Clin Rheumatol DOI 10.1007/s10067-015-2972-4
ORIGINAL ARTICLE
The age–risk relationship of hematologic malignancies in patients with rheumatoid arthritis: a nationwide retrospective cohort study Yu-Chih Lin 1 & Hui-Wen Chou 2 & Wen-Chan Tsai 2,3 & Jeng-Hsien Yen 3,4 & Shun-Jen Chang 5 & Yi-Ching Lin 2,6,7,8
Received: 27 October 2014 / Revised: 16 April 2015 / Accepted: 7 May 2015 # International League of Associations for Rheumatology (ILAR) 2015
Abstract The risk of hematologic malignancies in rheumatoid arthritis (RA) patients has been an important clinical concern. The information of age effect on the interval from the diagnosis of RA to that of hematologic malignancies is limited. This study aimed to define the age–risk relationship between hematologic malignancies and RA. A retrospective cohort study was conducted nationwide with 17,472 patients and 87,360 controls from the Taiwan National Health Insurance Database covering 1997–2008. The subsequent development of hematologic malignancy was observed. The age-adjusted standardized incidence ratios (SIRs), incidence per 1000 person-years, follow-up duration for the diagnosis of hematologic malignancies, and cumulative hazard rates of hematologic malignancies between RA and controls were analyzed. Significantly higher incidences of both lymphoid and myeloid malignancies were found in male RA patients compared with RA-free patients (SIR 3.36, 95 % CI=2.03–5.57, and SIR:
3.69, 95 % CI=2.46–5.53). A significantly increased overall incidence risk was found in lymphoid malignancies (SIR 3.00, 95 % CI=2.22–4.05) but not significantly increased in myeloid malignancies (SIR 1.54, 95 % CI=0.95–2.50) in female RA. The follow-up duration for hematologic malignancies was significantly shorter in RA patients than in RA-free patients in both males and females (70.70 vs. 103.80 months and 67.73 vs. 100.93 months, respectively). Additionally, higher cumulative hazard rates in both lymphoid and myeloid malignancies were found in male RA patients (p20, ≤30 >30, ≤40 >40, ≤50 >50, ≤60 >60, ≤70 >70 Total
805 (5.8) 1827 (13.2) 3403 (24.7) 3469 (25.2) 2863 (20.8) 1418 (10.3) 13,785 (100)
4025 (5.8) 9135 (13.2) 17,015 (24.7) 17,345 (25.2) 14,315 (20.8) 7090 (10.3) 68,925 (100)
162 (4.4) 317 (8.6) 664 (18.0) 811 (22.0) 1054 (28.6) 679 (18.4) 3687 (100)
810 (4.4) 1585 (8.6) 3320 (18.0) 4055 (22.0) 5270 (28.6) 3395 (18.4) 18,435 (100)
SD standard deviation
free controls was conducted using the log-rank test. The statistical analysis for the national database was performed using the PERL (v5.8) program, and the cumulative hazard rates and log-rank test were estimated by SAS (v9.2) program.
14.07 years, respectively. The age distributions between the RA patients and the control group were similar (Table 1).
The age–risk relationship for lymphoid malignancies in RA
Results Demographic data of the study subjects A total of 13,785 female and 3687 male RA patients without lymphoid and myeloid malignancies during enrollment and the first 2 years after RA diagnosis between July 1997 and December 2001 were identified as the baseline population in this study. The age distribution of the baseline population is shown in Table 1. The mean ages and SDs in the female and male RA patients were 52.89 ± 13.66 years and 57.44 ± Table 2
SIRs and 95 % CIs for lymphoid malignancies, according to age and duration of follow-up in RA patients in Taiwan
Age (mean±SD)
Age at enrollment
During the follow-up period, a total of 81 lymphoid malignancies (60 female and 21 male) were identified after the diagnosis of RA. Significantly increased risks of lymphoid malignancies were observed in both female (SIR 3.00, 95 % CI=2.22– 4.05) and male (SIR 3.36, 95 % CI=2.03–5.57) RA patients (Table 2). In females, the peak risk ratio was found in patients aged 41–50 years (rate ratio 4.11, 95 % CI=2.17–7.79), and the risk ratios were not significantly different in the patients older than 40 years or younger than 70 years during the follow-up period. In males, the peak risk ratio was found in patients who were 61–70 years old (rate ratio 5.33, 95 % CI=
Female
Male
Patients (n=13,785)
Controls (n=68,925)
Patients (n=3687)
Controls (n=18,435)
52.89±13.66
52.49±14.13
57.44±14.07
56.99±14.27
No.(Inc)a
No. (Inc)a
No. (Inc)a
No. (Inc)a
Mean followup months
Mean followup months
Rate ratio
95 % CI
Mean followup months
Mean followup months
Rate ratio
95 % CI
>20, ≤30
1 (0.16)
79.00
5 (0.12)
98.40
1.36
0.16–11.65
0 (0.00)
0.00
0 (0.00)
0.00
–
–
>30, ≤40
4 (0.28)
64.00
11 (0.11)
106.91
2.47
0.79–7.77
0 (0.00)
0.00
4 (0.24)
83.75
0.00
–
>40, ≤50
15 (0.56)
86.53
25 (0.14)
109.52
4.11
2.17–7.79
3 (0.65)
72.67
8 (0.23)
107.38
2.89
0.77–10.91
>50, ≤60
19 (0.72)
71.47
46 (0.25)
102.37
2.91
1.71–4.97
5 (0.87)
76.80
10 (0.23)
108.90
3.75
1.28–10.92
>60, ≤70
16 (0.80)
67.25
35 (0.23)
98.14
3.50
1.94–6.33
9 (1.42)
77.22
15 (0.27)
110.47
5.33
2.33–12.18
>70
5 (0.67)
50.40
27 (0.39)
107.11
1.73
0.67–4.49
4 (1.31)
49.25
18 (0.54)
105.11
2.41
0.81–7.11
Total
60 (0.59)
71.98
149 (0.20)
103.64
2.91
2.16–3.93
21 (0.91)
71.14
55 (0.29)
106.04
3.21
1.94–5.31
3.00
2.22–4.05
3.36
2.03–5.57
Age-adjusted standardized incidence ratio
SIRs standardized incidence ratios, 95% CI 95 % confidence interval a
Numbers of new diagnosed lymphoid malignancies (incidence per 1000 person-years)
2.29–5.15
2.46–5.53 3.69
3.44 104.39 36 (0.19) 61.54
a
Numbers of new diagnosed lymphoid malignancies (incidence per 1000 person-years)
0.95–2.50 1.54
SIRs standardized incidence ratios, 95% CI 95 % confidence interval
63.40
96 (0.13)
104.07
1.51
0.93–2.44
13 (0.56) 20 (0.20) Total
Age-adjusted standardized incidence ratio
2.06–7.58
0.80–5.40 2.08 103.20 10 (0.30) 49.50
13 (0.19)
107.15
2.86
0.93–8.76
2 (0) 4 (0.54) >70
43.50
2.25–12.32 5.27
3.95 108.47
101.33 6 (0.14)
15 (0.27) 55.25 0.45–2.96
0.95–4.58 2.08
1.15 102.76
106.74 27 (0.14)
33 (0.22) 74.60
67.25
4 (0.63) 5 (0.25) >60, ≤70
4 (0.69) 8 (0.30) >50, ≤60
51.75
–
1.23–10.22 3.55 98.20 5 (0.14) 53.00
17 (0.09)
107.18
1.20
0.35–4.11
2 (0.43) 3 (0.11) >40, ≤50
102.00
7.33
– 0.00
0.00 0 (0.00)
0 (0.00) 0.00 –
– –
0.00 83.83
0.00 0 (0.00)
6 (0.06) 0.00
0.00
0 (0.00) 0 (0.00) >30, ≤40
95 % CI Rate ratio Mean follow-up months Mean follow-up months
1 (0.92) 0 (0.00) >20, ≤30
81.00
Mean follow-up months No. (Inc)a No. (Inc)a No. (Inc)a Age at enrollment
Mean follow-up months 52.50±14.14 52.89±13.66
No. (Inc)a
Patients (n=3687) Controls (n=68,925) Patients (n=13,785)
57.44±14.07
Male Female
A comparison of the cumulative hazard rates of the lymphoid and myeloid malignancies between RA patients and the controls is shown in Figs. 1 and 2. For lymphoid malignancies, the 120-month cumulative hazard rate was 6.10/1000 patients in female RA patients vs. 1.74/1000 patients in female patients without RA, and it was 11.9/1000 vs. 2.81/1000 in male patients. RA patients had higher cumulative hazard rates of lymphoid malignancies in both female (χ2 =95.42, p
ORIGINAL ARTICLE
The age–risk relationship of hematologic malignancies in patients with rheumatoid arthritis: a nationwide retrospective cohort study Yu-Chih Lin 1 & Hui-Wen Chou 2 & Wen-Chan Tsai 2,3 & Jeng-Hsien Yen 3,4 & Shun-Jen Chang 5 & Yi-Ching Lin 2,6,7,8
Received: 27 October 2014 / Revised: 16 April 2015 / Accepted: 7 May 2015 # International League of Associations for Rheumatology (ILAR) 2015
Abstract The risk of hematologic malignancies in rheumatoid arthritis (RA) patients has been an important clinical concern. The information of age effect on the interval from the diagnosis of RA to that of hematologic malignancies is limited. This study aimed to define the age–risk relationship between hematologic malignancies and RA. A retrospective cohort study was conducted nationwide with 17,472 patients and 87,360 controls from the Taiwan National Health Insurance Database covering 1997–2008. The subsequent development of hematologic malignancy was observed. The age-adjusted standardized incidence ratios (SIRs), incidence per 1000 person-years, follow-up duration for the diagnosis of hematologic malignancies, and cumulative hazard rates of hematologic malignancies between RA and controls were analyzed. Significantly higher incidences of both lymphoid and myeloid malignancies were found in male RA patients compared with RA-free patients (SIR 3.36, 95 % CI=2.03–5.57, and SIR:
3.69, 95 % CI=2.46–5.53). A significantly increased overall incidence risk was found in lymphoid malignancies (SIR 3.00, 95 % CI=2.22–4.05) but not significantly increased in myeloid malignancies (SIR 1.54, 95 % CI=0.95–2.50) in female RA. The follow-up duration for hematologic malignancies was significantly shorter in RA patients than in RA-free patients in both males and females (70.70 vs. 103.80 months and 67.73 vs. 100.93 months, respectively). Additionally, higher cumulative hazard rates in both lymphoid and myeloid malignancies were found in male RA patients (p20, ≤30 >30, ≤40 >40, ≤50 >50, ≤60 >60, ≤70 >70 Total
805 (5.8) 1827 (13.2) 3403 (24.7) 3469 (25.2) 2863 (20.8) 1418 (10.3) 13,785 (100)
4025 (5.8) 9135 (13.2) 17,015 (24.7) 17,345 (25.2) 14,315 (20.8) 7090 (10.3) 68,925 (100)
162 (4.4) 317 (8.6) 664 (18.0) 811 (22.0) 1054 (28.6) 679 (18.4) 3687 (100)
810 (4.4) 1585 (8.6) 3320 (18.0) 4055 (22.0) 5270 (28.6) 3395 (18.4) 18,435 (100)
SD standard deviation
free controls was conducted using the log-rank test. The statistical analysis for the national database was performed using the PERL (v5.8) program, and the cumulative hazard rates and log-rank test were estimated by SAS (v9.2) program.
14.07 years, respectively. The age distributions between the RA patients and the control group were similar (Table 1).
The age–risk relationship for lymphoid malignancies in RA
Results Demographic data of the study subjects A total of 13,785 female and 3687 male RA patients without lymphoid and myeloid malignancies during enrollment and the first 2 years after RA diagnosis between July 1997 and December 2001 were identified as the baseline population in this study. The age distribution of the baseline population is shown in Table 1. The mean ages and SDs in the female and male RA patients were 52.89 ± 13.66 years and 57.44 ± Table 2
SIRs and 95 % CIs for lymphoid malignancies, according to age and duration of follow-up in RA patients in Taiwan
Age (mean±SD)
Age at enrollment
During the follow-up period, a total of 81 lymphoid malignancies (60 female and 21 male) were identified after the diagnosis of RA. Significantly increased risks of lymphoid malignancies were observed in both female (SIR 3.00, 95 % CI=2.22– 4.05) and male (SIR 3.36, 95 % CI=2.03–5.57) RA patients (Table 2). In females, the peak risk ratio was found in patients aged 41–50 years (rate ratio 4.11, 95 % CI=2.17–7.79), and the risk ratios were not significantly different in the patients older than 40 years or younger than 70 years during the follow-up period. In males, the peak risk ratio was found in patients who were 61–70 years old (rate ratio 5.33, 95 % CI=
Female
Male
Patients (n=13,785)
Controls (n=68,925)
Patients (n=3687)
Controls (n=18,435)
52.89±13.66
52.49±14.13
57.44±14.07
56.99±14.27
No.(Inc)a
No. (Inc)a
No. (Inc)a
No. (Inc)a
Mean followup months
Mean followup months
Rate ratio
95 % CI
Mean followup months
Mean followup months
Rate ratio
95 % CI
>20, ≤30
1 (0.16)
79.00
5 (0.12)
98.40
1.36
0.16–11.65
0 (0.00)
0.00
0 (0.00)
0.00
–
–
>30, ≤40
4 (0.28)
64.00
11 (0.11)
106.91
2.47
0.79–7.77
0 (0.00)
0.00
4 (0.24)
83.75
0.00
–
>40, ≤50
15 (0.56)
86.53
25 (0.14)
109.52
4.11
2.17–7.79
3 (0.65)
72.67
8 (0.23)
107.38
2.89
0.77–10.91
>50, ≤60
19 (0.72)
71.47
46 (0.25)
102.37
2.91
1.71–4.97
5 (0.87)
76.80
10 (0.23)
108.90
3.75
1.28–10.92
>60, ≤70
16 (0.80)
67.25
35 (0.23)
98.14
3.50
1.94–6.33
9 (1.42)
77.22
15 (0.27)
110.47
5.33
2.33–12.18
>70
5 (0.67)
50.40
27 (0.39)
107.11
1.73
0.67–4.49
4 (1.31)
49.25
18 (0.54)
105.11
2.41
0.81–7.11
Total
60 (0.59)
71.98
149 (0.20)
103.64
2.91
2.16–3.93
21 (0.91)
71.14
55 (0.29)
106.04
3.21
1.94–5.31
3.00
2.22–4.05
3.36
2.03–5.57
Age-adjusted standardized incidence ratio
SIRs standardized incidence ratios, 95% CI 95 % confidence interval a
Numbers of new diagnosed lymphoid malignancies (incidence per 1000 person-years)
2.29–5.15
2.46–5.53 3.69
3.44 104.39 36 (0.19) 61.54
a
Numbers of new diagnosed lymphoid malignancies (incidence per 1000 person-years)
0.95–2.50 1.54
SIRs standardized incidence ratios, 95% CI 95 % confidence interval
63.40
96 (0.13)
104.07
1.51
0.93–2.44
13 (0.56) 20 (0.20) Total
Age-adjusted standardized incidence ratio
2.06–7.58
0.80–5.40 2.08 103.20 10 (0.30) 49.50
13 (0.19)
107.15
2.86
0.93–8.76
2 (0) 4 (0.54) >70
43.50
2.25–12.32 5.27
3.95 108.47
101.33 6 (0.14)
15 (0.27) 55.25 0.45–2.96
0.95–4.58 2.08
1.15 102.76
106.74 27 (0.14)
33 (0.22) 74.60
67.25
4 (0.63) 5 (0.25) >60, ≤70
4 (0.69) 8 (0.30) >50, ≤60
51.75
–
1.23–10.22 3.55 98.20 5 (0.14) 53.00
17 (0.09)
107.18
1.20
0.35–4.11
2 (0.43) 3 (0.11) >40, ≤50
102.00
7.33
– 0.00
0.00 0 (0.00)
0 (0.00) 0.00 –
– –
0.00 83.83
0.00 0 (0.00)
6 (0.06) 0.00
0.00
0 (0.00) 0 (0.00) >30, ≤40
95 % CI Rate ratio Mean follow-up months Mean follow-up months
1 (0.92) 0 (0.00) >20, ≤30
81.00
Mean follow-up months No. (Inc)a No. (Inc)a No. (Inc)a Age at enrollment
Mean follow-up months 52.50±14.14 52.89±13.66
No. (Inc)a
Patients (n=3687) Controls (n=68,925) Patients (n=13,785)
57.44±14.07
Male Female
A comparison of the cumulative hazard rates of the lymphoid and myeloid malignancies between RA patients and the controls is shown in Figs. 1 and 2. For lymphoid malignancies, the 120-month cumulative hazard rate was 6.10/1000 patients in female RA patients vs. 1.74/1000 patients in female patients without RA, and it was 11.9/1000 vs. 2.81/1000 in male patients. RA patients had higher cumulative hazard rates of lymphoid malignancies in both female (χ2 =95.42, p
