`THE JOURNAL OF UROLOGY
`Copyright © 1990 by AMERICAN UROLOGICAL ASSOCIATION, INC.
`
`Vol. 144, Novembe~
`Prirtted in U.S.A.
`
`PROSTATE SPECIFIC ANTIGEN FOR ASSESSING RESPONSE TO
`KETOCONAZOLE AND PREDNISONE IN PATIENTS WITH
`HORMONE REFRACTORY METASTATIC PROSTATE CANCER
`
`GLENN S. GERBER AND GERALD W. CHODAK*
`From the Department of Surgery, Section of Urology, and Pritzker School of Medicine, The University of Chicago, Chicago, Illinois
`
`ABSTRACT
`
`Serial prostate specific antigen levels were assessed in 15 patients with hormone refractory
`metastatic prostate cancer treated with ketoconazole and prednisone. Of the men 12 (80%) with
`continually increasing prostate specific antigen levels before treatment had a decrease in prostate
`specific antigen with a median duration of response of 3 months. Three patients (20%) had a
`prolonged response (greater than or equal to 8 months) as seen by a persistently decreasing prostate
`specific antigen and improvement in bone pain. There appears to be a small subgroup of patients
`with progressive prostate cancer despite androgen ablation who will benefit from ketoconazole and
`glucocorticoid treatment. The use of serial prostate specific antigen levels appears to help define
`this subgroup and avoid the need for multiple radiological procedures to assess response. (J. Ural.,
`144: 1177-1179, 1990)
`
`Ketoconazole was originally developed as an antifungal agent
`effective against a wide variety of pathogenic fungi. 1 However,
`it also was noted that this drug is a potent inhibitor of gonadal
`and adrenocortical steroid synthesis.2
`3 In addition, an in vitro
`•
`cytotoxic effect on prostate cancer cells was demonstrated. 4
`These findings suggested a potential role for ketoconazole in
`the treatment of prostate cancer. Initially, the drug was used
`with favorable results in patients with previously untreated
`stage D2 disease.r.-7 However, in men with hormone refractory
`metastatic prostate cancer, results with ketoconazole have not
`been as promising.6-9 Jubelirer and Hogan reported findings in
`16 of their patients and they reviewed the literature concerning
`ketoconazole use in men with disease progression despite ade(cid:173)
`quate hormonal therapy.8 They noted that complete and partial
`responses to ketoconazole were seen in less than 1 and 14% of
`the patients, respectively. The criteria to assess response in
`these studies were varied but generally included reduction in
`measurable tumor size, improvement in bone scan abnormali(cid:173)
`ties and/or decrease in acid phosphatase level. None of the
`reports used serial prostate specific antigen (PSA) levels to
`assess clinical response. While changes in PSA are a good
`indicator of disease activity in men with metastatic prostate
`cancer treated with hormonal manipulation, 10 the role of PSA
`changes in patients treated with second line hormonal agents
`in unclear. Therefore, we were prompted to investigate the
`changes in PSA in men with hormone refractory metastatic
`prostate cancer treated with a combination of ketoconazole and
`prednisone.
`
`MATERIALS AND METHODS
`
`A total of 15 patients with hormone refractory metastatic
`prostate cancer was treated with ketoconazole and prednisone.
`Of the patients 14 had undergone orchiectomy a minimum of 5
`months before treatment, while 1 had been treated with injec(cid:173)
`tions of a luteinizing hormone-releasing hormone agonist for
`17 months before therapy. The latter patient had an anorchid
`serum testosterone level when treatment with ketoconazole and
`prednisone was begun and he was maintained on the luteinizing
`hormone-releasing hormone agonist while on ketoconazole and
`Accepted for publication May 8, 1990.
`Supported in part by the Whirlpool Foundation and the Cancer and
`Urology Research Endorsement Fund.
`*Requests for reprints: The University of Chicago, Box 403, 5841 S.
`Maryland Ave., Chicago, Illinois 60614.
`
`prednisone. All patients had histologically proved adenocarci(cid:173)
`noma of the prostate and bone scan evidence of metastatic
`disease at the onset of hormonal therapy. The median duration
`of response to hormonal therapy, as assessed by stable or
`decreasing PSA levels and no progression of bone pain, was 14
`months (range 2 to 50 months).
`Unresponsiveness to the initial hormonal therapy was de(cid:173)
`fined as an increasing PSA level on 2 consecutive determina(cid:173)
`tions that were at least 1 month apart. All 15 patients satisfied
`this requirement. In addition to increasing continually PSA
`levels 12 patients had bone pain and/or other symptoms of
`widespread malignancy at the onset of ketoconazole and pred(cid:173)
`nisone therapy. The remaining 3 patients were asymptomatic.
`Failure on ketoconazole and prednisone therapy was defined as
`an increasing PSA level on 2 consecutive determinations.
`Of 15 patients 10 had previously received radiotherapy. No
`patient received radiation concurrently with ketoconazole and
`prednisone. Of these 10 patients 9 had evidence of progressive
`disease (increasing PSA levels) after the completion of radio(cid:173)
`therapy and before initiation of ketoconazole and prednisone,
`while 1 had a lower PSA level after radiation and was treated
`with ketoconazole and prednisone because of continued bone
`pain. The median interval from completion of radiotherapy to
`the onset of treatment with ketoconazole and prednisone in the
`10 patients was 8 months (range 2 weeks to 5 years).
`Serum PSA levels were determined with the Hybritech Tan(cid:173)
`dem-R assay in most cases. In several patients the Yang Pros(cid:173)
`Check assay was used before a change in the methodology used
`by the laboratory at our hospitals. While each assay has a
`different range of normal values, both have been shown to be
`reliable with highly reproducible results. 11
`12 Some studies have
`·
`suggested that variables, such as bed rest and hospitalization,
`may cause a substantial decrease in PSA levels. 11
`13 In addition,
`•
`prostate manipulation, especially core biopsy, may cause in(cid:173)
`creases in serum PSA levels. 13 To control for these factors all
`PSA levels were determined in ambulatory patients seen in the
`outpatient clinic. Prostate manipulation or cystoscopy always
`was performed after a serum sample for PSA determination
`had been collected.
`Although some patients were initially analyzed with the Yang
`assay and then subsequently analyzed with the Hybritech
`method, none of the positive responses was the result of chang(cid:173)
`ing the assay method (figs. 1 to 3).
`Patients were initially treated with 600 to 900 mg. ketocon-
`
`1177
`
`<T>1,16<END1>1<END2>14<END3>(577,-14)<E4>22</E4>0<E5>1<E6>18<E7>11<E8>12/1/2015 12:00:00 AM14:54:28.7994914<E9></T>
`
`WCK1004
`Page 1
`
`
`
`GERBER AND CHODAK
`
`1178
`
`60
`
`_so
`E
`.......
`Ol40
`c:
`.....,
`;;j 30
`0..
`E 20
`::I
`......
`Q)
`en 10
`
`2
`
`6
`
`10
`
`14
`
`22
`18
`Months
`FIG. 1. Changes in PSA for patient 1 who was initially treated by
`bilateral orchiectomy and then ketoconazole and prednisone were ini(cid:173)
`tiated when PSA was 60 ng./ ml.
`
`26
`
`30
`
`34
`
`60
`
`-so
`E
`.......
`g> 40
`;;j 30
`0..
`E 20
`::I
`......
`Q) en 10
`
`2
`
`6
`
`10
`
`14
`
`18
`22
`Months
`FIG. 2. Patient 2 had increasing PSA despite bilateral orchiectomy
`and focal radiation. Ketoconazole and prednisone were initiated when
`PSA was 60 ng./ml.
`
`26
`
`30
`
`34
`
`240
`
`azole daily in 3 divided doses and 5 mg. prednisone twice per
`day. PSA levels were recommended monthly and the ketocon(cid:173)
`azole dosage was increased to 1,200 mg. daily in 3 divided doses
`if the PSA did not decrease. At the t ime of each visit patients
`also underwent history and physical examination, and they
`were specifically questioned regarding treatment-related side
`effects. Liver function fests were evaluated along with each
`serum PSA level.
`
`RESULTS
`Of the 15 patients 12 (80%) with hormone refractory prostate
`cancer showed a decrease in PSA in response to ketoconazole
`and prednisone with a median duration of response of 3 months.
`The mean decrease in PSA in the 12 responding patients was
`49% of the pre-treatment level (mean 312 ng./ml., range 38 to
`1,717, before treatment and 138 ng./ml., range 6 to 347, after
`treatment). Twelve patients also had bone pain and/or other
`symptoms of advancing malignancy, of whom 9 (75%) had
`subjective improvement. The decrease in PSA corresponded to
`subjective improvement in symptomatic patients with 1 excep(cid:173)
`tion. One patient had a 7% decrease in PSA during 4 months
`of treatment but he had no improvement in bone pain. One
`patient who had no decrease in PSA did have a decrease in
`bone pain. The mean peak PSA level before treatment in the
`nonresponding patients was 300 ng./ml.
`The duration of response to ketoconazole and prednisone
`was 4 months or less in 9 of the 12 responding men (75%).
`However, 3 patients had a prolonged response of 8 to 10 months
`(see table). Patient 1 underwent radical prostatectomy in 1984
`at which time he had stage D1 disease. Bone metastases sub(cid:173)
`sequently developed and he was treated with bilateral orchiec(cid:173)
`tomy in 1986. In December 1988 a steadily increasing PSA
`level as well as bone pain were noted and the patient was
`treated with ketoconazole and prednisone (fig. 1). The bone
`pain resolved and the PSA level decreased rapidly until Sep(cid:173)
`tember 1989 when it began to increase again. Patient 2 was
`diagnosed with stage D2 disease in 1983 and underwent bilat(cid:173)
`eral orchiectomy. Back pain developed due to bony metastases
`and he was treated with local radiotherapy through May. Before
`radiation the PSA level was 16 ng./ml. and 3 months after
`treatment it increased to 60 ng./ml. (fig. 2). In addition, the
`back pain recurred and a bone scan showed progressive disease.
`He was treated with ketoconazole and prednisone beginning in
`August 1988. Subsequently, the back pain improved and the
`PSA level decreased. In January 1989 the bone scan showed
`stable disease and he remained clinically well until Jun;,>, when
`the bone pain worsened and the PSA increased rapidly. He
`died shortly thereafter. Patient 3 underwent radiotherapy for
`localized prostate cancer in 1984. In 1987 bone metastases
`developed and he was treated with bilateral orchiectomy. He
`remained stable until November 1988, when the PSA level had
`increased to 180 ng./ml. and the bone scan showed progressive
`disease. The patient was asymptomatic at this time. He was
`treated with ketoconazole and prednisone, and after an initial
`decrease in PSA the level began to increase again 2 months
`later, at which time the drugs were discontinued (fig. 3). Sub(cid:173)
`sequently, bone pain developed and the patient self-adminis(cid:173)
`tered ketoconazole and prednisone. At the next outpatient
`clinic visit he reported complete resolution of the bone pain
`
`PSA levels in long-term (more than 6 months) responders to
`ketoconazole and prednisone
`Pre-Treatment
`Post· Treatment
`Peak PSA Level
`Nadir PSA Level
`Ng./Ml. (% decrease)
`Ng./ Ml.
`
`Mos. of
`Response
`
`Subjective
`Improvement
`
`P t.
`No.
`
`1
`2
`3*
`
`58
`60
`180
`206
`
`Yes
`10
`Yes
`10
`2
`Asymptomatic
`Yes
`8
`• Received second course of therapy after initial progression.
`
`6 (90)
`15 (75)
`122 (32)
`123 (40)
`
`_200
`E
`........
`Cl160
`
`c: -<t: 120
`
`(/)
`a..
`E so
`....
`::I
`Q)
`(/) 40
`
`> a.
`ttl
`....
`Q)
`.£;
`1-
`Q)
`:l
`,!::;
`
`Q)
`
`-~
`Cii
`c:
`'ii)
`a:
`
`Q)
`
`.~ -:s
`
`c
`0
`0
`en
`0
`~
`8
`10
`Months
`FIG. 3. Patient 3 initiated ketoconazole and prednisone after PSA
`increased from 100 to 180 ng./ml. After short response PSA again
`increased and medication was discontinued. Patient began therapy
`again because of increasing pain and PSA stabilized.
`
`2
`
`4
`
`6
`
`12
`
`14
`
`16
`
`WCK1004
`Page 2
`
`
`
`PROSTATE SPECIFIC ANTIGEN
`
`1179
`
`and the PSA level had decreased from 198 to 155 ng./ml. He
`remained asymptomatic with a steadily decreasing PSA until
`November 1989 when the pain recurred and the PSA increased
`slightly. Nevertheless, he was maintained on ketoconazole and
`prednisone and the PSA has continued to fluctuate near its
`nadir level.
`Ketoconazole was generally well to.lerated. One patient had
`nausea and vomiting requiring discontinuation of the drugs
`after 1 month of treatment. Two patients suffered minor bruis(cid:173)
`ing believed to be secondary to prednisone. No significant
`elevation in liver function studies was observed.
`
`DISCUSSION
`After the demonstration of diminished adrenocortical and
`gonadal steroid synthesis by ketoconazole,2 several investiga(cid:173)
`tors reported results with this drug in men with untreated
`metastatic prostate cancer.6
`7 The combined results of these
`-
`trials showed a response rate in excess of 80%. However, in
`patients with disease progression despite adequate hormonal
`therapy the objective response rate with ketoconazole has been
`less than 15%.8 Based on the latter results Jubelirer and Hogan
`concluded that ketoconazole as a single agent has limited use
`in patients who have failed prior hormonal therapy for ad(cid:173)
`vanced prostate cancer.8 In contrast to this conclusion Trump
`and associates stated that there may be a small subset of
`patients with hormone refractory disease who will benefit from
`ketoconazole treatment.14 They found that 5 of 36 patients had
`a greater than 50% decrease in tumor mass or a regression of
`disease on bone scan when treated with the combination of
`ketoconazole and physiological glucocorticoid replacement
`therapy. Since it is apparent that a small but reasonable per(cid:173)
`centage of men failing standard hormonal therapy will respond
`favorably to ketoconazole, it is important to identify these
`patients as objectively as possible. Recent studies with PSA
`suggested that this marker could be useful to identify the
`patients who respond to ketoconazole.
`While it generally is accepted that PSA changes correlate
`with disease activity10 the significance of these changes in men
`treated with second-line hormonal agents is unclear. Of the 5
`responding patients in the series reported by Trump and asso(cid:173)
`ciates 4 had a greater than 80% decrease in PSA. 14 However,
`several patients with clinical evidence of disease progression
`bad either stable or diminished PSA levels. These investigators
`concluded that over-all PSA changes did not reliably reflect
`disease response or progression.
`In our study 12 of 15 castrated patients (80%) with progres(cid:173)
`sively increasing PSA levels had a decrease in PSA in response
`to treatment with ketoconazole and prednisone. While this
`improvement in PSA was short-lived (less than or equal to 4
`months) and occasionally of small magnitude in 9 of the 12
`men, it did correlate with subjective improvement in sympto(cid:173)
`matic patients in all but 1 instance. The importance of short(cid:173)
`term decreases in PSA is unclear, although it is unlikely that
`significant impact on survival will be seen in these cases. These
`transient PSA decreases may partially explain the disease
`progression in patients with stable or diminished PSA levels
`seen by Trump and associates. 14 Of our 3 patients in whom the
`PSA level continued to increase despite treatment with keto(cid:173)
`conazole and prednisone, 1 reported a transient decrease in
`bone pain. In general, therefore, changes in PSA correlated
`well with symptomatology.
`Of our patients 3 (20%) have had prolonged (8 to 10 months)
`favorable response to ketoconazole and prednisone based on
`persistently decreasing PSA levels and symptomatic improve(cid:173)
`ment. This rate of response is similar to that found in studies
`
`that have used changes in measurable tumor size, bone scan
`abnormalities and acid phosphatase to assess response.8
`14
`·
`Thus, it appears that serial PSA levels may be useful to define
`the small subgroup of men failing standard hormonal therapy
`who will benefit from the combination of ketoconazole and
`prednisone. The use of PSA may then obviate the need for
`repeated bone scans, chest radiographs, and abdominal and
`pelvic computerized tomography scans to assess disease re(cid:173)
`sponse. In our experience approximately 10% of the patients
`will have a low PSA despite progressive metastatic disease and
`their response will have to be assessed by other methods.
`In summary, as in the report by Trump and associates14 there
`appears to be a small subgroup of patients with progressive
`prostate cancer despite hormonal therapy who will derive sig(cid:173)
`nificant benefit from the combination of ketoconazole and
`glucocorticoid replacement therapy. The presence of a persist(cid:173)
`ent {greater than or equal to 6 months) decrease in PSA, as
`well as prolonged relief of symptoms may help to define these
`patients and avoid the need for multiple radiological procedures
`to assess response. Short-term decreases in PSA are of unclear
`importance but probably do not reflect significant disease
`regression.
`
`REFERENCES
`1. Restrepo, A., Stevens, D. A. and Utz, J. P.: First International
`Symposium on Ketoconazole. Introduction. Rev. Infect. Dis., 2:
`519, 1980.
`2. Sonino, N.: The use of ketoconazole as an inhibitor of steroid
`production. New Engl. J. Med., 317: 812, 1987.
`3. Pont, A., Williams, P. L., Loose, D. S., Feldman, D., Reitz, R. E.,
`Bochra, C. and Stevens, D. A.: Ketoconazole inhibits adrenal
`steroid synthesis. Ann. Intern. Med., 97: 370, 1982.
`4. Eichenberger, T., Trachtenberg, J., Toor, P. and Keating, A.:
`Ketoconazole: a possible direct cytotoxic effect on prostate car(cid:173)
`cinoma cells. J. Urol., 141: 190, 1989.
`5. Pont, A.: Long-term experience with high dose ketoconazole ther(cid:173)
`apy in patients with stage D2 prostatic carcinoma. J. Urol., 137:
`902, 1987.
`6. Vanuytsel, L., Ang, K. K., Vantongelen, K., Drochmans, A., Baert,
`L. and VanDer Schueren, E.: Ketoconazole therapy for advanced
`prostatic cancer: feasibility and treatment results. J. Urol., 137:
`905, 1987.
`7. Debruyne, F. M. J . and Witjes, F. J.: Ketoconazole high dose (H.D.)
`in the management of metastatic prostatic carcinoma. J. Urol.,
`part 2, 135: 203A, abstract 397, 1986.
`8. Jubelirer, S. J. and Hogan, T.: High dose ketoconazole for the
`treatment of hormone refractory metastatic prostate carcinoma:
`16 cases and review of the literature. J. Urol., 142: 89, 1989.
`9. Williams, G., Kerle, D. J., Ware, H., Doble, A., Dunlop, H., Smith,
`C., Allen, J., Yeo, T. and Bloom, S. R.: Objective responses to
`ketoconazole therapy in patients with relapsed progressive pros(cid:173)
`tatic cancer. Brit. J. Urol., 58: 45, 1986.
`10. Hudson, M. A., Bahnsen, R. R. and Catalona, W. J.: CLinical use
`of prostate specific antigen in patients with prostate cancer. J.
`Urol., 142: 1011, 1989.
`11. Stamey, T. A., Yang, N., Hay, A. R., McNeal, J. E., Freiha, F. S.
`and Redwine, E.: Prostate-specific antigen as a serum marker
`for adenocarcinoma of the prostate. New Engl. J . Med., 317:
`909, 1987.
`12. Maatman, T. J.: The role of prostate specific antigen as a tumor
`marker in men with advanced adenocarcinoma of the prostate.
`J . Urol., 141: 1378, 1989.
`13. Stamey, T. A.: Prostate specific antigen in the diagnosis and
`treatment of adenocarcinoma of the prostate. Monogr. Urol.,
`10: 50, 1989.
`14. Trump, D. L., Havlin, K. H., Messing, E. M., Cummings, K. B.,
`Lange, P. H. and Jordan, V. C.: High-dose ketoconazole in
`advanced hormone-refractory prostate cancer: endocrinologic
`and clinical effects. J. Clin. Oncol., 7: 1093, 1989.
`
`r
`
`WCK1004
`Page 3