Genomic-adjusted Radiation Dose Research Articles

Abstract PS16-09: Analysis of the Radiosensitivity Index/Genomic-Adjusted Radiation Dose (RSI-GARD) in Paired Pre- and Post-Treatment TNBC Samples: Implications for Biomarker-Guided Radiotherapy

Abstract Objectives: Adjuvant radiotherapy is an important component of curative treatment for triple-negative breast cancer (TNBC). While there are several accepted variations in nominal dose and fractionation, these regimens are generally felt to be radiobiologically equivalent. Molecular radiosensitivity biomarkers have the potential to allow tailoring of physical radiotherapy dose for individual patients. The Radiosensitivity Index (RSI)/Genomic-Adjusted Radiation Dose (GARD) is a gene expression signature that predicts intrinsic radiosensitivity across multiple malignancies, with initial studies demonstrating clinical utility of RSI-GARD in TNBC. Since most TNBC patients will receive neoadjuvant systemic therapy, we sought to understand how neoadjuvant therapy-induced molecular adaptations could alter intrinsic radiosensitivity and RSI-GARD. Methods: Total RNA was isolated from pre-treatment and paired surgical specimens (for patients with residual disease (RD)) from TNBC patients treated with chemotherapy on the NeoSTOP (NCT02413320) or chemoimmunotherapy on the NeoPACT (NCT03639948) neoadjuvant trials and was subjected to RNA exome sequencing. The RSI was calculated according to the published algorithm using normalized expression of the 10 genes in this signature (AR, JUN1, STAT1, PRKCB, RELA, ABL1, SUMO1, PAK2, HDAC1, and IRF1). The resulting RSI score was substituted as the surviving fraction (SF) using the linear quadratic model of radiation-induced cellular lethality assuming a fractional dose of 2.0 Gy and a constant β=0.05 Gy−2 to derive a patient-specific α, which is a coefficient modifying dose-dependent cellular lethality resulting from binary mis-repair of double-strand breaks (DSBs) arising from a single particle track. Results: Pre-treatment sequencing data were available for N=200 patients and the overall pathologic complete response (pCR) rate was 56.5%. Paired pre- and post-treatment sequencing data were available for N=58 patients (N=27 from NeoSTOP and N=31 from NeoPACT). One NeoPACT patient had a negative RSI, which led to an undefined α. Under the assumptions of the RSI-substituted LQ model, the median α/β ratio (reflecting sensitivity to fractionation) for all pre-treatment samples was 11.7 Gy. TNBC patients who achieved pCR had a significantly higher pre-treatment α (i.e., higher intrinsic radiosensitivity) compared to patients who had residual disease (P=0.04), and this trend was the same for patients treated with chemotherapy (P=0.31) or chemoimmunotherapy (P=0.05). Importantly, α did not change significantly between paired pre- and post-treatment samples (median change post- vs. pre-treatment=0.03 Gy−1 (range -0.25 to +0.24 Gy−1; P=0.26) in patients with RD, with no difference in subgroups treated with chemotherapy (P=0.99) or chemoimmunotherapy (P=0.10). Conclusions: We found that predicted intrinsic radiosensitivity using the RSI-GARD is associated with pathologic response to neoadjuvant systemic therapy, suggesting that this signature also predicts intrinsic chemosensitivity in TNBC patients. The modeled α/β ratio for all pre-treatment samples was 11.7 Gy, which is substantially higher than empirically determined α/β ratios in studies that were enriched for HR+/HER2- breast cancers. Globally, the RSI-GARD score was not impacted by neoadjuvant systemic therapy, with no significant change in the score in paired pre- and post-treatment samples. These results can inform future testing and implementation of RSI-GARD into prospective trials. Citation Format: Shane Stecklein, Julia White, Rachel Yoder, Joshua Staley, Zachary Schmitt, Anne O'Dea, Lauren Nye, Deepti Satelli, Gregory Crane, Rashna Madan, Maura O'Neil, Andrew Godwin, Harsh Pathak, Qamar Khan, Joyce O'Shaughnessy, Priyanka Sharma. Analysis of the Radiosensitivity Index/Genomic-Adjusted Radiation Dose (RSI-GARD) in Paired Pre- and Post-Treatment TNBC Samples: Implications for Biomarker-Guided Radiotherapy [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PS16-09.

Habitat escalated adaptive therapy (HEAT): a phase 2 trial utilizing radiomic habitat-directed and genomic-adjusted radiation dose (GARD) optimization for high-grade soft tissue sarcoma

BackgroundSoft tissue sarcomas (STS), have significant inter- and intra-tumoral heterogeneity, with poor response to standard neoadjuvant radiotherapy (RT). Achieving a favorable pathologic response (FPR ≥ 95%) from RT is associated with improved patient outcome. Genomic adjusted radiation dose (GARD), a radiation-specific metric that quantifies the expected RT treatment effect as a function of tumor dose and genomics, proposed that STS is significantly underdosed. STS have significant radiomic heterogeneity, where radiomic habitats can delineate regions of intra-tumoral hypoxia and radioresistance. We designed a novel clinical trial, Habitat Escalated Adaptive Therapy (HEAT), utilizing radiomic habitats to identify areas of radioresistance within the tumor and targeting them with GARD-optimized doses, to improve FPR in high-grade STS.MethodsPhase 2 non-randomized single-arm clinical trial includes non-metastatic, resectable high-grade STS patients. Pre-treatment multiparametric MRIs (mpMRI) delineate three distinct intra-tumoral habitats based on apparent diffusion coefficient (ADC) and dynamic contrast enhanced (DCE) sequences. GARD estimates that simultaneous integrated boost (SIB) doses of 70 and 60 Gy in 25 fractions to the highest and intermediate radioresistant habitats, while the remaining volume receives standard 50 Gy, would lead to a > 3 fold FPR increase to 24%. Pre-treatment CT guided biopsies of each habitat along with clip placement will be performed for pathologic evaluation, future genomic studies, and response assessment. An mpMRI taken between weeks two and three of treatment will be used for biological plan adaptation to account for tumor response, in addition to an mpMRI after the completion of radiotherapy in addition to pathologic response, toxicity, radiomic response, disease control, and survival will be evaluated as secondary endpoints. Furthermore, liquid biopsy will be performed with mpMRI for future ancillary studies.DiscussionThis is the first clinical trial to test a novel genomic-based RT dose optimization (GARD) and to utilize radiomic habitats to identify and target radioresistance regions, as a strategy to improve the outcome of RT-treated STS patients. Its success could usher in a new phase in radiation oncology, integrating genomic and radiomic insights into clinical practice and trial designs, and may reveal new radiomic and genomic biomarkers, refining personalized treatment strategies for STS.Trial registrationNCT05301283.Trial statusThe trial started recruitment on March 17, 2022.

Using the genomic adjusted radiation dose (GARD) to personalize the radiation dose in nasopharyngeal cancer

BackgroundLocally advanced nasopharyngeal cancer (NPC) patients undergoing radiotherapy are at risk of treatment failure, particularly locoregional recurrence. To optimize the individual radiation dose, we hypothesize that the genomic adjusted radiation dose (GARD) can be used to correlate with locoregional control. MethodsA total of 92 patients with American Joint Committee on Cancer / International Union Against Cancer stage III to stage IVB recruited in a randomized phase III trial were assessed (NPC-0501) (NCT00379262). Patients were treated with concurrent chemo-radiotherapy plus (neo) adjuvant chemotherapy. The primary endpoint is locoregional failure free rate (LRFFR). ResultsDespite the homogenous physical radiation dose prescribed (Median: 70 Gy, range 66–76 Gy), there was a wide range of GARD values (median: 50.7, range 31.1–67.8) in this cohort. In multivariable analysis, a GARD threshold (GARDT) of 45 was independently associated with LRFFR (p = 0.008). By evaluating the physical dose required to achieve the GARDT (RxRSI), three distinct clinical subgroups were identified: (1) radiosensitive tumors that RxRSI at dose < 66 Gy (N = 59, 64.1 %) (b) moderately radiosensitive tumors that RxRSI dose within the current standard of care range (66–74 Gy) (N = 20, 21.7 %), (c) radioresistant tumors that need a significant dose escalation above the current standard of care (>74 Gy) (N = 13, 14.1 %). ConclusionGARD is independently associated with locoregional control in radiotherapy-treated NPC patients from a Phase 3 clinical trial. GARD may be a potential framework to personalize radiotherapy dose for NPC patients.

Genomic Adjusted Radiation Dose (GARD) Predicts Overall Survival and Outperforms AJCC 8th Edition in Prognostication of HPV-Positive Oropharyngeal Squamous Cell Carcinoma

<h3>Purpose/Objective(s)</h3> Treatment decision-making in oropharyngeal squamous cell carcinoma (OPSCC) includes clinical stage, HPV status, and smoking history. Despite improvements in staging with separation of HPV positive and negative OPSCC in AJCC 8^th edition (AJCC8), patients are largely treated with a uniform approach, with recent efforts on de-intensification in low-risk patients. We have previously shown, in a pooled analysis, that the genomic adjusted radiation dose (GARD) is predictive of radiation treatment benefit and can be used to guide dose selection. We hypothesize that GARD can be used to predict overall survival (OS) in HPV-positive OPSCC patients treated with radiotherapy (RT). <h3>Materials/Methods</h3> Gene expression profiles were analyzed for 234 formalin-fixed paraffin-embedded samples from HPV-positive OPSCC patients within an international, multi-institutional, prospective/retrospective observational study (Big Data and Models for Personalized Head and Neck Cancer Decision Support (BD2Decide, NCT02832102). GARD, a measure of the biological effect of RT, was calculated for each patient as previously described. In total, 203 patients received definitive treatment (chemoradiation or RT alone), and 31 patients received post-operative RT. Two RT dose fractionations were utilized for definitive cases (70 Gy in 35 fractions or 69.96 Gy in 33 fractions). Median RT dose was 70 Gy (mean 50.88-74) for definitive cases and 66 Gy (range 44-70) for post-operative cases. 29 patients received surgery and adjuvant chemoradiation, 14 received surgery and adjuvant RT, and 19 received RT alone. The median follow up was 46.2 months (95% CI, 41.2 to 50). Cox proportional hazards analyses were performed with GARD as a continuous variable and ROC analyses comparing the performance of GARD with AJCC8. <h3>Results</h3> Despite uniform radiation dose utilization, GARD showed significant heterogeneity (range 30-110), reflecting the underlying genomic differences in the cohort. On multivariate analysis, each unit increase in GARD was associated with an improvement in OS (HR= 0.97, 95% CI 0.94-1.0, p=0.040) compared to AJCC8 (HR=1.96, 95% CI 0.90-4.32, p=0.0952). ROC analysis for GARD at 24 months yielded an AUC of 75.6 (95% CI 61.8-89.4) compared with an AUC of 73.9 (95% CI 56.9-90.9) for AJCC8. GARD ≥64.97 was associated with improved OS (HR=0.29, p=0.005). <h3>Conclusion</h3> In this multi-institutional cohort of patients with HPV-positive OPSCC, GARD predicts OS as a continuous variable and outperforms AJCC8 in prognostication. We propose that GARD, which provides the first opportunity for genomic guided personalization of radiation dose, should be incorporated in the diagnostic workup of HPV-positive OPSCC patients and can be used to guide treatment decision making.

Genomic Adjusted Radiation Dose (GARD) Predicts Local Recurrence and Overall Survival in Triple-Negative Breast Cancer

<h3>Purpose/Objective(s)</h3> Although it is known that breast cancer is a heterogeneous disease with variable response to different treatments, radiation therapy (RT) is typically prescribed with a uniform approach. We have previously shown that the genomic adjusted radiation dose (GARD) is predictive of RT benefit in a pooled analysis which included hormone receptor positive and triple-negative breast cancer (TNBC) patients. Recent trials showed no benefit in overall survival (OS) with a randomized 16 Gy boost, but we hypothesize that GARD can predict outcomes in TNBC and could select patients who would benefit from dose escalation. <h3>Materials/Methods</h3> We used two IRB-approved, prospectively collected cohorts of TNBC patients treated with RT, 58 from the Netherlands Cancer Institute and 55 from Moffitt Cancer Center. Cox proportional hazards analysis was performed with GARD as a continuous variable for local recurrence (LR) and OS. We used the log-rank test to maximize the difference between groups above and below each GARD value and defined RxRSI as the physical radiation dose necessary to achieve that GARD cutpoint. Patients were stratified into 3 groups based on whether their RxRSI fell above, within, or below the standard-of-care dosing range of 50-66 Gy. To determine the minimum enrollment number needed to show a significant LR benefit in a clinical trial, we performed a power analysis by simulating an <i>in silico</i> trial with two arms. In one arm, all patients received 50 Gy, and in the other, dose was boosted to 66 Gy only for patients whose RxRSI fell within 50-66 Gy while the rest of the cohort received 50 Gy. Thus, selective dose escalation was only applied to patients who we predicted would benefit from it based on their individual RSI and GARD. The simulation was repeated with different enrollment numbers to find the smallest number needed to show a significant benefit in LR in a trial of this design. <h3>Results</h3> While the radiation dose delivered in the cohorts analyzed was between 43.5 to 75 Gy, GARD showed a much wider range (12 to 76). Cox analysis showed significant association between GARD and OS (HR=0.88, p=0.02) and strong association between GARD and LR (HR = 0.96, p=0.08). Receiving GARD of at least 22 maximized LR and OS benefit (p<0.01). By this measure, 50% of patients were optimized with 50 Gy, 22% would benefit from a boost, and 28% require higher radiation dose than is standard. Using our stochastic <i>in silico</i> method, we found that to see a significant difference in LR between boosted and standard-dosing groups in a trial of genomically stratified TNBC patients, only 30 patients would be needed. <h3>Conclusion</h3> Our analysis shows that for each unit GARD, a statistically significant increase in OS seen, and that using genomic stratification in trial design yields a small and quickly tractable trial to test the efficacy of this approach in tumor bed dose decision-making.

Exploring breast and prostate cancer RNA-seq derived radiosensitivity with the Genomic Adjusted Radiation Dose (GARD) model

The use of a 10 gene transcriptional signature as part of the GARD model has been shown to be predictive of radiotherapy benefit for a range of cancers, with the potential to determine an optimal overall dose per patient. We used publicly available RNA-seq transcriptomics data from a luminal B breast cancer patient and from 14 prostate cancer patients to explore the radiosensitivity indices (RSI) and so GARD estimates of both tumour and proximal normal biopsies from each individual. Clear differences of clinical relevance in derived radiobiological properties between tumour and proximal normal tissues were evident for the breast cancer patient, whilst such differences across the prostate cancer cohort were more equivocal. Using the prostate cancer cohort’s median tumour predicted GARD value as a threshold for high therapeutic effect for radiotherapy, we found evidence that a higher overall prescribed dose than the widely used 72 Gy/36fx could benefit half of these patients. This exploratory study demonstrates the potential combining the GARD model with sequencing based transcriptomics could have in informing personalised radiotherapeutic practise for both breast and prostate cancer patients.

Abstract MS2-2: Can genomic risk be used to tailor radiation therapy?

Abstract Radiation therapy (RT) reduces any-recurrence risks and breast cancer related mortality after mastectomy for node positive disease and after breast conserving surgery. The relative risk reduction is partially independent from patient, tumour and treatment related factors, while the absolute benefit depends to a large extent on the absolute risks without RT. Of note is an important interaction between systemic and locoregional treatments. The continuous improvements in outcomes after breast cancer diagnosis led to a quest for treatment de-escalation. In this, the decreased extent of axillary surgery led to an increased proportion of patients eligible for nodal RT. While most data concerning prognostic and predictive value of genomic data is related to metastatic disease and to systemic treatments, more recently also RT evolves from the “one-size fits all” to a much more “tailor-made approach”. Optimally, current tailoring of locoregional treatments involves decreasing volumes and extensiveness, maintaining treatment focus mainly at high-risk areas, while de-escalating treatment to low-risk areas. The multidisciplinary aspects call for integration of predictive and prognostic tools and looking into optimally combining less of both systemic and locoregional treatments to optimise tumour control while sparing normal tissues.Not only distant but also locoregional recurrences risks depend on the intrinsic tumour phenotype, which can be expressed by the genomic profile. Several post-hoc evaluations of datasets derived from prospective studies showed that the genomic profile score can assist in predicting locoregional recurrence risks (and thereby indirectly the possible benefits of RT) or, with more scarce data, the interaction between the genomically predicted risk and the risk reduction obtained by RT. Currently, several trials are evaluating further de-escalation of locoregional treatments in patients with favourable intrinsic tumour types. Of interest in patients with intrinsic high-risk tumour types is the response to primary systemic therapy, with trials ongoing to de-escalate both axillary and local management after a favourable tumour response. Preclinical data suggests that immunogenic cell death, mediated through activation of tumour infiltrating lymphocytes (TIL’s) might contribute to the efficacy of RT. However, high TIL’s levels are seen in only 10% of breast cancers, more commonly in basal-like- and Her2-enriched subtypes. The mechanism behind this association with TIL’s remain unclear. Of particular interest seems the recently validated genomic adjusted radiation dose (GARD), aimed to personalise RT dose prescription based on the biological effect rather than merely on physical dose. GARD combines a gene expression assay, assuming pan-tissue biological networks of radiosensitivity and radioresistance, with a linear quadratic model for estimation of total radiation dose in tissues. Up to now, clinical utility remains to be confirmed, preferably by integrating GARD in prospective clinical research. Summary: at present no validated biomarkers can reliably predict the benefit derived from RT at the individual patient’s level, with current evidence not yet robust enough to guide decision making. Available data are mainly based on retrospective analyses of older trials, necessitating prospective validation in contemporary cohorts. Research is ongoing to personalise RT using biological factors including gene expression profiles. Projects to identify genetic variants associated with susceptibility to radiotoxicity are required to proceed to a more personalised indication/target volume/dose approach for our patients. In the meantime, we should avoid jeopardising the impressive improvements obtained in the outcomes after breast cancer diagnosis by precipitated treatment de-escalating based on beliefs and assumptions. Citation Format: PMP Poortmans. Can genomic risk be used to tailor radiation therapy? [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr MS2-2.

Genomic-adjusted radiation dose to personalise radiotherapy.

In The Lancet Oncology, Jacob Scott and colleagues 1 Scott JG Sedor G Ellswroth P et al. Pan-cancer prediction of radiotherapy benefit using genomic-adjusted radiation dose (GARD): a cohort-based pooled analysis. Lancet Oncol. 2021; (published Aug 4.)https://doi.org/10.1016/S1470-2045(21)00347-8 Summary Full Text Full Text PDF Scopus (17) Google Scholar describe an algorithm using the genomic-adjusted radiation dose (GARD) to personalise radiotherapy dose prescription on the basis of the biological effect of the dose rather than on physical dose alone. GARD is a preclinically and clinically validated combination of a gene expression assay, which assumes pan-tissue biological networks of radiosensitivity and radioresistance, with a linear quadratic model for estimation of total radiation dose in tissues. In previous work, the investigators identified a tumour molecular signature based on the expression of ten genes (AR, cJun [JUN], STAT1, PKC-β [PRKCB], RelA [RELA], cABL [ABL], SUMO1, PAK2, HDAC1, and IRF1) that correlated with radiosensitivity (expressed as tumour cell survival fraction per 2 Gy fraction) in a number of cancer cell lines, called the radiosensitivity index (RSI). 2 Eschrich S Zhang H Zhao H et al. Systems biology modeling of the radiation sensitivity network: a biomarker discovery platform. Int J Radiat Oncol Biol Phys. 2009; 75: 497-505 Summary Full Text Full Text PDF PubMed Scopus (157) Google Scholar , 3 Eschrich SA Pramana J Zhang H et al. A gene expression model of intrinsic tumor radiosensitivity: prediction of response and prognosis after chemoradiation. Int J Radiat Oncol Biol Phys. 2009; 75: 489-496 Summary Full Text Full Text PDF PubMed Scopus (191) Google Scholar , 4 Scott JG Berglund A Schell MJ et al. A genome-based model for adjusting radiotherapy dose (GARD): a retrospective, cohort-based study. Lancet Oncol. 2017; 18: 202-211 Summary Full Text Full Text PDF PubMed Scopus (219) Google Scholar Later, RSI was validated on various tumour types that led to the GARD algorithm. 4 Scott JG Berglund A Schell MJ et al. A genome-based model for adjusting radiotherapy dose (GARD): a retrospective, cohort-based study. Lancet Oncol. 2017; 18: 202-211 Summary Full Text Full Text PDF PubMed Scopus (219) Google Scholar In the current publication, 1 Scott JG Sedor G Ellswroth P et al. Pan-cancer prediction of radiotherapy benefit using genomic-adjusted radiation dose (GARD): a cohort-based pooled analysis. Lancet Oncol. 2021; (published Aug 4.)https://doi.org/10.1016/S1470-2045(21)00347-8 Summary Full Text Full Text PDF Scopus (17) Google Scholar Scott and colleagues validated GARD in 1615 patients with seven cancer types from 11 study cohorts. They found that GARD was associated with time to first recurrence and overall survival for patients receiving radiotherapy and that, by contrast with physical dose, it predicted radiotherapy benefit. Consequently, GARD is a model that can potentially serve as a tool to individualise prescribed radiation doses in clinical practice. The efforts of Scott and colleagues need to be applauded worldwide, because radiotherapy is considerably lagging compared with the enormous progress done in the field of personalised medicine that currently mainly applies to decisions on the use of systemic therapy or targeted agents. Pan-cancer prediction of radiotherapy benefit using genomic-adjusted radiation dose (GARD): a cohort-based pooled analysisThe biological effect of radiotherapy, as quantified by GARD, is significantly associated with time to first recurrence and overall survival for patients with cancer treated with radiation. It is predictive of radiotherapy benefit, and physical dose of radiation is not. We propose integration of genomics into radiation dosing decisions, using a GARD-based framework, as the new paradigm for personalising radiotherapy prescription dose. Full-Text PDF

Pan-cancer prediction of radiotherapy benefit using genomic-adjusted radiation dose (GARD): a cohort-based pooled analysis.

Despite advances in cancer genomics, radiotherapy is still prescribed on the basis of an empirical one-size-fits-all paradigm. Previously, we proposed a novel algorithm using the genomic-adjusted radiation dose (GARD) model to personalise prescription of radiation dose on the basis of the biological effect of a given physical dose of radiation, calculated using individual tumour genomics. We hypothesise that GARD will reveal interpatient heterogeneity associated with opportunities to improve outcomes compared with physical dose of radiotherapy alone. We aimed to test this hypothesis and investigate the GARD-based radiotherapy dosing paradigm. We did a pooled, pan-cancer analysis of 11 previously published clinical cohorts of unique patients with seven different types of cancer, which are all available cohorts with the data required to calculate GARD, together with clinical outcome. The included cancers were breast cancer, head and neck cancer, non-small-cell lung cancer, pancreatic cancer, endometrial cancer, melanoma, and glioma. Our dataset comprised 1615 unique patients, of whom 1298 (982 with radiotherapy, 316 without radiotherapy) were assessed for time to first recurrence and 677 patients (424 with radiotherapy and 253 without radiotherapy) were assessed for overall survival. We analysed two clinical outcomes of interest: time to first recurrence and overall survival. We used Cox regression, stratified by cohort, to test the association between GARD and outcome with separate models using dose of radiation and sham-GARD (ie, patients treated without radiotherapy, but modelled as having a standard-of-care dose of radiotherapy) for comparison. We did interaction tests between GARD and treatment (with or without radiotherapy) using the Wald statistic. Pooled analysis of all available data showed that GARD as a continuous variable is associated with time to first recurrence (hazard ratio [HR] 0·98 [95% CI 0·97-0·99]; p=0·0017) and overall survival (0·97 [0·95-0·99]; p=0·0007). The interaction test showed the effect of GARD on overall survival depends on whether or not that patient received radiotherapy (Wald statistic p=0·011). The interaction test for GARD and radiotherapy was not significant for time to first recurrence (Wald statistic p=0·22). The HR for physical dose of radiation was 0·99 (95% CI 0·97-1·01; p=0·53) for time to first recurrence and 1·00 (0·96-1·04; p=0·95) for overall survival. The HR for sham-GARD was 1·00 (0·97-1·03; p=1·00) for time to first recurrence and 1·00 (0·98-1·02; p=0·87) for overall survival. The biological effect of radiotherapy, as quantified by GARD, is significantly associated with time to first recurrence and overall survival for patients with cancer treated with radiation. It is predictive of radiotherapy benefit, and physical dose of radiation is not. We propose integration of genomics into radiation dosing decisions, using a GARD-based framework, as the new paradigm for personalising radiotherapy prescription dose. None. VIDEO ABSTRACT.

Genomic identification of sarcoma radiosensitivity and the clinical implications for radiation dose personalization

BackgroundSoft-tissue sarcomas (STS) are heterogeneous with variable response to radiation therapy (RT). Utilizing the radiosensitivity index (RSI) we estimated the radiobiologic ratio of lethal to sublethal damage (α/β), genomic-adjusted radiation dose(GARD), and in-turn a biological effective radiation dose (BED). MethodsTwo independent cohorts of patients with soft-tissue sarcoma were identified. The first cohort included 217 genomically-profiled samples from our institutional prospective tissue collection protocol; RSI was calculated for these samples, which were then used to dichotomize the population as either highly radioresistant (HRR) or conventionally radioresistant (CRR). In addition, RSI was used to calculate α/β ratio and GARD, providing ideal dosing based on sarcoma genomic radiosensitivity. A second cohort comprising 399 non-metastatic-STS patients treated with neoadjuvant RT and surgery was used to validate our findings. ResultsBased on the RSI of the sample cohort, 84% would historically be considered radioresistant. We identified a HRR subset that had a significant difference in the RSI, and clinically a lower tumor response to radiation (2.4% vs. 19.4%), 5-year locoregional-control (76.5% vs. 90.8%), and lower estimated α/β (3.29 vs. 5.98), when compared to CRR sarcoma. Using GARD, the dose required to optimize outcome in the HRR subset is a BEDα/β=3.29 of 97 Gy. ConclusionsWe demonstrate that on a genomic scale, that although STS is radioresistant overall, they are heterogeneous in terms of radiosensitivity. We validated this clinically and estimated an α/β ratio and dosing that would optimize outcome, personalizing dose.

GARD is a Pan-Cancer Predictor of Outcome and Radiation Treatment Benefit

Background: Radiation therapy (RT) is prescribed using an empiric one-size-fits-all paradigm. We hypothesize that the Genomic Adjusted Radiation Dose (GARD), a quantification of personalized RT dose effect, will reveal interpatient heterogeneity associated with opportunities to improve outcomes compared to physical RT dose alone. To test this hypothesis, we performed a pan-cancer, pooled analysis of all available cohorts with information available to calculate GARD. Methods: Using 9 previously-published datasets, we defined two clinical endpoints: (i) time to first recurrence and (ii) overall survival, comprising 1,257 (972 +RT, 285 -RT) and 636 patients (414 +RT, 222 -RT), respectively. We used Cox regression stratified by disease site to test association between GARD and outcome with separate models using RT dose and sham-GARD for comparison. Interaction tests between GARD and treatment (+/- RT) were performed using the Wald statistic. Results: Pooled analysis reveals GARD is a continuous predictor of recurrence (HR = 0.984, CI [0.982,0.996], p=0.004) and survival (HR = 0.975, CI [0.958, 0.993], p=0.005). Interaction term was significant for OS (p =0.026) and for recurrence for patients who achieve GARD>19.2 (p=0.04). Physical RT dose and sham-GARD were not significantly associated with either outcome. Conclusions: Biologic effect, as quantified by GARD, is significantly associated with recurrence and survival: it is predictive of RT benefit; and physical RT dose is not. We propose integration of genomics into radiation dosing decisions, using a GARD-based framework, as the new paradigm for personalizing RT treatment. Funding: There was no funding for this work Declaration of Interest: JGS, SAE and JFTR hold IP on RSI, GARD and RxRSI, in addition to equity in Cvergenx, a company which seeks to commercialize these methods. All other authors have nothing to declare.

Personalizing Radiotherapy Prescription Dose Using Genomic Markers of Radiosensitivity and Normal Tissue Toxicity in NSCLC.

Cancer sequencing efforts have revealed that cancer is the most complex and heterogeneous disease that affects humans. However, radiation therapy (RT), one of the most common cancer treatments, is prescribed on the basis of an empirical one-size-fits-all approach. We propose that the field of radiation oncology is operating under an outdated null hypothesis: that all patients are biologically similar and should uniformly respond to the same dose of radiation. We have previously developed the genomic-adjusted radiation dose, a method that accounts for biological heterogeneity and can be used to predict optimal RT dose for an individual patient. In this article, we use genomic-adjusted radiation dose to characterize the biological imprecision of one-size-fits-all RT dosing schemes that result in both over- and under-dosing for most patients treated with RT. To elucidate this inefficiency, and therefore the opportunity for improvement using a personalized dosing scheme, we develop a patient-specific competing hazards style mathematical model combining the canonical equations for tumor control probability and normal tissue complication probability. This model simultaneously optimizes tumor control and toxicity by personalizing RT dose using patient-specific genomics. Using data from two prospectively collected cohorts of patients with NSCLC, we validate the competing hazards model by revealing that it predicts the results of RTOG 0617. We report how the failure of RTOG 0617 can beexplained by the biological imprecision of empirical uniform dose escalation which results in 80% of patients being overexposed to normal tissue toxicity without potential tumor control benefit. Our data reveal a tapestry of radiosensitivity heterogeneity, provide a biological framework that explains the failure of empirical RT dose escalation, and quantify the opportunity to improve clinical outcomes in lung cancer by incorporating genomics into RT.

A Multi-Outcome Meta-Analysis Method Reveals the Genomic Adjusted Radiation Dose is a Continuous Predictive Biomarker of Radiation Outcome

A Multi-Outcome Meta-Analysis Method Reveals the Genomic Adjusted Radiation Dose is a Continuous Predictive Biomarker of Radiation Outcome

Modeling precision genomic-based radiation dose response in rectal cancer.

Aim: Genomic-based risk stratification to personalize radiation dose in rectal cancer. Patients & methods: We modeled genomic-based radiation dose response using the previously validated radiosensitivity index (RSI) and the clinically actionable genomic-adjusted radiation dose. Results: RSI of rectal cancer ranged from 0.19 to 0.81 in a bimodal distribution. A pathologic complete response rate of 21% was achieved in tumors with an RSI <0.31 at a minimal genomic-adjusted radiation dose of 29.76 when modeling RxRSI to the commonly prescribed physical dose of 50 Gy. RxRSI-based dose escalation to 55 Gy in tumors with an RSI of 0.31-0.34 could increase pathologic complete response by 10%. Conclusion: This study provides a theoretical platform for development of an RxRSI-based prospective trial in rectal cancer.

Intrinsic radiosensitivity, genomic-based radiation dose and patterns of failure of penile cancer in response to adjuvant radiation therapy.

Optimal postoperative radiation therapy (PORT) dose is unclear in penile squamous cell carcinoma (PeSCC). Herein, we characterized the radiosensitivity index (RSI) and genomic-adjusted radiation dose (GARD) profiles in a cohort of patients with PeSCC, and assessed the application of GARD to personalize PORT. A total of 25 PeSCC samples were identified for transcriptomic profiling. The RSI score and GARD were derived for each sample. A cohort of 34 patients was reviewed for clinical correlation. The median RSI for PeSCC was 0.482 (range 0.215-0.682). The majority (n=21; 84%) of cases were classified as radioresistant. PeSCC GARD ranged from 9.56 to 38.39 (median 18.25), suggesting variable therapeutic effects from PORT. We further determined the optimal GARD-based RT doses to improve locoregional control. We found that therapeutic benefit was only achieved in 52% of PeSCC lesions with PORT of 50Gy, in contrast to 84% benefit from GARD-modeled PORT of 66Gy. In the clinical cohort, the majority of patients presented with pathological N2 or N3 disease (n=31; 91%) and was treated with adjuvant concurrent platinum-based chemoradiotherapy (CRT, n=30; 88%). Fourteen of the 34 patients (41%) had locoregional recurrence (LRR), of which half had LRR within six months of completion of PORT. The majority of PeSCC are intrinsically radioresistant with a low GARD-based therapeutic effect from PORT dose of 50Gy, consistent with the observed high rate of LRR in the clinical cohort. A GARD-based strategy will allow personalizing PORT dose prescription to individual tumor biology and improve outcomes.

Beyond Blind Dose-Escalation: Modeling Precision Genomic-Based Radiation Dose-Response In Rectal Cancer

Improved pathologic complete response with preoperative radiation dose escalation has been reported in locally advanced rectal cancer (LARC), but risk stratification to personalize radiation dose prescription has not been explored. We modeled genomic-based radiation dose-response utilizing the previously validated radiosensitivity index (RSI) and the clinically actionable genomic-adjusted radiation dose (GARD). We hypothesized that optimal therapeutic gain could be achieved in LARC stratified by genomic heterogeneity and RSI-based dose prescription (RxRSI). We assessed the patterns of pathologic response in a cohort of LARC patients treated with neoadjuvant chemoradiotherapy (NACRT) (cohort 1). Tumor regression grade (TRG) was analyzed for complete (TRG 0; pCR), favorable (TRG 1, moderate response), and poor (TRG 2 and 3) responders. We characterized the tumor genomics in a separate cohort of 113 rectal cancer tissue samples (cohort 2) to calculate tumor specific RSI, from which a tumor specific α was derived to generate GARD. RxRSI was modeled using the GARD threshold for which the percentile of RSI distribution mirrored the patterns of pathologic response in cohort 1. In cohort 1, the primary tumor stage was T3 (84%). The majority of patients (82%) received an RT dose of 50.4 Gy with concurrent 5-FU or Capecitabine. The rates of complete, favorable and poor responders were 21%, 53%, and 26%, respectively. The RSI of rectal cancer ranged from 0.19 to 0.81 in a bimodal distribution. A pCR rate of 21% would be achieved in tumors with RSI < 0.31 at a minimal GARD of 29.76 when modeling RxRSI to the commonly prescribed physical dose of 50.4 Gy. Modeling RxRSI dose escalation to 55 Gy in tumors with an RSI 0.31-0.34 would lead additional 10% tumors to achieve a pCR, whereas some radioresistant tumors would require much higher genomic-based dose to achieve pCR. For example, in our model, 105 Gy would be required for a tumor with an RSI of 0.57 to achieve a complete response. In regards to modeling RxRSI for a favorable response, tumors with RSI 0.32-0.51 would have TRG 1 in response to 50.4 Gy at a minimal GARD of 16.9. RxRSI dose escalation to 60 Gy could potentially bring 95% tumors (RSI <0.57) to a favorable response. We modeled personalized dose escalation in rectal cancer based on genomic expression profiling. This study could provide a theoretical platform for the development of an RxRSI-based prospective trial in rectal cancer. We anticipate individualizing radiation dose selection for each patient and optimizing patient selection for nonoperative management.

Personalizing Radiotherapy Prescription Dose Using Genomic Markers of Radiosensitivity

Purpose/Objective(s): Although radiotherapy (RT) remains a critical curative agent for cancer, it has yet to adapt a biological basis in the clinic. We previously proposed the gene expression-based radiosensitivity index (RSI) and the genomic-adjusted radiation dose (GARD), as a clinically-feasible approach to biology-based RT. We hypothesize that empiric one-size fits all RT dose is biologically imprecise and negatively impacts clinical outcome. Materials/Methods: We utilize the RSI/GARD formalism to calculate a personalized RT prescription dose that biologically optimizes prescription dose for each patient (RxRSI). We integrated RxRSI into a commercial treatment planning system to create the first approach to personalized biologically-optimized RT plans. We compare RxRSI with standard of care empiric radiotherapy dose in 2 cohorts of NSCLC patients (n=60 and n=1,747) from an IRB-approved de-identified institutional biorepository. To quantify the opportunity inherent in personalized RT prescription, we develop a first-in kind precision RxRSI model that estimates local control (LC) based on whether an optimal RT dose (RxRSI) is delivered and penalizes outcome based on exposure to excess normal tissue complication risk (penalized local control, pLC). To validate the precision RxRSI model, we ran an in silico clinical trial similar to RTOG 0617 in the modeling cohort. Results: We demonstrate that up to 75% of patients receive non-optimal radiotherapy doses, when treated with a one-size fits all empiric approach. Conversely, we show that personalized, radiotherapy prescription dose delivers optimal doses to up to 75% of the patients even when restricted to doses within the standard of care. These differences result in quantifiable excess doses to normal tissue that negatively impact clinical outcome. We validate the precision RxRSI model by demonstrating that it predicts that empiric radiotherapy dose escalation to 74 Gy results in no improvement in radiotherapy-associated outcome (pLC) as demonstrated in RTOG 0617. The precision RxRSI model demonstrates that only an additional 16.2% of lung cancer patients achieve RxRSI by receiving 74 Gy (beyond 60Gy). A genomic-based strategy to deliver 74 Gy to only those patients could have produced a 6.3% absolute improvement in clinical outcome (pLC) for the whole population (p<0.05). Conclusion: The current empiric approach to radiotherapy dose prescription is biologically imprecise and over or under-treats up to 75% of lung cancer patients. These inaccuracies negatively impact patient outcome by either an over-exposure to normal tissue complications or by achieving a sub-optimal tumor control. Personalized RT prescription dose may improve radiotherapy-associated outcome in lung cancer by an absolute 6.3% using radiotherapy doses within current standard of care without an expected increase in normal tissue toxicity. Funding Statement: DeBartolo Personalized Medicine Institute. Declaration of Interests: JTR, SAE and JGS report IP (RSI (JTR, SAE), GARD (JTR, JGS) and RxRSI (JTR, JGS). JTR and SAE report stock in Cvergenx. TF, AW report stock in Varian. Ethics Approval Statement: Data was compiled for two cohorts of patients from an IRB-approved de-identified institutional bio-repository.

OC-0268 Intrinsic radiosensitivity, genomic-adjusted radiation dose and patterns of failure of penile cancer

OC-0268 Intrinsic radiosensitivity, genomic-adjusted radiation dose and patterns of failure of penile cancer

Integrated profiling the patterns of pathologic response to neoadjuvant chemoradiation and the genomic-based radiation sensitivity in rectal cancer.

573 Background: Given the lack of biomarkers to predict a pathologic complete response (pCR) after neoadjuvant chemoradiation (NACRT) for rectal cancer, selection for non-operative management (NOM) mandates complete clinical response. We have previously developed/validated a model to assess genomic-based tumor radiosensitivity: the radiosensitivity index (RSI), which formulates a clinically actionable model to calculate genomic-adjusted radiation dose (GARD). We determined the profiles of RSI and GARD for rectal cancer and correlated these findings with the pathologic response patterns. Methods: One hundred seventeen rectal cancer patients treated from 2009 to 2018 with NACRT were assessed for the tumor regression grade (TRG) (0 = pCR; 1 = moderate response; 2 = partial response; 3 = poor response). RSI was analyzed in an independent tissue cohort of 113 resected rectal cancer samples. GARD was derived as described before, which shows a high GARD value indicated a superior therapeutic effect of radiation. Results: Median follow-up from completion of NACRT was 26 months. The primary tumor stages were 7% T2, 84% T3, and 9% T4. The majority of patients (82%) received concurrent 5-FU or Capecitabine and (83%) received RT dose of 50.4 Gy (range 45-56 Gy). Median time from end of NACRT to surgery was 61 days (range 36-105 days). The patterns of pathological response were TRG 0 (n = 24; 21%), 1 (n = 62; 53%), 2 (n = 25; 21%), and 3 (n = 6; 5%), suggesting heterogeneous sensitivity to treatment with similar tumor stage and treatment regimens. The median RSI for the tissue cohort was 0.46 (range 0.19-0.81) with 37% of the samples considered radiosensitive based on prior data. GARD values ranged from 5.17 to 41.79 (median 19.24), suggesting heterogeneous RT therapeutic effects. Conclusions: The findings from the clinical cohort were consistent with the tissue cohort showing significant heterogeneity in the individual tumor radiosensitivity and GARD-based RT therapeutic effects. With the development of GARD-based prospective trials, we anticipate more biology-based customized RT dosing which could optimize patient selection for NOM and individualize the most appropriate dose for each patient.

Personalizing Prescription of Radiation Therapy Utilizing Genomic Markers of Radiosensitivity

In previous studies, we developed the radio-sensitivity index (RSI) and genomic adjusted radiation dose (GARD) to capture genomic heterogeneity and personalize radiation therapy (RT) prescription. We hypothesize that the current one-size fits-all paradigm to RT prescription is inefficient for individual patients given observed distributions of tumor RSI across large cohorts of patients. We demonstrate a workflow for clinical use of genomic features in RT prescription. We assessed tumor genomics in a cohort of 1229 patients with non-small lung cancer (NSCLC, Cohort 1) using RSI. To model the impact of genomic heterogeneity on RT dose prescription, we utilized a cohort of 60 NSCLC patients treated with surgery and postoperative RT (45 – 70 Gy, median 54 Gy, Cohort 2). For all analyses, expression data (Affymetrix Hu-RSTA-2a520709) were used to calculate RSI. Using RSI, we derive a patient specific alfa to calculate GARD. Previously we demonstrated that lung cancer patients in cohort 2 who achieved a GARD >= 33 had an improved local control (HR=3.4, 1.3, 9.1 p=0.02). We defined the personalized RSI/GARD RT dose prescription (RxRSI) as the physical dose in Gy required to achieve the GARD threshold of 33. We integrated GARD into the planning workflow of a commercial treatment planning system to create personalized genomic radiation treatment plans (pGRT) using standardized plan optimizations for both standard of care (SoC) and RxRSI doses. RxRSI plans were evaluated against standard planning criteria and normal tissue constraints, and compared with SoC prescription plans using biological effective voxel based dose scaling. In cohort 1, RSI demonstrated a bimodal distribution, indicating a uniform treatment dose may be sub-optimal. A similar heterogeneity was observed in cohort 2, (RSI (IQR: 0.25, 0.40), GARD (IQR: 24.7, 38.62)) where we calculated a personalized RxRSI dose for each patient. In 24% of cohort 2, RxRSI dose was lower than the prescribed dose by more than 10%, with median RxRSI at 47Gy. In 60% of the cohort, RxRSI dose was greater than prescribed dose by more than 10%, ranging from 62Gy to 95Gy. pGRT planning achieved both desired target coverage for RxRSI and normal tissue constraints for patients under 74Gy, with esophageal dose falling below clinically acceptable levels for 80% of plans at 88Gy and above. One-size fits-all RT dose prescription may be sub-optimal for some patients given the genomic heterogeneity observed. Using a clinically-validated model (RSI/GARD) we identify that in up to 84% of patients the prescribed dose and personalized RxRSI did not match, suggesting that there is a large opportunity to optimize dose with genomics. A novel pGRT function within a commercial system can effectively optimize plans to present a choice between a personalized RxRSI and SoC treatment plans for consideration. To our knowledge this is the first demonstration of a feasible clinical approach to precision medicine in radiation oncology.