Objective: The objective of this study was todirectly assess the likelihood and timing of alphablocker discontinuation in patients receivingcombination therapy with dutasteride orfinasteride plus an alpha blocker.
Methods: A retrospective analysis of the PharMetricsIntegrated Medical and Pharmaceutical Database(Watertown, Mass) was conducted to assessdifferences in alpha blocker discontinuation ratesfor patients initiated on 5-alpha reductase inhibitor(5ARI) therapy. The database is nationally representative,encompassing more than 45 million patientsfrom 85 managed healthcare plans. Male patientsaged >50 years with a diagnosis of enlargedprostate (EP) who were receiving alpha blockertherapy and who began 5ARI treatment (dutasterideor finasteride) between January 1, 1999, andMarch 1, 2005, were included. Patients werestudied for up to 12 months to evaluate the likelihoodand timing of alpha blocker discontinuation.
Results: Overall, 56.7% of the patients remainedon alpha blocker therapy for 6 months. At 1 year,more dutasteride patients had discontinued alphablocker therapy (48.9% remained on alpha blocker)than finasteride patients (58.7% remained on alphablocker). After controlling for background covariates,dutasteride patients were 19.9% more likely todiscontinue alpha blocker therapy over 365 days.
Conclusion: Patients with EP who are taking analpha blocker and 5ARI in combination for urinarysymptom relief discontinue their alpha blocker19.9% earlier when taking dutasteride than whentaking finasteride. The ability to discontinue alphablocker therapy earlier could reduce the costs ofpharmacotherapy while continuing to provide anadequate level of symptom control and diseasemodification, which may result in cost savings tohealthcare plans.
(Am J Manag Care. 2007;13:S17-S22)
Prostate enlargement, also known as benign prostatic hyperplasia(BPH), is currently the fourth most prevalent diagnoseddisorder in men 50 years of age and older.1 The etiology ofthis disease has been linked to the hormone dihydrotestosterone(DHT) which promotes prostate growth.2 An enlarged prostate(EP) can lead to bladder outlet obstruction and bothersome urinarysymptoms (including hesitancy, urgency, frequent urination, nocturia,and inability to completely empty the bladder).3
Physicians often prescribe alpha blocker therapy to men who havebothersome urinary symptoms. Although alpha blockers are effective atproviding prompt symptom improvement, they do not reduce prostatevolume or lower the long-term risk of progression to acute urinaryretention (AUR) or prostate-related surgery.4
5-Alpha reductase inhibitors (5ARIs) have been shown to reducethe size of the prostate and to decrease the rate of progression to AURand prostate-related surgery.5,6 However, patients treated with 5ARIsmay need to wait 3 to 6 months to experience significant symptomimprovement; therefore, there is often a role for both alpha blockersand 5ARIs in men with bothersome symptoms who are also at risk forprogression.7 In fact, the Medical Therapy of Prostatic Symptoms(MTOPS) study found that combination therapy with an alpha blockerand a 5ARI lowered the overall risk of BPH progression better thaneither drug alone,4 suggesting a possible synergistic effect when thedrugs are used together. The American Urological Association recommendsconsideration of combination therapy with a 5ARI and analpha blocker in men with urinary symptoms and a demonstrably EP.8
After several months of successful treatment with combinationtherapy, many clinicians choose to discontinue the use of alpha blockers.A placebo-controlled clinical trial has demonstrated that, in amajority of patients, alpha blockers can be discontinued after 6 monthsof combination therapy without significant loss of symptom control.7
There are currently two 5ARIs available for the treatment of EP,finasteride and dutasteride. Dutasteride blocks both the type-1 andtype-2 5-alpha reductase isozymes, whereas finasteride blocks onlythe type-2 isozyme. As a result, dutasteride producesa greater reduction in serum DHT levels than finasteride,as demonstrated in a 24-week randomizedtrial in which serum DHT was reduced by 95% withdutasteride (0.5 mg/day) versus 71% with finasteride(5 mg/day).9 Dutasteride also has a longerhalf-life than finasteride.5,6 However, there havebeen no randomized head-to-head trials comparingclinical outcomes with the 2 drugs.
A recent study found that a significantly greaterpercentage of dutasteride-treated patients experiencedsymptom improvement within 3 monthscompared with finasteride-treated patients.10 Interms of combination therapy, this may mean thatclinicians could discontinue alpha blocker therapyearlier for patients treated with dutasteride, withoutaffecting the symptom burden of the patient.Although this theoretical premise is plausible, nostudies have directly compared the two 5ARIs withrespect to alpha blocker discontinuation. Theobjective of this retrospective comparative studywas to assess the likelihood and timing of successfulalpha blocker discontinuation in patients receivingcombination therapy with an alpha blocker plusdutasteride or finasteride.
Classification of Diseases, Ninth Revision, Clinical
Current Procedural Terminology
The PharMetrics Integrated Medical andPharmaceutical Database (Watertown, Mass) wasused to assess differences in alpha blocker discontinuationrates for patients initiated on 5ARI therapy.The database is nationally representative, encompassingmore than 45 million patients from 85 managedhealthcare plans. Data include inpatient andoutpatient diagnoses as determined by () codes,11 and proceduresas determined by 4(4) and Healthcare Common ProcedureCoding System (HCPCS) formats, as well as prescriptionrecords. Dates of service and both paidand charged amounts are available for all servicesrendered.
Male patients aged >50 years who began 5ARItreatment (dutasteride or finasteride, excludingdosage forms for male pattern baldness) betweenJanuary 1, 1999, and March 1, 2005, were identified.The index date for each patient was defined as thefirst date of a 5ARI prescription within this period.Identified patients were required to have a diagnosisof prostatic enlargement due to BPH or benign neoplasms,based on codes, and to be continuouslyeligible for medical and pharmacy services for6 months before and at least 12 months after theirindex dates. Patients were also required to have beenreceiving alpha blocker therapy on or before theindex date, and for at least 45 days after the indexdate. Patients were excluded if they had an code for prostate or bladder cancer during the18-month eligibility period; if they were less than30% adherent to 5ARI therapy (measured by calculatinga medication possession ratio over the followupperiod); or if they did not initiate alpha blockertherapy before or on the index date. Table 1 lists allrelevant inclusion and exclusion codes.
Assessment of Alpha Blocker Discontinuation
Patients meeting selection criteria were studiedfor up to 12 months to evaluate the likelihood ofdiscontinuing alpha blocker therapy. Time to alphablocker discontinuation was measured from eachpatient's index date to the last alpha blocker prescription.If the last alpha blocker prescription wasmore than 365 days past index, the patient wasassumed to be on therapy for 365 days.
To assess comorbidities in the study population,the Charlson Comorbidity Index with Dartmouth-Manitoba and Deyo modification was utilized.12,13This index is based on 19 medical conditions, eachassigned a weight ranging from 1 to 6. Possible totalscores range from 0 to 37, with higher numbersrepresenting a greater burden of comorbidity. Inthis study, Charlson Index scores were derived fromdiagnoses in the 6-month period beforeeach patient's index date.
Staging of EP
The Thomson Medstat Disease Staging codingcriteria were used to stage the severity of EP beforethe patient was placed on 5ARI therapy.14 Thismethod is based on electronic screening and identificationof a comprehensive map of diagnosiscodes. The proprietary coding criteria, developedby physicians and medical records professionalsemployed by Thomson Medstat, have been widelyused as a classification system for diagnostic categories—1 of 4 systems selected for disseminationwith the Healthcare Cost and Utilization ProjectNationwide Inpatient Sample. In the present study,each patient initiated on 5ARI therapy was placedinto 1 of 7 disease stages based on the presence ofcodes in the 6-month period before theindex date. The stages and corresponding criteria are presented in Table 2.
Additionally, patients having hematuria (code 599.7) and/or bladder stones (codes 592.0, 592.1, 592.9, and 594.1) in the6-month pre-period were identified, because theseoutcomes may also be indicative of disease severity.
Analysis of Outcomes
The likelihood and timing of alpha blocker discontinuationwere assessed via survival analysistechniques. Time to discontinuation was modeledas a function of age, Charlson Comorbidity Indexscore, Thomson Medstat stage, number of alphablocker prescription days before the index date,number of office visits related to EP, use of urologicalspecialty care, use of uroselective alpha blockers,initiating treatment before 2003, and presence ofhematuria and bladder stones, because all of thesevariables may be relevant severity indices for EP orconfounders for alpha blocker discontinuation.Univariate analyses of frequencies and means wereperformed to describe the demographic characteristicsof the study population. When appropriate, chisquaretests were used to compare differences indichotomous variables and -tests were used to comparedifferences in continuous variables.
A total of 2443 men >50 years of age met thestudy's inclusion criteria; of these, 81.4% (1989patients) had begun alpha blocker treatment beforeinitiating 5ARI therapy, while 18.6% (454patients) had begun both an alpha blocker and a5ARI on the index date. The mean age of the studypopulation was 62.8 years. Approximately 90%(2216) of the men were classified as having stage 1prostate enlargement (ie, without bladder outletobstruction, hydronephrosis, renal failure, sepsis, orshock). The mean Charlson Comorbidity Indexscore was 0.9, indicating a low level of comorbidity.
Dutasteride patients (n = 503) were an averageof 3 years younger (= .0711) than finasteridepatients (n = 1940). Charlson Comorbidity Indexscores, disease severity stage, the proportion ofpatients who initiated alpha blocker therapy beforethe 5ARI, and the mean duration of alpha blockertherapy before initiating the 5ARI were equivalentacross both cohorts. Demographics of the studypopulation are summarized in Table 3.
Alpha Blocker Utilization
Overall, 56.7% of the patients remained onalpha blocker therapy for 6 months; however, morefinasteride than dutasteride patients remained onalpha blocker therapy at 1 year (58.7% vs 48.9%,respectively). The survival curve is shown in theFigure. After controlling for background covariates,dutasteride patients were 19.9% more likely to discontinuealpha blocker therapy over 365 days,meaning that at any time point, dutasteride patientsdiscontinued alpha blocker therapy 19.9% fasterthan finasteride patients. The percentages ofpatients remaining on alpha blocker therapy at specific30-day intervals are presented in Table 4.
The Cox proportional hazards model controlledfor age, Charlson Comorbidity Index score,Thomson Medstat stage, number of alpha blockerprescriptions days before the index date, number ofoffice visits related to EP, use of urological specialtycare, use of uroselective alpha blockers, index year,and presence of hematuria and bladder stones.Results from the Cox regression model for days toalpha blocker discontinuation are presented inTable 5.
The purpose of this study was to assess the likelihoodand timing of alpha blocker discontinuationin patients initiating combination therapy withdutasteride or finasteride plus an alpha blocker. Theresults indicate that the rate of alpha blocker discontinuationwas 19.9% greater among the dutasteridepatients compared with the finasteridepatients. Although it is not possible to determinecausality in a retrospective database assessment, theresults of this study raise the possibility that patientsreceiving dutasteride may have faster symptom improvementthan finasteride patients, allowing themto discontinue alpha blocker therapy earlier. Thispremise is supported by a study by Hagerty et al, inwhich 86% more dutasteride patients had somesymptom improvement within 3 months whencompared with finasteride patients.10 The morerapid symptom improvement may be due to dutasteride'simproved (50% greater) suppression ofserum DHT compared with finasteride.9,15 BecauseDHT is the primary hormone responsible forprostate growth, one may surmise that greater DHTsuppression could lead to faster prostate volumereduction and symptom improvement; however,this hypothesis has yet to be confirmed.
In the present study, discontinuation curves separatesubstantially after 5 to 6 months of treatment.This supports work by Barkin et al, who found thatalpha blocker therapy can be successfully discontinuedin the majority of patients after 6 months ofcombination therapy with dutasteride withoutdiminishing the level of symptom control achievedwith combination therapy.7 In contrast, a study byBaldwin et al found that with finasteride, alphablocker therapy could not be discontinued earlierthan 9 months without affecting symptom control.16 Putting these 2 studies together, the 3-monthdifference suggests that alpha blocker therapy indutasteride patients can be stopped 25% earlier(over 1 year), which supports the 19.9% faster discontinuationrate in the present analysis.
For managed care decision makers, the resultsof this study may have important economic implications.The ability to discontinue alpha blockertherapy earlier could reduce the costs of pharmacotherapywhile continuing to provide an adequatelevel of symptom control and disease modification.Now that finasteride has become generically available,it is important for managed care decision makersto explore the economic differences between thetwo 5ARIs in terms of both 5ARI and concomitantalpha blocker acquisition costs.
There are several limitations with this study.First, patients initiating finasteride therapy before2003 were included in this analysis even thoughdutasteride was not available at this time. Theanalysis controlled for a timing and uroselectivealpha blocker use bias by including a covariate thatrepresented the index date of the patients and thetype of alpha blocker prescribed. Additionally, apost-hoc sensitivity analysis was conducted toensure the robustness of the results, stratifying theanalysis from 2003 forward. The stratification provideda parameter estimate that was similar to thatof the original analysis (hazard ratio, 1.201; 95%confidence interval, 1.002, 1.439; = .048). Also,patients were not required to be 100% adherentwith alpha blocker therapy. Time to alpha blockerdiscontinuation was defined as the time betweeninitiating a 5ARI and the last alpha blocker prescriptionwithin the patient's follow-up. During thistime period, patients could have stopped andrestarted alpha blocker therapy several times inunsuccessful attempts at discontinuation. Althoughpatients were not required to be 100% adherent,this type of pharmacotherapy pattern is reflective ofreal-world scenarios, and may increase the externalvalidity of the findings. Second, the exact reasonsfor alpha blocker discontinuation are not known;discontinuation may be for a result of reasons otherthan symptom control. Third, the results do notquantify the level of symptom control that wasestablished within each cohort. Finally, the data foranalysis in this study were obtained from a managedcare database and may not be generalizable to otherpopulations (eg, patients receiving care throughpublicly funded sources).
Even with these limitations, this study providesevidence that patients receiving combination medicaltherapy for BPH are likely to discontinue alphablocker therapy at an earlier time when takingdutasteride than when taking finasteride.
BPH patients who are taking an alpha blockerand 5ARI in combination for urinary symptomrelief discontinue their alpha blockers 20% earlierwhen taking dutasteride than when taking finasteride,which may result in cost savings to healthcareplans.
Address correspondence to: Michael Naslund, MD, MBA, Professor of Urology, University ofMaryland School of Medicine, 655 West Baltimore Street, Baltimore, MD 21201-1559. E-mail:firstname.lastname@example.org.
Am J Manag Care.
1. Issa MM, Fenter TC, Black L, Grogg AL, Kruep EJ. Anassessment of the diagnosed prevalence of diseases inmen 50 years of age or older. 2006;12(4 suppl):S83-S89.
2. Carson C 3rd, Rittmaster R. The role of dihydrotestosteronein benign prostatic hyperplasia. 2003;61(4 suppl 1):2-7.
3. Roehrborn CG. Benign prostatic hyperplasia: anoverview. 2005;7(suppl 9):S3-S14.
N Engl J Med.
4. McConnell JD, Roehrborn CG, Bautista OM, et al.Thelong-term effect of doxazosin, finasteride, and combinationtherapy on the clinical progression of benignprostatic hyperplasia. 2003;349:2387-2398.
5. Avodart® (dutasteride) [prescribing information].Research Triangle Park, NC: GlaxoSmithKline; May2005. Available at: http://us.gsk.com/products/assets/us_avodart.pdf. Accessed January 8, 2007.
6. Proscar® (finasteride) [prescribing information].Whitehouse Station, NJ: Merck & Co, Inc; January2006. Available at: http://www.merck.com/product/usa/pi_circulars/p/proscar_pi.pdf. Accessed January 8,2007.
7. Barkin J, Guimarães M, Jacobi G, et al. Alpha blockertherapy can be withdrawn in the majority of men followinginitial combination therapy with the dual 5alpha reductase inhibitor dutasteride. 2003;44:461-466.
8. Roehrborn CG, McConnell JD, Barry MJ, et al. Guidelineon the management of benign prostatic hyperplasia(BPH). 2003; updated 2006. American UrologicalAssociation. Available at: http://www.aua2004.org/guidelines/bph.cfm. Accessed January 8, 2007.
J Clin Endocrinol Metab.
9. Clark RV, Hermann DJ, Cunningham GR, et al. Markedsuppression of dihydrotestosterone in men withbenign prostatic hyperplasia by dutasteride, a dual 5alpha reductase inhibitor. 2004;89:2179-2184.
10. Hagerty JA, Ginsberg PC, Harkaway RC. A prospective,comparative study of the onset of symptomaticbenefit of dutasteride versus finasteride in men withbenign prostatic hyperplasia in clinical practice. Posterpresented at: the 2004 Annual Meeting of the AmericanUrological Association; May 8-13, 2004; San Francisco,Calif. Poster 1353.
International Classification of Diseases, Ninth
Revision, Clinical Modification
11. ()Hyattsville,Md: Centers for Disease Control and Prevention; 1979.
12. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A newmethod of classifying prognostic comorbidity in longitudinalstudies: development and validation. 1987;40:373-383.
J Clin Epidemiol.
13. Deyo RA, Cherkin DC, Ciol MA. Adapting a clinicalcomorbidity index for use with administrativedatabases. 1992;45:613-619.
Disease Staging: Coded Criteria.
14. 5th ed. Ann Arbor,Mich: Thomson Medstat; 2003.
15. Roehrborn C, Andriole G, Schalken J, et al.Dutasteride, a novel dual 5a-reductase inhibitor,reduces serum DHT to a greater extent versus finasterideand achieves finasteride maximal reductionin a larger proportion of patients. Poster presentedat: XVIIIth Congress of European Association ofUrology; March 12-March 15, 2003; Madrid, Spain.Poster 635.
16. Baldwin KC, Ginsberg PC, Roehrborn CG, HarkawayRC. Discontinuation of alpha-blockade after initial treatmentwith finasteride and doxazosin in men with lowerurinary tract symptoms and clinical evidence of benignprostatic hyperplasia. 2001;58:203-209.