Genitourinary Oncology Service, Sidney Kimmel Center for Prostate and Urologic Cancers, Department of Medicine, Memorial Sloan-Kettering Cancer Center, New York, New York
Department of Medicine, Weill Cornell Medical College, New York, New York
Corresponding author: Michael J. Morris, MD, Genitourinary Oncology Service, Department of Medicine, Memorial Sloan-Kettering Cancer Center, 1275 York Ave, New York, NY 10065; Fax: (212) 988-0701; firstname.lastname@example.org
We thank Margaret McPartland for her editorial support.
β-emitting bone-seeking radiopharmaceuticals have historically been administered for pain palliation whereas docetaxel prolongs life in patients with metastatic castration-resistant prostate cancer (mCRPC). In combination, these agents simultaneously target the bone stroma and cancer cell to optimize antitumor effects. The toxicity and efficacy when each agent is combined at full, recommended doses, in a repetitive fashion is not well established.
Patients with progressive mCRPC and ≥3 bone lesions received 153Sm-EDTMP (samarium-153 ethylene diamine tetramethylene phosphonate) at a dose of 1.0 mCi/kg every 9 weeks and docetaxel at a dose of 75 mg/m2 every 3 weeks. In the absence of unacceptable toxicity, patients were allowed to continue additional cycles, defined by 9 weeks of treatment, until intolerance or biochemical/radiographic disease progression.
Of the 30 patients treated, approximately 50% were considered to be taxane-naive, 36.7% were taxane-refractory, and 13.3% had previously been exposed to taxanes but were not considered refractory. Patients received on average 2.5 cycles of treatment (6.5 doses of docetaxel and 2.5 doses of 153Sm-EDTMP). Twelve patients (40%) demonstrated a decline in their prostate-specific antigen level of ≥50%. The median progression-free survival (biochemical or radiographic) was 7.0 months and the overall survival was 14.3 months. Nine patients (30%) did not recover platelet counts >100 K/mm3 after a median of 3 cycles to allow for additional treatment, with 4 patients experiencing prolonged thrombocytopenia. The most common reasons for trial discontinuation were progressive disease and hematologic toxicity.
Metastatic castration-resistant prostate cancer (mCRPC) has a strong predilection for bone, and therefore strategies using bone-seeking radiopharmaceuticals that target the bone microenvironment are scientifically relevant. However, these agents are likely clinically underutilized. Concerns for myelosuppression, lack of a demonstrated survival benefit of β-emitters, and the need for collaboration with nuclear medicine have all contributed to their limited use. Furthermore, although β-emitting radiopharmaceuticals have long been approved for the treatment of bone pain, they have not been extensively investigated as antitumor agents either alone or in combination with other agents.
Combining a radiopharmaceutical such as 153Sm-EDTMP (samarium-153 ethylene diamine tetramethylene phosphonate), which is known to palliate bone pain, with docetaxel, an agent known to prolong survival in patients with bone and soft tissue/visceral mCRPC, may enhance chemotherapeutic response and improve clinical outcomes. Each agent may offer independent effects that in combination provide additive benefit to the patient.
Previous trials have systematically combined these agents, but typically have limited either the dose or number of cycles of the β radiopharmaceutical. For such reasons, we previously conducted a dose escalation trial exploring a repetitively dosed regimen of 153Sm-EDTMP and docetaxel. 153Sm-EDTMP was given every 6 to 9 weeks and docetaxel was given every 3 weeks in 28 patients with mCRPC. The maximum tolerated dose, based on cycle 1 toxicity, was not reached in this study, thereby providing the rationale for using the full US Food and Drug Administration (FDA)-approved doses of each agent (1 millicurie [mCi]/kg of 153Sm-EDTMP and 75 mg/m2 of docetaxel) as ongoing therapy. To facilitate our understanding of the toxicity profile, explore preliminary efficacy and duration of effect, and create a reference database by which combinations of α-emitting radiopharmaceuticals and docetaxel could be compared, herein we present data regarding the 30 patients treated at full doses of docetaxel and 153Sm-EDTMP. These patients were treated repetitively until they experienced toxicity or disease progression, through a preplanned expansion cohort.
MATERIALS AND METHODS
Patients were treated at full FDA-approved doses of docetaxel (75 mg/m2 every 3 weeks) and 153Sm-EDTMP (1.0 mCi/kg every 9 weeks). The protocol was approved by the Memorial Sloan-Kettering Cancer Center Institutional Review Board and all patients provided written informed consent.
Eligible patients had histologic evidence of prostate adenocarcinoma, castrate levels of testosterone (<50 ng/dL), and biochemical or radiographic disease progression as previously described. All patients had to have at least 3 bone metastases noted on bone scintigraphy. Visceral metastases or symptomatic lymphadenopathy were allowed provided there was also bone disease.
Adequate hematologic function was defined as a white blood cell count ≥3.0 K/mm3, an absolute neutrophil count (ANC) ≥1.5 K/mm3, a platelet count ≥100 K/mm3, and hemoglobin ≥10 mg/dL. Adequate hepatic function required bilirubin levels of less than the upper limit of normal and alanine aminotransferase and aspartate aminotransferase levels ≤1.5 times the upper limit of normal. Adequate renal function was defined as serum creatinine ≤2.0 mg/dL. A Karnofsky performance status ≥60 was required.
Patients who received >1 course of external beam radiation targeted to bone, had circumferential or transcortical pathologic long bone fractures, or had spinal cord compression were excluded.
Cycles consisted of 153Sm-EDTMP administered on day 1 of each cycle, with docetaxel administered on days 1, 22, and 43, as seen in Figure 1. All patients received prednisone at a dose of 5 mg twice daily throughout the cycle, and dexamethasone at a dose of 4 mg twice daily on the day before, day of, and day after docetaxel treatment. On day 1, patients were treated with 1.0 mCi/kg of 153Sm-EDTMP followed by 75 mg/m2 of docetaxel between 6 and 30 hours later. Docetaxel was administered again on days 22 and 43 in the absence of grade 3 or 4 toxicity. Treatment was permitted to be held for 1 week for toxicity recovery, with retreatment permitted in the absence of grade 3 or 4 toxicity. Patients were eligible to receive additional cycles in the presence of grade 0 to 1 neutropenia, a platelet count ≥100 K/mm3, or ≤grade 2 nonhematologic toxicity. Recycling was allowed to be held for up to 2 weeks to allow for recovery of toxicity to permitted levels.
Patients were tested for hematologic parameters on day 1 and every 7 days thereafter. Prophylactic granulocyte–colony-stimulating factor (GCSF) was added after each docetaxel treatment for persistent grade 4 neutropenia lasting ≥1 week, or in the event of neutropenic fever. A dose reduction to 60 mg/m2 of docetaxel and 0.5 mCi/kg2 of 153Sm-EDTMP was allowed for grade 4 thrombocytopenia. For other serious nonhematologic toxicity, dose reductions of both drugs by 25% for a grade 2 toxicity and 50% for a grade 3 toxicity were allowed.
The primary endpoint was safety with chronic dosing. Dose-limiting toxicity (DLT) for the first 6 patients, as prespecified by the protocol, was defined as any neutropenic fever, ≥grade 3 nonhematologic toxicity, grade 3 neutropenia that lasted for ≥7 days despite growth factor support, ≥grade 3 anemia or thrombocytopenia that lasted for ≥5 days, or failure of count recovery to allow for therapy in cycle 1. Any DLT occurring in the first 6 patients was considered grounds for discontinuation. For the subsequent 24 patients, the definition of a DLT was specifically liberalized to test the tolerance of the regimen when administered repetitively. Therefore, DLT was not defined by cycle 1 toxicity but by failure to meet criteria to allow for recycling as described above in the treatment section. Details of toxicities and adverse events were assessed according to the Common Terminology Criteria for Adverse Events (version 3.0) for all patients at all cycles. Complete blood counts were performed weekly, and physical examination, vital signs, serum chemistry, and adverse event inquiries were performed once every 3 weeks. All patients who received at least 1 dose of study drug were included in the analysis.
Progression-free survival (PFS) was defined as the time between the start of treatment and the first documented disease progression or death. Disease progression was indicated by meeting at least 1 of 3 criteria: biochemical disease progression, soft-tissue progression, or a combination of a new bone lesion(s) with a rising prostate-specific antigen (PSA) or worsening bone pain. PSA was drawn at baseline and every 3 weeks. Biochemical disease progression was defined as a 50% increase in PSA over nadir in patients who had experienced a ≥50% decline in PSA on study. In patients with a <50% decline in PSA, disease progression was defined as a ≥25% increase over nadir or baseline, whichever was lower.
Radiographic assessments were performed at baseline and every 12 weeks. Soft-tissue response was assessed by Response Evaluation Criteria in Solid Tumors (RECIST) (version 1.0). The appearance of ≥1 new bone lesions in the setting of a rising PSA or worsening bone pain (based on clinical judgment) was also considered disease progression.
Bone marker analyses
Indices of bone deposition (bone-specific alkaline phosphatase, osteocalcin) and bone resorption (serum and urine N-telopeptide) were assessed every 3 weeks.
The dose escalations in this trial were conducted based on the probabilistic calculation that if the true risk of toxicity was 0.10, 0.20, 0.30, 0.40, 0.50, and 0.60 the probability of escalation was therefore 0.91, 0.71, 0.49, 0.31, 0.17, and 0.08, respectively. It was on this basis that doses were escalated to the full FDA-approved doses of each drug, as no maximum tolerated dose was reached on the basis of cycle 1 toxicity as previously reported. The last 6 patients of the dose escalation cohort were included in the expansion cohort analyses, in which 24 additional patients were treated. All 30 patients were treated until disease progression or toxicity, to confirm that the dose selected based on the above probabilities was in fact clinically safe. PFS and overall survival (OS) were computed using the Kaplan-Meier estimator.
A total of 31 patients were enrolled. One patient developed spinal cord compression after registration but before treatment and was excluded from the analysis. The first 6 patients were part of the published dose escalation portion of this trial, and were included in the current study because they received the same dose/schedule as the remaining patients reported herein (part of the expansion cohort of this same trial). Table 1 provides baseline patient characteristics. Fifteen patients (50%) were taxane-naive, 11 patients (36.7%) were taxane-refractory, and 4 patients (13.3%) had previous exposure to taxanes but were not considered refractory at the time of discontinuation of docetaxel because they were responding to treatment. For the purposes of the current study, taxane refractoriness was defined as definitive biochemical and/or radiographic disease progression while receiving taxane therapy. No patient received <3 cycles of docetaxel (225 mg/m2) who met the study definition of being taxane-refractory. Cabazitaxel was not approved at the time the current study was conducted and therefore a more contemporary definition of taxane refractoriness (ie, refractoriness in both the first-line and second-line setting) was not implemented.
Table 1. Patient Characteristics at Baseline (N = 30)
Patients received a median of 2.5 doses of 153Sm-EDTMP (range, 1 dose-10 doses) and 6.5 doses of docetaxel (range, 1 dose-32 doses) (Fig. 2 Top and Bottom). Taxane-naive patients (n = 15 patients) received a median cumulative exposure of 3.0 mCi/kg of 153Sm-EDTMP over 3 doses (range, 1 dose-10 doses) and 675 mg/m2 of docetaxel over 9 doses (range, 1 dose-29 doses). The 4 patients who were exposed to taxane but did not have taxane-refractory disease received a median cumulative exposure of 3.0 mCi/kg of 153Sm-EDTMP over 3 doses (range, 1 dose-6 doses) and 600 mg/m2 of docetaxel over 8 doses (range, 2 doses-18 doses). Taxane-refractory patients (n = 11 patients) received a median cumulative exposure of 2.0 mCi/kg of 153Sm-EDTMP over 2 doses (range, 1 dose-4 doses), and 375 mg/m2 of docetaxel over 5 doses (range, 1 dose-12 doses). One patient in the taxane-exposed group had the dose of both drugs reduced by 25% mid-cycle because of grade 2 fatigue. No other dose reductions occurred.
Ten patients failed to recycle on the basis of hematologic toxicity (DLT). Thrombocytopenia was the most common hematologic reason (9 patients) and 1 patient experienced neutropenic fever during the first cycle. Disease progression accounted for discontinuation of treatment in 10 patients. The remainder of the patients came off the study for comorbid conditions, patient request for a treatment break, or closure of the trial.
Tables 2 and 3 report adverse events from cycle 1 and all cycles. Neutropenia and thrombocytopenia were the predominant toxicities, as reviewed below and in Table 4.
Abbreviations: ALT, alanine aminotransferase; INR, International Normalized Ratio.
Adverse events were graded according to version 3.0 of the National Cancer Institute Common Terminology Criteria for Adverse Events. Events ≤ grade 2 were reported if they occurred in ≥ 25% of patients, unless a grade 3 or 4 toxicity was noted. All grade 3 or 4 events were included.
Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; INR, International Normalized Ratio; NOS, not otherwise specified.
Adverse events were graded according to version 3.0 of the National Cancer Institute Common Terminology Criteria for Adverse Events. Events of ≤ grade 2 were reported if they occurred in ≥ 25% of patients, unless a grade 3 or 4 toxicity was noted.
Median days to ANC recovery of patients with nadir ANC ≥ grade 3 (range)
No. of patients who did not recover ANC to allow recycling
No. of patients with nadir hemoglobin ≥grade 3
Grade 3 to 4 neutropenia occurred in 96% of patients. All patients recovered their ANC to allow for recycling (Table 4). One patient experienced an uncomplicated neutropenic fever and recovered without sequelae but was taken off the study because the protocol considered this to be grounds for discontinuation in the first 6 patients. Six other patients experienced a neutropenic fever, recovered their counts in time for recycling, and were permitted to stay on the study as per protocol. A total of 18 patients received GCSF prophylaxis.
Eight patients (27%) experienced grade 3 to 4 thrombocytopenia. Five were taxane-naive, 1 was previously exposed to taxanes, and 2 patients were taxane-refractory. The median platelet nadir in cycle 1 was 97 K/mm3 (range, 39 K/mm3−172 K/mm3) and the all-cycle median nadir was 64 K/mm3 (range, 19 K/mm3−172 K/mm3). Nine patients (30%) did not recover platelet counts that were sufficient to allow for recycling: 2 occurred in cycle 1, 2 occurred in cycle 2, 3 occurred in cycle 3, 1 occurred in cycle 5, and 1 occurred in cycle 6. Four of the 9 patients who did not recover their platelet count to allow for recycling (13% of the total) experienced prolonged thrombocytopenia (87 days-177 days) that failed to recover to 100 K/mm3 after 1 or 2 cycles. One of these patients was taxane-naive, 1 had received prior taxane exposure, and 2 were taxane-refractory. Three of these 4 patients were found to have new lesions on bone scintigraphy with the onset of thrombocytopenia and went on to receive external beam radiation to bone. The fourth patient did not have a bone scan performed; magnetic resonance imaging demonstrated stable disease, but the patient's pain increased.
Best-response PSA waterfall plots are presented in Figure 3. Overall, 24 patients (80%) had a decline in PSA, 12 of whom (40%) had a decline of at least 50%. Of the 10 patients with measurable soft-tissue disease, 2 (20%) achieved a partial response and 4 (40%) had stable disease.
The median PFS was 7.0 months (95% confidence interval [95% CI], 3.7 months-8.8 months) and the overall survival was 14.3 months (95% CI, 10.3 months-24.0 months) (Fig. 4). The longest PFS and OS was seen in the taxane-naive group, with a median PFS of 8.2 months (95% CI, 5.7 months-13.7 months) and a median OS of 23.4 months (95% CI, 13.8 months-42.7 months). Taxane-exposed patients had a median PFS of 7.0 months (95% CI, 3.2 months-11.5 months) and median OS of 19.4 months (95% CI, 5.6 months-30.3 months). The taxane-refractory group had a median PFS of 3.5 months (95% CI, 0.7 months-8.0 months) and a median OS of 9.6 months (95% CI, 4.8 months-13.5 months).
Eighteen patients (60%) received zoledronic acid during the trial and 1 patient (3%) received alendronate. The median percentage change from baseline to week 3 was −14.1% for urinary N-telopeptide, −17.6% for serum N-telopeptide, −82.1% for osteocalcin, and −6.25% for bone-specific alkaline phosphatase (Fig. 5).
β-emitting radiopharmaceuticals have long been a part of the armamentarium for treating bone pain in patients with mCRPC. However, there is growing evidence to support the development of bone-seeking radiopharmaceuticals as antitumor agents to confer benefit beyond palliation. The most definitive evidence has arisen from the large, randomized phase 3 study of the α-emitter radium-223, which demonstrated a 2.8-month prolongation in survival compared with placebo. Other strategies aimed at developing radiopharmaceuticals as antitumor agents have used a combination of β-emitting radiopharmaceuticals with chemotherapy known to prolong survival. However, these earlier phase studies have typically limited the dose and number of cycles of the β-emitting radiopharmaceutical used, primarily out of concern for hematologic toxicity.[5-8]
The current study defines the hematologic toxicity profile in what to our knowledge is the largest published cohort to date using repetitive, full dosing of each agent and in combination. Neutropenia rarely limited therapy and returned to acceptable levels at a median of 7 days from nadir, although 60% of patients received GCSF. Neutropenic fever was noted in 23% of patients and did not recur after GCSF prophylaxis; all patients recovered ANC in time for recycling.
Thrombocytopenia was a more significant limitation because recycling was precluded in 9 patients (30%) after a median of 3 cycles. Four patients did not recover their platelet count to >100 K/mm3 after a median of 1.5 cycles. These findings are consistent with those of Tu et al, in which 28% of patients experienced grade 3 hematologic toxicity after 2 cycles of combination therapy. In the current study, 3 of the 4 patients who never recovered a platelet count >100 K/mm3 had signs of bone-based radiographic disease progression, suggesting disease status as a possible contributor. In another study, bone marrow failure occurred in 9 of 43 responders to induction with weekly docetaxel at a dose of 20 mg/m2 and estramustine who then underwent consolidation with single-dose 153Sm-EDTMP and continued weekly docetaxel therapy. Involvement of prostate cancer in the bone marrow was the most common finding in these patients. Bone marrow involvement is common in patients with terminal stages of CRPC, making the distinction between drug effect and progressive disease challenging when considering a heavily pretreated population.
The use of 153Sm-EDTMP did not appear to limit the dose of docetaxel administered in this trial when compared with reference chemotherapy trials in patients with mCRPC.[9, 10] Chemotherapy-naive patients received the same number of cycles of chemotherapy with both drugs as patients receiving docetaxel alone in previous trials. Even patients who were previously treated with docetaxel (but were not refractory) received only 1 fewer cycle of docetaxel as chemotherapy-naive patients. Therefore, the addition of 153Sm-EDTMP did not limit the duration of treatment or total chemotherapy exposure of docetaxel, and perhaps allowed for additional treatment by enhancing response in chemotherapy-exposed or chemotherapy-resistant patients.
Preliminary efficacy was demonstrated in this trial in both chemotherapy-naive and previously treated patients. Overall, 40% of patients experienced a PSA decline ≥50%. It appears that the use of a radiopharmaceutical in the current study was able to resensitize previously refractory patients to the effects of docetaxel.
The future use of bone-seeking radiopharmaceuticals will likely emphasize α-emitting rather than β-emitting radiopharmaceuticals. α-Emitting radiopharmaceuticals such as radium-223 deliver energy with less tissue penetration than β-emitters, and subsequently have less hematologic toxicity. Radium-223 was developed as an antitumor agent, and has demonstrated a survival advantage. Therefore, we will not pursue 153Sm-EDTMP and docetaxel in efficacy-based studies, despite the tolerability and preliminary promising efficacy data noted in the current study. Rather, the design, safety, and efficacy data from the current study of 153Sm-EDTMP and docetaxel has resulted in a currently ongoing multicenter study combining radium-223 and docetaxel (NCT01106352). The rationale underlying that trial is that the combination involves 2 known life-prolonging therapies, each with a favorable toxicity profile when given as monotherapy that can target both tumor and bone. To the best of our knowledge, the current study is among the largest safety and efficacy comparison data sets for interpreting the results of trials combining chemotherapy with α-emitting radiopharmaceuticals.
Full doses of 153Sm-EDTMP and docetaxel can be given together as ongoing therapy in chemotherapy-naive, chemotherapy-exposed, and chemotherapy-refractory patients. The results of the current study provide encouraging preliminary efficacy to support the development of combination strategies of bone-seeking radiopharmaceuticals with chemotherapy. Although the hematologic toxicity noted in the current study was manageable, the anticipated availability of radium-223, which has been shown to be minimally toxic to the bone marrow, provides even greater enthusiasm for combination regimens targeting both the tumor and the bone stroma.
Funded in part by EUSA Inc and National Cancer Institute grant CA102544
CONFLICT OF INTEREST DISCLOSURES
Dr. Scher has acted as a consultant for Amgen, Bristol-Meyers Squibb, Dendreon, Endo Pharmaceuticals, Millennium Pharmaceuticals, Novartis, Ortho Biotech Oncology Research, Sanofi-Aventis, and Senior Scientific LLC. Dr. Morris has acted as a consultant for Millennium Pharmaceuticals, Bayer Inc., Cytogen Inc, and Sanofi-Aventis.