POTENTIALLY AVOIDED SURGERIES IN MEN WITH LOCALIZED, LOW-RISK PROSTATE CANCER UNDERGOING ACTIVE SURVEILLANCE: A MODEL-BASED ANALYSIS

INTRODUCTION AND OBJECTIVE: Active surveillance (AS) is emerging as an alternative to immediate treatment for patients with clinically localized, low-risk prostate cancer. Benefits from AS (avoidance of costs and risks of complications and mortality associated with treatment) must be balanced against potential harms (reduced life expectancy from metastatic progression, and worries of living with cancer). This study estimated the number of immediate surgeries avoidable by patients undergoing AS without any sacrifice of life expectancy. METHODS: We developed a Markov model comparing patients with clinically localized, low-risk prostate cancer (T 1-2b N0 M0, PSA <=10 ng/mL, Gleason sum <=6) undergoing AS with those receiving immediate surgery. We modeled patients’ annual progression through 4 health states (no clinical progression; clinical progression; metastatic cancer; dead), and surgery prompted by disease progression or patient choice in those with non-metastatic disease. Published literature, SEER*Stat and US life tables provided inputs. We derived the metastatic disease progression rates to equalize life expectancy under the two strategies (AS vs. immediate surgery), and then estimated the corresponding number of surgeries avoided by patients following AS starting at ages 65, 70 or 75. RESULTS: Simulated life expectancies under AS and immediate surgery were equal when annual rates of progression to metastatic cancer for the AS group were 1.6%, 2.1%, and 2.7% at ages 65, 70, and 75, respectively. With equalized life expectancies, the corresponding proportions of patients undergoing AS who would avoid surgery were 54.4%, 61.4%, and 68.6%. When life expectancy is quality-adjusted, 2-3% more patients in each age group could avoid surgery under AS. These results vary by the annual clinical progression rate and by the proportion of patients electing surgery without evidence of clinical progression. Based on the number of incident cases of localized, low-risk prostate cancer among US men aged 65-75, AS could avoid approximately 8,500-9,000 surgeries annually. CONCLUSIONS: Our simulations show that active surveillance has the potential to substantially reduce the number of patients with localized, low-risk prostate cancer who receive immediate surgery.