Jr010100104p

Copyright 2001 by AMERICAN UROLOGICAL ASSOCIATION, INC.® MEDICAL VERSUS SURGICAL ANDROGEN SUPRESSION THERAPY FOR PROSTATE CANCER: A 10-YEAR LONGITUDINAL COST STUDY ALBERT J. MARIANI, MONTY GLOVER AND SUZETTE ARITA From the Department of Urology, John A. Burns School of Medicine, University of Hawaii and Kaiser Medical Center, Honolulu, Hawaii Purpose: We provide a relative cost comparison of medical versus surgical androgen suppres- Materials and Methods: Comparison is based on a cohort of 96 patients who began androgen suppressive therapy for prostate cancer between 1988 and 1990. Patients were followed untildeath or the end point of study in June 2000 at which time 15% were alive. Current Medicareorchiectomy reimbursements were compared to 1999 wholesale drug costs.
Results: For an individual patient the cost of luteinizing hormone releasing hormone (LH-RH) agonist treatment surpassed the cost of surgery at less than 4.2 to 5.3 months, and for combinedandrogen blockade (LH-RH agonists and nonsteroidal antiandrogens) at less than 2.7 to 3.4months. For 5 (5.2%) patients on combined androgen blockade and 6 (6.3%) on LH-RH agonistsalone, medical therapy would have had a cost advantage over bilateral orchiectomy. For theandrogen suppression cohort the cost of LH-RH agonist treatment was 10.7 to 13.5 times andcombined androgen blockade was 17.3 to 20.9 times the cost of bilateral orchiectomy. Urologyresource use comparisons are provided. These findings significantly underestimate the costadvantage of surgery. A seventh of the patients were alive at study end point, and prostatespecific antigen induced stage shifting and changes in practice patterns resulted in earlier andmore frequent androgen suppressive treatment.
Conclusions: Except for patients with short anticipated survivals current medical androgen suppressive treatment options are more costly than bilateral orchiectomy. There is a need for acost comparable medical option to orchiectomy.
KEY WORDS: prostatic neoplasms, gonadorelin, costs and cost analysis, orchiectomy, androgen antagonists Prostate cancer is the second leading cause of cancer mor- tality in men. In 1999 an estimated 179,300 new cases of and By 1987 prostate specific antigen (PSA) was routinely per- 37,000 deaths from disease were predicted.1 In 1994 the total formed in patients enrolled in the Kaiser Foundation Health Medicare expenditure for treatment of prostate cancer was Plan of Hawaii. Between 1988 and 1990 from an average $1,411,687,900, and of this amount greater than a third was population of 381 patients with newly diagnosed or living for luteinizing hormone releasing hormone (LH-RH) agonists with biopsy proved prostate cancer, 96 began androgen sup- alone.2 LH-RH agonists, diethylstilbestrol and bilateral or- pressive therapy. These patients included all those with pros- chiectomy are equally effective against advanced prostate tate cancer from a closed panel, prepaid, nonprofit health cancer. Combined androgen blockade remains controversial, plan population averaging 170,000 patients. This population and its benefit over androgen suppressive monotherapy, if was a sixth and representative of the population of Hawaii.
any, is of short duration and questionable clinical signifi- Androgen suppressive therapy was initiated in 52 (54%) of these patients because of newly diagnosed metastatic dis- American society, with a combination of market forces and ease, in 18 (19%) for new onset metastatic disease, in 19 government intervention, has limited the percentage of na- (20%) as primary therapy for localized disease usually in tional resources that will be spent on health care.4 This “fixedbudget” has led to budget concepts, such as capitated care elderly patients, in 6 (6%) for local progression after defini- and global disease budgets. In this economic environment in tive treatment failed and in 1 (1%) for increasing PSA. All which 2 therapies are equally effective it is rational for the patients were followed until death or the end point of study in clinician to choose that which costs less, thus conserving resources for other treatment. It is also reasonable to restrict Therapy consisted of bilateral orchiectomy in 68 (70.8%), investigational treatments only to clinical studies in which leuprolide in 15 (15.6%) and diethylstilbestrol (DES) in 13 information is collected to direct future care. The magnitude (13.6%) of 96 patients. Of the 13 patients on diethylstilbestrol of the cost difference would increase the importance of these 4 (31%) were converted to bilateral orchiectomy, while 6 choices. We provide the clinician with a relative cost compar- (46%) were converted to leuprolide. Of the 15 patients on ison of medical versus surgical androgen suppressive therapy leuprolide 1 (7%) was converted to bilateral orchiectomy. No patient was converted to receive diethylstilbestrol.
Economic assumptions. Surgical treatment consisted of Accepted for publication July 28, 2000.
outpatient bilateral orchiectomy with the patient under gen- Presented at annual meeting of Western Section, American Uro- eral or spinal anesthesia. The surgical cost per case used was logical Association, Salt Lake City, Utah, September 14 –18, 1997.
$2,479, which was the total 1999 Medicare Part A allowable Editor’s Note: This article is the fourth of 5 published in this
facility charge of $1,869, which was the average of the last 8 issue for which category 1 CME credits can be earned. In-
bilateral orchiectomy Part A reimbursements and the Medi- structions for obtaining credits are given with the questions
on pages 230 and 231.
care Part B allowable professional charge of $610.
MEDICAL VERSUS SURGICAL ANDROGEN SUPPRESSIVE THERAPY COST Drug costs for LH-RH agonists and nonsteroidal antian- would have continued to accumulate costs if on medical ther- drogens were based on the 1999 Drug Topics Red Book apy. For 1 mg. diethylstilbestrol and 81 mg. acetylsalicylic Wholesale Prices (table 1). For LH-RH agonists an additional acid prophylaxis the cost was a tenth the cost of bilateral charge for a nursing service for injection every 3 months was orchiectomy if no patients received prophylactic breast irra- added using the 99201 code reimbursed by Medicare Part B diation, and greater than 1.4 ϫ the cost if all received pro- at $25. Breast irradiation cost was determined by totaling phylactic breast irradiation (table 2).
the local reimbursement for complex treatment plan (77263), Incidence and prevalence. The prevalence of prostate can- complex simulation (77290), complex isodose (77310), dosim- cer cases in the Kaiser Health Plan of Hawaii population etry (77300 ϫ 2), use of treatment device (77430) and weekly increased 215% from 2.24 in 1988 to 1990 to 4.82/1,000 in treatment with 3 fractions (77430), which totaled $2,224 per 1999. During the same interval the incidence of patients case. Past and current incidence data were obtained by re- started on hormonal suppressive therapy increased 29% from viewing membership data and all cases in the American 18.8 to 24.3/100,000 compared to the 1988 through 1990 Cancer Society certified internal tumor registry, which series of those started on androgen suppressive therapy. A tracks all patients with cancer until death whether or not modern series would have greater than twice as many pa- tients being followed with prostate cancer at risk for progres- Economic data. Data regarding staff budgets and phar- sion with a third more being started on hormonal therapy macy costs were obtained from computerized cost manage- with proportionally greater resource impact for the health ment information center reports and then cross-checked by hand against actual budgets and expenditures. Economic Evidence associating increasing PSA with early recurrence data reflect actual costs rather than charges. Personnel costs of prostate cancer has made early hormonal suppression included salaries and benefits. No attempt was made to fac- more popular.6 Proportionally fewer patients began hor- tor in costs of therapy complications despite limited evidence monal suppression for newly diagnosed metastases while that favors monotherapy over combined androgen block- more are being treated on the basis of increasing PSA after ade.3, 5 No attempt was made to factor in costs of followup, definitive treatment (table 3). The widespread use of PSA which were assumed to be similar for patients with advanced surveillance has led to stage shifting in this population. Of prostate cancer treated medically or surgically. No attempt 213 patients newly diagnosed with prostate cancer from 1988 was made to provide a rate of time discount to bring costs to to 1990, 53 (24.8%) presented with bone metastases. Of 117 present value terms because the cost of pharmaceuticals and patients presenting with prostate cancer in 1996 only 9 surgery has not followed stable economic models during the (7.6%) presented with bone metastases.
last decade.4 Finally, no attempt was made to apply the cost Current androgen suppression use. Of 1,040 patients alive of compliance monitoring for patients on medical androgen with prostate cancer as of December 1, 1999, 111 (10.7%) ablation therapy because assigning it would have been arbi- were being treated with hormonal suppressive therapy. Of trary. This assumption provided a cost advantage to medical these patients 76 (69%) had undergone bilateral orchiectomy, therapy as compliance is not an issue for patients treated 27 (24%) were being treated with LH-RH agonists, 6 (5%) were being treated with combined androgen blockade and 2(2%) continued to receive diethylstilbestrol. This group of patients must be distinguished from the 96 in the 1988through 1990 cohort, of whom 14 were alive and included in Cost comparison. For an individual patient the break-even the current group of 111 who received androgen suppressive cost ranged from less than 4.2 to 5.3 months, and less than 2.7 to 3.4 months for combined androgen blockade. For 5 Urology resource use comparison. The 2 studies that of- (5.2%) patients combined androgen blockade would have had fered orchiectomy versus medical therapy demonstrated a a cost advantage over surgery, while for 6 (6.3%) LH-RH patient preference of 70% for medical therapy.7 If 70% (76) of agonists would have had a cost advantage over bilateral the 111 patients receiving hormonal suppressive therapy orchiectomy. For 1 mg. diethylstilbestrol orally every day were treated with LH-RH agonists or combined androgen plus 81 mg. acetylsalicylic acid orally every day the break- suppression, the cost would represent 312% to 620% of the even cost was 45.3 years if no patients received prophylactic remaining urology outpatient pharmacy, 126% to 250% of the breast irradiation and 4.65 years if all received prophylactic urology department outpatient support staff and 22% to 44% radiation. Thus, for the individual patient expected to live less than these intervals, medical hormonal suppressionwould have a cost advantage over bilateral orchiectomy.
For the cohort of patients with prostate cancer treated with androgen ablation therapy between 1988 and 1990 the pop- Except for the individual patient with a short (2 to ulation cost of LH-RH agonists ranged from greater than 10.7 6-month) anticipated survival, bilateral orchiectomy had a to 13.5 times the cost of bilateral orchiectomy. For combined major cost advantage over any LH-RH agonist or combined androgen blockade the cost ranged from greater than 17.3 to androgen suppression. Only 5% to 6% of the prostate cancer 20.9 times the cost of bilateral orchiectomy. The cost indexes cases treated with androgen suppression in this series would are expressed as greater than ϫ to account for the 14.67% of have been in this category. Based on a cohort of patients with patients who survived until the end point of study and who prostate cancer who began hormonal therapy 9.5 to 11.5 22.5 Mg. leuprolide intramuscularly every 3 mos.
10.8 Mg. goserelin intramuscularly every 3 mos.
1 Mg. diethylstilbestrol orally every day* 1999 Drug Topics Red Book. Montvale, New Jersey: Medical Economics, 1999.
* Last entry in 1997 Red Book when Eli Lilly discontinued manufacture of diethylstilbestrol.
MEDICAL VERSUS SURGICAL ANDROGEN SUPPRESSIVE THERAPY COST TABLE 2. Relative cost of therapeutic options 22.5 Mg. leuprolide intramuscularly every 3 mos.
10.8 Mg. goserelin intramuscularly every 3 mos.
22.5 Mg. leuprolide intramuscularly every 3 mos. ϩ 250 mg. flutamide 3ϫ daily orally 22.5 Mg. leuprolide intramuscularly every 3 mos. ϩ 50 mg. bicalutamide orally every day 22.5 Mg. leuprolide intramuscularly every 3 mos. ϩ 150 mg. nilutamide orally every day 10.8 Mg. goserelin intramuscularly every 3 mos. ϩ 250 mg. flutamide 3ϫ daily orally 10.8 Mg. goserelin intramuscularly every 3 mos. ϩ 50 mg. bicalutamide 3ϫ daily orally 10.8 Mg. goserelin intramuscularly every 3 mos. ϩ 150 mg. nilutamide 3ϫ daily orally 1 Mg. diethylstilbestrol orally every day, 81 mg. aspirin orally every day ϩ 100% 900 cGy.
1 Mg. diethylstilbestrol orally every day ϩ 81 mg. aspirin orally every day, no prophylac- * Relative cost of 1 androgen suppressive therapy against another for this population of patients with prostate cancer who began androgen suppressive therapy.
† The length of treatment at which the cost of medical treatment exceeds the cost of surgical treatment (for patients expected to survive for less than this period medical treatment is the lower cost alternative).
‡ Prophylactic breast irradiation not included in cost as it would not be appropriate for short-term treatment.
§ Cardiovascular toxicity safety relative to other medical treatment options not proved.
TABLE 3. Changing patterns of care search demonstrating the safe and effective use of diethyl-stilbestrol in select low risk populations with effective anti- coagulation or parenteral depot preparations that bypassliver metabolism would provide a low cost medical alterna- tive to orchiectomy. To our knowledge such an alternative is While it is estimated that 70% would choose the medical option in patient preference studies comparing medical ver-sus surgical hormonal ablation,7 preference may not reflectoutcome. In a recent quality of life study of asymptomatic years ago, the cost of LH-RH agonist or combined androgen men with prostate cancer, except for sexual function, pa- blockade would exceed the cost of bilateral orchiectomy by tients who underwent orchiectomy had quality of life indexes greater than 10 to 20 times using modern androgen suppres- similar to those who had no treatment or local therapy.5 sive therapy costs. This comparison seriously understates the Patients treated with LH-RH agonist or combined androgen cost advantage of bilateral orchiectomy over LH-RH agonists blockade had significantly lower quality of life scores. While or combined androgen blockade. At the end point of study a orchiectomy may be preferred over LH-RH agonist or com- seventh of the patients were alive and would have continued bined androgen blockade medical therapy, there are wide to accrue medical therapy costs until death. In our population differences in individual perception of castration.5 For pa- the incidence of patients being treated hormonally for pros- tients who have difficulty accepting orchiectomy a subcapsu- tate cancer increased 29% from 1988 through 1990 to 1996.
lar orchiectomy could be offered to alleviate body image prob- While there continues to be a vigorous controversy surround- lems.18 The subcapsular orchiectomy has been found to be as ing early versus late hormonal treatment of prostate cancer, effective as standard orchiectomy for androgen suppres- during the last decade the association of increasing PSA with sion.19 Intermittent hormonal therapy appears promising subclinical disease progression has resulted in changing but the issue of effectiveness relative to continuous androgen practice patterns toward earlier hormonal treatment.6 In suppression remains unresolved.20 Should the effectiveness 1996 a third of our patients began hormonal treatment for and quality of life benefit be proved, patients who have un- asymptomatic elevations of PSA alone.
dergone standard or subcapsular orchiectomy could be ad- For greater than 94% of our patients with prostate cancer ministered intermittent androgen supplementation.
treated hormonally, there is currently no accepted medical We are unaware of any validated assessment instruments therapy with a cost comparable to bilateral orchiectomy.
that would identify patients who are not candidates for or- Before the availability of LH-RH agonists and nonsteroidal chiectomy for psychological reasons. At our institution the androgens diethylstilbestrol was the most commonly used decision was made that preference alone was not a sufficient medical hormonal treatment for prostate cancer. Availability reason to administer a therapeutically equivalent modality of clinical alternatives and increasing awareness of the car- that costs more than orchiectomy. Thus, a cost sharing pro- diovascular toxicity of diethylstilbestrol have resulted in its gram was instituted in which patients electing a LH-RH fall from favor. Diethylstilbestrol had several advantages, agonist paid 20% of the cost of the pharmaceutical unless including its low cost, patients could take it once a day and they were exempted as a poor surgical risk for medical rea- not only did it inhibit LH-RH production by the hypothala- sons. This standard was liberally applied. Thus, while 70% of mus but there was evidence of a reduction in adrenal andro- patients who received androgen suppression might have been gen production through hepatic mechanisms8 and direct cy- expected to prefer LH-RH agonists or combined androgen totoxic effects in prostate cancer animal models.9 Unlike blockade, when cost sharing was factored in only 24% elected LH-RH agonists or orchiectomy, which promote osteoporo- this option. An additional 5% of patients were receiving sis, diethylstilbestrol had a protective effect.10, 11 Unfortu- LH-RH agonists because they were classified as poor medical nately, diethylstilbestrol was also associated with excess car- diovascular mortality most likely due to excess hepatic Whether use of LH-RH agonists or combined androgen production of coagulation factors.12 Attempts at lowering this blockade based solely on patient preference for 7.5% of this mortality by lowering the dose13 or combining it with antico- prostate cancer population is worth the equivalent of approx- agulants were unsuccessful.14 In randomized controlled stud- imately a third of the cost of maintaining a staff of urological ies patients treated with diethylstilbestrol have less prostate surgeons, twice the cost of maintaining urology support clinic cancer mortality but more cardiovascular mortality than staff or 3 to 6 times the cost of all other outpatient pharma- those treated with orchiectomy or LH-RH agonist.3, 15–17 Re- ceuticals ordered by the urology department is a complex MEDICAL VERSUS SURGICAL ANDROGEN SUPPRESSIVE THERAPY COST issue. If society is willing to allocate a larger portion of its 7. Iversen, P., Tyrrell, C. J., Kaisary, A. V. et al: Casodex (bicalu- production to medical care there is more flexibility. In the tamide) 150-mg. monotherapy compared with castration in United States the trend is toward a smaller portion of pro- patients with previously untreated nonmetastatic prostate duction being devoted to medical care.2 In an insurance en- cancer: results from two multicenter randomized trials at a vironment in which there is such a fixed budget, expendi- median follow-up of 4 years. Urology, 51: 389, 1998
tures in 1 area result in reductions elsewhere. Thus, it is 8. Poussette, A., Carlstro¨m, K. and Stege, R.: Androgens during reasonable for clinicians, when faced with nearly equivalent different modes of endocrine treatment of prostatic cancer.
Urol Res, 17: 95, 1989
treatment options, to choose that which costs the least. They 9. Landstrom, M., Damber, J. E. and Bergh, A.: Estrogen treat- should not treat patients with expensive investigational pro- ment postpones the castration-induced dedifferentiation of tocols unless data are being collected to resolve a clinical Dunning R3327-PAP prostatic adenocarcinoma. Prostate, 25:
issue. This practice is strictly true only in an economic envi- ronment in which conserved resources are returned to med- 10. Daniell, H. W., Dunn, S. R., Ferguson, D. W. et al: Progressive osteoporosis during androgen deprivation therapy for prostate It has also been argued that the clinician should focus cancer. J Urol, 163: 181, 2000
exclusively on the individual patient and ignore cost consid- 11. Eriksson, S., Eriksson, A., Stege, R. et al: Bone mineral density erations.21 In the current medical economic environment in patients with prostatic cancer treated with orchiectomy and such an approach does not appear realistic. Resource alloca- with estrogens. Calcif Tissue Int, 57: 97, 1995
tion decisions will be made. In our opinion if clinicians abdi- 12. Cox, R. L. and Crawford, E. D.: Estrogens in the treatment of cate their role in resource allocation decision making, these prostate cancer. J Urol, 154: 1991, 1995
decisions will be made by default by relatively uninformed 13. Robinson, M. R., Smith, P. H., Richards, B. et al: The final and often crude market, medical industrial and political analysis of the EORTC Genito-Urinary Tract Cancer Co- Operative Group phase III clinical trial (protocol 30805) com-paring orchidectomy, orchidectomy plus cyproterone acetate Carol Tom and Barbara Kashiwabara provided assistance and low dose stilboestrol in the management of metastatic with the tumor registry and pharmaceutical services.
carcinoma of the prostate. Eur Urol, 28: 273, 1995
14. Klotz, L., McNeill, I. and Fleshner, N.: A phase 1-2 trial of diethylstilbestrol plus low dose warfarin in advanced prostate
carcinoma. J Urol, 161: 169, 1999
1. Landis, S. H., Murray, T., Bolden, S. et al: Cancer statistics, 15. Carcinoma of the prostate: treatment comparisons. The Veter- 1999. CA Cancer J Clin, 49: 8, 1999
ans Administration Cooperative Urological Research Group.
2. Holtgrewe, H. L., Bay-Nielsen, H., Bouffioux, C. et al: The eco- J Urol, 98: 516, 1967
nomics of prostate cancer. In: First International Consultation 16. Johansson, J.-E., Andersson, S.-O., Holmberg, L. et al: Primary on Prostate Cancer. Edited by G. Murphy, K. Griffiths, L.
orchiectomy versus estrogen therapy in advanced prostatic Denis et al. London: Scientific Communication International, cancer—a randomized study: results after 7 to 10 years of followup. J Urol, 145: 519, 1991
3. Agency for Health Policy and Research: Relative effectiveness 17. Leuprolide versus diethylstilbestrol for metastatic prostate can- and cost effectiveness of methods of androgen suppression in cer. The Leuprolide Study Group. N Engl J Med, 311: 1281,
the treatment of advanced prostatic cancer. Summary, evi-dence report/technology assessment: No. 4. Rockville, Mary- 18. Riba, L. W.: Subcapsular castration for carcinoma of the pros- 4. Levit, K., Cowan, C., Lazenby, H. et al: Health spending in 1998: tate. J Urol, 48: 384, 1942
signals of change. The Health Accounts Team. Health Aff 19. Bergman, B., Damber, J. E. and Tomic, R.: Effects of total and (Millwood), 19: 124, 2000
subcapsular orchidectomy on serum concentrations of testos- 5. Herr, H. W. and O’Sullivan, M.: Quality of life of asymptomatic terone and pituitary hormones in patients with carcinoma of men with nonmetastatic prostate cancer on androgen depriva- the prostate. Urol Int, 37: 139, 1982
tion therapy. J Urol, 163: 1743, 2000
20. Gleave, M., Bruchovsky, N., Goldenberg, S. L. et al: Intermittent 6. Wasson, J. H., Fowler, F. J., Jr. and Barry, M. J.: Androgen androgen suppression for prostate cancer: rationale and clin- deprivation therapy for asymptomatic advanced prostate can- ical experience. Eur Urol, suppl., 34: 37, 1998
cer in the prostate specific antigen era: a national survey of 21. Kassirer, J. P.: Managing care—should we adopt a new ethic? urologist beliefs and practices. J Urol, 159: 1993, 1998
N Engl J Med, 339: 397, 1998

Source: http://www.sunu.hacettepe.edu.tr/ankara-uro-asistan/1.pdf