Scientific Research
Oncobox Applications in Clinical Oncology
The case outlines a female patient with a recurring granulosa cell tumor in her right ovary, unresponsive to multiple chemotherapy treatments. Following an Oncobox test, imatinib was identified as a potential experimental therapy. Administration of imatinib resulted in a partial treatment response according to RECIST criteria, enhancing the patient's quality of life and leading to sustained disease stabilization.
Poddubskaya EV, Baranova MP, Allina DO, Sekacheva MI, Makovskaia LA, Kamashev DE, Suntsova MV, Barbara VS, Kochergina-Nikitskaya IN, Aleshin AA
Authors:
Cold Spring Harbor Molecular Case Studies
Journal:
Personalized prescription of imatinib in recurrent granulosa cell tumor of the ovary: case report.
The case details a patient with advancing metastatic unresectable cholangiocarcinoma. Following the results of the Oncobox test, the patient received experimental therapy, which entailed the sequential use of sorafenib and pazopanib. This approach halted disease progression and achieved long-term disease stabilization, leading to an enhanced quality of life for the patient and the elimination of pain syndrome.
Poddubskaya EV, Baranova MP, Allina DO, Smirnov PY, Albert EA, Kirilchev AP, Aleshin AA, Sekacheva MI, Suntsova MV
Authors:
Experimental Hematology & Oncology
Journal:
Personalized prescription of tyrosine kinase inhibitors in unresectable metastatic cholangiocarcinoma.
Anti-cancer-targeted drugs (TDs) specifically bind to and inhibit molecular targets crucial for cancer development and progression. Currently, hundreds of TDs are certified for clinical use worldwide. While TDs generally have fewer side effects than other cancer chemotherapy methods, they often come with a higher cost. Each TD operates through unique mechanisms and proves effective for distinct patient populations. However, their effectiveness for advanced-stage tumors is relatively limited without personalized prescription. Consequently, there's a growing demand for a personalized approach to selecting candidate drugs from this group. This review highlights a new generation of tumor response biomarkers to TDs—activation of molecular pathways—and their clinical application.
A.Buzdin, M.Sorokin, A.Garazha, E.Sekacheva, E.Kim, N.Zhukov, Y.Wang, X.Li, S.Kar, C.Hartmann, A.Samii, A.Giese, N.Borisov
Authors:
Seminars in Cancer Biology
Journal:
Molecular pathway activation – new type of biomarkers for tumor morphology and personalized selection of target drugs.
Machine Learning for Cancer Genetic Data
The developers of Oncobox have introduced a novel data harmonizer named Shambhala. This tool is platform-agnostic, meaning it operates independently of the experimental platform used to gather data. It facilitates the conversion of multiple datasets into a universal format.
N.Borisov, I.Shabalina, V.Tkachev, M.Sorokin, A.Aliper, A.Anisenko, A.Pulin, I.I. Eremin, A.Garazha, A.Buzdin
Authors:
BMC Bioinformatics
Journal:
Shambhala: a platform-agnostic data harmonizer for gene expression data.
The developers at Oncobox have introduced a novel heuristic approach for data filtering using the SVM method, called the FLOating Window Projective Separator (FloWPS). This method is tailored to personalize forecasts regarding the success of tumor chemotherapy, leveraging "big" molecular data.
V.Tkachev, M.Sorokin, A.Mescheryakov, A.Simonov, A.Garazha, A.Buzdin, I.Muchnik, N.Borisov
Authors:
Frontiers in Genetics
Journal:
FLOating-Window Projective Separator (FloWPS): a data trimming tool for support vector machines (SVM) to improve robustness of the classifier
Personalized medicine involves tailoring therapy to each patient based on specific diagnostic criteria, such as gene expression profiles. However, most machine learning methods face a challenge due to a lack of prior cases, like combined data on patient gene expression and their response to a particular therapy. Meanwhile, thousands of cancer cell lines have had their gene expression profiles determined alongside quantitative characterization of their response to hundreds of drugs, assessing their ability to inhibit cancer cell growth. The innovative approach by Oncobox authors introduces a new machine learning method where drug effectiveness assessment relies on features transferred from the cell line database. This method's effectiveness has been validated for targeted therapy — tyrosine kinase inhibitors — across three datasets representing different tumor types: chronic myeloid leukemia, lung adenocarcinoma, and renal cell carcinoma.
Borisov N, Tkachev V, Suntsova M, Kovalchuk O, Zhavoronkov A, Muchnik I, Buzdin A
Authors:
Cell Cycle
Journal:
A method of gene expression data transfer from cell lines to cancer patients for machine-learning prediction of drug efficiency.
Predicting Cancer Drug Effectiveness
Drawing from molecular pathways, Oncobox authors have developed an algorithm aimed at creating a personalized ranking of targeted drugs to optimize their efficacy through thorough mutation analysis. The effectiveness of this algorithm has been confirmed for 128 targeted drugs across 3,800 whole-exome mutation profiles obtained as part of the international Cancer Genome Atlas (TCGA) project.
M.A. Zolotovskaia, M. Sorokin, A.A. Emelianova, N.Borisov, D.V.Kuzmin, A. Buzdin
Authors:
Frontiers in Pharmacology
Journal:
Pathway based analysis of mutation data is efficient for scoring target cancer drugs.
Oncobox has unveiled an atlas of RNA sequencing profiles derived from normal human tissues collected from postmortem healthy donors. This data is featured in a new database called the Oncobox Atlas of Normal Tissue Expression (ANTE). This initiative highlights the importance of normalizing gene expression data for thorough analysis in biomedical research.
M. Suntsova, N. Gaifullin, D. Allina, A. Reshetun, X. Li, L. Mendeleeva, V. Surin, A. Sergeeva, P. Spirin, V. Prasolov, A.A. Morgan, A.Garazha, M.Sorokin, A.Buzdin
Authors:
Scientific Data
Journal:
Atlas of RNA sequencing profiles for normal human tissues.
A recently published bioinformatics algorithm calculates the activation levels of molecular pathways in tumors compared to their normal tissue counterparts. By leveraging knowledge about specific drug molecular targets, the algorithm quantitatively assesses their potential to inhibit tumor growth by blocking abnormally activated regulatory pathways. Validation of the algorithm involved comparing predicted effectiveness levels for five drugs (sorafenib, bevacizumab, cetuximab, imatinib, sunitinib) across seven cancer types (clear cell renal cell carcinoma, colorectal cancer, lung adenocarcinoma, non-Hodgkin's lymphoma, thyroid cancer, and sarcoma) with clinical trial data for each cancer type and drug. The percentage of patients responding to treatment correlated positively with the number of tumors for which the calculated drug rating was highest (Pearson correlation 0.77, p = 0.023).
Artem Artemov, Alexander Aliper, Michael Korzinkin, Ksenia Lezhnina, Leslie Jellen, Nicolay Zhukov, Sergey Roumiantsev, Nurshat Gaifullin, Alex Zhavoronkov, Nicolas Borisov, Anton Buzdin
Authors:
Oncotarget
Journal:
A method for predicting target drug efficiency in cancer based on the analysis of signaling pathway activation.
Combination Therapy and New Molecular Drug Targets
The Oncobox platform was employed to identify signaling pathways associated with resistance to tyrosine kinase inhibitor drugs in ovarian carcinoma and neuroblastoma cell lines. An analysis of 378 molecular pathways was conducted, along with modeling of potential efficacy and synergistic effects of 13 different drug combinations. Experimental validation of the model confirmed that this approach outperforms random selection of drug combinations and can facilitate the more effective discovery of new synergistically acting combinations of anti-cancer targeted drugs.
M.Sorokin, R.Kholodenko, M.Suntsova, G.Malakhova, A.Garazha, R.Vasilov, E.Poddubskaya, I.S. Stilidi, P.Arhirii and A.Buzdin
Authors:
Cancers
Journal:
Oncobox bioinformatical platform for selecting potentially effective combinations of target drugs using high-throughput gene expression data.
Due to one of the most common chromosomal translocations in acute myeloid leukemia, abnormal transcripts encoding the RUNX1-ETO hybrid protein are produced. This protein is a promising target for treating leukemia with this mutation. Our research has shown that inhibiting the activity of RUNX1-ETO causes leukemia cells to develop resistance by activating intracellular signaling pathways that promote cell survival. We have also discovered that the ERK2 protein regulates the activation of most of these pathways. Combining oridonin (a natural inhibitor of RUNX1-ETO) with ERK2 inhibitors has a synergistic effect and hampers the survival of leukemia cells.
PSpirin P, Lebedev T, Orlova N, Morozov A, Poymenova N, Dmitriev SE, Buzdin A, Stocking C, Olga K, Vladimir P
Authors:
Oncotarget
Journal:
Synergistic suppression of t(8;21)-positive leukemia cell growth by combining oridonin and MAPK1/ERK2 inhibitors.
The Oncobox platform was used to simulate the combined impact of angiogenesis inhibitors and the mutant form of the BRAF protein. Experimental validation in a mouse model showed that simultaneous inhibition resulted in tumor cell apoptosis, normalization of the vascularization process, decreased hypoxia, remodeling of the extracellular matrix, infiltration of M1 macrophages, and reduced levels of cancer-associated fibroblasts.
Corà D, Orso F, Consonni FM, Middonti E, Di Nicolantonio F, Buzdin A, Sica A, Medico E, Sangiolo D, Taverna D,
Bussolino F
Authors:
EMBO Molecular Medicine
Journal:
VEGF blockade enhances the antitumor effect of BRAFV600E inhibition.
In this study, researchers from Oncobox compared experimental data collected in their own laboratory with information from the "Genomics of Drug Sensitivity in Cancer" (GDS) project to investigate the relationship between transcriptomes and response to anti-cancer medications. The analysis focused on four targeted drugs—pazopanib, sorafenib, sunitinib, and temsirolimus—across 238 cell lines. By examining transcriptomic data related to approximately 600 molecular pathways, they pinpointed pathways where the strength of activation correlated significantly with toxicity values (IC50) for the drugs under scrutiny. Notably, these pathways exhibited substantial correlation coefficients in both their own experimental investigations and the GDS project data. Furthermore, molecular interaction models with the molecular targets of the respective drugs were developed for these pathways for the first time.
Larisa Venkova, Alexander Aliper, Maria Suntsova, Roman Kholodenko, Denis Shepelin, Nicolas Borisov, Galina Malakhova, Raif Vasilov, Sergey Roumiantsev, Alex Zhavoronkov, Anton Buzdin
Authors:
Oncotarget
Journal:
Combinatorial high-throughput experimental and bioinformatic approach identifies molecular pathways linked with the sensitivity to anticancer target drugs.
Molecular Pathways and Cancer Biomarkers
Mutations are pivotal in cancer development and progression, with profiles varying widely among different cancer types and individual tumors. While mutations in specific genes can act as cancer biomarkers, their limited number restricts their use for most oncological diagnoses. We present a novel method that substantially improves the search for cancer progression biomarkers using DNA mutation data. This method employs a quantitative measure called "Pathway Instability" (PI), which assesses the enrichment of mutations within intracellular molecular pathways.

Testing the method involved analyzing 5,956 mutation profiles from tumors across 15 cancer types, sourced from The Cancer Genome Atlas (TCGA) database. In total, 2,316,670 mutations spanning 19,872 genes and 1,748 molecular pathways were examined. The results underscore the significant advantage of utilizing pathway-based biomarkers over individual mutation profiles. This improvement is evident in the enhanced quality of biomarkers as measured by the AUC metric, along with a substantial increase in the number of high-quality markers (AUC>0.75) by several orders of magnitude. These findings suggest that PI holds promise as a comprehensive next-generation cancer biomarker, offering significantly greater effectiveness than biomarkers relying solely on individual gene mutations.
Zolotovskaia, M.Sorokin, S.Roumiantsev, N.Borisov, A.Buzdin
Authors:
Frontiers in Oncology
Journal:
Pathway instability is an effective new mutation-based type of cancer biomarkers.
The iPANDA method is introduced for assessing the activation levels of molecular pathways. It has been validated to stratify patients with breast cancer according to their sensitivity to neoadjuvant therapy.
Ozerov IV, Lezhnina KV, Izumchenko E, Artemov AV, Medintsev S, Vanhaelen Q, Aliper A, Vijg J, Osipov AN, Labat I, West MD, Buzdin A, Cantor CR, Nikolsky Y, Borisov N, Irincheeva I, Khokhlovich E, Sidransky D, Camargo ML, Zhavoronkov A
Authors:
Nature Communications
Journal:
In silico Pathway Activation Network Decomposition Analysis (iPANDA) as a method for biomarker development.
The effectiveness of cetuximab therapy in colorectal cancer patients can be accurately predicted using a set of advanced molecular markers known as molecular pathway activation levels. Analysis of transcriptomic data gathered from mouse xenograft models of patient tumor cells revealed that these pathway activation levels are reliable predictors of cetuximab response. This suggests that, alongside testing for mutations in the RAS gene family, evaluating pathway activation could serve as an additional criterion for selecting cetuximab treatment for metastatic colorectal cancer patients.
Qingsong Zhu, Evgeny Izumchenko, Alexander Aliper, Evgeny Makarev, Keren Paz, Anton Buzdin, Alex Zhavoronkov, and David Sidransky
Authors:
Human Genome Variation
Journal:
Pathway Activation Strength (PAS) is a Novel Independent Prognostic Biomarker for Cetuximab Sensitivity in Colorectal Cancer Patients.
It has been shown that a set of next-generation molecular markers—molecular pathway activation levels—can be highly effective in diagnosing bladder cancer. These markers are comprised of mathematical functions that describe the expression levels of multiple gene products, setting them apart from "traditional" expression biomarkers, which only evaluate the concentrations of transcripts from individual genes.
K. Lezhnina, O. Kovalchuk, A.A. Zhavoronkov, M.B. Korzinkin, A.A. Zabolotneva, P.V. Shegay, D.G. Sokov, N.M. Gaifullin, I.G. Rusakov, A.M. Aliper, S.A., Roumiantsev, B.Y. Alekseev, N.M. Borisov, and A.A. Buzdin
Authors:
Oncotarget
Journal:
Novel robust biomarkers for human bladder cancer based on activation of intracellular signaling pathways.
This study demonstrates that the activation level of molecular pathways serves as a significantly more qualitative biomarker compared to the expression levels of individual genes. An analysis of pathway activation profiles, involving over 2700 human gene products across 82 signaling pathways, was conducted using 292 samples of cancerous tissues representing nine different cancer types and 128 samples of normal human tissues. Across all cancer types studied, the activation level values of signaling pathways exhibited higher-quality statistical criterion values—area under the ROC curve (AUC)—compared to individual genes participating in the same pathways. These findings suggest that activation level values of molecular pathways can serve as next-generation biomarkers in oncology.
N.M. Borisov, N.V. Terekhanova, A.M. Aliper, L.S. Venkova, P.Yu. Smirnov, S.Roumiantsev, M.B. Korzinkin, A.A. Zhavoronkov, A.A. Buzdin
Authors:
Oncotarget
Journal:
Signaling pathway activation profiles make better markers of cancer than expression of individual genes.
A novel bioinformatics method has been introduced, designed for both quantitative and qualitative analysis of intracellular signaling pathway activation. It identifies the functional role of each gene product within the molecular pathway. This method is versatile and applicable for analyzing various physiological changes, such as stress, aging, oncological diseases, and other pathologies.
Buzdin AA, Zhavoronkov AA, Korzinkin MB, Venkova LS, Zenin AA, Smirnov PY, Borisov NM
Authors:
Oncotarget
Journal:
OncoFinder, a new method for the analysis of intracellular signaling pathway activation using transcriptomic data.
Fundamental Cancer Mechanisms
Acquired resistance to chemotherapy and radiation therapy presents a significant obstacle, diminishing the efficacy of cancer treatment. Oncobox researchers simulated acquired resistance to five targeted anticancer drugs in two cell lines (SKOV-3 ovarian carcinoma and NGP-127 neuroblastoma). Cells were exposed to incrementally escalating concentrations of three tyrosine kinase inhibitors—sorafenib, pazopanib, and sunitinib—as well as mTOR inhibitors—everolimus and temsirolimus. Subsequently, the cells underwent exposure to a standard therapeutic radiation dose of 10 Gy. In the SKOV-3 cell line, but not in NGP-127, a statistically significant increase in the capacity of cells resistant to sorafenib, pazopanib, and sunitinib to repair double-strand DNA breaks was observed compared to control cells that did not develop resistance to tyrosine kinase inhibitors. These characteristics were linked to heightened activation of the DNA repair pathway facilitated by the ATM protein. Our findings suggest the development of a novel model for assessing the effectiveness of anticancer therapy and underscore the potential for tissue-specific development of resistance to radiation therapy, possibly arising as a consequence of tyrosine kinase inhibitor treatment.
Sorokin M, Kholodenko R, Grekhova A, Suntsova M, Pustovalova M, Vorobyeva N, Kholodenko I, Malakhova G, Garazha A, Nedoluzhko A, Vasilov R, Poddubskaya E, Kovalchuk O, Adamyan L, Prassolov V, Allina D, Kuzmin D, Ignatev K, Osipov A, Buzdin A
Authors:
Oncotarget
Journal:
Acquired resistance to tyrosine kinase inhibitors may be linked with the decreased sensitivity to X-ray irradiation.
The chromosomal rearrangement t(8;21)(q22;q22) is frequently observed in acute myeloid leukemia (AML). This genetic alteration results in the creation of a transcript that codes for the hybrid protein AML1-ETO (AE), which functions as a transcription factor. AE represents a promising therapeutic target for treating leukemia positive for t(8;21). However, the presence of AE alone is inadequate to trigger the transformation of cells into tumors. AML cells commonly display dysregulation of various genes, including the c-KIT gene, which encodes a tyrosine kinase receptor.

In this study, Oncobox researchers illustrated that AML cells with artificially suppressed AE expression exhibited a significant reduction in growth rate, attributed to the induction of apoptosis and dysregulation of the cell cycle. Moreover, a decrease in c-KIT mRNA levels was noted. Artificial inhibition of c-KIT expression attenuated the growth intensity of tumor cells but did not induce apoptosis. Examination of the transcriptional profiles of cells that evaded cell death and acquired resistance to AE suppression revealed the activation of multiple signaling pathways that promote cell survival and proliferation. Specifically, it was discovered that the ERK2 protein actively regulates 23 out of 29 (79%) of these molecular pathways, suggesting its pivotal role in the development of resistance to AE suppression in leukemia.
Spirin PV, Lebedev TD, Orlova NN, Gornostaeva AS, Prokofjeva MM, Nikitenko NA, Dmitriev SE,Buzdin AA, Borisov NM, Aliper AM, Garazha AV, Rubtsov PM, Stocking C, Prassolov VS
Authors:
Leukemia (Nature Publishing Group)
Journal:
Silencing AML1-ETO gene expression leads to simultaneous activation of both pro-apoptotic and proliferation signaling.