Patients were enrolled between May, 2016 and July, 2019. The data cutoff date for this primary analysis was August, 2020. Of the 84 patients assessed for eligibility, 40 patients randomly assigned to ZA group and only 16 patients completed the study protocol, while 13 patients among the 44 patients in the control group (figure 1). Patients were off the study generally had once ZA infusion, did not regularly take calcium and vitamin D3 orally, were lack of BMD reports or serum BMD turnover biomarkers. Baseline characteristics were generally balanced between the two groups (Table 1). Overall, the median age was 54 years old with a median BMI 23.29, the ratio of male to female was 1:1, and 73% of female were postmenopausal. There were 82.75% of patients diagnosed aggressive NHL and majority staged I-II (68.97%). The primary chemotherapy regimens were including CHOP (like), VIPD, DDGP, VDLP, R2-EPOCH and BFM 90[17], which all contained high does glucocorticoid. At baseline, the BMD biomarker of serum calcium, 25-hydroxy vitamin D and PTH were all in the normal range.
Table 1
Baseline characteristics of patients for randomization.
|
Total patients
N=29
n (%)
|
Zoledronic Acid group
N=16
n (%)
|
Control group N=13
n (%)
|
Zoledronic Acid group vs. control group
p value
|
Median age (y)
|
54 (29-76)
|
53 (37-70)
|
55 (29-76)
|
0.948
|
BMI
|
23.39
(20.14-29.76)
|
22.68
(20.20-27.11)
|
23.88
(20.14-29.76)
|
0.351
|
Sex
|
0.462
|
Male
|
14 (48.27%)
|
9 (56.25%)
|
5 (38.46%)
|
|
Female
|
15 (51.72%)
|
7 (43.75%)
|
8 (61.54%)
|
|
Postmenopausal
|
11 (37.93%)
|
4 (57.14%)
|
7 (87.5%)
|
|
Diagnosis
|
|
|
|
1.000
|
Aggressive
|
24 (82.75%)
|
13 (81.25%)
|
11 (84.62%)
|
|
Indolent
|
5 (17.24%)
|
3 (18.75%)
|
2 (15.38%)
|
|
Ann Arbor staging
|
0.688
|
I-II
|
20 (68.97%)
|
12 (75%)
|
8 (61.54%)
|
|
III-IV
|
9 (30.03%)
|
4 (25%)
|
5 (38.46%)
|
|
Treatment
|
0.130
|
CHOP (like)
|
20 (68.97%)
|
9 (56.25%)
|
11 (84.62%)
|
|
VIPD
|
1 (3.44%)
|
1 (5.25%)
|
0
|
|
DDGP
|
3 (10.34%)
|
3 (18.75%)
|
0
|
|
VDLP
|
3 (10.34%)
|
2 (12.5%)
|
1 (7.69%)
|
|
R2-EPOCH
|
1 (3.44%)
|
1 (6.25%)
|
0
|
|
BFM-90
|
1 (3.44%)
|
0
|
1 (7.69%)
|
|
Calcium
|
2.27
(2.14-4.38)
|
2.28
(2.14-2.43)
|
2.27
(2.19-4.38)
|
0.714
|
25-hydroxy vitamin D
|
37.2
(11.77-71.62)
|
39.60
(11.77-71.68)
|
36.68
(28.28-50.68)
|
0.746
|
PTH ()
|
4.60
(1.36-11.07)
|
4.33
(1.36-7.17)
|
4.85
(3.06-11.07)
|
0.398
|
BMI: Body Mass Index; CHOP: cyclophosphamide, doxorubicin, vincristine, and prednisone. VIPD: etoposide, ifosfamide, cisplatin, dexamethasone; DDGP: dexamethasone, cisplatin, gemcitabline, pegaspargase; VDLP; etoposide, dexamethasone, L-asparaginase, cisplatin; R2-EPOCH: rituximab, lenalidomide etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin; BFM-90: see reference 17; PTH: parathyroid hormone, pmol/L. Calcium: mmol/L. 25-hydroxy vitamin D: nmol/L. |
With regard the BMD T score, it didn’t display statistic significantly difference at any location from baseline data between ZA and control group (Table 2). In total, the average T score of L1-L4 were -0.44, -0.58, -0.34, -0.19, respectively; for the left femoral neck, left Ward’s triangle, left trochanter major, left hip were -0.47, -1.08, -0.87, -0.44, separately. It was 52.38% that diagnosed osteopenia at baseline (n=84), among them there were 72.73% of patients older than 50 years old, of which 50% were women (supplementary table 1).
Table 2
T score at baseline in randomized patients.
Location
|
Total patients
N=29
(mean±SD)
|
Zoledronic Acid group
N=16
(mean±SD)
|
Control group
N=13
(mean±SD)
|
Zoledronic Acid group vs control group
p value
|
L1
|
-0.44±1.36
|
-0.56±1.54
|
-0.31±1.15
|
0.634
|
L2
|
-0.58±1.39
|
-0.90±1.15
|
-0.19±1.60
|
0.176
|
L3
|
-0.34±1.59
|
-0.55±1.30
|
-0.09±1.92
|
0.452
|
L4
|
-0.19±1.51
|
-0.48±1.19
|
0.18±1.81
|
0.250
|
Left femoral neck
|
-0.47±1.08
|
-0.70±1.19
|
-0.18±0.89
|
0.206
|
Left Ward’s triangle
|
-1.08±1.24
|
-1.37±1.16
|
-0.73±1.28
|
0.172
|
Left trochanter major
|
-0.87±1.17
|
-1.24±0.95
|
-0.41±1.29
|
0.056
|
Left hip
|
-0.44±0.99
|
-0.68±0.98
|
-0.14±0.96
|
0.147
|
SD: standard deviation. L1/2/3/4: lumbar spine 1/2/3/4. |
The primary endpoint in this trial was the alternation of T score from enrollment to the 12th month at any location. As shown in Table 3, we found a marked improvement of T score in ZA group compared to the control in a year after enrollment at lumbar vertebra (L1-L4). However, in the left femoral neck, Ward’s triangle, trochanter major and hip, the data showed no statistic difference between the two groups. Additionally, comparing ZA group with control group, there was no significant difference at the 12 months of the T score (supplementary table 2). Interestingly, we could detect a pronounced deterioration of T score in the control group patients at the 12 months in contrast to that at the enrollment, especially in L1, L2, L4 and left femoral neck, Ward’s triangle (supplementary table 3). On the contrary, it presented that a profound increase of T score measured at the 12 months in ZA group after twice infusion of ZA at the L2, L3, L4 locations (supplementary table 4). Moreover, no patient appeared new bone fractures or other bone related symptoms during the treatment and observation periods in both groups.
Table 3
T score alternations from baseline at 12th month.
Location
|
Total patients
N=29
(mean±SD)
|
Zoledronic Acid group
N=16
(mean±SD)
|
Control group N=13
(mean±SD)
|
Zoledronic Acid group vs control group
p value
|
L1
|
-0.07±0.57
|
0.14±0.61
|
-0.33±0.40
|
0.009
|
L2
|
-0.01±0.56
|
0.28±0.41
|
-0.35±0.54
|
0.003
|
L3
|
0.02±0.58
|
0.23±0.42
|
-0.23±0.66
|
0.020
|
L4
|
0±0.57
|
0.28±0.44
|
-0.35±0.54
|
0.020
|
Left femoral neck
|
-0.16±0.23
|
-0.13±0.22
|
-0.19±0.24
|
0.619
|
Left Ward’s triangle
|
-0.18±0.32
|
-0.07±0.21
|
-0.32±0.38
|
0.092
|
Left trochanter major
|
0.05±0.38
|
0.11±0.32
|
-0.02±0.46
|
0.170
|
Left hip
|
-0.44±0.33
|
-0.02±0.42
|
0.08±0.16
|
0.144
|
SD: standard deviation. |
BMD serum turnover biomarkers could reflect the bone metabolism when glucocorticoid was administered. We dynamically monitored levels of serum β-CTX and BALP at 0, 3, 6, 9, 12 months in 29 patients. As shown in the Table 4, we observed the β-CTX and BALP levels at enrollment were 0.76±1.58 ng/mL and 13.70±3.84 ug/L for all patients. In addition, comparing the ZA group to the control group, either β-CTX or BALP level was similar at baseline. Of note, it started with the 3th month that there appeared significantly reduction of serum β-CTX and BALP levels in the ZA group in comparison to control group, which continued to the 12th month, p value at all time points (3, 6, 9, 12 months) were all < 0.05.
Table 4
Serum bone turnover biochemical markers of patients at different times.
|
Total patients
N=29
(mean±SD)
|
Zoledronic Acid group
N=16
(mean±SD)
|
Control group N=13
(mean±SD)
|
Zoledronic Acid group vs. control group
p value
|
β-CTX
|
baseline
|
0.76±1.58
|
0.49±0.31
|
0.47±0.21
|
0.879
|
3 months
|
0.36±0.27
|
0.19±0.10
|
0.56±0.28
|
<0.001
|
6 months
|
0.37±0.29
|
0.21±0.12
|
0.56±0.32
|
0.002
|
9 months
|
0.33±0.24
|
0.20±0.14
|
0.33±0.24
|
<0.001
|
12 months
|
0.31±0.23
|
0.20±0.11
|
0.45±0.25
|
0.004
|
BALP
|
baseline
|
13.70±3.84
|
13.12±3.62
|
14.42±4.12
|
0.374
|
3 months
|
11.78±5.26
|
8.95±1.62
|
15.26±6.11
|
0.003
|
6 months
|
12.91±6.19
|
9.79±2.37
|
16.76±7.30
|
0.005
|
9 months
|
13.47±6.59
|
10.14±3.16
|
17.56±7.48
|
0.004
|
12 months
|
13.13±5.66
|
10.43±3.48
|
16.46±6.16
|
0.006
|
β-CTX: type I collagen carboxy-terminal peptide, ng/mL. BALP: bone alkaline phosphatase, µg/L. |
ZA related severe adverse events were not detected during the treatment and follow-up periods. A minority of patients experienced Flu-like symptoms, such as transient fever, rigor, which relieved quickly after non-steroidal anti-inflammatory drug (NSAIDS) administered. Notably, an extra-nodal NK/T-cell lymphoma, nasal type (ENKTL) patient in ZA group appeared with osteonecrosis of mandible in the follow-up, considering the ZA side effect, we performed tissue biopsy for the lesions that diagnosed as ENKTL involved, excluded the ZA related osteonecrosis.