Project description:Purpose: Despite advances in radical surgery and chemotherapy delivery, ovarian cancer is the most lethal gynecologic malignancy. Most of these patients are treated with platinum-based chemotherapies, but there is no biomarker model to guide their responses to these therapeutic agents. We have developed and independently tested our novel multivariate molecular predictors for forecasting patients' responses to individual drugs on a cohort of 58 ovarian cancer patients. Experimental Design: We adapted and applied the previously-published COXEN algorithm to develop molecular predictors for therapeutic responses of patients' tumors based on expression signatures derived from the NCI-60 in vitro drug activities and genomic expression data. Genome-wide candidate biomarkers were first triaged by examining expression patterns of frozen and formalin-fixed paraffin embedded (FFPE) tissue samples. We then identify initial drug sensitivity biomarkers for carboplatin and paclitaxel, respectively. These biomarkers were further narrowed by examining concordant expression patterns between cell lines and a historical set of ovarian cancer patients. Multivariate predictors were obtained from the NCI-60 cell lines and refined using historical patient cohorts. To independent validate these molecular predictors, we performed genome-wide profiling on FFPE samples of 58 ovarian cancer patients obtained prior to adjuvant chemotherapy. Results: Carboplatin predictor significantly stratified platinum sensitive and resistant patients (p = 0.019) with sensitivity = 93%, specificity = 33%, PPV = 65%, and NPV = 78%. Paclitaxel predictor also significantly stratified patients' responses (p = 0.033) with sensitivity = 96%, specificity = 26%, PPV = 61%, and NPV = 86%. The combination predictor for platinum-taxane combination demonstrated a significant survival difference between the predicted responders and nonresponders with median survival of 12.9 months vs. 8.1 months (p = 0.045). Conclusions: COXEN predictors successfully stratified platinum resistance and taxane response in this retrospective cohort, especially based on their FFPE tumor samples. Accurate prediction of chemotherapeutic response, especially to platinum agents is highly clinically relevant and could alter primary management of ovarian cancer. Gene expression data from 58 stage III-IV ovarian cancer patients treated with Carboplatin and Taxol agents

Project description:Ovarian cancers are still largely treated with platinum-based chemotherapy as the standard of care, yet few biomarkers of clinical response have had an impact on clinical decision making as of yet. Two particular challenges faced in mechanistically deciphering platinum responsiveness in ovarian cancer have been the suitability of cell line models for ovarian cancer subtypes and the availability of information on comparatively how sensitive ovarian cancer cell lines are to platinum. We performed one of the most comprehensive profiles to date on 36 ovarian cancer cell lines across over seven subtypes and integrated drug response and multiomic data to improve on our understanding of the best cell line models for platinum responsiveness in ovarian cancer. RNAseq analysis of the 36 cell lines in a single batch experiment largely conform with the currently accepted subtyping of ovarian cancers, further supporting other studies that demonstrate that commonly used cell lines are poor models of high-grade serous ovarian carcinoma. We performed drug dose response assays in the 32 of these cell lines for cisplatin and carboplatin, as previous studies have not comprehensively reported this quantity of cell lines and/or the actual IC50s for these drugs. Our results demonstrate that cell lines largely fall either well above or below the equivalent dose of the clinical maximally achievable dose (Cmax) of each compound, allowing designation of cell lines as sensitive or resistant. We performed differential expression analysis for high-grade serous ovarian carcinoma cell lines to identify gene expression correlating with platinum-response. Further, we generated two platinum-resistant derivatives each for OVCAR3 and OVCAR4, as well as leveraged clinically-resistant PEO1/PEO4/PEO6 and PEA1/PEA2 isogenic models to perform differential expression analysis for seven total isogenic pairs of platinum resistant cell lines. While gene expression changes overall were heterogeneous and vast, themes of innate immunity/STAT activation, epithelial to mesenchymal transition and stemness, and platinum influx/efflux regulators. In addition to gene expression analyses, we performed copy number signature analysis and orthogonal measures of HRD scar scores and copy number burden, which is the first report to our knowledge applying field-standard copy number signatures to ovarian cancer cell lines, which revealed unique features of some ovarian cancer models. We also examined markers and functional readouts of stemness that revealed that cell lines are poor models for examination of stemness feature contributions to platinum resistance, as this is likely a transient state. Overall this study serves as a resource to determine the best cell lines to utilize for ovarian cancer research on certain subtypes and platinum response studies, as well as sparks new hypotheses for future study in ovarian cancer.

Project description:Platinum drugs (Pt drugs) are one of the most widely used chemotherapy drugs in cancer treatment. Although the cytotoxic and resistant mechanisms of Pt drugs have been thoroughly explored, it remains elusive what factors affect the receptiveness of DNA to Pt drug-induced damage. A large fraction of the genome shows significantly altered Pt drug-induced DNA damage susceptibility in the in vivo nuclear environment in comparison to isolated DNA, which cannot be explained only by the transcriptional status and DNA accessibility of the chromatin regions in vivo. Here, we demonstrate that, besides local chromatin structure, the global nuclear locations of genomic regions play a key role in modulating Pt drug-induced DNA damage susceptibility in vivo. By integrating data from damage-seq experiments with 3D genome structure information we show that the preferential nuclear locations of genomic regions relative to specific nuclear bodies and nuclear compartments can explain patterns of Pt drug DNA damage susceptibility. Thus, our observations are consistent with recent in vitro observations of an enrichment of Pt drugs in biomolecular condensates linked to certain nuclear bodies. Finally, when mapping observed DNA damage-seq signals onto 3D genome structures, we found that the 3D nuclear distribution of Pt-drug induced DNA damage differs in drug resistant cells in comparison to drug sensitve cells. In particular DNA damage increases in gene poor chromatin preferentially located in the lamina compartment, while DNA damage is decreased in gene rich regions preferentially located at nuclear speckles. This change may deter Pt-drug induced DNA damage from more viable gene dense chromatin regions that are crucial for short term cell survival. This observation suggests a selective spatial redistribution of Pt drug action in the nucleus during the emergence of chemoresistance. These observations are relevant for a better understanding of Pt drug action and the development of cancer resistant cells.

Dataset Information

Genetic Susceptibility and Biomarkers of Platinum-Related Toxicities

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets