Table 3. Final Multivariable Flexible Parametric Survival Model for Drug Discontinuation Associated With Ineffectivenessa.
| Covariate | Hazard ratio (95% CI) |
|---|---|
| Age | 0.99 (0.99-1.00) |
| Female sex | 1.11 (1.04-1.19)b |
| Baseline Psoriasis Area and Severity Index | 1.01 (1.01-1.02)b |
| BMIc | 0.01 (0.01-0.03)b |
| Palmoplantar psoriasis | 1.10 (1.01-1.20)b |
| Diabetes | 1.22 (1.09-1.35)b |
| COPD | 1.26 (1.00-1.58) |
| Flexural psoriasis | 1.06 (0.99-1.14) |
| Methotrexate use | 1.19 (1.05-1.33)b |
| Cyclosporine use | 3.00 (2.55-3.53)b |
| Psoriatic arthritis (in patients receiving treatment with ustekinumab) | 1.18 (1.03-1.35)b |
| Nail involvement (in patients receiving treatment with ustekinumab) | 1.23 (1.09-1.38)b |
| Previous biologic experience (compared with biologic-naive patients) | |
| 1 Previous biologic (in patients receiving treatment with ustekinumab) | 1.62 (1.42-1.85)b |
| 2 Previous biologics (in patients receiving treatment with ustekinumab) | 1.69 (1.43-2.00)b |
| Race and ethnicity (compared with White individuals) | |
| Asian | 1.21 (0.96-1.52) |
| Otherd | 1.20 (0.89-1.64) |
| Biologic therapies | |
| Biologic-naive White patients with no PsA and nail involvement | |
| Ustekinumab | 1 [Reference] |
| Adalimumab | 2.36 (2.02-2.76)b |
| Secukinumab | 1.22 (0.96-1.53) |
| Ixekizumab | 0.51 (0.22-1.20) |
| Guselkumab | 0.14 (0.03-0.61)b |
| In patients with PsA | |
| Ustekinumab | 1 [Reference] |
| Adalimumab | 1.81 (1.46-2.23)b |
| Secukinumab | 0.99 (0.74-1.32) |
| Ixekizumab | 0.41 (0.16-1.02) |
| Guselkumab | 0.07 (0.02-0.37)b |
| In patients with 1 previous biologic | |
| Ustekinumab | 1 [Reference] |
| Adalimumab | 1.51 (1.26-1.82)b |
| Secukinumab | 1.38 (1.11-1.72)b |
| Ixekizumab | 0.84 (0.50-1.41) |
| Guselkumab | 0.20 (0.08-0.55)b |
| In patients with ≥2 previous biologics | |
| Ustekinumab | 1 [Reference] |
| Adalimumab | 2.36 (1.85-3.03)b |
| Secukinumab | 1.85 (1.46-2.35)b |
| Ixekizumab | 2.07 (1.39-3.08)b |
| Guselkumab | 0.78 (0.41-1.50) |
| In patients with nail involvement | |
| Ustekinumab | 1 [Reference] |
| Adalimumab | 1.91 (1.65-2.21)b |
| Secukinumab | 1.09 (0.87-1.37) |
| Ixekizumab | 0.40 (0.17-0.92)b |
| Guselkumab | 0.10 (0.02-0.44)b |
| In Asian patients | |
| Ustekinumab | 1 [Reference] |
| Adalimumab | 3.10 (2.24-4.29)b |
| Secukinumab | 0.89 (0.57-1.37) |
| Ixekizumab | 0.43 (0.14-1.26) |
| Guselkumab | 0.04 (0.00-0.46)b |
| In patients of other racial and ethnic minority groups | |
| Ustekinumab | 1 [Reference] |
| Adalimumab | 2.09 (1.39-3.15)b |
| Secukinumab | 0.73 (0.43-1.25) |
| Ixekizumab | 0.46 (0.15-1.42) |
| Guselkumab | 0.13 (0.02-0.89)b |
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); COPD, chronic obstructive pulmonary disease; PsA, psoriatic arthritis.
a
Comparisons of model fit statistics suggested that 4 knots and 1 knot be placed for the restricted cubic splines to model the baseline hazard and the time-dependent effect of biologic treatment, respectively. For the interactions, the effects of the biologic therapies are compared with each other in each covariate strata.
b
Statistically significant results (P < .05).
c
BMI transformation = (BMI/10)−2.
d
Other race and ethnicity included Black British individuals.