This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Schmid, H.-P. Articles by Dawson, N. A. Search for Related Content PubMed PubMed Citation Articles by Schmid, H.-P. Articles by Dawson, N. A. Social Bookmarking                            What's this?

Prostate-Specific Antigen Doubling Time: A Potential Surrogate End Point in Hormone-Refractory Prostate Cancer

Westfaelische Wilhelms University, Muenster, Germany
Swiss Group for Clinical Cancer Research, Bern, Switzerland

To the Editor: In hormone-refractory prostate cancer (HRPC), standardized end points in clinical trials are urgently needed. According to a 125-question survey, 33 (94%) of 35 established investigators of HRPC accepted a decrease of prostate-specific antigen (PSA) level as evidence of response to chemotherapy.¹ However, there was no consensus on what degree of PSA level decrease is necessary for a response. Furthermore, the interpretation of the time at which a patient both responds and progresses based on changes in PSA was subject to considerable variability.

To further assess the antiproliferative efficacy of cytotoxic agents, we suggest PSA doubling time (PSADT) as a novel surrogate end point in HRPC. In untreated patients with prostate cancer, as well as in patients who relapse either after radical prostatectomy or radiation therapy, the increase of serum PSA over time is exponential (log-linear) in the vast majority of cases.² This observation allows for calculation of PSADT according to the following formula:

where t is the time from initial to final PSA determination. In a recent multicenter phase II trial, the Swiss Group for Clinical Cancer Research reported for the first time on PSADT in HRPC.³

In that study, 30 patients were enrolled to receive oral idarubicin (4-demethoxy-daunorubicin). Patients were administered 35 mg of idarubicin on days 1 and 8 of each cycle, and treatment was repeated every 3 weeks. Twenty-six patients were fully assessable. Three patients with measurable lesions had stable disease as their best response; their PSA also remained stable. In the other 23 patients (10 with measurable and 13 with nonmeasurable disease), PSA increased exponentially over time. The median PSADT was 2.1 months (range, 0.7 to 6.1 months, mean, 2.6 months) which is very similar to that in patients who progress under antiandrogen therapy alone,⁴ suggesting that idarubicin has no effect at all in these patients.

We conclude that PSADT may be a potential surrogate end point in HRPC. This dynamic parameter could be used to assess the antiproliferative activity of cytotoxic agents and to compare the efficacy of available therapies, keeping in mind that PSA changes do not necessarily always reflect tumor behavior.⁵

REFERENCES

1. Dawson NA: Apples and oranges: Building a consensus for standardized eligibility criteria and end points in prostate cancer clinical trials. J Clin Oncol 16:3398-3405, 1998[Abstract]

2. Schmid H-P: Prostate specific antigen doubling time in diagnosis and follow-up of patients with prostate cancer. Tumour Marker Update 8:71-77, 1996 (review)

3. Schmid H-P, Maibach R, Bernhard J, et al: A phase II study of oral idarubicin as a treatment for metastatic hormone-refractory prostate carcinoma with special focus on prostate specific antigen doubling time. Cancer 79:1703-1709, 1997[Medline]

4. Fowler JE Jr Pandey P, Seaver LE, et al: Prostate specific antigen regression and progression after androgen deprivation for localized and metastatic prostate cancer. J Urol 153:1860-1865, 1995[Medline]

5. Schroeder FH, Norming U, Blumenstein BA, et al: Phase II studies on prostate cancer. Urology 49:3-14, 1997 (suppl 4A)[Medline]

Walter Reed Army Medical Center, Washington, DC

In Reply: Drs Schmid, Semjonow, and Maibach are correct in pointing out that PSA is a moving target and that a dynamic model may potentially improve its usefulness as a surrogate end point. The role of PSA as a response criterion in clinical trials in HRPC is evolving. Initial retrospective analyses suggested a decline in PSA, usually 50%, at a fixed time, as correlated with prolonged survival.^1,2 As pointed out in the referenced survey,³ investigators rank survival as the most critical end point. Hence, the validity of PSA or a PSA-based model as a surrogate end point will be determined by how well it predicts the probability of dying.

PSADT as defined by these authors takes into consideration that increase in PSA over time is exponential in most patients⁴ and that the effect of any therapy on PSADT may correlate with whether the therapy being tested is efficacious. As a test of this hypothesis, investigators in the Cancer and Leukemia Group B (CALGB) have tried to refine a dynamic PSA model, with an eye toward assessing the effect of a given therapeutic intervention.^5,6 We have recently analyzed PSA data on 148 men with HRPC treated with either low-dose (160 mg/d) or high-dose (640 mg/d) megestrol acetate (CALGB 9181).⁵ Both log (PSA) and average relative velocity of PSA (rva), defined as (dy/dt)/y, with y symbolizing serum PSA and t symbolizing time, were significantly correlated with survival (P = .0001 and P = .0008, respectively). Further analysis with a Cox proportional hazards model resulted in the following formula for a PSA hazard score:

This score is closely correlated with the probability of death.⁵ We have subsequently validated and improved this model based on 239 additional men treated on a second clinical trial of men randomized to low-dose hydrocortisone with or without mitoxantrone (CALGB 9182).⁶ The new model incorporates serial measurements of two additional known prognostic variables, patient weight (wgt) and serum hemoglobin (hgb). The prognostic score for the four variables at time t is 0.156 _* (log PSA(t)) - mean log(PSA)) + 17.2 _* (rva(t) - mean rva) - 3.16 _* (log(hgb(t)) - mean log(hgb)) - 1.69 _* (log(wgt(t)) - mean log(wgt)). This score, which can be calculated with an available computer program, is dynamic, is closely correlated with survival, can be determined at any time in the course of the disease, and has potential to determine the early effects of treatment. Additionally, by incorporation of rva, this model can account for declining, as well as rising or stable, PSA values.

In summary, I agree that the use of a dynamic PSA model has theoretical advantages over a static assessment, but that these scores must be prospectively validated and must be easily determined for practical use in the care of patients.

REFERENCES

1. Kelly WK, Scher HI, Mazumdar M, et al: Prostate-specific antigen as a measure of disease outcome in metastatic hormone-refractory prostate cancer. J Clin Oncol 11:607-615, 1993[Abstract]

2. Smith DC, Dunn RL, Strawderman MS, et al: Change in serum prostate-specific antigen as a marker of response to cytotoxic therapy for hormone-refractory prostate cancer. J Clin Oncol 16:1835-1843, 1998[Abstract]

3. Dawson NA: Apples and oranges: Building a consensus for standardized eligibility criteria and end points in prostate cancer clinical trials. J Clin Oncol 16:3398-3405, 1998

4. Schmid H-P: Prostate specific antigen doubling time in diagnosis and follow-up of patients with prostate cancer. Tumor Marker Update 8:71-77, 1996 (review)

5. Vollmer RT, Dawson NA, Vogelzang NJ: The dynamics of prostate specific antigen in hormone refractory prostate cancer: An analysis of Cancer and Leukemia Group B study 9181 of megestrol acetate. Cancer 83:1989-1994, 1998[Medline]

6. Vollmer RT, Kantoff PW, Dawson NA, et al: A prognostic score for hormone refractory prostate cancer: Analysis of two CALGB studies. Clin Cancer Res (in press)

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Schmid, H.-P. Articles by Dawson, N. A. Search for Related Content PubMed PubMed Citation Articles by Schmid, H.-P. Articles by Dawson, N. A. Social Bookmarking                            What's this?