Phase II Trial Assessing Toxicity of Personalized Response-Based Radiation Treatment in Patients with Locally Advanced Non-Small Cell Lung Cancer.

INTRODUCTION: Local failure rates after treatment for locally advanced non-small-cell lung cancer (NSCLC) remain high. Efforts to improve local control with uniform dose-escalation or dose-escalation to mid-treatment PET-avid residual disease have been limited by heightened toxicity. This trial aimed to refine response-based adaptive radiation (RT) and minimize toxicity by incorporating FDG-PET and V/Q SPECT imaging mid-treatment. METHODS: 47 patients with Stage IIA-III unresectable NSCLC were prospectively enrolled in this single-institution trial (NCT02492867). Patients received concurrent chemoradiation with personalized response-based adaptive RT over 30 fractions incorporating V/Q SPECT and FDG-PET. The first 21 fractions (46.2Gy at 2.2 Gy/fraction) were delivered to the tumor while minimizing dose to SPECT-defined functional lung. The plan was then adapted for the final 9 fractions (2.2-3.8Gy/fraction) up to a total of 80.4Gy, based on mid-treatment FDG-PET tumor response to escalate dose to residual tumor while minimizing dose to SPECT-defined functional lung. Non-progressing patients received consolidative carboplatin/paclitaxel or durvalumab. The primary endpoint of the study was ≥ grade 2 lung and esophageal toxicities. Secondary endpoints included time to local progression, tumor response, and overall survival. RESULTS: At one year post-treatment, the rates of grade 2 and grade 3 pneumonitis were 21.3% and 2.1%, respectively, with no difference in pneumonitis rates among patients who received and did not receive adjuvant durvalumab (p=0.74). While there were no grade 3 esophageal-related toxicities, 66.0% of patients experienced grade 2 esophagitis. 1- and 2-year local control rates were 94.5% (95% CI, 87.4% - 100%) and 87.5% (95% CI, 76.7% - 100%), respectively. Overall survival was 82.8% (95% CI, 72.6% -94.4%) at 1 year and 62.3% (95% CI, 49.6%-78.3%) at 2 years. CONCLUSIONS: Response-based adaptive dose-escalation accounting for tumor change and normal tissue function during treatment provided excellent local control, comparable toxicity to standard chemoradiation, and did not increase toxicity with adjuvant immunotherapy.

Artificial intelligence to unlock real-world evidence in clinical oncology: A primer on recent advances.

PURPOSE: Real world evidence is crucial to understanding the diffusion of new oncologic therapies, monitoring cancer outcomes, and detecting unexpected toxicities. In practice, real world evidence is challenging to collect rapidly and comprehensively, often requiring expensive and time-consuming manual case-finding and annotation of clinical text. In this Review, we summarise recent developments in the use of artificial intelligence to collect and analyze real world evidence in oncology. METHODS: We performed a narrative review of the major current trends and recent literature in artificial intelligence applications in oncology. RESULTS: Artificial intelligence (AI) approaches are increasingly used to efficiently phenotype patients and tumors at large scale. These tools also may provide novel biological insights and improve risk prediction through multimodal integration of radiographic, pathological, and genomic datasets. Custom language processing pipelines and large language models hold great promise for clinical prediction and phenotyping. CONCLUSIONS: Despite rapid advances, continued progress in computation, generalizability, interpretability, and reliability as well as prospective validation are needed to integrate AI approaches into routine clinical care and real-time monitoring of novel therapies.

ISG15/GRAIL1/CD3 axis influences survival of patients with esophageal adenocarcinoma.

Immunosuppression is a common feature of esophageal adenocarcinoma (EAC) and has been linked to poor overall survival (OS). We hypothesized that upstream factors might negatively influence CD3 levels and T cell activity, thus promoting immunosuppression and worse survival. We used clinical data and patient samples of those who progressed from Barrett's to dysplasia to EAC, investigated gene (RNA-Seq) and protein (tissue microarray) expression, and performed cell biology studies to delineate a pathway impacting CD3 protein stability that might influence EAC outcome. We showed that the loss of both CD3-ε expression and CD3+ T cell number correlated with worse OS in EAC. The gene related to anergy in lymphocytes isoform 1 (GRAIL1), which is the prominent isoform in EACs, degraded (ε, γ, δ) CD3s and inactivated T cells. In contrast, isoform 2 (GRAIL2), which is reduced in EACs, stabilized CD3s. Further, GRAIL1-mediated CD3 degradation was facilitated by interferon-stimulated gene 15 (ISG15), a ubiquitin-like protein. Consequently, the overexpression of a ligase-dead GRAIL1, ISG15 knockdown, or the overexpression of a conjugation-defective ISG15-leucine-arginine-glycine-glycine mutant could increase CD3 levels. Together, we identified an ISG15/GRAIL1/mutant p53 amplification loop negatively influencing CD3 levels and T cell activity, thus promoting immunosuppression in EAC.

Isoform alterations in the ubiquitination machinery impacting gastrointestinal malignancies.

The advancement of RNAseq and isoform-specific expression platforms has led to the understanding that isoform changes can alter molecular signaling to promote tumorigenesis. An active area in cancer research is uncovering the roles of ubiquitination on spliceosome assembly contributing to transcript diversity and expression of alternative isoforms. However, the effects of isoform changes on functionality of ubiquitination machineries (E1, E2, E3, E4, and deubiquitinating (DUB) enzymes) influencing onco- and tumor suppressor protein stabilities is currently understudied. Characterizing these changes could be instrumental in improving cancer outcomes via the identification of novel biomarkers and targetable signaling pathways. In this review, we focus on highlighting reported examples of direct, protein-coded isoform variation of ubiquitination enzymes influencing cancer development and progression in gastrointestinal (GI) malignancies. We have used a semi-automated system for identifying relevant literature and applied established systems for isoform categorization and functional classification to help structure literature findings. The results are a comprehensive snapshot of known isoform changes that are significant to GI cancers, and a framework for readers to use to address isoform variation in their own research. One of the key findings is the potential influence that isoforms of the ubiquitination machinery have on oncoprotein stability.

Potentiating the radiation-induced type I interferon antitumoral immune response by ATM inhibition in pancreatic cancer.

Radiotherapy induces a type I interferon-mediated (T1IFN-mediated) antitumoral immune response that we hypothesized could be potentiated by a first-in-class ataxia telangiectasia mutated (ATM) inhibitor, leading to enhanced innate immune signaling, T1IFN expression, and sensitization to immunotherapy in pancreatic cancer. We evaluated the effects of AZD1390 or a structurally related compound, AZD0156, on innate immune signaling and found that both inhibitors enhanced radiation-induced T1IFN expression via the POLIII/RIG-I/MAVS pathway. In immunocompetent syngeneic mouse models of pancreatic cancer, ATM inhibitor enhanced radiation-induced antitumoral immune responses and sensitized tumors to anti-PD-L1, producing immunogenic memory and durable tumor control. Therapeutic responses were associated with increased intratumoral CD8+ T cell frequency and effector function. Tumor control was dependent on CD8+ T cells, as therapeutic efficacy was blunted in CD8+ T cell-depleted mice. Adaptive immune responses to combination therapy provided systemic control of contralateral tumors outside of the radiation field. Taken together, we show that a clinical candidate ATM inhibitor enhances radiation-induced T1IFN, leading to both innate and subsequent adaptive antitumoral immune responses and sensitization of otherwise resistant pancreatic cancer to immunotherapy.

Defining Minimum Treatment Parameters of Ablative Radiation Therapy in Patients With Hepatocellular Carcinoma: An Expert Consensus.

PURPOSE: External beam radiation therapy (EBRT) is a highly effective treatment in select patients with hepatocellular carcinoma (HCC). However, the Barcelona Clinic Liver Cancer system does not recommend the use of EBRT in HCC due to a lack of sufficient evidence and intends to perform an individual patient level meta-analysis of ablative EBRT in this population. However, there are many types of EBRT described in the literature with no formal definition of what constitutes "ablative." Thus, we convened a group of international experts to provide consensus on the parameters that define ablative EBRT in HCC. METHODS AND MATERIALS: Fundamental parameters related to dose, fractionation, radiobiology, target identification, and delivery technique were identified by a steering committee to generate 7 Key Criteria (KC) that would define ablative EBRT for HCC. Using a modified Delphi (mDelphi) method, experts in the use of EBRT in the treatment of HCC were surveyed. Respondents were given 30 days to respond in round 1 of the mDelphi and 14 days to respond in round 2. A threshold of ≥70% was used to define consensus for answers to each KC. RESULTS: Of 40 invitations extended, 35 (88%) returned responses. In the first round, 3 of 7 KC reached consensus. In the second round, 100% returned responses and consensus was reached in 3 of the remaining 4 KC. The distribution of answers for one KC, which queried the a/b ratio of HCC, was such that consensus was not achieved. Based on this analysis, ablative EBRT for HCC was defined as a BED10 ≥80 Gy with daily imaging and multiphasic contrast used for target delineation. Treatment breaks (eg, for adaptive EBRT) are allowed, but the total treatment time should be ≤6 weeks. Equivalent dose when treating with protons should use a conversion factor of 1.1, but there is no single conversion factor for carbon ions. CONCLUSIONS: Using a mDelphi method assessing expert opinion, we provide the first consensus definition of ablative EBRT for HCC. Empirical data are required to define the a/b of HCC.

GTP Signaling Links Metabolism, DNA Repair, and Responses to Genotoxic Stress.

How cell metabolism regulates DNA repair is incompletely understood. Here, we define a GTP-mediated signaling cascade that links metabolism to DNA repair and has significant therapeutic implications. GTP, but not other nucleotides, regulates the activity of Rac1, a guanine nucleotide-binding protein, which promotes the dephosphorylation of serine 323 on Abl-interactor 1 (Abi-1) by protein phosphatase 5 (PP5). Dephosphorylated Abi-1, a protein previously not known to activate DNA repair, promotes nonhomologous end joining. In patients and mouse models of glioblastoma, Rac1 and dephosphorylated Abi-1 mediate DNA repair and resistance to standard-of-care genotoxic treatments. The GTP-Rac1-PP5-Abi-1 signaling axis is not limited to brain cancer, as GTP supplementation promotes DNA repair and Abi-1-S323 dephosphorylation in nonmalignant cells and protects mouse tissues from genotoxic insult. This unexpected ability of GTP to regulate DNA repair independently of deoxynucleotide pools has important implications for normal physiology and cancer treatment. SIGNIFICANCE: A newly described GTP-dependent signaling axis is an unexpected link between nucleotide metabolism and DNA repair. Disrupting this pathway can overcome cancer resistance to genotoxic therapy while augmenting it can mitigate genotoxic injury of normal tissues. This article is featured in Selected Articles from This Issue, p. 5.

Predictors of Acute and Late Toxicity in Patients Receiving Chemoradiation for Unresectable Pancreatic Cancer.

Purpose: Patients with pancreatic cancer undergoing chemoradiation therapy may experience acute and chronic side effects. We conducted an exploratory analysis of patients with locally advanced pancreatic cancer (LAPC) undergoing definitive chemoradiation to identify factors influencing the occurrence of gastrointestinal (GI) bleeding, short-term radiation side effects, patterns of failure, and survival. Methods and Materials: Under an institutional review board-approved protocol, we retrospectively studied patients with LAPC treated with chemoradiation. Statistical models were used to test associations between clinical characteristics and outcomes, including upper GI bleeding, radiation treatment breaks, and weight loss during therapy. Results: Between 1999 and 2012, 211 patients were treated with radiation for pancreatic cancer. All patients received concurrent chemotherapy with either gemcitabine (174) or 5-fluorouracil (27), and 67 received intensity modulated radiation therapy (IMRT). Overall, 18 patients experienced an upper GI bleed related to treatment, with 70% of bleeds occurring in the stomach or duodenum, and among those patients, 11 (61%) patients had a pancreatic head tumor and 17 (94%) patients had a metallic biliary stent. IMRT was associated with decreased risk of postradiation nausea (odds ratio, 0.27 [0.11, 0.67], P = .006) compared with 3-dimensional conformal radiation. Regarding long-term toxicities, patients with a metallic biliary stent at the time of radiation therapy were at a significantly higher risk of developing upper GI bleeding (unadjusted hazard ratio [HR], 15.41 [2.02, 117.42], P = .008), even after controlling for radiation treatment modality and prescribed radiation dose (adjusted HR, 17.38 [2.26, 133.58], P = .006). Furthermore, biliary stent placement was associated with a higher risk of death (HR, 1.99 [1.41, 2.83], P < .001) after adjusting for demographic, treatment-related, and patient-related variables. Conclusions: Metallic biliary stents may be associated with an increased risk of upper GI bleeding and mortality. Furthermore, IMRT was associated with less nausea and short-term toxicity compared with 3-dimensional conformal therapy.

Rewiring of cortical glucose metabolism fuels human brain cancer growth.

The brain avidly consumes glucose to fuel neurophysiology. Cancers of the brain, such as glioblastoma (GBM), lose aspects of normal biology and gain the ability to proliferate and invade healthy tissue. How brain cancers rewire glucose utilization to fuel these processes is poorly understood. Here we perform infusions of 13 C-labeled glucose into patients and mice with brain cancer to define the metabolic fates of glucose-derived carbon in tumor and cortex. By combining these measurements with quantitative metabolic flux analysis, we find that human cortex funnels glucose-derived carbons towards physiologic processes including TCA cycle oxidation and neurotransmitter synthesis. In contrast, brain cancers downregulate these physiologic processes, scavenge alternative carbon sources from the environment, and instead use glucose-derived carbons to produce molecules needed for proliferation and invasion. Targeting this metabolic rewiring in mice through dietary modulation selectively alters GBM metabolism and slows tumor growth. Significance: This study is the first to directly measure biosynthetic flux in both glioma and cortical tissue in human brain cancer patients. Brain tumors rewire glucose carbon utilization away from oxidation and neurotransmitter production towards biosynthesis to fuel growth. Blocking these metabolic adaptations with dietary interventions slows brain cancer growth with minimal effects on cortical metabolism.

Evolution of Response-Based Radiotherapy for Hepatocellular Cancer.

ABSTRACT: Stereotactic body radiation therapy has emerged as a safe and effective treatment modality for properly selected hepatocellular cancer (HCC) patients with normal liver function. However, many HCC patients have reduced baseline liver function due to underlying cirrhosis or prior liver-directed therapies. Therefore, because of the increased risk of hepatotoxicity, the use of stereotactic body radiation therapy for patients with reduced liver function has been approached with caution. Individualized, response-based radiotherapy incorporates models, imaging tools, and biomarkers that determine the dose-response relationship of the liver before, during, and after treatment and has been useful in reducing the likelihood of liver damage without sacrificing tumor control. This review discusses the evolution of response-based radiotherapy for HCC and highlights areas for further investigation.

A Phase II Study of Optimized Individualized Adaptive Radiotherapy for Hepatocellular Carcinoma.

PURPOSE: We hypothesized that optimizing the utility of stereotactic body radiotherapy (SBRT) based on the individual patient's probability for tumor control and risk of liver injury would decrease toxicity without sacrificing local control in patients with impaired liver function or tumors not amenable to thermal ablation. PATIENTS AND METHODS: Patients with Child-Pugh (CP) A to B7 liver function with aggregate tumor size >3.5 cm, or CP ≥ B8 with any size tumor were prospectively enrolled on an Institutional Review Board-approved phase II clinical trial to undergo SBRT with baseline and midtreatment dose optimization using a quantitative, individualized utility-based analysis. Primary endpoints were change in CP score of ≥2 points within 6 months and local control. Protocol-treated patients were compared with patients receiving conventional SBRT at another cancer center using overlap weighting. RESULTS: A total of 56 patients with 80 treated tumors were analyzed with a median follow-up of 11.2 months. Two-year cumulative incidence of local progression was 6.4% [95% confidence interval (CI, 2.4-13.4)]. Twenty-one percent of patients experienced treatment-related toxicity within 6 months, which is similar to the rate for SBRT in patients with CP A liver function. An analysis using overlap weighting revealed similar local control [HR, 0.69; 95% CI (0.25-1.91); P = 0.48] and decreased toxicity [OR, 0.26; 95% CI (0.07-0.99); P = 0.048] compared with conventional SBRT. CONCLUSIONS: Treatment of individuals with impaired liver function or tumors not amenable to thermal ablation with a treatment paradigm designed to optimize utility may decrease treatment-related toxicity while maintaining tumor control.

Is Apparent Diffusion Coefficient Established as an Imaging Biomarker for Stereotactic Body Radiation Therapy Assessment in Hepatocellular Carcinoma?

ABSTRACT: In this article, as part of this special issue on biomarkers of early response, we review currently available reports regarding magnetic resonance imaging apparent diffusion coefficient (ADC) changes in hepatocellular carcinoma (HCC) in response to stereotactic body radiation therapy. We compare diffusion image acquisition, ADC analysis, methods for HCC response assessment, and statistical methods for prediction of local tumor progression by ADC metrics. We discuss the pros and cons of these studies. Following detailed analyses of existing investigations, we cannot conclude that ADC is established as an imaging biomarker for stereotactic body radiation therapy assessment in HCC.

Does Vascular Collapse Occur After Treatment of Hepatocellular Cancer With Stereotactic Body Radiation Therapy?

There is debate about why stereotactic body radiation therapy (SBRT) produces superior control of hepatocellular cancer (HCC) compared to fractionated treatment. Both preclinical and clinical evidence has been presented to support a "classic" biological explanation: the greater BED of SBRT produces more DNA damage and tumor cell kill. More recently, preclinical evidence has supported the concept of a "new biology", particularly radiation-induced vascular collapse, which increases hypoxia and free radical activation. This is hypothesized to cause much greater tumor cell death than was produced by the initial radiation-induced DNA damage to the tumor. We decided to investigate if vascular collapse occurs after standard SBRT for patients with HCC. Eight patients with 10 lesions underwent dynamic contrast enhanced MRI at the time of simulation and either 48 or 96 hours after the first fraction. Only three of 10 tumors showed a decrease in blood flow. These findings suggest that vascular collapse does not typically occur after SBRT for HCC.

Thermal ablation compared to stereotactic body radiation therapy for hepatocellular carcinoma: A multicenter retrospective comparative study.

BACKGROUND AIMS: Early-stage HCC can be treated with thermal ablation or stereotactic body radiation therapy (SBRT). We retrospectively compared local progression, mortality, and toxicity among patients with HCC treated with ablation or SBRT in a multicenter, US cohort. APPROACH RESULTS: We included adult patients with treatment-naïve HCC lesions without vascular invasion treated with thermal ablation or SBRT per individual physician or institutional preference from January 2012 to December 2018. Outcomes included local progression after a 3-month landmark period assessed at the lesion level and overall survival at the patient level. Inverse probability of treatment weighting was used to account for imbalances in treatment groups. The Cox proportional hazard modeling was used to compare progression and overall survival, and logistic regression was used for toxicity. There were 642 patients with 786 lesions (median size: 2.1 cm) treated with ablation or SBRT. In adjusted analyses, SBRT was associated with a reduced risk of local progression compared to ablation (aHR 0.30, 95% CI: 0.15-0.60). However, SBRT-treated patients had an increased risk of liver dysfunction at 3 months (absolute difference 5.5%, aOR 2.31, 95% CI: 1.13-4.73) and death (aHR 2.04, 95% CI: 1.44-2.88, p < 0.0001). CONCLUSIONS: In this multicenter study of patients with HCC, SBRT was associated with a lower risk of local progression compared to thermal ablation but higher all-cause mortality. Survival differences may be attributable to residual confounding, patient selection, or downstream treatments. These retrospective real-world data help guide treatment decisions while demonstrating the need for a prospective clinical trial.

Efficacy of a Second Course of Radiation for Patients With Metachronous Hepatocellular Carcinoma.

PURPOSE: Liver-directed radiation therapy is an effective treatment for hepatocellular carcinoma (HCC), but metachronous lesions develop outside the irradiated field in >50% of patients. We hypothesized that irradiation of these new lesions would produce an outcome like that of patients receiving a first course (C1) of treatment. METHODS AND MATERIALS: We included patients with HCC who received a second course (C2) of radiation therapy >1 month after C1. Toxicity was defined as Child-Pugh score increase ≥2 within 6 months posttreatment (binary model) and as the change in albumin-bilirubin during the year after treatment (longitudinal model). Overall survival (OS) and local failure (LF) were captured at the patient and lesion level, respectively; both were summarized with Kaplan-Meier estimates. Predictors of toxicity and OS were assessed using generalized linear mixed and Cox regression models, respectively. RESULTS: Of 340 patients with HCC, 47 underwent irradiation for metachronous HCC, receiving similar prescription dose in C1/C2. Median follow-up was 17 months after C1 and 15 months after C2. Twenty-two percent of patients experienced toxicity after C1, and 25% experienced toxicity after C2. Worse baseline albumin-bilirubin predicted toxicity in both binary (odds ratio, 2.40; 95% CI, 1.46-3.94; P = .0005) and longitudinal models (P < .005). Two-year LF rate was 11.2% after C1 and 8.3% after C2; tumor dose (hazard ratio [HR], 0.982; 95% CI, 0.969-0.995; P = .007) and tumor size (HR, 1.135; 95% CI, 1.068-1.206; P < .005) predicted LF. Two-year OS was 46.0% after C1 and 42.6% after C2; tumor dose (HR, 0.986; 95% CI, 0.979-0.992; P < .005) and tumor size (HR, 1.049; 95% CI, 1.010-1.088; P = .0124) predicted OS. Reirradiation was not associated with toxicity (P > .7), LF (P = .79), or OS (P = .39). CONCLUSIONS: In this largest series in the Western hemisphere, we demonstrate that irradiation for metachronous HCC offers low rates of LF with acceptable toxicity and OS like that of patients receiving a C1. These findings support judicious selection of patients for reirradiation in metachronous HCC.

The adapter protein FADD provides an alternate pathway for entry into the cell cycle by regulating APC/C-Cdh1 E3 ubiquitin ligase activity.

The E3 ubiquitin ligase APC/C-Cdh1 maintains the G0/G1 state, and its inactivation is required for cell cycle entry. We reveal a novel role for Fas-associated protein with death domain (FADD) in the cell cycle through its function as an inhibitor of APC/C-Cdh1. Using real-time, single-cell imaging of live cells combined with biochemical analysis, we demonstrate that APC/C-Cdh1 hyperactivity in FADD-deficient cells leads to a G1 arrest despite persistent mitogenic signaling through oncogenic EGFR/KRAS. We further show that FADDWT interacts with Cdh1, while a mutant lacking a consensus KEN-box motif (FADDKEN) fails to interact with Cdh1 and results in a G1 arrest due to its inability to inhibit APC/C-Cdh1. Additionally, enhanced expression of FADDWT but not FADDKEN, in cells arrested in G1 upon CDK4/6 inhibition, leads to APC/C-Cdh1 inactivation and entry into the cell cycle in the absence of retinoblastoma protein phosphorylation. FADD's function in the cell cycle requires its phosphorylation by CK1α at Ser-194 which promotes its nuclear translocation. Overall, FADD provides a CDK4/6-Rb-E2F-independent "bypass" mechanism for cell cycle entry and thus a therapeutic opportunity for CDK4/6 inhibitor resistance.

GTP signaling links metabolism, DNA repair, and responses to genotoxic stress.

How cell metabolism regulates DNA repair is incompletely understood. Here, we define a GTP-mediated signaling cascade that links metabolism to DNA repair and has significant therapeutic implications. GTP, but not other nucleotides, regulates the activity of Rac1, a G protein, that promotes the dephosphorylation of serine 323 on Abl-interactor 1 (Abi-1) by protein phosphatase 5 (PP5). Dephosphorylated Abi-1, a protein previously not known to activate DNA repair, promotes non-homologous end joining. In patients and mouse models of glioblastoma, Rac1 and dephosphorylated Abi-1 mediate DNA repair and resistance to standard of care genotoxic treatments. The GTP-Rac1-PP5-Abi-1 signaling axis is not limited to brain cancer, as GTP supplementation promotes DNA repair and Abi-1-S323 dephosphorylation in non-malignant cells and protects mouse tissues from genotoxic insult. This unexpected ability of GTP to regulate DNA repair independently of deoxynucleotide pools has important implications for normal physiology and cancer treatment.

A clinical decision support system for AI-assisted decision-making in response-adaptive radiotherapy (ARCliDS).

Involvement of many variables, uncertainty in treatment response, and inter-patient heterogeneity challenge objective decision-making in dynamic treatment regime (DTR) in oncology. Advanced machine learning analytics in conjunction with information-rich dense multi-omics data have the ability to overcome such challenges. We have developed a comprehensive artificial intelligence (AI)-based optimal decision-making framework for assisting oncologists in DTR. In this work, we demonstrate the proposed framework to Knowledge Based Response-Adaptive Radiotherapy (KBR-ART) applications by developing an interactive software tool entitled Adaptive Radiotherapy Clinical Decision Support (ARCliDS). ARCliDS is composed of two main components: Artifcial RT Environment (ARTE) and Optimal Decision Maker (ODM). ARTE is designed as a Markov decision process and modeled via supervised learning. Given a patient's pre- and during-treatment information, ARTE can estimate treatment outcomes for a selected daily dosage value (radiation fraction size). ODM is formulated using reinforcement learning and is trained on ARTE. ODM can recommend optimal daily dosage adjustments to maximize the tumor local control probability and minimize the side effects. Graph Neural Networks (GNN) are applied to exploit the inter-feature relationships for improved modeling performance and a novel double GNN architecture is designed to avoid nonphysical treatment response. Datasets of size 117 and 292 were available from two clinical trials on adaptive RT in non-small cell lung cancer (NSCLC) patients and adaptive stereotactic body RT (SBRT) in hepatocellular carcinoma (HCC) patients, respectively. For training and validation, dense data with 297 features were available for 67 NSCLC patients and 110 features for 71 HCC patients. To increase the sample size for ODM training, we applied Generative Adversarial Networks to generate 10,000 synthetic patients. The ODM was trained on the synthetic patients and validated on the original dataset. We found that, Double GNN architecture was able to correct the nonphysical dose-response trend and improve ARCliDS recommendation. The average root mean squared difference (RMSD) between ARCliDS recommendation and reported clinical decisions using double GNNs were 0.61 [0.03] Gy/frac (mean [sem]) for adaptive RT in NSCLC patients and 2.96 [0.42] Gy/frac for adaptive SBRT HCC compared to the single GNN's RMSDs of 0.97 [0.12] Gy/frac and 4.75 [0.16] Gy/frac, respectively. Overall, For NSCLC and HCC, ARCliDS with double GNNs was able to reproduce 36% and 50% of the good clinical decisions (local control and no side effects) and improve 74% and 30% of the bad clinical decisions, respectively. In conclusion, ARCliDS is the first web-based software dedicated to assist KBR-ART with multi-omics data. ARCliDS can learn from the reported clinical decisions and facilitate AI-assisted clinical decision-making for improving the outcomes in DTR.

Deep learning prediction of post-SBRT liver function changes and NTCP modeling in hepatocellular carcinoma based on DGAE-MRI.

BACKGROUND: Stereotactic body radiation therapy (SBRT) produces excellent local control for patients with hepatocellular carcinoma (HCC). However, the risk of toxicity for normal liver tissue is still a limiting factor. Normal tissue complication probability (NTCP) models have been proposed to estimate the toxicity with the assumption of uniform liver function distribution, which is not optimal. With more accurate regional liver functional imaging available for individual patient, we can improve the estimation and be more patient-specific. PURPOSE: To develop normal tissue complication probability (NTCP) models using pre-/during-treatment (RT) dynamic Gadoxetic Acid-enhanced (DGAE) MRI for adaptation of RT in a patient-specific manner in hepatocellular cancer (HCC) patients who receive SBRT. METHODS: 24 of 146 HCC patients who received SBRT underwent DGAE MRI. Physical doses were converted into EQD2 for analysis. Voxel-by-voxel quantification of the contrast uptake rate (k1) from DGAE-MRI was used to quantify liver function. A logistic dose-response model was used to estimate the fraction of liver functional loss, and NTCP was estimated using the cumulative functional reserve model for changes in Child-Pugh (C-P) scores. Model parameters were calculated using maximum-likelihood estimations. During-RT liver functional maps were predicted from dose distributions and pre-RT k1 maps with a conditional Wasserstein generative adversarial network (cWGAN). Imaging prediction quality was assessed using root-mean-square error (RMSE) and structural similarity (SSIM) metrics. The dose-response and NTCP were fit on both original and cWGAN predicted images and compared using a Wilcoxon signed-rank test. RESULTS: Logistic dose response models for changes in k1 yielded D50 of 35.2 (95% CI: 26.7-47.5) Gy and k of 0.62 (0.49-0.75) for the whole population. The high baseline ALBI (poor liver function) subgroup showed a significantly smaller D50 of 11.7 (CI: 9.06-15.4) Gy and larger k of 0.96 (CI: 0.74-1.22) compared to a low baseline ALBI (good liver function) subgroup of 54.8 (CI: 38.3-79.1) Gy and 0.59 (CI: 0.48-0.74), with p-values of < 0.001 and = 0.008, respectively, which indicates higher radiosensitivity for the worse baseline liver function cohort. Subset analyses were also performed for high/low baseline CP subgroups. The corresponding NTCP models showed good agreement for the fit parameters between cWGAN predicted and the ground-truth during-RT images with no statistical differences for low ALBI subgroup. CONCLUSIONS: NTCP models which incorporate voxel-wise functional information from DGAE-MRI k1 maps were successfully developed and feasibility was demonstrated in a small patient cohort. cWGAN predicted functional maps show promise for estimating localized patient-specific response to RT and warrant further validation in a larger patient cohort.

Metabolomic Profiles of Human Glioma Inform Patient Survival.

Aims: Targeting tumor metabolism may improve the outcomes for patients with glioblastoma (GBM). To further preclinical efforts targeting metabolism in GBM, we tested the hypothesis that brain tumors can be stratified into distinct metabolic groups with different patient outcomes. Therefore, to determine if tumor metabolites relate to patient survival, we profiled the metabolomes of human gliomas and correlated metabolic information with clinical data. Results: We found that isocitrate dehydrogenase-wildtype (IDHwt) GBMs are metabolically distinguishable from IDH mutated (IDHmut) astrocytomas and oligodendrogliomas. Survival of patients with IDHmut gliomas was expectedly more favorable than those with IDHwt GBM, and metabolic signatures can stratify IDHwt GBMs subtypes with varying prognoses. Patients whose GBMs were enriched in amino acids had improved survival, while those whose tumors were enriched for nucleotides, redox molecules, and lipid metabolites fared more poorly. These findings were recapitulated in validation cohorts using both metabolomic and transcriptomic data. Innovation: Our results suggest the existence of metabolic subtypes of GBM with differing prognoses, and further support the concept that metabolism may drive the aggressiveness of human gliomas. Conclusions: Our data show that metabolic signatures of human gliomas can inform patient survival. These findings may be used clinically to tailor novel metabolically targeted agents for GBM patients with different metabolic phenotypes. Antioxid. Redox Signal. 39, 942-956.