Population pharmacokinetics of PF-06438179/GP1111 (an infliximab biosimilar) and reference infliximab in patients with moderately to severely active rheumatoid arthritis

Figures & data

Figure 1. Study design. (Cohen et al. Arthritis Res Ther. 2018;20:155. © The Author(s). Reprinted with permission [https://creativecommons.org/licenses/by/4.0/]). ^a A sample size of approximately 614 patients was planned for enrolment; the actual number of patients randomized was 650. ^b Intravenous PF-SZ-IFX or ref-IFX-EU 3 mg/kg was given as an induction regimen at Weeks 0, 2, and 6, followed by maintenance treatment with a 3-mg/kg dose starting at Week 14 and continuing every 8 weeks thereafter. Dose escalation to 5 mg/kg PF-SZ-IFX or ref-IFX-EU was permitted at or after Week 14 for patients with inadequate RA response. EOT, end of treatment; PF-SZ-IFX, PF-06438179/GP1111; RA, rheumatoid arthritis; ref-IFX-EU, infliximab authorized in the European Union.

Figure 2. PopPK model development schematic. PF-SZ-IFX, PF-06438179/GP1111; PopPK, population pharmacokinetics; ref-IFX-EU, infliximab authorized in the European Union; VPC, visual predictive check.

Table 1. Covariates included in population pharmacokinetic analysis.

PK parameter	Covariate
CL	SEX, BALB, BCRP, ADATL, MTXDC, BWBC
V1	SEX, BWT, ADATL

ADA, antidrug antibody; ADATL, ADA titers by LOCF; BALB, baseline albumin; BCRP, baseline C-reactive protein; BWBC, baseline white blood cell count; BWT, baseline body weight; CL, clearance; LOCF, last observation carried forward; MTXDC, dose of methotrexate categories; PK, pharmacokinetic; SEX, gender; V₁, volume of distribution in central compartment.

Table 2. Summary of demographics and key parameters by treatment group^a.

	PF-SZ-IFX (N = 323)	ref-IFX-EU (N = 326)
Baseline continuous covariates	Median (range)	Median (range)
Age (y)	54.0 (21–86)	53.5 (23–81)
Body weight (kg)	70.3 (32.5–179.4)	70.8 (36.2–162.7)
Albumin (g/dL)	3.9 (2.6–4.8)	3.9 (2.3–4.7)
hs-CRP (mg/L)	17.9 (0.47–135)	16.5 (0.75–203)
WBC (x 10^3/mm^3)	8.1 (2.6–18.1)	7.8 (2.9–18.1)
Categorical covariates	n (%)	n (%)
Sex
Males	65 (20.1)	62 (19.0)
Females	258 (79.9)	264 (81.0)
Race
White	257 (79.6)	247 (75.8)
Black	5 (1.5)	9 (2.8)
Asian	46 (14.2)	45 (13.8)
Other	15 (4.6)	25 (7.7)
ADA status^b
Positive	157 (48.6)	167 (51.2)
Negative	163 (50.5)	158 (48.5)
Japanese ethnicity
Yes	23 (7.1)	23 (7.1)
No	300 (92.9)	303 (92.9)
MTXDC
<10 mg/wk	20 (6.2)	28 (8.6)
Between 10 and 20 mg/wk	250 (77.4)	232 (71.2)
≥20 mg/wk	53 (16.4)	66 (20.2)

^a Summary statistics were calculated prior to any covariate imputations.

^b A positive subject was defined as having at least one post-dose sample tested positive during Treatment Period 1 (Week 0 to Week 30), regardless of the pre-dose ADA status.

ADA, antidrug antibody; hs-CRP, high-sensitivity C-reactive protein; MTXDC, dose of methotrexate categories; N, number of patients; PF-SZ-IFX, PF-06438179/GP1111; ref-IFX-EU, infliximab authorized in the European Union; WBC, white blood cell count.

Table 3. Final model parameter estimates and CIs for the ref-IFX-EU and PF-SZ-IFX arms.

Parameter	ref-IFX-EU			PF-SZ-IFX
	NONMEM results OFV = 2062.429		Nonparametric bootstrap results	NONMEM results OFV = 2109.553		Nonparametric bootstrap results
	Estimate [%RSE]^a	Shrinkage, %	Estimate, median [95% CI]^b	Estimate [%RSE]^a	Shrinkage, %	Estimate, median [95% CI]^e
CL, L/h	0.015 [5.31]	^d	0.015 [0.014, 0.016]	0.014 [5.29]	^d	0.014 [0.013, 0.015]
ADA titers on CL	0.478 [4.48]		0.479 [0.381, 0.607]	0.394 [4.01]		0.394 [0.316, 0.509]
Sex on CL	0.255 [36.5]		0.252 [0.145, 0.375]	0.283 [29.4]		0.280 [0.145, 0.477]
V1, L	3.57 [6.64]	^d	3.57 [3.37, 3.81]	3.38 [6.07]	^d	3.39 [3.19, 3.61]
BWT on V1	0.669 [20.2]		0.676 [0.496, 0.862]	0.425 [32.2]		0.425 [0.222, 0.634]
Q, L/h^c	0.2	^d	0.2	0.2	^d	0.2
V2, L	1.65 [18.3]	^d	1.65 [1.33, 1.99]	1.70 [17.9]	^d	1.69 [1.38, 2.00]
CL ω^2 (%CV)	0.161 (40.1) [9.94]	10.2	0.162 (40.2) [0.135, 0.190]	0.186 (43.1) [8.23]	10.2	0.181 (42.5) [0.142, 0.226]
V1 ω^2 (%CV)	0.056 (23.7) [42.7]	51.2	0.053 (23.0) [0.013, 0.107]	0.079 (28.1) [36.3]	51.2	0.071 (26.6) [0.029, 0.116]
Res prop error	0.770 [0.935]	23.9	0.767 [0.685, 0.857]	0.776 [1.01]	23.9	0.776 [0.690, 0.859]

^a %RSE; relative standard error of the parameter was manually calculated as (SE/final parameter estimate)*100.

^b All 1000 of 1000 bootstraps had successful minimization, with 0 of those bootstraps with estimates near the boundary; hence, a total of 1000 bootstraps were included in the calculation of the 95% CI.

^c Fixed to published literature data (Buurman DJ, et al. Aliment Pharmacol Ther 2015;42(5): 529–39; Fasanmade AA, et al. Clin Ther 2011;33(7): 946–64; Fasanmade AA, et al. Eur J Clin Pharmacol 2009;65(12): 1211–28; Fasanmade AA, et al. Int J Clin Pharmacol Ther 2010;48(5): 297–308).

^d Not estimated or not applicable.

^e 999 of 1000 bootstraps had successful minimization, with 0 of those bootstraps with estimates near the boundary; hence, a total of 999 bootstraps were included in the calculation of the 95% CI.

ADA, antidrug antibody; BWT, baseline body weight; CI, confidence interval; CL, clearance; CV, coefficient of variation; NONMEM, nonlinear mixed-effects modeling; OFV, objective function value; PF-SZ-IFX, PF-06438179/GP1111; Q, intercompartmental clearance; ref-IFX-EU, infliximab authorized in the European Union; Res prop error, residual proportional error; SE, standard error; V₁, volume of distribution in central compartment; V₂, volume of distribution in peripheral compartment.

Figure 3. Diagnostic plots for final model for the ref-IFX-EU arm (a) and the PF-SZ-IFX arm (b). In the scatter plots of observations vs. predictions, the solid red line and dashed blue line show the reference line (diagonal line) and linear regression line based on the individual data points. In the scatter plots of residuals, the solid and dotted lines show the reference line (y = 0) and the locally weighted scatter-plot smoothing trend line (LOESS), respectively. Observed concentrations and individual predictions were log-transformed. PF-SZ-IFX, PF-06438179/GP1111; ref-IFX-EU, infliximab authorized in the European Union.

Figure 4. VPC for final model for the ref-IFX-EU arm (a) and the PF-SZ-IFX arm (b). Blue circles represent the observed data, and red lines represent the median (solid line), 5th percentile (dashed line), and 95th percentile (dashed line) of the observed data. For 1000 simulated trials, the median, 5th percentile, and 95th percentile of simulated concentrations were calculated for each time point and are represented by black lines. The 95% CI for simulated median and each percentile are shown by light pink and light blue shaded areas. CI, confidence interval; h, hours; LLOQ lower limit of quantification; PF-SZ-IFX, PF-06438179/GP1111; ref-IFX-EU, infliximab authorized in the European Union; VPC, visual predictive check.