Targeting cancer stem cell OXPHOS with tailored ruthenium complexes as a new anti-cancer strategy

BACKGROUND

Previous studies by our group have shown that oxidative phosphorylation (OXPHOS) is the main pathway by which pancreatic cancer stem cells (CSCs) meet their energetic requirements; therefore, OXPHOS represents an Achille's heel of these highly tumorigenic cells. Unfortunately, therapies that target OXPHOS in CSCs are lacking.

METHODS

The safety and anti-CSC activity of a ruthenium complex featuring bipyridine and terpyridine ligands and one coordination labile position (Ru1) were evaluated across primary pancreatic cancer cultures and in vivo, using 8 patient-derived xenografts (PDXs). RNAseq analysis followed by mitochondria-specific molecular assays were used to determine the mechanism of action.

RESULTS

We show that Ru1 is capable of inhibiting CSC OXPHOS function in vitro, and more importantly, it presents excellent anti-cancer activity, with low toxicity, across a large panel of human pancreatic PDXs, as well as in colorectal cancer and osteosarcoma PDXs. Mechanistic studies suggest that this activity stems from Ru1 binding to the D-loop region of the mitochondrial DNA of CSCs, inhibiting OXPHOS complex-associated transcription, leading to reduced mitochondrial oxygen consumption, membrane potential, and ATP production, all of which are necessary for CSCs, which heavily depend on mitochondrial respiration.

CONCLUSIONS

Overall, the coordination complex Ru1 represents not only an exciting new anti-cancer agent, but also a molecular tool to dissect the role of OXPHOS in CSCs. Results indicating that the compound is safe, non-toxic and highly effective in vivo are extremely exciting, and have allowed us to uncover unprecedented mechanistic possibilities to fight different cancer types based on targeting CSC OXPHOS.

Clinico-Pathological Features, Outcomes and Impacts of COVID-19 Pandemic on Patients with Early-Onset Colorectal Cancer: A Single-Institution Experience

BACKGROUND

The rising incidence of colorectal cancer (CRC) among young patients is alarming. We aim to characterize the clinico-pathological features and outcomes of patients with early-onset CRC (EOCRC), as well as the impacts of COVID-19 pandemic.

METHODS

We included all patients with pathologically confirmed diagnoses of CRC at Hospital Universitario La Paz from October 2016 to December 2021. The EOCRC cut-off age was 50 years old.

RESULTS

A total of 1475 patients diagnosed with CRC were included, eighty (5.4%) of whom had EOCRC. Significant differences were found between EOCRC and later-onset patients regarding T, N stage and metastatic presentation at diagnosis; perineural invasion; tumor budding; high-grade tumors; and signet ring cell histology, with all issues having higher prevalence in the early-onset group. More EOCRC patients had the RAS/ BRAF wild type. Chemotherapy was administered more frequently to patients with EOCRC. In the metastatic setting, the EOCRC group presented a significantly longer median OS. Regarding the COVID-19 pandemic, more patients with COVID-19 were diagnosed with metastatic disease (61%) in the year after the lockdown (14 March 2020) than in the pre-pandemic EOCRC group (29%).

CONCLUSIONS

EOCRC is diagnosed at a more advanced stage and with worse survival features in localized patients. More patients with EOCRC were diagnosed with metastatic disease in the year after the COVID-19 pandemic lockdown. The long-term consequences of COVID-19 are yet to be determined.

Abstract 5416: RNA-seq and proteomics to identify response to immunotherapy in advanced melanoma: a Spanish Melanoma Group Study

Background: Prediction of response to immunotherapy remains an unmet need in the field of advanced melanoma.

Methods: Computational analyses, including probabilistic graphical models, sparse k-means, and consensus cluster, were used to characterize melanoma TCGA samples. The implication of the identified processes in response to immunotherapy was then studied in an independent cohort of 53 patients with advanced melanoma and treated with PD-1 inhibitors. Paraffin samples from this cohort were analyzed using RNA-seq and mass-spectrometry proteomics.

Results: In the TCGA cohort, there were two different layers of information: one related to molecular features of the tumor (based on keratinization, melanogenesis, and extracellular space), and one related to immune status. Therefore, two independent classifications of TCGA melanoma samples were established: molecular and immune. The immune classification distinguished between responders and not responders to immunotherapy in the second cohort (p= 0.0006, HR=6.52). Finally, high-throughput proteomics was used to characterize molecular mechanisms involved in response to immunotherapy, identifying several biological processes and proteins that may be relevant in treatment selection for patients who do not respond.

Conclusions: We established that the immune information was independent of tumor molecular features in melanomas included in the TCGA. An immune classification of these tumors was established. This immune classification predicted response to immunotherapy in a new cohort of patients with advanced melanoma treated with PD-1 inhibitors. Finally, proteomics was used to identify possible targets in those patients who did not respond to immunotherapy.

Citation Format: Lucia Trilla-Fuertes, Guillermo Prado-Vazquez, Angelo Gámez-Pozo, Rocio Lopez-Vacas, Maria Isabel Lumbreras Herrera, Virtudes Soriano, Fernando Garicano, Maria Jose Lecumberri, Maria Rodriguez, Margarita Majem, Elisabeth Perez, Maria Gonzalez-Cao, Juana Oramas, Alejandra Magdaleno, Joaquin Fra, Alfonso Martin, Monica Corral, Teresa Puertolas, Juan Angel Fresno Vara, Enrique Espinosa. RNA-seq and proteomics to identify response to immunotherapy in advanced melanoma: a Spanish Melanoma Group Study [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 5416.

Abstract P3-08-42: Disease-free survival prognostic signature in triple-negative breast cancer based on high-throughput proteomics data

Introduction Triple negative breast cancer (TNBC) accounts for 15-20% of the breast cancer and is characterized by an aggressive phenotype and worst prognosis. TNBC does not benefit from any targeted therapy, so further characterization would be needed to define subgroups with potential therapeutic value. Material and methods 125 TNBC paraffin samples were analyzed using high-throughput proteomics based on SWATH-MS. Survival analyses and a prognostic predictor were done using BRB Array Tools. Proteins related with disease-free survival were established and, then, a prognostic signature was built based on their p-values. Results and discussion Using SWATH-MS, 1,206 proteins were identified in a cohort of 125 TNBC tumors. Of these 1,206 proteins, 29 proteins were related with disease-free survival. In addition, a prognostic signature based on the expression of two proteins, RMB3 and NIPSNAP1, was defined. The predictor split our population into a low-risk and a high-risk group (p=0.0002, HR= 6.519). Multivariate analysis showed that the prognostic signature based on the expression of these two proteins supplied significant information to the clinical parameters. Conclusion SWATH-MS proteomics demonstrates its utility in the analysis of TNBC paraffin samples. Moreover, this proteomics data allows us to build a prognostic signature based on the expression of two proteins (RBM3 and NIPSNAP1). This prognostic signature could be used in the future to identify a population with a high-risk of relapse that may be directed to a clinical trial.

Citation Format: Pilar Zamora Auñon, Silvia García Adrián, Lucia Trilla-Fuertes, Angelo Gámez-Pozo, Guillermo Prado-Vázquez, Andrea Zapater-Moros, Mariana Díaz- Almirón, Rocío López Vacas, Cristina Chiva, Cristina Chiva, Eduard Sabidó, Enrique Espinosa Arranz, Juan Ángel Fresno Vara. Disease-free survival prognostic signature in triple-negative breast cancer based on high-throughput proteomics data [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P3-08-42.

Abstract P5-12-14: A pilot study of metabolomics biomarkers in breast cancer tumors treated with neoadjuvant therapy

Introduction

Breast cancer is one of the most prevalent cancers in the world. Traditionally, early breast cancer treatment is based on surgery and, after surgery, hormone treatment or chemotherapy. However, the neoadjuvant treatment is increasingly used. Metabolomics is the most recent “omics” which allows quantify metabolites into blood patient samples. Coupled with computational analyses it could be possible to study differential metabolomics patterns and associate them with neoadjuvant response.

Material and methods

Blood plasma samples from patients with breast cancer treated with neoadjuvant chemotherapy were used to perform metabolomics experiments. One sample before the treatment (basal) and one sample after the chemotherapy (post-treatment) were analyzed and clinical data regarding response (complete response or partial response) was also collected. Metabolomics experiments were performed using liquid chromatography coupled with mass-spectrometry. Bayesian network and class comparison analyses were used to establish differential metabolic patterns between conditions. Additionally, a response prediction model based on logistic regression was build using metabolomics data from basal samples.

Results and discussion

A network showing the relationships between metabolites was build. Comparing metabolite measurements between complete response and partial response tumors in basal samples, 19 metabolites showed a differential quantification between both types of responses. Moreover, one of these metabolites is linoleic acid, previously described as a biomarker of complete response in neoadjuvant treatment in breast cancer. Using these 19 differential metabolites, a response predictive model was build. According to this model, it is possible to predict response to neoadjuvant treatment based on the amount of one metabolite, still only identified by its mass and charge. On the other hand, comparing basal and post-treatment samples, the network showed differential metabolomics patterns. These differential metabolites could be used as predictive biomarkers of response.

Conclusion

This study is a proof of concept that using a new “omics” technique such as metabolomics in blood samples, coupled with computational analyses, it is possible to identify differential metabolomics patterns between complete and partial response or basal and post-treatment samples and design predictive models of response These results could facilitate in the future the implementation of blood-based tests into the clinical routine.

Citation Format: Zamora P, Trilla-Fuertes L, Zapater-Moros A, Gámez-Pozo A, Prado-Vázquez G, Ferrer-Gómez M, Díaz- Almirón M, López Vacas R, Espinosa E, Fresno Vara JA. A pilot study of metabolomics biomarkers in breast cancer tumors treated with neoadjuvant therapy [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P5-12-14.

Abstract P2-02-13: Computational metabolism modeling predicts risk of relapse in breast cancer patients

Introduction

Breast cancer is one of the most prevalent cancers in the world. In previous works we observed differences in glucose metabolism between breast cancer subtypes, suggesting that metabolism plays an important role in this disease. Flux Balance Analysis (FBA) is widely used to study metabolic networks, allowing predicting growth rates or the rate of production of a given metabolite.

Material and methods

Proteomics data from 96 breast cancer tumors were obtained applying a high-throughput proteomics approach to routinely archive formalin-fixed, paraffin-embedded tumor tissue. Proteomics tumor data were analyzed using the human metabolic reconstruction Recon2 and FBA. The tumor growth rate for each tumor was calculated. In order to analyze fluxes from the different metabolic pathways, flux activities were calculated as the sum of the fluxes of each reaction in each pathway defined in the Recon2. Then, flux activities were used to build prognostic models.

Results and discussion

Using the results obtained from FBA in the proteomics dataset, flux activities were calculated for each pathway. Employing these flux activities, a prognostic signature was built. Flux activities of vitamin A, tetrahydrobiopterin metabolism and beta-alanine metabolism pathways split our population into a low and a high risk group (p=0.044).

Conclusion

Vitamine A, beta-alanine and tetrahydrobiopterin metabolism flux activities could be used to predict relapse risk. Flux activities is a method proposed in a previous work to study response against drugs that now also demonstrated its utility in summarizing FBA data and is associated with prognosis.

Citation Format: Zamora P, Trilla-Fuertes L, Gámez-Pozo A, Prado-Vázquez G, Zapater-Moros A, Ferrer-Gómez M, Díaz- Almirón M, López Vacas R, Espinosa E, Fresno Vara JA. Computational metabolism modeling predicts risk of relapse in breast cancer patients [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P2-02-13.

Abstract P6-15-12: A functional approach to the molecular basis of neoadjuvant treatment response in breast cancer

BACKGROUND

Breast cancer is a diverse and heterogeneous disease. The use of neoadjuvant treatments has improved the prognosis of localized breast cancer. However, molecular basis of neoadjuvant treatment response and resistance remains unknown. Clinical data has uncovered the existence of different tumor responses to neoadjudvant chemotherapy, allowing the classification of patients in different groups. Gene expression profile description of the different patient groups provide essential information in the clinical decision making as well as to allow a deeper knowledge of this disease.

MATERIALS AND METHODS

A breast cancer tumor dataset was obtained from the Gene Expression Omnibus (GSE41998) and from a phase II trial (NCT00455533). 279 tumors from previously untreated women with primary invasive breast adenocarcinoma were included in this study. Whole genome gene expression profiling was performed using Affymetrix GeneChip gene expression microarrays.Differentially expressed genes were chosen selecting 3000 more variable probes among all patients and were used to construct four networks of gene functional interactions, one for all tumors and three for each molecular subtype independently. Functional structure was performed using probabilistic graphical models with local minimum Bayesian Information Criterion. Data analyses were carried out using MeV, BRB Array Tools, R, Cytoscape software suites and DAVID web tools.

RESULTS

Regardless of tumor molecular subtype, tumors showing a complete response to treatment showed higher "Immune response (MHCII)", "Immune response (chemotaxis)", "Immune response (B cell)“ and "Immune response (Interferon)” nodes activities compared to resistant tumors (stable disease tumors). These differences are also observed when analyzing tumor molecular subgroups (Luminal A, Luminal B and Basal-like) separately. Moreover, complete response tumors, showed significantly higher levels of lymphocytic cell lineage markers (CD4, CD8 and CD20).

CONCLUSION

This type of approach allows seeing differences at biological process levels rather than at the individual gene level.Tumors that respond to neoadjuvant treatment showed higher “Immune” nodes activity than resistant tumors and these differences were also showed in analyses stratified by molecular subtype. Besides, complete response tumors presented higher values of lymphocyte cell lineage markers which might suggest a greater amount of tumor-infiltrating lymphocytes (TILs). These results can suggest that patients' immune system could play an important role in the response to neoadjuvant chemotherapy treatment.

Citation Format: Zamora Auñón P, Zapater-Moros A, Trilla-Fuertes L, Gamez-Pozo A, Prado-Vázquez G, Llorente-Armijo S, Lopez-Vacas R, Main P, Espinosa Arranz E, Fresno-Vara JA. A functional approach to the molecular basis of neoadjuvant treatment response in breast cancer [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P6-15-12.

Abstract P6-07-07: Triple negative breast cancer classification according to cancer stem cell hypothesis

Background:

Triple negative breast cancer (TNBC) is described by lack of ER and PR expression and HER2 overexpression. This subgroup has no targeted therapies and its prognosis its worse than other breast cancer subtypes. The Cancer Stem Cell hypothesis lies in two ideas; first, that breast cancers can initiate in different cell types, which can be any epithelial stem cells or any of their progeny, and, that this original cell will give the tumor a specific molecular profile. Our work proposes a division in molecular groups of TNBC looking for this tissue of origin molecular profile through gene expression data and probabilistic graphical models analyses.

Material and methods:

TNBC gene expression data was obtained from GSE31519 (n=494). 2000 most variable genes were selected for subsequent analysis. A functional network was built using a probabilistic graphical model approach. Functional nodes were defined, and its function was explored by Gene Ontology using DAVID. Then, a new molecular classification was generated using the activity of functional nodes and k-means. Subgroups were characterized and compared with previous TNBC molecular classifications.

Results:

Probabilistic graphical models defined a functional structure comprising 27 functional nodes. We found some Luminal, some Basal and a claudin-enriched node. Based on these nodes molecular subgroups were defined, following the cancer stem cell hypothesis. Thus, four subtypes: (Luminal Androgen receptor (LAR), basal, claudin-low and claudin-high were defined matching tumor origin characteristics with those in the actual tumor sample. Tumors with low expression of claudins (CLDN-low subtype) had been differentiated in the first steps of the mammary epithelial development. Tumors with high expression of claudins (CLDN-High) had been originated at the second step of the development. Next step in the development are basal-epithelial cells, which will generate Basal subtype of TNBC. And finally, the last step in development is the differentiation to luminal cell, which is the origin of the Luminal subtype. Immune status, determined by immune functional nodes, showed prognostic value (p>0.05). Finally, we compared our classification with previous ones defined by Lehmann, Burstein and the PAM50.

Table 1Cellular ClassificacionTumor sizeGradeNodal T1RestG3G1 or G2N0N1Basal761631994516835CLDN-High227315194CLDN-Low103221202811LAR115429333618Total9927628010325168

Conclusion:

Functional networks can provide a relevant molecular knowledge which complements the TNBC classification. From this approach we establish a new classification taking into account the cancer stem cell hypothesis. Besides, this deep knowledge will allow a more accurate prediction of outcome and can also be used for diagnostic purposes and therapy selection.

Citation Format: Zamora-Auñón P, Trilla-Fuertes L, Diaz-Almiron M, Gamez-Pozo A, Prado-Vázquez G, Zapater-Moros A, Llorente-Armijo S, Gaya Romero F, Espinosa Arranz E, Fresno-Vara JA. Triple negative breast cancer classification according to cancer stem cell hypothesis [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P6-07-07.

Abstract P1-02-06: Computational modeling predicts drugs response to targeting metabolism in breast cancer cells

Background

Reprogramming of metabolism is a hallmark in cancer. In previous works we observed differences in glucose metabolism between tumors from different breast cancer subtypes, suggesting the possibility to use drugs against metabolism in this disease. Flux Balance Analysis (FBA) is widely used to study biochemical networks, allowing to predict growth rates and to simulate drug response.

Material and methods

Breast cancer cell lines and different drugs against metabolic targets were evaluated with dose-response curves, and pharmacological parameters for each condition were calculated. Proteomics data from breast cancer cells lines treated with sub-lethal doses and controls were obtained applying a mass spectrometry-based approach. Differences in protein expression between treated vs. control were assessed. An FBA approach using the human metabolic reconstruction Recon2 and including the protein expression values from perturbation experiments was also applied. Model predictions were validated using dynamic FBA and growth rate for each sample was estimated. With the aim to compare the activity of the different pathway fluxes between control and treated cells, flux activity was calculated for each condition and for each pathway and response predictive models were performed.

Results

Drug response was diverse across different breast cancer cells. Mass spectrometry from cell samples allows identifying and quantifying 4,114 proteins. FBA predicted that growth rates decrease in treated cells vs. control, as observed in cell viability assays. Dynamic FBA showed that our model correctly reflects cell growth rates. Finally, using flux activities, it is possible to build models which could predict response against these drugs.

Conclusions

Proteomics provide insights of the mechanisms responsible of cells' response to metabolism drugs. A validated computational model able to predict tumor growth using data from proteomics was developed. Model predicts growth rates and also dysregulation of biological processes triggered by drug treatment. Moreover, these computational approaches could be used to propose new mechanisms of action and effects of metabolic drugs.

Acknowledgments

This work was supported by grant PI15/01310 from Instituto de Salud Carlos III, Spanish Economy and Competitiveness Ministry, Spain and co-funded by FEDER program, “Una forma de hacer Europa”. L.T.-F is supported by Spanish Economy and Competitiveness Ministry grant DI-15-07614.

Competing interest

JAFV, AG-P and EE are stakeholders of Biomedica Molecular Medicine S.L. and Biomedica Molecular Medicine Ltd. LT-F is an employee of Biomedica Molecular Medicine S.L. The authors have declared no other conflict of interest.

Citation Format: Zamora Auñón P, Trilla-Fuertes L, Díaz-Almirón M, Gamez-Pozo A, Prado-Vázquez G, Zapater-Moros A, Llorente-Armijo S, Gaya Romero F, Espinosa Arranz E, Fresno-Vara JA. Computational modeling predicts drugs response to targeting metabolism in breast cancer cells [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P1-02-06.

Urothelial cancer proteomics provides both prognostic and functional information

Traditionally, bladder cancer has been classified based on histology features. Recently, some works have proposed a molecular classification of invasive bladder tumors. To determine whether proteomics can define molecular subtypes of  muscle invasive urothelial cancer (MIUC) and allow evaluating the status of biological processes and its clinical value. 58 MIUC patients who underwent curative surgical resection at our institution between 2006 and 2012 were included. Proteome was evaluated by high-throughput proteomics in routinely archive FFPE tumor tissue. New molecular subgroups were defined. Functional structure and individual proteins prognostic value were evaluated and correlated with clinicopathologic parameters. 1,453 proteins were quantified, leading to two MIUC molecular subgroups. A protein-based functional structure was defined, including several nodes with specific biological activity. The functional structure showed differences between subtypes in metabolism, focal adhesion, RNA and splicing nodes. Focal adhesion node has prognostic value in the whole population. A 6-protein prognostic signature, associated with higher risk of relapse (5 year DFS 70% versus 20%) was defined. Additionally, we identified two MIUC subtypes groups. Prognostic information provided by pathologic characteristics is not enough to understand MIUC behavior. Proteomics analysis may enhance our understanding of prognostic and classification. These findings can lead to improving diagnosis and treatment selection in these patients.