Deep segmentation networks predict survival of non-small cell lung cancer

Non-small-cell lung cancer (NSCLC) represents approximately 80-85% of lung cancer diagnoses and is the leading cause of cancer-related death worldwide. Recent studies indicate that image-based radiomics features from positron emission tomography/computed tomography (PET/CT) images have predictive power for NSCLC outcomes. To this end, easily calculated functional features such as the maximum and the mean of standard uptake value (SUV) and total lesion glycolysis (TLG) are most commonly used for NSCLC prognostication, but their prognostic value remains controversial. Meanwhile, convolutional neural networks (CNN) are rapidly emerging as a new method for cancer image analysis, with significantly enhanced predictive power compared to hand-crafted radiomics features. Here we show that CNNs trained to perform the tumor segmentation task, with no other information than physician contours, identify a rich set of survival-related image features with remarkable prognostic value. In a retrospective study on pre-treatment PET-CT images of 96 NSCLC patients before stereotactic-body radiotherapy (SBRT), we found that the CNN segmentation algorithm (U-Net) trained for tumor segmentation in PET and CT images, contained features having strong correlation with 2- and 5-year overall and disease-specific survivals. The U-Net algorithm has not seen any other clinical information (e.g. survival, age, smoking history, etc.) than the images and the corresponding tumor contours provided by physicians. In addition, we observed the same trend by validating the U-Net features against an extramural data set provided by Stanford Cancer Institute. Furthermore, through visualization of the U-Net, we also found convincing evidence that the regions of metastasis and recurrence appear to match with the regions where the U-Net features identified patterns that predicted higher likelihoods of death. We anticipate our findings will be a starting point for more sophisticated non-intrusive patient specific cancer prognosis determination. For example, the deep learned PET/CT features can not only predict survival but also visualize high-risk regions within or adjacent to the primary tumor and hence potentially impact therapeutic outcomes by optimal selection of therapeutic strategy or first-line therapy adjustment.

Estimating survival for renal cell carcinoma patients with brain metastases: an update of the Renal Graded Prognostic Assessment tool

Background

Brain metastases are a common complication of renal cell carcinoma (RCC). Our group previously published the Renal Graded Prognostic Assessment (GPA) tool. In our prior RCC study (n = 286, 1985-2005), we found marked heterogeneity and variation in outcomes. In our recent update in a larger, more contemporary cohort, we identified additional significant prognostic factors. The purpose of this study is to update the original Renal-GPA based on the newly identified prognostic factors.

Methods

A multi-institutional retrospective institutional review board-approved database of 711 RCC patients with new brain metastases diagnosed from January 1, 2006 to December 31, 2015 was created. Clinical parameters and treatment were correlated with survival. A revised Renal GPA index was designed by weighting the most significant factors in proportion to their hazard ratios and assigning scores such that the patients with the best and worst prognoses would have a GPA of 4.0 and 0.0, respectively.

Results

The 4 most significant factors were Karnofsky performance status, number of brain metastases, extracranial metastases, and hemoglobin. The overall median survival was 12 months. Median survival for GPA groups 0-1.0, 1.5-2.0, 2.5-3, and 3.5-4.0 (% n = 25, 27, 30 and 17) was 4, 12, 17, and 35 months, respectively.

Conclusion

The updated Renal GPA is a user-friendly tool that will help clinicians and patients better understand prognosis, individualize clinical decision making and treatment selection, provide a means to compare retrospective literature, and provide more robust stratification of future clinical trials in this heterogeneous population. To simplify use of this tool in daily practice, a free online application is available at brainmetgpa.com.

Phase IIb, Randomized, Double-Blind Trial of GC4419 Versus Placebo to Reduce Severe Oral Mucositis Due to Concurrent Radiotherapy and Cisplatin For Head and Neck Cancer

PURPOSE

Oral mucositis (OM) remains a common, debilitating toxicity of radiation therapy (RT) for head and neck cancer. The goal of this phase IIb, multi-institutional, randomized, double-blind trial was to compare the efficacy and safety of GC4419, a superoxide dismutase mimetic, with placebo to reduce the duration, incidence, and severity of severe OM (SOM).

PATIENTS AND METHODS

A total of 223 patients (from 44 institutions) with locally advanced oral cavity or oropharynx cancer planned to be treated with definitive or postoperative intensity-modulated RT (IMRT; 60 to 72 Gy [≥ 50 Gy to two or more oral sites]) plus cisplatin (weekly or every 3 weeks) were randomly assigned to receive 30 mg (n = 73) or 90 mg (n = 76) of GC4419 or to receive placebo (n = 74) by 60-minute intravenous administration before each IMRT fraction. WHO grade of OM was assessed biweekly during IMRT and then weekly for up to 8 weeks after IMRT. The primary endpoint was duration of SOM tested for each active dose level versus placebo (intent-to-treat population, two-sided α of .05). The National Cancer Institute Common Terminology Criteria for Adverse Events, version 4.03, was used for adverse event grading.

RESULTS

Baseline patient and tumor characteristics as well as treatment delivery were balanced. With 90 mg GC4419 versus placebo, SOM duration was significantly reduced (P = .024; median, 1.5 v 19 days). SOM incidence (43% v 65%; P = .009) and severity (grade 4 incidence, 16% v 30%; P = .045) also were improved. Intermediate improvements were seen with the 30-mg dose. Safety was comparable across arms, with no significant GC4419-specific toxicity nor increase of known toxicities of IMRT plus cisplatin. The 2-year follow-up for tumor outcomes is ongoing.

CONCLUSION

GC4419 at a dose of 90 mg produced a significant, clinically meaningful reduction of SOM duration, incidence, and severity with acceptable safety. A phase III trial (ROMAN; ClinicalTrials.gov identifier: NCT03689712) has begun.

First-in-Human Phase I Clinical Trial of Pharmacologic Ascorbate Combined with Radiation and Temozolomide for Newly Diagnosed Glioblastoma

PURPOSE

Standard treatment for glioblastoma (GBM) includes surgery, radiation therapy (RT), and temozolomide (TMZ), yielding a median overall survival (OS) of approximately 14 months. Preclinical models suggest that pharmacologic ascorbate (P-AscH^-) enhances RT/TMZ antitumor effect in GBM. We evaluated the safety of adding P-AscH^- to standard RT/TMZ therapy.

PATIENTS AND METHODS

This first-in-human trial was divided into an RT phase (concurrent RT/TMZ/P-AscH^-) and an adjuvant (ADJ) phase (post RT/TMZ/P-AscH^- phase). Eight P-AscH^- dose cohorts were evaluated in the RT phase until targeted plasma ascorbate levels were achieved (≥20 mmol/L). In the ADJ phase, P-AscH^- doses were escalated in each subject at each cycle until plasma concentrations were ≥20 mmol/L. P-AscH^- was infused 3 times weekly during the RT phase and 2 times weekly during the ADJ phase continuing for six cycles or until disease progression. Adverse events were quantified by CTCAE (v4.03).

RESULTS

Eleven subjects were evaluable. No dose-limiting toxicities occurred. Observed toxicities were consistent with historical controls. Adverse events related to study drug were dry mouth and chills. Targeted ascorbate plasma levels of 20 mmol/L were achieved in the 87.5 g cohort; diminishing returns were realized in higher dose cohorts. Median progression-free survival (PFS) was 9.4 months and median OS was 18 months. In subjects with undetectable MGMT promoter methylation (n = 8), median PFS was 10 months and median OS was 23 months.

CONCLUSIONS

P-AscH^-/RT/TMZ is safe with promising clinical outcomes warranting further investigation.

Estimating survival in patients with gastrointestinal cancers and brain metastases: An update of the graded prognostic assessment for gastrointestinal cancers (GI-GPA)

Background

Patients with gastrointestinal cancers and brain metastases (BM) represent a unique and heterogeneous population. Our group previously published the Diagnosis-Specific Graded Prognostic Assessment (DS-GPA) for patients with GI cancers (GI-GPA) (1985-2007, n = 209). The purpose of this study is to update the GI-GPA based on a larger contemporary database.

Methods

An IRB-approved consortium database analysis was performed using a multi-institutional (18), multi-national (3) cohort of 792 patients with gastrointestinal (GI) cancers, with newly-diagnosed BM diagnosed between 1/1/2006 and 12/31/2017. Survival was measured from date of first treatment for BM. Multiple Cox regression was used to select and weight prognostic factors in proportion to their hazard ratios. These factors were incorporated into the updated GI-GPA.

Results

Median survival (MS) varied widely by primary site and other prognostic factors. Four significant factors (KPS, age, extracranial metastases and number of BM) were used to formulate the updated GI-GPA. Overall MS for this cohort remains poor; 8 months. MS by GPA was 3, 7, 11 and 17 months for GPA 0-1, 1.5-2, 2.5-3.0 and 3.5-4.0, respectively. >30% present in the worst prognostic group (GI-GPA of ≤1.0).

Conclusions

Brain metastases are not uncommon in GI cancer patients and MS varies widely among them. This updated GI-GPA index improves our ability to estimate survival for these patients and will be useful for therapy selection, end-of-life decision-making and stratification for future clinical trials. A user-friendly, free, on-line app to calculate the GPA score and estimate survival for an individual patient is available at brainmetgpa.com.

Machine learning with the TCGA-HNSC dataset: improving usability by addressing inconsistency, sparsity, and high-dimensionality

Background

In the era of precision oncology and publicly available datasets, the amount of information available for each patient case has dramatically increased. From clinical variables and PET-CT radiomics measures to DNA-variant and RNA expression profiles, such a wide variety of data presents a multitude of challenges. Large clinical datasets are subject to sparsely and/or inconsistently populated fields. Corresponding sequencing profiles can suffer from the problem of high-dimensionality, where making useful inferences can be difficult without correspondingly large numbers of instances. In this paper we report a novel deployment of machine learning techniques to handle data sparsity and high dimensionality, while evaluating potential biomarkers in the form of unsupervised transformations of RNA data. We apply preprocessing, MICE imputation, and sparse principal component analysis (SPCA) to improve the usability of more than 500 patient cases from the TCGA-HNSC dataset for enhancing future oncological decision support for Head and Neck Squamous Cell Carcinoma (HNSCC).

Results

Imputation was shown to improve prognostic ability of sparse clinical treatment variables. SPCA transformation of RNA expression variables reduced runtime for RNA-based models, though changes to classifier performance were not significant. Gene ontology enrichment analysis of gene sets associated with individual sparse principal components (SPCs) are also reported, showing that both high- and low-importance SPCs were associated with cell death pathways, though the high-importance gene sets were found to be associated with a wider variety of cancer-related biological processes.

Conclusions

MICE imputation allowed us to impute missing values for clinically informative features, improving their overall importance for predicting two-year recurrence-free survival by incorporating variance from other clinical variables. Dimensionality reduction of RNA expression profiles via SPCA reduced both computation cost and model training/evaluation time without affecting classifier performance, allowing researchers to obtain experimental results much more quickly. SPCA simultaneously provided a convenient avenue for consideration of biological context via gene ontology enrichment analysis.

Survival and prognostic factors in patients with gastrointestinal cancers and brain metastases: have we made progress?

The literature describing the prognosis of patients with gastrointestinal (GI) cancers and brain metastases (BM) is sparse. Our group previously published a prognostic index, the Graded Prognostic Assessment (GPA) for GI cancer patients with BM, based on 209 patients diagnosed from 1985-2005. The purpose of this analysis is to identify prognostic factors for GI cancer patients with newly diagnosed BM in a larger contemporary cohort. A multi-institutional retrospective IRB-approved database of 792 GI cancer patients with new BM diagnosed from 1/1/2006 to 12/31/2016 was created. Demographic data, clinical parameters, and treatment were correlated with survival and time from primary diagnosis to BM (TPDBM). Kaplan-Meier median survival (MS) estimates were calculated and compared with log-rank tests. The MS from time of first treatment for BM for the prior and current cohorts were 5 and 8 months, respectively (P < 0.001). Eight prognostic factors (age, stage, primary site, resection of primary tumor, Karnofsky Performance Status (KPS), extracranial metastases, number of BM and Hgb were found to be significant for survival, in contrast to only one (KPS) in the prior cohort. In this cohort, the most common primary sites were rectum (24%) and esophagus (23%). Median TPDBM was 22 months. Notably, 37% (267/716) presented with poor prognosis (GPA 0-1.0). Although little improvement in overall survival in this cohort has been achieved in recent decades, survival varies widely and multiple new prognostic factors were identified. Future work will translate these factors into a prognostic index to facilitate clinical decision-making and stratification of future clinical trials.

FDG PET based prediction of response in head and neck cancer treatment: Assessment of new quantitative imaging features

Introduction

18 F-fluorodeoxyglucose (FDG) positron emission tomography (PET) is now a standard diagnostic imaging test performed in patients with head and neck cancer for staging, re-staging, radiotherapy planning, and outcome assessment. Currently, quantitative analysis of FDG PET scans is limited to simple metrics like maximum standardized uptake value, metabolic tumor volume, or total lesion glycolysis, which have limited predictive value. The goal of this work was to assess the predictive potential of new (i.e., nonstandard) quantitative imaging features on head and neck cancer outcome.

Methods

This retrospective study analyzed fifty-eight pre- and post-treatment FDG PET scans of patients with head and neck squamous cell cancer to calculate five standard and seventeen new features at baseline and post-treatment. Cox survival regression was used to assess the predictive potential of each quantitative imaging feature on disease-free survival.

Results

Analysis showed that the post-treatment change of the average tracer uptake in the rim background region immediately adjacent to the tumor normalized by uptake in the liver represents a novel PET feature that is associated with disease-free survival (HR 1.95; 95% CI 1.27, 2.99) and has good discriminative performance (c index 0.791).

Conclusion

The reported findings define a promising new direction for quantitative imaging biomarker research in head and neck squamous cell cancer and highlight the potential role of new radiomics features in oncology decision making as part of precision medicine.

The Use of Quantitative Imaging in Radiation Oncology: A Quantitative Imaging Network (QIN) Perspective

Modern radiation therapy is delivered with great precision, in part by relying on high-resolution multidimensional anatomic imaging to define targets in space and time. The development of quantitative imaging (QI) modalities capable of monitoring biologic parameters could provide deeper insight into tumor biology and facilitate more personalized clinical decision-making. The Quantitative Imaging Network (QIN) was established by the National Cancer Institute to advance and validate these QI modalities in the context of oncology clinical trials. In particular, the QIN has significant interest in the application of QI to widen the therapeutic window of radiation therapy. QI modalities have great promise in radiation oncology and will help address significant clinical needs, including finer prognostication, more specific target delineation, reduction of normal tissue toxicity, identification of radioresistant disease, and clearer interpretation of treatment response. Patient-specific QI is being incorporated into radiation treatment design in ways such as dose escalation and adaptive replanning, with the intent of improving outcomes while lessening treatment morbidities. This review discusses the current vision of the QIN, current areas of investigation, and how the QIN hopes to enhance the integration of QI into the practice of radiation oncology.

Pharmacologic Ascorbate Reduces Radiation-Induced Normal Tissue Toxicity and Enhances Tumor Radiosensitization in Pancreatic Cancer

: Chemoradiation therapy is the mainstay for treatment of locally advanced, borderline resectable pancreatic cancer. Pharmacologic ascorbate (P-AscH^-, i.e., intravenous infusions of ascorbic acid, vitamin C), but not oral ascorbate, produces high plasma concentrations capable of selective cytotoxicity to tumor cells. In doses achievable in humans, P-AscH^- decreases the viability and proliferative capacity of pancreatic cancer via a hydrogen peroxide (H₂O₂)-mediated mechanism. In this study, we demonstrate that P-AscH^- radiosensitizes pancreatic cancer cells but inhibits radiation-induced damage to normal cells. Specifically, radiation-induced decreases in clonogenic survival and double-stranded DNA breaks in tumor cells, but not in normal cells, were enhanced by P-AscH^-, while radiation-induced intestinal damage, collagen deposition, and oxidative stress were also reduced with P-AscH^- in normal tissue. We also report on our first-in-human phase I trial that infused P-AscH^- during the radiotherapy "beam on." Specifically, treatment with P-AscH^- increased median overall survival compared with our institutional average (21.7 vs. 12.7 months, P = 0.08) and the E4201 trial (21.7 vs. 11.1 months). Progression-free survival in P-AscH^--treated subjects was also greater than our institutional average (13.7 vs. 4.6 months, P < 0.05) and the E4201 trial (6.0 months). Results indicated that P-AscH^- in combination with gemcitabine and radiotherapy for locally advanced pancreatic adenocarcinoma is safe and well tolerated with suggestions of efficacy. Because of the potential effect size and minimal toxicity, our findings suggest that investigation of P-AscH^- efficacy is warranted in a phase II clinical trial. SIGNIFICANCE: These findings demonstrate that pharmacologic ascorbate enhances pancreatic tumor cell radiation cytotoxicity in addition to offering potential protection from radiation damage in normal surrounding tissue, making it an optimal agent for improving treatment of locally advanced pancreatic adenocarcinoma.

Effect of Targeted Therapies on Prognostic Factors, Patterns of Care, and Survival in Patients With Renal Cell Carcinoma and Brain Metastases

PURPOSE

To identify prognostic factors, define evolving patterns of care, and the effect of targeted therapies in a larger contemporary cohort of renal cell carcinoma (RCC) patients with new brain metastases (BM).

METHODS AND MATERIALS

A multi-institutional retrospective institutional review board-approved database of 711 RCC patients with new BM diagnosed from January 1, 2006, to December 31, 2015, was created. Clinical parameters and treatment were correlated with median survival and time from primary diagnosis to BM. Multivariable analyses were performed.

RESULTS

The median survival for the prior/present cohorts was 9.6/12 months, respectively (P < .01). Four prognostic factors (Karnofsky performance status, extracranial metastases, number of BM, and hemoglobin b) were significant for survival after the diagnosis of BM. Of the 6 drug types studied, only cytokine use after BM was associated with improved survival. The use of whole-brain radiation therapy declined from 50% to 22%, and the use of stereotactic radiosurgery alone increased from 46% to 58%. Nonneurologic causes of death were twice as common as neurologic causes.

CONCLUSIONS

Additional prognostic factors refine prognostication in this larger contemporary cohort. Patterns of care have changed, and survival of RCC patients with BM has improved over time. The reasons for this improvement in survival remain unknown but may relate to more aggressive use of local brain metastasis therapy and a wider array of systemic treatment options for those patients with progressive extracranial tumor.

Automated model-based quantitative analysis of phantoms with spherical inserts in FDG PET scans

PURPOSE

Quality control plays an increasingly important role in quantitative PET imaging and is typically performed using phantoms. The purpose of this work was to develop and validate a fully automated analysis method for two common PET/CT quality assurance phantoms: the NEMA NU-2 IQ and SNMMI/CTN oncology phantom. The algorithm was designed to only utilize the PET scan to enable the analysis of phantoms with thin-walled inserts.

METHODS

We introduce a model-based method for automated analysis of phantoms with spherical inserts. Models are first constructed for each type of phantom to be analyzed. A robust insert detection algorithm uses the model to locate all inserts inside the phantom. First, candidates for inserts are detected using a scale-space detection approach. Second, candidates are given an initial label using a score-based optimization algorithm. Third, a robust model fitting step aligns the phantom model to the initial labeling and fixes incorrect labels. Finally, the detected insert locations are refined and measurements are taken for each insert and several background regions. In addition, an approach for automated selection of NEMA and CTN phantom models is presented. The method was evaluated on a diverse set of 15 NEMA and 20 CTN phantom PET/CT scans. NEMA phantoms were filled with radioactive tracer solution at 9.7:1 activity ratio over background, and CTN phantoms were filled with 4:1 and 2:1 activity ratio over background. For quantitative evaluation, an independent reference standard was generated by two experts using PET/CT scans of the phantoms. In addition, the automated approach was compared against manual analysis, which represents the current clinical standard approach, of the PET phantom scans by four experts.

RESULTS

The automated analysis method successfully detected and measured all inserts in all test phantom scans. It is a deterministic algorithm (zero variability), and the insert detection RMS error (i.e., bias) was 0.97, 1.12, and 1.48 mm for phantom activity ratios 9.7:1, 4:1, and 2:1, respectively. For all phantoms and at all contrast ratios, the average RMS error was found to be significantly lower for the proposed automated method compared to the manual analysis of the phantom scans. The uptake measurements produced by the automated method showed high correlation with the independent reference standard (R² ≥ 0.9987). In addition, the average computing time for the automated method was 30.6 s and was found to be significantly lower (P ≪ 0.001) compared to manual analysis (mean: 247.8 s).

CONCLUSIONS

The proposed automated approach was found to have less error when measured against the independent reference than the manual approach. It can be easily adapted to other phantoms with spherical inserts. In addition, it eliminates inter- and intraoperator variability in PET phantom analysis and is significantly more time efficient, and therefore, represents a promising approach to facilitate and simplify PET standardization and harmonization efforts.

Phase 1b/2a Trial of the Superoxide Dismutase Mimetic GC4419 to Reduce Chemoradiotherapy-Induced Oral Mucositis in Patients With Oral Cavity or Oropharyngeal Carcinoma

PURPOSE

To assess the safety of the superoxide dismutase mimetic GC4419 in combination with radiation and concurrent cisplatin for patients with oral cavity or oropharyngeal cancer (OCC) and to assess the potential of GC4419 to reduce severe oral mucositis (OM).

PATIENTS AND METHODS

Patients with locally advanced OCC treated with definitive or postoperative intensity modulated radiation therapy (IMRT) plus cisplatin received GC4419 by 60-minute intravenous infusion, ending <60 minutes before IMRT, Monday through Friday for 3 to 7 weeks, in a dose and duration escalation study. Oral mucositis was assessed twice weekly during and weekly after IMRT.

RESULTS

A total of 46 patients received GC4419 in 11 separate dosing and duration cohorts: dose escalation occurred in 5 cohorts receiving 15 to 112 mg/d over 3 weeks (n=20), duration escalation in 3 cohorts receiving 112 mg/d over 4 to 6 weeks (n=12), and then 3 additional cohorts receiving 30 or 90 mg/d over 6 to 7 weeks (n=14). A maximum tolerated dose was not reached. One dose-limiting toxicity (grade 3 gastroenteritis and vomiting with hyponatremia) occurred in each of 2 separate cohorts at 112 mg. Nausea/vomiting and facial paresthesia during infusion seemed to be GC4419 dose-related. Severe OM occurred through 60 Gy in 4 of 14 patients (29%) dosed for 6 to 7 weeks, with median duration of only 2.5 days.

CONCLUSIONS

The safety of GC4419 concurrently with chemoradiation for OCC was acceptable. Toxicities included nausea/vomiting and paresthesia. Doses of 30 and 90 mg/d administered for 7 weeks were selected for further study. In an exploratory analysis, severe OM seemed less frequent and briefer than expected.

SBRT to adrenal metastases provides high local control with minimal toxicity

Purpose

The adrenal glands are a common site of metastases because of their rich blood supply. Previously, adrenal metastases were treated with systemic chemotherapy or, more rarely, with surgical resection or palliative radiation therapy. Stereotactic body radiation therapy (SBRT) has recently emerged as an attractive noninvasive approach to definitively treat these lesions. We present our experience in treating adrenal metastases using SBRT and review the current literature.

Methods and materials

This is a single-institution retrospective review of patients who received SBRT to adrenal metastases originating from various primary malignancies. Patients who were eligible for SBRT included those with limited metastatic disease (≤5 sites) with otherwise controlled metastatic disease and uncontrolled adrenal metastases.

Results

Ten patients met the study's inclusion criteria and received SBRT doses of 30 to 48 Gy in 3 to 5 fractions. Acute sequelae of SBRT treatment included 4 patients with grades 1 or 2 nausea, 3 patients with grade 1 fatigue, and 1 with grade 1 diarrhea. The median follow-up was 6 months with a median overall survival of 9.9 months. One patient demonstrated progressive adrenal gland disease 18.8 months after SBRT treatment. Seven patients developed new distant metastases after treatment, with a median progression-free survival of 3.4 months. Three months after SBRT to the adrenal gland, 1 patient developed a gastrointestinal bleed.

Conclusions

These results complement the limited existing body of literature by demonstrating that SBRT provides good control of treated adrenal gland metastasis; however, high-grade late toxicities may occur. More stringent dose constraint limits may prevent associated serious adverse events.

Mitochondrial Superoxide Increases Age-Associated Susceptibility of Human Dermal Fibroblasts to Radiation and Chemotherapy

Elderly cancer patients treated with ionizing radiation (IR) or chemotherapy experience more frequent and greater normal tissue toxicity relative to younger patients. The current study demonstrates that exponentially growing fibroblasts from elderly (old) male donor subjects (70, 72, and 78 years) are significantly more sensitive to clonogenic killing mediated by platinum-based chemotherapy and IR (∼70%-80% killing) relative to young fibroblasts (5 months and 1 year; ∼10%-20% killing) and adult fibroblasts (20 years old; ∼10%-30% killing). Old fibroblasts also displayed significantly increased (2-4-fold) steady-state levels of O₂^•-, O₂ consumption, and mitochondrial membrane potential as well as significantly decreased (40%-50%) electron transport chain (ETC) complex I, II, IV, V, and aconitase (70%) activities, decreased ATP levels, and significantly altered mitochondrial structure. Following adenoviral-mediated overexpression of SOD2 activity (5-7-fold), mitochondrial ETC activity and aconitase activity were restored, demonstrating a role for mitochondrial O₂^•- in these effects. Old fibroblasts also demonstrated elevated levels of endogenous DNA damage that were increased following treatment with IR and chemotherapy. Most importantly, treatment with the small-molecule, superoxide dismutase mimetic (GC4419; 0.25 μmol/L) significantly mitigated the increased sensitivity of old fibroblasts to IR and chemotherapy and partially restored mitochondrial function without affecting IR or chemotherapy-induced cancer cell killing. These results support the hypothesis that age-associated increased O₂^•- and resulting DNA damage mediate the increased susceptibility of old fibroblasts to IR and chemotherapy that can be mitigated by GC4419. Cancer Res; 77(18); 5054-67. ©2017 AACR.

Radioiodine Ablation following Thyroidectomy for Differentiated Thyroid Cancer: Literature Review of Utility, Dose, and Toxicity

Management recommendations for differentiated thyroid cancer are evolving. Total thyroidectomy is the backbone of curative-intent therapy, with radioiodine ablation (RAI) of the thyroid remnant routinely performed, in order to facilitate serologic surveillance and reduce recurrence risk. Several single-institution series have identified patient subsets for whom recurrence risk is sufficiently low that RAI may not be indicated. Further, the appropriate dose of RAI specific to variable clinicopathologic presentations remains poorly defined. While recent randomized trials demonstrated equivalent thyroid remnant ablation rates between low- and high-dose RAI, long-term oncologic endpoints remain unreported. While RAI may be employed to facilitate surveillance following total thyroidectomy, cancer recurrence risk reduction is not demonstrated in favorable-risk patients with tumor size ≤1 cm without high-risk pathologic features. When RAI is indicated, in patients without macroscopic residual disease or metastasis, the evidence suggests that the rate of successful remnant ablation following total thyroidectomy is equivalent between doses of 30-50 mCi and doses ≥100 mCi, with fewer acute side effects; however, in the setting of subtotal thyroidectomy or when preablation diagnostic scan uptake is >2%, higher doses are associated with improved ablation rates. Historical series demonstrate conflicting findings of long-term cancer control rates between dose levels; long-term results from modern series have yet to be reported. For high-risk patients, including those with positive surgical margins, gross extrathyroidal extension, lymph node involvement, subtotal thyroidectomy, or >5% uptake, higher-dose RAI therapy appears to provide superior rates of ablation and cancer control.

Development of a radiobiological evaluation tool to assess the expected clinical impacts of contouring accuracy between manual and semi-automated segmentation algorithms

RADEval is a tool developed to assess the expected clinical impact of contouring accuracy when comparing manual contouring and semi-automated segmentation. The RADEval tool, designed to process large scale datasets, imported a total of 2,760 segmentation datasets, along with a Simultaneous Truth and Performance Level Estimation (STAPLE) to act as ground truth tumor segmentations. Virtual dose-maps were created within RADEval and two different tumor control probability (TCP) values using a Logistic and a Poisson TCP models were calculated in RADEval using each STAPLE and each dose-map. RADEval also virtually generated a ring of normal tissue. To evaluate clinical impact, two different uncomplicated TCP (UTCP) values were calculated in RADEval by using two TCP-NTCP correlation parameters (δ = 0 and 1). NTCP values showed that semi-automatic segmentation resulted in lower NTCP with an average 1.5 - 1.6 % regardless of STAPLE design. This was true even though each normal tissue was created from each STAPLE (p <; 0.00001). TCP and UTCP presented no statistically significant differences (p ≥ 0.1884). The intra-operator standard deviations (SDs) for TCP, NTCP and UTCP were significantly lower for the semi-automatic segmentation method regardless of STAPLE design (p <; 0.0331). Both intra-and inter-operator SDs of TCP, NTCP and UTCP were significantly lower for semi-automatic segmentation for the STAPLE 1 design (p <;0.0331). RADEval was able to efficiently process 4,920 datasets of two STAPLE designs and successfully assess the expected clinical impact of contouring accuracy.

Consuming a Ketogenic Diet while Receiving Radiation and Chemotherapy for Locally Advanced Lung Cancer and Pancreatic Cancer: The University of Iowa Experience of Two Phase 1 Clinical Trials

Ketogenic diets are low in carbohydrates and high in fat, which forces cells to rely more heavily upon mitochondrial oxidation of fatty acids for energy. Relative to normal cells, cancer cells are believed to exist under a condition of chronic mitochondrial oxidative stress that is compensated for by increases in glucose metabolism to generate reducing equivalents. In this study we tested the hypothesis that a ketogenic diet concurrent with radiation and chemotherapy would be clinically tolerable in locally advanced non-small cell lung cancer (NSCLC) and pancreatic cancer and could potentially exploit cancer cell oxidative metabolism to improve therapeutic outcomes. Mice bearing MIA PaCa-2 pancreatic cancer xenografts were fed either a ketogenic diet or standard rodent chow, treated with conventionally fractionated radiation (2 Gy/fraction), and tumor growth rates were assessed daily. Tumors were assessed for immunoreactive 4-hydroxy-2-nonenal-(4HNE)-modfied proteins as a marker of oxidative stress. Based on this and another previously published preclinical study, phase 1 clinical trials in locally advanced NSCLC and pancreatic cancer were initiated, combining standard radiation and chemotherapy with a ketogenic diet for six weeks (NSCLC) or five weeks (pancreatic cancer). The xenograft experiments demonstrated prolonged survival and increased 4HNE-modfied proteins in animals consuming a ketogenic diet combined with radiation compared to radiation alone. In the phase 1 clinical trial, over a period of three years, seven NSCLC patients enrolled in the study. Of these, four were unable to comply with the diet and withdrew, two completed the study and one was withdrawn due to a dose-limiting toxicity. Over the same time period, two pancreatic cancer patients enrolled in the trial. Of these, one completed the study and the other was withdrawn due to a dose-limiting toxicity. The preclinical experiments demonstrate that a ketogenic diet increases radiation sensitivity in a pancreatic cancer xenograft model. However, patients with locally advanced NSCLC and pancreatic cancer receiving concurrent radiotherapy and chemotherapy had suboptimal compliance to the oral ketogenic diet and thus, poor tolerance.

O2⋅- and H2O2-Mediated Disruption of Fe Metabolism Causes the Differential Susceptibility of NSCLC and GBM Cancer Cells to Pharmacological Ascorbate

Pharmacological ascorbate has been proposed as a potential anti-cancer agent when combined with radiation and chemotherapy. The anti-cancer effects of ascorbate are hypothesized to involve the autoxidation of ascorbate leading to increased steady-state levels of H2O2; however, the mechanism(s) for cancer cell-selective toxicity remain unknown. The current study shows that alterations in cancer cell mitochondrial oxidative metabolism resulting in increased levels of O2⋅- and H2O2 are capable of disrupting intracellular iron metabolism, thereby selectively sensitizing non-small-cell lung cancer (NSCLC) and glioblastoma (GBM) cells to ascorbate through pro-oxidant chemistry involving redox-active labile iron and H2O2. In addition, preclinical studies and clinical trials demonstrate the feasibility, selective toxicity, tolerability, and potential efficacy of pharmacological ascorbate in GBM and NSCLC therapy.

Multi-site quality and variability analysis of 3D FDG PET segmentations based on phantom and clinical image data

PURPOSE

Radiomics utilizes a large number of image-derived features for quantifying tumor characteristics that can in turn be correlated with response and prognosis. Unfortunately, extraction and analysis of such image-based features is subject to measurement variability and bias. The challenge for radiomics is particularly acute in Positron Emission Tomography (PET) where limited resolution, a high noise component related to the limited stochastic nature of the raw data, and the wide variety of reconstruction options confound quantitative feature metrics. Extracted feature quality is also affected by tumor segmentation methods used to define regions over which to calculate features, making it challenging to produce consistent radiomics analysis results across multiple institutions that use different segmentation algorithms in their PET image analysis. Understanding each element contributing to these inconsistencies in quantitative image feature and metric generation is paramount for ultimate utilization of these methods in multi-institutional trials and clinical oncology decision making.

METHODS

To assess segmentation quality and consistency at the multi-institutional level, we conducted a study of seven institutional members of the National Cancer Institute Quantitative Imaging Network. For the study, members were asked to segment a common set of phantom PET scans acquired over a range of imaging conditions as well as a second set of head and neck cancer (HNC) PET scans. Segmentations were generated at each institution using their preferred approach. In addition, participants were asked to repeat segmentations with a time interval between initial and repeat segmentation. This procedure resulted in overall 806 phantom insert and 641 lesion segmentations. Subsequently, the volume was computed from the segmentations and compared to the corresponding reference volume by means of statistical analysis.

RESULTS

On the two test sets (phantom and HNC PET scans), the performance of the seven segmentation approaches was as follows. On the phantom test set, the mean relative volume errors ranged from 29.9 to 87.8% of the ground truth reference volumes, and the repeat difference for each institution ranged between -36.4 to 39.9%. On the HNC test set, the mean relative volume error ranged between -50.5 to 701.5%, and the repeat difference for each institution ranged between -37.7 to 31.5%. In addition, performance measures per phantom insert/lesion size categories are given in the paper. On phantom data, regression analysis resulted in coefficient of variation (CV) components of 42.5% for scanners, 26.8% for institutional approaches, 21.1% for repeated segmentations, 14.3% for relative contrasts, 5.3% for count statistics (acquisition times), and 0.0% for repeated scans. Analysis showed that the CV components for approaches and repeated segmentations were significantly larger on the HNC test set with increases by 112.7% and 102.4%, respectively.

CONCLUSION

Analysis results underline the importance of PET scanner reconstruction harmonization and imaging protocol standardization for quantification of lesion volumes. In addition, to enable a distributed multi-site analysis of FDG PET images, harmonization of analysis approaches and operator training in combination with highly automated segmentation methods seems to be advisable. Future work will focus on quantifying the impact of segmentation variation on radiomics system performance.

Semiautomated segmentation of head and neck cancers in 18F-FDG PET scans: A just-enough-interaction approach

Purpose:

The purpose of this work was to develop, validate, and compare a highly computer-aided method for the segmentation of hot lesions in head and neck 18F-FDG PET scans.

Methods:

A semiautomated segmentation method was developed, which transforms the segmentation problem into a graph-based optimization problem. For this purpose, a graph structure around a user-provided approximate lesion centerpoint is constructed and a suitable cost function is derived based on local image statistics. To handle frequently occurring situations that are ambiguous (e.g., lesions adjacent to each other versus lesion with inhomogeneous uptake), several segmentation modes are introduced that adapt the behavior of the base algorithm accordingly. In addition, the authors present approaches for the efficient interactive local and global refinement of initial segmentations that are based on the “just-enough-interaction” principle. For method validation, 60 PET/CT scans from 59 different subjects with 230 head and neck lesions were utilized. All patients had squamous cell carcinoma of the head and neck. A detailed comparison with the current clinically relevant standard manual segmentation approach was performed based on 2760 segmentations produced by three experts.

Results:

Segmentation accuracy measured by the Dice coefficient of the proposed semiautomated and standard manual segmentation approach was 0.766 and 0.764, respectively. This difference was not statistically significant (p = 0.2145). However, the intra- and interoperator standard deviations were significantly lower for the semiautomated method. In addition, the proposed method was found to be significantly faster and resulted in significantly higher intra- and interoperator segmentation agreement when compared to the manual segmentation approach.

Conclusions:

Lack of consistency in tumor definition is a critical barrier for radiation treatment targeting as well as for response assessment in clinical trials and in clinical oncology decision-making. The properties of the authors approach make it well suited for applications in image-guided radiation oncology, response assessment, or treatment outcome prediction.