Gaps in Validated Quality Indicator Use Are Evident in Active Surveillance of Low-Risk Prostate Cancer

Image Credit: © SciePro - stock.adobe.com

Varied levels of compliance were observed in terms of using validated quality indicators to determine the quality of active surveillance for patients with low-risk prostate cancer, according to findings from a population-based study published in the Journal of the Comprehensive Cancer Network.

Findings showed that in a population of men with low-risk prostate cancer (n = 33,454), compliance with the 10 quality indicators concerning process ranged from 36.6% to 100%, with 6 being adhered to more than 80% of the time. Additionally, significant variations in terms of outcome indicators were reported between patient groups, including a 10-year metastasis-free survival (MFS) rate of 95.0% for patients aged 65 to 74 years vs 97.5% for those younger than 55 years old. Notably, higher 10-year MFS rates were observed for patients of physicians with at least 6 patients undergoing active surveillance annually compared with those who treated 1 to 2 patients with active surveillance, at 95.8% vs 94.5%, respectively.

Moreover, outcome indicators demonstrated 5-year rates for treatment-free survival (TFS; 50.6%), MFS (98.5%), and prostate cancer-specific survival (99.6%). The 10-year rates for MFS, prostate cancer-specific survival, and overall survival (OS) were 95.5%, 98.4%, and 90.7%, respectively.

Additional findings from the study revealed that, regarding quality indicator 1, nearly all patients (99.9%) were cared for by prostate cancer specialists. Process of care indicators occurred as follows: diagnostic biopsy with at least 8 cores (91.9%), initial active surveillance uptake in low-risk patients (36.6%), active surveillance initiation in patients with low-volume disease (85.1%), prostate cancer specialist follow-up (81.3%), confirmatory biopsy within 1 year (42.6%), repeat biopsy between 2 to 5 years (76.2%), prostate cancer specialist visits prior to switching to definitive treatment (91.5%), biopsy within 6 months prior to definitive treatment (76.2%), and active treatment following an upgrade in Gleason score (88.9%). Further, the process of care indicators demonstrated that bone scans were not performed in any patients.

The use of 8 or more cores in a diagnostic biopsy increased steadily over time; in diagnosis years it was 72.9% from 2002 to 2005, 90.4% from 2006 to 2008, 97.3% in 2009 to 2011, and 98.5% between 2012 to 2014. Rates of confirmatory biopsy within 1 year decreased over the same time periods, at 55.3%, 44.6%, 38.8%, and 36.7%, respectively. For process quality indicators overall, 20% declined, 20% changed minimally, and 60% increased in adherence over time during the study period.

A subgroup of analysis of outcome indicators showed that patients who were less than 55 years old (n = 1335), 55 to 64 years old (n = 5506), 65 to 74 years old (n = 4678), and at least 75 years old (n = 724) experienced 5-year TFS rates of 52.7%, 51.3%, 47.0%, and 45.6%, respectively (P = .001).

The 5-year MFS rates were 99.3%, 98.7%, 98.3%, and 97.3%, respectively (P < .001); the 10-year MFS rates were 97.5%, 96.3%, 95.0%, and 89.3%, respectively (P < .001).

The 5-year prostate cancer-specific survival rates were 99.9%, 99.9%, 99.6%, and 97.4%, respectively (P < .001); the 10-year prostate cancer-specific survival rates were 99.7%, 98.8%, 98.5%, and 92.9%, respectively (P < .001).

Ten-year OS rates were 97.7%, 94.8%, 88.5%, and 67.5%, respectively (P < .001).

In terms of physician average annual active surveillance volume, clinicians who managed 1 to 2, 3 to 5, and 6 or more patients with active surveillance had 5-years TFS rates of 49.9%, 47.4%, and 50.3%, respectively (P = .001).

The 5-year MFS rates were 98.1%, 98.4%, 98.6%, respectively (P = .051); the 10-year MFS rates were 94.5%, 95.3%, and 95.8%,respectively (P = .019); the 5-year prostate cancer-specific survival rates were 99.0%, 99.5%, and 99.8%, respectively (P < .001); the 10-year prostate cancer-specific survival rates were 96.9%, 97.9%, and 98.8%, respectively (P < .001); and the 10-year OS rates were 87.6%, 89.5%, and 91.8%, respectively (P < .001).

The retrospective cohort study included patients with prostate cancer whose disease was managed via active surveillance from 2002 to 2014 in Ontario, Canada. Data was gathered using the Ontario Cancer Registry, which includes information on all patients with newly diagnosed cancer as well as cancer mortality rates.

Investigators determined a total of 20 quality indicators to focus their study on using a comprehensive literature review followed by a modified Delphi methodology. Patients with low-risk (Gleason score ≤ 6) nonmetastatic prostate cancer were then identified to determine the feasibility of the quality indicators in an appropriate dataset. This cohort of patients was then linked to biopsy pathology results for all men with prostate cancer who underwent biopsies or had surgical pathology data in the Ontario Cancer Registry.

Quality indicator 1 was a structure indicator that determined if a patient was managed by a prostate cancer specialist vs another clinician. Process indicators 4 through 7 were concerned with diagnosis and eligibility criteria, and indicators 8 through 13 all occurred during the follow-up phase, such as the frequency of biopsies, specialist visits prior to switching therapy, and initiation of active treatment after upgrade in clinical stage or Gleason score. Outcome indicators were represented in quality indicators 15 through 20, including metrics like 5-year TFS, 5-year MFS, and 10-year OS.

Process indicators 2, 3, and 9 were unable to be measured at the population level. These indicators dealt with initial prostate-specific antigen (PSA) measured at diagnosis, measurement of T stage, and the frequency of PSA testing every 3 to 6 months, respectively.

Patients who did not have retrievable information were excluded from the numerator and denominator of each quality indicator and were classified as missing from that indicator. Time-to-event outcome quality indicators were estimated using Kaplan-Meier estimates of cumulative incidence function.

The goal of the study was “…to apply these structure-, process-, and outcomes-based quality indicators to describe variations in quality of care during active surveillance among patients with low-risk prostate cancer using population-based data.”

The median age of patients included in the study was 65 years (IQR, 59-71) with a median comorbidity ACG score of 10 (IQR, 3-19). The median PSA at diagnosis was 6.2 ng/mL (IQR, 4.7-8.5) among 18,194 evaluable patients. The median number of cores at diagnosis was 11 (IQR, 10-12) and the median number of biopsies during active surveillance was 2 (IQR, 1-6). Most patients were treated by a urologist (79.7%) and 20.2% were seen by a radiation oncologist; patients were cared for in the academic (39.9%), nonacademic (46.1%), or regional cancer center (14.0%), settings.

The ranges during which patients received a diagnosis were from 2002 to 2005 (26.8%), 2006 to 2008 (28.5%), 2009 to 2011 (26.8%), or 2012 to 2014 (17.9%). The average annual active surveillance volume for clinicians who provided the therapy (n = 12,147) was primarily 6 or more (62.5%), although a portion of providers had an annual volume of 1 to 2 (8.8%) or 3 to 5 (28.7%).

“This study established a foundation on which to build benchmarking for quality-of-care assessment and monitor the quality of care for patients receiving active surveillance. Future efforts can focus on quality improvement initiatives and measure appropriateness of care according to active surveillance guidelines. Further research is needed to evaluate the causes of variation in quality of care by hospital type (academic vs community, cancer center vs other), which may also improve outcomes for patients receiving active surveillance,” study authors wrote in conclusion.

Reference

Timilshina N, Finelli A, Tomlinson G, Sander B, Alibhai SMH. Applying quality indicators to examine quality of care during active surveillance in low-risk prostate cancer: a population-based study. J Natl Compr Canc Netw. 2023;21(5):465-472.e9. doi:10.6004/jnccn.2022.7256