Christelle Rodrigues1, Sylvain Fouliard1
1Quantitative Pharmacology, Translational Medicine, Servier
Objectives The dosing of monoclonal antibodies (mAbs) in oncology is often based on body size, despite evidence suggesting that fixed dosing may be equally effective [1]. Body weight is commonly used as a covariate on central volume (Vc) and clearance (CL) [2] but may not be the best predictor of blood volume and mAbs elimination [1]. This study aims to identify the most relevant body size-related covariate for describing variability in Vc and CL using data from three therapeutic antibodies in clinical development. Methods Population PK models were developed for three mAbs using nonlinear mixed-effects modeling with data of three phase I dose-escalation studies: 39 patients for mAb1 (dose range of 150-fold), 60 for mAb2 (dose range of 667-fold), and 70 for mAb3 (dose range of 10-fold) . Two-compartment models with linear elimination were identified for mAb1 and mAb2, while mAb3 exhibited both linear and nonlinear elimination. The following body size-related covariates were tested on Vc and CL: body weight (BW), height (HT), body mass index (BMI), body surface area (BSA) – calculated using both the Mosteller formula (BSA1) [3] and the DuBois and DuBois formula (BSA2) [4], lean body weight (LBW) – calculated using the formulas of Janmahasatian et al. (LBW1) [5] and Boer (LBW2) [6], and fat mass index (FMI), derived from LBW1 and LBW2. No other covariates were tested. A forward inclusion approach was performed following exploratory eta-covariate analysis, with selection based on the change in objective function value (?OFV), reduction in inter-individual variability (IIV), and covariate effect size (ß). Results All size-related covariates were highly correlated. In the eta-covariate analysis, BMI and FMI were not significantly associated with Vc for mAb1 and mAb3, as well as BMI, FMI, HT and LBW1 on CL of mAb1 and HT and LBW on CL of mAb2 (Pearson correlation, p > 0.05), and thus were not included in the forward inclusion step. Of note, significant correlations were stronger with Vc. For Vc, the most significant covariates were LBW2 for mAb1 (ß = 1.25), BSA1 for mAb2 (ß = 1.15) and BSA2 for mAb3 (ß = 1.16), with reductions in OFV of -23.36, -22.43 and -18.43, respectively. For each of the three mAbs, effect sizes (ß) for these covariates had values >1, suggesting a more than proportional increase of Vc with body size. Incorporating size covariates led to 28.20%, 17.11% and 19.12% reductions in IIV for Vc across mAb1, mAb2 and mAb3, respectively. For CL, most impactful covariates were BSA1 for mAb1 (ß = 1.19), BW for mAb2 (ß = 0.74) and BSA1 for mAb3 (ß = 1.53) but with impacts less pronounced (?OFV of -4.72, -6.43 and -9.16, respectively). The decrease in IIV of those additions on CL were 10.64%, 15.88% and 16.29% respectively. Of note, power coefficient beta for BW effect was between 0.57 and 0.70 on Vc and between 0.59 and 0.85 on CL in line with values reported in [1]. Conclusions These results suggest that body size has a more pronounced impact on Vc than CL, supporting the hypothesis that distribution is more size-dependent than elimination for mAbs. Indeed, the weaker association between CL and body size might reflect the complexity of mAb clearance mechanisms, which involve both target-mediated drug disposition (TMDD) and nonspecific FcRn-mediated recycling [6] that are not related to body size [1]. Both BSA and LBW seemed relevant for covariates for Vc. The volume of distribution of mAbs is primarily restricted to plasma and extracellular fluids, making body composition less relevant than for small molecules [1]. BSA includes contributions from both fat and lean tissue, with a higher correlation with lean mass [8]. However, blood volume scales more closely with LBW rather BW [5] or BSA [9]. Thus, LBW seems to be a more physiologically relevant covariate for Vc than BW or BSA. Further investigations across a broader set of mAbs are warranted to refine body size covariate selection in PK modeling.
[1] Hendrikx JJMA, Haanen JBAG, Voest EE, Schellens JHM, Huitema ADR, Beijnen JH. Fixed Dosing of Monoclonal Antibodies in Oncology. Oncologist. 2017 Oct;22(10):1212-1221 [2] Bajaj G, Suryawanshi S, Roy A, Gupta M. Evaluation of covariate effects on pharmacokinetics of monoclonal antibodies in oncology. Br J Clin Pharmacol. 2019; 85: 2045–2058 [3] Mosteller RD. Simplified calculation of body-surface area. N Engl J Med. 1987 Oct 22;317(17):1098 [4] Du Bois D, Du Bois EF. A formula to estimate the approximate surface area if height and weight be known. 1916. Nutrition. 1989 Sep-Oct;5(5):303-11; discussion 312-3 [5] Janmahasatian S, Duffull SB, Ash S, Ward LC, Byrne NM, Green B. Quantification of lean bodyweight. Clin Pharmacokinet. 2005;44(10):1051-65 [6] Boer P. Estimated lean body mass as an index for normalization of body fluid volumes in humans. Am J Physiol. 1984 Oct;247(4 Pt 2):F632-6 [7] Dostalek M, Gardner I, Gurbaxani BM, Rose RH, Chetty M. Pharmacokinetics, pharmacodynamics and physiologically-based pharmacokinetic modelling of monoclonal antibodies. Clin Pharmacokinet. 2013 Feb;52(2):83-124 [8] Zanforlini BM, Alessi A, Pontarin A, De Rui M, Zoccarato F, Seccia DM, Trevisan C, Brunello A, Basso U, Manzato E, Sergi G. Association of body surface area with fat mass, free fat mass and total weight in healthy individuals, and implications for the dosage of cytotoxic drugs. Clin Nutr ESPEN. 2021 Jun;43:471-477 [9] Oberholzer L, Montero D, Robach P, et al. Determinants and reference values for blood volume and total hemoglobin mass in women and men. Am J Hematol. 2024; 99(1): 88-98
Reference: PAGE 33 (2025) Abstr 11648 [www.page-meeting.org/?abstract=11648]
Poster: Methodology - Covariate/Variability Models