Abstract 2451

Background.

Biomedical research is trending to the development of increasingly more sophisticated antibody-based biologics, such as bispecific antibodies, immunocytokines and antibody-drug conjugates. Compared to traditional mAbs, development of more complex, and less natural, fusion proteins are challenged by problems with yield, stability, toxicity, immunogenicity and Pk. Previously, we reported potent anti-lymphoma activity for both anti-CD22/CD20 bispecific hexavalent antibodies (bsHexAbs; Rossi et al., Blood 2009; 113: 6161–6171) and immunocytokines comprising anti-CD20 mAb and tetrameric IFNα2b (IgG-IFN; Rossi et al., Blood 2009;114:3864-71). For each class of immunoconjugate that were produced with the Dock-and-Lock (DNL) method using an IgG module having an AD2 peptide fused at the C-terminal end of the Fc, we identified Pk and in vivo stability as potentially limiting parameters.

Methods.

Using the DNL method, we generated a new class of IgG modules, which have an AD2 peptide fused at the C-terminal end of the kappa light chain and were used to produce Ck-based (indicated by *) bsHexAbs and IgG-IFNα2b, for comparison with homologous Fc-based constructs. The Ck-based immunocytokine 20*-2b, has a similar molecular size and composition to its Fc-based homolog, 20-2b, each comprising the humanized anti-CD20 mAb, veltuzumab, and 4 IFNα2b groups that are fused at the C-terminal ends of the light or heavy chains, respectively. The Ck-based bsHexAb 22*-(20)-(20) and its Fc-based homolog 22-(20)-(20), each comprise the humanized anti-CD22 mAb, epratuzumab, and 4 Fabs of veltuzumab, which are fused at the C-terminal ends of the light and heavy chains, respectively.

Results.

The Ck-based constructs exhibited superior Pk (longer T1/2) in mice and rabbits, with either IV or SC injection (Table 1). Although the bsHexAbs and IgG-IFN are considerably stable in sera, analysis of Pk samples indicated that some dissociation occurs in vivo, presumably by intracellular processing. The in vivo dissociation rate for 20*-2b (0.18%/h) was 5.4-fold slower than 20-2b (0.97%/h). Similarly, 22*-(20)-(20) (0.19%/h) was more stable in vivo than 22-(20)-(20) (0.55%/h). Fc effector functions were markedly enhanced for the Ck-based constructs. Where 20*-2b induced strong CDC, which approached the potency of veltuzumab, no activity was evident for 20-2b. Epratuzumab did not have CDC, while 22-(20)-(20) achieved modestly increased activity, and 22*-(20)-(20) induced even greater CDC. In vitro, epratuzumab induced minimal ADCC and 22-(20)-(20) did not show a statistically significant improvement. However, 22*-(20)-(20) exhibited potent ADCC, which was similar to that of veltuzumab. Finally, the Ck-based conjugates were more effective than the Fc-based counterparts for therapy of disseminated NHL (Daudi) xenografts. Using a single low dose of 0.25 mg, superiority (P=0.0351) was demonstrated for 20*-2b (median survival time [MST]>189 days, 87% cures), compared to 20-2b (MST=134.5 days, 37.5% cures). At a high (1 mg) dose, 22*-(20)-(20) (MST>98 days, 100% survival) was superior (P<0.0001) to 22-(20)-(20) (MST=71 days, 10% survival). With low-dose (10 μg) treatment, the MST was 91 days for 22*-(20)-(20), compared to 50.5 days for 22-(20-(20) (P=0.0014).

Conclusions.

These new constructs demonstrate the further enhancement of two different classes of fusion proteins with already potent anti-lymphoma efficacy. Due to extended Pk, improved stability and enhanced effector function, the Ck-based design is superior for in vivo applications. The strategy of designing antibody fusion proteins at the C-terminus of the light chain, instead of at the commonly used Fc, may improve the in vivo efficacy of most immunoconjugates in general, and the DNL conjugates in particular.

Table 1

Summary of pharmacokinetic parameters

SpeciesRouteDose (mg)ConstructT1/2 (h)Tmax(h)Cmax(μg/mL)AUC (h*μg/ml)MRT (h)
Mouse IV 1.0 20*-2b 36.2 6.0 649.0 32516.5 55.2 
   20-2b 17.1 6.0 629.8 15514.0 19.1 
Mouse SC 1.0 20*-2b 37.9 16.0 312.1 18318.2 62.1 
   20-2b 16.0 16.0 146.0 6498.6 30.9 
Mouse SC 0.5 22*-(20)-(20) 106.5 24.0 50.6 6704.7 153.1 
   22-(20)-(20) 54.5 16.0 26.5 1752.9 85.2 
Rabbit SC 18 22*-(20)-(20) 117.9 53.3 31.6 6079.1 179.6 
   22-(20)-(20) 51.1 37.3 17.8 1838.4 89.2 
SpeciesRouteDose (mg)ConstructT1/2 (h)Tmax(h)Cmax(μg/mL)AUC (h*μg/ml)MRT (h)
Mouse IV 1.0 20*-2b 36.2 6.0 649.0 32516.5 55.2 
   20-2b 17.1 6.0 629.8 15514.0 19.1 
Mouse SC 1.0 20*-2b 37.9 16.0 312.1 18318.2 62.1 
   20-2b 16.0 16.0 146.0 6498.6 30.9 
Mouse SC 0.5 22*-(20)-(20) 106.5 24.0 50.6 6704.7 153.1 
   22-(20)-(20) 54.5 16.0 26.5 1752.9 85.2 
Rabbit SC 18 22*-(20)-(20) 117.9 53.3 31.6 6079.1 179.6 
   22-(20)-(20) 51.1 37.3 17.8 1838.4 89.2 

T1/2, elimination half-life; Tmax, time of maximal concentration; Cmax, maximal concentration; AUC, area under the curve; MRT, mean residence time

Disclosures:

Rossi:Immunomedics, Inc.: Employment; IBC Pharmaceuticals Inc.: Employment. Cardillo:Immunomedics, Inc: Employment. Goldenberg:Immunomedics: Employment, Equity Ownership. Chang:Immunomedics, Inc.: Employment.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution