonstrated no metastatic disease.

Figure 1.

Prostate MRI demonstrating (A-C) a bilateral anterior mid gland very high suspicion, PI-RADS 5, region of suspicion (single arrow) with suggestion of extraprostatic extension at the anterior fibromuscular stroma. The lesion demonstrates a circumscribed moderately hypointense mass on T2-weighted image (A), markedly restricted diffusion on ADC map (B), and early enhancement on dynamic contrast enhancement (C) following the administration of gadolinium. Irregularity of the anterior capsule suggests extraprostatic extension (double arrow). Apical images (D, E) demonstrate extension of the anterior lesion (single arrow), with an intermediate suspicion, PI-RADS 3, abnormality in the left posterior apex (double arrow) on T2-weighted (D), ADC map (E), and dynamic contrast-enhanced (F) images. ADC, apparent diffusion coefficient; MRI, magnetic resonance imaging; PI-RADS, Prostate Imaging Reporting and Data System.

A bilateral nerve-sparing robotic-assisted radical prostatectomy was performed on November 17, 2015. Despite the presence of a large volume of high-grade cancer, nerve sparing was considered feasible because of the very anterior location of the dominant tumor within the prostate, allowing a wide anterior margin without resection of neurovascular bundles (Figure 2A-C). Surgical pathology revealed high-volume Gleason 4+5 cancer, occupying 75% of the gland, with bladder neck invasion. The final stage was pT3a, N0 (9 nodes), Mx, with a positive bladder neck margin. Following surgery, the first postoperative PSA value was 0.05 ng/dL, on February 20, 2016. Salvage intensity-modulated radiotherapy was completed on June 23, 2017, with a total dose of 6600 cGy. In follow-up, the patient’s PSA level declined to <0.01 ng/dL and remained undetectable on April 15, 2017. At last followup, he is continent and does not wear protective pads. He notes occasional leaking during strenuous exercise. He is potent, with erections sufficient for penetrative intercourse, on tadalafil, 5 mg daily.

Figure 2.

The Taneja technique of retrograde neurovascular bundle dissection in robotic-assisted radical prostatectomy. (A) Prior to division of the urethra or the bladder neck, the endopelvic fascia is incised and the prostate is freed from the sidewall to the level of the perirectal fascia. The lateral fascia is incised and reflected off the posterolateral surface of the prostate. The neurovascular bundle is reflected laterally in an extrafascial fashion (double arrow), exposing Denonvilliers’ fascia medial to the nerve bundle (single arrow). (B) Denonvilliers’ fascia is sharply incised, exposing the perirectal fat in the retroprostatic space (arrow). (C) A cottonoid sponge is packed into the defect to tamponade any bleeding along the neurovascular bundle, until the prostate is excised. Image courtesy of A. Mass, MD and SS. Taneja, MD.

Discussion

Active surveillance has become a well-accepted management option for men diagnosed with low-risk disease.^1,2 Previous concerns regarding the use of surveillance in young men have been greatly reduced by the demonstration of extremely low rates of metastatic progression and cancer-specific mortality at 10 and 15 years of follow-up.^3-5 Despite low rates of mortality in general, 50% to 60% of men on surveillance require treatment within 5 to 10 years because of evidence of clinical progression in the form of increasing grade, cancer volume, or serum PSA.³ Whether delay in treatment worsens treatment outcome is not well demonstrated, and this remains controversial.⁵

Candidate selection remains a primary challenge of implementing active surveillance in clinical practice. Among men with low-risk pathology demonstrated on systematic biopsy (men who would have been candidates for surveillance), up to 45% demonstrate adverse pathology when they choose radical prostatectomy.⁶ As such, progression on follow-up biopsy may reflect true disease progression in some cases,⁷ but likely reflects missed occult high-grade and/or locally advanced disease on base-line biopsy in many cases. Because systematic biopsy is such an inaccurate tool for candidate selection, a number of tools have emerged for better selection of low-risk patients prior to surveillance, including risk assessment nomograms, genomic tests, and multiparametric MRI.^8-17 Controversy remains regarding whether or not each deposit of cancer within the prostate is likely to demonstrate genomic expression patterns representative of the overall aggressiveness of the patient’s cancer.¹⁸ In this case, it did not. In this patient, despite the presence of occult high-grade disease, genomic testing suggested a very low risk of unfavorable pathology, demonstrating the imperfection of the testing.

Management

In the Department of Urology at NYU Langone Medical Center (New York, NY), we have utilized multiparametric MRI as a primary risk assessment tool among men referred for consideration of surveillance of low-risk prostate cancer since 2011.^19-22 Our experience,^23-26 along with that of several other groups,^27-32 has demonstrated that upon repeat MRI-targeted and systematic biopsy with MRI-ultrasound coregistration, up to 40% of men are found to have increased risk compared with base-line biopsy, allowing earlier treatment for those who need it, fewer follow-up biopsies among those selected for surveillance, and, in theory, a lower likelihood of progression on surveillance. Prostate MRI has the advantage of both assessing the relative risk of occult high-grade prostate cancer through the suspicion score (Table 2) and determining the location and extent of cancer, allowing directed MRI-targeted biopsy or staging prior to treatment. In this case, based on our experience, the presence of a very high suspicion, PI-RADS 5, abnormality could have allowed for treatment without repeat biopsy, based on the very high rate of treatment following repeat biopsy (Figure 3).

Table 2.

Outcomes of MRI-targeted Biopsy in 252 Men Presenting With a History of Previously Diagnosed Gleason 313 Prostate Cancer, Segregated by MRI Suspicion Score/PI-RADS

mSS/PI-RADS	N	GS 6	GS ≥7	Any Cancer
Overall	252	35.3% (89)	29.8% (75)	65.1% (164)
5	14	14.3% (2)	85.7% (12)	100% (14)
4	65	33.8% (22)	52.3% (34)	86.2% (56)
3	73	46.6% (34)	27.4% (20)	74.0% (54)
2	91	30.7% (28)	9.8% (9)	40.6% (37)
1	9	33.3% (3)	0.0% (0)	33.3% (3)

GS, Gleason score; MRI, magnetic resonance imaging; mSS, MRI suspicion score; PI-RADS, Prostate Imaging Reporting and Data System. Unpublished data from Department of Urology, NYU Langone Medical Center (New York, NY).

Figure 3.

Likelihood of remaining on active surveillance following MRI-targeted biopsy, segregated by MRI suspicion score/PI-RADS in 252 men. MRI, magnetic resonance imaging; PI-RADS, Prostate Imaging Reporting and Data System. Unpublished data from Department of Urology, NYU Langone Medical Center (New York, NY).

Upon diagnosis of high-risk prostate cancer, surveillance was no longer a reasonable option for this patient and treatment was indicated. Patients with high-risk prostate cancer should be counseled that single-modality treatment might not be curative because of the high risk of systemic recurrence.^33-36

Management options for high-risk prostate cancer include radical prostatectomy, with or without adjuvant/salvage radiation, or radiotherapy in combination with 2 years of androgen deprivation therapy. In treating high-risk patients surgically, wide margin control is essential, but margin control can be improved through the consideration of MRI results.

The patient’s pathology demonstrated locally advanced, Gleason 5+4 prostate cancer with positive surgical margins. Two large, randomized studies have demonstrated that the use of adjuvant radiotherapy reduces the risk of recurrence and cancer-specific mortality among men with extraprostatic extension, seminal vesical invasion, and/or positive surgical margins.³⁷ On this basis, the recently published American Urological Association/American Society for Radiation Oncology guidelines suggest that all men with high-risk pathology should be counseled regarding the option of adjuvant radiotherapy.³⁸ There is no direct comparison of adjuvant versus early salvage radiotherapy; therefore, timing of radiation following surgery has become controversial.^39,40 The use of adjuvant radiotherapy does result in a reduction in recovery of potency and continence following surgery; for this reason, the managing physician should consider the relative risk of mortality associated with relapse for the individual patient.⁴¹ Men with Gleason score 8-10 disease and extraprostatic extension are at a high relative risk of mortality following disease relapse; in these men, early adjuvant or salvage radiation therapy should be a consideration.