Long-term Outcomes in Men with Stable PSA During 5-α Reductase Inhibitor Therapy After Negative Prostate Biopsy: Population-based Study Authors: Eugenio Ventimiglia, Marcus Westerberg, Andri Wilberg Orasson, Mats Ahlberg, David Robinson, Rolf Gedeborg, Pär Stattin, Hans Garmo Published: 2026 ABSTRACT Evidence on long-term outcomes after a negative biopsy, particularly prostate cancer mortality, remains limited, and optimal follow-up strategies are unclear. This nationwide population-based study assessed whether prostate-specific antigen (PSA) decline or stability during 5-α reductase inhibitor (5-ARI) therapy after a negative biopsy identifies men at very low risk of prostate cancer and prostate cancer death. Using Prostate Cancer data Base Sweden (PCBaSe Xtend), we identified men with longitudinal PSA and biopsy data who had at least one negative prostate biopsy between 2007 and 2018 and who initiated 5-ARI therapy with defined exposure and adherence criteria. PSA change during the first 2 yr of treatment was evaluated using PSA velocity and classified as stable or increasing. Follow-up started 2 yr after treatment initiation. Among 1248 men, 901 had stable PSA and 347 had increasing PSA. After 10 yr, the cumulative incidence proportion of prostate cancer was 8.3% (95% confidence interval [CI] = 6.0–12%) in men with stable PSA and 18% (95% CI = 13–24%) in men with increasing PSA; most cancers in men with stable PSA were low grade. Prostate cancer mortality was very low among men with stable PSA (standardized mortality ratio [SMR] = 0.19; 95% CI = 0.04–0.56) and higher in men with increasing PSA (SMR = 1.20; 95% CI = 0.52–2.37. Stable PSA during 5-ARI therapy after a negative biopsy was associated with very low prostate cancer mortality; whether this should influence PSA surveillance intensity requires further study. ADVANCING PRACTICE What does this study add? This nationwide, population-based study provides long-term evidence that prostate-specific antigen (PSA) stability or decline during 5-α-reductase inhibitor (5-ARI) therapy after a negative prostate biopsy is associated with a substantially lower risk of prostate cancer diagnosis and an especially low risk of prostate cancer death. Unlike prior studies with shorter follow-up, this analysis includes prostate cancer mortality as an endpoint. Men with increasing PSA during 5-ARI treatment had consistently higher risks of prostate cancer across Gleason categories. These findings support using PSA trends during 5-ARI therapy to tailor surveillance intensity after a negative biopsy. Clinical Relevance Stable PSA levels during 5-α reductase inhibitor therapy following a negative biopsy indicate a very low long-term risk of prostate cancer mortality. These findings suggest that monitoring PSA trajectory during 5-α reductase inhibitor therapy can help clinicians differentiate low- versus high-risk patients and potentially reduce the intensity of follow-up surveillance. Associate Editor: Roderick C.N. van den Bergh, MD, PhD. Patient Summary Men with a negative prostate biopsy and stable PSA while on finasteride or dutasteride have a very low risk of prostate cancer death. What does this study add? This nationwide, population-based study provides long-term evidence that prostate-specific antigen (PSA) stability or decline during 5-α-reductase inhibitor (5-ARI) therapy after a negative prostate biopsy is associated with a substantially lower risk of prostate cancer diagnosis and an especially low risk of prostate cancer death. Unlike prior studies with shorter follow-up, this analysis includes prostate cancer mortality as an endpoint. Men with increasing PSA during 5-ARI treatment had consistently higher risks of prostate cancer across Gleason categories. These findings support using PSA trends during 5-ARI therapy to tailor surveillance intensity after a negative biopsy. Clinical Relevance Stable PSA levels during 5-α reductase inhibitor therapy following a negative biopsy indicate a very low long-term risk of prostate cancer mortality. These findings suggest that monitoring PSA trajectory during 5-α reductase inhibitor therapy can help clinicians differentiate low- versus high-risk patients and potentially reduce the intensity of follow-up surveillance. Associate Editor: Roderick C.N. van den Bergh, MD, PhD. Patient Summary Men with a negative prostate biopsy and stable PSA while on finasteride or dutasteride have a very low risk of prostate cancer death. Men with stable or declining prostate-specific antigen (PSA) levels have a low risk, whereas an increase is associated with an increased risk of prostate cancer (PCa) [1], [2], [3], [4], [5], [6]. However, these trials have been limited by short follow-up, leaving uncertainty regarding long-term outcomes, and no study has used PCa death as an endpoint. Moreover, the optimal duration of follow-up remains unknown, and current guidelines provide little clarity on the best surveillance strategy [7]. The aim of this study was to assess whether, following a negative biopsy, decreasing PSA during therapy with 5-α-reductase inhibitors (5-ARIs) can identify men at very low risk of a PCa diagnosis and PCa death. The Prostate Cancer data Base Sweden extended treatments and endpoints data (PCBase Xtend) holds data on men with and without PCa from the National Prostate Cancer Register (which captures >98% of PCa diagnoses in the Cancer Register, to which reporting is mandatory) and other registers, such as the Patient Register, Cause of Death Register (which records date and cause of death), and Prescribed Drug Register, by use of the Swedish personal identity number [8], as well as from individual-level results from PSA tests and prostate biopsies collected from regional clinical laboratory information systems. Filled prescriptions for 5-ARI were retrieved from the Prescribed Drug Register using the Anatomical Therapeutic Chemical (ATC) codes for finasteride (G04CB01) and dutasteride (G04CB02). Men with longitudinal PSA and biopsy data who had at least one negative prostate biopsy were identified. To include those initiating 5-ARI therapy soon after biopsy, we selected men with negative biopsies from January 1, 2007, onward, since the Prescribed Drug Register is complete from July 2005. To ensure 2 yr of PSA follow-up after biopsy, inclusion was limited to men with negative biopsies through December 31, 2018. Although no predefined PSA testing schedule existed, the Confederation of Regional Cancer Centers has launched an organized PCa testing program [9]. Men diagnosed with PCa between biopsy and the start of follow-up were excluded (Supplementary Table 1). Life expectancy was calculated based on age, a multidimensional diagnosis-based comorbidity index (MDCI), and a drug comorbidity index (DCI) [10]. During the 2 yr following the negative biopsy, men were considered to be on 5-ARI treatment if they received ≥60 defined daily doses (DDDs) within a 90-day period, with treatment adherence ≥50% of the prescribed DDDs. A PSA measurement not preceded by a 90-day period with ≥60 DDD was considered not to be affected by 5-ARI. PSA measurements were collected before the biopsy, after the biopsy but before 5-ARI initiation, and during treatment. Men were included if a PSA test was recorded within 180 d before biopsy and if their prebiopsy PSA was <50 ng/ml. At least two PSA measurements during 5-ARI treatment were required before follow-up. PSA changes during the 2 yr following 5-ARI initiation were analyzed using PSA velocity, calculated for each man as the slope of a linear model fitted to the log-transformed PSA values and time (d) since the first PSA affected by 5-ARI. From a clinical perspective, a negative velocity can be interpreted as decreasing PSA and was classified as stable PSA, whereas a positive velocity was classified as increasing PSA [11]. Follow-up began 2 yr after 5-ARI start. The cumulative incidence proportion of PCa was estimated for men with stable and increasing PSA, stratified by Gleason score, with death from any cause considered a competing event. The last day of follow-up for this analysis was December 31, 2023. Follow-up was split into age categories (35–39, 40–44, …, 95–99, or ≥100 yr) and calendar periods (2009 until 2023). We created contingency tables with the observed number of PCa cases. Similarly, the accumulated person-time for the corresponding age groups and calendar periods was calculated. The expected number of men diagnosed with PCa was obtained from the corresponding age- and time-specific annual incidence of PCa for the entire Swedish population, and PCa death was assessed in a similar manner [12], [13]. The standardized mortality ratios (SIRs) and standardized mortality ratios (SMRs) were thus defined as the ratio of the observed number of PCa diagnoses and deaths [8] to their expected numbers, that is, using indirect standardization. The 95% confidence intervals [CIs] for the SIRs and SMRs were estimated by assuming that the observed number of cases had a Poisson distribution using Byar’s normal approximation [14]. To take into account differences among detected PCa, SIR calculations were also stratified according to Gleason score. All analyses were performed using R version 4.3.2. We evaluated the following alternative definition of PSA increase and decrease in relation to the risk of receiving a PCa diagnosis and for the estimation of SMRs during follow-up:–PSA decrease >10%, stable PSA (decrease ≤ 10% or increase ≤ 20%), and PSA increase >20% according to the ratio between the last and the first PSA affected by 5-ARI before the start of follow-up –PSA decrease >10%, stable PSA (decrease ≤ 10% or increase ≤ 20%), and PSA increase >20% according to the ratio between the last and the first PSA affected by 5-ARI before the start of follow-up PSA decrease >10%, stable PSA (decrease ≤ 10% or increase ≤ 20%), and PSA increase >20% according to the ratio between the last and the first PSA affected by 5-ARI before the start of follow-up A total of 1248 men had a negative prostate biopsy and fulfilled the criteria for 5-ARI treatment (Supplementary Fig. 1). Median age at follow-up start was 69 yr (interquartile range [IQR] = 65–73). Median prebiopsy PSA was 8.1 ng/ml (IQR = 5.7–12.0), similar to the level immediately before 5-ARI initiation. During the first 2 yr of treatment, the first and last PSA values were 4.3 and 3.1 ng/ml, respectively, in men with stable PSA (n = 901), and 3.7 and 4.6 ng/ml, respectively, in those with increasing PSA. Baseline characteristics were similar between groups (Supplementary Table 2). After 10 yr, the cumulative incidence proportion of PCa was 8.3% (95% CIs = 6.0–12%) in men with stable PSA and 18% (95% CI = 13–24%) in those with increasing PSA (Fig. 1). Most cancers in the stable group were Gleason 6 or 3 + 4 (accounting for 66% of all diagnoses). Median follow-up time was 6.5 yr. The biopsy rate was 0.048 biopsies per person-yr in men with increasing PSA, whereas it was 0.025 biopsies per person-yr in men with stable PSA.Fig. 1Cumulative incidence of prostate cancer stratified by PSA increase in men with a negative prostate biopsy and treated with 5-alpha reductase inhibitors. PSA = prostate-specific antigen. Cumulative incidence of prostate cancer stratified by PSA increase in men with a negative prostate biopsy and treated with 5-alpha reductase inhibitors. PSA = prostate-specific antigen. In men with stable PSA, we observed 43 PCa diagnoses (expected 36.8), corresponding to an SIR of 1.17 (95% CI = 0.84–1.57). SIRs attenuated when analysis was restricted to advanced disease: 0.73 (95% CI = 0.20–1.87) for Gleason scores 9–10. Conversely, men with increasing PSA had consistently elevated SIRs across Gleason categories (Supplementary Table 3). We observed 3 PCa deaths (expected 15.6) in men with stable PSA, corresponding to an SMR of 0.19 (95% CI = 0.04–0.56). We observed 8 PCa deaths (expected 6.6) in men with increasing PSA, corresponding to an SMR of 1.20 (95% CI = 0.52–2.37). PCa mortality was very low among men with stable PSA levels (Supplementary Table 4). Alternative definitions of PSA increase and decrease showed that fluctuating PSA around the initial value (between −10% and +20%) resulted in a similar PCa cumulative incidence proportion compared with men with decreasing PSA: 11% (95% CI = 8−16%; Supplementary Fig. 2), whereas an increase >20% was associated with a higher cumulative incidence proportion of PCa, 24% (95% CI = 15–36%). SMRs for these alternative definitions are provided in Supplementary Table 5, ranging from 0.16 (95% CI = 0.02–0.57) for men with a PSA decrease > 10% to 1.6 (95% CI = 0.43–4.1) for men with a PSA increase > 20%. In men with a negative prostate biopsy who initiated 5-ARI therapy, stable PSA during treatment was associated with a very low risk of PCa diagnosis and mortality. This study is the first population-based analysis of long-term PSA trends during 5-ARI therapy following a negative biopsy and uniquely assesses PCa death as an endpoint. Previous trials have largely focused on short-term PSA changes and cancer detection, typically with ≤4 yr of follow-up [15]. In the ARIA3001/3002 trials, median PSA declined initially in men on dutasteride regardless of later cancer diagnosis [5]. However, an increase in PSA after the first treatment year was observed in men subsequently diagnosed with PCa, while PSA remained low in those without cancer. Among men without a future diagnosis, the median maximum PSA increase from nadir was 0.1 ng/ml versus 0.9 ng/ml in those later diagnosed with PCa [5]. A more recent observational study of men with previous negative biopsies on 5-ARI therapy and repeat biopsies (median = 23 mo apart) reported a PSA increase in 21% of men later diagnosed with PCa versus 13% of those with a second negative biopsy [16]. These findings are consistent with our results, suggesting that PSA rise under 5-ARI treatment predicts higher PCa risk. Stable PSA during 5-ARI therapy following a negative biopsy was associated with very low PCa mortality. We hypothesize that this finding may influence PSA surveillance intensity, although investigation is needed. Integrating PSA kinetics during 5-ARI therapy into follow-up algorithms can guide testing intervals. We hypothesize that men with stable PSA can have longer intervals between tests than men with rising PSA. This would reduce unnecessary biopsies, patient anxiety, and health care costs, ultimately improving the efficiency of PCa detection programs. A major strength of our study is its nationwide, population-based design, which uses high-quality data [17]. PSA testing in our cohort was performed in clinical practice, with large variation in testing frequency and follow-up biopsies. Nevertheless, adherence to guideline recommendations among Swedish urologists is generally high [17]. We also lacked information about the indication for 5-ARI and family history of PCa, as well as information regarding negative biopsies before 2007. Moreover, we acknowledge that men with increased PSA underwent a more intense PCa diagnostic strategy during follow-up, and this may introduce ascertainment bias amplifying differences between the two groups. Despite the virtually complete capture of the population-based registers that we used in this study, we cannot exclude selection bias, potentially limiting the generalizability of the results. Moreover, there is a sticky diagnosis bias for PCa death in the Cause of Death Register, particularly for older men with low-risk PCa at diagnosis, and this could potentially also inflate the risk of PCa death. More recent cohorts are needed to enable comparison with men who underwent biopsy without cancer detection in the MRI era. Stable PSA during 5-ARI therapy after a negative biopsy was associated with very low PCa mortality; whether this should influence PSA surveillance intensity requires further study. Author contributions: Eugenio Ventimiglia had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Ventimiglia, Garmo, Stattin, Gedeborg. Acquisition of data: Garmo, Stattin, Westerberg, Gedeborg. Analysis and interpretation of data: Garmo, Ventimiglia, Gedeborg. Drafting of the manuscript: Ventimiglia, Stattin, Gedeborg. Critical revision of the manuscript for important intellectual content: Stattin, Robinson, Ahlberg, Orasson. Statistical analysis: Garmo, Westerberg. Obtaining funding: Stattin. Administrative, technical, or material support: Stattin. Supervision: Stattin, Robinson, Gedeborg. Other: None. Financial disclosures: Eugenio Ventimiglia certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: Rolf Gedeborg is employed by the Medical Products Agency (MPA) in Sweden. The MPA is a Swedish Government Agency. The views expressed in this article may not represent the views of the MPA. Funding/Support and role of the sponsor: None.