1
|
Yu W, Xu N, Huang N, Chen H. Bridging the gap: Geometry-centric discriminative manifold distribution alignment for enhanced classification in colorectal cancer imaging. Comput Biol Med 2024; 170:107998. [PMID: 38266468 DOI: 10.1016/j.compbiomed.2024.107998] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/18/2023] [Revised: 12/19/2023] [Accepted: 01/13/2024] [Indexed: 01/26/2024]
Abstract
The early detection of colorectal cancer (CRC) through medical image analysis is a pivotal concern in healthcare, with the potential to significantly reduce mortality rates. Current Domain Adaptation (DA) methods strive to mitigate the discrepancies between different imaging modalities that are critical in identifying CRC, yet they often fall short in addressing the complexity of cancer's presentation within these images. These conventional techniques typically overlook the intricate geometrical structures and the local variations within the data, leading to suboptimal diagnostic performance. This study introduces an innovative application of the Discriminative Manifold Distribution Alignment (DMDA) method, which is specifically engineered to enhance the medical image diagnosis of colorectal cancer. DMDA transcends traditional DA approaches by focusing on both local and global distribution alignments and by intricately learning the intrinsic geometrical characteristics present in manifold space. This is achieved without depending on the potentially misleading pseudo-labels, a common pitfall in existing methodologies. Our implementation of DMDA on three distinct datasets, involving several unique DA tasks, has consistently demonstrated superior classification accuracy and computational efficiency. The method adeptly captures the complex morphological and textural nuances of CRC lesions, leading to a significant leap in domain adaptation technology. DMDA's ability to reconcile global and local distributional disparities, coupled with its manifold-based geometrical structure learning, signals a paradigm shift in medical imaging analysis. The results obtained are not only promising in terms of advancing domain adaptation theory but also in their practical implications, offering the prospect of substantially improved diagnostic accuracy and faster clinical workflows. This heralds a transformative approach in personalized oncology care, aligning with the pressing need for early and accurate CRC detection.
Collapse
Affiliation(s)
- Weiwei Yu
- Department of Gastroenterology, Wenzhou TCM Hospital of Zhejiang Chinese Medical University, Wenzhou, Zhejiang, 325000, China.
| | - Nuo Xu
- Department of Medical Oncology, Wenzhou TCM Hospital of Zhejiang Chinese Medical University, Wenzhou, Zhejiang, 325000, China.
| | - Nuanhui Huang
- Department of Medical Oncology, Wenzhou TCM Hospital of Zhejiang Chinese Medical University, Wenzhou, Zhejiang, 325000, China.
| | - Houliang Chen
- Department of Medical Oncology, Wenzhou TCM Hospital of Zhejiang Chinese Medical University, Wenzhou, Zhejiang, 325000, China.
| |
Collapse
|
2
|
R I A, Vatsyayan A, Damodaran D, Sivadas A, Van der Speeten K. Multi-omics Analysis Classifies Colorectal Cancer into Distinct Methylated Immunogenic and Angiogenic Subtypes Based on Anatomical Laterality. Indian J Surg Oncol 2023; 14:209-219. [PMID: 37359923 PMCID: PMC10284779 DOI: 10.1007/s13193-023-01760-6] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/19/2022] [Accepted: 04/25/2023] [Indexed: 06/28/2023] Open
Abstract
We employed supervised machine learning algorithms to a cohort of colorectal cancer patients from the NCI to differentiate and classify the heterogenous disease based on anatomical laterality and multi-omics stratification, in a first of its kind. Multi-omics integrative analysis shows distinct clustering of left and right colorectal cancer with disentangled representation of methylome and delineation of transcriptome and genome. We present novel multi-omics findings consistent with augmented hypermethylation of genes in right CRC, epigenomic biomarkers on the right in conjunction with immune-mediated pathway signatures, and lymphocytic invasion which unlocks unique therapeutic avenues. Contrarily, left CRC multi-omics signature is found to be marked by angiogenesis, cadherins, and epithelial-mesenchymal transition (EMT). An integrated multi-omics molecular signature of RNF217-AS1, hsa-miR-10b, and panel of FBX02, FBX06, FBX044, MAD2L2, and MIIP copy number altered genes have been found by the study. Overall survival analysis reveals genomic biomarkers ABCA13 and TTN in 852 LCRC cases, and SOX11 in 170 RCRC cases that predicts a significant survival benefit. Our study exemplifies the translational competence and robustness of machine learning in effective translational bridging of research and clinic. Supplementary Information The online version contains supplementary material available at 10.1007/s13193-023-01760-6.
Collapse
Affiliation(s)
- Anu R I
- Department of Cancer Biology and Therapeutics, MVR Cancer Center and Research Institute, Calicut, Kerala India
- Department of Clinical Biochemistry, MVR Cancer Center and Research Institute, Calicut, Kerala India
| | - Aastha Vatsyayan
- CSIR Institute of Genomics and Integrative Biology (CSIR-IGIB), New Delhi, India
- Academy of Scientific and Innovative Research (AcSIR), Ghaziabad, India
| | - Dileep Damodaran
- Department of Surgical Oncology, MVR Cancer Center and Research Institute, Calicut, Kerala India
| | - Ambily Sivadas
- Division of Nutrition, St. John’s Research Institute, Bangalore, India
| | - Kurt Van der Speeten
- Department of Surgical Oncology, Ziekenhuis Oost-Limburg, Genk, Belgium
- Faculty of Medicine and Life Sciences, BIOMED Research Institute, University Hasselt, Hasselt, Belgium
| |
Collapse
|
3
|
Hormuth DA, Farhat M, Christenson C, Curl B, Chad Quarles C, Chung C, Yankeelov TE. Opportunities for improving brain cancer treatment outcomes through imaging-based mathematical modeling of the delivery of radiotherapy and immunotherapy. Adv Drug Deliv Rev 2022; 187:114367. [PMID: 35654212 DOI: 10.1016/j.addr.2022.114367] [Citation(s) in RCA: 11] [Impact Index Per Article: 5.5] [Reference Citation Analysis] [What about the content of this article? (0)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/10/2022] [Revised: 04/25/2022] [Accepted: 05/25/2022] [Indexed: 11/01/2022]
Abstract
Immunotherapy has become a fourth pillar in the treatment of brain tumors and, when combined with radiation therapy, may improve patient outcomes and reduce the neurotoxicity. As with other combination therapies, the identification of a treatment schedule that maximizes the synergistic effect of radiation- and immune-therapy is a fundamental challenge. Mechanism-based mathematical modeling is one promising approach to systematically investigate therapeutic combinations to maximize positive outcomes within a rigorous framework. However, successful clinical translation of model-generated combinations of treatment requires patient-specific data to allow the models to be meaningfully initialized and parameterized. Quantitative imaging techniques have emerged as a promising source of high quality, spatially and temporally resolved data for the development and validation of mathematical models. In this review, we will present approaches to personalize mechanism-based modeling frameworks with patient data, and then discuss how these techniques could be leveraged to improve brain cancer outcomes through patient-specific modeling and optimization of treatment strategies.
Collapse
|
4
|
Terterov IN, Chubenko VA, Knyazev NA, Klimenko VV, Bogdanov AA, Moiseyenko VM, Bogdanov AA. Minimal PK/PD model for simultaneous description of the maximal tolerated dose and metronomic treatment outcomes in mouse tumor models. Cancer Chemother Pharmacol 2021; 88:867-878. [PMID: 34351468 DOI: 10.1007/s00280-021-04326-x] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/04/2021] [Accepted: 07/06/2021] [Indexed: 01/08/2023]
Abstract
PURPOSE Metronomic chemotherapy (MC) is a promising approach where, in contrast to the conventional maximal tolerated dose (MTD) strategy, regular fractionated doses of the drug are used. This approach has proven its efficacy, although drug dosing and scheduling are often chosen empirically. Pharmacokinetic/pharmacodynamic (PK/PD) models provide a way to choose optimal protocols with computational methods. Existing models are usually too complicated and are valid for only a subset of drug schedules. To address this issue, we propose herein a simple model that can describe MC and MTD regimens simultaneously. METHODS The minimal model comprises tumor suppression due to antiangiogenic drug effect together with a cell-kill term, responsible for its cytotoxicity. The model was tested on data obtained on tumor-bearing mice treated with gemcitabine in ether MTD, MC, or combined (MTD + MC) regimens. RESULTS We conducted a number of tests in which data were divided in various ways into training and validation sets. The model successfully described different trends in the MTD and MC regimens. With parameters obtained by fitting the model to MTD data, the simulations correctly predicted trends in both the MC and combined therapy groups. CONCLUSION Our results demonstrate that the proposed model presents a minimal yet efficient tool for modeling outcomes in different treatment regimens in mice. We hope that this model has the potential for use in clinical practice in the development of patient-specific chemotherapy scheduling protocols based on observed treatment response.
Collapse
Affiliation(s)
- Ivan N Terterov
- Saint-Petersburg Clinical Scientific and Practical Center of Specialized Types of Medical Care (Oncological), St. Petersburg, Russia.
| | - Vyacheslav A Chubenko
- Saint-Petersburg Clinical Scientific and Practical Center of Specialized Types of Medical Care (Oncological), St. Petersburg, Russia
| | - Nikolay A Knyazev
- Saint-Petersburg Clinical Scientific and Practical Center of Specialized Types of Medical Care (Oncological), St. Petersburg, Russia
| | - Vladimir V Klimenko
- Saint-Petersburg Clinical Scientific and Practical Center of Specialized Types of Medical Care (Oncological), St. Petersburg, Russia
| | - Andrei A Bogdanov
- Saint-Petersburg Clinical Scientific and Practical Center of Specialized Types of Medical Care (Oncological), St. Petersburg, Russia
| | - Vladimir M Moiseyenko
- Saint-Petersburg Clinical Scientific and Practical Center of Specialized Types of Medical Care (Oncological), St. Petersburg, Russia
| | - Alexey A Bogdanov
- Saint-Petersburg Clinical Scientific and Practical Center of Specialized Types of Medical Care (Oncological), St. Petersburg, Russia
| |
Collapse
|
5
|
Hormuth DA, Jarrett AM, Lorenzo G, Lima EA, Wu C, Chung C, Patt D, Yankeelov TE. Math, magnets, and medicine: enabling personalized oncology. Expert Rev Precis Med Drug Dev 2021; 6:79-81. [PMID: 34027102 PMCID: PMC8133535 DOI: 10.1080/23808993.2021.1878023] [Citation(s) in RCA: 9] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/15/2020] [Accepted: 01/15/2021] [Indexed: 02/02/2023]
Affiliation(s)
- David A. Hormuth
- Oden Institute for Computational Engineering and Sciences, The University of Texas at Austin, United States of America
- Livestrong Cancer Institutes, The University of Texas at Austin, United States of America
| | - Angela M. Jarrett
- Oden Institute for Computational Engineering and Sciences, The University of Texas at Austin, United States of America
- Livestrong Cancer Institutes, The University of Texas at Austin, United States of America
| | - Guillermo Lorenzo
- Oden Institute for Computational Engineering and Sciences, The University of Texas at Austin, United States of America
- Department of Civil Engineering and Architecture, University of Pavia, Via Ferrata 3, 27100 Pavia, Italy
| | - Ernesto A.B.F. Lima
- Oden Institute for Computational Engineering and Sciences, The University of Texas at Austin, United States of America
- Texas Advanced Computing Center, The University of Texas at Austin, United States of America
| | - Chengyue Wu
- Oden Institute for Computational Engineering and Sciences, The University of Texas at Austin, United States of America
| | - Caroline Chung
- Department of Radiation Oncology, The University of Texas at Austin, United States of America
| | - Debra Patt
- Texas Oncology, Austin, Texas, United States of America
| | - Thomas E. Yankeelov
- Oden Institute for Computational Engineering and Sciences, The University of Texas at Austin, United States of America
- Livestrong Cancer Institutes, The University of Texas at Austin, United States of America
- Department of Radiation Oncology, The University of Texas at Austin, United States of America
- Departments of Biomedical Engineering, The University of Texas at Austin, United States of America
- Department of Diagnostic Medicine, The University of Texas at Austin, United States of America
- Department of Oncology, The University of Texas at Austin, Austin, Texas, United States of America
- Departments of Imaging Physics, The University of Texas MD Anderson Cancer Center, Houston, Texas, United States of America
| |
Collapse
|
6
|
Kim H, Lee AJ, Lee J, Chun H, Ju YS, Hong D. FIREVAT: finding reliable variants without artifacts in human cancer samples using etiologically relevant mutational signatures. Genome Med 2019; 11:81. [PMID: 31847917 PMCID: PMC6916105 DOI: 10.1186/s13073-019-0695-x] [Citation(s) in RCA: 7] [Impact Index Per Article: 1.4] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/25/2019] [Accepted: 11/18/2019] [Indexed: 12/22/2022] Open
Abstract
Background Accurate identification of real somatic variants is a primary part of cancer genome studies and precision oncology. However, artifacts introduced in various steps of sequencing obfuscate confidence in variant calling. Current computational approaches to variant filtering involve intensive interrogation of Binary Alignment Map (BAM) files and require massive computing power, data storage, and manual labor. Recently, mutational signatures associated with sequencing artifacts have been extracted by the Pan-cancer Analysis of Whole Genomes (PCAWG) study. These spectrums can be used to evaluate refinement quality of a given set of somatic mutations. Results Here we introduce a novel variant refinement software, FIREVAT (FInding REliable Variants without ArTifacts), which uses known spectrums of sequencing artifacts extracted from one of the largest publicly available catalogs of human tumor samples. FIREVAT performs a quick and efficient variant refinement that accurately removes artifacts and greatly improves the precision and specificity of somatic calls. We validated FIREVAT refinement performance using orthogonal sequencing datasets totaling 384 tumor samples with respect to ground truth. Our novel method achieved the highest level of performance compared to existing filtering approaches. Application of FIREVAT on additional 308 The Cancer Genome Atlas (TCGA) samples demonstrated that FIREVAT refinement leads to identification of more biologically and clinically relevant mutational signatures as well as enrichment of sequence contexts associated with experimental errors. FIREVAT only requires a Variant Call Format file (VCF) and generates a comprehensive report of the variant refinement processes and outcomes for the user. Conclusions In summary, FIREVAT facilitates a novel refinement strategy using mutational signatures to distinguish artifactual point mutations called in human cancer samples. We anticipate that FIREVAT results will further contribute to precision oncology efforts that rely on accurate identification of variants, especially in the context of analyzing mutational signatures that bear prognostic and therapeutic significance. FIREVAT is freely available at https://github.com/cgab-ncc/FIREVAT
Collapse
Affiliation(s)
- Hyunbin Kim
- Bioinformatics Analysis Team, National Cancer Center, 323 Ilsan-ro, Ilsandong-gu, Goyang-si, Gyeonggi-do, 10408, Republic of Korea
| | - Andy Jinseok Lee
- Bioinformatics Analysis Team, National Cancer Center, 323 Ilsan-ro, Ilsandong-gu, Goyang-si, Gyeonggi-do, 10408, Republic of Korea
| | - Jongkeun Lee
- Bioinformatics Analysis Team, National Cancer Center, 323 Ilsan-ro, Ilsandong-gu, Goyang-si, Gyeonggi-do, 10408, Republic of Korea
| | - Hyonho Chun
- Department of Mathematics and Statistics, Boston University, Boston, MA, 02215, USA
| | - Young Seok Ju
- Graduate School of Medical Science and Engineering, Korea Advanced Institute of Science and Technology, Daejeon, 34141, Republic of Korea
| | - Dongwan Hong
- Bioinformatics Analysis Team, National Cancer Center, 323 Ilsan-ro, Ilsandong-gu, Goyang-si, Gyeonggi-do, 10408, Republic of Korea.
| |
Collapse
|
7
|
Elharrar X, Barbolosi D, Ciccolini J, Meille C, Faivre C, Lacarelle B, André N, Barlesi F. A phase Ia/Ib clinical trial of metronomic chemotherapy based on a mathematical model of oral vinorelbine in metastatic non-small cell lung cancer and malignant pleural mesothelioma: rationale and study protocol. BMC Cancer 2016; 16:278. [PMID: 27094927 DOI: 10.1186/s12885-016-2308-z] [Citation(s) in RCA: 20] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [What about the content of this article? (0)] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/14/2015] [Accepted: 04/11/2016] [Indexed: 11/17/2022] Open
Abstract
Background Metronomic oral vinorelbine is effective in metastatic NSCLC and malignant pleural mesothelioma, but all the studies published thus far were based upon a variety of empirical and possibly suboptimal schedules, with inconsistent results. Mathematical modelling showed by simulation that a new metronomic protocol could lead to a better safety and efficacy profile. Design This phase Ia/Ib trial was designed to confirm safety (phase Ia) and evaluate efficacy (phase Ib) of a new metronomic oral vinorelbine schedule. Patients with metastatic NSCLC or malignant pleural mesothelioma in whom standard treatments failed and who exhibited ECOG performance status 0–2 and adequate organ function will be eligible. Our mathematical PK-PD model suggested an alternative weekly D1, D2 and D4 schedule (named Vinorelbine Theoretical Protocol) with a respective dose of 60, 30 and 60 mg. Trial recruitment will be two-staged, as 12 patients are planned to participate in phase Ia to confirm safety and consolidate the calibration of the model parameters. Depending on the phase Ia results and after a favourable decision from a consultative committee, the extension phase (phase Ib) will be an efficacy study including 20 patients who will receive the Optimal Vinorelbine Theoretical Protocol. The primary endpoint is the tolerance (assessed by CTC v4.0) for the phase Ia and the objective response according to RECIST 1.1 for phase Ib. An ancillary study on circulating angiogenesis biomarkers will be a subproject of the trial. Discussion This ongoing trial is the first to prospectively test a mathematically optimized schedule in metronomic chemotherapy. As such, this trial can be considered as a proof-of-concept study demonstrating the feasibility to run a computational-driven protocol to ensure an optimal efficacy/toxicity balance in patients with cancer. Trial registration EudraCT N°: 2015-000138-31
Collapse
|
8
|
Argyri KD, Dionysiou DD, Misichroni FD, Stamatakos GS. Numerical simulation of vascular tumour growth under antiangiogenic treatment: addressing the paradigm of single-agent bevacizumab therapy with the use of experimental data. Biol Direct 2016; 11:12. [PMID: 27005569 PMCID: PMC4804544 DOI: 10.1186/s13062-016-0114-9] [Citation(s) in RCA: 11] [Impact Index Per Article: 1.4] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/13/2016] [Accepted: 03/14/2016] [Indexed: 11/30/2022] Open
Abstract
Background Antiangiogenic agents have been recently added to the oncological armamentarium with bevacizumab probably being the most popular representative in current clinical practice. The elucidation of the mode of action of these agents is a prerequisite for personalized prediction of antiangiogenic treatment response and selection of patients who may benefit from this kind of therapy. To this end, having used as a basis a preexisting continuous vascular tumour growth model which addresses the targeted nature of antiangiogenic treatment, we present a paper characterized by the following three features. First, the integration of a two-compartmental bevacizumab specific pharmacokinetic module into the core of the aforementioned preexisting model. Second, its mathematical modification in order to reproduce the asymptotic behaviour of tumour volume in the theoretical case of a total destruction of tumour neovasculature. Third, the exploitation of a range of published animal datasets pertaining to antitumour efficacy of bevacizumab on various tumour types (breast, lung, head and neck, colon). Results Results for both the unperturbed growth and the treatment module reveal qualitative similarities with experimental observations establishing the biologically acceptable behaviour of the model. The dynamics of the untreated tumour has been studied via a parameter analysis, revealing the role of each relevant input parameter to tumour evolution. The combined effect of endogenous proangiogenic and antiangiogenic factors on the angiogenic potential of a tumour is also studied, in order to capture the dynamics of molecular competition between the two key-players of tumoural angiogenesis. The adopted methodology also allows accounting for the newly recognized direct antitumour effect of the specific agent. Conclusions Interesting observations have been made, suggesting a potential size-dependent tumour response to different treatment modalities and determining the relative timing of cytotoxic versus antiangiogenic agents administration. Insight into the comparative effectiveness of different antiangiogenic treatment strategies is revealed. The results of a series of in vivo experiments in mice bearing diverse types of tumours (breast, lung, head and neck, colon) and treated with bevacizumab are successfully reproduced, supporting thus the validity of the underlying model. Reviewers This article was reviewed by L. Hanin, T. Radivoyevitch and L. Edler.
Collapse
Affiliation(s)
- Katerina D Argyri
- In Silico Oncology and In Silico Medicine Group, Laboratory of Microwaves and Fiber Optics, Institute of Communication and Computer Systems, School of Electrical and Computer Engineering, National Technical University of Athens, 9 Iroon Polytechniou, Zografos, GR 157 80, Athens, Greece
| | - Dimitra D Dionysiou
- In Silico Oncology and In Silico Medicine Group, Laboratory of Microwaves and Fiber Optics, Institute of Communication and Computer Systems, School of Electrical and Computer Engineering, National Technical University of Athens, 9 Iroon Polytechniou, Zografos, GR 157 80, Athens, Greece
| | - Fay D Misichroni
- In Silico Oncology and In Silico Medicine Group, Laboratory of Microwaves and Fiber Optics, Institute of Communication and Computer Systems, School of Electrical and Computer Engineering, National Technical University of Athens, 9 Iroon Polytechniou, Zografos, GR 157 80, Athens, Greece
| | - Georgios S Stamatakos
- In Silico Oncology and In Silico Medicine Group, Laboratory of Microwaves and Fiber Optics, Institute of Communication and Computer Systems, School of Electrical and Computer Engineering, National Technical University of Athens, 9 Iroon Polytechniou, Zografos, GR 157 80, Athens, Greece.
| |
Collapse
|