Kurnik & Artul - Amended report

Upper decile vs Others

Table 1

Characteristic Overall, N = 1461 No aGVHD (0,1), N = 801 aGVHD (2,3,4), N = 661 p-value2
patient_height 168 (161, 175) 168 (161, 173) 170 (161, 177) 0.092
    Unknown 3 0 3
patient_abo 0.12
    A 52 (36%) 33 (41%) 19 (29%)
    AB 9 (6.2%) 7 (8.8%) 2 (3.0%)
    B 23 (16%) 12 (15%) 11 (17%)
    O 62 (42%) 28 (35%) 34 (52%)
donor_sex 0.3
    F 45 (31%) 22 (28%) 23 (36%)
    M 98 (69%) 57 (72%) 41 (64%)
    Unknown 3 1 2
death_before_100_days 22 (15%) 14 (18%) 8 (12%) 0.4
    Unknown 2 1 1
donor_age 35 (22, 48) 34 (23, 51) 35 (22, 44) 0.5
    Unknown 8 2 6
death 60 (42%) 30 (39%) 30 (45%) 0.5
    Unknown 4 4 0
sex_disparity 0.5
    different sex 84 (58%) 48 (60%) 36 (55%)
    same sex 62 (42%) 32 (40%) 30 (45%)
relapse 28 (20%) 16 (22%) 12 (18%) 0.6
    Unknown 9 8 1
related_unrelated 0.7
    Related 69 (47%) 39 (49%) 30 (45%)
    Unrelated 76 (52%) 41 (51%) 35 (53%)
    UNRELATED 1 (0.7%) 0 (0%) 1 (1.5%)
donor_cmv_status 88 (68%) 48 (67%) 40 (70%) 0.7
    Unknown 17 8 9
hla_disparity 0.7
    MRD 69 (47%) 39 (49%) 30 (45%)
    MUD 77 (53%) 41 (51%) 36 (55%)
conditioning_regimen 0.7
    BuCy 14 (9.7%) 8 (10%) 6 (9.2%)
    BuFlu 54 (37%) 27 (34%) 27 (42%)
    BuFLuThio 12 (8.3%) 7 (8.8%) 5 (7.7%)
    FLuThioMel 1 (0.7%) 0 (0%) 1 (1.5%)
    others 64 (44%) 38 (48%) 26 (40%)
    Unknown 1 0 1
donor_abo 0.8
    A 47 (36%) 28 (38%) 19 (33%)
    AB 9 (6.9%) 4 (5.4%) 5 (8.8%)
    B 16 (12%) 10 (14%) 6 (11%)
    O 59 (45%) 32 (43%) 27 (47%)
    Unknown 15 6 9
patient_cmv_status 122 (85%) 66 (85%) 56 (86%) 0.8
    Unknown 3 2 1
patient_age 54 (40, 63) 55 (38, 66) 54 (44, 62) 0.8
atg 86 (62%) 48 (62%) 38 (63%) 0.8
    Unknown 8 2 6
patient_weight 74 (63, 86) 73 (63, 86) 74 (64, 84) 0.8
    Unknown 3 0 3
indication_for_allohsct 0.9
    acute_leukemia 101 (69%) 56 (70%) 45 (68%)
    chronic_leukemia 5 (3.4%) 2 (2.5%) 3 (4.5%)
    others 40 (27%) 22 (28%) 18 (27%)
patient_sex >0.9
    F 58 (40%) 32 (40%) 26 (39%)
    M 88 (60%) 48 (60%) 40 (61%)
graft_source >0.9
    BM 2 (1.4%) 1 (1.3%) 1 (1.6%)
    PBSC 137 (99%) 76 (99%) 61 (98%)
    Unknown 7 3 4
1 n (%); Median (IQR)
2 Pearson’s Chi-squared test; Wilcoxon rank sum test; Fisher’s exact test

Figures

Figure 1 - Daily WA of CsA conc. + Normal approximation confidence interval

Survival data

Dataframe ka_surv contains time to death and time to agvhd.

The end of follow up for all patients is March 1 - 2022.

Upper vs. others

Each time frame has the following analyses:

  • Fisher test - tests whether there is an association between the CSA quantile during period \(t_1-t_2\) and agvhd outcome (yes or no).

  • Odds ratio - as requested, computes the OR of D10/Others. The contingency table corresponding to the OR is printed as well.

  • Crosstable - there used to be a nice function that outputs crosstables in SPSS format, but I could not find it, hence the ugly tables.

Day 0

Fisher test


    Fisher's Exact Test for Count Data

data:  table(dat)
p-value = 0.2878
alternative hypothesis: true odds ratio is not equal to 1
95 percent confidence interval:
  0.5167773 30.9338737
sample estimates:
odds ratio 
    2.9535

Odds ratio

No aGVHD (0,1) CSA in Rest : 41 Upper : 7
aGVHD (2,3,4) Rest : 35 Upper : 2
odds ratio     lwr.ci     upr.ci 
0.33469388 0.06525467 1.71665871

Crosstable

ID in Upper: decile 136 147 18 35 43 60 92 130 142

Simple \(\chi^2\) test

Upper vs others

Characteristic Overall, N = 851 Rest, N = 761 Upper, N = 91 p-value2
conc_wa 187 (124, 227) 174 (123, 213) 318 (310, 338) <0.001
agvhd_level 0.3
    aGVHD (2,3,4) 37 (44%) 35 (46%) 2 (22%)
    No aGVHD (0,1) 48 (56%) 41 (54%) 7 (78%)
1 Median (IQR); n (%)
2 Wilcoxon rank sum test; Fisher’s exact test

Day 0-7

Fisher test


    Fisher's Exact Test for Count Data

data:  table(dat)
p-value = 0.1717
alternative hypothesis: true odds ratio is not equal to 1
95 percent confidence interval:
  0.69265 11.29676
sample estimates:
odds ratio 
    2.4926

Odds ratio

No aGVHD (0,1) CSA in Rest : 68 Upper : 11
aGVHD (2,3,4) Rest : 62 Upper : 4
odds ratio     lwr.ci     upr.ci 
 0.3988270  0.1207258  1.3175558

Crosstable

ID in Upper: decile 6 12 25 48 60 74 86 92 109 130 136 139 142 146 147

Simple \(\chi^2\) test

Upper vs others

Characteristic Overall, N = 1451 Rest, N = 1301 Upper, N = 151 p-value2
conc_wa 210 (159, 255) 198 (152, 245) 322 (316, 340) <0.001
agvhd_level 0.2
    aGVHD (2,3,4) 66 (46%) 62 (48%) 4 (27%)
    No aGVHD (0,1) 79 (54%) 68 (52%) 11 (73%)
1 Median (IQR); n (%)
2 Wilcoxon rank sum test; Fisher’s exact test

Day 0-14

Fisher test


    Fisher's Exact Test for Count Data

data:  table(dat)
p-value = 0.1729
alternative hypothesis: true odds ratio is not equal to 1
95 percent confidence interval:
  0.6829939 11.1305739
sample estimates:
odds ratio 
  2.456845

Odds ratio

No aGVHD (0,1) CSA in Rest : 69 Upper : 11
aGVHD (2,3,4) Rest : 62 Upper : 4
odds ratio     lwr.ci     upr.ci 
 0.4046921  0.1225431  1.3364737

Crosstable

ID in Upper: decile 6 25 43 48 60 64 86 92 93 109 130 139 142 146 147

Simple \(\chi^2\) test

Upper vs others

Characteristic Overall, N = 1461 Rest, N = 1311 Upper, N = 151 p-value2
conc_wa 229 (182, 273) 222 (177, 257) 319 (317, 326) <0.001
agvhd_level 0.2
    aGVHD (2,3,4) 66 (45%) 62 (47%) 4 (27%)
    No aGVHD (0,1) 80 (55%) 69 (53%) 11 (73%)
1 Median (IQR); n (%)
2 Wilcoxon rank sum test; Fisher’s exact test

Day 0-30

Fisher test


    Fisher's Exact Test for Count Data

data:  table(dat)
p-value = 0.4165
alternative hypothesis: true odds ratio is not equal to 1
95 percent confidence interval:
 0.5069219 6.8429476
sample estimates:
odds ratio 
  1.736444

Odds ratio

No aGVHD (0,1) CSA in Rest : 70 Upper : 10
aGVHD (2,3,4) Rest : 61 Upper : 5
odds ratio     lwr.ci     upr.ci 
 0.5737705  0.1858905  1.7710025

Crosstable

ID in Upper: decile 6 43 47 48 64 86 92 93 109 115 120 139 142 146 147

Upper vs others

Characteristic Overall, N = 1461 Rest, N = 1311 Upper, N = 151 p-value2
conc_wa 261 (224, 292) 250 (222, 280) 327 (321, 332) <0.001
agvhd_level 0.4
    aGVHD (2,3,4) 66 (45%) 61 (47%) 5 (33%)
    No aGVHD (0,1) 80 (55%) 70 (53%) 10 (67%)
1 Median (IQR); n (%)
2 Wilcoxon rank sum test; Fisher’s exact test

Concentration by agvhd severity

Median + IQR + Wilcoxon-Mann-Whitney test

days 0-3
Characteristic aGvHD, N = 1181 no aGvHD, N = 1361 p-value2
time 0.8
    0 33 (28%) 39 (29%)
    1 31 (26%) 41 (30%)
    2 22 (19%) 19 (14%)
    3 32 (27%) 37 (27%)
conc 191 (135, 242) 207 (149, 278) 0.071
1 n (%); Median (IQR)
2 Pearson’s Chi-squared test; Wilcoxon rank sum test
days 0-7
Characteristic aGvHD, N = 2291 no aGvHD, N = 2791 p-value2
time >0.9
    0 33 (14%) 39 (14%)
    1 31 (14%) 41 (15%)
    2 22 (9.6%) 19 (6.8%)
    3 32 (14%) 37 (13%)
    4 17 (7.4%) 25 (9.0%)
    5 29 (13%) 37 (13%)
    6 34 (15%) 43 (15%)
    7 31 (14%) 38 (14%)
conc 223 (155, 288) 238 (172, 309) 0.10
1 n (%); Median (IQR)
2 Pearson’s Chi-squared test; Wilcoxon rank sum test
days 0-14
Characteristic aGvHD, N = 4271 no aGvHD, N = 5161 p-value2
time 7.0 (3.0, 11.0) 7.0 (3.0, 11.0) >0.9
conc 253 (191, 321) 268 (194, 332) 0.081
1 Median (IQR)
2 Wilcoxon rank sum test
days 0-21
Characteristic aGvHD, N = 5911 no aGvHD, N = 7251 p-value2
time 10 (5, 15) 10 (5, 15) 0.6
conc 276 (203, 345) 283 (209, 350) 0.3
1 Median (IQR)
2 Wilcoxon rank sum test
days 0-28
Characteristic aGvHD, N = 6921 no aGvHD, N = 8631 p-value2
time 12 (6, 18) 12 (6, 19) 0.3
conc 280 (206, 350) 283 (210, 352) 0.7
1 Median (IQR)
2 Wilcoxon rank sum test

Mead + SD + t.test

days 0-3
Characteristic aGvHD, N = 661 no aGvHD, N = 791 Difference2 95% CI2,3 p-value2
csa_wa 201 (73) 212 (80) -12 -37, 14 0.4
1 Mean (SD)
2 Welch Two Sample t-test
3 CI = Confidence Interval
days 0-7
Characteristic aGvHD, N = 661 no aGvHD, N = 801 Difference2 95% CI2,3 p-value2
csa_wa 229 (63) 246 (72) -16 -39, 5.7 0.14
1 Mean (SD)
2 Welch Two Sample t-test
3 CI = Confidence Interval
days 0-14
Characteristic aGvHD, N = 661 no aGvHD, N = 801 Difference2 95% CI2,3 p-value2
csa_wa 258 (53) 269 (58) -11 -29, 7.0 0.2
1 Mean (SD)
2 Welch Two Sample t-test
3 CI = Confidence Interval
days 0-21
Characteristic aGvHD, N = 661 no aGvHD, N = 801 Difference2 95% CI2,3 p-value2
csa_wa 281 (52) 287 (50) -5.2 -22, 12 0.5
1 Mean (SD)
2 Welch Two Sample t-test
3 CI = Confidence Interval
days 0-28
Characteristic aGvHD, N = 661 no aGvHD, N = 801 Difference2 95% CI2,3 p-value2
csa_wa 287 (48) 289 (46) -2.1 -18, 13 0.8
1 Mean (SD)
2 Welch Two Sample t-test
3 CI = Confidence Interval