JU INSIGHT: Outcomes of Active Surveillance of Low-grade Prostate Cancer: A Landmark Approach

By: N. Timilshina, MPH, PhD, University Health Network, Toronto, Ontario, Canada, University of Toronto, Ontario, Canada; S. M. H. Alibhai, MD, University Health Network, Toronto, Ontario, Canada, University of Toronto, Ontario, Canada; G. Tomlinson, PhD, University Health Network, Toronto, Ontario, Canada, University of Toronto, Ontario, Canada; B. Sander, PhD, University Health Network, Toronto, Ontario, Canada, University of Toronto, Ontario, Canada; D. C. Cheung, MD, University Health Network, Toronto, Ontario, Canada, University of Toronto, Ontario, Canada; A. Finelli, MD, MSc, University Health Network, Toronto, Ontario, Canada, University of Toronto, Ontario, Canada | Posted on: 08 Mar 2023

Timilshina N, Alibhai SMH, Tomlinson G, Sander B, Cheung DC, Finelli A. Long-term outcomes following active surveillance of low-grade prostate cancer: a population-based study using a landmark approach. J Urol. 2023;209(3):540-548.

Study Need and Importance

Treatment options for low-risk prostate cancer (PC) commonly include active surveillance (AS), surgery, and radiation therapy, but no high-level evidence has shown any one treatment strategy to be superior to another in low-grade PC. Although reports from single academic institutions on long-term outcomes for AS have been promising, with excellent PC-specific survival of 99%-100% at 10 years, cogent population-level data are lacking. Generally, published studies feature highly selected patients with very low–risk PC, protocol- based follow-up, and median follow-up of roughly 5 years. No study has published long-term oncologic outcomes of AS at a population level.

What We Found

In this retrospective, population-based study, we compared long-term cancer outcomes of 21,282 men with low-grade PC managed with AS to those with initial treatment. We used 2 complementary analytical approaches to reduce bias. Our results (with a median follow-up of 10 years) suggest that AS was associated with slightly worse long-term metastasis-free survival, overall survival, and PC-specific survival compared with initial treatment (see Table).

Limitations

Table. Cox Proportional Survival Model of Metastasis, Overall Mortality, and Prostate Cancer-specific Mortality (Landmark Analysis and Propensity Score Matched Analysis)

Variables	Primary landmark analysis		Propensity score matched analysis
	Hazard ratio (95%CI)	P value	Hazard ratio (95%CI)	P value
*Outcome 1: Metastasis*
Active surveillance vs initial treatment	1.34 (1.15-1.57)	< .001	1.28 (1.08-1.51)	< .001
*Outcome 2: Overall mortality*
Active surveillance vs initial treatment	1.12 (1.01-1.24)	.038	1.12 (1.01-1.25)	.036
*Outcome 3: Prostate cancer–specific mortality*
Active surveillance vs initial treatment	1.66 (1.15-2.39)	.007	1.87 (1.24-2.82)	.003
Abbreviation: CI, confidence interval. Each analysis was adjusted for age, diagnosis year, Johns Hopkins Adjusted Clinical Groups comorbidity score, disease characteristics at diagnosis (PSA, positive core, and maximum % of core), rurality, patient income, and provider-related characteristics (hospital type, physician type, physician and institution volume tertile).

Our study lacks data on some prognostic variables such as cancer staging and PSA that may influence treatment choice and outcome.

Interpretation for Patient Care

In this real-world study of long-term outcomes in men with low-grade PC, AS is associated with excellent long-term metastasis-free survival and overall survival. However, long-term PC-specific survival was slightly inferior to initial treatment (1% worse at 10 years with AS); this must be balanced against known harms of overtreatment. Minor differences in long-term outcomes may be due to less restrictive inclusion criteria, less rigid follow-up, or the effects of initial treatment. With more contemporary use of MRI and transperineal biopsy, the results of AS will also likely improve and reduce any differences with radical treatment.