Active surveillance (AS) is now considered the preferred management option for men with low-risk prostate cancer. The guidelines for all the major organizations relating to prostate cancer recommend a discussion of AS with patients diagnosed on biopsy with low-risk disease.
In Sweden AS has become the dominant management regimen for low-volume disease. By 2014 91% of men with very low-risk cancer were managed with AS, and the figure was 74% for low-risk patients. (Loeb et al. JAMA Oncol Oct 2016) However, recent reporting from 91 practices across the US finds that current practice in this country falls far short of those high numbers. At the 2017 AUA meeting Dr. Cooperberg presented data from the AUA Quality registry showing that only “28.1% of patient with low-risk prostate cancer are being managed with active surveillance or watchful waiting.”
The current literature on this important subject is nearly overwhelming. To engage in a full discussion of this issue in the brief confines of a follow-up office visit with a man recently diagnosed is a substantial challenge for a busy urologist. This Commentary is an effort to succinctly highlight which men are optimal candidates for AS and what outcome they might aspire to… and also to suggests ways in which this regimen can be improved.
The study to most definitively address both these issues was “Intermediate and Longer-Term Outcomes from a Prospective Active Surveillance Program for Favorable-Risk Prostate Cancer, by Tosoian, Epstein, Carter and colleagues from Johns Hopkins University School of Medicine (JH), J Clin Oncol. 2015 Oct 20. This study validates that carefully selected men can experience a median treatment-free survival of 8 1/2 years.
The projected prostate cancer-specific survival for the 1298 men in the trial was 99.9% at both 10 and 15 years; and 99.4% were projected to be metastases-free at 10 and 15. These projections were based on 650 men followed for 5 years and 184 for 10 years. Further follow-up is planned. The authors note that the 15-year data is based on a small number of patients, but feel that at the time of this report in 2015 the 10-year data is valid.
What were the characteristics of the men studied in the Johns Hopkins trial?
The men in the very low-risk group (71% of total) had biopsy Gleason score 6 in a gland normal on DRE (cT1c) with 1 or 2 biopsy cores positive and no core showing greater than 50% cancer. They restricted eligibility to those with a prostate specific antigen density (PSAD) of <0.15. (PSAD is calculated by dividing the PSA by the size of the gland in cubic centimeters based on measurement by ultrasound or MRI.) The median PSA in this group was 4.5 ng/mL, but no PSA threshold was set. The median PSAD was 0.9. The adverse outcome of a PSAD more than 0.15 was established in prior studies from Johns Hopkins.
The second group (29%), designated low-risk included older men with clinical stage <T2a (a gland with a nodule occupying less than 1/2 of one side of the gland), biopsy Gleason score 6, and PSA < 10 ng/ml. A PSA density of more than 0.15 was allowed and 14% of men exceeded that value. The median PSA in this group was 5.6 ng/mL; the median PSAD was 0.18.
Both of these cohorts were optimal candidates for AS. The median number of positive cores in the 926 men with very low-risk was 1 and the maximal involvement of any core with cancer was 1-10%, whereas for the 372 men with low risk disease the median number of cores positive was between 1-2 with 1-20% cancer involvement.
Follow-up consisted of PSA testing at 6 months and yearly biopsies were recommended. Men were reclassified and curative treatment recommended for a rise in PSAD, more than 2 cores positive, and a higher Gleason score.
To what extent was treatment with curative intent recommended during the course of active surveillance?
In the JH study intervention with curative intent occurred in 37% of men by 5 years, 50% by 10 years, 57% by 15 years. The median time to intervention was 3 years. In this study three quarters of those men who did progress based Gleason score upgrading or increased tumor volume did so within 1 – 2 years of the original biopsy. This suggests that intervention was the result of undersampling in the initial biopsy as opposed to progression of an indolent cancer. The upgrade on reclassification occurred to an similar extent in both groups: to Gleason 3+4 in 6%, 10% and 11%, and to Gleason score 4+3 in 3%, 5%, and 5% at 5, 10, and 15 years for both.
Other large scale trials with less restrictive eligibility requirements report long-term outcomes inferior to the JH trial.
The international PRIAS study of AS in low-risk prostate cancer allowed PSAD to be as high as 0.2. The criteria of switching to intervention in this trial is under progressive review since the investigators found that too many men based on the protocol’s switching criteria were being advised toward intervention and would have qualified to remain on AS. The current criteria for intervening in the PRIAS study are finding disease around the prostate on clinical exam (cT3) and an upgrade in the Gleason score, but not the number of positive cores or a fast PSA doubling time (Bokhorst et al., Eur Urol, 2016).
The Klotz-led study from Toronto (JCO, 2014) admitted men with favorable intermediate-risk disease with Gleason score of 3+4. “The 10- and 15-year actuarial cause-specific survival rates were 98.1% and 94.3%, respectively. The allowance of 4 – 10% secondary pattern Gleason 4 (i.e. 3+4) is used in various programs especially for men older of 75 years with favorable intermediate-risk disease who have a limited lifespan.
Inaccurate initial biopsy classification of the Gleason score is the Achilles heel of current AS strategy and likely the cause of early reclassification and recommendation to switch to intervention.
This issue was studied and reported as early as 2012 in “Transrectal ultrasound (TRUS) and TRUS-biopsy accuracy in potential candidates for PRIAS active surveillance protocol,” Lacetera et al., Italian Archives of Andrology. Their conclusion: “The transrectal bioptic mapping in these patients has an insufficient accuracy of staging and correct assessment of a low Gleason score that implies a risk of upstaging and upgrading respectively of 20% and 38%.”
How might this inherent inaccuracy be addressed?
Consensus has developed over recent years that as compared to the conventional “systematic” random 12-core TRUS biopsy, a biopsy targeted into the index lesion as imaged on a multiparametric MRI more accurately identifies significant disease. In this technique the MRI image is digitally superimposed onto the ultrasound screen for biopsy guidance. This scheme has repeatedly been documented to identify with the greatest probability the prostatic lesion of the highest Gleason score. By using this fusion technique once a decision has been made to conduct a biopsy, undergrading can be reduced. However, both techniques may be necessary for optimal diagnosis. Ballentine Carter reported that “11% of high-grade cancers were identified by systematic biopsy with no cancer on targeted biopsy,” (Curr Opin Urol 2015).
A caveat is always appropriate: MRI interpretation is critical and requires an experienced and expert radiologist for optimal results.
Another common method for augmenting staging accuracy is conducting a second “confirmatory” biopsy within the first year of AS.
Genomic analysis in the future may importantly improve risk stratification.
The shortcoming of basing AS eligibility on the standard clinical metrics, and the potential for improvement by performing genomic analysis was addressed by Daniel Lin, M.D. and multiple national collaborators in “Outcome of Active Surveillance for Clinically Localized Prostate Cancer in the Prospective, Multiinstitutional Canary PASS Cohort,” (J Urol 2016). In their study of 905 men with very low-, low-, and intermediate-risk cancer they found, as did so many other studies, that “At a median follow-up of 28 months 24% of participants experienced adverse reclassification based on increased Gleason grade or >34% positive cores.”
Their findings: “Of 103 participants subsequently treated with radical prostatectomy 34% had adverse pathology, defined as primary pattern 4 or 5 (i.e. Gleason score 4+3 or greater) or non-organ confined disease. Their conclusion: “However, clinical factors had only a modest association with disease reclassification, supporting the need for approaches that improve the prediction of that outcome.” Reclassification occurred in nearly similar rates across the risk-stratified groups: 37% of very low-risk, 32% of low-risk, and 40% of intermediate-risk men.
Based on their data the Canary trialists offer a “Canary-edrn active surveillance risk calculator” http://prostate-cancer-risk-calculator.org. (Control+click link to follow) For example a 65 year old man with a PSA of 6.5 ng/mL with 3 of 12 cores positive would have a 7% chance of progression on AS. (Lin et al. Eur Urol. 2015 Dec)
The PASS authors believe that a more biologically based assessment is needed at baseline. The study is collecting an extensive data set of not only clinical factors but also genomic abnormalities that might predict more accurately which men at initial diagnosis harbor the seeds of adverse progression. Dr. Lin informs me that the “calculator is being updated [and] addresses the role of PSA kinetics and the effect on reclassification. The Canary PASS cohort is now over 1500 patient from 10 centers.”
The Prolaris Genomic Classifier:
This biopsy based genomic classifier assesses the probability of aggressive cancer based on a mRNA profile assessing cell cycle progression (CCP). Its application has recently has been reported on low-risk, Gleason 6, cancers, and found informative. The Prolaris platform predicts for the risk of death in 10 years with risk rising from -1 to 5. Application of Prolaris to Gleason 6 cancers was reported at the ASCO 2017 meeting. (Tosoian et al. JCO, Supp May 13, 2017 as MP28-13). The Tosoian study was based on biochemical recurrence post prostatectomy but carries implication for AS selection. The sweet spots on the scale were scores of -1 and 0-1 corresponding to a 5-yr biochemical free survival post prostatectomy of 89.2% and 80.4% for the low-risk group. They conclude that the CCP “would help improve the assessment of candidacy for active surveillance.”
How do men who are reclassified and undergo curative treatment fare?
The overriding goal of AS is to avoid overtreatment. A secondary and related goal is to reduce the rate of reclassification. However, as clearly noted above, reclassification and treatment occurs. What is the outcome of treatment of these men? What risk do they incur in their quest to avoid the known adverse effects of intervention?
This is the obvious question asked by men contemplating AS … and it has been well studied. In the absence of the requisite long-term follow-up studies, this issue has been addressed by comparing pathology findings of matched cohorts of men who were candidates eligible for AS but chose initial surgery to men operated upon after progressing on AS.
Peter Carroll and colleagues at UCSF concluded in “Immediate Versus Delayed Radical Prostatectomy: Updated Outcomes Following Active Surveillance of Prostate Cancer” (Eur Urol June 2015). “The rate of adverse pathology did not differ between immediate and delayed RP in men matched for pretreatment characteristics.”
Matoso, Carter, Epstein et al. arrived at a similar conclusion in “Radical Prostatectomy Findings in Men on Active Surveillance” (J Urol. 2015 Sept): “The majority of men with biopsy progression still had tumors with features of curable disease..”
Another analysis, “Pathologic Outcomes in Favorable-risk Prostate Cancer: Comparative Analysis of Men Electing Active Surveillance and Immediate Surgery,” (Tosoian et al. from Johns Hopkins, Eur Urol. 2016 Apr) found that adverse pathology was similar in matched cohorts. They concluded: “Our finding suggest that men who are closely followed with surveillance may have similar outcomes to men who elect immediate surgery, but additional research is needed.”
The rationale and safety of active surveillance in carefully selected men with very low-risk and low-risk prostate cancer has been validated. Analyses of men with similar characteristics who were treated after progressing on AS have shown similar outcomes to those who were treated at diagnosis. Active surveillance as a management strategy merits greater participation in the US.