Is lung cancer heterogeneity a chaotic process?

Speculative fractal explanation

James Fleck, MD, PhD: Anticancerweb 07(09), 2021

The fractal pattern is present throughout nature. These two leopards may not be exactly alike, but self-similarity leads to immediate species identification. In fact, the leopard skin pattern is fractal, a rough or fragmented geometric shape that can be divided into parts, each of which is a reduced-size copy of the whole, as defined by Benoît Mandelbrot. In 1980, the mathematician moved to IBM, where he used complex numbers and quadratic polynomials to build an infinite iteration image. This image, known as Mandelbrot set, was the first to reproduce in an electronic system the fractal geometry found in nature. Fractal geometry is defined by multiple levels of organization, irregular shapes and self-similarity. Transposed to biological models, this phenotypic expression is present in both health and disease. Tumor cells have fractal expression, which allows the identification of specific patterns used in pathological diagnosis. Let’s take lung adenocarcinoma as an experimental model. Tumor heterogeneity was identified in light microscopy in former described five histologic subtypes (lepidic, acinar, papillary, micropapillary and solid). Each subtype is an iteration image making possible specific morphologic diagnosis. 


Since the beginning of the 21st  century, an incipient tumor molecular expression was described in most solid tumors. In 2014, the Lung Cancer Mutation Consortium brought together 20 institutions in the US and identified driver mutations in 62% of lung adenocarcinoma. Patients who expressed a driver mutation and were treated with biological targeted drugs had a median survival of 3.5 years. Patients who expressed a driver mutation who did not receive biological targeted therapy had a median survival of 2.4 years. The median survival decreased even further (2.1 years) in those patients who did not express any driver mutation and were not treated with targeted therapy. Therefore, the identification of a driver mutation became a parameter of quality of care in the treatment of lung cancer.Currently, several driver mutations have been described in lung adenocarcinoma, creating genomic patterns, whose molecular spatial geometry has not yet been entirely identified. The figure represents nine driver mutations found in lung adenocarcinoma, showing their respective percentage of expression. 


Each specific molecular phenotypes described in lung adenocarcinoma, is an actionable mutation and has been used to turn off addictive oncogenic cell behavior. The table below describe some therapeutic options for advanced lung adenocarcinoma according its molecular phenotype. Grade of recommendation varies according the evidence-based already provided in clinical trial.


Unfortunately, this is a dynamic process and the opportunistic molecular phenotypic plasticity of lung adenocarcinoma easily leads to drug resistance. The recognition of a supposed spatial molecular geometry and its multidimensional behavior would be a very important step to improve therapeutic results in lung adenocarcinoma.

 

References:

1.     Benoît Mandelbrot: The Fractal Geometry of Nature, 1983
2.     Lung Cancer Mutation Consortium, 2014: https://www.lungcancerresearchfoundation.org/research/lung-cancer-mutation-consortium/
3.     Ian Stewart: Mathematics of Life: Unlocking the Secrets of Existence, Profile Books, 368p, London, 2012

4.     Photo by Roberta Doyle on Unsplash (modified)