Testing randomness for cancer risk

TL;DR

This paper examines stochastic models of cancer risk, testing key hypotheses about cell division probabilities and independence, and finds recent data consistent with these models.

Contribution

It provides empirical validation that certain simple probabilistic hypotheses align with observed cancer risk data.

Findings

Data supports fixed probability of cancerous cell development per division

Cell divisions appear to be independent in the context of cancer risk

Models based on these hypotheses are consistent with recent epidemiological data

Abstract

There are numerous stochastic models for cancer risk for a given tissue. Many rely on the following two hypotheses. 1. There is a fixed probability that a given cell division will eventually lead to a cancerous cell. 2. Cell divisions are nefarious or not independently of each other. We show that recent data on cancer risk and number of stem divisions is consistent with hypotheses 1 and 2.