Javier F. Torres-Roca

A Gene Expression Model of Intrinsic Tumor Radiosensitivity: Prediction of Response and Prognosis After Chemoradiation

PURPOSE Development of a radiosensitivity predictive assay is a central goal of radiation oncology. We reasoned a gene expression model could be developed to predict intrinsic radiosensitivity and treatment response in patients. METHODS AND MATERIALS Radiosensitivity (determined by survival fraction at 2 Gy) was modeled as a function of gene expression, tissue of origin, ras status (mut/wt), and p53 status (mut/wt) in 48 human cancer cell lines. Ten genes were identified and used to build a rank-based linear regression algorithm to predict an intrinsic radiosensitivity index (RSI, high index = radioresistance). This model was applied to three independent cohorts treated with concurrent chemoradiation: head-and-neck cancer (HNC, n = 92); rectal cancer (n = 14); and esophageal cancer (n = 12). RESULTS Predicted RSI was significantly different in responders (R) vs. nonresponders (NR) in the rectal (RSI R vs. NR 0.32 vs. 0.46, p = 0.03), esophageal (RSI R vs. NR 0.37 vs. 0.50, p = 0.05) and combined rectal/esophageal (RSI R vs. NR 0.34 vs. 0.48, p = 0.001511) cohorts. Using a threshold RSI of 0.46, the model has a sensitivity of 80%, specificity of 82%, and positive predictive value of 86%. Finally, we evaluated the model as a prognostic marker in HNC. There was an improved 2-year locoregional control (LRC) in the predicted radiosensitive group (2-year LRC 86% vs. 61%, p = 0.05). CONCLUSIONS We validate a robust multigene expression model of intrinsic tumor radiosensitivity in three independent cohorts totaling 118 patients. To our knowledge, this is the first time that a systems biology-based radiosensitivity model is validated in multiple independent clinical datasets.

Prediction of Radiation Sensitivity Using a Gene Expression Classifier

The development of a successful radiation sensitivity predictive assay has been a major goal of radiation biology for several decades. We have developed a radiation classifier that predicts the inherent radiosensitivity of tumor cell lines as measured by survival fraction at 2 Gy (SF2), based on gene expression profiles obtained from the literature. Our classifier correctly predicts the SF2 value in 22 of 35 cell lines from the National Cancer Institute panel of 60, a result significantly different from chance (P = 0.0002). In our approach, we treat radiation sensitivity as a continuous variable, significance analysis of microarrays is used for gene selection, and a multivariate linear regression model is used for radiosensitivity prediction. The gene selection step identified three novel genes (RbAp48, RGS19, and R5PIA) of which expression values are correlated with radiation sensitivity. Gene expression was confirmed by quantitative real-time PCR. To biologically validate our classifier, we transfected RbAp48 into three cancer cell lines (HS-578T, MALME-3M, and MDA-MB-231). RbAp48 overexpression induced radiosensitization (1.5- to 2-fold) when compared with mock-transfected cell lines. Furthermore, we show that HS-578T-RbAp48 overexpressors have a higher proportion of cells in G2-M (27% versus 5%), the radiosensitive phase of the cell cycle. Finally, RbAp48 overexpression is correlated with dephosphorylation of Akt, suggesting that RbAp48 may be exerting its effect by antagonizing the Ras pathway. The implications of our findings are significant. We establish that radiation sensitivity can be predicted based on gene expression profiles and we introduce a genomic approach to the identification of novel molecular markers of radiation sensitivity.

Systems Biology Modeling of the Radiation Sensitivity Network: A Biomarker Discovery Platform

PURPOSE The discovery of effective biomarkers is a fundamental goal of molecular medicine. Developing a systems-biology understanding of radiosensitivity can enhance our ability of identifying radiation-specific biomarkers. METHODS AND MATERIALS Radiosensitivity, as represented by the survival fraction at 2 Gy was modeled in 48 human cancer cell lines. We applied a linear regression algorithm that integrates gene expression with biological variables, including ras status (mut/wt), tissue of origin and p53 status (mut/wt). RESULTS The biomarker discovery platform is a network representation of the top 500 genes identified by linear regression analysis. This network was reduced to a 10-hub network that includes c-Jun, HDAC1, RELA (p65 subunit of NFKB), PKC-beta, SUMO-1, c-Abl, STAT1, AR, CDK1, and IRF1. Nine targets associated with radiosensitization drugs are linked to the network, demonstrating clinical relevance. Furthermore, the model identified four significant radiosensitivity clusters of terms and genes. Ras was a dominant variable in the analysis, as was the tissue of origin, and their interaction with gene expression but not p53. Overrepresented biological pathways differed between clusters but included DNA repair, cell cycle, apoptosis, and metabolism. The c-Jun network hub was validated using a knockdown approach in 8 human cell lines representing lung, colon, and breast cancers. CONCLUSION We have developed a novel radiation-biomarker discovery platform using a systems biology modeling approach. We believe this platform will play a central role in the integration of biology into clinical radiation oncology practice.

Validation of a Radiosensitivity Molecular Signature in Breast Cancer

Purpose: Previously, we developed a radiosensitivity molecular signature [radiosensitivity index (RSI)] that was clinically validated in 3 independent datasets (rectal, esophageal, and head and neck) in 118 patients. Here, we test RSI in radiotherapy (RT)-treated breast cancer patients. Experimental Design: RSI was tested in 2 previously published breast cancer datasets. Patients were treated at the Karolinska University Hospital (n = 159) and Erasmus Medical Center (n = 344). RSI was applied as previously described. Results: We tested RSI in RT-treated patients (Karolinska). Patients predicted to be radiosensitive (RS) had an improved 5-year relapse-free survival when compared with radioresistant (RR) patients (95% vs. 75%, P = 0.0212), but there was no difference between RS/RR patients treated without RT (71% vs. 77%, P = 0.6744), consistent with RSI being RT-specific (interaction term RSI × RT, P = 0.05). Similarly, in the Erasmus dataset, RT-treated RS patients had an improved 5-year distant metastasis-free survival over RR patients (77% vs. 64%, P = 0.0409), but no difference was observed in patients treated without RT (RS vs. RR, 80% vs. 81%, P = 0.9425). Multivariable analysis showed RSI is the strongest variable in RT-treated patients (Karolinska, HR = 5.53, P = 0.0987, Erasmus, HR = 1.64, P = 0.0758) and in backward selection (removal α of 0.10), RSI was the only variable remaining in the final model. Finally, RSI is an independent predictor of outcome in RT-treated ER+ patients (Erasmus, multivariable analysis, HR = 2.64, P = 0.0085). Conclusions: RSI is validated in 2 independent breast cancer datasets totaling 503 patients. Including prior data, RSI is validated in 5 independent cohorts (621 patients) and represents, to our knowledge, the most extensively validated molecular signature in radiation oncology. Clin Cancer Res; 18(18); 5134–43. ©2012 AACR.

A genome-based model for adjusting radiotherapy dose (GARD): a retrospective, cohort-based study

BACKGROUND Despite its common use in cancer treatment, radiotherapy has not yet entered the era of precision medicine, and there have been no approaches to adjust dose based on biological differences between or within tumours. We aimed to assess whether a patient-specific molecular signature of radiation sensitivity could be used to identify the optimum radiotherapy dose. METHODS We used the gene-expression-based radiation-sensitivity index and the linear quadratic model to derive the genomic-adjusted radiation dose (GARD). A high GARD value predicts for high therapeutic effect for radiotherapy; which we postulate would relate to clinical outcome. Using data from the prospective, observational Total Cancer Care (TCC) protocol, we calculated GARD for primary tumours from 20 disease sites treated using standard radiotherapy doses for each disease type. We also used multivariable Cox modelling to assess whether GARD was independently associated with clinical outcome in five clinical cohorts: Erasmus Breast Cancer Cohort (n=263); Karolinska Breast Cancer Cohort (n=77); Moffitt Lung Cancer Cohort (n=60); Moffitt Pancreas Cancer Cohort (n=40); and The Cancer Genome Atlas Glioblastoma Patient Cohort (n=98). FINDINGS We calculated GARD for 8271 tissue samples from the TCC cohort. There was a wide range of GARD values (range 1·66-172·4) across the TCC cohort despite assignment of uniform radiotherapy doses within disease types. Median GARD values were lowest for gliomas and sarcomas and highest for cervical cancer and oropharyngeal head and neck cancer. There was a wide range of GARD values within tumour type groups. GARD independently predicted clinical outcome in breast cancer, lung cancer, glioblastoma, and pancreatic cancer. In the Erasmus Breast Cancer Cohort, 5-year distant-metastasis-free survival was longer in patients with high GARD values than in those with low GARD values (hazard ratio 2·11, 95% 1·13-3·94, p=0·018). INTERPRETATION A GARD-based clinical model could allow the individualisation of radiotherapy dose to tumour radiosensitivity and could provide a framework to design genomically-guided clinical trials in radiation oncology. FUNDING None.

Integration of a Radiosensitivity Molecular Signature Into the Assessment of Local Recurrence Risk in Breast Cancer

PURPOSE Recently, we developed radiosensitivity (RSI), a clinically validated molecular signature that estimates tumor radiosensitivity. In the present study, we tested whether integrating RSI with the molecular subtype refines the classification of local recurrence (LR) risk in breast cancer. METHODS AND MATERIALS RSI and molecular subtype were evaluated in 343 patients treated with breast-conserving therapy that included whole-breast radiation therapy with or without a tumor bed boost (dose range 45-72 Gy). The follow-up period for patients without recurrence was 10 years. The clinical endpoint was LR-free survival. RESULTS Although RSI did not uniformly predict for LR across the entire cohort, combining RSI and the molecular subtype identified a subpopulation with an increased risk of LR: triple negative (TN) and radioresistant (reference TN-radioresistant, hazard ratio [HR] 0.37, 95% confidence interval [CI] 0.15-0.92, P=.02). TN patients who were RSI-sensitive/intermediate had LR rates similar to those of luminal (LUM) patients (HR 0.86, 95% CI 0.47-1.57, P=.63). On multivariate analysis, combined RSI and molecular subtype (P=.004) and age (P=.001) were the most significant predictors of LR. In contrast, integrating RSI into the LUM subtype did not identify additional risk groups. We hypothesized that radiation dose escalation was affecting radioresistance in the LUM subtype and serving as a confounder. An increased radiation dose decreased LR only in the luminal-resistant (LUM-R) subset (HR 0.23, 95% CI 0.05-0.98, P=.03). On multivariate analysis, the radiation dose was an independent variable only in the LUMA/B-RR subset (HR 0.025, 95% CI 0.001-0.946, P=.046), along with age (P=.008), T stage (P=.004), and chemotherapy (P=.008). CONCLUSIONS The combined molecular subtype-RSI identified a novel molecular subpopulation (TN and radioresistant) with an increased risk of LR after breast-conserving therapy. We propose that the combination of RSI and molecular subtype could be useful in guiding radiation therapy-based decisions in breast cancer.

Differences Between Colon Cancer Primaries and Metastases Using a Molecular Assay for Tumor Radiation Sensitivity Suggest Implications for Potential Oligometastatic SBRT Patient Selection

PURPOSE We previously developed a multigene expression model of tumor radiation sensitivity index (RSI) with clinical validation in multiple independent cohorts (breast, rectal, esophageal, and head and neck patients). The purpose of this study was to assess differences between RSI scores in primary colon cancer and metastases. METHODS AND MATERIALS Patients were identified from our institutional review board-approved prospective observational protocol. A total of 704 metastatic and 1362 primary lesions were obtained from a de-identified metadata pool. RSI was calculated using the previously published rank-based algorithm. An independent cohort of 29 lung or liver colon metastases treated with 60 Gy in 5 fractions stereotactic body radiation therapy (SBRT) was used for validation. RESULTS The most common sites of metastases included liver (n=374; 53%), lung (n=116; 17%), and lymph nodes (n=40; 6%). Sixty percent of metastatic tumors, compared with 54% of primaries, were in the RSI radiation-resistant peak, suggesting metastatic tumors may be slightly more radiation resistant than primaries (P=.01). In contrast, when we analyzed metastases based on anatomical site, we uncovered large differences in RSI. The median RSIs for metastases in descending order of radiation resistance were ovary (0.48), abdomen (0.47), liver (0.43), brain (0.42), lung (0.32), and lymph nodes (0.31) (P<.0001). These findings were confirmed when the analysis was restricted to lesions from the same patient (n=139). In our independent cohort of treated lung and liver metastases, lung metastases had an improved local control rate compared to that in patients with liver metastases (2-year local control rate of 100% vs 73.0%, respectively; P=.026). CONCLUSIONS Assessment of radiation sensitivity between primary and metastatic tissues of colon cancer histology revealed significant differences based on anatomical location of metastases. These initial results warrant validation in a larger clinical cohort.

Radiosensitivity Differences Between Liver Metastases Based on Primary Histology Suggest Implications for Clinical Outcomes After Stereotactic Body Radiation Therapy

PURPOSE/OBJECTIVES Evidence from the management of oligometastases with stereotactic body radiation therapy (SBRT) reveals differences in outcomes based on primary histology. We have previously identified a multigene expression index for tumor radiosensitivity (RSI) with validation in multiple independent cohorts. In this study, we assessed RSI in liver metastases and assessed our clinical outcomes after SBRT based on primary histology. METHODS AND MATERIALS Patients were identified from our prospective, observational protocol. The previously tested RSI 10 gene assay was run on samples and calculated using the published algorithm. An independent cohort of 33 patients with 38 liver metastases treated with SBRT was used for clinical correlation. RESULTS A total of 372 unique metastatic liver lesions were identified for inclusion from our prospective, institutional metadata pool. The most common primary histologies for liver metastases were colorectal adenocarcinoma (n=314, 84.4%), breast adenocarcinoma (n=12, 3.2%), and pancreas neuroendocrine (n=11, 3%). There were significant differences in RSI of liver metastases based on histology. The median RSIs for liver metastases in descending order of radioresistance were gastrointestinal stromal tumor (0.57), melanoma (0.53), colorectal neuroendocrine (0.46), pancreas neuroendocrine (0.44), colorectal adenocarcinoma (0.43), breast adenocarcinoma (0.35), lung adenocarcinoma (0.31), pancreas adenocarcinoma (0.27), anal squamous cell cancer (0.22), and small intestine neuroendocrine (0.21) (P<.0001). The 12-month and 24-month Kaplan-Meier rates of local control (LC) for colorectal lesions from the independent clinical cohort were 79% and 59%, compared with 100% for noncolorectal lesions (P=.019), respectively. CONCLUSIONS In this analysis, we found significant differences based on primary histology. This study suggests that primary histology may be an important factor to consider in SBRT radiation dose selection.

Predicting response to clinical radiotherapy: past, present, and future directions.

BACKGROUND Personalized radiation therapy holds the promise that the diagnosis, prevention, and treatment of cancer will be based on individual assessment of risk. Although advances in personalized radiation therapy have been achieved, the biological parameters that define individual radiosensitivity remain unclear. METHODS This review focuses on discussing the field of radiosensitivity predictive assays, a technology central to the concept of personalized medicine in radiation oncology. Two novel approaches, DNA end-binding complexes and gene expression classifiers, show promise in solving some of the logistic problems associated with previous assays. RESULTS Current data suggest that predicting clinical response to radiotherapy is possible. The delivery of this promise depends on the ability to define the variables that define response to clinical radiotherapy. A successful predictive assay is key to the development of personalized treatment strategies in radiation oncology. CONCLUSIONS Novel technologies need to be developed that will improve our understanding of the biological variables that define clinical tumor response and will lead to the development of a clinically useful assay.

The radiosensitivity index predicts for overall survival in glioblastoma

We have previously developed a multigene expression model of tumor radiosensitivity (RSI) with clinical validation in multiple cohorts and disease sites. We hypothesized RSI would identify glioblastoma patients who would respond to radiation and predict treatment outcomes. Clinical and array based gene expression (Affymetrix HT Human Genome U133 Array Plate Set) level 2 data was downloaded from the cancer genome atlas (TCGA). A total of 270 patients were identified for the analysis: 214 who underwent radiotherapy and temozolomide and 56 who did not undergo radiotherapy. Median follow-up for the entire cohort was 9.1 months (range: 0.04–92.2 months). Patients who did not receive radiotherapy were more likely to be older (p < 0.001) and of poorer performance status (p < 0.001). On multivariate analysis, RSI is an independent predictor of OS (HR = 1.64, 95% CI 1.08–2.5; p = 0.02). Furthermore, on subset analysis, radiosensitive patients had significantly improved OS in the patients with high MGMT expression (unmethylated MGMT), 1 year OS 84.1% vs. 53.7% (p = 0.005). This observation held on MVA (HR = 1.94, 95% CI 1.19–3.31; p = 0.008), suggesting that RT has a larger therapeutic impact in these patients. In conclusion, RSI predicts for OS in glioblastoma. These data further confirm the value of RSI as a disease-site independent biomarker.