Clinical Application of Computational Methods in Precision Oncology: A Review

Reproducibility of prediction models in health services research

The field of health services research studies the health care system by examining outcomes relevant to patients and clinicians but also health economists and policy makers. Such outcomes often include health care spending, and utilization of care services. Building accurate prediction models using reproducible research practices for health services research is important for evidence-based decision making. Several systematic reviews have summarized prediction models for outcomes relevant to health services research, but these systematic reviews do not present a thorough assessment of reproducibility and research quality of the prediction modelling studies. In the present commentary, we discuss how recent advances in prediction modelling in other medical fields can be applied to health services research. We also describe the current status of prediction modelling in health services research, and we summarize available methodological guidance for the development, update, external validation and systematic appraisal of prediction models.

A smart hospital-driven approach to precision pharmacovigilance

Researchers, regulatory agencies, and the pharmaceutical industry are moving towards precision pharmacovigilance as a comprehensive framework for drug safety assessment, at the service of the individual patient, by clustering specific risk groups in different databases. This article explores its implementation by focusing on: (i) designing a new data collection infrastructure, (ii) exploring new computational methods suitable for drug safety data, and (iii) providing a computer-aided framework for distributed clinical decisions with the aim of compiling a personalized information leaflet with specific reference to a drug's risks and adverse drug reactions. These goals can be achieved by using 'smart hospitals' as the principal data sources and by employing methods of precision medicine and medical statistics to supplement current public health decisions.

A computational method for prioritizing targeted therapies in precision oncology: performance analysis in the SHIVA01 trial

Precision oncology is currently based on pairing molecularly targeted agents (MTA) to predefined single driver genes or biomarkers. Each tumor harbors a combination of a large number of potential genetic alterations of multiple driver genes in a complex system that limits the potential of this approach. We have developed an artificial intelligence (AI)-assisted computational method, the digital drug-assignment (DDA) system, to prioritize potential MTAs for each cancer patient based on the complex individual molecular profile of their tumor. We analyzed the clinical benefit of the DDA system on the molecular and clinical outcome data of patients treated in the SHIVA01 precision oncology clinical trial with MTAs matched to individual genetic alterations or biomarkers of their tumor. We found that the DDA score assigned to MTAs was significantly higher in patients experiencing disease control than in patients with progressive disease (1523 versus 580, P = 0.037). The median PFS was also significantly longer in patients receiving MTAs with high (1000+ <) than with low (<0) DDA scores (3.95 versus 1.95 months, P = 0.044). Our results indicate that AI-based systems, like DDA, are promising new tools for oncologists to improve the clinical benefit of precision oncology.

Impact of deep learning-determined smoking status on mortality of cancer patients: never too late to quit

Background

Persistent smoking after cancer diagnosis is associated with increased overall mortality (OM) and cancer mortality (CM). According to the 2020 Surgeon General's report, smoking cessation may reduce CM but supporting evidence is not wide. Use of deep learning-based modeling that enables universal natural language processing of medical narratives to acquire population-based real-life smoking data may help overcome the challenge. We assessed the effect of smoking status and within-1-year smoking cessation on CM by an in-house adapted freely available language processing algorithm.

Materials and methods

This cross-sectional real-world study included 29 823 patients diagnosed with cancer in 2009-2018 in Southwest Finland. The medical narrative, International Classification of Diseases-10th edition codes, histology, cancer treatment records, and death certificates were combined. Over 162 000 sentences describing tobacco smoking behavior were analyzed with ULMFiT and BERT algorithms.

Results

The language model classified the smoking status of 23 031 patients. Recent quitters had reduced CM [hazard ratio (HR) 0.80 (0.74-0.87)] and OM [HR 0.78 (0.72-0.84)] compared to persistent smokers. Compared to never smokers, persistent smokers had increased CM in head and neck, gastro-esophageal, pancreatic, lung, prostate, and breast cancer and Hodgkin's lymphoma, irrespective of age, comorbidities, performance status, or presence of metastatic disease. Increased CM was also observed in smokers with colorectal cancer, men with melanoma or bladder cancer, and lymphoid and myeloid leukemia, but no longer independently of the abovementioned covariates. Specificity and sensitivity were 96%/96%, 98%/68%, and 88%/99% for never, former, and current smokers, respectively, being essentially the same with both models.

Conclusions

Deep learning can be used to classify large amounts of smoking data from the medical narrative with good accuracy. The results highlight the detrimental effects of persistent smoking in oncologic patients and emphasize that smoking cessation should always be an essential element of patient counseling.

•

Deep learning/universal language modeling was used to extract smoking status of cancer patients. Good accuracy was observed.

•

Those who continue smoking after cancer diagnosis had increased CM compared to never smokers.

•

Recent within-1-year cessation reduced this mortality.

•

Detrimental effects of smoking were observed in multiple types of early- and advanced-stage cancers, including the elderly.

•

We conclude that smoking cessation support should always be included in cancer care.

MRPL15 is a novel prognostic biomarker and therapeutic target for epithelial ovarian cancer

PURPOSE

To analyze the role of six human epididymis protein 4 (HE4)-related mitochondrial ribosomal proteins (MRPs) in ovarian cancer and selected MRPL15, which is most closely related to the tumorigenesis and prognosis of ovarian cancer, for further analyses.

METHODS

Using STRING database and MCODE plugin in Cytoscape, six MRPs were identified among genes that are upregulated in response to HE4 overexpression in epithelial ovarian cancer cells. The Cancer Genome Atlas (TCGA) ovarian cancer, GTEX, Oncomine, and TISIDB were used to analyze the expression of the six MRPs. The prognostic impact and genetic variation of these six MRPs in ovarian cancer were evaluated using Kaplan-Meier Plotter and cBioPortal, respectively. MRPL15 was selected for immunohistochemistry and GEO verification. TCGA ovarian cancer data, gene set enrichment analysis, and Enrichr were used to explore the mechanism of MRPL15 in ovarian cancer. Finally, the relationship between MRPL15 expression and immune subtype, tumor-infiltrating lymphocytes, and immune regulatory factors was analyzed using TCGA ovarian cancer data and TISIDB.

RESULTS

Six MRPs (MRPL10, MRPL15, MRPL36, MRPL39, MRPS16, and MRPS31) related to HE4 in ovarian cancer were selected. MRPL15 was highly expressed and amplified in ovarian cancer and was related to the poor prognosis of patients. Mechanism analysis indicated that MRPL15 plays a role in ovarian cancer through pathways such as the cell cycle, DNA repair, and mTOR 1 signaling. High expression of MRPL15 in ovarian cancer may be associated with its amplification and hypomethylation. Additionally, MRPL15 showed the lowest expression in C3 ovarian cancer and was correlated with proliferation of CD8⁺ T cells and dendritic cells as well as TGFβR1 and IDO1 expression.

CONCLUSION

MRPL15 may be a prognostic indicator and therapeutic target for ovarian cancer. Because of its close correlation with HE4, this study provides insights into the mechanism of HE4 in ovarian cancer.

The intervention, the patient and the illness - Personalizing non-invasive brain stimulation in psychiatry

Current hypotheses on the therapeutic action of non-invasive brain stimulation (NIBS) in psychiatric disorders build on the abundant data from neuroimaging studies. This makes NIBS a very promising tool for developing personalized interventions within a precision medicine framework. NIBS methods fundamentally vary in their neurophysiological properties. They comprise repetitive transcranial magnetic stimulation (rTMS) and its variants (e.g. theta burst stimulation - TBS) as well as different types of transcranial electrical stimulation (tES), with the largest body of evidence for transcranial direct current stimulation (tDCS). In the last two decades, significant conceptual progress has been made in terms of NIBS targets, i.e. from single brain regions to neural circuits and to functional connectivity as well as their states, recently leading to brain state modulating closed-loop approaches. Regarding structural and functional brain anatomy, NIBS meets an individually unique constellation, which varies across normal and pathophysiological states. Thus, individual constitutions and signatures of disorders may be indistinguishable at a given time point, but can theoretically be parsed along course- and treatment-related trajectories. We address precision interventions on three levels: 1) the NIBS intervention, 2) the constitutional factors of a single patient, and 3) the phenotypes and pathophysiology of illness. With examples from research on depressive disorders, we propose solutions and discuss future perspectives, e.g. individual MRI-based electrical field strength as a proxy for NIBS dosage, and also symptoms, their clusters, or biotypes instead of disorder focused NIBS. In conclusion, we propose interleaved research on these three levels along a general track of reverse and forward translation including both clinically directed research in preclinical model systems, and biomarker guided controlled clinical trials. Besides driving the development of safe and efficacious interventions, this framework could also deepen our understanding of psychiatric disorders at their neurophysiological underpinnings.

Building a precision oncology workforce by multidisciplinary and case-based learning

BACKGROUND

Participants in two recent National Academy of Medicine workshops identified a need for more multi-disciplinary professionals on teams to assist oncology clinicians in precision oncology.

METHODS

We developed a graduate school course to prepare biomedical students and pharmacy students to work within a multidisciplinary team of oncology clinicians, pathologists, radiologists, clinical pharmacists, and genetic counselors. Students learned precision oncology skills via case-based learning, hands-on data analyses, and presentations to peers. After the course, a focus group session was conducted to gain an in-depth student perspective on their interprofessional training experience, achievement of the course learning outcomes, ways to improve the course design in future offerings, and how the course could improve future career outcomes. A convenience sampling strategy was used for recruitment into the focus group session. A thematic content analysis was then conducted using the constant comparative method.

RESULTS

Major themes arising from student feedback were (1) appreciation of a customized patient case-based teaching approach, (2) more emphasis on using data analysis tools, (3) valuing interdisciplinary inclusion, and (4) request for more student discussion with advanced preparation materials.

CONCLUSIONS

Feedback was generally positive and supports the continuation and expansion of the precision oncology course to include more hands-on instruction on the use of clinical bioinformatic tools.

Machine learning to predict mortality after rehabilitation among patients with severe stroke

Stroke is among the leading causes of death and disability worldwide. Approximately 20–25% of stroke survivors present severe disability, which is associated with increased mortality risk. Prognostication is inherent in the process of clinical decision-making. Machine learning (ML) methods have gained increasing popularity in the setting of biomedical research. The aim of this study was twofold: assessing the performance of ML tree-based algorithms for predicting three-year mortality model in 1207 stroke patients with severe disability who completed rehabilitation and comparing the performance of ML algorithms to that of a standard logistic regression. The logistic regression model achieved an area under the Receiver Operating Characteristics curve (AUC) of 0.745 and was well calibrated. At the optimal risk threshold, the model had an accuracy of 75.7%, a positive predictive value (PPV) of 33.9%, and a negative predictive value (NPV) of 91.0%. The ML algorithm outperformed the logistic regression model through the implementation of synthetic minority oversampling technique and the Random Forests, achieving an AUC of 0.928 and an accuracy of 86.3%. The PPV was 84.6% and the NPV 87.5%. This study introduced a step forward in the creation of standardisable tools for predicting health outcomes in individuals affected by stroke.