Patient characteristics
A total of 180 patients were enrolled in the study, among whom five were excluded due to noncompliance with the participation criteria (four patients had grade 0 CTCAE and STTA and one had neuroendocrine carcinoma) and three patients in the no intervention group changed to zinc therapy. Thus, the final total of patients eligible for evaluation in the present study included 53 in the no intervention group, 60 in the zinc acetate hydrate group, and 59 in the polaprezinc group. The characteristics of the 172 patients who participated in the study are summarized in Table 1. There were more females in the no intervention group (male/female: 25/28) and more males in the zinc acetate hydrate and polaprezinc groups (male/female: 42/18, 26/23, respectively). The median age of the polaprezinc group was 67 years, which was slightly younger than that of the no intervention and zinc acetate hydrate groups, which was 70 years. The mean serum zinc levels at baseline were 67.3 µg/dL [95% confidence interval (CI), 63.0–71.6 µg/dL] in the no intervention group, 67.5 µg/dL (95% CI, 63.9–71.1 µg/dL) in the zinc acetate hydrate group, and 66.6 µg/dL (95% CI, 63.0–70.2 µg/dL) in the polaprezinc group. The mean serum zinc levels at baseline were similar in each group and below the normal limit of 80 µg/dL.
Table 1
n (%)
|
|
No intervention
(n = 53)
|
Zinc acetate hydrate
(n = 60)
|
Polaprezinc
(n = 59)
|
Sex
|
Male
Female
|
25 (47)
28 (53)
|
42 (70)
18 (30)
|
36 (61)
23 (39)
|
Age
|
Median (Range)
|
70 (27–86)
|
70 (42–81)
|
67 (40–79)
|
Body mass index
|
Median (Range)
|
21.0 (14.6–33.5)
|
21.9 (16.2–30.7)
|
21.9 (15.4–30.5)
|
ECOG PS
|
0
1
2
|
28 (53)
24 (45)
1 (2)
|
37 (62)
23 (38)
0 (0)
|
35 (59)
24 (41)
0 (0)
|
Professional oral care*
|
Yes
No
|
18 (34)
35 (66)
|
26 (43)
34 (57)
|
20 (34)
39 (66)
|
Mouth rinses
|
Yes
No
|
27 (51)
26 (49)
|
39 (65)
21 (35)
|
32 (54)
27 (46)
|
Duration of dysgeusia (day)
|
Median (range)
|
36 (1–1,933)
|
23 (1–1,167)
|
29 (1–3,697)
|
Cancer type
|
Esophagus
Gastric
Colorectal
Pancreatic
Biliary tract
Others
|
2 (4)
15 (28)
26 (49)
6 (11)
4 (8)
0 (0)
|
1 (2)
12 (20)
28 (47)
12 (20)
6 (10)
1 (2)
|
0 (0)
14 (24)
30 (51)
7 (12)
7 (12)
1 (2)
|
Chemotherapy drugs
|
Platinum
Fluoropyrimidines
Taxanes
|
28 (53)
38 (72)
10 (19)
|
30 (50)
48 (80)
8 (13)
|
35 (59)
45 (76)
7 (12)
|
Number of pretreatments
|
0
1
≥2
|
37 (70)
11 (21)
5 (9)
|
41 (68)
17 (28)
2 (3)
|
41 (69)
12 (20)
6 (10)
|
Serum zinc (µg/dL)
|
Median (range)
|
68 / (29–113)
|
64 / (43–109)
|
66 (25–109)
|
Hb (g/dL)
|
Median (range)
|
11.2 (8.3–16.1)
|
11.5 (8.1–16.4)
|
11.0 (8.0–16.2)
|
Alb (g/dL)
|
Median (range)
|
3.6 (2.1–4.5)
|
3.7 (2.6–4.5)
|
3.5 (2.4–4.5)
|
Ferritin (ng/mL)
|
Median (range)
|
129.5 (3.7–1,855)
|
155.1 (7–1,343)
|
97.2 (4–1,051)
|
Vit B12 (pg/mL)
|
Median (range)
|
457 (110–1,500)
|
478 (108–1,500)
|
513 (144–1,500)
|
*Oral care delivered by dental professionals |
Zinc supplementation
Zinc supplementation was successfully achieved for 12 weeks by 53 patients in the polaprezinc group and 46 patients in the zinc acetate hydrate group. The reasons for discontinuation were mostly unrelated to zinc therapy and included exacerbation of the primary disease (four patients in each group), cholangitis (two patients in the zinc acetate hydrate group), self-decision (two patients in the poraprezinc group and six patients in the zinc acetate hydrate group), or physician’s opinion (one patient in zinc acetate hydrate group). Discontinuation due to chest discomfort related to zinc supplementation was observed in one patient in the zinc acetate hydrate group. The mean number of days on medication for patients who discontinued was 57.8 days (95% CI, 39.9–75.7) in the polaprezinc group and 34.1 days (95% CI, 24.8–43.5) in the zinc acetate hydrate group.
Changes in serum zinc levels
The mean serum zinc levels at each evaluation time point are shown in Figure 1. In the zinc acetate hydrate and polaprezinc groups, there was a significant increase in serum zinc levels at 6 weeks (P < 0.001, respectively; Wilcoxon’s signed rank test), which was maintained until 12 weeks. The mean change in serum zinc levels at 12 weeks in the no intervention, zinc acetate, and polaprezinc groups were −3.8 µg/dL (95% CI, −8.7–1.1 µg/dL), 46.6 µg/dL (95% CI, 34.5–58.7µg/dL), and 14.3 µg/dL (95% CI, 9.0–19.7 µg/dL), respectively. These results clearly demonstrate that serum zinc levels increased in a dose-dependent manner, and the increase in serum zinc in the group receiving zinc acetate hydrate and polaprezinc was statistically significant compared with that of the no intervention group (P < 0.001; Steel’s multiple comparison test).
Changes in dysgeusia score
Changes in dysgeusia determined using CTCAE and STTA score are shown in Figure 2. The numbers of patients with missing questionnaires at week 12 were 2, 8, and 2 in the no intervention, zinc acetate hydrate, and polaprezinc groups, respectively. The efficiency rate of the STTA score was 33.3% (17/51), 32.7% (17/52), and 52.6% (30/57) in the no intervention, zinc acetate hydrate, and polaprezinc groups, respectively, and there was no statistically significant difference between the groups (P = 0.053; chi-square test). The odds ratios of zinc acetate hydrate and polaprezinc to no intervention were 0.971 (95% CI, 0.427–2.209; P = 0.945) and 2.222 (95% CI, 1.018–3.850; P = 0.045). There was a statistically significant improvement in STTA score in the polaprezinc group compared with the no intervention group, but no statistical difference in the zinc acetate hydrate group compared with the no intervention group. The efficiency rates using the CTCAE score were 33.3% (17/51), 34.6% (18/52), and 36.8% (21/57) in the no intervention, zinc acetate hydrate, and polaprezinc groups, respectively, and there were no statistically significant differences between the groups (P = 0.927; chi-square test). The odds ratios of zinc acetate hydrate and polaprezinc compared with no intervention were 1.059 (95% CI, 0.468–2.394; P = 0.891) and 1.167 (95% CI, 0.528–2.578; P = 0.703), respectively. Changes in dysgeusia determined using VAS are shown in Table 2. The mean changes in VAS score were 6.6 mm (95% CI, 1.0–12.1), 6.1 mm (95% CI, −1.6–13.7), and 10.8 mm (95% CI, 4.7–16.9) in the no intervention, zinc acetate hydrate, and polaprezinc groups, respectively, and there were no statistically significant differences between the zinc acetate hydrate and polaprezinc groups compared with the no intervention group (P = 0.994, P = 0.669, respectively; Steel’s multiple comparison test). Changes in taste perception measured using CiTAS are shown in Table 3. The mean changes of decline in basic taste using the CiTAS subscore were 0.16 (95% CI, −0.04–0.37), 0.36 (95% CI, 0.05–0.67), and 0.47 (95% CI, 0.26–0.68) in the no intervention, zinc acetate hydrate, and polaprezinc groups, respectively. The mean changes in discomfort determined using the CiTAS subscore were −0.03 (95% CI, −0.21–0.15), 0.12 (95% CI, −0.09–0.33), and 0.22 (95% CI, 0.04–0.40), phantogeusia and parageusia were 0.03 (95% CI, −0.16–0.22), 0.13 (95% CI, −0.21–0.47), and 0.21 (95% CI, −0.06–0.48), and general taste alterations were 0.08 (95% CI, −0.09–0.26), 0.35 (95% CI, 0.04–0.65), and 0.44 (95% CI, 0.18–0.71) in the no intervention, zinc acetate hydrate, and polaprezinc groups, respectively. There were no statistically significant differences between the groups. However, CTCAE, VAS, and CiTAS scores showed a trend toward improvement in dysgeusia in the polaprezinc group, but not the zinc acetate hydrate group, compared with the no intervention group.
Table 2
Changes in taste perception using VAS score.
(mm)
|
|
No intervention
(n = 52)
|
Zinc acetate hydrate (n = 52)
|
Polaprezinc (n = 57)
|
|
P-value†
|
|
P-value†
|
Baseline
|
Average
(95% CI)
|
35.6
(29.1–42.1)
|
44.5
(38.2–50.8)
|
0.106*
|
37.6
(30.6–44.7)
|
0.931*
|
12 weeks
|
Average
(95% CI)
|
29.3
(22.7–35.8)
|
38.4
(30.9–46.0)
|
0.170*
|
26.9
(20.1–32.7)
|
0.800**
|
ΔBase – 12 weeks
|
Average
(95% CI)
|
6.6
(1.0–12.1)
|
6.1
(−1.6–13.7)
|
0.994*
|
10.8
(4.7–16.9)
|
0.669*
|
†vs no intervention * Steel's multiple comparison test |
Table 3
Changes in taste perception using CiTAS score.
Average (95% CI)
|
No intervention
(n = 51)
|
Zinc acetate hydrate (n = 52)
|
Polaprezinc (N=57)
|
|
p-value†
|
|
p-value†
|
Decline in basic taste
|
0.16 (−0.04–0.37)
|
0.36 (0.05–0.67)
|
0.359*
|
0.47 (0.26–0.68)
|
0.102*
|
Discomfort
|
−0.03 (−0.21–0.15)
|
0.12 (−0.09–0.33)
|
0.689*
|
0.22 (0.04–0.40)
|
0.109*
|
Phantogeusia and parageusia
|
0.03 (−0.16–0.22)
|
0.13 (−0.21–0.47)
|
0.968*
|
0.21 (−0.06–0.48)
|
0.445*
|
General taste alterations
|
0.08 (−0.09–0.26)
|
0.35 (0.04–0.65)
|
0.205*
|
0.44 (0.18–0.71)
|
0.089*
|
†vs no intervention * Steel's multiple comparison test |
Factors related to improvement of dysgeusia
Multivariate analysis was conducted using multiple regression analysis and logistic regression analysis to investigate independent factors related to improvement of dysgeusia. The results of logistic regression analysis for STTA are shown in Online Resource 1. Polaprezinc was analyzed as an independent significant factor for improvement in taste disorder (P = 0.013).
Quality of life
The results of the QOL assessment by QOL-ACD are shown in Table 4. There was a significant improvement in psychological condition of QOL-ACD subdomain in the polaprezinc group compared with the no intervention group. There was a significant improvement in face scale in the QOL-ACD subdomain in the zinc acetate hydrate group compared with the no intervention group. However, in the other subdomains, there was no significant improvement in the zinc-treated group compared with the non-intervention group, and there was no significant improvement in total score. Therefore, QOL was not improved by zinc administration.
Table 4
The results of QOL assessment by QOL-ACD
Average (95%CI)
|
No intervention
(N=51)
|
Zinc acetate hydrate (N=52)
|
Polaprezinc (N=57)
|
|
p-value†
|
|
p-value†
|
Daily activity
|
−1.41 (−2.80–−0.02)
|
−0.50 (−1.77–0.77)
|
0.686*
|
0.00 (−1.37–1.37)
|
0.414*
|
Physical condition
|
0.08 (−1.11–1.27)
|
0.46 (−0.64–1.57)
|
0.938*
|
1.12 (−0.02–2.26)
|
0.454*
|
Psychological condition
|
−0.63 (−1.66–0.41)
|
0.19 (−0.45–0.84)
|
0.360*
|
1.12 (0.10–2.14)
|
0.049*
|
Social attitude
|
0.14 (−0.91–1.19)
|
0.63 (−0.27–1.54)
|
0.817*
|
0.54 (−0.42–1.51)
|
0.921*
|
Face scale
|
−0.31 (−0.60–−0.02)
|
0.19 (−0.04–0.42)
|
0.021*
|
0.04 (−0.21–0.28)
|
0.190*
|
Total
|
−2.14 (−5.50–1.22)
|
0.98 (−1.64–3.60)
|
0.237*
|
2.82 (−0.36–6.01)
|
0.083*
|
†vs no intervention * Steel's multiple comparison test |