# The impact of model assumptions in interpreting cell kinetic studies

## TL;DR

This paper examines how simplifying assumptions affect the interpretation of cell dynamics measured using stable isotope labelling in humans.

## Contribution

The paper evaluates the impact of three common assumptions in cell kinetic studies and proposes ways to reduce errors from these assumptions.

## Key findings

- Assumptions about closed populations and homogeneity can lead to errors in interpreting cell dynamics.
- Pragmatic approaches are suggested to reduce errors caused by these assumptions.
- Stable isotope labelling is highlighted as a preferred method for measuring cell dynamics in humans.

## Abstract

Stable isotope labelling is one of the best methods currently available for quantifying cell dynamics in vivo, particularly in humans where the absence of toxicity makes it preferable over other techniques such as CFSE or BrdU. Interpretation of stable isotope labelling data (as for BrdU and CFSE) necessitates simplifying assumptions. Here we investigate the impact of three of the most commonly used simplifying assumptions: (i) that the cell population of interest is closed, (ii) that the population of interest is kinetically homogeneous, and (iii) that the population is spatially homogeneous and suggest pragmatic ways in which the resulting errors can be reduced.

Our immune response is responsible for protecting us from infectious disease and cancer. The immune system is not static, immune cells are consistent proliferating and dying even in healthy individuals. Understanding the rate at which immune cells proliferate and die is an important to understanding how our immune systems function; for example how immune memory is maintained or how a diverse repertoire of cells capable of responding to any invading pathogen is generated. Cell dynamics are usually measured using a tracer dye. A method called stable isotope labelling is one of the best methods for quantifying cell dynamics in humans in vivo as it is nontoxic. However, interpretation of any labelling experiment requires assumptions. In this paper we discuss the impact of some commonly used assumptions and suggest ways in which the resulting errors can be reduced.