Cancer Treatment Response Assessment Research Articles

Imaging for assessment of cancer treatment response to immune checkpoint inhibitors can be complementary in identifying hypophysitis.

Hypophysitis is reported in 8.5%-14% of patients receiving combination immune checkpoint inhibition (cICI) but can be a diagnostic challenge. This study aimed to assess the role of routine diagnostic imaging performed during therapeutic monitoring of combination anti-CTLA-4/anti-PD-1 treatment in the identification of hypophysitis and the relationship of imaging findings to clinical diagnostic criteria. This retrospective cohort study identified patients treated with cICI between January 2016 and January 2019 at a quaternary melanoma service. Medical records were reviewed to identify patients with a documented diagnosis of hypophysitis based on clinical criteria. Available structural brain imaging with magnetic resonance imaging (MRI) or computed tomography (CT) of the brain and 2-deoxy-2-[18F]fluoro-D-glucose positron emission tomography with computed tomography (FDG-PET/CT) were assessed retrospectively. The main radiological outcome measures were a relative change in pituitary size or FDG uptake temporally attributed to cICI. There were 162 patients (median age 60 years, 30% female) included. A total of 100 and 134 had serial CT/MRI of the brain and FDG-PET/CT, respectively. There were 31 patients who had a documented diagnosis of hypophysitis and an additional 20 who had isolated pituitary imaging findings. The pituitary gland enlargement was mild, and the largest absolute gland size was 13 mm, with a relative increase of 7 mm from baseline. There were no cases of optic chiasm compression. Pituitary enlargement and increased FDG uptake were universally transient. High-dose glucocorticoid treatment for concurrent irAEs prevented assessment of the pituitary-adrenal axis in 90% of patients with isolated imaging findings. Careful review of changes in pituitary characteristics on imaging performed for assessment of therapeutic response to iICI may lead to increased identification and more prompt management of cICI-induced hypophysitis.

Follow-up imaging of anal cancer after treatment.

Anal cancer treatment response assessment can be challenging with both magnetic resonance imaging (MRI) and clinical evaluation considered essential. MRI, in particular, has shown to be useful for the assessment of treatment response, the detection of recurrent disease in follow up and surveillance, and the evaluation of possible post-treatment complications as well as complications from the tumor itself. In this review, we focus on the role of imaging, mainly MRI, in anal cancer treatment response assessment. We also describe the treatment complications that can occur, and the imaging findings associated with those complications.

Prompt Assessment of Tumor Load and Treatment Response in Patients with Lymphoma By a Blood-Based Multi-Omics Approach

Introduction Lymphoma is a class of highly heterogeneous hematological malignancy derived from malignant clonal proliferation of lymphocytes. Periodic imaging examinations are the current standard for clinical assessment of lymphoma treatment response. Liquid biopsy is an emerging, non-invasive method for quantitative assessment of cancer treatment response. In this prospective study, we applied a noninvasive multi-omics assay named SeekInClarity to assess lymphoma treatment response and predict patient outcomes. Methods 149 patients meeting the WHO diagnosis criteria of lymphoid neoplasm were prospectively recruited from two clinical sites. The study is in progress, and 114 patients were included in the current interim analysis. 8.8% of patients who were lost to follow-up or quickly died were excluded. SeekInClarity assay integrates the analysis of genomic feature and common plasma tumor protein markers (PTMs, including AFP, CEA, CA153, CA125, CA199, CYFRA21-1 and CA724). Genomic features including copy number aberration (CNA) and the fragment size (FS) that are pan-cancer hallmarks were analyzed by shallow whole genome sequencing. Patients received SeekInClarity test before and after 1, 2, 4 and 6 cycles of treatment to generate the molecular tumor burden score (MTB). MTB positive/negative suggests that molecular signal of tumor is detectable/undetectable. Results The value of each dimension (protein marker, CNA and FS) in the prior-treatment (baseline) lymphoma samples were significantly higher than that in healthy control samples (Mann-Whitney test: P<0.001). 72.8% base line samples were MTB positive. After 2 cycles of treatment, the samples from patients who experienced disease progression showed higher MTB positivity rate than that from those without disease progression (85.7% vs. 26.7). MTB- patients showed a significantly longer progression free survival (PFS) time than MTB+ patients [HR: 12.6, 95% CI:4.1-38.6, P<0.0001] regardless of tumor stage and lymphoma subtype. Patients with reduced sample MTB values compared to baseline sample MTB values after two cycles of treatment had significantly better disease-free survival than patients with increased sample MTB values after treatment [HR: 5.0, 95% CI:0.48-52.5; P<0.0001]. We further divided the patients into two groups according to the degree of reduction in MTB values in the pre- and post-treatment samples (>50% & <50%), and we found that patients with higher MTB decrease (>50%) showed a better PFS than those with lower MTB decrease (<50%, P < 0.001). Conclusions This prospective study demonstrated that SeekInClarity, a non-invasive, multidimensional blood-based assay, can promptly evaluate the treatment response of lymphoma patients, and the MTB status showed an excellent correlation with clinical outcome. The dynamic change of MTB exhibited a good prognostic power of lymphoma. It challenged the status quo of employing imaging as a sole standard for evaluating early treatment response.

Computerized Decision Support for Bladder Cancer Treatment Response Assessment in CT Urography: Effect on Diagnostic Accuracy in Multi-Institution Multi-Specialty Study.

This observer study investigates the effect of computerized artificial intelligence (AI)-based decision support system (CDSS-T) on physicians’ diagnostic accuracy in assessing bladder cancer treatment response. The performance of 17 observers was evaluated when assessing bladder cancer treatment response without and with CDSS-T using pre- and post-chemotherapy CTU scans in 123 patients having 157 pre- and post-treatment cancer pairs. The impact of cancer case difficulty, observers’ clinical experience, institution affiliation, specialty, and the assessment times on the observers’ diagnostic performance with and without using CDSS-T were analyzed. It was found that the average performance of the 17 observers was significantly improved (p = 0.002) when aided by the CDSS-T. The cancer case difficulty, institution affiliation, specialty, and the assessment times influenced the observers’ performance without CDSS-T. The AI-based decision support system has the potential to improve the diagnostic accuracy in assessing bladder cancer treatment response and result in more consistent performance among all physicians.

Intraobserver Variability in Bladder Cancer Treatment Response Assessment With and Without Computerized Decision Support.

We evaluated the intraobserver variability of physicians aided by a computerized decision-support system for treatment response assessment (CDSS-T) to identify patients who show complete response to neoadjuvant chemotherapy for bladder cancer, and the effects of the intraobserver variability on physicians' assessment accuracy. A CDSS-T tool was developed that uses a combination of deep learning neural network and radiomic features from computed tomography (CT) scans to detect bladder cancers that have fully responded to neoadjuvant treatment. Pre- and postchemotherapy CT scans of 157 bladder cancers from 123 patients were collected. In a multireader, multicase observer study, physician-observers estimated the likelihood of pathologic T0 disease by viewing paired pre/posttreatment CT scans placed side by side on an in-house-developed graphical user interface. Five abdominal radiologists, 4 diagnostic radiology residents, 2 oncologists, and 1 urologist participated as observers. They first provided an estimate without CDSS-T and then with CDSS-T. A subset of cases was evaluated twice to study the intraobserver variability and its effects on observer consistency. The mean areas under the curves for assessment of pathologic T0 disease were 0.85 for CDSS-T alone, 0.76 for physicians without CDSS-T and improved to 0.80 for physicians with CDSS-T (P = .001) in the original evaluation, and 0.78 for physicians without CDSS-T and improved to 0.81 for physicians with CDSS-T (P = .010) in the repeated evaluation. The intraobserver variability was significantly reduced with CDSS-T (P < .0001). The CDSS-T can significantly reduce physicians' variability and improve their accuracy for identifying complete response of muscle-invasive bladder cancer to neoadjuvant chemotherapy.

Diagnostic Accuracy of CT for Prediction of Bladder Cancer Treatment Response with and without Computerized Decision Support

To evaluate whether a computed tomography (CT)-based computerized decision-support system for muscle-invasive bladder cancer treatment response assessment (CDSS-T) can improve identification of patients who have responded completely to neoadjuvant chemotherapy. Following Institutional Review Board approval, pre-chemotherapy and post-chemotherapy CT scans of 123 subjects with 157 muscle-invasive bladder cancer foci were collected retrospectively. CT data were analyzed with a CDSS-T that uses a combination of deep-learning convolutional neural network and radiomic features to distinguish muscle-invasive bladder cancers that have fully responded to neoadjuvant treatment from those that have not. Leave-one-case-out cross-validation was used to minimize overfitting. Five attending abdominal radiologists, four diagnostic radiology residents, two attending oncologists, and one attending urologist estimated the likelihood of pathologic T0 disease (complete response) by viewing paired pre/post-treatment CT scans placed side-by-side on an internally-developed graphical user interface. The observers provided an estimate without use of CDSS-T and then were permitted to revise their estimate after a CDSS-T-derived likelihood score was displayed. Observer estimates were analyzed with multi-reader, multi-case receiver operating characteristic methodology. The area under the curve (AUC) and the statistical significance of the difference were estimated. The mean AUCs for assessment of pathologic T0 disease were 0.80 for CDSS-T alone, 0.74 for physicians not using CDSS-T, and 0.77 for physicians using CDSS-T. The increase in the physicians' performance was statistically significant (P < .05). CDSS-T improves physician performance for identifying complete response of muscle-invasive bladder cancer to neoadjuvant chemotherapy.

Bladder Cancer Treatment Response Assessment in CT using Radiomics with Deep-Learning

Cross-sectional X-ray imaging has become the standard for staging most solid organ malignancies. However, for some malignancies such as urinary bladder cancer, the ability to accurately assess local extent of the disease and understand response to systemic chemotherapy is limited with current imaging approaches. In this study, we explored the feasibility that radiomics-based predictive models using pre- and post-treatment computed tomography (CT) images might be able to distinguish between bladder cancers with and without complete chemotherapy responses. We assessed three unique radiomics-based predictive models, each of which employed different fundamental design principles ranging from a pattern recognition method via deep-learning convolution neural network (DL-CNN), to a more deterministic radiomics feature-based approach and then a bridging method between the two, utilizing a system which extracts radiomics features from the image patterns. Our study indicates that the computerized assessment using radiomics information from the pre- and post-treatment CT of bladder cancer patients has the potential to assist in assessment of treatment response.

The Attenuation Distribution Across the Long Axis (ADLA): Preliminary Findings for Assessing Response to Cancer Treatment

Novel image analysis methods may be useful adjuncts to standard cancer treatment response assessment techniques. The attenuation distribution across the long axis (ADLA) is a simple measure of lesion heterogeneity that can be obtained while measuring the long axis diameter of a target lesion. The purpose of this study was to obtain preliminary validation of the ADLA method for predicting treatment response in a small clinical trial. Under an Institutional Review Board waiver, we obtained de-identified imaging and clinical data from a phase 2 trial of an investigational anticancer therapy at our institution. We retrospectively analyzed all patients with at least one liver metastasis measuring ≥15 mm on baseline contrast-enhanced computed tomography. For each patient at every imaging time point, up to two target liver lesions were evaluated using Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 and ADLA measurements. The ADLA was obtained as the standard deviation of the post-contrast computed tomography attenuation values in the portal venous phase across a linear function spanning the long-axis diameter. Using Kaplan-Meier survival analysis, the log-rank test was used to evaluate the ability of RECIST 1.1 and ADLA measurements to discriminate patients with longer overall survival (OS). Fifteen patients met inclusion criteria. Median survival was 149 days (range 57-487). Best overall response by the ADLA method successfully separated patients with longer OS (p = .04). Best overall response by RECIST 1.1 did not discriminate patients with longer survival (P > .05). In retrospective data analysis from a phase 2 clinical trial, the ADLA method was more predictive of OS than RECIST 1.1. Further studies are needed to explore the utility of this measurement in predicting response to cancer treatment.

Automated temporal tracking and segmentation of lymphoma on serial CT examinations

It is challenging to reproducibly measure and compare cancer lesions on numerous follow-up studies; the process is time-consuming and error-prone. In this paper, we show a method to automatically and reproducibly identify and segment abnormal lymph nodes in serial computed tomography (CT) exams. Our method leverages initial identification of enlarged (abnormal) lymph nodes in the baseline scan. We then identify an approximate region for the node in the follow-up scans using nonrigid image registration. The baseline scan is also used to locate regions of normal, non-nodal tissue surrounding the lymph node and to map them onto the follow-up scans, in order to reduce the search space to locate the lymph node on the follow-up scans. Adaptive region-growing and clustering algorithms are then used to obtain the final contours for segmentation. We applied our method to 24 distinct enlarged lymph nodes at multiple time points from 14 patients. The scan at the earlier time point was used as the baseline scan to be used in evaluating the follow-up scan, resulting in 70 total test cases (e.g., a series of scans obtained at 4 time points results in 3 test cases). For each of the 70 cases, a "reference standard" was obtained by manual segmentation by a radiologist. Assessment according to response evaluation criteria in solid tumors (RECIST) using our method agreed with RECIST assessments made using the reference standard segmentations in all test cases, and by calculating node overlap ratio and Hausdorff distance between the computer and radiologist-generated contours. Compared to the reference standard, our method made the correct RECIST assessment for all 70 cases. The average overlap ratio was 80.7 ± 9.7% s.d., and the average Hausdorff distance was 3.2 ± 1.8 mm s.d. The concordance correlation between automated and manual segmentations was 0.978 (95% confidence interval 0.962, 0.984). The 100% agreement in our sample between our method and the standard with regard to RECIST classification suggests that the true disagreement rate is no more than 6%. Our automated lymph node segmentation method achieves excellent overall segmentation performance and provides equivalent RECIST assessment. It potentially will be useful to streamline and improve cancer lesion measurement and tracking and to improve assessment of cancer treatment response.

Computational approaches to assist in the evaluation of cancer treatment response

Cancer treatment-response assessment is an important clinical and research task that utilizes imagebased methods for tracking changes in tumor lesions over time. This article reviews the role of imaging in cancer treatment-response assessment, the limitations of current approaches, and the emerging and future imaging informatics methods to support treatment-response assessment. Imaging informatics methods described include: image repositories for applying and developing response criteria; methods to support the acquisition and storage of image metadata, including ontologies, data standards, image annotation tools and metadata databases; and methods to visualize images and image metadata for treatment decision making. Future directions are discussed, including the development of linked clinical, image and image metadata repositories, and the integration of image metadata into the electronic health record.

TU‐C‐201C‐04: Informatics Infrastructure for Quantitative Assessment of Cancer Treatment Response

There have been substantial advances in creating novel imaging platforms and quantitative imaging methods to provide accurate information about the status of patients with cancer. Quantitative imaging biomarkers promise improved sensitivity in assessing cancer treatment response and in identifying new therapeutics. These methods have tremendous potential to improve the accuracy and efficiency of cancer research; however, the lack of software infrastructure to enable validating and adopting these methods in cancer clinical trials hampers progress. Validating and using new quantitative imaging techniques is hindered by several challenges: (1) lack of tools to record quantitative imaging information about tumor burden reproducibly and efficiently in the clinical workflow; (2) lack of coordination and effective communication between oncologists and radiologists in making quantitative imaging assessments; (3) lack of integration of multiple different quantitative measures of tumor burden that, when taken together, would be more informative than individual indicators, and (4) lack of infrastructure to enable sharing of data derived from quantitative imaging methods among centers and data mining.We are tackling these challenges with the goal of enabling cancer researchers to reproducibly measure tumor burden, to better assess the effectiveness of therapies in patient cohorts, potentially leading to shorter clinical trials, and to make the best treatment choices for individual cancer patients. Our work, being conducted as one of the research centers in NCI's new Quantitative Imaging Network (QIN), has four objectives: (1) to develop software infrastructure to harmonize data collection and analysis across different image viewing platforms, (2) to create tools leveraging caBIG technologies that permit comprehensive and reproducible assessment of the quantitative imaging features of tumor burden, and that will improve the coordination of radiologists and oncologists in collecting quantitative image data, (3) to develop methods to analyze quantitative image data to provide decision support, helping oncologists to evaluate and recognize whether a treatment is effective, and (4) to evaluate the utility of this infrastructure in prospective clinical trials, thereby demonstrating improvement in researchers' ability to assess response to treatment in patient cohorts and in individual patients. As a result of this work, the oncology community will be able to share approaches to validating and standardizing imaging biomarkers and quantitative imaging measurements of tumor responses to cancer therapies, which in turn will lead to robust comparison of quantitative methods and to the identification of more accurate approaches for assessing tumor response. Ultimately, this work will enable translational research, accelerating incorporation of quantitative imaging methods into cancer research and improving patient care.Learning Objectives:1. Understand the opportunity and current limitations of quantitative imaging in cancer research.2. Learn about new developments in imaging informatics to improve quantitative image‐based assessment of cancer treatment response.3. Recognize future opportunities of imaging informatics to use quantitative imaging for assessing new cancer therapies, optimizing treatment decisions, and personalizing patient care.

SU‐FF‐J‐125: Therapy Assessment Using a Full Time Point (fTP) Pharmacokinetic Analysis of Dynamic Contrast‐Enhanced Magnetic Resonance Imaging (DCE‐MRI): Role of Region of Interest (ROI) Selection in Three Tumor Sites

Background: DCE‐MRI has great potential to provide routine assessment of cancer treatment response. The contrast agent preferentially accumulates in tumors due to increased tumor vascular permeability. The MR based parameters that can be used for treatment assessment (permeability and leakage space) can be averaged over the whole tumor or over a maximum enhancement ROI. The optimal method of analysis is controversial. Material and Methods: DCE‐MR images of selective head and neck, breast and extremity sarcoma patients were acquired on a 1.5T GE Signa Exite scanner before and after different therapy regimens (concurrent chemoradiation for head and neck, neoadjuvant and hyperthermia for breast, and radiation and hyperthermia for extremity sarcomas). The images were analyzed using an fTP pharmacokinetic analysis implemented by CAD Sciences® (White Plains, NY) that determines the tumor's permeability (PERM) and extracellular volume fraction (EVF). ROIs were defined over the entire extent of the tumor (Tumor ROI) and over areas that show maximum enhancement (MaxEnhROI). For head and neck cases, MaxEnhROI were carefully selected not to include fast enhancing arteries which can bias the averaging. Results: For all three tumor types, the trends of PERM and EVF change are the same, but the absolute changes are not. Larger differences are seen in tumors that have peripheral enhancement and large necrotic areas (found often in head and neck and extremity sarcomas). As expected, for homogeneous enhancing tumors (rarely encountered), the differences are not significant. Conclusions: Preliminary results show that ROI selection is important in MR parameter averaging, particularly if the quantitative analysis is used for therapy assessment. This statement is true for all the three sites considered. Although DCE‐MRI has great potential to provide routine assessment of cancer treatment response, its widespread application should be standardized for clinical use. Supported in part by grant NCI CA42745.

Expanding the Use of Magnetic Resonance in the Assessment of Tumor Response to Therapy: Workshop Report

Although dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and magnetic resonance spectroscopy (MRS) have great potential to provide routine assessment of cancer treatment response, their widespread application has been hampered by a lack of standards for use. Thus, the National Cancer Institute convened a workshop to assess developments and applications of these methods, develop standards for methodology, and engage relevant partners (drug and device industries, researchers, clinicians, and government) to encourage sharing of data and methodologies. Consensus recommendations were reached for DCE-MRI methodologies and the focus for initial multicenter trials of MRS. In this meeting report, we outline the presentations, the topics discussed, the ongoing challenges identified, and the recommendations made by workshop participants for the use of DCE-MRI and 1H MRS in the clinical assessment of antitumor therapies.

Rapid and Quantitative Assessment of Cancer Treatment Response Using In Vivo Bioluminescence Imaging

Current assessment of orthotopic tumor models in animals utilizes survival as the primary therapeutic end point. In vivo bioluminescence imaging (BLI) is a sensitive imaging modality that is rapid and accessible, and may comprise an ideal tool for evaluating antineoplastic therapies [1 ]. Using human tumor cell lines constitutively expressing luciferase, the kinetics of tumor growth and response to therapy have been assessed in intraperitoneal [2], subcutaneous, and intravascular [3] cancer models. However, use of this approach for evaluating orthotopic tumor models has not been demonstrated. In this report, the ability of BLI to noninvasively quantitate the growth and therapeuticinduced cell kill of orthotopic rat brain tumors derived from 9L gliosarcoma cells genetically engineered to stably express firefly luciferase (9LLuc) was investigated. Intracerebral tumor burden was monitored over time by quantitation of photon emission and tumor volume using a cryogenically cooled CCD camera and magnetic resonance imaging (MRI), respectively. There was excellent correlation (r=0.91) between detected photons and tumor volume. A quantitative comparison of tumor cell kill determined from serial MRI volume measurements and BLI photon counts following 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU) treatment revealed that both imaging modalities yielded statistically similar cell kill values (P=.951). These results provide direct validation of BLI imaging as a powerful and quantitative tool for the assessment of antineoplastic therapies in living animals.