Machine Learning-Based Prediction of Responsiveness to Neoadjuvant Chemoradiotheapy in Locally Advanced Rectal Cancer Patients from Endomicroscopy.

The protocol for treating locally advanced rectal cancer consists of the application of chemoradiotherapy (neoCRT) followed by surgical intervention. One issue for clinical oncologists is predicting the efficacy of neoCRT in order to adjust the dosage and avoid treatment toxicity in cases when surgery should be conducted promptly. Biomarkers may be used for this purpose along with in vivo cell-level images of the colorectal mucosa obtained by probe-based confocal laser endomicroscopy (pCLE) during colonoscopy. The aim of this article is to report our experience with Motiro, a computational framework that we developed for machine learning (ML) based analysis of pCLE videos for predicting neoCRT response in locally advanced rectal cancer patients. pCLE videos were collected from 47 patients who were diagnosed with locally advanced rectal cancer (T3/T4, or N+). The patients received neoCRT. Response to treatment by all patients was assessed by endoscopy along with biopsy and magnetic resonance imaging (MRI). Thirty-seven patients were classified as non-responsive to neoCRT because they presented a visible macroscopic neoplastic lesion, as confirmed by pCLE examination. Ten remaining patients were considered responsive to neoCRT because they presented lesions as a scar or small ulcer with negative biopsy, at post-treatment follow-up. Motiro was used for batch mode analysis of pCLE videos. It automatically characterized the tumoral region and its surroundings. That enabled classifying a patient as responsive or non-responsive to neoCRT based on pre-neoCRT pCLE videos. Motiro classified patients as responsive or non-responsive to neoCRT with an accuracy of ~ 0.62 when using images of the tumor. When using images of regions surrounding the tumor, it reached an accuracy of ~ 0.70. Feature analysis showed that spatial heterogeneity in fluorescence distribution within regions surrounding the tumor was the main contributor to predicting response to neoCRT. We developed a computational framework to predict response to neoCRT by locally advanced rectal cancer patients based on pCLE images acquired pre-neoCRT. We demonstrate that the analysis of the mucosa of the region surrounding the tumor provides stronger predictive power.

Limited inhibition of multiple nodes in a driver network blocks metastasis.

Metastasis suppression by high-dose, multi-drug targeting is unsuccessful due to network heterogeneity and compensatory network activation. Here, we show that targeting driver network signaling capacity by limited inhibition of core pathways is a more effective anti-metastatic strategy. This principle underlies the action of a physiological metastasis suppressor, Raf Kinase Inhibitory Protein (RKIP), that moderately decreases stress-regulated MAP kinase network activity, reducing output to transcription factors such as pro-metastastic BACH1 and motility-related target genes. We developed a low-dose four-drug mimic that blocks metastatic colonization in mouse breast cancer models and increases survival. Experiments and network flow modeling show limited inhibition of multiple pathways is required to overcome variation in MAPK network topology and suppress signaling output across heterogeneous tumor cells. Restricting inhibition of individual kinases dissipates surplus signal, preventing threshold activation of compensatory kinase networks. This low-dose multi-drug approach to decrease signaling capacity of driver networks represents a transformative, clinically relevant strategy for anti-metastatic treatment.

A comparative analysis of noise properties of stochastic binary models for a self-repressing and for an externally regulating gene.

This manuscript presents a comparison of noise properties exhibited by two stochastic binary models for: (i) a self-repressing gene; (ii) a repressed or activated externally regulating one. The stochastic models describe the dynamics of probability distributions governing two random variables, namely, protein numbers and the gene state as ON or OFF. In a previous work, we quantify noise in protein numbers by means of its Fano factor and write this quantity as a function of the covariance between the two random variables. Then we show that distributions governing the number of gene products can be super-Fano, Fano or sub-Fano if the covariance is, respectively, positive, null or negative. The latter condition is exclusive for the self-repressing gene and our analysis shows the conditions for which the Fano factor is a sufficient classifier of fluctuations in gene expression. In this work, we present the conditions for which the noise on the number of gene products generated from a self-repressing gene or an externally regulating one are quantitatively similar. That is important for inference of gene regulation from noise in gene expression quantitative data. Our results contribute to a classification of noise function in biological systems by theoretically demonstrating the mechanisms underpinning the higher precision in expression of a self-repressing gene in comparison with an externally regulated one.

Lessons and perspectives for applications of stochastic models in biological and cancer research.

The effects of randomness, an unavoidable feature of intracellular environments, are observed at higher hierarchical levels of living matter organization, such as cells, tissues, and organisms. Additionally, the many compounds interacting as a well-orchestrated network of reactions increase the difficulties of assessing these systems using only experiments. This limitation indicates that elucidation of the dynamics of biological systems is a complex task that will benefit from the establishment of principles to help describe, categorize, and predict the behavior of these systems. The theoretical machinery already available, or ones to be discovered to help solve biological problems, might play an important role in these processes. Here, we demonstrate the application of theoretical tools by discussing some biological problems that we have approached mathematically: fluctuations in gene expression and cell proliferation in the context of loss of contact inhibition. We discuss the methods that have been employed to provide the reader with a biologically motivated phenomenological perspective of the use of theoretical methods. Finally, we end this review with a discussion of new research perspectives motivated by our results.

Lessons and perspectives for applications of stochastic models in biological and cancer research

The effects of randomness, an unavoidable feature of intracellular environments, are observed at higher hierarchical levels of living matter organization, such as cells, tissues, and organisms. Additionally, the many compounds interacting as a well-orchestrated network of reactions increase the difficulties of assessing these systems using only experiments. This limitation indicates that elucidation of the dynamics of biological systems is a complex task that will benefit from the establishment of principles to help describe, categorize, and predict the behavior of these systems. The theoretical machinery already available, or ones to be discovered to help solve biological problems, might play an important role in these processes. Here, we demonstrate the application of theoretical tools by discussing some biological problems that we have approached mathematically: fluctuations in gene expression and cell proliferation in the context of loss of contact inhibition. We discuss the methods that have been employed to provide the reader with a biologically motivated phenomenological perspective of the use of theoretical methods. Finally, we end this review with a discussion of new research perspectives motivated by our results.

Stochastic model of contact inhibition and the proliferation of melanoma in situ.

Contact inhibition is a central feature orchestrating cell proliferation in culture experiments; its loss is associated with malignant transformation and tumorigenesis. We performed a co-culture experiment with human metastatic melanoma cell line (SKMEL- 147) and immortalized keratinocyte cells (HaCaT). After 8 days a spatial pattern was detected, characterized by the formation of clusters of melanoma cells surrounded by keratinocytes constraining their proliferation. In addition, we observed that the proportion of melanoma cells within the total population has increased. To explain our results we propose a spatial stochastic model (following a philosophy of the Widom-Rowlinson model from Statistical Physics and Molecular Chemistry) which considers cell proliferation, death, migration, and cell-to-cell interaction through contact inhibition. Our numerical simulations demonstrate that loss of contact inhibition is a sufficient mechanism, appropriate for an explanation of the increase in the proportion of tumor cells and generation of spatial patterns established in the conducted experiments.