Appendix 1 Review of ~~Zoledronate~~Zoledronate Safety in Prostate Cancer, Study 039

Study # CGP 42446-03-039: “A randomized, double-blind, placebo-controlled, multicenter, comparative, safety and efficacy study of intravenous ~~zoledronate~~zoledronate (4 and 8 mg) in prostate cancer patients with metastatic bone lesions receiving antineoplastic therapy.”

This is a multicenter (136 sites), international study from June 22, 1998, until January 26, 2001. The study duration was 96 weeks, of which Phase 1, the 60-week Safety and Efficacy portion, is the subject of this review. Phase 2 was a 36-week Extension phase.

The study population consists of men with rising serum prostate specific antigen (PSA) while on hormonal therapy for prostate cancer metastatic to bone. No prior chemotherapy was allowed, although patients could receive antineoplastic therapy concomitant with the study, which could be hormonal or chemotherapy. Three sequentially rising PSA’s were required, within 8 weeks of visit 1, and patients had to demonstrate castrate levels of testosterone. The creatinine was to be < 3. Corrected serum calcium was required to be in the range of 8.0-11.6 mg/dL. Patients were stratified according to the presence or absence of metastatic disease at time of initial diagnosis.

Patients were randomized to receive ~~zoledronate~~zoledronate 4 mg or 8 mg or placebo by intravenous (i.v.) infusion q 3 weeks for 24 months. Initially ~~zoledronate~~zoledronate or placebo was given as a 5 minute i.v. infusion in 50 ml every 3 weeks. After Amendment 3, on June 24, 1999, the infusion time and volume were increased to 15 minutes and 100 ml, respectively. This was instituted in response to SAEs of renal failure in 3 patients receiving 8 mg dosing of ~~zoledronate~~zoledronate. By recommendation of the Data Safety Monitoring Board (DSMB) and the Renal Advisory Board (RAB), Amendment 4 was instituted on June 7, 2000. This required that all patients who received ~~zoledronate~~zoledronate would receive only 4 mg. Serum creatinine would be measured before each dose, and ~~zoledronate~~zoledronate held for worsening of creatinine, until the level was within 10% of the baseline creatinine. (Prior to the amendment, chemistries were required only at 3 weeks, 3months and then at 6 week intervals.)

The primary efficacy analysis was at the end of the study (month 15). The main efficacy endpoint was the “proportion of patients having at least one skeletal related event (SRE), which were defined as “pathologic bone fractures, spinal cord compression, surgery to bone, radiation therapy to bone and change in antineoplastic therapy to treat bone pain.” Hypercalcemia (HCM) was not included as an SRE.

Applicant assessment and analysis of safety

Adverse events, serious adverse events, laboratory studies and survival data were the main safety variables. Baseline and end of study physical examination, EKG, and laboratory evaluations were done, including hematology, blood chemistry, urine. Interim physical examination, vital signs, assessment for adverse events and laboratory studies were repeated every 3 weeks, except urine studies were approximately every 3 months. Serum creatinine was measured prior to each dose of study drug per amendment 4 (June 7, 2000). The time to discontinuation of study drug and duration of survival were assessed.

“For laboratory and adverse event analysis, data were cut at the end of the study drug period,” either the end of phase 1 or the last date of study medication plus 28 days. For time to death and renal deterioration analysis, all available data was included, up to the date of data base lock~~, including from phase 2.~~ For other safety parameters, the last visit date was used.

Renal toxicity was assessed by the number of patients experiencing “a renal adverse event using selected terms and the number of patients who met the predefined criteria of renal function deterioration.” Kaplan-Meier curves were used to define the time course of renal function deterioration.

Study population

There were 643 patients randomized to the following groups:

Zometa 4 mg # 214

Zometa 8/4 mg # 222

Placebo # 208

“The safety evaluable population included all patients who were randomized and received study drug.” The number of patients in each arm is shown in applicant’s table 7-2. Three of the randomized patients did not receive study drug and were not included in the safety analysis. One patient was randomized to the 4 mg group and 2 patients were randomized to the 8/4 mg group. One patient was randomized to the 8/4 mg group but actually received 4 mg for all treatments. This patient (USA/1891/11002) was included in the 8/4 mg group for efficacy, but was included in the 4 mg dose for safety analysis.

Applicant Table 7-2

Table 7-2.	Number (%) of patients in analysis populations by treatment group
	Zol 4 mg	Zol 8/4 mg	Placebo
Populations
Randomized	214 ( 100)	221 ( 100)	208 ( 100)
ITT population	214 ( 100)	221 ( 100)	208 ( 100)
Safety evaluable population	214	218	208
Source: Post-text table 7.1-1.

Reviewer’s Note: From data provided by the applicant, for patients assigned to the 8/4 mg group, 247 of 2400 administrations of ~~zoledronate~~zoledronate were actually 4 mg rather than 8 mg, or approximately 10%.

The following is an abbreviated, composite version of the applicant’s table 7-3, “Demographic summary by treatment group” and table 7-5, “Baseline disease specific variable by treatment”:

Reviewer table 1X

Zol 4 mg Zol 8/4 mg Placebo

Age (years)

N 214 218 208 Mean ± SD 71.8±7.91 71.2±8.04 72.2±7.89 Median 72.0 72.0 73.0 Min-max 45-90 43-90 37-90

Age

£ 60 19 ( 8.9) 19 ( 8.7) 15 ( 7.2) >60 195 ( 91.1) 199 ( 91.3) 193 ( 92.8)

Serum creatinine

Normal (< 1.4 mg/dL) 173 ( 80.8%) 168 ( 77.1%) 170 (81.7%) Abnormal (³ 1.4 mg/dL) 41 ( 19.2) 47 ( 21.6) 33 (15.9) Missing 0 ( 0.0) 3 ( 1.4) 5 ( 2.4)

~~The~~ Ppatient factors which m~~would be expected to cause~~ight increased susceptibility to renal toxicity of ~~zoledronate~~zoledronate are baseline renal function, age (which may relate to renal function reserve), and exposure to other nephrotoxic therapy. The above tables suggest similar age and baseline serum creatinine for the treatment groups. The applicant states that concomitant medication was similar for all treatment grooups, and this seem to be the case (Post-text tables 8.2-1,2,3,4), with patients experiencing little exposure to potentially nephrotoxic drugs.

.~~{Type of malignancy might be a variable in an integrated review of safety, with malignancies more commonly associated with renal dysfunction including multiple myeloma and prostate cancer.~~)~~Treatment-related factors for renal toxicity include dose, duration of infusion, exposure over time (# of months of treatment.)~~

Overall Exposure

The applicant’s table 8-1 demonstrates the overall exposure to study drug by treatment group for the safety evaluable patients.

Applicant table 8-1

For patients who did not present with metastatic disease at diagnosis,
the exposure (in months) to study drug is similar, particularly for the 4 mg and placebo treatment groups. For patients who presented at diagnosis with metastatic disease, the duration of exposure was greater for ~~zoledronate~~zoledronate 4 mg than for placebo or ~~zoledronate~~zoledronate 8/4 mg groups. The mean and median exposure was greater for the 4 mg group when the exposure for both strata was totaled.

For patients who did not present with metastatic disease at diagnosis, the exposure (in months) to study drug is similar, particularly for the 4 mg and placebo treatment groups. For patients who presented at diagnosis with metastatic disease, the duration of exposure was greater for zoledronate 4 mg than for placebo or zoledronate 8/4 mg groups. The mean and median duration exposure was greater for the 4 mg group when the exposure for both strata was totaled.

The following table, applicant table 8-2, demonstrates exposure to the study drug by treatment group, separating pre and post 15-minute infusion amendment patients. Data includes core (Phase 1) and extension (Phase 2) treatment periods. “The distribution of duration from randomization was similar before and after the amendment.”

Applicant table 8-2

Overall incidence and severity of adverse events

Clincal study reports used the IMN dictionary to code adverse events but the data for Study 039 is presented using the MedDRA dictionary. Adverse events were mapped from IMN preferred terms to the corresponding MedDRA terms, prior to pooling of data for analysis.

Almost all patients in each study group experienced at least one adverse event. The applicant’s Table 10-2 lists the frequency of AEs with an incidence of at least 15% in each treatment group. Bone pain, nausea, constipation, and fatigue were noted most often. Fatigue, anemia, myalgia, pyrexia, and lower limb edema were more frequent in patients receiving ~~zoledronate~~zoledronate versus placebo, but there was no apparent correlation with dose. Nausea, anorexia and vomiting were more common in the ~~zoledronate~~zoledronate 8/4 mg group. Bone pain was less in the 4 mg group. Dizziness was higher in the 4 mg group (17.8%) compared with the 8/4 mg group (10.1%) and placebo (11.5%). The significance of these differences is uncertain, and there may be no direct relationship with treatment.

Applicant table 10-2

Adverse events were thought to be study drug-related in 41.6%, 50.5% and 21.6% (Post-text table 10.1-5).

The incidence of grade 4 events was similar, 27.1%, 32.1%, and 25.0% for the 4 mg, 8/4 mg and placebo groups, respectively (Post-text table 10.1-4, volume 109). ~~Adverse events were thought to be study drug-related in 41.6%, 50.5% and 21.6% (Post-text table 10.1-5).~~

Reviewer Table 2X. Selected (more frequent) grade 4 adverse events by body system and treatment group:

Body system	Preferred term	Zoledr 4 mg N (%)	Zoledr 8/4 mg N (%)	Placebo N (%)
Any	Total	58 (27.1)	70 (32.1)	52 (25.0)
Blood and lymph	Total	8 (3.7)	9 (4.1)	5 (2.4)
	Anemia	6 (2.8)	6(2.8)	2 (1.0)
Cardiac	Total	10 (4.7)	10 (4.6)	8 (3.8)
GI	Total	8 (3.7)	3 (1.4)	6 (2.9)
Metab and Nutrit	Total	7 (3.3)	13 (6.0)	7 (3.4)
	Dehydration	4 (1.9)	2 (0.9)	2 (1.0)
Musculoskeletal	Bone pain	3 (1.4)	4 (1.8)	5 (2.4)
Neoplasms	Aggravated mal	10 (4.7)	14 (6.4)	6 (2.9)
Nervous system	Total	6 (2.8)	7 (3.2)	7 (3.4)
Renal, urinary	Total	7 (3.3)	15 (6.9)	7 (3.4)
	RF Acute	6 (2.8)	6 (2.8)	1 (0.5)

There is no clear signal suggesting a relation of grade 4 events to treatment arm, except for renal adverse events.

Renal adverse events

A Renal Advisory Board (RAB) was established in November of 1999 to monitor the renal safety of ~~zoledronate~~zoledronate, because of concerns about renal dysfunction associated with treatment. Amendment 3 had been instituted June 24, 1999, in response to SAEs of renal failure in 3 patients receiving 8 mg ~~zoledronate~~zoledronate. This changed administration volume from 50 ml to 100 ml and administration time from 5 minutes to 15 minutes. The pre-amendment data (applicant’s table 10-4), suggests a possible dose-related renal toxicity for overall events, “renal failure acute,” “ renal impairment NOS,” and “blood creatinine increased.” Also of interest is the marked increase in “urinary retention” in the ~~zoledronate~~zoledronate 8/4 group, which could have exaggerated the apparent renal-toxic effect of study drug. One would anticipate that urinary retention is disease-, rather than drug-related.

Reviewer table 3X.

Selected renal AEs by preferred term and treatment group for pre 15-minute infusion amendment patients

	Zoledr 4 mg	Zoledr 8/4 mg	Placebo
	N (%)	N (%)	N (%)
Total # patients	117	125	124
Total # with renal AE	36 (30.8)	54 (43.2)	33 (26.6)
RF, Acute	6 (5.1)	9 (7.2)	4 (3.2)
Renal impair, NOS	6 (5.1)	8 (6.4)	3 (2.4)
Urinary retention	5 (4.3)	18 (14.4)	11 (8.9)
Blood creatinine increas	4 (3.4)	6 (4.8)	0

The following table demonstrates renal AEs following amendment 3. The incidence of “Acute renal failure” is still higher for the ~~zoledronate~~zoledronate patients compared with placebo. However, “renal failure NOS” and “increased creatinine” are no longer reported in the 4 mg group. “Renal impairment NOS” is highest in the 4 mg group, with the 8/4 mg and placebo groups being similar.

Applicant table 10-5

Reviewer comment: Amendment 3 may have resulted in some slight improvement in the renal toxicity profile, but the important analysis of safety is for patients who were randomized following amendment 3 and amendment 4 (see below). Also note that the number of patients with urinary retention is less disproportionately distributed to the 8/4 mg group compared with the pre-third amendment patients.

The following table (applicant’s table 10-7) summarizes events which the applicant suggests are associated with bisphosphonates as a class. Electrolyte abnormalities are increased in ~~zoledronate~~zoledronate patients compared with placebo, possibly in a dose-dependant way.

Applicant table 10-7

Deaths and other serious and other significant adverse events:

In Section 3.5.3.2 of the protocol, the applicant defines a serious adverse event (SAE) as an event which:

1. Is fatal or life threatening.

2. Requires or prolongs hospitalization

3. Is significantly or persistently disabling or incapacitating

4. Constitutes a congenital anomaly or a birth defect

5. Encompasses any other clinically significant event

Item 5 is not clearly explained. “Clinically significant AEs” are defined as events which were not SAEs, but “resulted in withdrawal of study drug or were considered to be clinically important and required concomitant therapy.”

Applicant table 10-8 lists “patients who died, had other serious or clinically significant AEs or discontinued therapy because of them.”

Applicant table 10-8

SAEs occurred in fewer of the ~~zoledronate~~zoledronate 4mg patients (49.5%), compared with 8/4 (58.7%) or placebo groups (56.3%). However, deaths were greatest in the ~~zoledronate~~zoledronate 8/4 mg group, followed by placebo. Similarly, discontinuation due to SAEs was greatest in the 8/4 mg group.

SAEs

Applicant’s Table 10-9 shows the body systems most often involved in SAEs, with greater than 10% incidence in any treatment group.

Applicant table 10-9

The incidence of SAEs for ~~zoledronate~~zoledronate 4 mg seems favorable when compared to placebo, except for the category of “metabolism and nutrition”. In contrast, the incidence of SAEs was higher for ~~zoledronate~~zoledronate 8/4 compared with both the 4 mg and placebo groups, for “blood and lymphatic” and “renal and urinary”. “Anemia NOS” occurred in 5.1% of the ~~zoledronate~~zoledronate 4 mg group, 8.3% of the 8/4 mg group, and 3.8% of placebo patients. It should be noted that skeletal related events (SRE) were not included as SAEs.

Applicant’s post-text table 10.2-1 (volume 109) summarizes SAEs, regardless of cause, by body system, preferred term and treatment group. Reviewer’s Table X4 is abstracted from this table, and shows SAEs occurring in at least 5% of patients in any treatment group. These events include anemia, nausea, vomiting, weakness, dehydration, bone pain, malignant neoplasm aggravated, acute renal failure, and urinary retention.

Reviewer Table 4X

SAE’s occurring in at least 5% of patients in any treatment group

Preferred term	Zoled 4 mg N (%)	Zoled 8/4 mg N (%)	Placebo N (%)
Anemia NOS	11 (5.1)	18 (8.3)	8 (3.8)
Nausea	13 (6.1)	13 (6.0)	7 (3.4)
Vomiting NOS	12 (5.6)	11 (5.0)	7 (3.4)
Weakness	14 (6.5)	7 (3.2)	7 (3.4)
Dehydration	16 (7.5)	9 (4.1)	7 (3.4)
Bone pain	11 (5.1)	21 (9.6)	25(12.0)
Malignant neopl. aggravated	13 (6.1)	20 (9.2)	13 (6.3)
Acute renal failure	10 (4.7)	12 (5.5)	3 (1.4)
Urinary retention	3 (1.4)	14 (6.4)	9 (4.3)

Reviewer comment: The incidence of anemia in the ~~zoledronate~~zoledronate 8/4 population seems significant compared with the placebo population, although unexplained. The distribution of bone pain for the 3 treatment groups does not suggest a dose-dependant benefit of the study drug for this parameter. Acute renal failure is comparably increased in both ~~zoledronate~~zoledronate groups compared with placebo.

Deaths

One hundred fourteen patients died during the study or within 28 days of the last dose of study drug. The incidence of deaths was similar for placebo (16.8%) and ~~zoledronate~~zoledronate 4 mg (14.5%), but there was a higher per cent deaths in the ~~zoledronate~~zoledronate 8/4 mg group (22%). The most frequent cause of death was “aggravated malignant neoplasm.”

Post-text table 10.2-3 lists the cause of death in the 97 patients who died before completing the study, who discontinued because of death. Of these patients, there were

10 (4.7%) in the 4 mg group, 16 (7.3%) in the 8/4 mg group and 11 (5.3%) in the placebo group. Of the 4 patients (1.8%) who died during the study due to urinary or renal disorders, all were in the ~~zoledronate~~zoledronate 8/4 mg group, with 2 deaths due to acute renal failure, 1 death due to “renal failure NOS” and 1 death due to hematuria.

When deaths were analyzed by stratum (patients without or with metastases), those with metastases at diagnosis had a shorter time to death if they were in the ~~zoledronate~~zoledronate 8/4 mg group compared with ~~zoledronate~~zoledronate 4 mg. In patients with no metastases at diagnosis, the time to death was similar in the 3 treatment groups. This data appears in post-text table 10.2-7. For the overall group, the median survival in days was 464 for placebo, 546 for the 4 mg group, and 407 for the 8/4 mg group.

Treatment discontinuations

Applicant table 10-11 lists the number of patients discontinued for adverse events by body system.

Applicant table 10-11

The discontinuations for “General disorders” were predominantly for nonspecific symptoms such as weakness and fatigue and seem unlikely to be study-drug related. The discontinuations for “Renal and urinary disorders” are of concern. Reviewer’s Table 5 X (from post-text table 10.2-4) shows selected renal AEs associated with discontinuation.

Reviewer table 5X.

Number (%) of patients discontinued for (selected) renal AEs by preferred terms and treatment group.

Preferred term	Zoledr 4mg, N=214 N (%)	Zoledr 8/4mg, N=218, N (%)	Placebo N=208, N (%)
Total renal and urinary	11 (5.1)	12 (5.5)	4 (1.9)
Renal impairment NOS	6 (2.8)	4 (1.8)	0
Renal failure acute	3 (1.4)	5 (2.3)	1 (0.5)
Oliguria	1 (0.5)	0	0
Urinary retention	1 (0.5)	2 (0.9)	2 (1.0)

Laboratory parameters

The treatment groups were similar when hematology tests were analyzed by number of patients reaching selected “notable” values of neutrophil count <0.5 x 10⁹, platelets < 25 x 10⁹, and hemoglobin < 6.5 g/dL (applicant’s table 10-13). However, applicant’s table 10-14 demonstrates that more patients in the ~~zoledronate~~zoledronate arms had a greater than 25% decrease in hemoglobin, compared with placebo, possible in a dose-dependant fashion.

Applicant table 10-12

The following table shows grade 3 or 4 hematologic toxicity for the treatment groups.

Applicant table 10-15

Grade 3 hemoglobin values were more than twice as common in the ~~zoledronate~~zoledronate 8/4 mg group than either the 4 mg or placebo groups. The significance of this is uncertain, but it could relate to the higher incidence of renal dy~~sfunction~~dysfunction in the higher dose ~~zoledronate~~zoledronate group.

When patients were analyzed for notable biochemistry values (other than creatinine), there was almost twice the incidence in the ~~zoledronate~~zoledronate groups of hypermagnesemia > 3.0 (4 mg =6.4 %; 4/8 mg = 5.8%) compared with placebo (3%). In the ~~zoledronate~~zoledronate groups, there was more hypophosphatemia, and only a slight increase in hypokalemia and hyponatremia compared with placebo. Applicant table 10-19 shows grade 3 or 4 chemistry abnormalities by treatment group, demonstrating excess grade 3 hypermagnesemia, hypophosphatemia, and hypokalemia for the ~~zoledronate~~zoledronate groups compared with placebo.

Applicant table 10-19

Serum creatinine and renal function deterioration

Baseline serum creatinine was considered normal if <1.4 mg/dL and abnormal if > 1.4 mg/dL. Renal function deterioration was defined as any of the following:

· Normal baseline with change from baseline > 0.5 mg/dL

· Abnormal baseline with change from baseline > 1.0 mg/dL

· Post-baseline value > 2 time the baseline value.

Subsequent to amendment 3 (15 minute infusion), renal function deterioration as defined in applicant table 10-16, was 17.4% for ~~zoledronate~~zoledronate 4 mg, 25.3 % for ~~zoledronate~~zoledronate 25.3% and 11.5 % for placebo.

Applicant table 10-16

No patient experienced grade 4 creatinine elevation (> 6xULN). Grade 3 (>3xULN) creatinine values occurred in 5 ~~zoledronate~~zoledronate 4 mg patients and in 2 of the 8/4 mg group following amendment 3, but none in the placebo group. (Prior to amendment 3, 2 placebo patients had grade 3 elevation of creatinine).

The incidence of renal function deterioration was increased in the ~~zoledronate~~zoledronate treatment groups for patients with a baseline serum creatinine of > 1.4, compared with those with a normal baseline creatinine. The incidence of deterioration decreased after amendment 3, but still was higher in the ~~zoledronate~~zoledronate groups. The following table is abstracted from applicant table 10-21.

Reviewer table 6X

Patients who experienced renal function deterioration by baseline serum creatinine (15-minute infusion)

	Zoledr 4 mg N (%)	Zoledr 8/4 mg N (%)	Placebo N (%)
Patients with normal baseline creatinine	82	68	68
- - - with deterioration	10 (12.2)	14 (20.6)	7 (10.3)
Patients with abnormal baseline creatinine	10	19	20
- - - with deterioration	4 (40.0)	4 (21.1)	2 (20.0)
Total # patients at baseline	92	87	78
- - - with deterioration	14 (15.2)	18 (20.7)	9 (11.5)

When time to renal function deterioration was analyzed by Cox regression for patients treated post amendment 3, there were no statistically significant differences in pairwise comparisons between any treatment groups. The Kaplan-Meier curves of first renal function deterioration by treatment group are similar for patients treated by 15-minute infusion (Applicant’s figure 10.2). The risk ratio between ~~zoledronate~~zoledronate 4 mg and placebo was 1.066 for patients randomized after amendment 3. (The risk ratio between ~~zoledronate~~zoledronate 8/4 mg and placebo was 1.764.) The differences were not statistically significant.

Applicant figure 10.2

Urinalysis

The percentage of patients with elevated urine protein (> 2+) post-baseline was greater in the ~~zoledronate~~zoledronate treatment groups than in the placebo group (Post-text table 10.3-2).

Vital signs

More patients in the ~~zoledronate~~zoledronate groups had > 10% weight decrease. In the 4mg group, the incidence was 30.5 %. 32.7 in the 8/4 mg group, and 19.4% in the placebo group.

ECGs

For those patients who had follow-up ECGs, there were more abnormalities compared with baseline in all study groups

Special safety topics-PSA

For patients who had serum PSA measured within 30 days of disease progression, changes in PSA were similar for the ~~zoledronate~~zoledronate 4 mg and placebo groups. The applicant suggests that ~~zoledronate~~zoledronate 4 mg has “no effect on the secretion, clearance, or measurement of PSA”. This seems a reasonable interpretation of the data.

Reviewer comment and conclusions

Increased risk of renal function deterioration is the only significant safety issue for ~~zoledronate~~zoledronate in this and related clinical trials of long-term treatment for patients with metastatic malignancy to bone. This toxicity was dose and time-related.

Adverse events, seen with other bisphosphonates, and possibly class effects of bisphosphonates, were reported more frequently with ~~zoledronate~~zoledronate than placebo. These include fever, arthralgias/myalgias, and electrolyte abnormalities, but do not present a safety issue.

Anemia was slightly more common in the ~~zoledronate~~zoledronate groups (approximately 27%) compared with placebo (17.8%). This is of uncertain, if any, significance. In the ~~zoledronate~~zoledronate groups, there was an excess of hypermagnesemia and hypophosphatemia compared with placebo.

Subsequent to the occurrence of acute renal failure in a total of 3 patients receiving ~~zoledronate~~zoledronate 8 mg in this and related trials for disease metastatic to bone, a renal advisory board (RAB) of nephrologists was convened and several amendments to the protocol were made. The incidence of renal function deterioration was decreased by prolonging the infusion time from 5 to 15 minutes (amendment 3). Amendment 4 further improved the safety profile by eliminating the 8 mg dose. The safety of the 4 mg dose was improved by requiring assessment of serum creatinine before each dose. ~~Zoledronate~~Zoledronate was held for renal deterioration, as defined in the amended protocol. Therapy was not reinstituted until creatinine returned to within 10% of the baseline value.

Following amendment 3, the overall incidence of renal related AEs was comparable for ~~zoledronate~~zoledronate 4 mg (30.9%) and placebo (31.0%). However, the incidence of “renal impairment NOS” was 7.2% for ~~zoledronate~~zoledronate 4 mg, 3.2% for ~~zoledronate~~zoledronate 8/4 mg and 3.6% for placebo. The incidence of renal failure acute” was 5.2% for ~~zoledronate~~zoledronate 4 mg, 4.3% for ~~zoledronate~~zoledronate 8/4 mg, and 0 for placebo. “Blood creatinine increased was reported at 0 for ~~zoledronate~~zoledronate 4 mg and 3.6% for placebo (4.3% for ~~zoledronate~~zoledronate 8/4 mg).

The incidence of renal function deterioration was higher in the ~~zoledronate~~zoledronate 4 mg group (15.2%) and ~~zoledronate~~zoledronate 8m (20.7%) compared with placebo (11.5%), even following amendment 3.

Cox regression analysis of time to renal function deterioration showed no difference between the groups. Kaplan-Meier curves of first renal function deterioration by treatment group showed no difference for patients treated by 15-minute infusion. The risk ratio between ~~zoledronate~~zoledronate 4 mg and placebo was 1.066 for these patients, but the difference was not statistically significant.

Reviewer conclusion: ~~Zoledronate~~Zoledronate 4 mg i.v. over 15 minutes every 3 weeks has an acceptable risk profile. The risk of renal deterioration with ~~zoledronate~~zoledronate is greater than placebo, but acceptable with clinical monitoring of renal function before each dose. The risk of renal toxicity increases with duration of therapy. Caution is indicated for patients with elevated baseline creatinine, particularly since the study population excluded patients with creatinine > 3 and the drug is excreted unchanged by the kidneys. This study population (prostate cancer) did not have extensive concomitant exposure to other potentially renal toxic drugs. As the treatment population is expanded, it will be necessary to monitor for possible enhancement of renal toxicity in patients who are treated concomitantly with drugs of known nephrotoxic potential.

Appendix 2 Review of ~~Zoledronate~~Zoledronate Safety in Solid Tumors, Study 011

Study # CGP 42446-03-011: “A randomized, double-blind, placebo-controlled, multicenter trial to evaluate the safety and efficacy of ~~zoledronate~~zoledronate (4 and 8 mg) administered intravenously as an adjuvant to anticancer therapy to patients with any cancer with bone metastases other than breast cancer, multiple myeloma or prostate cancer.”

This is a multicenter, international study (141 sites, in US 72, others in Europe, Australia/New Zealand). The first patient enrolled 27 August 1998 and the last patient completed therapy 30 January 2001. The study population consisted of adults with solid tumors other than breast or prostate cancer. Patients could receive antineoplastic chemotherapy. Patients were to have objective evidence of at least 1 site of metastatic disease to bone. ECOG performance status was to be 0.1, or 2. If ECOG 2, patients needed to have been diagnosed within 6 weeks of the screening visit.

Criteria for exclusion included creatinine >3. Corrected calcium < 8 or > 12 at visit 1 was an indication for exclusion.

From December 1998 to May 1999 there were 3 renal failure SAEs in patients receiving ~~zoledronate~~zoledronate 8mg, manifested by progressively rising serum creatinine. “These 3 cases represent approximately 1% of the patients who have been exposed to the 8 mg dose in the ~~zoledronate~~zoledronate program.” This occurred in 1 patient each of a Phase II myeloma study (cycle #11), protocol 010 (diagnosis breast cancer, cycle #1), and protocol 039 (prostate cancer, after cycle #7). The patients were, respectively, ages 75, 80, and 71, with baseline creatinines of 1.6, 1.9, and 1.4. These events led to Amendment 3 on 25 June 1999 and Amendment 5 on 7 June 2000.

The duration of therapy was 9 months (36 weeks). Patients were randomized to receive ~~zoledronate~~zoledronate 4 mg or 8 mg or placebo i.v. over 5 minutes q 3 weeks. After amendment 3, the infusion duration was prolonged to 15 minutes. After amendment 5, all ~~zoledronate~~zoledronate patients received 4 mg per dose. Serum creatinine was to be evaluated before each dose, and treatment held for deterioration of creatinine, as defined in the amendment, until the level was within 10% of the baseline creatinine.

Applicant assessment and analysis of safety

Adverse events, serious adverse events, laboratory studies, and survival data were the main safety variables. Baseline and end of study (week 36) physical examination, EKG, and laboratory evaluations were done, including hematology, blood chemistries, urine chemistries and urinalysis. Interim physical examination, vital signs, assessment for adverse events, and laboratory studies were repeated every 3 weeks, except urine studies were approximately every 3 months. After amendment 5, the results of creatinine levels were required prior to each dose of ~~zoledronate~~zoledronate, so that the drug could be held for renal deterioration.

Safety analysis was based on the type and frequency of adverse events and the laboratory values outside of pre-determined ranges. The data were tabulated.

The effect on renal function was analyzed according to the number of patients who experienced renal adverse events using selected terms suggested by the renal advisory board (RAB) and the number of patients who met pre-defined criteria of renal deterioration. Kaplan-Meier curves were used to describe the time course of renal function deterioration.

“For laboratory and adverse event analysis, data were cut at the end of the study drug period,” either at the end of the core study phase or the last date of study medication plus 28 days. For time to death and renal deterioration analysis, all available data was included, up to the date of the data base lock. For other safety parameters, the last visit date was used.

Study population

Seven hundred seventy-three patients with osteolytic bone metastases from solid tumors other than breast and prostate cancer were randomized to the following groups:

~~Zoledronate~~Zoledronate 4 mg #257

~~Zoledronate~~Zoledronate 8 mg #266

Placebo #250

Patients were stratified into two groups, patients with non-small cell lung cancer (NSCLC) and patients with all other types of solid tumors (except breast and prostate).

The safety population consisted of all randomized patients who received study medication, and had documented evidence of at least one post-baseline safety evaluation.

Seven randomized patients were not included in the safety analysis because they failed to receive any study treatment. The number of patients in each arm is shown in applicant’s table 7-2.

Applicant Table 7-2

Reviewer’s Note: Analysis of data provided by the applicant, demonstrates that for patients assigned to the 8/4 mg group, 391 of 1809 ~~zoledronate~~zoledronate doses administered were actually 4mg rather than 8 mg, or approximately 22%.

Thirty-eight patients were randomized to the incorrect stratum.

· 3 patients with NSCLC were randomized into the other solid tumors stratum, one in each of the 3 treatment groups.

· 35 patients with small cell lung cancer were randomized in the NSCL cancer stratum with 12 in the 4 mg group, 10 in the 8/4 mg group and 13 in the placebo group.

For the efficacy analysis, the patients were not reassigned, but for the safety analysis, “such patients were reassigned to the correct stratum.”

Reviewer comment: The protocol (section 3.4.1) describes stratification for “patients with lung cancer” and “patients with all other cancers”. NSCLC and SCLC are not distinguished in the protocol or its amendments and, it seems there was no clear distinction between NSCLC and SCLC until the study report.

The following table abstracts data from applicant’s tables 7-3 and 7-4, to illustrate baseline age and serum creatinine. These are patient factors which might increase the susceptibility to the renal toxic effects of ~~zoledronate~~zoledronate.

Reviewer Table 1x. Baseline patient age and serum creatinine by treatment group.

This data shows similar age for treatment groups whether viewed as median age or % older than 60 years of age. The baseline serum creatinine values are similar, particularly for the ~~zoledronate~~zoledronate 4 mg and placebo groups.

Reviewer’s comment: The treatment related factors that might predict for renal toxicity of ~~zoledronate~~zoledronate include dose, duration of infusion, exposure over time (# of months of treatment), and concomitant exposure to renal toxic drugs.

Overall exposure

The duration of exposure to the study drug was similar for all treatment groups (applicant’s table 8-1). The mean duration of exposure was 4.51 months for ~~zoledronate~~zoledronate 4 mg, 4.45 months for the 8/4 mg group and 4.62 months for placebo. The median duration of exposure was 4.04 months for ~~zoledronate~~zoledronate 4mg, 3.21 months for ~~zoledronate~~zoledronate 8/4 mg, and 3.79 months for placebo.

Most patients in all groups were treated following amendment 3 and, therefore, were treated by infusion over 15 (rather than 5) minutes. This was 73% of the ~~zoledronate~~zoledronate 4 mg patients, 76% of the 8/4 mg group and 74.5% of the placebo group. Many patients were treated fewer than 3 months. For the 15 minute infusion patients, this includes 37.1% of the 4mg group, 40.8% of the 8/4 mg group, and 41.8% of the placebo patients.

Concomitant medication

“Approximately 80% in each group were receiving antineoplastic therapy at study entry and during the study with the most common agents being carboplatin, cisplatin, paclitaxel, gemcitabine, fluorouracil, etoposide, folinic acid, vinorelbine and doxorubicin.” Post-text table 8.2-4 shows that the incidence of treatment with the nephrotoxic drug, cisplatin, concomitant with study medication was similar among the treatment groups. The incidence of cisplatin therapy was 14.6% for the ~~zoledronate~~zoledronate 4mg group, 17.4% for the 8/4 mg group, and 16.6% for placebo.

Overall incidence and severity of adverse events:

Clinical study reports used the IMN dictionary to code adverse events (AEs.) The data is summarized in the study report using MedDRA preferred terms. Adverse events were mapped from the IMN preferred terms to the corresponding MedDRA terms. The number of patients with AEs is provided by body system, preferred term, and treatment group. If a patient had more than one type of event within a body system, the patient was counted only once.

“Almost all patients in all treatment groups experienced at least one adverse event.” There was a slightly higher incidence among the ~~zoledronate~~zoledronate 4 mg patients of AEs affecting gastrointestinal, nervous, and respiratory systems, and the eye, compared with the other 2 groups. Applicant table 10-1 shows the % of patients with adverse events overall and by system, for events occurring > 15% in any group.

Applicant Table 10-1

The patients with renal and urinary disorders include all patients, regardless of whether they were treated by 5- or 15-minute infusion. The incidence is similar for ~~zoledronate~~zoledronate 4 mg and placebo, slightly higher for the 8/4 mg group.

The most frequent adverse events were bone pain, nausea, anemia, vomiting, constipation, dyspnea, fatigue, pyrexia, weakness, anorexia, lower extremity edema, and malignant neoplasm aggravated. Nausea, vomiting, and dyspnea were higher in the ~~zoledronate~~zoledronate 4 mg group than in the placebo group. There was no clear dose relationship. This data appears in applicant table 10-2.

Applicant Table 10-2

Reviewer comment: The applicant also presented the number of patients with the most common AEs by stratum (NSCLC or not) and treatment group (table 10.3) but this did not seem to add significantly to the analysis.

The overall incidence of grade 3 and 4 adverse events, regardless of cause, is similar for the treatment groups. (Post text table 10.1-4, volume 95.) The incidence of grade 3 AEs for the ~~zoledronate~~zoledronate 4 mg group was 33.5% and 38.9% for the placebo group. The incidence of grade 4 AEs was 43.3% and 39.8%, respectively.

Renal adverse events

A Renal Advisory Board (RAB) was established in November 1999 to monitor the renal safety of ~~zoledronate~~zoledronate, due to concerns about renal toxicity. Three patients in other studies had renal failure after receiving ~~zoledronate~~zoledronate 8 mg. Applicant table 10-4 shows renal AEs using the preferred terms specified by the RAB for patients randomized before the 15-minute infusion amendment and in table 10-5 for patients randomized after the 15-minute infusion amendment. For the pre-amendment patients, the incidence of renal AEs was 16.2 % in the ~~zoledronate~~zoledronate 4 mg group, 20.3% in the 8/4 mg group and 12.7% in the placebo group. The incidence of acute renal failure was 5.9% (4 patients) and 6.3% for ~~zoledronate~~zoledronate 4 mg and 8/4 mg, respectively, and 0 for placebo. The incidence of “renal impairment NOS” was 2.9% (2 patients) and 3.1% in the ~~zoledronate~~zoledronate groups, and 0 in the placebo group.

Renal AEs decreased in the ~~zoledronate~~zoledronate arms after extending the infusion time to 15 minutes, and was stable for the placebo patients. Applicant table 5 shows the renal AEs by preferred term and treatment group for the patients randomized after the 15-minute infusion amendment.

Applicant Table 10.5

Even after institution of the 15-minute amendment, the incidence of “renal failure acute” is higher for ~~zoledronate~~zoledronate (4mg, 2.2 ) compared with placebo (0.5), but particularly for the 8/4 mg group (3.5). The category of “blood creatinine increased” is higher in the post 15-minute group, 5% for 4 mg, 8% for 8/4 mg, and 1% for placebo, compared with the pre 15-minute incidence of 0, 1.6% and 0 respectively.

Reviewer comment: The higher incidence of elevated creatinine post- compared with pre-amendment 3 may actually be a higher incidence of discovery, relating to the requirement to assess creatinine prior to each treatment, following amendment 5 in June 2000.

In table 10-6 of the study report, the applicant reviews chemotherapy-associated AEs to determine if these were increased by the addition of ~~zoledronate~~zoledronate. The most common AEs were nausea, vomiting, anorexia and appetite decrease. Each of these, except anorexia, was more frequent in the ~~zoledronate~~zoledronate groups than in the placebo group.

Reviewer comment: Since the incidence of these AEs was higher for ~~zoledronate~~zoledronate 4 mg than 8/4 mg, makes it less likely that ~~zoledronate~~zoledronate caused the symptoms. Of interest, but probably not of causal significance, is the fact that stomatitis is present at higher incidence in the ~~zoledronate~~zoledronate 4 mg (5.9%) and 8/4 mg (6.4%) groups than for placebo (3.6%).

AEs (other than renal) tabulated as “most clinically relevant for bisphosphonates” showed no overall increased incidence for ~~zoledronate~~zoledronate patients when “any body system” was specified. However, the ~~zoledronate~~zoledronate 4 mg group did have excess AEs compared to placebo for “electrolytes” (20.1% vs. 12.1%) and “eye abnormalities” (12.6% vs. 6.9%). There was no dose-related increase for the ~~zoledronate~~zoledronate 8/4 mg group, which was similar to the 4 mg group for these AEs (19.2% “electrolytes” and 11.3% “eye abnormalities.”

Deaths, SAEs and other “significant adverse events”

In section 3.5.3.2 of the protocol, the applicant defines a serious adverse event (SAE) as an event which:

1. Is fatal or life threatening

2. Requires or prolongs hospitalization

3. Is significantly or persistently disabling or incapacitating

4. Constitutes a congenital anomaly or a birth defect

5. Encompasses any other clinically significant event

Item 5 is not clearly explained. “Clinically significant AEs” are defined as events which were no SAEs, but “resulted in withdrawal of study drug or were considered to be clinically important and required concomitant therapy.

The applicant’s table 10.8 lists “patients who died (up to the end of the core study phase or within 28 days following last treatment), had other serious or clinically significant AEs or discontinued because of them”.

Applicants table 10-8

There is a slight excess of SAEs for ~~zoledronate~~zoledronate 4mg compared with placebo, but not for the other events. There were 261 deaths, with a similar incidence in each treatment group. The most frequent cause of death was progression of cancer.

Post-text table 10.2-3 summarizes number of patients who discontinued due to death by body system, preferred term and treatment group. This shows 10.6% of the ~~zoledronate~~zoledronate 4 mg group and 17.4% of the placebo group dying from “malignant neoplasm aggravated”. Death was attributed to “renal and urinary disorders” for only 1 patient in the ~~zoledronate~~zoledronate 4 mg group, and no patients in the placebo group.

Reviewer note: There is a discrepancy between the per cent of patients in each treatment group for table 10-8 and post-text-table 10.2-3. The latter cites 26.0%, 28.3% and 26.7% incidence of deaths in the ~~zoledronate~~zoledronate 4 and 8/4 mg groups, compared with 35.0%, 33.6%, and 33.6% deaths in table 10-8.

The most frequent SAEs, regardless of study drug relationship, were anemia, dehydration, malignant neoplasm aggravated and dyspnea. Anemia occurred in 7.9% of the ~~zoledronate~~zoledronate 4 mg group and in 3.6% of the placebo group. Dyspnea was present in 10.2% and 7.7%, respectively. The incidence of “renal and urinary disorders” was 5.1% in ~~zoledronate~~zoledronate 4mg, 7.2% in 8/4 mg, and 3.6% in placebo. For acute renal failure, the incidence was 3.1%, 4.2%, and 0.4%, respectively.

Note that the efficacy endpoint of SREs (skeletal related events) was not to be included in the SAE tabulation.

Applicant’s table 10-9 summarizes the number of patients discontinued from study for an adverse event.

Applicant’s table 10-9

There is no increase in discontinuations for ~~zoledronate~~zoledronate 4 mg compared with placebo. Discontinuation for renal and urinary disorders is similar for the 2 groups, but slightly higher for the ~~zoledronate~~zoledronate 8/4 mg group.

Laboratory parameters

Hematology

The treatment groups were similar when hematology tests were analyzed by number of patients reaching selected “notable” values of neutrophil count <0.5 x 10⁹, platelets < 25 x 10⁹ and hemoglobin < 6.5 G/dL (applicant’s table 10-12). The percentage of patients with >25% decrease in hemoglobin is also similar for the groups.

There was a slightly increased incidence of grade 3, but not grade 4, post-baseline hemoglobin (<8 g/dL) for the ~~zoledronate~~zoledronate 4 mg group compared with 8/4 mg and placebo. These percentages were 3.6%, 2.2%, and 1.4% respectively.

Serum chemistry

The tests performed include sodium, potassium, SGOT, SGPT, alkaline phosphatase, LDH, creatinine, BUN, calcium, phosphorus and magnesium. When patients were analyzed for notable biochemistry values (other than creatinine), there were no major differences between the ~~zoledronate~~zoledronate 4 mg and placebo groups. The incidence of phosphorus < 1.5 mg/dL was 2.7% for ~~zoledronate~~zoledronate 4 mg and 0.9% for placebo (applicant table 10-15). Grade 3 and 4 abnormalities are shown in table 10-17 and reveal little difference between ~~zoledronate~~zoledronate groups and control for chemistry abnormalities post baseline. except for more hypophosphatemia with ~~zoledronate~~zoledronate. Phosphorus <2 (grade 3 hypophosphatemia) was present 8.9%, 12.0% and 1.4%, respectively for zolodronate 4 mg, 8/4 mg and placebo. Grade 3 hypokalemia was slightly higher (4.9%) in the zolodronate 4 mg group compared with placebo (3.2%).

Serum creatinine and renal function deterioration

Baseline serum creatinine was considered normal if <1.4 mg/dL and abnormal if > 1.4 mg/dL. Renal function deterioration was defined as any of the following:

· Normal baseline with change from baseline > 0.5 mg/dL

· Abnormal baseline with change from baseline > 1.0 mg/dL

· Post-baseline value > 2 time the baseline value.

Applicant table 10-14 shows the incidence of post-baseline notable serum creatinine. Subsequent to amendment 3 (15 minute infusion), 2 patients (1.2%) in the ~~zoledronate~~zoledronate 4 mg group developed creatinine > 4.5 , 2 patients in the 8/4 mg group, and 0 in the placebo group. Excluding these patients, the incidence of creatinine increase > 0.5 was 10.3% for ~~zoledronate~~zoledronate 4 mg, 11.6% for ~~zoledronate~~zoledronate 8/4 mg and 7.4 % for placebo. This represents an improvement for the ~~zoledronate~~zoledronate groups, compared with pre 15-minute infusion amendment.

Applicant table 10-14

Applicant table 10-16 shows the distribution of grade 3 or 4 creatinine elevations.

Applicant table 10-16

For the post amendment patients, the incidence of grade 3 and 4 creatinine elevations was similar across the 3 treatment groups, representing an increase in incidence for ~~zoledronate~~zoledronate 4 mg and placebo groups compared with pre amendment.

Reviewer comment: There were too few patients with abnormal creatinine in the pre amendment group to make quantitative comparisons with the post amendment group. In the 5-minute infusion group, 3 of 6 patients with abnormal baseline serum creatinine experienced renal function deterioration after ~~zoledronate~~zoledronate 4 mg, compared with 2 of 8 in the 8/4 mg group and 0 of 4 in the placebo group. The following table is abstracted from applicant’s table 10-19, demonstrating renal deterioration in the patients randomized following amendment 3.

Reviewer table 2X

Patients who experienced renal function deterioration by baseline serum creatinine (15- minute infusion)

	Zoledr 4 mg N (%)	Zoledr 8/4 mg N (%)	Placebo N (%)
Patients with normal baseline creatinine	154	160	143
---with deterioration	17 (11.0%)	19 (11.9%)	10 (7.0%)
Patients with abnormal baseline creatinine	11	21	20
---with deterioration	1 (9.1%)	2 (9.5%)	1 (5.0%)
TOTAL # patients at baseline	165	181	163
---with deterioration	18 (10.9%)	21 (11.6%)	11 (6.7%)

When time to renal function deterioration was analyzed by Cox regression for patients treated post amendment 3, there was no statistically significant difference between

~~zoledronate~~zoledronate and placebo groups. The risk ratio was 1.571 between ~~zoledronate~~zoledronate 4 mg and placebo, indicating a higher risk of renal function deterioration for ~~zoledronate~~zoledronate 4 mg compared with placebo. However, the p-value was not statistically significant.

Applicant’s figure 10-2 is the Kaplan-Meier curve for time to first renal function deterioration by treatment group for patients randomized after the 15-minute infusion amendment.

Applicant figure 10-2

Urinalysis

The percentage of patients with elevated urine protein (> 2) post-baseline was similar in all groups (6-8%). Post-text table 10.3-2.

Vital signs

Systolic and diastolic blood pressure

The incidence of patients with notably high or low systolic or diastolic blood pressures is similar in all groups.

Weight

The incidence of patients with > 10% change in weight was similar for zoledronatezoledronate 4 mg and placebo groups.

ECGs

“The proportion of evaluated patients who had an abnormality was similar at baseline and end of study, but most patients were not evaluated at the end of the study.”

Special safety topics

The ~~zoledronate~~zoledronate 4 mg treatment group had the longest median time to death (203 days), compared to the 8/4 mg (189 days) and placebo (183) groups. This date is from post-text table10.2-7.

Reviewer comment and conclusions

Of the more common adverse events, nausea, vomiting, and dyspnea were higher in the ~~zoledronate~~zoledronate 4 mg group than in the placebo group. There was no clear dose relationship.

The overall incidence of grade 3 and 4 adverse events, regardless of cause, is similar for all the treatment groups. The most frequent SAEs, regardless of study drug relationship, were anemia, dehydration, malignant neoplasm aggravated and dyspnea. Anemia occurred in 7.9% of the ~~zoledronate~~zoledronate 4 mg group and in 3.6% of the placebo group. Dyspnea was present in 10.2% and 7.7%, respectively.

Hypophosphatemia was more common with ~~zoledronate~~zoledronate than placebo. The incidence of grade 3 hypophosphatemia was 8.9%, 12.0% and 1.4%, respectively for zolodronate 4 mg, 8/4 mg and placebo. Grade 3 hypokalemia was slightly higher (4.9%) in the zolodronate 4 mg group compared with placebo (3.2%).

Subsequent to the occurrence of acute renal failure in 3 patients receiving ~~zoledronate~~zoledronate 8 mg in related trials for disease metastatic to bone, several amendments were made to the protocol and a renal advisory board (RAB) of nephrologists was convened. Prolonging the infusion time from 5 to 15 minutes (amendment 3), decreased the incidence of renal function deterioration. Amendment 5 further improved the safety profile by eliminating the 8 mg dose. The safety of the 4 mg dose was improved by requiring assessment of serum creatinine before each dose. ~~Zoledronate~~Zoledronate was held for renal deterioration, as defined in the amended protocol, and therapy was not reinstituted until creatinine returned to within 10% of the baseline value.

Following amendment 3, the overall incidence of renal related AEs was comparable for ~~zoledronate~~zoledronate 4 mg (11.8%) and placebo (12.5%). However, the incidence of “blood creatinine increased” was 2.7% for ~~zoledronate~~zoledronate 4mg and 0.5 for placebo. “Renal failure acute” was 2.2% and 0.5%, respectively.

The incidence of renal function deterioration was increased in the ~~zoledronate~~zoledronate treatment groups for patients with normal baseline creatinine and creatinine > 1.4 compared with placebo pre and post amendment. For patients randomized following amendment 3, the overall incidence of renal deterioration was 10.9% for ~~zoledronate~~zoledronate 4 mg, 11.6% for 4/8 mg and 6.7% for placebo. (From applicant table 10-19)

When time to renal function deterioration was analyzed by Cox regression for patients treated post amendment 3, there was no statistically significant difference between

~~zoledronate~~zoledronate and placebo groups. The risk ratio was 1.571 between ~~zoledronate~~zoledronate 4 mg and placebo, indicating a higher risk of renal function deterioration for ~~zoledronate~~zoledronate 4 mg compared with placebo. However, the p-value was not statistically significant.

Reviewer Conclusion: ~~Zoledronate~~Zoledronate 4 mg i.v. over 15 minutes every 3 weeks has an acceptable risk profile. The risk of renal deterioration with ~~zoledronate~~zoledronate is greater than placebo, but acceptable with clinical monitoring of renal function before each dose. Risk of renal toxicity increases with cumulative time under therapy. Caution is indicated for patients with elevated baseline creatinine, particularly since the study population excluded patients with creatinine>3.0 and the drug is excreted by the kidneys.

AAppendix 3 Review of ~~Zoledronate~~Zoledronate Safety in Breast Cancer and Myeloma,

Study 010

~~Study # ZOL42446-03-010:~~ “A randomized, double-blind, multi-center, comparative trial of i.v. zoledronic acid (4 mg or 8 mg) versus i.v. Aredia (90 mg), as an adjunct to standard therapies, in the treatment of multiple myeloma and breast cancer patients with cancer-related bone lesions.”

This is a multi-center, international study (248 centers in 20 countries). The first patient was enrolled 16 October 1998 and the last patient completed therapy 12 January 2001. The study duration was 13 months (12 months treatment, 1 month observation), with an optional 12 month extension period.

The study population consisted of adults with Stage III multiple myeloma or stage IV breast cancer. Patients were to be receiving anti-cancer therapy at the time of randomization. Breast cancer patients could be receiving chemotherapy and/or first or second line hormonal therapy. Myeloma patients were required to have at least one osteolytic bone lesion on conventional (plain film) radiograph. Breast cancer patients were required to have a lytic, sclerotic or mixed lesion on plain film. Patients were to be performance status < 2.

~~Exclusions included serum creatinine > 3, bilirubin > 2.5, and corrected serum calcium < 8 mg/dL or~~ > ~~12 mg/dL.~~

~~Patients were stratified according to the diagnosis of multiple myeloma or breast cancer on chemotherapy or breast cancer on hormonal therapy.~~

~~Patients were randomized to receive Zometa 4 mg or 8 mg or Aredia 90 mg i.v. every~~ 3 ~~or 4 weeks~~ ~~for 12 months. Initially, Zometa was to be infused over 5 minutes. Aredia was infused over 2 hours. Patients were not taken off study for progressive disease.~~

From December 1998 to May 1999 there were 3 renal failure SAEs in patients receiving Zometa 8mg, manifested by progressively rising serum creatinine. This occurred in 1 patient each of a Phase II myeloma study (cycle #11), protocol 010 (diagnosis breast cancer, cycle #1), and protocol 039 (prostate cancer, after cycle #7). The patients were, respectively, ages 75, 80, and 71, with baseline creatinines of 1.6, 1.9, and 1.4. These events led to Amendment 2 on 25 June 1999 and Amendment 5 on 7 June 2000 and the formation of a Renal Advisory Board (RAB) in November 1999.

After amendment 2, the infusion duration was prolonged to 15 minutes for Zometa. After amendment 5, all Zometa patients received 4 mg per dose. Serum creatinine was to be evaluated before each dose, and treatment held for deterioration of creatinine, as defined in the amendment, until the level was within 10% of the baseline creatinine.

~~Applicant assessment and analysis of safety~~

Adverse events, serious adverse events, laboratory studies, and survival data were the main safety variables. Physical examination, assessment for adverse events and laboratory studies were done at baseline and then every 3-4 weeks, and at the end of study. Chemistries were repeated every 3-4 weeks. Hematology and urine chemistries and urinalyses were repeated at 3-4 weeks, then at month 3, then every 3 months. Serum creatinine was measured prior to each dose of study drug following amendment 5, so that treatment could be held for renal deterioration. The time to discontinuation of study and duration of survival were assessed.

Safety analysis was based on the type and frequency of adverse events and laboratory values outside of pre-determined ranges. The data were tabulated. Data were cut at the end of the study drug period for laboratory and adverse event analysis. This was the later of the end of the core study phase or the last date of study medication plus 28 days. All data was included up to the date of the data base lock for time to death and renal deterioration. The last visit date was used for other safety parameters.

~~Study population~~

~~There were 1648 patients randomized to the following treatment groups:~~

Zometa 4 mg #564

Zometa 8/4 mg #526

Aredia 90 mg #558

Five patients were not included in the safety analysis because they did not receive study drug. The safety population includes all patents that were randomized and received study drug. “There were 9, 9, and 10 patients who were randomized to the incorrect stratum” for Zometa 4 mg, 8/4 mg, and Aredia, respectively. They were re-assigned for safety analysis.

A similar % of patients did not complete the study for all treatment groups: 37.3% for Zometa 4 mg, 40.3% for 8/4 mg and 39.2% for Aredia 90 mg. Discontinuation was most frequently due to adverse events and death. The % of discontinuation for death in each group was 10.8%, 10.7% and 11.5%, respectively.

~~The number of patients in each arm is shown in applicant’s table 7-2.~~

~~Applicant table 7-2~~

~~Reviewer Note~~: Analysis of data provided by the applicant demonstrates that for patients assigned to the 8/4 mg group, 1394 of 7072 Zometa doses delivered were actually 4 mg rather than 8 mg doses, or approximately 20% were 4 mg doses.

The following table is an abbreviated, composite version of applicant’s table 7-3 and 7-4, to illustrate baseline age and serum creatinine. These are some of the patient factors that might affect susceptibility to the renal toxic effects of treatment.

~~Reviewer table X.~~

~~Baseline patient age and serum creatinine by treatment group~~

This data shows similar age for treatment groups whether viewed as median age or % older than 60 years of age. The baseline serum creatinine values are similar for the Zometa groups, but less favorable compared with the Aredia patients, of whom only 13.2% had abnormal creatinine > ~~1.4.~~

~~Reviewer comment~~~~: The treatment related factors that might predict for renal toxicity include dose, duration of infusion, exposure over time (# of months of treatment), and concomitant exposure to renal toxic drugs.~~

~~Overall exposure~~

The duration of exposure to study drug was similar for the treatment groups. The mean duration of exposure for the multiple myeloma patients and hormonal breast cancer patients was 10 months. The mean for chemotherapy breast cancer patients was 9 months. For the overall group, the mean was approximately 10 months and the median exposure approximately 12 months for each of the 3 treatment groups. (This information is found in applicant’s table 8-1).

~~Concomitant medication~~

Most patients received antineoplastic therapy during the study. The most common chemotherapy drugs were cyclophosphamide and doxorubicin. Post-text table 8.2-2 and 8.2-3 show # of patients who received antineoplastic therapy prior to and after the start of the study respectively. The incidence of prior therapy with cisplatin was 2%, 1.0% and 0.9% for Zometa 4, 8/4 and Aredia. The incidence of therapy with cisplatin after the start of the study was 2.0%, 1.0%, and 0.7%, respectively.

~~Overall incidence and severity of adverse events:~~

Clinical study reports used the IMN dictionary to code adverse events (AEs). The data is summarized in the study report using MedDRA preferred terms. Adverse events were mapped from the IMN preferred terms to the corresponding MedDRA terms. The number of patients with AEs is provided by body system, preferred term, and treatment group. If a patient had more than one type of event within a body system, the patient was counted only once.

~~Most patients experienced at least one adverse event.~~

~~Applicant table 10-1 shows the % patients with adverse events overall and by system for events occurring~~ > ~~15% in any group.~~

~~Applicant table 10-1~~

There is no major imbalance in AEs by system, although the incidence of “metabolism and nutrition disorders” and “neoplasms benign and malignant” is slightly higher for Zometa 4 than Aredia. The incidence of renal AEs is similar for these treatment groups, including for patients treated with 5 minute and 15 minute infusion.

The most frequent adverse events were bone pain, nausea, fatigue, pyrexia, and vomiting. Of these, only pyrexia was present more frequently in the Zometa 4 mg and 8/4 mg groups than in the Aredia group. There was a slight increase in the Zometa 4 mg group compared with the Aredia group of cough, arthralgia, weakness and anorexia. This information is summarized in applicant table 10-2.

~~Applicant table 10-2~~

~~The overall incidence of grade 3 and 4 AEs was similar for the treatment groups. Post-text table 10.1-4 was reviewed.~~

~~Renal adverse events~~

A Renal Advisory Board (RAB) was established in November 1999 to monitor the renal safety of Zometa, due to concerns about renal toxicity when a total of 3 patients in this and other studies developed renal failure following treatment with Zometa 8 mg. Renal adverse events with preferred terms specified by the RAB are presented in applicant table 10-3 for patients randomized prior to the 15-minute infusion amendment and table 10-4 for patients randomized following the amendment. The overall incidence of renal adverse events is similar for Zometa 4 mg and Aredia 90 mg both pre and post amendment. The incidence for Zometa 8/4 mg is approximately 30% higher than the other treatment groups for the 5-minute and 15-minute infusions. For the post amendment patients, the incidence of “blood creatinine increased” is 2.8% for Zometa 4 mg, 4.3% for Zometa 8/4 mg and l.8% for Aredia 90 mg. The incidence of “renal impairment NOS is identical for the Zometa 4 mg and Aredia groups, as it for renal failure acute. The incidence for both events is higher for Zometa 8/4 mg. The incidence of 4.7% “renal failure acute” is significantly higher for the 8/4 mg dose Zometa, than for the lower dose Zometa or Aredia, which are both 0.7%.

~~Applicant table 10.4~~

The sponsor selected and reviewed adverse events previously reported with bisphosphonates. These are reported in table 10-5. The incidence of arthralgias and fever was higher for Zometa 4 mg than Aredia, but the incidence for Zometa 8/4 mg was intermediate, suggesting that there is no causal relationship for these events. The incidence of electrolyte AEs was higher for both Zometa groups (15.3% and 19.1%), possibly in a dose-dependant way, compared with Aredia (11.9%)

~~Applicant table 10-5~~

~~Events believed to be chemotherapy-associated were not increased in incidence for Zometa patients compared with those that received Aredia.~~

~~Deaths, SAEs and other significant adverse events~~

~~In Section 3.5.3 of the protocol, a serious adverse event (SAE) is defined as an untoward event which:~~

~~Is fatal or life-threatening,~~

~~Required or prolonged hospitalization,~~

~~Was significantly or permanently disabling or incapacitating,~~

~~Constitutes a congenital anomaly or a birth defect,~~

~~May jeopardize the subject and may require medical or surgical intervention to prevent one of the outcomes listed above.~~

Applicant table 10-7 lists patients who died, had other SAE or “clinically significant AEs” or discontinued because of them. Deaths are included if they occurred up to the end of the core study phase or within 28 day following the last treatment.

~~Applicant table 10-7~~

There were 217 deaths during the core study phase or within 28 days of last study medication. The percentage of deaths in each treatment arm was similar. The most frequent cause was progression of cancer. There were no deaths due to renal causes in the Zometa 4 mg or Aredia 90 groups. There were 4 such deaths in Zometa 8/4 mg.

SAEs occurred in a similar proportion of patients in each treatment group (from post-text table 10.2-1, 48.8%, 48.3% and 49.8%). The most frequent SAE was infection for all groups. The incidence of “renal and urinary” SAEs was 2.8% for Zometa 4 mg, 5.7% for 8/4 mg and 2.9% for Aredia. The incidence for “renal failure acute” was 0.5%, 3.4% and 1.4%, respectively. The incidence for which “renal and urinary” SAEs were suspected to be treatment related were 0.5%, 1.9% and 0.2%, respectively.

The study report provides specific information about 27 SAEs that were not included in the database at the time of the lock, but the SAEs were not felt to be study drug related. Of these, one breast cancer patient was treated with Aredia and had unexplained sudden death. One additional breast patient (Zometa 4 mg) had azotemia and hypercalcemia at study day 400, for which she was hospitalized. A myeloma patient (Zometa 8/4 mg) had “renal impairment NOS” at study day 80.

~~In this study, skeletal related events (SRE) were reported as SAEs. The percentage was 5.0%, 3.8%, and 7.7%, respectively for Zometa 4 and 8/4 mg and for Aredia.~~

The incidence of patients discontinuing therapy for adverse events was 10.1% for Zometa 4 mg, 12.2% for 8/4 mg, and 8.8% for Aredia. Discontinuations for “renal and urinary disorders” was 0.5% for Zometa 4 mg and Aredia 90 mg. It was 2.5% for Zometa 8/4 mg.

~~Laboratory parameters~~

~~Hematology~~

~~The treatment groups were similar when hematology tests were analyzed by number of patients reaching selected “notable” values of neutrophil count < 0.5 x 10~~⁹~~, platelets < 25 x 10~~⁹~~, and hemoglobin < 6.5 gm/dL. However, a greater percentage of patients in the~~ ~~Zometa 4 mg (10.2%) and 8/4 mg (11.1%) had a decrease of >25% from baseline hemoglobin~~ ~~compared with the Aredia group (7.8%). The incidence of grade 3 and grade 4 hematologic toxicity is similar for the treatment groups.~~

~~Serum chemistry~~

~~hypophosphatemia and grade 3 hypokalemia for Zometa 8/4 mg.~~

~~Serum creatinine and renal function deterioration~~

~~Baseline serum creatinine was considered normal if <1.4 mg/dL and abnormal if~~ > ~~1.4 mg/dL. Renal function deterioration was defined as any of the following:~~

~~Normal baseline with change from baseline~~ > ~~0.5 mg/dL~~

~~Abnormal baseline with change from baseline~~ > ~~1.0 mg/dL~~

~~Post-baseline value~~ > ~~2 time the baseline value.~~

Applicant table 10-14 shows the incidence of post-baseline notable serum creatinine for patients randomized before and after the 15 minute infusion amendment. For post-amendment patients, 10.7%, 18.6% and 7.5% of Zometa 4, 8/4 and Aredia patients, respectively, developed creatinine increase from baseline > ~~0.5. Patients in each group who developed creatinine~~ > ~~4.5 were 0 for Zometa 4 mg, 1.1 % (3 patients) for 8/4 mg and 0.8% (2 patients) for Aredia.~~

~~Applicant table 10-14~~

There was a decrease in the post-baseline notable serum creatinine for both Zometa groups when the infusion time was prolonged from 5 minutes to 15 minutes. The incidence remains significantly elevated for Zometa 8/4 mg but compared with the two other treatment groups, which are similar.

~~Applicant table 10-16 shows the distribution of post-baseline grade 3 or 4 creatinine by treatment group.~~

~~Applicant table 10-16~~

~~For the patients randomized post amendment, the incidence of grade 3 and 4 creatinine was least in the Zometa 4 mg group. The incidence seems only slightly higher for Zometa 8/4 mg than Aredia.~~

~~Reviewer comment~~

~~Reviewer table X~~

~~Patients who experienced renal function deterioration by baseline serum creatinine (15-minute infusion)~~

	~~Zoledr 4 mg~~ ~~N (%)~~	~~Zoledr 8/4 mg~~ ~~N (%)~~	~~Aredia 90 mg~~ ~~N (%)~~
~~Patients with normal baseline creatinine~~	~~246~~	~~242~~	~~246~~
~~---with deterioration~~	~~23 (9.3%)~~	~~43 (17.8%)~~	~~20 (8.1%)~~
~~Patients with abnormal baseline creatinine~~	26	21	22
~~---with deterioration~~	~~1 (3.8%)~~	~~6 (28.6%)~~	~~2 (9.1%)~~
~~TOTAL # patients at baseline~~	~~272~~	~~263~~	~~268~~
~~---with deterioration~~	~~24 (8.8%)~~	~~49 (18.6%)~~	~~22 (8.2%)~~

Study # ZOL42446-03-010: “A randomized, double-blind, multi-center, comparative trial of i.v. zoledronic acid (4 mg or 8 mg) versus i.v. Aredia (90 mg), as an adjunct to standard therapies, in the treatment of multiple myeloma and breast cancer patients with cancer-related bone lesions.”

Exclusions included serum creatinine > 3, bilirubin > 2.5, and corrected serum calcium < 8 mg/dL or > 12 mg/dL.

Patients were stratified according to the diagnosis of multiple myeloma or breast cancer on chemotherapy or breast cancer on hormonal therapy.

Patients were randomized to receive zoledronate 4 mg or 8 mg or Aredia 90 mg i.v. every 3 or 4 weeks for 12 months. Initially, zoledronate was to be infused over 5 minutes. Aredia was infused over 2 hours. Patients were not taken off study for progressive disease.

From December 1998 to May 1999 there were 3 renal failure SAEs in patients receiving zoledronate 8mg, manifested by progressively rising serum creatinine. This occurred in 1 patient each of a Phase II myeloma study (cycle #11), protocol 010 (diagnosis breast cancer, cycle #1), and protocol 039 (prostate cancer, after cycle #7). The patients were, respectively, ages 75, 80, and 71, with baseline creatinines of 1.6, 1.9, and 1.4. These events led to Amendment 2 on 25 June 1999 and Amendment 5 on 7 June 2000 and the formation of a Renal Advisory Board (RAB) in November 1999.

After amendment 2, the infusion duration was prolonged to 15 minutes for zoledronate. After amendment 5, all zoledronate patients received 4 mg per dose. Serum creatinine was to be evaluated before each dose, and treatment held for deterioration of creatinine, as defined in the amendment, until the level was within 10% of the baseline creatinine.

Applicant assessment and analysis of safety

Study population

There were 1648 patients randomized to the following treatment groups:

Zoledronate 4 mg #564

Zoledronate 8/4 mg #526

Aredia 90 mg #558

Five patients were not included in the safety analysis because they did not receive study drug. The safety population includes all patents that were randomized and received study drug. “There were 9, 9, and 10 patients who were randomized to the incorrect stratum” for zoledronate 4 mg, 8/4 mg, and Aredia, respectively. They were re-assigned for safety analysis.

A similar % of patients did not complete the study for all treatment groups: 37.3% for zoledronate 4 mg, 40.3% for 8/4 mg and 39.2% for Aredia 90 mg. Discontinuation was most frequently due to adverse events and death. The % of discontinuation for death in each group was 10.8%, 10.7% and 11.5%, respectively.

The number of patients in each arm is shown in applicant’s table 7-2.

Applicant table 7-2

Reviewer Note: Analysis of data provided by the applicant demonstrates that for patients assigned to the 8/4 mg group, 1394 of 7072 zoledronate doses delivered were actually 4 mg rather than 8 mg doses, or approximately 20% were 4 mg doses.

Reviewer table 1.

Baseline patient age and serum creatinine by treatment group

This data shows similar age for treatment groups whether viewed as median age or % older than 60 years of age. The baseline serum creatinine values are similar for the zoledronate groups, but less favorable compared with the Aredia patients, of whom only 13.2% had abnormal creatinine > 1.4.

Reviewer comment: The treatment related factors that might predict for renal toxicity include dose, duration of infusion, exposure over time (# of months of treatment), and concomitant exposure to renal toxic drugs.

Overall exposure

Concomitant medication

Most patients received antineoplastic therapy during the study. The most common chemotherapy drugs were cyclophosphamide and doxorubicin. Post-text table 8.2-2 and 8.2-3 show # of patients who received antineoplastic therapy prior to and after the start of the study respectively. The incidence of prior therapy with cisplatin was 2%, 1.0% and 0.9% for zoledronate 4, 8/4 and Aredia. The incidence of therapy with cisplatin after the start of the study was 2.0%, 1.0%, and 0.7%, respectively.

Overall incidence and severity of adverse events:

Clinical study reports used the IMN dictionary to code adverse events (AEs). The data is summarized in the study report using MedDRA preferred terms. Adverse events were mapped from the IMN preferred terms to the corresponding MedDRA terms. The number of patients with AEs is provided by body system, preferred term, and treatment group. If a patient had more than one type of event within a body system, the patient was counted only once.

Most patients experienced at least one adverse event.

Applicant table 10-1 shows the % patients with adverse events overall and by system for events occurring > 15% in any group.

Applicant table 10-1

There is no major imbalance in AEs by system, although the incidence of “metabolism and nutrition disorders” and “neoplasms benign and malignant” is slightly higher for zoledronate 4 than Aredia. The incidence of renal AEs is similar for these treatment groups, including for patients treated with 5 minute and 15 minute infusion.

The most frequent adverse events were bone pain, nausea, fatigue, pyrexia, and vomiting. Of these, only pyrexia was present more frequently in the zoledronate 4 mg and 8/4 mg groups than in the Aredia group. There was a slight increase in the zoledronate 4 mg group compared with the Aredia group of cough, arthralgia, weakness and anorexia. This information is summarized in applicant table 10-2.

Applicant table 10-2

The overall incidence of grade 3 and 4 AEs was similar for the treatment groups. Post-text table 10.1-4 was reviewed.

Renal adverse events

A Renal Advisory Board (RAB) was established in November 1999 to monitor the renal safety of zoledronate, due to concerns about renal toxicity when a total of 3 patients in this and other studies developed renal failure following treatment with zoledronate 8 mg. Renal adverse events with preferred terms specified by the RAB are presented in applicant table 10-3 for patients randomized prior to the 15-minute infusion amendment and table 10-4 for patients randomized following the amendment. The overall incidence of renal adverse events is similar for zoledronate 4 mg and Aredia 90 mg both pre and post amendment. The incidence for zoledronate 8/4 mg is approximately 30% higher than the other treatment groups for the 5-minute and 15-minute infusions. For the post amendment patients, the incidence of “blood creatinine increased” is 2.8% for zoledronate 4 mg, 4.3% for zoledronate 8/4 mg and l.8% for Aredia 90 mg. The incidence of “renal impairment NOS is identical for the zoledronate 4 mg and Aredia groups, as it for renal failure acute. The incidence for both events is higher for zoledronate 8/4 mg. The incidence of 4.7% “renal failure acute” is significantly higher for the 8/4 mg dose zoledronate, than for the lower dose zoledronate or Aredia, which are both 0.7%.

Applicant table 10.4

The sponsor selected and reviewed adverse events previously reported with bisphosphonates. These are reported in table 10-5. The incidence of arthralgias and fever was higher for zoledronate 4 mg than Aredia, but the incidence for zoledronate 8/4 mg was intermediate, suggesting that there is no causal relationship for these events. The incidence of electrolyte AEs was higher for both zoledronate groups (15.3% and 19.1%), possibly in a dose-dependant way, compared with Aredia (11.9%)

Applicant table 10-5

Events believed to be chemotherapy-associated were not increased in incidence for zoledronate patients compared with those that received Aredia.

Deaths, SAEs and other significant adverse events

In Section 3.5.3 of the protocol, a serious adverse event (SAE) is defined as an untoward event which:

1. Is fatal or life-threatening,

2. Required or prolonged hospitalization,

3. Was significantly or permanently disabling or incapacitating,

4. Constitutes a congenital anomaly or a birth defect,

5. May jeopardize the subject and may require medical or surgical intervention to prevent one of the outcomes listed above.

Applicant table 10-7

There were 217 deaths during the core study phase or within 28 days of last study medication. The percentage of deaths in each treatment arm was similar. The most frequent cause was progression of cancer. There were no deaths due to renal causes in the zoledronate 4 mg or Aredia 90 groups. There were 4 such deaths in zoledronate 8/4 mg.

SAEs occurred in a similar proportion of patients in each treatment group (from post-text table 10.2-1, 48.8%, 48.3% and 49.8%). The most frequent SAE was infection for all groups. The incidence of “renal and urinary” SAEs was 2.8% for zoledronate 4 mg, 5.7% for 8/4 mg and 2.9% for Aredia. The incidence for “renal failure acute” was 0.5%, 3.4% and 1.4%, respectively. The incidence for which “renal and urinary” SAEs were suspected to be treatment related were 0.5%, 1.9% and 0.2%, respectively.

The study report provides specific information about 27 SAEs that were not included in the database at the time of the lock, but the SAEs were not felt to be study drug related. Of these, one breast cancer patient was treated with Aredia and had unexplained sudden death. One additional breast patient (zoledronate 4 mg) had azotemia and hypercalcemia at study day 400, for which she was hospitalized. A myeloma patient (zoledronate 8/4 mg) had “renal impairment NOS” at study day 80.

In this study, skeletal related events (SRE) were reported as SAEs. The percentage was 5.0%, 3.8%, and 7.7%, respectively for zoledronate 4 and 8/4 mg and for Aredia.

The incidence of patients discontinuing therapy for adverse events was 10.1% for zoledronate 4 mg, 12.2% for 8/4 mg, and 8.8% for Aredia. Discontinuations for “renal and urinary disorders” was 0.5% for zoledronate 4 mg and Aredia 90 mg. It was 2.5% for zoledronate 8/4 mg.

Laboratory parameters

Hematology

The treatment groups were similar when hematology tests were analyzed by number of patients reaching selected “notable” values of neutrophil count < 0.5 x 10⁹, platelets < 25 x 10⁹, and hemoglobin < 6.5 gm/dL. However, a greater percentage of patients in the zoledronate 4 mg (10.2%) and 8/4 mg (11.1%) had a decrease of >25% from baseline hemoglobin compared with the Aredia group (7.8%). The incidence of grade 3 and grade 4 hematologic toxicity is similar for the treatment groups.

Serum chemistry

The tests performed include sodium, potassium, SGOT, SGPT, alkaline phosphatase, LDH, creatinine, BUN, calcium, phosphorus and magnesium. When patients were analyzed for notable biochemistry values (other than creatinine), there were few differences between the treatment groups. There was a higher incidence of hypercalcemia > 12.0 mg/dL in the Aredia group (2.24%) than in the zoledronate 4 (0.37%) or 8/4 mg groups (0.60%). There was more hypokalemia < 3 mEq/L in the zoledronate 8/4 mg group (9.84%) and more hyperkalemia > 6 mEq/L (4.68%) in the zoledronate 4 mg group compared with the other 2 treatment groups. The incidence of grade 3 and grade 4 toxicity was similar for the treatment groups, except for excess grade 3 hypophosphatemia and grade 3 hypokalemia for zoledronate 8/4 mg. Zoledronate 4 mg patients had 3% grade 4 hyperkalemia, but only 0.8% for zoledronate 8/4 mg and 1.3% for Aredia. The following table shows notable chemistry abnormalities by treatment group.

Applicant table 10-15 (abbreviated)

Reviewer comment : There were more electrolyte abnormalities for zoledronate 8/4 mg. However, comparing zoledronate 4 mg and Aredia, there are probably no significant differences. The lack of dose relationship for hyperkalemia in the zoledronate groups makes it unlikely to be a treatment–related occurrence.

Serum creatinine and renal function deterioration

Baseline serum creatinine was considered normal if <1.4 mg/dL and abnormal if > 1.4 mg/dL. Renal function deterioration was defined as any of the following:

· Normal baseline with change from baseline > 0.5 mg/dL

· Abnormal baseline with change from baseline > 1.0 mg/dL

· Post-baseline value > 2 time the baseline value.

Applicant table 10-14 shows the incidence of post-baseline notable serum creatinine for patients randomized before and after the 15-minute infusion amendment. For post-amendment patients, 10.7%, 18.6% and 7.5% of zoledronate 4, 8/4 and Aredia patients, respectively, developed creatinine increase from baseline > 0.5. Patients in each group who developed creatinine > 4.5 were 0 for zoledronate 4 mg, 1.1 % (3 patients) for 8/4 mg and 0.8% (2 patients) for Aredia.

Applicant table 10-14

There was a decrease in the frequency of post-baseline notable serum creatinine for both zoledronate groups when the infusion time was prolonged from 5 minutes to 15 minutes. The incidence remains significantly elevated for zoledronate 8/4 mg compared with the two other treatment groups. The incidence of post-baseline notable serum creatinine elevation is 10.7% for zoledronate 4 mg and 8.3% for Aredia (7.5% + 0.8%). Applicant table 10-16 shows the distribution of post-baseline grade 3 or 4 creatinine by treatment group.

Applicant table 10-16

For the patients randomized post amendment, the incidence of grade 3 and 4 creatinine was least in the zoledronate 4 mg group. The incidence of grade 3 plus grade 4 creatinine is slightly higher for zoledronate 8/4 mg (2.7%) than Aredia (1.9%).

For the pre 15-minute infusion amendment patients, renal function deterioration was experience by 13.2% of the zoledronate 4 mg patients, 20.4% of the zoledronate 8/4 mg patients, and by 6.7% of the Aredia patients. Extending the time of the infusion significantly decreased the incidence of creatinine deterioration for the zoledronate 4 mg patients. The following table is abstracted from applicant’s table 10-19 and shows renal function deterioration by baseline creatinine for post 15-minute infusion amendment patients.

Reviewer table 2

Patients who experienced renal function deterioration by baseline serum creatinine (15-minute infusion)

	Zoledr 4 mg N (%)	Zoledr 8/4 mg N (%)	Aredia 90 mg N (%)
Patients with normal baseline creatinine	246	242	246
---with deterioration	23 (9.3%)	43 (17.8%)	20 (8.1%)
Patients with abnormal baseline creatinine	26	21	22
---with deterioration	1 (3.8%)	6 (28.6%)	2 (9.1%)
TOTAL # patients at baseline	272	263	268
---with deterioration	24 (8.8%)	49 (18.6%)	22 (8.2%)

Zoledronate 4 mg was comparable to Aredia 90 mg for the entire population and for patients with normal baseline creatinine. For the small number of patients with abnormal baseline creatinine, zoledronate 4 mg had a lower incidence of creatinine deterioration (3.8%) than Aredia 90 mg (9.1%). Zoledronate 8/4 mg was more renal toxic than either zoledronate 4 mg or Aredia 90 mg for patients with normal and abnormal baseline creatinine.

When time to renal function deterioration was analyzed by Cox regression for patients randomized after the 15-minute infusion amendment, there was no statistically significant difference between zoledronate 4 mg and Aredia 90 mg. The risk ratio was 0.984 between zoledronate 4 mg and Aredia 90 mg groups. However, there was a statistically significant difference in terms of time to renal function deterioration for each of zoledronate 4 mg and Aredia 90 mg compared with zoledronate 8/4 mg. The risk ratios were 2.372 and 2.379, respectively.

Applicant’s figure 10-2 is the Kaplan-Meier curve for time to first renal function deterioration by treatment group for patients randomized after the 15-minute infusion amendment.

Applicant figure 10-2

Urinalysis

The percentage of patients with elevated post-baseline urine protein was similar for zoledronate 4 mg and Aredia 90 mg.

Vital signs, other safety evaluations, special safety topics

Vital signs were not collected. There were no other clinically relevant findings.

Reviewer comment and conclusions

Increased risk of renal function deterioration is the only significant safety issue for zoledronate in this and related clinical trials of long-term treatment for patients with metastatic malignancy to bone. Toxicity was dose-related and exposure over time-related.

The more common adverse events were bone pain, nausea, fatigue, pyrexia, and vomiting. Pyrexia, cough, arthralgia, weakness and anorexia were slightly increased in the Zoledronate 4 mg group compared with the Aredia 90 mg group. The incidence of grade 3 and 4 AEs was similar.

More patients in the zoledronate 4 mg group had a decrease of > 25% from baseline hemoglobin compared with the Aredia patients, although there was no difference in notable hematology values or grade 3 and 4 hematologic toxicity.

There were no major differences in blood chemistry abnormalities for zoledronate 4 mg, compared with Aredia.

Subsequent to the occurrence of acute renal failure in 3 patients receiving zoledronate 8 mg in trials for malignancy metastatic to bone, several amendments were made to the protocol, which improved safety. The infusion time of zoledronate was increased from 5 to 15 minutes (amendment 2), which decreased the incidence of renal function deterioration in the zoledronate groups. Amendment 5 further improved the safety profile by eliminating the 8 mg dose. The safety of the 4 mg dose was improved by requiring assessment of serum creatinine before each dose. Zoledronate was held for renal deterioration, as defined in the amended protocol, and therapy was not reinstituted until creatinine returned to within 10% of the baseline value.

The overall incidence of renal adverse events was similar for zoledronate 4 mg and Aredia 90 mg both pre and post the amendment which prolonged the infusion time. The incidence of renal function deterioration with zoledronate 4 mg was higher than Aredia pre-amendment, but was similar (8.8% and 8.2%, respectively) when zoledronate was infused over 15 minutes. Zoledronate was comparable to Aredia for the entire population of patients and for patients with normal baseline creatinine. For the small number of patients with abnormal baseline creatinine, zoledronate 4 mg had a lower incidence of creatinine deterioration (3.8%) than Aredia 90 mg (9.1%). There was no statistically significant difference in time to first renal function deterioration for zoledronate 4 mg compared with Aredia 90 mg.

Reviewer conclusions: Zoledronate 4 mg i.v. over 15 minutes every 3-4 weeks has an acceptable risk profile that is comparable to Aredia. To minimize renal risk, zoledronate must be infused over 15 minutes and clinical monitoring of serum creatinine should be performed before each dose. Caution is indicated for patient with elevated serum creatinine, particularly since the study population excluded patients with creatinine > 3.0 and the drug is excreted unchanged by the kidneys.

~~Applicant’s figure 10-2 is the Kaplan-Meier curve for time to first renal function deterioration by treatment group for patients randomized after the 15-minute infusion amendment.~~

~~Applicant figure 10-2~~