Long-term efficacy and safety of zoledronic acid compared with pamidronate disodium in the treatment of skeletal complications in patients with advanced multiple myeloma or breast carcinoma

A randomized, double-blind, multicenter, comparative trial

Abstract

BACKGROUND

The goal of the current study was to compare the long-term (25-month) safety and efficacy of zoledronic acid with pamidronate in patients with bone lesions secondary to advanced breast carcinoma or multiple myeloma.

METHODS

Patients (n = 1648) were randomized to receive 4 mg or 8 mg (reduced to 4 mg) zoledronic acid as a 15-minute infusion or to receive 90 mg pamidronate as a 2-hour infusion every 3–4 weeks for 24 months. The primary endpoint was the proportion of patients with at least 1 skeletal-related event (SRE), defined as pathologic fracture, spinal cord compression, radiation therapy, or surgery to bone. Secondary analyses included time to first SRE, skeletal morbidity rate, and multiple-event analysis. Hypercalcemia of malignancy (HCM) was included as an SRE in some secondary analyses.

RESULTS

After 25 months of follow-up, zoledronic acid reduced the overall proportion of patients with an SRE and reduced the skeletal morbidity rate similar to pamidronate. Compared with pamidronate, zoledronic acid (4 mg) reduced the overall risk of developing skeletal complications (including HCM) by an additional 16% (P = 0.030). In patients with breast carcinoma, zoledronic acid (4 mg) was significantly more effective than pamidronate, reducing the risk of SREs by an additional 20% (P = 0.025) compared with pamidronate and by an additional 30% in patients receiving hormonal therapy (P = 0.009). Zoledronic acid (4 mg) and pamidronate were tolerated equally well. The most common adverse events included bone pain, nausea, and fatigue.

CONCLUSIONS

Long-term follow-up data confirm that zoledronic acid was more effective than pamidronate in reducing the risk of skeletal complications in patients with bone metastases from breast carcinoma and was of similar efficacy in patients with multiple myeloma. Cancer 2003. © 2003 American Cancer Society.

DOI 10.1002/cncr.11701

Every year, almost 800,000 women throughout the world develop breast carcinoma,1 and approximately 200,000 breast cancer diagnoses occur annually in the United States alone. Metastatic bone lesions are common during disease progression and affect an estimated 65–75% of patients with breast carcinoma.1–3 Bone lesions also occur in virtually all patients with advanced-stage multiple myeloma, which afflicts almost 60,000 people globally each year.4 Bone lesions can result in substantial morbidity, including severe bone pain and debilitating complications such as pathologic fractures, spinal cord compression, and potentially life-threatening hypercalcemia of malignancy (HCM).3, 5 Long-bone and hip fractures are extremely painful, often require orthopedic surgery, and can result in impaired mobility and function. In addition, vertebral collapse and subsequent spinal cord compression may lead to paralysis.3, 6–8 The median survival of patients with multiple myeloma, which colonizes the bone marrow, may reach 6–7 years,9 and the median survival for patients with breast carcinoma is approximately 18–26 months after the initial diagnosis of bone metastases.10, 11 Therefore, patients with bone lesions from breast carcinoma or multiple myeloma are at long-term risk of developing skeletal complications. Preventing or delaying the occurrence of these skeletal complications would provide important clinical benefits for these patients.

Before the introduction of zoledronic acid, patients with radiologically documented osteolytic bone lesions secondary to breast carcinoma or multiple myeloma generally were treated with intravenous pamidronate (90 mg), which significantly delayed the onset and reduced the incidence of skeletal complications in randomized, placebo-controlled trials in these patient populations.12–16 However, pamidronate required a lengthy (2-hour or 4-hour) infusion every 3–4 weeks.

Zoledronic acid is a highly potent, new-generation, nitrogen-containing bisphosphonate that has demonstrated greater potency compared with pamidronate in preclinical testing and can be administered safely as a 15-minute infusion.17, 18 Zoledronic acid has demonstrated superiority over pamidronate in the treatment of HCM19 and is the first bisphosphonate to demonstrate efficacy in patients with bone metastases from solid tumors other than breast carcinoma, including prostate carcinoma, nonsmall cell lung carcinoma, and a variety of other tumor types.20–22

The 13-month efficacy and safety data from a randomized, Phase III, multicenter trial of zoledronic acid compared with pamidronate in patients with bone lesions secondary to breast carcinoma or multiple myeloma were reported previously.23 At the end of the core phase of the trial (13 months), zoledronic acid demonstrated efficacy and safety comparable with pamidronate. Herein, we report the 25-month final analysis of that trial.23

MATERIALS AND METHODS

Patients

The study enrolled adult patients with at least 1 osteolytic bone lesion secondary to Durie–Salmon Stage III multiple myeloma or at least 1 bone metastasis (osteolytic, osteoblastic, or mixed) secondary to Stage IV breast carcinoma, Eastern Cooperative Oncology Group (ECOG) performance status ≤ 2, corrected serum calcium < 12 mg/dL, serum creatinine ≤ 3 mg/dL (265 μmol/L), and serum bilirubin ≤ 2.5 mg/dL (43 μmol/L). Patients were not eligible if they had received prior treatment with bisphosphonates within 12 months of the screening visit, were pregnant or lactating, or had a history of noncompliance with medical regimens. Ongoing antineoplastic therapy or hormonal therapy was required for all patients at the time of randomization. Hormonal therapy was first-line or second-line therapy; patients receiving combination antineoplastic and hormonal therapy already may have received two or more lines of therapy. Antineoplastic therapy could be changed or discontinued during the study at the discretion of the treating physician. All patients provided written informed consent.

Treatment

Patients were randomized to treatment with zoledronic acid (4 mg or 8 mg) or pamidronate (90 mg) as an infusion in conjunction with standard antineoplastic therapy. Infusions were administered every 3–4 weeks for 24 months. Double-dummy infusions were used to maintain the double-blind nature of the study. For the duration of the study, all patients received a daily 500-mg calcium supplement and 400–500 International Units of vitamin D. Pamidronate was infused over 2 hours in 250 mL of hydration solution (0.9% sodium chloride). Zoledronic acid initially was administered as a 5-minute infusion in 50 mL of hydration solution. However, because of concerns over renal safety, two protocol amendments were introduced. The infusion duration was amended to 15 minutes, and the infusate volume was increased to 100 mL. Subsequently, after all patients were enrolled, the 8-mg zoledronic acid dose was reduced to 4 mg (henceforth referred to as the 8/4 mg zoledronic acid treatment group), and serum creatinine was monitored before each dose of study drug.

Study Design and Schedule

This was an international, multicenter, stratified, randomized, double-blind, double-dummy, parallel-group study. Patients were divided into three strata based on their primary cancers and ongoing therapies at the time of randomization: 1) patients with multiple myeloma; 2) patients with breast carcinoma receiving chemotherapy (with or without hormonal therapy); and 3) patients with breast carcinoma receiving hormonal therapy (first-line or second-line). Patients were randomized using an automated system, with only the pharmacist at each site aware of the assigned treatment.

Patient evaluations at baseline and during the core phase of this study have been described previously.23 During the extension phase of this trial, ECOG performance status was assessed, and complete and differential blood counts were performed at 15 months, 18 months, 21 months, and 25 months. Patients who discontinued study treatment received follow-up assessments according to the protocol on the same schedule as patients who continued treatment. Urinalyses were performed at 18 months and 25 months. Serum chemistry analyses, interim physical examinations, and assessment of skeletal complications were performed at each visit before the infusion of study medication.

Efficacy and Safety Assessment

The primary objective of the 25-month extension phase was to determine the safety and efficacy of long-term therapy with zoledronic acid compared with pamidronate. Adverse events, serious adverse events (SAEs), and laboratory data were summarized by each treatment group. An SAE was defined as any fatal or life-threatening event; any event that required prolonged hospitalization; any event that was significantly or permanently disabling or incapacitating; or any event that required medical or surgical intervention to prevent death, disability, or incapacitation. Deterioration in renal function was defined as serum creatinine levels that increased by ≥ 0.5 mg/dL or 1.0 mg/dL from baseline for patients with baseline serum creatinine < 1.4 mg/dL or > 1.4 mg/dL, respectively, or that became ≥ 2 times the baseline value. All treatment groups were included in the safety analysis.

The primary efficacy endpoint was the proportion (percentage) of patients in each treatment group who experienced at least 1 skeletal-related event (SRE) during the 25-month study period. SREs were defined as pathologic fracture, spinal cord compression, radiation therapy to bone, and surgery to bone. The specifications for diagnosing each of these events were described previously.23 For the primary efficacy analysis, HCM was not included in the definition of SREs, because zoledronic acid already has demonstrated efficacy in treating HCM.24 However, HCM is a clinically important event that can be life-threatening; therefore, some secondary efficacy analyses included HCM, defined as corrected serum calcium ≥ 12 mg/dL. Secondary efficacy endpoints included the proportion of patients experiencing any SRE, the proportion of patients experiencing each type of SRE, time to first SRE, time to each type of SRE, skeletal morbidity rate, multiple-event analysis, overall survival, and ECOG performance status changes.

The protocol amendment to assure renal safety in all patients caused heterogeneity in the treatment of patients in the 8/4 mg zoledronic acid group. Therefore, although data for this group are reported, no conclusions were drawn.

Statistical Analysis

The statistical design of this noninferiority trial was published with the core data.23 The proportion of patients with at least 1 SRE was compared among treatment groups using the Cochran-Mantel-Haenszel test. The Cochran-Mantel-Haenszel test was used to compare secondary efficacy variables, including skeletal morbidity rate (defined as the number of SREs per year) and change from baseline in ECOG performance status between treatment groups. The Kaplan-Meier method and the stratified log-rank test were used to compare time to first SRE, time to progression of bone lesions, time to overall disease progression, time to first serum creatinine increase, and overall survival among treatment groups. A preplanned multiple-event analysis was performed using the Andersen-Gill approach, and the robust estimate of variance was used to calculate P values.25 For the skeletal morbidity rate calculations and multiple-event analysis, a 21-day event window was used for counting SREs, such that any event occurring within 21 days of a previous event was not counted. The 21-day event window and robust estimate of variance were used to minimize the impact of correlations between events that occurred closely in time or in clusters.

In addition to the overall patient population and the patient strata outlined in the study design (patients with multiple myeloma, patients with breast carcinoma receiving chemotherapy, and patients with breast carcinoma receiving hormonal therapy), exploratory subset analyses were performed on the population of all patients with breast carcinoma in this trial by combining the data for the two breast carcinoma strata (patients with breast carcinoma receiving hormonal therapy and patients with breast carcinoma receiving cytotoxic chemotherapy).

RESULTS

Patients

A total of 1648 patients with ≥ 1 bone lesion secondary to either Durie–Salmon Stage III multiple myeloma or Stage IV breast carcinoma were enrolled between October, 1998 and January, 2000. Table 1 shows that patients were randomized to receive treatment with 4 mg zoledronic acid (n = 564), 8/4 mg zoledronic acid (n = 526), or 90 mg pamidronate (n = 558). The intent-to-treat population excluded 8 patients (3 patients in the 4 mg zoledronic acid group, 2 patients in the 8/4 mg zoledronic acid group, and 3 patients in the pamidronate group) because they were treated at 1 study center that did not follow good clinical practice. Slightly less than 40% of the enrolled patients (n = 606) completed the 13-month core phase of this trial and extended their double-blind treatment for an additional 12 months (4 mg zoledronic acid: n = 212; 8/4 mg zoledronic acid: n = 189; 90 mg pamidronate: n = 205). More than 50% of patients who enrolled in the extension phase completed the full 25 months on study. Reasons for discontinuation are shown in Table 2.

Table 1. Patient Disposition by Treatment Group: Evaluable Patients

Patient disposition	No of patients (%)
	Zoledronic acid 4 mg	Zoledronic acid 8/4 mg	Pamidronate 90 mg	Total
Randomized	564 (100)	526 (100)	558 (100)	1648
Safety evaluable (core and extension study)	563 (99.8)	524 (99.6)	556 (99.6)	1643
Completed core study	353 (62.7)	313 (59.7)	338 (60.8)	1004
Entered extension study	212 (37.6)	189 (35.9)	205 (36.7)	606
Multiple myeloma stratum	73	56	65	194
Breast carcinoma	139	133	140	412
Completed extension study	123 (21.8)	105 (20.0)	123 (22.0)	351
Intent-to-treat (core and extension studies)	561	524	555	1640

Table 2. Reasons for Premature Discontinuation of Study Medications for Patients in the Extension Study: Safety-Evaluable Patients

Reason for discontinuation	No. of patients (%)
	Zoledronic acid 4 mg (n = 212)	Zoledronic acid 8/4 mg (n = 189)	Pamidronate 90 mg (n = 205)
Total discontinued	89 (42.0)	84 (44.4)	82 (40.0)
Patient withdrew consent	21 (9.9)	23 (12.2)	16 (7.8)
Death	21 (9.9)	13 (6.9)	21 (10.2)
Adverse event(s)	19 (9.0)	20 (10.6)	19 (9.3)
Abnormal laboratory value(s)	14 (6.6)	11 (5.8)	6 (2.9)
Unsatisfactory therapeutic effect	4 (1.9)	6 (3.2)	7 (3.4)
Protocol violation	3 (1.4)	2 (1.1)	4 (2.0)
Lost to follow-up	3 (1.4)	0 (0.0)	2 (1.0)
Administration problems	2 (0.9)	6 (3.2)	4 (2.0)
No longer required study drug	2 (0.9)	3 (1.6)	2 (1.0)
Abnormal procedure	0 (0.0)	0 (0.0)	1 (0.5)

The mean time from the initial cancer diagnosis to study entry was approximately 1.5 years and 6 years for patients with multiple myeloma and breast carcinoma, respectively. Treatment groups were well balanced with respect to all patient and baseline disease characteristics. The median age was approximately 60 years, and the mean bone pain index (BPI) composite pain score was 3.0 in all groups.

The primary disease characteristics and disease histories were similar within each of the study populations. In the overall population of patients with breast carcinoma, > 50% of patients had progressed after first-line therapy, and study medication was initiated an average of 4 months after the initial diagnosis of bone metastases (Table 3). Roughly 60% of the patients with breast carcinoma and > 80% of the patients with multiple myeloma already had experienced an SRE before study entry.

Table 3. Demographics for the Population of Patients with Breast Carcinoma: Safety-Evaluable Patients

Characteristic	No. of patients (%)
	Zoledronic acid 4 mg (n = 377)	Zoledronic acid 8/4 mg (n = 364)	Pamidronate 90 mg (n = 389)	Total (n = 1130)
SRE: skeletal-related event; SD: standard deviation. a One month equaled 28 days in the current study.
Treatment stratum
Chemotherapy	176 (47)	169 (46)	179 (46)	524 (46)
Hormonal therapy	201 (53)	195 (54)	210 (54)	606 (54)
Mean age (yrs)	58.0	57.5	57.0	57.5
First-line therapy
Yes	161 (43)	180 (49)	182 (47)	523 (47)
No	216 (57)	184 (51)	207 (53)	607 (54)
Previous SREs
Yes	232 (62)	207 (57)	244 (63)	683 (60)
No	145 (38)	157 (43)	145 (37)	447 (40)
Time from first diagnosis of bone metastases to study treatment (mos)a
Mean ± SD	17.5 ± 33.85	14.1 ± 22.87	12.6 ± 21.68	—
Median	4.0	4.4	3.6	—

Efficacy

The primary analysis of efficacy was based on the comparison of 4 mg zoledronic acid and 90 mg pamidronate. The primary endpoint was the proportion of patients with at least 1 SRE.

Proportion of patients with at least 1 SRE

The percentage of patients with an SRE at 25 months by stratum and treatment group (intent-to-treat analysis) is shown in Figure 1. The percentage of patients who experienced an SRE other than HCM was similar between treatment groups (47% for 4 mg zoledronic acid and 51% for pamidronate). The percentage of patients with each individual type of SRE consistently was lower in the 4 mg zoledronic acid treatment group than in the pamidronate group. This difference was significant for the percentage of patients in the overall population who required radiotherapy to bone (19% for 4 mg zoledronic acid vs. 24% for pamidronate; P = 0.037).

Figure 1. Proportion of patients with skeletal-related events. For the analysis in (A), hypercalcemia of malignancy was not included as a skeletal-related event. For the analysis in (B), hypercalcemia of malignancy was included as a skeletal-related event. Black bar: zoledronic acid 4 mg group; gray bar: zoledronic acid 8/4 mg group; white bar: pamidronate 90 mg group; SRE: skeletal-related event; BC-C: breast carcinoma with chemotherapy; BC-H: breast carcinoma with hormonal therapy.

Time to first SRE

The median time to first SRE (including HCM) was comparable in the 4 mg zoledronic acid and 90 mg pamidronate groups for the overall population (376 days vs. 356 days, respectively; P = 0.151), in the multiple myeloma stratum (380 days vs. 286 days, respectively; P = 0.538), and in the chemotherapy-treated breast carcinoma stratum (349 days for 4 mg zoledronic acid vs. 366 days for pamidronate; P = 0.826) (Fig. 2). However, in the hormonal therapy breast carcinoma stratum, 4 mg zoledronic acid significantly extended the time to first SRE (415 days vs. 370 days for pamidronate; P = 0.047).

Figure 2. Kaplan-Meier estimate of the median time to first skeletal-related event (including hypercalcemia of malignancy) in each treatment group.

Skeletal morbidity rate

In the overall patient population, zoledronic acid (4 mg) reduced the mean annual incidence of skeletal complications (skeletal morbidity rate) by 25% compared with pamidronate (1.04 events per year for 4 mg zoledronic acid vs. 1.39 events per year for pamidronate; P = 0.084). These reductions were significant in the hormonal therapy breast carcinoma stratum (0.83 SREs per year vs. 1.37 SREs per year, respectively; P = 0.039). The mean skeletal morbidity rate in the 4 mg zoledronic acid group consistently was lower compared with the pamidronate group in all patient strata and for all individual types of SREs (Table 4). These differences were significant in the overall patient population for radiation to bone (0.47 events per year for 4 mg zoledronic acid vs. 0.71 events per year for pamidronate; P = 0.015).

Table 4. Skeletal Morbidity Rate (Events per Year) by Treatment Group: Intent-to-Treat Analysisa

Treatment group	Zoledronic acid 4 mg (n = 561)	Zoledronic acid 8/4 mg (n = 524)	Pamidronate 90 mg (n = 555)
SRE: skeletal-related event; HCM: hypercalcemia of malignancy. a All entries are presented as mean ± standard deviation. b P ≤ 0.05 for 4 mg zoledronic acid versus 90 mg pamidronate.
All SREs (without HCM)	1.04 ± 3.95	0.99 ± 1.93	1.33 ± 4.29
All SREs (with HCM)	1.04 ± 3.95	1.00 ± 1.93	1.39 ± 4.38
Pathologic fracture	0.53 ± 0.91	0.54 ± 1.17	0.60 ± 1.12
Radiation to boneb	0.47 ± 3.83	0.45 ± 1.50	0.71 ± 4.11
Nonvertebral fracture	0.35 ± 0.74	0.33 ± 0.72	0.41 ± 0.90
Vertebral fracture	0.23 ± 0.57	0.25 ± 0.83	0.24 ± 0.66
Surgery to boneb	0.05 ± 0.28	0.04 ± 0.25	0.10 ± 0.54
Spinal cord compression	0.03 ± 0.21	0.04 ± 0.30	0.09 ± 0.80
Hypercalcemiab	0.02 ± 0.16	0.02 ± 0.20	0.07 ± 0.94

Multiple-event analysis (including HCM)

Treatment with 4 mg zoledronic acid reduced the risk of developing a skeletal complication by an additional 16% compared with pamidronate in the overall patient population (Fig. 3). The risk ratio derived from the multiple-event analysis was 0.841 (zoledronic acid vs. pamidronate; 95% confidence interval [95% CI], 0.719–0.983; P = 0.030). The risk of developing a skeletal complication was comparable in the 4 mg zoledronic acid and 90 mg pamidronate groups in the multiple myeloma stratum (risk ratio, 0.932; P = 0.593) and breast carcinoma chemotherapy stratum (risk ratio, 0.955; P = 0.749). However, patients in the breast carcinoma hormonal therapy stratum who received 4 mg zoledronic acid had an approximately 30% additional reduction in the risk of developing a skeletal complication compared with the pamidronate group (zoledronic acid vs. pamidronate; risk ratio, 0.693; 95% CI, 0.527–0.911; P = 0.009).

Figure 3. Relative risk ratios for skeletal-related events (including hypercalcemia of malignancy) in patients treated with 4 mg zoledronic acid versus those treated with 90 mg pamidronate.

The 8/4 mg zoledronic acid treatment group

In the 8/4 mg zoledronic acid group, the proportion of patients in all patient strata with an SRE (with and without HCM) was 49%. Compared with the pamidronate group, fewer patients in this group required surgery to bone (4% for 8/4 mg zoledronic acid vs. 6% for pamidronate; P = 0.049) or developed HCM (1% for 8/4 mg zoledronic acid vs. 3% for pamidronate; P = 0.025). The overall population of patients in the 8/4 mg group had a median time to first skeletal complication (including HCM) of 351 days, which was not significantly different from the pamidronate group (356 days). However, the 8/4 mg zoledronic acid group had a 28% lower mean annual incidence of skeletal events (including HCM) compared with the pamidronate group (1.00 vs. 1.39 events per year; P = 0.227), and multiple-event analysis showed a trend toward a reduction in the risk of developing skeletal complications (including HCM) in the overall population for the 8/4 mg treatment group compared with pamidronate (risk ratio, 0.854; 95% CI, 0.728–1.001; P = 0.052).

Subset analysis in patients with breast carcinoma

The proportion of all patients with breast carcinoma who experienced at least 1 SRE (not including HCM) in the 4 mg zoledronic acid and pamidronate treatment groups was similar (46% vs. 49%, respectively). The skeletal morbidity rate was reduced by 40% in patients with breast carcinoma who were treated with 4 mg zoledronic acid compared with pamidronate (0.9 vs. 1.49 events per year, respectively; P = 0.125). This difference was even greater when HCM was included as an SRE in the assessment of the skeletal morbidity rate (0.91 vs. 1.57 events per year, respectively; P = 0.102), although statistical significance was not reached. However, multiple-event analysis showed that, among all breast carcinoma patients, 4 mg zoledronic acid significantly reduced the risk of developing any skeletal complications (including HCM) by an additional 20% compared with pamidronate (risk ratio, 0.799; 95% CI, 0.657–0.972; P = 0.025) (Fig. 3).

Survival and performance status

Consistent with previous reports of life expectancy in patients with advanced breast carcinoma and multiple myeloma,10 patients in the current study had a median survival > 2 years. The Kaplan–Meier estimates of median survival were comparable among all treatment groups but differed substantially for patients in each stratum. The median survival was longest for patients with multiple myeloma (> 30 months for all treatment groups). Patients in the hormonal therapy breast carcinoma stratum had a median survival > 2 years. The median survival was shortest among patients in the chemotherapy breast carcinoma stratum (range, 18.7–22.3 months).

The mean ECOG performance status scores for patients in all 3 treatment groups were significantly increased compared with baseline between 15 months and 25 months in each of the patient strata. However, there were no significant differences in the mean changes from baseline between any of the treatment groups.

Safety

Although no more than 58% of the patients enrolled in the extension study completed the full course of therapy, adverse events were responsible for < 10% of patient discontinuations (Table 2). The most common reasons for discontinuation were the withdrawal of consent and death.

The most frequently reported adverse events in all treatment groups after 25 months of therapy (regardless of relation to study drug) were bone pain, nausea, fatigue, pyrexia, and emesis (Table 5). Similar to other bisphosphonate studies, the most frequent adverse events reported as drug-related were infections, arthralgia/myalgias, cytopenias, fever, eye disorders, electrolyte abnormalities, and injection site reactions. During the extension phase of this study, no unexpected toxicities or adverse drug interactions emerged. Most adverse events were mild to moderate in severity and were managed easily with supportive care. The incidence of SAEs was similar in all treatment groups, with malignant neoplasm aggravated, pyrexia, bone pain, anemia, and pneumonia reported most frequently.

Table 5. Reported Adverse Events by Treatment Group (≥ 15% Frequency for Any Group) Regardless of Study Drug Relation

Adverse event	No. of patients (%)
	Zoledronic acid 4 mg (n = 563)	Zoledronic acid 8/4 mg (n = 524)	Pamidronate 90 mg (n = 556)
Any event	556 (98.8)	521 (99.4)	548 (98.6)
Bone pain	325 (57.7)	295 (56.3)	316 (56.8)
Nausea	270 (48.0)	265 (50.6)	266 (47.8)
Fatigue	241 (42.8)	209 (39.9)	240 (43.2)
Pyrexia	213 (37.8)	195 (37.2)	172 (30.9)
Emesis	187 (33.2)	178 (34.0)	183 (32.9)
Anemia	181 (32.1)	183 (34.9)	175 (31.5)
Diarrhea	163 (29.0)	157 (30.0)	162 (29.1)
Dyspnea	158 (28.1)	129 (24.6)	155 (27.9)
Myalgia	153 (27.2)	133 (25.4)	143 (25.7)
Constipation	150 (26.6)	145 (27.7)	162 (29.1)
Cough	150 (26.6)	116 (22.1)	129 (23.2)
Arthralgia	148 (26.3)	119 (22.7)	131 (23.6)
Weakness	127 (22.6)	100 (19.1)	108 (19.4)
Headache	123 (21.8)	123 (23.5)	149 (26.8)
Anorexia	120 (21.3)	97 (18.5)	81 (14.6)
Aggravation of tumor	114 (20.2)	93 (17.7)	97 (17.4)
Edema, lower limb	111 (19.7)	92 (17.6)	126 (22.7)
Pain in limb	102 (18.1)	82 (15.6)	84 (15.1)
Alopecia	99 (17.6)	86 (16.4)	80 (14.4)
Paraesthesia	98 (17.4)	76 (14.5)	85 (15.3)
Back pain	97 (17.2)	102 (19.5)	106 (19.1)
Insomnia	97 (17.2)	90 (17.2)	111 (20.0)
Dizziness (except vertigo)	97 (17.2)	82 (15.6)	91 (16.4)
Neutropenia	92 (16.3)	84 (16.0)	83 (14.9)
Depression	90 (16.0)	78 (14.9)	95 (17.1)
Weight decrease	86 (15.3)	66 (12.6)	50 (9.0)
Abdominal pain	85 (15.1)	70 (13.4)	81 (14.6)

There were no significant differences in renal safety profiles between the 4 mg zoledronic acid group and the 90 mg pamidronate group during long-term treatment. However, for the 8/4 mg zoledronic acid treatment group, the risk of decreased renal function was roughly double that in the pamidronate group (risk ratio, 2.187; P = 0.001). The proportion of patients who experienced increases in serum creatinine did not increase in any of the treatment groups after 25 months of treatment compared with the 13-month safety data.23 Before the infusion-time amendment, 14.3% of patients in the 4 mg zoledronic acid group, 6.7% of patients in the pamidronate group, and 21.3% of patients in the 8/4 mg zoledronic acid group had increased serum creatinine levels. However, after the infusion time for zoledronic acid was increased to 15 minutes, 10.7% of patients in the 4 mg zoledronic acid group, 9.3% of patients in the pamidronate group, and 19.4% of patients in the 8/4 mg zoledronic acid group had increased serum creatinine levels. After 25 months, Grade 3–4 serum creatinine increases had occurred in 0.4% of patients in the 4 mg zoledronic acid group, in 1.9% of patients in the pamidronate group, and in 2.7% of patients in the 8/4 mg zoledronic acid group. Moreover, the pairwise comparison for the difference in renal safety between 4 mg zoledronic acid and 90 mg pamidronate revealed no significant differences for patients who were randomized after the infusion-time amendment (risk ratio, 1.057; P = 0.839). Therefore, 4 mg zoledronic acid by 15-minute infusion demonstrated a renal safety profile similar to that of 90 mg pamidronate by 2-hour infusion.

DISCUSSION

Patients with breast carcinoma or multiple myeloma suffer significant skeletal morbidity during the natural course of their disease. Bone metastases occur in 65–75% of patients with advanced breast carcinoma, and most bone metastases have an osteolytic appearance on radiographs.3 Without bisphosphonate therapy, patients with osteolytic lesions from advanced breast carcinoma suffer skeletal complications at an average rate of 3–4 events per year and experience their first pathologic fracture a median of 7 months after the diagnosis of bone metastases.12 Furthermore, in the placebo arm of an earlier pamidronate trial, during a 2-year study period, 52% of patients with predominantly osteolytic bone lesions secondary to breast carcinoma developed pathologic fractures, 37% required palliative radiation therapy, 13% developed HCM, and 11% required orthopedic bone surgery.12 Skeletal complications also occur frequently in patients with multiple myeloma. At the initial diagnosis of multiple myeloma, one-third of all patients have experienced their first pathologic fracture, and roughly 75% of patients report bone pain.26, 27 Therefore, skeletal complications place a heavy burden of illness on patients with malignant bone disease. Because patients with breast carcinoma survive for an average of 18–26 months after the initial diagnosis of bone metastases,11 and median survivals of 6–7 years have been reported in some multiple myeloma treatment centers,9 long-term palliative therapy is needed to reduce the chronic risk of skeletal complications.

Zoledronic acid has demonstrated favorable efficacy and safety for the treatment of skeletal complications from bone lesions secondary to multiple myeloma or a broad range of solid tumors, including breast, prostate, and lung carcinoma.20–22 Based on these favorable results, zoledronic acid received regulatory approval from the United States Food and Drug Administration in February, 2002 for the treatment of osteolytic lesions associated with multiple myeloma and bone metastases secondary to a wide variety of solid tumors in conjunction with standard antineoplastic therapy. Shortly thereafter (July 2002), the European regulatory authorities also granted zoledronic acid marketing approval for the prevention of skeletal complications in patients with advanced malignancies involving bone. The results from the core phase of this clinical trial in patients with multiple myeloma and breast carcinoma demonstrated that zoledronic acid was at least as effective as pamidronate in preventing or delaying the occurrence of skeletal complications from malignant bone disease.23 The 25-month extension phase of this trial has confirmed the long-term efficacy and safety of zoledronic acid in these patients and has demonstrated the durable clinical benefits of zoledronic acid. Furthermore, the long-term safety profile of zoledronic acid (4 mg by 15-minute infusion) was comparable to that of pamidronate (90 mg by 2-hour infusion), with no increase in the risk of renal impairment during long-term therapy.

Most importantly, analyses of secondary endpoints and patient subsets provide preliminary evidence of a differential clinical benefit favoring zoledronic acid over pamidronate. One such preplanned analysis was the Andersen-Gill multiple-event analysis, which incorporates the total number of skeletal complications that a patient experiences and the timing of these events into a robust model of skeletal morbidity.25 The Andersen-Gill multiple-event analysis can demonstrate treatment effects over the full course of the study and accommodates the differences in event rates and clustering of SREs that often occur in individual patients by adjusting the event rate parameters for individual patients throughout the course of the study.28 Using this robust assessment, it was found that zoledronic acid (4 mg) was more effective than pamidronate in reducing the risk of developing skeletal complications in the overall population over 25 months of therapy (risk ratio, 0.841; P = 0.030), with most of the advantage of zoledronic acid over pamidronate attributable to its higher efficacy in patients with breast carcinoma (risk ratio, 0.799; P = 0.025).

In the subpopulation of patients with breast carcinoma, 4 mg zoledronic acid demonstrated superior efficacy compared with pamidronate. The annual incidence of skeletal complications was reduced by 40%, and the risk of developing skeletal complications was reduced by 20% (P = 0.025), as assessed by multiple-event analysis. Furthermore, in the stratum of patients with breast carcinoma who were treated with hormonal therapy, zoledronic acid demonstrated significant benefit compared with pamidronate. Zoledronic acid (4 mg) significantly lowered the mean skeletal morbidity rate (P = 0.039), increased the median time to first SRE (P = 0.047), and reduced the risk of developing an SRE by 30% (P = 0.009) compared with pamidronate. The greater benefit of 4 mg zoledronic acid compared with pamidronate in the hormonal therapy breast carcinoma stratum may be due in part to the apparent earlier stage of disease or better prognosis of these patients. The median survival of patients with breast carcinoma who were receiving hormonal therapy was roughly 8 months longer than for patients who were receiving chemotherapy. Therefore, although the skeletal morbidity rate was similar in the hormonal therapy stratum and in the population of all patients with breast carcinoma, this stratum had a longer average duration of time at risk for skeletal complications. Hence, greater cumulative effects of therapy would be detected by the clinical endpoints in this stratum. These results suggest that earlier and sustained intervention with zoledronic acid may provide greater clinical benefits. Further clinical trials are needed to fully investigate this observation.

Pamidronate already has exhibited clear benefits in the treatment of patients with predominantly osteolytic lesions from multiple myeloma or breast carcinoma metastatic to bone in randomized, placebo-controlled trials.12–16 It now has been demonstrated that zoledronic acid has efficacy that is at least comparable to that of pamidronate in these patient populations. In addition, zoledronic acid has the added benefit of a convenient, 15-minute infusion. A recent report comparing infusion protocols for zoledronic acid and pamidronate found that pamidronate took an average of over 1.5 hours longer to infuse than zoledronic acid (66 minutes for zoledronic acid vs. 172 minutes for pamidronate).29 Therefore, the added convenience of the 15-minute infusion for zoledronic acid may have a positive impact on patient quality of life and may reduce utilization of healthcare resources.

In summary, the final analysis of this 25-month clinical trial has confirmed the long-term safety and efficacy of zoledronic acid (4 mg by 15-minute infusion). We also demonstrated that zoledronic acid (4 mg) is more effective than pamidronate in reducing the risk of developing skeletal complications in the overall population and in patients with breast carcinoma.

Acknowledgements

The authors thank the study investigators and the study participants. In addition to the authors, the principal investigators at centers that enrolled six or more patients included the following: Dr. J. Apffelstaedt, Dr. Andrew Belch, Dr. Richard Bell, Dr. H. Blacklock, Prof. Jean Brody, Dr. Adam Brufsky, Dr. Howard Burris III, Dr. John Carpenter, Dr. Ronald Carroll, Dr. R. Coleman, Dr. Robert Dewar, Dr. Ingo Diel, Dr. William Dugan, Dr. Peter Eisenberg, Dr. Benjamin Esparaz, Dr. Thierry Facon, Dr. Sebastian George, Dr. John Hamm, Prof. Anthony Howell, Dr. Mohamad Hussein, Dr. David Irwin, Dr. Andrew Jacobs, Dr. Leonard Kalman, Dr. Mary Kaminski, Dr. Dusan Kotasek, Dr. Eric Kraut, Dr. Steven Kuross, Dr. Allan Lipton, Dr. Charles Lui, Dr. Alan Lyss, Dr. Pierre Major, Dr. Jacinta Meharchand, Dr. Glenn Mills, Doctora Montserrat Muñoz, Dr. Lawrence Pawl, Dr. Brin Pressnail, Dr. Louise Provencher, Dr. Lee Rosen, Dr. Lee Schwartzberg, Dr. Ross Siemers, Dr. Keith Stewart, Dr. John Strupp, Dr. Simon Tchekmedyian, Dr. Richard Theriault, Dr. B. Thuerlimann, Dr. Ann Traynor, Dr. Rostislav Vyzula, and Dr. Furhan Yunus.