Pharmaceutical platform technology for the development of natural products
AssigneeSinoveda Canada, Inc.
InventorsYun Kau Tam, Jack Adam Tuszynski
Life of this patent
Estimating the term…
Abstract
The present invention provides a set of in vitro and in silico methodologies for predicting in vivo pharmacokinetics and pharmacodynamics of multiple components; the methodologies comprise mathematical models for solving multiple unknowns which are linearly independent and/or interacting with each other. The present invention can be applied to develop phytomedicines which contain multiple active ingredients without prior identification, isolation and purification of these components.
Claims9
1-22. (canceled)
A method of preparing a composition comprising multiple components for drug formulations, wherein the identities of or interactions among the components are not known a priori, the method comprises the steps of:
- a) preparing multiple samples of a mixture, said mixture comprises the same multiple components at different concentrations, wherein the identities of or interactions among the components are not known a priori;
- b) using in vitro assays or in silico methods to obtain parameters describing the rate of elimination of said components and their active metabolites in a plurality of mammalian tissue systems;
- c) using in vitro assays or in silico methods to obtain parameters describing distribution of said components and their active metabolites in a plurality of mammalian tissue systems;
- d) obtaining pharmacodynamic parameters describing the response of said samples of mixture;
- e) performing Subset-Selection Principal Component Analysis to reduce dimensionality of the parameters from (b) to (d), thereby generating parameters for active components in the mixture;
- f) inputting the parameters for active components from (e) into a first computer-based modeling system to estimate optimal weights of the active components, said modeling system comprises
- r
- ≈
- r
- _
- +
- ∑
- i
-
- w
- i
-
- (
- d
- i
- -
- d
- _
- i
- )
- +
- ∑
- i
-
- w
- i
- ′
-
- (
- d
- i
- -
- d
- _
- i
- )
- 2
- +
- ∑
- i
- ,
- j
-
- w
- i
- ,
- j
-
- (
- d
- i
- -
- d
- _
- i
- )
-
- (
- d
- j
- -
- d
- _
- j
- )
-
- min
- w
- i
- ,
- w
- i
- ′
- ,
- w
- i
- ,
- j
-
- {
- r
- -
- [
- r
- _
- +
- ∑
- i
-
- w
- i
-
- (
- d
- i
- -
- d
- _
- i
- )
- +
- ∑
- i
-
- w
- i
- ′
-
- (
- d
- i
- -
- d
- _
- i
- )
- 2
- +
- ∑
- i
- ,
- j
-
- w
- i
- ,
- j
-
- (
- d
- i
- -
- d
- _
- i
- )
-
- (
- d
- j
- -
- d
- _
- j
- )
- ]
- }
- 2
- wherein r is linearized response, r is average linearized response;
- wi is weight of the ith component, w′i is weight of the non-linear self-interaction behavior of the ith component, di is the dose of the ith component, di and dj are average doses of the ith and jth component, and wi,j is the weight of the interacting pair,
- min
- w
- i
- ,
- w
- i
- ′
- ,
- w
- i
- ,
- j
- is a minimization procedure to minimize the difference between r and said first modeling system by varying wi, w′i, wi,j,
- wherein said first modeling system generates outputs comprising (i) optimal weights αi and βi,j, which are optimal values of wi and w′i and wi,j obtained from said minimization procedure, and (ii) total dose of the components D, which is defined as Σi=1ndi;
- g) inputting the optimal weights, αi and βi,j, from (f) into a second computer-based modeling system that comprises
- A
- =
- α
- 0
- +
- ∑
- i
- =
- 1
- n
-
- α
- i
-
- x
- i
- +
- ∑
- i
- =
- 1
- n
-
- ∑
- j
- =
- 1
- n
-
- β
- i
- ,
- j
-
- x
- i
-
- x
- j
-
-
- max
- x
-
-
- α
- 0
- +
- ∑
- i
- =
- 1
- n
-
- α
- i
-
- x
- i
- +
- ∑
- i
- =
- 1
- n
-
- ∑
- j
- =
- 1
- n
-
- β
- i
- ,
- j
-
- x
- i
-
- x
- j
-
- s
- .
- t
- .
-
-
- ∑
- i
- =
- 1
- n
-
- x
- i
- =
- 1
-
- x
- i
- ≥
- 0
- ,
- ∀
- i
- wherein, A is estimated response generated by said active components per total dose, α0 is an average linearized response per total dose, xi and xj are defined as d′i/D and d′j/D respectively, which are fractions of said total dose D contributed by the ith and jth component respectively, and wherein
- D
- =
- ∑
- i
- =
- 1
- n
-
- d
- i
- ′
- ,
-
- max
- x
- is a maximization procedure achieved by varying xi, ∀i is for all i values,
- wherein said second modeling system generates outputs comprising dosages of active components that generate maximal response;
- and
- h) using dosages of active components that generate the maximal response from (g) to prepare a composition comprising multiple components, wherein the dosages for said multiple components are obtained from said dosages of active components that generate the maximal response.
Dependencies are inferred from the claim text.