Translational considerations for cancer nanomedicine

Abstract

There are many important considerations during preclinical development of cancer nanomedicines, including: 1) unique aspects of animal study design; 2) the difficulties in evaluating biological potency, especially for complex formulations; 3) the importance of analytical methods that can determine platform stability in vivo, and differentiate bound and free active pharmaceutical ingredient (API) in biological matrices; and 4) the appropriateness of current dose scaling techniques for estimation of clinical first-in-man dose from preclinical data. Biologics share many commonalities with nanotechnology products with regard to complexity and biological attributes, and can, in some cases, provide context for dealing with these preclinical issues. In other instances, such as the case of in vivo stability analysis, new approaches are required. This paper will discuss the significance of these preclinical issues, and present examples of current methods and best practices for addressing them. Where possible, these recommendations are justified using the existing regulatory guidance literature.

Fig. 1

Nanoplatform-drug stability.

Fig. 2

¹⁴C-C6 ceramide and ³H-DSPC plasma profiles. Reproduced with permission from Drug Metabolism and Disposition, Zolnik et al., 2008 [39]. Plasma time profiles are expressed as the percent injected dose per mL. (◆), ³H-DSPC of ceramide liposome; (□), ³H-DSPC of control liposome on y−1 axis; (●) ¹⁴C-ceramide on y−2 axis. Each point represents the mean ± std. dev. from n=4–5 rats.

Fig. 3

Blood concentration-time profiles for TNF-Au, Au and native TNF. Blood profiles of colloidal gold bound TNF (TNF-Au), gold (Au) and native TNF (Native TNF), are expressed as % injected dose/mL. The TNF concentration was determined by ELISA and gold concentration was determined by ICP-MS. Each symbol represents the mean±SD for that time point (n=4–5).

Fig. 4

Au–TNF/Au-hyperbolic stability model. The rectangular hyperbolic Michaelis-Menten type equation, ratio=1–(1*t/(t+t₅₀)), was fit to the pooled Au–TNF to gold (Au) concentration ratio-time profile. Data are presented as mean (n=4). Lines represent the fit of the hyperbolic stability model to the blood concentration-time data.

Fig. 5

Pharmacokinetic model for TNF release. Pharmacokinetic parameters were obtained by fitting the blood concentration-time data for gold (C_Au) and Au–TNF (C_T). The TNF release was estimated using the hyperbolic stability equation, C_Rel=C_T(0)*exp (−k₁*t)*(t/(t+t₅₀). The first-order elimination constant (k₂) and concentration at time zero (C_N(0)) for native TNF (C_N) were fixed. The first-order elimination rate for gold (k₁), hyperbolic time to 50% TNF release (t₅₀), and time zero concentrations for gold (C_Au(0)) and Au–TNF (C_T(0)) were obtained from fitting the kinetic model to the data.

Fig. 6

Au–TNF release model. The Au–TNF release model was fit to the pooled Au–TNF and gold(Au) blood concentration-time data, expressed as %ID/mL. Data are presented as mean (n=4–5). Lines represent the fit of the Au–TNF release model to the blood concentration time data.

Fig. 7

Au–TNF release model simulations. The native TNF and released free TNF blood concentration data were simulated using the Au–TNF release model and estimated pharmacokinetic parameters. Lines represent the simulated blood concentration-time data for the free TNF and native TNF comparison.

Fig. 8

Hypothetical dose–response. Preclinical dose-efficacy/dose-toxicity profiles for hypothetical drug X are displayed: MTD, maximum tolerated dose ED50; effective dose 50=%; LD50, lethal dose 50%.

Fig. 9

Allometric Scaling of Au–TNF by the Power Model. The clearance data for the CytImmune Au–TNF formulation in multiple species were fit to the power equations CL=a*BW^b. Data points are presented as the mean (n=2) of two clinical and rabbit dose levels, and the single estimate from a single rat dose level. Lines represent the fit of the model to the preclinical data, excluding the clinical data.

Fig. 10

Allometric scaling of Au–TNF by the brain weight (BrW) product model. The clearance data were fit to the power equations BrW×CL=a*BW^b. Data points are presented as the mean (n=2) for both human and rabbit dose levels, and the single estimate for rat. Lines represent the fit of the model to all data.