|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2003 American Society for Clinical Oncology Prostate-Specific Antigen Kinetics as a Measure of the Biologic Effect of Granulocyte-Macrophage Colony-Stimulating Factor in Patients With Serologic Progression of Prostate Cancer
From The University of California San Francisco Comprehensive Cancer Center, San Francisco, CA. Address reprint requests to Brian I. Rini, MD, UCSF Comprehensive Cancer Center, 1600 Divisadero, 3rd floor, San Francisco, CA 94115; email: brini{at}medicine.ucsf.edu.
Purpose: To determine the biologic effect of granulocyte-macrophage colony-stimulating factor (sangramostim, GM-CSF; Immunex Corporation, Seattle, WA) as measured by prostate-specific antigen (PSA) kinetics in patients with serologic progression of prostate cancer after definitive local therapy. Patients and Methods: Patients with prostate cancer who had undergone previous definitive surgical or radiation therapy with nonmetastatic, recurrent disease as manifested by a rising PSA between 0.4 ng/mL and 6.0 ng/mL were enrolled on this phase II trial. Patients received 250 µg/m2/day of subcutaneous GM-CSF on days 1 through 14 of a 28-day cycle. PSA was measured at day 1 of each cycle. Results: Thirty patients with serologic progression of prostate cancer were treated. The median pretreatment PSA was 2.9 ng/mL. Of the 29 evaluable patients, three patients (10%; 95% confidence interval, 2% to 27%) achieved a 50% reduction in PSA. For the patients (n = 26) in whom the on-treatment PSA doubling time could be calculated, the median PSA doubling time increased from 8.4 months to 15 months (P = .001), and the median slope of the PSA versus time curve decreased with treatment (P = .004). Therapy was well tolerated by all patients, with an average treatment duration of 16.5 cycles (range, 5 to 33). Conclusion: GM-CSF has a biologic effect in patients with serologic progression of prostate cancer after definitive local therapy, as measured by PSA declines and modulation of PSA kinetics.
DESPITE ADVANCES in early detection and treatment of localized prostate cancer, 20% to 60% of patients treated with radiation or prostatectomy will eventually develop recurrent disease.1,2 This recurrence is often detected in patients who are asymptomatic and who have rising prostate-specific antigen (PSA) values as the sole manifestation of relapse (biochemical relapse). Although androgen deprivation is a treatment option, the associated loss of libido and chronic sequelae including fatigue, anemia, and osteoporosis make this therapy unacceptable to some patients. Novel treatments are needed, and immune-based treatments may ideally be applied in these otherwise healthy patients with intact immune function and minimal disease burden. The evaluation of therapy for patients with biochemically relapsed prostate cancer is hampered by a lack of defined clinical end points. Tremendous variability in the natural history of patients with serologic progression and the long time to development of metastatic disease make it difficult to use standard clinical end points such as objective progression or survival. It is clear that a detectable and progressively rising PSA after radical prostatectomy or radiation therapy portends a worse outcome. For example, a more rapid PSA doubling time after local therapy is predictive of the development of metastatic disease.3,4 It is unknown, however, whether altering PSA kinetics as a consequence of a therapeutic intervention translates into an altered clinical outcome. Nevertheless, novel agents that result in a slowing or reversal of a PSA rise might be considered for further study. In this manner, the effect of a novel agent on PSA kinetics can be viewed as a screen for biologic effect, but not necessarily for clinical efficacy. Granulocyte-macrophage colony-stimulating factor (GM-CSF) is a cytokine that regulates the differentiation and function of granulocytes and macrophages. GM-CSF has also been demonstrated to be a potent growth factor and stimulator of dendritic cells (DCs), leading to T-cell cross-priming. GM-CSF has been tested as an immune modulator in several prostate cancer studies.
Cellular approaches to local GM-CSF delivery have been explored. In one study, autologous prostate cancer cells removed at surgery were modified to secrete high levels of GM-CSF via ex vivo retroviral transduction with cDNA encoding GM-CSF. These cells were then irradiated and injected subcutaneously into patients.5,6 This approach was based on the rationale that GM-CSF promotes uptake of tumor antigens by DCs that subsequently stimulate antitumor T cells and thus promote systemic anticancer immunity. Eight patients with locally advanced prostate cancer were vaccinated every 21 days (three to six vaccinations). Effective induction of a cellular immune response was demonstrated as measured by increases in delayed-type hypersensitivity (DTH) responses against irradiated, unmodified autologous tumor cells. Another approach used irradiated, allogeneic prostate cancer cell lines (PC-3 and LNCaP) transduced ex vivo with the human GM-CSF gene. Ninety-six patients with hormone-naive or hormone-refractory prostate cancer received a single priming dose followed by 12 booster doses at 2-week intervals.7 One hormone-naive patient had a PSA decline On the basis of the hypothesis that apoptotic tumor cells would provide a source of tumor antigen and that systemic GM-CSF could also lead to T-cell cross-priming by DC, systemic GM-CSF has been studied in hormone-refractory prostate cancer (HRPC)8 A cohort of 23 patients with a median pretreatment PSA of 276 ng/mL received subcutaneous GM-CSF 250 µg/m2/d for the first 14 consecutive days of a 28-day cycle. PSA declines were seen in 10 patients during GM-CSF administration, with increases during the 14-day rest period. The median PSA decline was 37% (range, -5.8% to -64%), with five patients experiencing a more than 50% PSA decline on at least one occasion. PSA declines were not sustained, and all patients eventually progressed objectively or by PSA criteria. Because of the oscillating nature of PSA values in this first cohort, an additional cohort of 13 men with HRPC was treated. This second cohort received an initial 14 days of 250 µg/m2/d of SC GM-CSF followed by a maintenance period of GM-CSF 250 µg/m2 three times weekly. All but one patient in the second cohort experienced a PSA decline (median decline, 32.4%), and a PSA decline of more than 50% sustained for more than 14 months was seen in one patient, persistent for 14+ months. Of note, this PSA decline (from 77 to 0.1 ng/mL) was associated with an improved bone scan. GM-CSF was tested in vitro in the androgen-dependent cell line LNCaP to evaluate effects on PSA secretion. No significant decreases in PSA secretion were observed, not accounting for the magnitude of PSA declines observed in the clinical trial. In addition, there was no evidence that the presence of GM-CSF affected the assay for PSA.
Systemic GM-CSF has also been investigated in the setting of biochemically relapsed prostate cancer. Dreicer et al9 treated 16 men with androgen-dependent (n = 7) or androgen-independent (n = 9) prostate cancer progressive after local therapy. Baseline PSA was 7.8 ng/mL, and 12 patients had PSA-only disease. GM-CSF 250 µg was given subcutaneously three times weekly for a maximum of 24 weeks. Whereas 10 patients had an initial PSA decline, a On the basis of the biologic effect of systemically administered GM-CSF in patients with HRPC, a prospective phase II trial of GM-CSF in patients with serologic progression of prostate cancer was conducted. The primary end point of this trial was to determine the effect of this therapy in promoting PSA declines and modulating PSA kinetics.
Patients with histologically diagnosed adenocarcinoma of the prostate who had undergone prior definitive therapy for prostate cancer consisting of external beam radiotherapy, brachytherapy (with or without pelvic external beam radiotherapy), or radical prostatectomy were eligible. Patients treated with adjuvant or salvage radiation therapy after radical prostatectomy were eligible provided postprostatectomy PSA was never more than 6.0 ng/mL. Patients were required to have had a therapeutic PSA response to primary therapy, defined arbitrarily as a PSA below 1.0 ng/mL after radiation therapy and below 0.4 ng/mL after radical prostatectomy. Progressive disease, defined as two climbing PSA values between 0.4 ng/mL and 6.0 ng/mL measured at least 1 week apart, was required. Patients with a local recurrence or evidence of metastases on bone scan or computed topography (CT) scan were excluded. Adequate renal and hepatic function (defined as total bilirubin/serum glutamic-oxaloacetic [SGOT] less than two times the upper limit of normal and blood urea nitrogen [BUN]/serum creatinine less than two times the upper limit of normal) were required. All patients signed a written, informed consent approved by the University of California San Francisco Committee on Human Research. Prior hormonal therapy for the treatment of progressive disease was not permitted (prior hormonal therapy used in an adjuvant or neoadjuvant setting was permitted, but the last day of effective androgen deprivation must have been at least 3 months before study entry). Prior chemotherapy, immunotherapy, or investigational therapy for prostate cancer was not permitted. In addition, patients with clinically significant cardiac or pulmonary disease, uncontrolled infection, surgery within the prior 4 weeks, or current systemic steroid therapy were excluded. Concurrent use of PC-SPES or Saw Palmetto was prohibited, and prior use of these agents for progressive disease was also not allowed.
Patients received 250 µg/m2/d of GM-CSF administered subcutaneously on days 1 through 14 of a 28-day cycle. PSA was measured at day 1 and day 15 of the first three cycles, then on day 1 of all subsequent cycles. Assessment of PSA changes first occurred after three cycles of therapy, then monthly thereafter. Disease progression was arbitrarily defined as a change from nadir PSA value or pretreatment PSA value (whichever was lowest) that was maintained for at least two measurements at least 2 weeks apart as follows: for patients with maximum pretreatment PSA The National Cancer Institute Cancer Clinical Trials common toxicity criteria version 2.0 were used to assess toxicity. GM-CSF therapy was discontinued if the absolute neutrophil count (ANC) exceeded 20,000/µL at any point during therapy or if there were signs or symptoms attributed to hyperviscosity. If GM-CSF was discontinued because of ANC more than 20,000/µL, therapy was held until the ANC had decreased to less than 15,000/µL and then resumed with a 50% reduction of the previous dose. Subsequent GM-CSF dosage on all ensuing cycles continued at the reduced dose. Patients were taken off the study for any grade 4 toxicity, any grade 3 toxicity persisting more than 4 weeks, or any recurrent grade 3 toxicity.
Statistical Considerations
For analysis of the data, the linear rate of increase of the PSA on the natural log (Ln) scale over time before entry on protocol as well as the change in Ln PSA calculated from the start of treatment were estimated by the slope. The PSA doubling time both pre- and on-treatment was estimated using the relationship of Ln 2 divided by the slope. If a decline in the rate of PSA with treatment was observed, then this resulted in a negative slope, and a doubling time could not be estimated. The probability of remaining free of progressive disease and time to nadir were estimated using the Kaplan-Meier product limit method. The
Patient Characteristics Thirty patients with serologic progression of prostate cancer were enrolled from February 1998 to December 1999. Table 1  lists the pretreatment patient characteristics. The median age of patients was 66 (range, 52 to 78). Primary therapy included radiation (XRT) in 11 patients, radical prostatectomy (RP) in eight patients, or RP followed by adjuvant radiation in 11 patients. Fifteen patients (12 who received prior XRT and three who received prior RP) had received neoadjuvant or adjuvant hormonal therapy. These patients received a median of 4 months of hormone therapy (range, 2 to 12 months) that was discontinued for a median of 34 months (range, 16 to 96 months) before treatment with GM-CSF. The median maximum pretreatment PSA was 2.9 ng/mL (range, 0.4 to 6.1 ng/mL). Twenty-eight patients had maximum pretreatment PSA values between 0.4 ng/mL and 6.0 ng/mL, inclusive. Two patients were eligible for the study based on two climbing pretreatment PSA values less than 6.0 ng/mL and had a PSA value on the first day of treatment of 6.1 ng/mL. Eighteen patients (64%) had a GS of 7 or higher. There was a suggestion that patients treated with RP had a lower maximum pretreatment PSA value than patients treated with XRT (median, 1.6 v 3.8; P = .07). In addition, it was more common for patients treated with prior adjuvant or neoadjuvant hormone therapy to have a shorter pretreatment PSA doubling time (median, 6.5 v 9.2 months; P = .09).
PSA Changes Twenty-nine patients were evaluable for assessment of posttreatment PSA changes with a median follow-up of 11.1 months. One patient withdrew consent after receiving only one cycle of GM-CSF and thus was unevaluable for this assessment. This paitent was, however, included in the toxicity evaluation. A PSA decline of  50% was observed in three patients (10%; 95% CI, 2% to 27%). Two of these patients were previously treated with XRT and hormonal therapy and the third patient underwent RP and adjuvant XRT. The pretreatment PSA doubling times were 6, 6, and 16 months for these patients. The first patient declined from a maximum pretreatment PSA value of 1.7 ng/mL to 0.7 ng/mL after 5 months of GM-CSF therapy, achieving a nadir of 0.5 ng/mL (71% PSA decline) at month 28. The patient remains on the study at 30 months with a most recent PSA value of 0.6 ng/mL. The second patient had a maximum pretreatment PSA of 1.5 ng/mL that declined continuously with treatment, reaching a nadir 13 months after starting therapy at a PSA value of 0.3 ng/mL (80% PSA decline). This patient was removed from the study at month 22 with a PSA of 0.7 ng/mL as a result of grade 4 toxicity (vide infra). The third patient demonstrated a PSA decline from a maximum pretreatment value of 1.6 ng/mL to 0.8 ng/mL at cycle 3. A nadir of 0.7 ng/mL (56% PSA decline) was reached after 8 months of therapy, and the patient continues on treatment at month 22 with PSA values ranging between 0.86 and 1.1 ng/mL.
Whereas the three patients with PSA declines
The PSA doubling time, an estimate of the rate of change of PSA, was calculated before and during treatment as an additional measure of treatment effect (PSA doubling time = Ln 2/slope of rate of change of Ln PSA). For determination of pretreatment PSA doubling time, a median of five PSA values were used (range, 4 to 9) and were collected over a median of 8.1 months (range, 2.5 to 25.1 months). The median estimated pretreatment doubling time was 8.4 months, yet there was a broad range (2.3 to 24.7 months). The pretreatment doubling time was not correlated with the maximum pretreatment PSA value. Three patients, including two patients with PSA declines 50% and an additional patient with a 32% PSA decline, could not have on-treatment PSA doubling times calculated because of declining PSA values. One patient who achieved a 50% PSA reduction now has increasing PSA values and, thus, an on-treatment PSA doubling time can be calculated. For the 26 patients who could have the on-treatment PSA doubling time calculated, the median PSA doubling time increased from 8.4 months pretreatment to 15.0 months on-treatment (P = .001; Table 2). Figure 2 demonstrates the paired pretreatment and on-treatment PSA doubling times for each patient.
Overall, 16 patients had a twofold increase in on-treatment doubling time compared with the pretreatment PSA doubling time. Figure 3a shows a graphic representation of the Ln PSA versus time curve for a patient who experienced such an increase in PSA doubling time. On-treatment doubling time could not be calculated because of declining PSA values. Nine patients had an on-treatment doubling time between one and two times the pretreatment PSA doubling time (Fig 3b, increase in doubling time from 8.2 months to 12.7 months), and four patients had a decrease in the PSA doubling time (Fig 3c, decrease in doubling time from 15.8 months to 7.1 months).
Seventeen patients developed progressive disease by PSA criteria (median time to progression of 15.2 months) and were removed from protocol therapy. Fifteen of these patients had a follow-up bone scan and CT scan, and two patients demonstrated a positive bone scan at the time of PSA progression. Two patients were removed from the study for toxicity (vide infra). Two patients chose to stop protocol therapy after 9 and 11 months of treatment before meeting criteria for PSA progression. Both of these patients had slowly rising PSA values increasing from 0.8 to 3.7 ng/mL and from 1.0 to 2.1 ng/mL, respectively. Eight patients, including two patients with PSA declines of more than 50%, remain on study with treatment duration ranging from 20 to 32 months.
Toxicity
There were also two grade 3 toxicities. The first patient is 68 years of age who presented on cycle 21, day 15 with left upper extremity edema. An ultrasound revealed a deep venous thrombosis. The patient was anticoagulated with heparin followed by warfarin. The patient had no prior history of thrombosis nor any identifiable hypercoaguable risk factors. Given the patient’s prolonged PSA nonprogression on GM-CSF, therapy was continued without further toxicity at full GM-CSF dose once he was adequately anticoagulated. The second grade 3 toxicity was observed in a patient 74 years of age during cycle 5 of GM-CSF. The patient had a history of extensive coronary artery disease and coronary bypass surgery and experienced symptomatic bradycardia requiring pacemaker placement. This toxicity was felt to be unlikely related to the study drug, and the patient was taken off the study.
Biochemical-only progression of prostate cancer represents a unique scenario for the testing of novel agents. Whereas measurable disease is lacking and significance of PSA reductions is debated, evaluation of PSA changes as a result of treatment can serve as a screen for biologic effect. That is, those agents that decrease or reverse the PSA rise could be considered for further investigation, whereas agents that fail to modulate the PSA versus time curve would be less likely to be developed further.
Systemic GM-CSF in this population of patients had a biologic effect as measured by PSA declines and change in PSA doubling time. Three patients demonstrated a
Several methodologic issues with regard to evaluation of a novel agent such as GM-CSF in biochemically relapsed prostate cancer deserve mention. The use of PSA doubling time as an indicator of biologic effect has the advantage of using all PSA values and allows for determination of PSA declines that have established significance in other disease states.11 Yet determination of the clinical significance of altering the PSA doubling time will require association of PSA doubling time prolongation with clinical end points, such as development of bone metastases or overall survival. This trial enrolled patients with a rising PSA after definitive local therapy without regard to pretreatment PSA doubling time. It may be that evaluation of novel therapies in patients with a shorter PSA doubling time who are at higher risk for the development of clinically significant events such as bone metastases would allow for more rapid evaluation, or that a certain range of PSA doubling times should be targeted for a given agent. As well, standardizing measurement of pretreatment and on-treatment PSA doubling time in regards to number of PSA values, time interval, and duration of PSA measurements could provide a more consistent basis on which to evaluate treatment-induced changes in PSA kinetics. Examination of the Ln PSA versus time curves in Fig 3 Therapy was well tolerated by patients. Minor cytokine-related and local injection-site toxicity were observed in the majority of patients. Tolerance of therapy is further evident by the number of patients who continued on therapy for a prolonged period of time. Median time on study is 11 months, with 12 patients treated longer than 18 months and eight patients treated for more than 24 months.
Systemic GM-CSF has been previously investigated in prostate cancer patients. Administration of GM-CSF to advanced HRPC patients in two different schedules resulted in
Although the results of this study are encouraging, several caveats hold. First, no measurement of antiprostate cancer immunity was undertaken in this study, and thus no correlation of GM-CSF-induced antiprostate cancer immunity and biologic effect can be made. Therefore, the mechanism by which GM-CSF modulates PSA expression is not known. We hypothesize that GM-CSF induced uptake and the processing of relevant prostate cancer antigens by DCs that, in turn, cross-primed antiprostate cancer T cells, thus leading to the biologic effect observed. This study allowed prior neo/adjuvant hormonal therapy that theoretically could have affected the cell population present on PSA recurrence. However, a median duration of 4 months of hormone therapy, a median of 34 months since last hormonal therapy, and a normal testosterone level at entry make this brief and distant hormonal therapy unlikely to affect the results of the study. In addition, the low pretreatment PSA values make determination of PSA changes imprecise. Further, a PSA decline of
The biologic effect of GM-CSF has been demonstrated in this and other trials with a subset of patients experiencing either PSA declines of
Supported in part by the Immunex Corporation, Seattle, WA, CaPCURE, and the UCSF Comprehensive Cancer Center (P30CA82103). Presented in part at the American Society of Clinical Oncology meeting, May 2001.
1. Powell CR, Huisman TK, Riffenburgh RH, et al: Outcome for surgically staged localized prostate cancer treated with external beam radiation therapy. J Urol 157:1754–1759, 1997[CrossRef][Medline] 2. Moul JW, Connelly RR, Lubeck DP, et al: Predicting risk of prostate specific antigen recurrence after radical prostatectomy with the Center for Prostate Disease Research and Cancer of the Prostate Strategic Urologic Research Endeavor Databases. J Urol 166:1322–1327, 2001[CrossRef][Medline]
3. Pound CR, Partin AW, Eisenberger MA, et al: Natural history of progression after PSA elevation following radical prostatectomy. JAMA 281:1591–1597, 1999 4. Zagars GK, Pollack A: Kinetics of serum prostate-specific antigen after external beam radiation for clinically localized prostate cancer. Radiother Oncol 44:213–221, 1997[CrossRef][Medline] 5. Nelson WG, Simons JW, Mikhak B, et al: Cancer cells engineered to secrete granulocyte-macrophage colony-stimulating factor using ex vivo gene transfer as vaccines for the treatment of genitourinary malignancies. Cancer Chemother Pharmacol 46:S67–72, 2000
6. Simons JW, Mikhak B, Chang JF, et al: Induction of immunity to prostate cancer antigens: Results of a clinical trial of vaccination with irradiated autologous prostate tumor cells engineered to secrete granulocyte-macrophage colony-stimulating factor using ex vivo gene transfer. Cancer Res 59:5160–5168, 1999 7. Simons JW, Small EJ, Nelson W, et al: Phase II trials of a GM-CSF gene-transduced prostate cancer cell line vaccine (GVAX) demonstrate anti-tumor activity. Program/Proc Am Soc Clin Oncol 20, 2001 (abstr 1073)
8. Small EJ, Reese DM, Um B, et al: Therapy of advanced prostate cancer with granulocyte macrophage colony-stimulating factor. Clin Cancer Res 5:1738–1744, 1999 9. Dreicer R, See WA, Klein EA: Phase II trial of GM-CSF in advanced prostate cancer. Invest New Drugs 19:261–265, 2001[Medline] 10. Wasilenko WJ, Palad AJ, Somers KD, et al: Effects of the calcium influx inhibitor carboxyamido-triazole on the proliferation and invasiveness of human prostate tumor cell lines. Int J Cancer 68:259–264, 1996[CrossRef][Medline]
11. Bubley GJ, Carducci M, Dahut W, et al: Eligibility and response guidelines for phase II clinical trials in androgen-independent prostate cancer: Recommendations from the Prostate-Specific Antigen Working Group. J Clin Oncol 17:3461–3467, 1999 Submitted April 22, 2002; accepted September 13, 2002.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2003 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
2 ng/mL, progression occurred at PSA more than 4 ng/mL; for patients with maximum pretreatment PSA more than 2 ng/mL, progression occurred at PSA more than 100% above the maximum pretreatment value. Time to progression was defined as the time from the start of therapy to the first PSA meeting criteria for disease progression or the development of objective disease progression, whichever came first. Treatment continued until patients had evidence of disease progression by PSA criteria or had development of distant metastases or until unacceptable toxicity. Patients had a repeat bone scan and CT scan of the abdomen and pelvis at the time of disease progression by PSA criteria and as clinically indicated. 
2 statistic for categorical variables (eg, Gleason score [GS]) and the nonparametric Mann-Whitney test for continuous variables (eg, nadir PSA) were used for analysis of the comparability of subsets. The Wilcoxon matched pairs test was used to compare the distributions for the pre- and posttreatment doubling times. 
