- Email: [email protected]
PD47-08 VARIATION AND TRENDS IN PROSTATE CANCER CARE AT COMMISSION ON CANCER DESIGNATED FACILITIES
THE JOURNAL OF UROLOGYâ
e898
Vol. 197, No. 4S, Supplement, Sunday, May 14, 2017
85% of younger patients. The vast majority of cT3 patients received radiotherapy in combination with hormonal therapy, regardless of age. However, in the eldest patients (80 years) hormonal monotherapy was applied most frequently (>60%). In the patients with cT4/N+/M+ PC, the use of hormonal therapy as monotherapy increased strongly with increasing age. The 5-year relative survival decreased with increasing age: 92% for the patients aged <70 years, 87% for patients aged 70-80 years, and 68% for patients aged 80 years. The 5-year relative survival of low stage PC appeared to be similar for patients aged <70 years versus 70, whereas the survival of advanced PC (T3) was worse for older patients: 5-year relative survival is 97% versus 91% for the cT3 patients and 53% versus 43% for the cT4/N+/M+, respectively for patients aged <70 years versus 70 years. CONCLUSIONS: Elderly men with PC are more often diagnosed with advanced disease, possibly as a result of patients’ or doctors’ delay. After adjusting for disease stage, older patients have a worse prognosis than younger patients. Further research should elucidate whether elderly PC patients are treated optimally while taking the increased life expectancy and the trade-off between the beneficial effects and adverse events of the treatments into account. Source of Funding: none
PD47-06 LOWER RISK OF PROSTATE CANCER IN ASIAN MEN: FROM LESS SCREENING OR TRUE RACIAL DIFFERENCES? Tom Feng*, Alexis Freedland, Los Angeles, CA; Lauren Howard, Durham, NC; Adriana Vidal, Los Angeles, CA; Daniel Moreira, Chicago, IL; Ramiro Castro-Santamaria, King of Prussia, PA; Gerald Andriole, St. Louis, MO; Stephen Freeland, Los Angeles, CA INTRODUCTION AND OBJECTIVES: Global prostate cancer incidence rates are lower in Asian men than white men. To what degree this relates to less aggressive screening in Asian men or inherent differences by race remains to be determined. Our aim was to determine whether Asian race was associated with lower prostate cancer diagnosis in a study of all men who received prostate-specific antigen (PSA) screening and systematic prostate biopsies independent of PSA levels. METHODS: REDUCE was a 4-year, multicenter, randomized, double-blind, placebo-controlled study that followed biopsy-negative men with protocol-dictated PSA-independent biopsies at 2- and 4years. Eligible men were aged 50-75 years, had serum PSA between 2.5-10 ng/mL, and a prior negative prostate biopsy. We tested the association between race and receipt of prostate biopsy as well as race and prostate cancer diagnosis using multivariable logistic regression. RESULTS: Of 8,122 men in REDUCE, 7,296 were of white or Asian race and had complete data for analysis. Asian men had lower BMI (24.8 vs 26.9, p<0.001) and smaller prostate volumes (34.2 vs 43.4 cc, p<0.001) but were similar in baseline age, PSA, family history of prostate cancer, and digital rectal exam findings compared to white men. There was no difference in rate of receiving a prostate biopsy between Asian and white men (p¼0.634). After adjusting for various clinical and demographic characteristics, Asian men were less likely to be diagnosed with cancer during the 4-year study compared to white men (OR 0.56, p¼0.011). When testing for differences in cancer grade, Asian race was significantly associated with decreased risk of lowgrade cancer compared to white race (OR 0.43, p¼0.016). This risk reduction was also observed for high-grade cancer (OR 0.63, p¼0.290) though the association was not statistically significant. CONCLUSIONS: Among men with a negative pre-study biopsy who all underwent biopsies largely independent of PSA, Asian race was associated with reduced risk of prostate cancer diagnosis. These data suggest less screening among Asian men globally cannot completely explain the lower risk of prostate cancer among Asian men. Further studies are needed to explore the inherent differences attributed by race in prostate cancer diagnosis. Source of Funding: GlaxoSmithKline (GSK).
This
study
was
supported
by
PD47-07 THE EFFECT OF A PREFERENCE SENSITIVE ONLINE DECISION AID ON LOCALIZED PROSTATE CANCER TREATMENT: FIRST RESULTS OF A RANDOMIZED CLUSTER CONTROLLED TRIAL Romy Lamers*, Maarten Cuypers, Tilburg, Netherlands; Marieke de Vries, Nijmegen, Netherlands; Lonneke vd Poll-Franse, Tilburg, Netherlands; Ruud Bosch, Utrecht, Netherlands; Paul Kil, Tilburg, Netherlands INTRODUCTION AND OBJECTIVES: Decision aids (DAs) support shared decision making for the treatment of localized prostate cancer by providing balanced evidence based information, eliciting preferences and by structuring the decision making process. Studies have shown that the use of a DA may increase minimal/non-invasive treatment options. However, the current literature is uncertain about the effect on treatment choice in localized prostate cancer using DAs. Our objective is to study the effect of a preference sensitive web based DA on treatment decision making. METHODS: A randomized cluster controlled trial was performed among 18 hospitals between 2014-2016. In the intervention arm (nine hospitals) the DA was offered following diagnosis (N¼332). Patients in the control arm (another nine hospitals) received care and information as usual (N¼128). After treatment decision-making but before treatment start, patients in both arms received a questionnaire measuring treatment choice, decisional conflict and knowledge. Analyses were performed using t-tests, ANOVA and Pearson correlations. RESULTS: Response rate was 72% (intervention N¼273, control N¼109). No differences were found in PSA and Gleason score between groups. The decision aid led to more patients choosing active surveillance (AS) in comparison with standard information (28% vs. 17%, P¼0.03). In the control arm we found significantly more external beam radiotherapy (EBRTx) (16% vs 8%, P¼0.02) and brachytherapy (BT) ( 33% vs. 19%, P¼0.005). No differences were found in decisional conflict and knowledge. One in five patients indicated to prefer a decision aid in a paper form instead of online. CONCLUSIONS: Patients made different treatment choices after DA use. The online DA supported shared decision making and this may lead to significantly more AS. More radiotherapy was found in the control arm. The decision aid did not lead to more decisional conflict. However, a minority of the patients prefers a paper form of the decision aid. The most optimal DA form needs to be determined. We currently collect post-treatment measures to determine regret and treatment satisfaction levels. Source of Funding: none
PD47-08 VARIATION AND TRENDS IN PROSTATE CANCER CARE AT COMMISSION ON CANCER DESIGNATED FACILITIES € rn Lo € ppenberg*, Akshay Sood, Deepansh Dalela, Patrick Karabon, Bjo Detroit, MI; Jesse Sammon, Portland, ME; Malte Vetterlein, Detroit, MI; Joachim Noldus, Herne, Germany; James Peabody, Detroit, MI; Quoc-Dien Trinh, Boston, MA; Mani Menon, Firas Abdollah, Detroit, MI INTRODUCTION AND OBJECTIVES: Contemporary treatment trends for prostate cancer show increased rates of active surveillance. However, nationwide applicability of these reports is limited. Additionally, the impact of Commission on Cancer facility type on prostate cancer treatment patterns is unknown. METHODS: We used the National Cancer Data Base between 2004 and 2013 to identify men diagnosed with prostate cancer. Our cohort was stratified based on the National Comprehensive Cancer Network prostate cancer risk-classes. Cochran-Armitage tests evaluated temporal trends. Random effects hierarchical logit models assessed treatment variation at Commission on Cancer-facility and institution level.
THE JOURNAL OF UROLOGYâ
Vol. 197, No. 4S, Supplement, Sunday, May 14, 2017
RESULTS: In 825,707 men, utilization of radiation therapy declined and utilization of radical prostatectomy increased for all prostate cancer risk-groups between 2004-2013 (p<0.0001). Observation for low-risk prostate cancer increased from 16.3% in 2004-2005 to 32.0% in 2012-2013 (p<0.0001). Significant treatment variation was observed based on Commission on Cancer-facility type. For all riskgroups, rates of treatment according to facility type ranged from 28.4% to 76.9% for radical prostatectomy, 3.6% to 16.2% for brachytherapy, 13.7% to 28.1% for external beam radiation therapy, 1.3% to 7.3% for androgen deprivation therapy, 4.6% to 19.1% for observation, and 0% to 2.1% for cryotherapy. The highest rates of observation for low-risk disease were observed in academic centers. After adjusting for sociodemographic and facility factors, the highest proportions of treatment variation attributable to the single institution were observed for CT (59%, 95%CI 0.45-0.73) and BT (46%, 95%CI 38-53%), while the lowest proportion of treatment variation was observed for ADT (14%, 95%CI 12-15%), and Observation (15%, 95%CI 14-17%). The results were consistent in the sensitivity analysis and in all NCCN risk-groups. CONCLUSIONS: Regardless of tumor characteristics, significant variations in treatment modality exist among different facility types and institutions. The increased utilization of observation in low-risk prostate cancer is an encouraging finding, which appears to be mainly derived by a decrease in radiotherapy utilization in this risk group.
e899
RESULTS: Over a follow-up period of 9.6 years men with statin (n¼761) exposure had insignificantly lower risk to be diagnosed with PCa ([stat+] hazard ratio (HR) 0.77, 95 % confidence interval (CI) 0.58 to 1.02. Statin users had less low risk PCa compared to non-users (p<0.05) at baseline visit while there was no difference in other PCa risk groups (according to d’Amico risk groups classification) or at follow-up visit. Interestingly, total PSA values were lower in statin users both for baseline (1.5 vs. 1.8 ng/ml, p<0.001) and follow-up-visits (after four years) (1.8 vs. 2.1ng/ml, p<0.001). Overall mortality was higher among statin users compared to non-users ([stat+] HR 1.67, 95% CI 1.36 to 2.04, however the competing risk analysis could demonstrate that PCa incidence was not influenced by overall-mortality. CONCLUSIONS: In our study population we could demonstrate that statins intake did not alter overall PCa risk in a statistically significant manner. However, the finding of persistently lower PSA values in statin users is of potential clinical importance. It suggests that PSA cutoff values should be lowered in statin users otherwise it may introduce potential bias towards delayed PCa detection in this group, especially outside screening setting. On the other hand lower PSA values may suggest a durable protective effect of statins on PCa development. Source of Funding: None
PD47-10 THE RESEARCH IMPLICATIONS OF PSA REGISTRY ERRORS David Guo*, I-Chung Thomas, Harsha Mittakanti, Stanford, CA; Jeremy Shelton, Los Angeles, CA; Danil Makarov, New York, NY; Ted Skolarus, Ann Arbor, MI; Matthew Cooperberg, San Francisco, CA; Geoffrey Sonn, Benjamin Chung, James Brooks, John Leppert, Stanford, CA
Source of Funding: none
PD47-09 INFLUENCE OF STATIN INTAKE ON PSA VALUES, RISK OF PROSTATE CANCER DEVELOPMENT AND SURVIVAL IN A PROSPECTIVE SCREENING TRIAL COHORT (ERSPC AARAU) Maciej Kwiatkowski*, Aarau, Switzerland; Elena Lang, Ashkan Mortezavi, Zurich, Switzerland; Lukas Prause, Stephen Wyler, Rainer Grobholz, Andreas Huber, Aarau, Switzerland; Lukas Manka, Braunschweig, Germany; Tullio Sulser, Zurich, Switzerland; Franz Recker, Aarau, Switzerland; Daniel Eberli, Zurich, Switzerland INTRODUCTION AND OBJECTIVES: Chemoprevention of prostate cancer (PCa) has been extensively investigated in the last decades. So far only 5-alpha-reductase-inhibitors (5-ARI) are supported by clinical evidence to have chemopreventive effect on PCa incidence, hence unclear in terms of prevention of aggressive PCa. Evidence for an effect of statins on PCa is conflicting. The interaction between dyslipidemia and carcinogenesis is still to be established. The aim of the study was to analyse the influence of statins intake on PSA values and PCa development. METHODS: A population-based analysis including 4314 men from the European Randomized Study of Screening for Prostate Cancer (ERSPC) database was conducted. Data about drug intake, age, family history and symptoms was obtained by a self-administered questionnaire. A transrectal ultrasound guided prostate biopsy was performed in men with a PSA-level > 3ng/ml. Tumor stage and grade were registered, incidence and mortality data were obtained through registry linkages. PCa incidence and grade, total PSA value, free-to-total PSA and overall survival were compared between statin users and non-users, respectively.
INTRODUCTION AND OBJECTIVES: Errors in prostate specific antigen (PSA) values included in prostate cancer registries have called into question clinical research studies that rely on this information. We sought to characterize the potential effects of PSA registry errors on clinical research by comparing cohorts based on registry PSA values with those based on laboratory values extracted from an integrated national health care system. METHODS: We defined three example cohorts of men with prostate cancer using data from the VA integrated health care system: those with 00 very low00 (<4.0 ng/mL), 00 low00 (<10.0 ng/mL), and 00 high00 (20-100 ng/mL) PSA values. We compared the composition of each cohort when using the cancer registry versus the electronic health record PSA values. We compared overall survival for each cohort as an example clinical outcome. We fit multivariable proportional hazards models to determine the importance of the PSA source in each cohort. RESULTS: There was significant discordance when using cancer registry versus electronic health record PSA values to identify a cohort of patients with 00 very low PSA00 values. While 7,286 were included in both cohorts, one third (n¼3,515) of the cohort defined using cancer registry PSA values was misclassified and 1,800 additional patients were identified when using electronic health record data. The concordance was highest for patients with 00 low00 PSA values, with 21,860 (98%) of patients identified in both the cancer registry and electronic health record based cohorts. Cancer registry PSA values misclassified 41% (604) of the 00 high00 PSA cohort, and 133 additional patients were identified using electronic health record data. Comparisons of overall survival in the examples cohorts identified a difference in overall survival in the 00 very low00 (log rank P¼0.03), but not the 00 low00 or 00 high00 PSA cohorts. CONCLUSIONS: Patient cohorts based on cancer registry PSA values may have high rate of misclassification, particularly among patients with 00 very low00 or 00 high00 PSA values. In some cases, differences in cohorts resulted in measurable differences in overall survival. Attempts should be made to validate cancer registry PSA data to ensure accurate and reproducible results.
e898
Vol. 197, No. 4S, Supplement, Sunday, May 14, 2017
85% of younger patients. The vast majority of cT3 patients received radiotherapy in combination with hormonal therapy, regardless of age. However, in the eldest patients (80 years) hormonal monotherapy was applied most frequently (>60%). In the patients with cT4/N+/M+ PC, the use of hormonal therapy as monotherapy increased strongly with increasing age. The 5-year relative survival decreased with increasing age: 92% for the patients aged <70 years, 87% for patients aged 70-80 years, and 68% for patients aged 80 years. The 5-year relative survival of low stage PC appeared to be similar for patients aged <70 years versus 70, whereas the survival of advanced PC (T3) was worse for older patients: 5-year relative survival is 97% versus 91% for the cT3 patients and 53% versus 43% for the cT4/N+/M+, respectively for patients aged <70 years versus 70 years. CONCLUSIONS: Elderly men with PC are more often diagnosed with advanced disease, possibly as a result of patients’ or doctors’ delay. After adjusting for disease stage, older patients have a worse prognosis than younger patients. Further research should elucidate whether elderly PC patients are treated optimally while taking the increased life expectancy and the trade-off between the beneficial effects and adverse events of the treatments into account. Source of Funding: none
PD47-06 LOWER RISK OF PROSTATE CANCER IN ASIAN MEN: FROM LESS SCREENING OR TRUE RACIAL DIFFERENCES? Tom Feng*, Alexis Freedland, Los Angeles, CA; Lauren Howard, Durham, NC; Adriana Vidal, Los Angeles, CA; Daniel Moreira, Chicago, IL; Ramiro Castro-Santamaria, King of Prussia, PA; Gerald Andriole, St. Louis, MO; Stephen Freeland, Los Angeles, CA INTRODUCTION AND OBJECTIVES: Global prostate cancer incidence rates are lower in Asian men than white men. To what degree this relates to less aggressive screening in Asian men or inherent differences by race remains to be determined. Our aim was to determine whether Asian race was associated with lower prostate cancer diagnosis in a study of all men who received prostate-specific antigen (PSA) screening and systematic prostate biopsies independent of PSA levels. METHODS: REDUCE was a 4-year, multicenter, randomized, double-blind, placebo-controlled study that followed biopsy-negative men with protocol-dictated PSA-independent biopsies at 2- and 4years. Eligible men were aged 50-75 years, had serum PSA between 2.5-10 ng/mL, and a prior negative prostate biopsy. We tested the association between race and receipt of prostate biopsy as well as race and prostate cancer diagnosis using multivariable logistic regression. RESULTS: Of 8,122 men in REDUCE, 7,296 were of white or Asian race and had complete data for analysis. Asian men had lower BMI (24.8 vs 26.9, p<0.001) and smaller prostate volumes (34.2 vs 43.4 cc, p<0.001) but were similar in baseline age, PSA, family history of prostate cancer, and digital rectal exam findings compared to white men. There was no difference in rate of receiving a prostate biopsy between Asian and white men (p¼0.634). After adjusting for various clinical and demographic characteristics, Asian men were less likely to be diagnosed with cancer during the 4-year study compared to white men (OR 0.56, p¼0.011). When testing for differences in cancer grade, Asian race was significantly associated with decreased risk of lowgrade cancer compared to white race (OR 0.43, p¼0.016). This risk reduction was also observed for high-grade cancer (OR 0.63, p¼0.290) though the association was not statistically significant. CONCLUSIONS: Among men with a negative pre-study biopsy who all underwent biopsies largely independent of PSA, Asian race was associated with reduced risk of prostate cancer diagnosis. These data suggest less screening among Asian men globally cannot completely explain the lower risk of prostate cancer among Asian men. Further studies are needed to explore the inherent differences attributed by race in prostate cancer diagnosis. Source of Funding: GlaxoSmithKline (GSK).
This
study
was
supported
by
PD47-07 THE EFFECT OF A PREFERENCE SENSITIVE ONLINE DECISION AID ON LOCALIZED PROSTATE CANCER TREATMENT: FIRST RESULTS OF A RANDOMIZED CLUSTER CONTROLLED TRIAL Romy Lamers*, Maarten Cuypers, Tilburg, Netherlands; Marieke de Vries, Nijmegen, Netherlands; Lonneke vd Poll-Franse, Tilburg, Netherlands; Ruud Bosch, Utrecht, Netherlands; Paul Kil, Tilburg, Netherlands INTRODUCTION AND OBJECTIVES: Decision aids (DAs) support shared decision making for the treatment of localized prostate cancer by providing balanced evidence based information, eliciting preferences and by structuring the decision making process. Studies have shown that the use of a DA may increase minimal/non-invasive treatment options. However, the current literature is uncertain about the effect on treatment choice in localized prostate cancer using DAs. Our objective is to study the effect of a preference sensitive web based DA on treatment decision making. METHODS: A randomized cluster controlled trial was performed among 18 hospitals between 2014-2016. In the intervention arm (nine hospitals) the DA was offered following diagnosis (N¼332). Patients in the control arm (another nine hospitals) received care and information as usual (N¼128). After treatment decision-making but before treatment start, patients in both arms received a questionnaire measuring treatment choice, decisional conflict and knowledge. Analyses were performed using t-tests, ANOVA and Pearson correlations. RESULTS: Response rate was 72% (intervention N¼273, control N¼109). No differences were found in PSA and Gleason score between groups. The decision aid led to more patients choosing active surveillance (AS) in comparison with standard information (28% vs. 17%, P¼0.03). In the control arm we found significantly more external beam radiotherapy (EBRTx) (16% vs 8%, P¼0.02) and brachytherapy (BT) ( 33% vs. 19%, P¼0.005). No differences were found in decisional conflict and knowledge. One in five patients indicated to prefer a decision aid in a paper form instead of online. CONCLUSIONS: Patients made different treatment choices after DA use. The online DA supported shared decision making and this may lead to significantly more AS. More radiotherapy was found in the control arm. The decision aid did not lead to more decisional conflict. However, a minority of the patients prefers a paper form of the decision aid. The most optimal DA form needs to be determined. We currently collect post-treatment measures to determine regret and treatment satisfaction levels. Source of Funding: none
PD47-08 VARIATION AND TRENDS IN PROSTATE CANCER CARE AT COMMISSION ON CANCER DESIGNATED FACILITIES € rn Lo € ppenberg*, Akshay Sood, Deepansh Dalela, Patrick Karabon, Bjo Detroit, MI; Jesse Sammon, Portland, ME; Malte Vetterlein, Detroit, MI; Joachim Noldus, Herne, Germany; James Peabody, Detroit, MI; Quoc-Dien Trinh, Boston, MA; Mani Menon, Firas Abdollah, Detroit, MI INTRODUCTION AND OBJECTIVES: Contemporary treatment trends for prostate cancer show increased rates of active surveillance. However, nationwide applicability of these reports is limited. Additionally, the impact of Commission on Cancer facility type on prostate cancer treatment patterns is unknown. METHODS: We used the National Cancer Data Base between 2004 and 2013 to identify men diagnosed with prostate cancer. Our cohort was stratified based on the National Comprehensive Cancer Network prostate cancer risk-classes. Cochran-Armitage tests evaluated temporal trends. Random effects hierarchical logit models assessed treatment variation at Commission on Cancer-facility and institution level.
THE JOURNAL OF UROLOGYâ
Vol. 197, No. 4S, Supplement, Sunday, May 14, 2017
RESULTS: In 825,707 men, utilization of radiation therapy declined and utilization of radical prostatectomy increased for all prostate cancer risk-groups between 2004-2013 (p<0.0001). Observation for low-risk prostate cancer increased from 16.3% in 2004-2005 to 32.0% in 2012-2013 (p<0.0001). Significant treatment variation was observed based on Commission on Cancer-facility type. For all riskgroups, rates of treatment according to facility type ranged from 28.4% to 76.9% for radical prostatectomy, 3.6% to 16.2% for brachytherapy, 13.7% to 28.1% for external beam radiation therapy, 1.3% to 7.3% for androgen deprivation therapy, 4.6% to 19.1% for observation, and 0% to 2.1% for cryotherapy. The highest rates of observation for low-risk disease were observed in academic centers. After adjusting for sociodemographic and facility factors, the highest proportions of treatment variation attributable to the single institution were observed for CT (59%, 95%CI 0.45-0.73) and BT (46%, 95%CI 38-53%), while the lowest proportion of treatment variation was observed for ADT (14%, 95%CI 12-15%), and Observation (15%, 95%CI 14-17%). The results were consistent in the sensitivity analysis and in all NCCN risk-groups. CONCLUSIONS: Regardless of tumor characteristics, significant variations in treatment modality exist among different facility types and institutions. The increased utilization of observation in low-risk prostate cancer is an encouraging finding, which appears to be mainly derived by a decrease in radiotherapy utilization in this risk group.
e899
RESULTS: Over a follow-up period of 9.6 years men with statin (n¼761) exposure had insignificantly lower risk to be diagnosed with PCa ([stat+] hazard ratio (HR) 0.77, 95 % confidence interval (CI) 0.58 to 1.02. Statin users had less low risk PCa compared to non-users (p<0.05) at baseline visit while there was no difference in other PCa risk groups (according to d’Amico risk groups classification) or at follow-up visit. Interestingly, total PSA values were lower in statin users both for baseline (1.5 vs. 1.8 ng/ml, p<0.001) and follow-up-visits (after four years) (1.8 vs. 2.1ng/ml, p<0.001). Overall mortality was higher among statin users compared to non-users ([stat+] HR 1.67, 95% CI 1.36 to 2.04, however the competing risk analysis could demonstrate that PCa incidence was not influenced by overall-mortality. CONCLUSIONS: In our study population we could demonstrate that statins intake did not alter overall PCa risk in a statistically significant manner. However, the finding of persistently lower PSA values in statin users is of potential clinical importance. It suggests that PSA cutoff values should be lowered in statin users otherwise it may introduce potential bias towards delayed PCa detection in this group, especially outside screening setting. On the other hand lower PSA values may suggest a durable protective effect of statins on PCa development. Source of Funding: None
PD47-10 THE RESEARCH IMPLICATIONS OF PSA REGISTRY ERRORS David Guo*, I-Chung Thomas, Harsha Mittakanti, Stanford, CA; Jeremy Shelton, Los Angeles, CA; Danil Makarov, New York, NY; Ted Skolarus, Ann Arbor, MI; Matthew Cooperberg, San Francisco, CA; Geoffrey Sonn, Benjamin Chung, James Brooks, John Leppert, Stanford, CA
Source of Funding: none
PD47-09 INFLUENCE OF STATIN INTAKE ON PSA VALUES, RISK OF PROSTATE CANCER DEVELOPMENT AND SURVIVAL IN A PROSPECTIVE SCREENING TRIAL COHORT (ERSPC AARAU) Maciej Kwiatkowski*, Aarau, Switzerland; Elena Lang, Ashkan Mortezavi, Zurich, Switzerland; Lukas Prause, Stephen Wyler, Rainer Grobholz, Andreas Huber, Aarau, Switzerland; Lukas Manka, Braunschweig, Germany; Tullio Sulser, Zurich, Switzerland; Franz Recker, Aarau, Switzerland; Daniel Eberli, Zurich, Switzerland INTRODUCTION AND OBJECTIVES: Chemoprevention of prostate cancer (PCa) has been extensively investigated in the last decades. So far only 5-alpha-reductase-inhibitors (5-ARI) are supported by clinical evidence to have chemopreventive effect on PCa incidence, hence unclear in terms of prevention of aggressive PCa. Evidence for an effect of statins on PCa is conflicting. The interaction between dyslipidemia and carcinogenesis is still to be established. The aim of the study was to analyse the influence of statins intake on PSA values and PCa development. METHODS: A population-based analysis including 4314 men from the European Randomized Study of Screening for Prostate Cancer (ERSPC) database was conducted. Data about drug intake, age, family history and symptoms was obtained by a self-administered questionnaire. A transrectal ultrasound guided prostate biopsy was performed in men with a PSA-level > 3ng/ml. Tumor stage and grade were registered, incidence and mortality data were obtained through registry linkages. PCa incidence and grade, total PSA value, free-to-total PSA and overall survival were compared between statin users and non-users, respectively.
INTRODUCTION AND OBJECTIVES: Errors in prostate specific antigen (PSA) values included in prostate cancer registries have called into question clinical research studies that rely on this information. We sought to characterize the potential effects of PSA registry errors on clinical research by comparing cohorts based on registry PSA values with those based on laboratory values extracted from an integrated national health care system. METHODS: We defined three example cohorts of men with prostate cancer using data from the VA integrated health care system: those with 00 very low00 (<4.0 ng/mL), 00 low00 (<10.0 ng/mL), and 00 high00 (20-100 ng/mL) PSA values. We compared the composition of each cohort when using the cancer registry versus the electronic health record PSA values. We compared overall survival for each cohort as an example clinical outcome. We fit multivariable proportional hazards models to determine the importance of the PSA source in each cohort. RESULTS: There was significant discordance when using cancer registry versus electronic health record PSA values to identify a cohort of patients with 00 very low PSA00 values. While 7,286 were included in both cohorts, one third (n¼3,515) of the cohort defined using cancer registry PSA values was misclassified and 1,800 additional patients were identified when using electronic health record data. The concordance was highest for patients with 00 low00 PSA values, with 21,860 (98%) of patients identified in both the cancer registry and electronic health record based cohorts. Cancer registry PSA values misclassified 41% (604) of the 00 high00 PSA cohort, and 133 additional patients were identified using electronic health record data. Comparisons of overall survival in the examples cohorts identified a difference in overall survival in the 00 very low00 (log rank P¼0.03), but not the 00 low00 or 00 high00 PSA cohorts. CONCLUSIONS: Patient cohorts based on cancer registry PSA values may have high rate of misclassification, particularly among patients with 00 very low00 or 00 high00 PSA values. In some cases, differences in cohorts resulted in measurable differences in overall survival. Attempts should be made to validate cancer registry PSA data to ensure accurate and reproducible results.