382 patients receiving chemotherapy were randomized, 185 to pamidronate disodium and 197 to placebo. 372 patients receiving hormonal therapy were randomized, 182 to pamidronate disodium and 190 to placebo. All but three patients were eva luable for efficacy. Patients were followed for 24 months of therapy or until they went off study. Median duration of follow-up was 13 months in patients receiving chemotherapy and 17 months in patients receiving hormone therapy. Twenty-five percent of the patients in the chemotherapy study and 37% of the patients in the hormone therapy study received pamidronate disodium for 24 months. The efficacy results are shown in the table below:

*
Fractures and radiation to bone were two of several secondary endpoints. The statistical significance of these analyses may be overestimated since numerous analyses were performed.
†
NR = Not Reached
 Breast Cancer Patients
Receiving Chemotherapy
 Breast Cancer Patients
Receiving Hormonal Therapy
 
Any SRE
 Radiation
 Fractures
 Any SRE
 Radiation
 Fractures
 
Pamidronate disodium
  Placebo
 Pamidronate disodium
  Placebo
 Pamidronate disodium
  Placebo
 Pamidronate disodium
  Placebo
 Pamidronate disodium
  Placebo
 Pamidronate disodium
  Placebo
 
N
 185
 195
 185
 195
 185
 195
 182
 189
 182
 189
 182
 189
 
Skeletal
Morbidity Rate
(#SRE/year)
Mean
 2.5
 3.7
 0.8
 1.3
 1.6
 2.2
 2.4
 3.6
 0.6
 1.2
 1.6
 2.2
 
P-Value
 <.001
   <.001*
 .018*
 .021
 .013*
 .040*
 
Proportion of
patients having
 
an SRE
 46%
  65%
 28%
  45%
 36%
  49%
 55%
  63%
 31%
  40%
 45%
  55%
 
P-Value
 <.001
 <.001*
 .014*
 .094
 .058*
 .054*
 
Median Time to
SRE (months)
 13.9
  7.0
 NR†
  14.2
 25.8
  13.3
 10.9
  7.4
 NR†
  23.4
 20.6
  12.8
 
P-Value
 <.001
 <.001*
 .009*
 .118
 .016*
 .113*
 

Bone lesion response was radiographically assessed at baseline and at 3, 6, and 12 months. The complete + partial response rate was 33% in pamidronate disodium patients and 18% in placebo patients treated with chemotherapy (P=0.001). No difference was seen between pamidronate disodium and placebo in hormonally-treated patients.

Pain and analgesic scores, ECOG performance status and Spitzer quality of life index were measured at baseline and periodically during the trials. The changes from baseline to the last measurement carried forward are shown in the table below:

Mean Change (Δ) from Baseline at Last Measurement  *
The statistical significance of analyses of