Cytokines and panels tested
| Name . | Cytokines . | Formula . | Cutoff . |
|---|---|---|---|
| ST219 | ST2 | — | Q3* |
| Panel-215 | ST2, REG3α | log[–log(1 – p)] = –11.263 + 1.844 × log10(ST2) + 0.577 × log10(REG3α) | P = .20 |
| Panel-316 | ST2, REG3α, TNFR1 | log[–log(1 – p)] = −9.169 + 0.598 × log2(TNFR1) −0.028 × log2(REG3α) + 0.189 × log2(ST2) | Q3*,† |
| Panel-418 | IL-2Rα, TNFR1, HGF, IL-8 | −4656 + 3.43 × IL-2Rα + 0.48 × TNFR1+ 0.11 × HGF –0.45 × IL-8 | Q3*,‡ |
| Panel-617 | Elafin, IL-2Rα, TNFR1, HGF, IL-8, REG3α | 0.31 × log(elafin) + 0.13 × log(IL-2Rα) + 0.80 × log(TNFR1) + 0.20 × log(IL-8) + 0.02 × log(HGF) + 0.19 × log(REG3α) | 12.8 |
| Name . | Cytokines . | Formula . | Cutoff . |
|---|---|---|---|
| ST219 | ST2 | — | Q3* |
| Panel-215 | ST2, REG3α | log[–log(1 – p)] = –11.263 + 1.844 × log10(ST2) + 0.577 × log10(REG3α) | P = .20 |
| Panel-316 | ST2, REG3α, TNFR1 | log[–log(1 – p)] = −9.169 + 0.598 × log2(TNFR1) −0.028 × log2(REG3α) + 0.189 × log2(ST2) | Q3*,† |
| Panel-418 | IL-2Rα, TNFR1, HGF, IL-8 | −4656 + 3.43 × IL-2Rα + 0.48 × TNFR1+ 0.11 × HGF –0.45 × IL-8 | Q3*,‡ |
| Panel-617 | Elafin, IL-2Rα, TNFR1, HGF, IL-8, REG3α | 0.31 × log(elafin) + 0.13 × log(IL-2Rα) + 0.80 × log(TNFR1) + 0.20 × log(IL-8) + 0.02 × log(HGF) + 0.19 × log(REG3α) | 12.8 |
Third quartile.
In the original publication, 2 cutoffs at P = .4 and P = .66 were used, but in the current data, values of P were > .66.
In the original publication, the panel score was 1000 times the linear combination presented, then back-transformed on a probability scale through reverse logit, but in the current data the range of values obtains makes it impossible for most observations.