Three-Dimensional Margin as a Predictor of Local Tumor Progression after Microwave Ablation: Intraprocedural versus 4-8-Week Postablation Assessment

PURPOSE

To evaluate the prognostic accuracy of intraprocedural and 4-8-week (current standard) post-microwave ablation zone (AZ) and margin assessments for prediction of local tumor progression (LTP) using 3-dimensional (3D) software.

MATERIALS AND METHODS

Data regarding 100 colorectal liver metastases (CLMs) in 75 patients were collected from 2 prospective fluorodeoxyglucose positron emission tomography (PET)/computed tomography (CT)-guided microwave ablation (MWA) trials. The target CLMs and theoretical 5- and 10-mm margins were segmented and registered intraprocedurally and at 4-8 weeks after MWA contrast-enhanced CT (or magnetic resonance [MR] imaging) using the same methodology and 3D software. Tumor and 5- and 10-mm minimal margin (MM) volumes not covered by the AZ were defined as volumes of insufficient coverage (VICs). The intraprocedural and 4-8-week post-MWA VICs were compared as predictors of LTP using receiver operating characteristic curve analysis.

RESULTS

The median follow-up time was 19.6 months (interquartile range, 7.97-36.5 months). VICs for 5- and 10-mm MMs were predictive of LTP at both time assessments. The highest accuracy for the prediction of LTP was documented with the intra-ablation 5-mm VIC (area under the curve [AUC], 0.78; 95% confidence interval, 0.66-0.89). LTP for a VIC of 6-10-mm margin category was 11.4% compared with 4.3% for >10-mm margin category (P < .001).

CONCLUSIONS

A 3D 5-mm MM is a critical endpoint of thermal ablation, whereas optimal local tumor control is noted with a 10-mm MM. Higher AUCs for prediction of LTP were achieved for intraprocedural evaluation than for the 4-8-week postablation 3D evaluation of the AZ.

CALGB 80802 (Alliance): Impact of Sorafenib with and without Doxorubicin on Hepatitis C Infection in Patients with Advanced Hepatocellular Carcinoma

Sorafenib blocks nonstructural protein 5A (NS5A)-recruited c-Raf-mediated hepatitis C virus (HCV) replication and gene expression. Release of Raf-1-Ask-1 dimer and inhibition of Raf-1 via sorafenib putatively differ in the presence or absence of doxorubicin. Cancer and Leukemia Group B (CALGB) 80802 (Alliance) randomized phase III trial of doxorubicin plus sorafenib versus sorafenib in patients with advanced hepatocellular carcinoma (HCC), showed no improvement in median overall survival (OS). Whether HCV viral load impacts therapy and whether any correlation between HCV titers and outcome based on HCV was studied. In patients with HCV, HCV titer levels were evaluated at baseline and at multiple postbaseline timepoints until disease progression or treatment discontinuation. HCV titer levels were evaluated in relation to OS and progression-free survival (PFS). Among 53 patients with baseline HCV data, 12 patients had undetectable HCV (HCV-UN). Postbaseline HCV titer levels did not significantly differ between treatment arms. One patient in each arm went from detectable to HCV-UN with greater than 2 log-fold titer levels reduction. Aside from these 2 HCV-UN patients, HCV titers remained stable on treatment. Patients who had HCV-UN at baseline were 3.5 times more likely to progress and/or die from HCC compared with HCV detectable (HR = 3.51; 95% confidence interval: 1.58-7.78; P = 0.002). HCV titer levels remained unchanged, negating any sorafenib impact onto HCV titer levels. Although an overall negative phase III study, patients treated with doxorubicin plus sorafenib and sorafenib only, on CALGB 80802 had worse PFS if HCV-UN. Higher levels of HCV titers at baseline were associated with significantly improved PFS.

SIGNIFICANCE

Sorafenib therapy for HCC may impact HCV replication and viral gene expression. In HCV-positive patients accrued to CLAGB 80802 phase III study evaluating the addition of doxorubicin to sorafenib, HCV titer levels were evaluated at baseline and different timepoints. Sorafenib did not impact HCV titer levels. Despite an improved PFS in patients with detectable higher level HCV titers at baseline, no difference in OS was noted.

Artificial intelligence in immunotherapy PET/SPECT imaging

OBJECTIVE

Immunotherapy has dramatically altered the therapeutic landscape for oncology, but more research is needed to identify patients who are likely to achieve durable clinical benefit and those who may develop unacceptable side effects. We investigated the role of artificial intelligence in PET/SPECT-guided approaches for immunotherapy-treated patients.

METHODS

We performed a scoping review of MEDLINE, CENTRAL, and Embase databases using key terms related to immunotherapy, PET/SPECT imaging, and AI/radiomics through October 12, 2022.

RESULTS

Of the 217 studies identified in our literature search, 24 relevant articles were selected. The median (interquartile range) sample size of included patient cohorts was 63 (157). Primary tumors of interest were lung (n = 14/24, 58.3%), lymphoma (n = 4/24, 16.7%), or melanoma (n = 4/24, 16.7%). A total of 28 treatment regimens were employed, including anti-PD-(L)1 (n = 13/28, 46.4%) and anti-CTLA-4 (n = 4/28, 14.3%) monoclonal antibodies. Predictive models were built from imaging features using univariate radiomics (n = 7/24, 29.2%), radiomics (n = 12/24, 50.0%), or deep learning (n = 5/24, 20.8%) and were most often used to prognosticate (n = 6/24, 25.0%) or describe tumor phenotype (n = 5/24, 20.8%). Eighteen studies (75.0%) performed AI model validation.

CONCLUSION

Preliminary results suggest broad potential for the application of AI-guided immunotherapy management after further validation of models on large, prospective, multicenter cohorts.

CLINICAL RELEVANCE STATEMENT

This scoping review describes how artificial intelligence models are built to make predictions based on medical imaging and explores their application specifically in the PET and SPECT examination of immunotherapy-treated cancers.

KEY POINTS

• Immunotherapy has drastically altered the cancer treatment landscape but is known to precipitate response patterns that are not accurately accounted for by traditional imaging methods. • There is an unmet need for better tools to not only facilitate in-treatment evaluation but also to predict, a priori, which patients are likely to achieve a good response with a certain treatment as well as those who are likely to develop side effects. • Artificial intelligence applied to PET/SPECT imaging of immunotherapy-treated patients is mainly used to make predictions about prognosis or tumor phenotype and is built from baseline, pre-treatment images. Further testing is required before a true transition to clinical application can be realized.

Machine learning-based identification of contrast-enhancement phase of computed tomography scans

Contrast-enhanced computed tomography scans (CECT) are routinely used in the evaluation of different clinical scenarios, including the detection and characterization of hepatocellular carcinoma (HCC). Quantitative medical image analysis has been an exponentially growing scientific field. A number of studies reported on the effects of variations in the contrast enhancement phase on the reproducibility of quantitative imaging features extracted from CT scans. The identification and labeling of phase enhancement is a time-consuming task, with a current need for an accurate automated labeling algorithm to identify the enhancement phase of CT scans. In this study, we investigated the ability of machine learning algorithms to label the phases in a dataset of 59 HCC patients scanned with a dynamic contrast-enhanced CT protocol. The ground truth labels were provided by expert radiologists. Regions of interest were defined within the aorta, the portal vein, and the liver. Mean density values were extracted from those regions of interest and used for machine learning modeling. Models were evaluated using accuracy, the area under the curve (AUC), and Matthew's correlation coefficient (MCC). We tested the algorithms on an external dataset (76 patients). Our results indicate that several supervised learning algorithms (logistic regression, random forest, etc.) performed similarly, and our developed algorithms can accurately classify the phase of contrast enhancement.

Ethnic diversity in treatment response for colorectal cancer: proof of concept for radiomics-driven enrichment trials

BACKGROUND

Several barriers hamper recruitment of diverse patient populations in multicenter clinical trials which determine efficacy of new systemic cancer therapies.

PURPOSE

We assessed if quantitative analysis of computed tomography (CT) scans of metastatic colorectal cancer (mCRC) patients using imaging features that predict overall survival (OS) can unravel the association between ethnicity and efficacy.

METHODS

We retrospectively analyzed CT images from 1584 mCRC patients in two phase III trials evaluating FOLFOX ± panitumumab (n = 331, 350) and FOLFIRI ± aflibercept (n = 437, 466) collected from August 2006 to March 2013. Primary and secondary endpoints compared RECIST1.1 response at month-2 and delta tumor volume at month-2, respectively. An ancillary study compared imaging phenotype using a peer-reviewed radiomics-signature combining 3 imaging features to predict OS landmarked from month-2. Analysis was stratified by ethnicity.

RESULTS

In total, 1584 patients were included (mean age, 60.25 ± 10.57 years; 969 men). Ethnicity was as follows: African (n = 50, 3.2%), Asian (n = 66, 4.2%), Caucasian (n = 1413, 89.2%), Latino (n = 27, 1.7%), Other (n = 28, 1.8%). Overall baseline tumor volume demonstrated Africans and Caucasians had more advanced disease (p < 0.001). Ethnicity was associated with treatment response. Response per RECIST1.1 at month-2 was distinct between ethnicities (p = 0.048) with higher response rate (55.6%) in Latinos. Overall delta tumor volume at month-2 demonstrated that Latino patients more likely experienced response to treatment (p = 0.021). Radiomics phenotype was also distinct in terms of tumor radiomics heterogeneity (p = 0.023).

CONCLUSION

This study highlights how clinical trials that inadequately represent minority groups may impact associated translational work. In appropriately powered studies, radiomics features may allow us to unravel associations between ethnicity and treatment efficacy, better elucidate mechanisms of resistance, and promote diversity in trials through predictive enrichment.

CLINICAL RELEVANCE STATEMENT

Radiomics could promote clinical trial diversity through predictive enrichment, hence benefit to historically underrepresented racial/ethnic groups that may respond variably to treatment due to socioeconomic factors and built environment, collectively referred to as social determinants of health.

KEY POINTS

•Findings indicate ethnicity was associated with treatment response across all 3 endpoints. First, response per RECIST1.1 at month-2 was distinct between ethnicities (p = 0.048) with higher response rate (55.6%) in Latinos. •Second, the overall delta tumor volume at month-2 demonstrated that Latino patients were more likely to experience response to treatment (p = 0.021). Radiomics phenotype was also distinct in terms of tumor radiomics heterogeneity (p = 0.023).

Recent Advances in the Field of Artificial Intelligence for Precision Medicine in Patients with a Diagnosis of Metastatic Cutaneous Melanoma

Standard-of-care medical imaging techniques such as CT, MRI, and PET play a critical role in managing patients diagnosed with metastatic cutaneous melanoma. Advancements in artificial intelligence (AI) techniques, such as radiomics, machine learning, and deep learning, could revolutionize the use of medical imaging by enhancing individualized image-guided precision medicine approaches. In the present article, we will decipher how AI/radiomics could mine information from medical images, such as tumor volume, heterogeneity, and shape, to provide insights into cancer biology that can be leveraged by clinicians to improve patient care both in the clinic and in clinical trials. More specifically, we will detail the potential role of AI in enhancing detection/diagnosis, staging, treatment planning, treatment delivery, response assessment, treatment toxicity assessment, and monitoring of patients diagnosed with metastatic cutaneous melanoma. Finally, we will explore how these proof-of-concept results can be translated from bench to bedside by describing how the implementation of AI techniques can be standardized for routine adoption in clinical settings worldwide to predict outcomes with great accuracy, reproducibility, and generalizability in patients diagnosed with metastatic cutaneous melanoma.

Multiparametric MRI in Era of Artificial Intelligence for Bladder Cancer Therapies

This review focuses on the principles, applications, and performance of mpMRI for bladder imaging. Quantitative imaging biomarkers (QIBs) derived from mpMRI are increasingly used in oncological applications, including tumor staging, prognosis, and assessment of treatment response. To standardize mpMRI acquisition and interpretation, an expert panel developed the Vesical Imaging-Reporting and Data System (VI-RADS). Many studies confirm the standardization and high degree of inter-reader agreement to discriminate muscle invasiveness in bladder cancer, supporting VI-RADS implementation in routine clinical practice. The standard MRI sequences for VI-RADS scoring are anatomical imaging, including T2w images, and physiological imaging with diffusion-weighted MRI (DW-MRI) and dynamic contrast-enhanced MRI (DCE-MRI). Physiological QIBs derived from analysis of DW- and DCE-MRI data and radiomic image features extracted from mpMRI images play an important role in bladder cancer. The current development of AI tools for analyzing mpMRI data and their potential impact on bladder imaging are surveyed. AI architectures are often implemented based on convolutional neural networks (CNNs), focusing on narrow/specific tasks. The application of AI can substantially impact bladder imaging clinical workflows; for example, manual tumor segmentation, which demands high time commitment and has inter-reader variability, can be replaced by an autosegmentation tool. The use of mpMRI and AI is projected to drive the field toward the personalized management of bladder cancer patients.

A Quantitative Multiparametric MRI Analysis Platform for Estimation of Robust Imaging Biomarkers in Clinical Oncology

There is a need to develop user-friendly imaging tools estimating robust quantitative biomarkers (QIBs) from multiparametric (mp)MRI for clinical applications in oncology. Quantitative metrics derived from (mp)MRI can monitor and predict early responses to treatment, often prior to anatomical changes. We have developed a vendor-agnostic, flexible, and user-friendly MATLAB-based toolkit, MRI-Quantitative Analysis and Multiparametric Evaluation Routines ("MRI-QAMPER", current release v3.0), for the estimation of quantitative metrics from dynamic contrast-enhanced (DCE) and multi-b value diffusion-weighted (DW) MR and MR relaxometry. MRI-QAMPER's functionality includes generating numerical parametric maps from these methods reflecting tumor permeability, cellularity, and tissue morphology. MRI-QAMPER routines were validated using digital reference objects (DROs) for DCE and DW MRI, serving as initial approval stages in the National Cancer Institute Quantitative Imaging Network (NCI/QIN) software benchmark. MRI-QAMPER has participated in DCE and DW MRI Collaborative Challenge Projects (CCPs), which are key technical stages in the NCI/QIN benchmark. In a DCE CCP, QAMPER presented the best repeatability coefficient (RC = 0.56) across test-retest brain metastasis data, out of ten participating DCE software packages. In a DW CCP, QAMPER ranked among the top five (out of fourteen) tools with the highest area under the curve (AUC) for prostate cancer detection. This platform can seamlessly process mpMRI data from brain, head and neck, thyroid, prostate, pancreas, and bladder cancer. MRI-QAMPER prospectively analyzes dose de-escalation trial data for oropharyngeal cancer, which has earned it advanced NCI/QIN approval for expanded usage and applications in wider clinical trials.

Integrating [18F]-Fluorodeoxyglucose Positron Emission Tomography with Computed Tomography with Radiation Therapy and Immunomodulation in Precision Therapy for Solid Tumors

[18F]-FDG positron emission tomography with computed tomography (PET/CT) imaging is widely used to enhance the quality of care in patients diagnosed with cancer. Furthermore, it holds the potential to offer insight into the synergic effect of combining radiation therapy (RT) with immuno-oncological (IO) agents. This is achieved by evaluating treatment responses both at the RT and distant tumor sites, thereby encompassing the phenomenon known as the abscopal effect. In this context, PET/CT can play an important role in establishing timelines for RT/IO administration and monitoring responses, including novel patterns such as hyperprogression, oligoprogression, and pseudoprogression, as well as immune-related adverse events. In this commentary, we explore the incremental value of PET/CT to enhance the combination of RT with IO in precision therapy for solid tumors, by offering supplementary insights to recently released joint guidelines.

Immunotherapy in the Treatment of Localized Genitourinary Cancers

Importance

A true revolution in the management of advanced genitourinary cancers has occurred with the discovery and adoption of immunotherapy (IO). The therapeutic benefits of IO were recently observed not to be solely confined to patients with disseminated disease but also in select patients with localized and locally advanced genitourinary neoplasms.

Observations

KEYNOTE-057 demonstrated the benefit of pembrolizumab monotherapy for treating high-risk nonmuscle invasive bladder cancer unresponsive to bacillus Calmette-Guérin (BCG), resulting in recent US Food and Drug Administration approval. Furthermore, a current phase 3 trial (Checkmate274) demonstrated a disease-free survival benefit with the administration of adjuvant nivolumab vs placebo in muscle-invasive urothelial carcinoma after radical cystectomy. In addition, the recent highly publicized phase 3 KEYNOTE 564 trial demonstrated a recurrence-free survival benefit of adjuvant pembrolizumab in patients with high-risk localized/locally advanced kidney cancer.

Conclusions and Relevance

The adoption and integration of IO in the management of localized genitourinary cancers exhibiting aggressive phenotypes are becoming an emerging therapeutic paradigm. Clinical oncologists and scientists should become familiar with these trials and indications because they are likely to dramatically change our treatment strategies in the months and years to come.

Modeling Tumor Growth Using Partly Conditional Survival Models: A Case Study in Colorectal Cancer

PURPOSE

There are multiple approaches to modeling the relationship between longitudinal tumor measurements obtained from serial imaging and overall survival. Many require strong assumptions that are untestable and debatable. We illustrate how to apply a novel, more flexible approach, the partly conditional (PC) survival model, using images acquired during a phase III, randomized clinical trial in colorectal cancer as an example.

METHODS

PC survival approaches were used to model longitudinal volumetric computed tomography data of 1,025 patients in the completed VELOUR trial, which evaluated adding aflibercept to infusional fluorouracil, leucovorin, and irinotecan for treating metastatic colorectal cancer. PC survival modeling is a semiparametric approach to estimating associations of longitudinal measurements with time-to-event outcomes. Overall survival was our outcome. Covariates included baseline tumor burden, change in tumor burden from baseline to each follow-up time, and treatment. Both unstratified and time-stratified models were investigated.

RESULTS

Without making assumptions about the distribution of the tumor growth process, we characterized associations between the change in tumor burden and survival. This change was significantly associated with survival (hazard ratio [HR], 1.04; 95% CI, 1.02 to 1.05; P < .001), suggesting that aflibercept works at least in part by altering the tumor growth trajectory. We also found baseline tumor size prognostic for survival even when accounting for the change in tumor burden over time (HR, 1.02; 95% CI, 1.01 to 1.02; P < .001).

CONCLUSION

The PC modeling approach offers flexible characterization of associations between longitudinal covariates, such as serially assessed tumor burden, and survival time. It can be applied to a variety of data of this nature and used as clinical trials are ongoing to incorporate new disease assessment information as it is accumulated, as indicated by an example from colorectal cancer.

A machine-learning-based combination of criteria to detect bladder cancer lymph node metastasis on [18F]FDG PET/CT: a pathology-controlled study

OBJECTIVES

Initial pelvic lymph node (LN) staging is pivotal for treatment planification in patients with muscle-invasive bladder cancer (MIBC), but [¹⁸F]FDG PET/CT provides insufficient and variable diagnostic performance. We aimed to develop and validate a machine-learning-based combination of criteria on [¹⁸F]FDG PET/CT to accurately identify pelvic LN involvement in bladder cancer patients.

METHODS

Consecutive patients with localized MIBC who performed preoperative [¹⁸F]FDG PET/CT between 2010 and 2017 were retrospectively assigned to training (n = 129) and validation (n = 44) sets. The reference standard was the pathological status after extended pelvic LN dissection. In the training set, a random forest algorithm identified the combination of criteria that best predicted LN status. The diagnostic performances (AUC) and interrater agreement of this combination of criteria were compared to a consensus of experts.

RESULTS

The overall prevalence of pelvic LN involvement was 24% (n = 41/173). In the training set, the top 3 features were derived from pelvic LNs (SUVmax of the most intense LN, and product of diameters of the largest LN) and primary bladder tumor (product of diameters). In the validation set, diagnostic performance did not differ significantly between the combination of criteria (AUC = 0.59 95%CI [0.43-0.73]) and the consensus of experts (AUC = 0.64 95%CI [0.48-0.78], p = 0.54). The interrater agreement was equally good with Κ = 0.66 for both.

CONCLUSION

The developed machine-learning-based combination of criteria performs as well as a consensus of experts to detect pelvic LN involvement on [¹⁸F]FDG PET/CT in patients with MIBC.

KEY POINTS

• The developed machine-learning-based combination of criteria performs as well as experts to detect pelvic LN involvement on [¹⁸F]FDG PET/CT in patients with muscle-invasive bladder cancer. • The top 3 features to predict LN involvement were the SUVmax of the most intense LN, the product of diameters of the largest LN, and the product of diameters of the primary bladder tumor.

Is local review of positron emission tomography scans sufficient in diffuse large B-cell lymphoma clinical trials? A CALGB 50303 analysis

BACKGROUND

Quantitative methods of Fluorodeoxyglucose Positron Emission Tomography (FDG-PET) interpretation, including the percent change in FDG uptake from baseline (ΔSUV), are under investigation in lymphoma to overcome challenges associated with visual scoring systems (VSS) such as the Deauville 5-point scale (5-PS).

METHODS

In CALGB 50303, patients with DLBCL received frontline R-CHOP or DA-EPOCH-R, and although there were no significant associations between interim PET responses assessed centrally after cycle 2 (iPET) using 5-PS with progression-free survival (PFS) or overall survival (OS), there were significant associations between central determinations of iPET ∆SUV with PFS/OS. In this patient cohort, we retrospectively compared local vs central iPET readings and evaluated associations between local imaging data and survival outcomes.

RESULTS

Agreement between local and central review was moderate (kappa = 0.53) for VSS and high (kappa = 0.81) for ∆SUV categories (<66% vs. ≥66%). ∆SUV ≥66% at iPET was significantly associated with PFS (p = 0.03) and OS (p = 0.002), but VSS was not. Associations with PFS/OS when applying local review vs central review were comparable.

CONCLUSIONS

These data suggest that local PET interpretation for response determination may be acceptable in clinical trials. Our findings also highlight limitations of VSS and call for incorporation of more objective measures of response assessment in clinical trials.

Baseline Radiomic Signature to Estimate Overall Survival in Patients With NSCLC

INTRODUCTION

We aimed to define a baseline radiomic signature associated with overall survival (OS) using baseline computed tomography (CT) images obtained from patients with NSCLC treated with nivolumab or chemotherapy.

METHODS

The radiomic signature was developed in patients with NSCLC treated with nivolumab in CheckMate-017, -026, and -063. Nivolumab-treated patients were pooled and randomized to training, calibration, or validation sets using a 2:1:1 ratio. From baseline CT images, volume of tumor lesions was semiautomatically segmented, and 38 radiomic variables depicting tumor phenotype were extracted. Association between the radiomic signature and OS was assessed in the nivolumab-treated (validation set) and chemotherapy-treated (test set) patients in these studies.

RESULTS

A baseline radiomic signature was identified using CT images obtained from 758 patients. The radiomic signature used a combination of imaging variables (spatial correlation, tumor volume in the liver, and tumor volume in the mediastinal lymph nodes) to output a continuous value, ranging from 0 to 1 (from most to least favorable estimated OS). Given a threshold of 0.55, the sensitivity and specificity of the radiomic signature for predicting 3-month OS were 86% and 77.8%, respectively. The signature was identified in the training set of patients treated with nivolumab and was significantly associated (p < 0.0001) with OS in patients treated with nivolumab or chemotherapy.

CONCLUSIONS

The radiomic signature provides an early readout of the anticipated OS in patients with NSCLC treated with nivolumab or chemotherapy. This could provide important prognostic information and may support risk stratification in clinical trials.

Emerging and Evolving Concepts in Cancer Immunotherapy Imaging

Criteria based on measurements of lesion diameter at CT have guided treatment with historical therapies due to the strong association between tumor size and survival. Clinical experience with immune checkpoint modulators shows that editing immune system function can be effective in various solid tumors. Equally, novel immune-related phenomena accompany this novel therapeutic paradigm. These effects of immunotherapy challenge the association of tumor size with response or progression and include risks and adverse events that present new demands for imaging to guide treatment decisions. Emerging and evolving approaches to immunotherapy highlight further key issues for imaging evaluation, such as dissociated response following local administration of immune checkpoint modulators, pseudoprogression due to immune infiltration in the tumor environment, and premature death due to hyperprogression. Research that may offer tools for radiologists to meet these challenges is reviewed. Different modalities are discussed, including immuno-PET, as well as new applications of CT, MRI, and fluorodeoxyglucose PET, such as radiomics and imaging of hematopoietic tissues or anthropometric characteristics. Multilevel integration of imaging and other biomarkers may improve clinical guidance for immunotherapies and provide theranostic opportunities.

High serum LDH and liver metastases are the dominant predictors of primary cancer resistance to anti-PD(L)1 immunotherapy

AIM

Anti-PD-(L)1 immunotherapies improve survival in multiple cancers but remain ineffective for most patients. We applied machine-learning algorithms and multivariate analyses on baseline medical data to estimate their relative impact on overall survival (OS) upon anti-PD-(L)1 monotherapies.

METHOD

This prognostic/predictive study retrospectively analysed 33 baseline routine medical variables derived from computed tomography (CT) images, clinical and biological meta-data. 695 patients with a diagnosis of advanced cancer were treated in prospective clinical trials in a single tertiary cancer centre in 3 cohorts including systemic anti-PD-(L)1 (251, 235 patients) versus other systemic therapies (209 patients). A random forest model combined variables to identify the combination (signature) which best estimated OS in patients treated with immunotherapy. The performance for estimating OS [95%CI] was measured using Kaplan-Meier Analysis and Log-Rank test.

RESULTS

Elevated serum lactate dehydrogenase (LDHhi) and presence of liver metastases (LM+) were dominant and independent predictors of short OS in independent cohorts of melanoma and non-melanoma solid tumours. Overall, LDHhiLM+ patients treated with anti-PD-(L)1 monotherapy had a poorer outcome (median OS: 3.1[2.4-7.8] months]) compared to LDHlowLM-patients (median OS: 15.3[8.9-NA] months; P < 0.0001). The OS of LDHlowLM-patients treated with immunotherapy was 28.8[17.9-NA] months (vs 13.1[10.8-18.5], P = 0.02) in the overall population and 30.3[19.93-NA] months (vs 14.1[8.69-NA], P = 0.0013) in patients with melanoma.

CONCLUSION

LDHhiLM+ status identifies patients who shall not benefit from anti-PD-(L)1 monotherapy. It could be used in clinical trials to stratify patients and eventually address this specific medical need.

The Impact of Image Acquisition Parameters and ComBat Harmonization on the Predictive Performance of Radiomics: A Renal Cell Carcinoma Model

Radiomics, one of the potential methods for developing clinical biomarker, is one of the exponentially growing research fields. In addition to its potential, several limitations have been identified in this field, and most importantly the effects of variations in imaging parameters on radiomic features (RFs). In this study, we investigate the potential of RFs to predict overall survival in patients with clear cell renal cell carcinoma, as well as the impact of ComBat harmonization on the performance of RF models. We assessed the robustness of the results by performing the analyses a thousand times. Publicly available CT scans of 179 patients were retrospectively collected and analyzed. The scans were acquired using different imaging vendors and parameters in different medical centers. The performance was calculated by averaging the metrics over all runs. On average, the clinical model significantly outperformed the radiomic models. The use of ComBat harmonization, on average, did not significantly improve the performance of radiomic models. Hence, the variability in image acquisition and reconstruction parameters significantly affect the performance of radiomic models. The development of radiomic specific harmonization techniques remain a necessity for the advancement of the field.

Artificial intelligence and radiomics: fundamentals, applications, and challenges in immunotherapy

Immunotherapy offers the potential for durable clinical benefit but calls into question the association between tumor size and outcome that currently forms the basis for imaging-guided treatment. Artificial intelligence (AI) and radiomics allow for discovery of novel patterns in medical images that can increase radiology’s role in management of patients with cancer, although methodological issues in the literature limit its clinical application. Using keywords related to immunotherapy and radiomics, we performed a literature review of MEDLINE, CENTRAL, and Embase from database inception through February 2022. We removed all duplicates, non-English language reports, abstracts, reviews, editorials, perspectives, case reports, book chapters, and non-relevant studies. From the remaining articles, the following information was extracted: publication information, sample size, primary tumor site, imaging modality, primary and secondary study objectives, data collection strategy (retrospective vs prospective, single center vs multicenter), radiomic signature validation strategy, signature performance, and metrics for calculation of a Radiomics Quality Score (RQS). We identified 351 studies, of which 87 were unique reports relevant to our research question. The median (IQR) of cohort sizes was 101 (57–180). Primary stated goals for radiomics model development were prognostication (n=29, 33.3%), treatment response prediction (n=24, 27.6%), and characterization of tumor phenotype (n=14, 16.1%) or immune environment (n=13, 14.9%). Most studies were retrospective (n=75, 86.2%) and recruited patients from a single center (n=57, 65.5%). For studies with available information on model testing, most (n=54, 65.9%) used a validation set or better. Performance metrics were generally highest for radiomics signatures predicting treatment response or tumor phenotype, as opposed to immune environment and overall prognosis. Out of a possible maximum of 36 points, the median (IQR) of RQS was 12 (10–16). While a rapidly increasing number of promising results offer proof of concept that AI and radiomics could drive precision medicine approaches for a wide range of indications, standardizing the data collection as well as optimizing the methodological quality and rigor are necessary before these results can be translated into clinical practice.

Prospective evaluation of 18F-FDG positron emission tomography in the preoperative staging of patients with hepatic colorectal metastases

Background

Despite considerable advances in preoperative imaging, up to one-third of patients operatively explored for hepatic colorectal metastases are unexpectedly found to harbor unresectable intrahepatic or extrahepatic disease.

Methods

The current study is a prospective, blinded study comparing utility of [¹⁸F]2-fluoro-2-deoxyglucose positron emission tomography (¹⁸F-FDG-PET) to computed tomography (CT) and CT arterial portography (CTAP) as preoperative staging.

Results

The 125 planned subjects were enrolled. Findings seen on FDG-PET alone changed therapy for 23 of 125 patients (18%). FDG-PET confirmed other radiologic findings in 16 cases (13%), for an overall influence on therapy in 39 cases (31%). FDG-PET was the most sensitive diagnostic imaging test for extrahepatic cancer; it was 80-90% sensitive for extrahepatic cancer and 70-90% specific. For the 28 cases of unresectable disease due to extrahepatic disease, FDG-PET findings solely changed therapies in 16 cases (57%) and influenced therapy in seven other cases (25%). Of the 21 unresectable cases due to extent of intrahepatic disease, FDG-PET did not solely change therapy in any. Overall, FDG-PET had the lowest sensitivity for hepatic sites compared with CT or CTAP. In particular, small (<1 cm) liver tumors were particularly poorly detected by FDG-PET. The area under the receiver operating characteristic (ROC) curve for small tumors was 0.58 and for patients on chemotherapy it was 0.66, a modest improvement over no imaging.

Conclusions

FDG-PET is an important test for preoperative staging of patients with hepatic colorectal metastases, affecting treatment decisions in nearly one-third of patients. The high yield is due mainly to detection of extrahepatic disease. It is therefore recommended in patients with extrahepatic lesions suspected to be disseminated cancer or those with high risk for extrahepatic disease. It is not a good test for identification of small tumors in the liver.

Novel biomarkers in NSCLC: Radiomic analysis, kinetic analysis, and circulating tumor DNA

Current radiographic methods of measuring treatment response for patients with nonsmall cell lung cancer have significant limitations. Recently, new modalities using standard of care images or minimally invasive blood-based DNA tests have gained interest as methods of evaluating treatment response. This article highlights three emerging modalities: radiomic analysis, kinetic analysis and serum-based measurement of circulating tumor DNA, with a focus on the clinical evidence supporting these methods. Additionally, we discuss the possibility of combining these modalities to develop a robust biomarker with strong correlation to clinically meaningful outcomes that could impact clinical trial design and patient care. At Last, we focus on how these methods specifically apply to a Veteran population.

Society for Immunotherapy of Cancer (SITC) clinical practice guideline on immunotherapy for the treatment of lung cancer and mesothelioma

Immunotherapy has transformed lung cancer care in recent years. In addition to providing durable responses and prolonged survival outcomes for a subset of patients with heavily pretreated non-small cell lung cancer (NSCLC), immune checkpoint inhibitors (ICIs)— either as monotherapy or in combination with other ICIs or chemotherapy—have demonstrated benefits in first-line therapy for advanced disease, the neoadjuvant and adjuvant settings, as well as in additional thoracic malignancies such as small-cell lung cancer (SCLC) and mesothelioma. Challenging questions remain, however, on topics including therapy selection, appropriate biomarker-based identification of patients who may derive benefit, the use of immunotherapy in special populations such as people with autoimmune disorders, and toxicity management. Patient and caregiver education and support for quality of life (QOL) is also important to attain maximal benefit with immunotherapy. To provide guidance to the oncology community on these and other important concerns, the Society for Immunotherapy of Cancer (SITC) convened a multidisciplinary panel of experts to develop a clinical practice guideline (CPG). This CPG represents an update to SITC’s 2018 publication on immunotherapy for the treatment of NSCLC, and is expanded to include recommendations on SCLC and mesothelioma. The Expert Panel drew on the published literature as well as their clinical experience to develop recommendations for healthcare professionals on these important aspects of immunotherapeutic treatment for lung cancer and mesothelioma, including diagnostic testing, treatment planning, immune-related adverse events, and patient QOL considerations. The evidence- and consensus-based recommendations in this CPG are intended to give guidance to cancer care providers using immunotherapy to treat patients with lung cancer or mesothelioma.

Early Readout on Overall Survival of Patients With Melanoma Treated With Immunotherapy Using a Novel Imaging Analysis

IMPORTANCE

Existing criteria to estimate the benefit of a therapy in patients with cancer rely almost exclusively on tumor size, an approach that was not designed to estimate survival benefit and is challenged by the unique properties of immunotherapy. More accurate prediction of survival by treatment could enhance treatment decisions.

OBJECTIVE

To validate, using radiomics and machine learning, the performance of a signature of quantitative computed tomography (CT) imaging features for estimating overall survival (OS) in patients with advanced melanoma treated with immunotherapy.

DESIGN, SETTING, AND PARTICIPANTS

This prognostic study used radiomics and machine learning to retrospectively analyze CT images obtained at baseline and first follow-up and their associated clinical metadata. Data were prospectively collected in the KEYNOTE-002 (Study of Pembrolizumab [MK-3475] Versus Chemotherapy in Participants With Advanced Melanoma; 2017 analysis) and KEYNOTE-006 (Study to Evaluate the Safety and Efficacy of Two Different Dosing Schedules of Pembrolizumab [MK-3475] Compared to Ipilimumab in Participants With Advanced Melanoma; 2016 analysis) multicenter clinical trials. Participants included 575 patients with a diagnosis of advanced melanoma who were randomly assigned to training and validation sets. Data for the present study were collected from November 20, 2012, to June 3, 2019, and analyzed from July 1, 2019, to September 15, 2021.

INTERVENTIONS

KEYNOTE-002 featured trial groups testing intravenous pembrolizumab, 2 mg/kg or 10 mg/kg every 2 or every 3 weeks based on randomization, or investigator-choice chemotherapy; KEYNOTE-006 featured trial groups testing intravenous ipilimumab, 3 mg/kg every 3 weeks and intravenous pembrolizumab, 10 mg/kg every 2 or 3 weeks based on randomization.

MAIN OUTCOMES AND MEASURES

The performance of the signature CT imaging features for estimating OS at the month 6 posttreatment landmark in patients who received pembrolizumab was measured using an area under the time-dependent receiver operating characteristics curve (AUC).

RESULTS

A random forest model combined 25 imaging features extracted from tumors segmented on CT images to identify the combination (signature) that best estimated OS with pembrolizumab in 575 patients. The signature combined 4 imaging features, 2 related to tumor size and 2 reflecting changes in tumor imaging phenotype. In the validation set (287 patients treated with pembrolizumab), the signature reached an AUC for estimation of OS status of 0.92 (95% CI, 0.89-0.95). The standard method, Response Evaluation Criteria in Solid Tumors 1.1, achieved an AUC of 0.80 (95% CI, 0.75-0.84) and classified tumor outcomes as partial or complete response (93 of 287 [32.4%]), stable disease (90 of 287 [31.3%]), or progressive disease (104 of 287 [36.2%]).

CONCLUSIONS AND RELEVANCE

The findings of this prognostic study suggest that the radiomic signature discerned from conventional CT images at baseline and on first follow-up may be used in clinical settings to provide an accurate early readout of future OS probability in patients with melanoma treated with single-agent programmed cell death 1 blockade.

Predicting death or recurrence of portal hypertension symptoms after TIPS procedures

BACKGROUND

Accurate prediction of portal hypertension recurrence after transjugular intrahepatic portosystemic shunt (TIPS) placement will improve clinical decision-making.

PURPOSE

To evaluate if perioperative variables could predict disease-free survival (DFS) in cirrhotic patients with portal hypertension (PHT) treated with TIPS.

MATERIALS AND METHODS

We recruited 206 cirrhotic patients with PHT treated with TIPS, randomly assigned to training (n = 138) and validation (n = 68) sets. We recorded 7 epidemiological, 4 clinical, and 9 radiological variables. TIPS-distal end positioning (TIPS-DEP) measured the distance between the distal end of the stent and the hepatocaval junction on contrast-enhanced CT scans. In the training set, the signature was defined as the random forest for survival algorithm achieving the lowest error rate for the prediction of DFS which was landmarked 4 weeks after the TIPS procedure. In the training set, a simple to use scoring system was derived from variables selected by the signature. The primary endpoint was to assess if TIPS-DEP was associated with DFS. The secondary endpoint was to validate the scoring system in the validation set.

RESULTS

Overall, patients with TIPS-DEP ≥ 6 mm (n = 49) had a median DFS of 24.5 months vs. 72.8 months otherwise (n = 157, p = 0.004). In the training set, the scoring system was calculated by adding age ≥ 60 years old, Child-Pugh B or C, and TIPS-DEP ≥ 6 mm (1 point each) since the signature showed high DFS probability at 6.5 months post-landmark in patients that did not meet these criteria: 86%, 80%, and 78%, respectively. The hazard ratio [95 CI] between patients determined to be low-risk (< 2 points) and high-risk (≥ 2 points) was 2.30 [1.35-3.93] (p = 0.002) in the training set and 2.01 [0.94-4.32] (p = 0.072) in the validation set.

CONCLUSION

TIPS-DEP is an actionable radiological biomarker which can be combined with age and Child-Pugh score to predict death or PHT symptom recurrence after TIPS procedure.

KEY POINTS

• TIPS-DEP measurement was the third most important but only actionable variable for predicting DFS. • TIPS-DEP < 6 mm was associated with a DFS probability of 78% at 6.5 months post-landmark. • A simple scoring system calculated using age, Child-Pugh score, and TIPS-DEP predicted DFS after TIPS.

An imaging signature to predict outcome in metastatic colorectal cancer using routine computed tomography scans

BACKGROUND & AIMS

Quantitative analysis of computed tomography (CT) scans of patients with metastatic colorectal cancer (mCRC) can identify imaging signatures that predict overall survival (OS).

METHODS

We retrospectively analysed CT images from 1584 mCRC patients on two phase III trials evaluating FOLFOX ± panitumumab (n = 331, 350) and FOLFIRI ± aflibercept (n = 437, 466). In the training set (n = 720), an algorithm was trained to predict OS landmarked from month 2; the output was a signature value on a scale from 0 to 1 (most to least favourable predicted OS). In the validation set (n = 864), hazard ratios (HRs) evaluated the association of the signature with OS using RECIST1.1 as a benchmark of comparison.

RESULTS

In the training set, the selected signature combined three features - change in tumour volume, change in tumour spatial heterogeneity, and tumour volume - to predict OS. In the validation set, RECIST1.1 classified patients in three categories: response (n = 166, 19.2%), stable disease (n = 636, 73.6%), and progression (n = 62, 7.2%). The HR was 3.93 (2.79-5.54). Using the same distribution for the signature, the HR was 21.04 (14.88-30.58), showing an incremental prognostic separation. Stable disease by RECIST1.1 was reclassified by the signature along a continuum where patients belonging to the most and least favourable signature quartiles had a median OS of 40.73 (28.49 to NA) months (n = 94) and 7.03 (5.66-7.89) months (n = 166), respectively.

CONCLUSIONS

A signature combining three imaging features provides early prognostic information that can improve treatment decisions for individual patients and clinical trial analyses.

Toward Generalizability in the Deployment of Artificial Intelligence in Radiology: Role of Computation Stress Testing to Overcome Underspecification

The clinical deployment of artificial intelligence (AI) applications in medical imaging is perhaps the greatest challenge facing radiology in the next decade. One of the main obstacles to the incorporation of automated AI-based decision-making tools in medicine is the failure of models to generalize when deployed across institutions with heterogeneous populations and imaging protocols. The most well-understood pitfall in developing these AI models is overfitting, which has, in part, been overcome by optimizing training protocols. However, overfitting is not the only obstacle to the success and generalizability of AI. Underspecification is also a serious impediment that requires conceptual understanding and correction. It is well known that a single AI pipeline, with prescribed training and testing sets, can produce several models with various levels of generalizability. Underspecification defines the inability of the pipeline to identify whether these models have embedded the structure of the underlying system by using a test set independent of, but distributed identically, to the training set. An underspecified pipeline is unable to assess the degree to which the models will be generalizable. Stress testing is a known tool in AI that can limit underspecification and, importantly, assure broad generalizability of AI models. However, the application of stress tests is new in radiologic applications. This report describes the concept of underspecification from a radiologist perspective, discusses stress testing as a specific strategy to overcome underspecification, and explains how stress tests could be designed in radiology-by modifying medical images or stratifying testing datasets. In the upcoming years, stress tests should become in radiology the standard that crash tests have become in the automotive industry. Keywords: Computer Applications-General, Informatics, Computer-aided Diagnosis © RSNA, 2021.

Convolutional Neural Network Addresses the Confounding Impact of CT Reconstruction Kernels on Radiomics Studies

Achieving high feature reproducibility while preserving biological information is one of the main challenges for the generalizability of current radiomics studies. Non-clinical imaging variables, such as reconstruction kernels, have shown to significantly impact radiomics features. In this study, we retrain an open-source convolutional neural network (CNN) to harmonize computerized tomography (CT) images with various reconstruction kernels to improve feature reproducibility and radiomic model performance using epidermal growth factor receptor (EGFR) mutation prediction in lung cancer as a paradigm. In the training phase, the CNN was retrained and tested on 32 lung cancer patients’ CT images between two different groups of reconstruction kernels (smooth and sharp). In the validation phase, the retrained CNN was validated on an external cohort of 223 lung cancer patients’ CT images acquired using different CT scanners and kernels. The results showed that the retrained CNN could be successfully applied to external datasets with different CT scanner parameters, and harmonization of reconstruction kernels from sharp to smooth could significantly improve the performance of radiomics model in predicting EGFR mutation status in lung cancer. In conclusion, the CNN based method showed great potential in improving feature reproducibility and generalizability by harmonizing medical images with heterogeneous reconstruction kernels.

Deep learning for the prediction of early on-treatment response in metastatic colorectal cancer from serial medical imaging

In current clinical practice, tumor response assessment is usually based on tumor size change on serial computerized tomography (CT) scan images. However, evaluation of tumor response to anti-vascular endothelial growth factor therapies in metastatic colorectal cancer (mCRC) is limited because morphological change in tumor may occur earlier than tumor size change. Here we present an analysis utilizing a deep learning (DL) network to characterize tumor morphological change for response assessment in mCRC patients. We retrospectively analyzed 1,028 mCRC patients who were prospectively included in the VELOUR trial (NCT00561470). We found that DL network was able to predict early on-treatment response in mCRC and showed better performance than its size-based counterpart with C-Index: 0.649 (95% CI: 0.619,0.679) vs. 0.627 (95% CI: 0.567,0.638), p = 0.009, z-test. The integration of DL network with size-based methodology could further improve the prediction performance to C-Index: 0.694 (95% CI: 0.661,0.720), which was superior to size/DL-based-only models (all p < 0.001, z-test). Our study suggests that DL network could provide a noninvasive mean for quantitative and comprehensive characterization of tumor morphological change, which may potentially benefit personalized early on-treatment decision making.

Nivolumab Plus Ipilimumab vs Nivolumab for Previously Treated Patients With Stage IV Squamous Cell Lung Cancer: The Lung-MAP S1400I Phase 3 Randomized Clinical Trial

IMPORTANCE

Nivolumab plus ipilimumab is superior to platinum-based chemotherapy in treatment-naive advanced non-small cell lung cancer (NSCLC). Nivolumab is superior to docetaxel in advanced pretreated NSCLC.

OBJECTIVE

To determine whether the addition of ipilimumab to nivolumab improves survival in patients with advanced, pretreated, immunotherapy-naive squamous (Sq) NSCLC.

DESIGN, SETTING, AND PARTICIPANTS

The Lung Cancer Master Protocol (Lung-MAP) S1400I phase 3, open-label randomized clinical trial was conducted from December 18, 2015, to April 23, 2018, randomizing patients in a 1:1 ratio to nivolumab alone or combined with ipilimumab. The median follow-up in surviving patients was 29.5 months. The trial was conducted through the National Clinical Trials Network and included patients with advanced immunotherapy-naive SqNSCLC and a Zubrod score of 0 (asymptomatic) to 1 (symptomatic but completely ambulatory) with disease progression after standard platinum-based chemotherapy. Randomization was stratified by sex and number of prior therapies (1 vs 2 or more). Data were analyzed from May 3, 2018, to February 1, 2021.

INTERVENTIONS

Nivolumab, 3 mg/kg intravenously every 2 weeks, with or without ipilimumab, 1 mg/kg intravenously every 6 weeks, until disease progression or intolerable toxic effects.

MAIN OUTCOMES AND MEASURES

The primary end point was overall survival (OS). Secondary end points included investigator-assessed progression-free survival (IA-PFS) and response per Response Evaluation Criteria in Solid Tumors (RECIST) guidelines, version 1.1.

RESULTS

Of 275 enrolled patients, 252 (mean age, 67.5 years [range 41.8-90.3 years]; 169 men [67%]; 206 White patients [82%]) were deemed eligible (125 randomized to nivolumab/ipilimumab and 127 to nivolumab). The study was closed for futility at a planned interim analysis. Overall survival was not significantly different between the groups (hazard ratio [HR], 0.87; 95% CI, 0.66-1.16; P = .34). Median survival was 10 months (95% CI, 8.0-14.4 months) in the nivolumab/ipilimumab group and 11 months (95% CI, 8.6-13.7 months) in the nivolumab group. The IA-PFS HR was 0.80 (95% CI, 0.61-1.03; P = .09); median IA-PFS was 3.8 months (95% CI, 2.7-4.4 months) in the nivolumab/ipilimumab group and 2.9 months (95% CI, 1.8-4.0 months) in the nivolumab alone group. Response rates were 18% (95% CI, 12%-25%) with nivolumab/ipilimumab and 17% (95% CI, 10%-23%) with nivolumab. Median response duration was 28.4 months (95% CI, 4.9 months to not reached) with nivolumab/ipilimumab and 9.7 months with nivolumab (95% CI, 4.2-23.1 months). Grade 3 or higher treatment-related adverse events occurred in 49 of 124 patients (39.5%) who received nivolumab/ipilimumab and in 41 of 123 (33.3%) who received nivolumab alone. Toxic effects led to discontinuation in 31 of 124 patients (25%) on nivolumab/ipilimumab and in 19 of 123 (15%) on nivolumab.

CONCLUSIONS AND RELEVANCE

In this phase 3 randomized clinical trial, ipilimumab added to nivolumab did not improve outcomes in patients with advanced, pretreated, immune checkpoint inhibitor-naive SqNSCLC.

TRIAL REGISTRATION

ClinicalTrials.gov Identifier: NCT02785952.

Phase II study of durvalumab plus tremelimumab as therapy for patients with previously treated anti-PD-1/PD-L1 resistant stage IV squamous cell lung cancer (Lung-MAP substudy S1400F, NCT03373760)

INTRODUCTION

S1400F is a non-match substudy of Lung Cancer Master Protocol (Lung-MAP) evaluating the immunotherapy combination of durvalumab and tremelimumab to overcome resistance to anti-programmed death ligand 1 (PD-(L)1) therapy in patients with advanced squamous lung carcinoma (sq non-small-cell lung cancer (NSCLC)).

METHODS

Patients with previously treated sqNSCLC with disease progression after anti-PD-(L)1 monotherapy, who did not qualify for any active molecularly targeted Lung-MAP substudies, were eligible. Patients received tremelimumab 75 mg plus durvalumab 1500 mg once every 28 days for four cycles then durvalumab alone every 28 days until disease progression. The primary endpoint was the objective response rate (RECIST V.1.1). Primary and acquired resistance cohorts, defined as disease progression within 24 weeks versus ≥24 weeks of starting prior anti-PD-(L)1 therapy, were analyzed separately and an interim analysis for futility was planned after 20 patients in each cohort were evaluable for response.

RESULTS

A total of 58 eligible patients received drug, 28 with primary resistance and 30 with acquired resistance to anti-PD-(L)1 monotherapy. Grade ≥3 adverse events at least possibly related to treatment were seen in 20 (34%) patients. The response rate in the primary resistance cohort was 7% (95% CI 0% to 17%), with one complete and one partial response. No responses were seen in the acquired resistance cohort. In the primary and resistance cohorts the median progression-free survival was 2.0 months (95% CI 1.6 to 3.0) and 2.1 months (95% CI 1.6 to 3.2), respectively, and overall survival was 7.7 months (95% CI 4.0 to 12.0) and 7.6 months (95% CI 5.3 to 10.2), respectively.

CONCLUSION

Durvalumab plus tremelimumab had minimal activity in patients with advanced sqNSCLC progressing on prior anti-PD-1 therapy.Trial registration numberNCT03373760.

Distinguishing benign and malignant lesions on contrast-enhanced breast cone-beam CT with deep learning neural architecture search

PURPOSE

To utilize a neural architecture search (NAS) approach to develop a convolutional neural network (CNN) method for distinguishing benign and malignant lesions on breast cone-beam CT (BCBCT).

METHOD

165 patients with 114 malignant and 86 benign lesions were collected by two institutions from May 2012 to August 2014. The NAS method autonomously generated a CNN model using one institution's dataset for training (patients/lesions: 71/91) and validation (patients/lesions: 20/23). The model was externally tested on another institution's dataset (patients/lesions: 74/87), and its performance was compared with fine-tuned ResNet-50 models and two breast radiologists who independently read the lesions in the testing dataset without knowing lesion diagnosis.

RESULTS

The lesion diameters (mean ± SD) were 18.8 ± 12.9 mm, 22.7 ± 10.5 mm, and 20.0 ± 11.8 mm in the training, validation, and external testing set, respectively. Compared to the best ResNet-50 model, the NAS-generated CNN model performed three times faster and, in the external testing set, achieved a higher (though not statistically different) AUC, with sensitivity (95% CI) and specificity (95% CI) of 0.727, 80% (66-90%), and 60% (42-75%), respectively. Meanwhile, the performances of the NAS-generated CNN and the two radiologists' visual ratings were not statistically different.

CONCLUSIONS

Our preliminary results demonstrated that a CNN autonomously generated by NAS performed comparably to pre-trained ResNet models and radiologists in predicting malignant breast lesions on contrast-enhanced BCBCT. In comparison to ResNet, which must be designed by an expert, the NAS approach may be used to automatically generate a deep learning architecture for medical image analysis.

Predicting Tyrosine Kinase Inhibitor Treatment Response in Stage IV Lung Adenocarcinoma Patients With EGFR Mutation Using Model-Based Deep Transfer Learning

Background

For stage IV patients harboring EGFR mutations, there is a differential response to the first-line TKI treatment. We constructed three-dimensional convolutional neural networks (CNN) with deep transfer learning to stratify patients into subgroups with different response and progression risks.

Materials and Methods

From 2013 to 2017, 339 patients with EGFR mutation receiving first-line TKI treatment were included. Progression-free survival (PFS) time and progression patterns were confirmed by routine follow-up and restaging examinations. Patients were divided into two subgroups according to the median PFS (<=9 months, > 9 months). We developed a PFS prediction model and a progression pattern classification model using transfer learning from a pre-trained EGFR mutation classification 3D CNN. Clinical features were fused with the 3D CNN to build the final hybrid prediction model. The performance was quantified using area under receiver operating characteristic curve (AUC), and model performance was compared by AUCs with Delong test.

Results

The PFS prediction CNN showed an AUC of 0.744 (95% CI, 0.645–0.843) in the independent validation set and the hybrid model of CNNs and clinical features showed an AUC of 0.771 (95% CI, 0.676–0.866), which are significantly better than clinical features-based model (AUC, 0.624, P<0.01). The progression pattern prediction model showed an AUC of 0.762(95% CI, 0.643–0.882) and the hybrid model with clinical features showed an AUC of 0.794 (95% CI, 0.681–0.908), which can provide compensate information for clinical features-based model (AUC, 0.710; 95% CI, 0.582–0.839).

Conclusion

The CNN exhibits potential ability to stratify progression status in patients with EGFR mutation treated with first-line TKI, which might help make clinical decisions.

Don't overlook spleen glucose metabolism on [18F]-FDG PET/CT for cancer drug discovery and development

Fluorine-18-fluorodeoxyglucose ([18F]-FDG) positron emission tomography/computed tomography (PET/CT) is a useful tool that assesses glucose metabolism in tumor cells to help guide management of cancer patients. However, the clinical relevance of glucose metabolism in healthy tissues, including hematopoietic tissues such as the spleen, has been potentially overlooked. Recent studies suggested that spleen glucose metabolism could improve the management of different cancers. Overall, the current literature includes 1,157 patients, with a wide range of tumor types. The prognostic and/or predictive value of spleen metabolism have been demonstrated in a broad spectrum of therapies including surgery and systemic cancer therapies. Most of these studies showed that high spleen glucose metabolism at baseline is associated with a poor outcome while treatment-induce change in spleen glucose metabolism is a multi-faceted surrogate of cancer-related inflammation, which correlates with immunosuppressive tumor microenvironment as well as with immune activation. In this systematic review, we seek to unravel the prognostic/predictive significance of spleen glucose metabolism on [18F]-FDG PET/CT and discuss how it could potentially guide cancer patient management in the future.

Multi-site, multi-platform comparison of MRI T1 measurement using the system phantom

Recent innovations in quantitative magnetic resonance imaging (MRI) measurement methods have led to improvements in accuracy, repeatability, and acquisition speed, and have prompted renewed interest to reevaluate the medical value of quantitative T₁. The purpose of this study was to determine the bias and reproducibility of T₁ measurements in a variety of MRI systems with an eye toward assessing the feasibility of applying diagnostic threshold T₁ measurement across multiple clinical sites. We used the International Society of Magnetic Resonance in Medicine/National Institute of Standards and Technology (ISMRM/NIST) system phantom to assess variations of T₁ measurements, using a slow, reference standard inversion recovery sequence and a rapid, commonly-available variable flip angle sequence, across MRI systems at 1.5 tesla (T) (two vendors, with number of MRI systems n = 9) and 3 T (three vendors, n = 18). We compared the T₁ measurements from inversion recovery and variable flip angle scans to ISMRM/NIST phantom reference values using Analysis of Variance (ANOVA) to test for statistical differences between T₁ measurements grouped according to MRI scanner manufacturers and/or static field strengths. The inversion recovery method had minor over- and under-estimations compared to the NMR-measured T₁ values at both 1.5 T and 3 T. Variable flip angle measurements had substantially greater deviations from the NMR-measured T₁ values than the inversion recovery measurements. At 3 T, the measured variable flip angle T₁ for one vendor is significantly different than the other two vendors for most of the samples throughout the clinically relevant range of T₁. There was no consistent pattern of discrepancy between vendors. We suggest establishing rigorous quality control procedures for validating quantitative MRI methods to promote confidence and stability in associated measurement techniques and to enable translation of diagnostic threshold from the research center to the entire clinical community.

Radiomics Prediction of EGFR Status in Lung Cancer-Our Experience in Using Multiple Feature Extractors and The Cancer Imaging Archive Data

We investigated the performance of multiple radiomics feature extractors/software on predicting epidermal growth factor receptor mutation status in 228 patients with non–small cell lung cancer from publicly available data sets in The Cancer Imaging Archive. The imaging and clinical data were split into training (n = 105) and validation cohorts (n = 123). Two of the most cited open-source feature extractors, IBEX (1563 features) and Pyradiomics (1319 features), and our in-house software, Columbia Image Feature Extractor (CIFE) (1160 features), were used to extract radiomics features. Univariate and multivariate analyses were performed sequentially to predict EGFR mutation status using each individual feature extractor. Our univariate analysis integrated an unsupervised clustering method to identify nonredundant and informative candidate features for the creation of prediction models by multivariate analyses. In training, unsupervised clustering-based univariate analysis identified 5, 6, and 4 features from IBEX, Pyradiomics, and CIFE as candidate features, respectively. Multivariate prediction models using these features from IBEX, Pyradiomics, and CIFE yielded similar areas under the receiver operating characteristic curve of 0.68, 0.67, and 0.69. However, in validation, areas under the receiver operating characteristic curve of multivariate prediction models from IBEX, Pyradiomics, and CIFE decreased to 0.54, 0.56 and 0.64, respectively. Different feature extractors select different radiomics features, which leads to prediction models with varying performance. However, correlation between those selected features from different extractors may indicate these features measure similar imaging phenotypes associated with similar biological characteristics. Overall, attention should be paid to the generalizability of individual radiomics features and radiomics prediction models.

Current and Future Role of Medical Imaging in Guiding the Management of Patients With Relapsed and Refractory Non-Hodgkin Lymphoma Treated With CAR T-Cell Therapy

Chimeric antigen receptor (CAR) T-cells are a novel immunotherapy available for patients with refractory/relapsed non-Hodgkin lymphoma. In this indication, clinical trials have demonstrated that CAR T-cells achieve high rates of response, complete response, and long-term response (up to 80%, 60%, and 40%, respectively). Nonetheless, the majority of patients ultimately relapsed. This review provides an overview about the current and future role of medical imaging in guiding the management of non-Hodgkin lymphoma patients treated with CAR T-cells. It discusses the value of predictive and prognostic biomarkers to better stratify the risk of relapse, and provide a patient-tailored therapeutic strategy. At baseline, high tumor volume (assessed on CT-scan or on [18F]-FDG PET/CT) is a prognostic factor associated with treatment failure. Response assessment has not been studied extensively yet. Available data suggests that current response assessment developed on CT-scan or on [18F]-FDG PET/CT for cytotoxic systemic therapies remains relevant to estimate lymphoma response to CAR T-cell therapy. Nonetheless, atypical patterns of response and progression have been observed and should be further analyzed. The potential advantages as well as limitations of artificial intelligence and radiomics as tools providing high throughput quantitative imaging features is described.

Uncontrolled Confounders May Lead to False or Overvalued Radiomics Signature: A Proof of Concept Using Survival Analysis in a Multicenter Cohort of Kidney Cancer

Purpose

We aimed to explore potential confounders of prognostic radiomics signature predicting survival outcomes in clear cell renal cell carcinoma (ccRCC) patients and demonstrate how to control for them.

Materials and Methods

Preoperative contrast enhanced abdominal CT scan of ccRCC patients along with pathological grade/stage, gene mutation status, and survival outcomes were retrieved from The Cancer Imaging Archive (TCIA)/The Cancer Genome Atlas-Kidney Renal Clear Cell Carcinoma (TCGA-KIRC) database, a publicly available dataset. A semi-automatic segmentation method was applied to segment ccRCC tumors, and 1,160 radiomics features were extracted from each segmented tumor on the CT images. Non-parametric principal component decomposition (PCD) and unsupervised hierarchical clustering were applied to build the radiomics signature models. The factors confounding the radiomics signature were investigated and controlled sequentially. Kaplan-Meier curves and Cox regression analyses were performed to test the association between radiomics signatures and survival outcomes.

Results

183 patients of TCGA-KIRC cohort with available imaging, pathological, and clinical outcomes were included in this study. All 1,160 radiomics features were included in the first radiomics signature. Three additional radiomics signatures were then modelled in successive steps removing redundant radiomics features first, removing radiomics features biased by CT slice thickness second, and removing radiomics features dependent on tumor size third. The final radiomics signature model was the most parsimonious, unbiased by CT slice thickness, and independent of tumor size. This final radiomics signature stratified the cohort into radiomics phenotypes that are different by cancer-specific and recurrence-free survival; HR (95% CI) = 3.0 (1.5-5.7), p <0.05 and HR (95% CI) = 6.6 (3.1-14.1), p <0.05, respectively.

Conclusion

Radiomics signature can be confounded by multiple factors, including feature redundancy, image acquisition parameters like slice thickness, and tumor size. Attention to and proper control for these potential confounders are necessary for a reliable and clinically valuable radiomics signature.

Phase 2 Study of Talazoparib in Patients With Homologous Recombination Repair-Deficient Squamous Cell Lung Cancer: Lung-MAP Substudy S1400G

PURPOSE

This signal finding study (S1400G) was designed to evaluate the efficacy of talazoparib in advanced stage squamous cell lung cancer harboring homologous recombination repair deficiency.

PATIENTS AND METHODS

The full eligible population (FEP) had tumors with a deleterious mutation in any of the study-defined homologous recombination repair genes and without prior exposure to a PARP inhibitor. The primary analysis population (PAP) is a subset of FEP with alteration in ATM, ATR, BRCA1, BRCA2, or PALB2. Treatment consisted of talazoparib 1 mg daily continuously in 21-day cycles. A 2-stage design with exact 93% power and 1-sided 0.07 type I error required enrollment of 40 patients in the PAP in order to rule out an overall response rate (ORR) of 15% or less if the true ORR is ≥ 35%.

RESULTS

The study enrolled 47 patients in the FEP, of whom 24 were in the PAP. The median age for the FEP was 66.7 years; 83% were male and 85% white. ORR in the PAP was 4% (95% confidence interval [CI], 0, 21) with disease control rate of 54% (95% CI, 33, 74). Median progression-free survival and overall survival were 2.4 months (95% CI, 1.5-2.8) and 5.2 months (95% CI, 4.0-10), respectively. In the FEP, ORR was 11% (95% CI, 3.6, 23), the disease control rate was 51% (95% CI, 36, 66), and the median duration of response was 1.8 months (95% CI, 1.3, 4.2). Median progression-free and overall survival were 2.5 months and 5.7 months, respectively.

CONCLUSIONS

S1400G failed to show sufficient level of efficacy for single agent talazoparib in a biomarker defined subset of squamous lung cancer with homologous recombination repair deficiency.

Radiomics Response Signature for Identification of Metastatic Colorectal Cancer Sensitive to Therapies Targeting EGFR Pathway

BACKGROUND

The authors sought to forecast survival and enhance treatment decisions for patients with liver metastatic colorectal cancer by using on-treatment radiomics signature to predict tumor sensitiveness to irinotecan, 5-fluorouracil, and leucovorin (FOLFIRI) alone (F) or in combination with cetuximab (FC).

METHODS

We retrospectively analyzed 667 metastatic colorectal cancer patients treated with F or FC. Computed tomography quality was classified as high (HQ) or standard (SD). Four datasets were created using the nomenclature (treatment) - (quality). Patients were randomly assigned (2:1) to training or validation sets: FCHQ: 78:38, FCSD: 124:62, FHQ: 78:51, FSD: 158:78. Four tumor-imaging biomarkers measured quantitative radiomics changes between standard of care computed tomography scans at baseline and 8 weeks. Using machine learning, the performance of the signature to classify tumors as treatment sensitive or treatment insensitive was trained and validated using receiver operating characteristic (ROC) curves. Hazard ratio and Cox regression models evaluated association with overall survival (OS).

RESULTS

The signature (area under the ROC curve [95% confidence interval (CI)]) used temporal decrease in tumor spatial heterogeneity plus boundary infiltration to successfully predict sensitivity to antiepidermal growth factor receptor therapy (FCHQ: 0.80 [95% CI = 0.69 to 0.94], FCSD: 0.72 [95% CI = 0.59 to 0.83]) but failed with chemotherapy (FHQ: 0.59 [95% CI = 0.44 to 0.72], FSD: 0.55 [95% CI = 0.43 to 0.66]). In cetuximab-containing sets, radiomics signature outperformed existing biomarkers (KRAS-mutational status, and tumor shrinkage by RECIST 1.1) for detection of treatment sensitivity and was strongly associated with OS (two-sided P < .005).

CONCLUSIONS

Radiomics response signature can serve as an intermediate surrogate marker of OS. The signature outperformed known biomarkers in providing an early prediction of treatment sensitivity and could be used to guide cetuximab treatment continuation decisions.

Quantitative imaging metrics derived from magnetic resonance fingerprinting using ISMRM/NIST MRI system phantom: An international multicenter repeatability and reproducibility study

PURPOSE

To compare the bias and inherent reliability of the quantitative (T₁ and T₂ ) imaging metrics generated from the magnetic resonance fingerprinting (MRF) technique using the ISMRM/NIST system phantom in an international multicenter setting.

METHOD

ISMRM/NIST MRI system phantom provides standard reference T₁ and T₂ relaxation values (vendor-provided) for each of the 14 vials in T₁ and T₂ arrays. MRF-SSFP scans repeated over 30 days on GE 1.5 and 3.0 T scanners at three collaborative centers. MRF estimated T_1, and T₂ values averaged over 30 days were compared with the phantom vendor-provided and spin-echo (SE) based convention gold standard (GS) method. Repeatability and reproducibility were characterized by the within-case coefficient of variation (wCV) of the MRF data acquired over 30 days, along with the biases.

RESULT

For the wide ranges of MRF estimated T₁ values, vials #1-8 (T₁ relaxation time between 2033 and 184 ms) exhibited a wCV less than 3% and 4%, respectively, on 3.0 and 1.5 T scanners. T₂ values in vials #1-8 (T₂ relaxation, 1044-45 ms) have shown wCV to be <7% on both 3.0 and 1.5 T scanners. A stronger linear correlation overall for T₁ (R²  = 0.9960 and 0.9963 at center-1 and center-2 on 3.0 T scanner, and R²  = 0.9951 and 0.9988 at center-1 and center-3 on 1.5 T scanner) compared to T₂ (R²  = 0.9971 and 0.9972 at center-1 and center-2 on 3.0 T scanner, and R²  = 0.9815 and 0.9754 at center-1 and center-3 on 1.5 T scanner). Bland-Altman (BA) analysis showed MRF based T₁ and T₂ values were within the limit of agreement (LOA) except for one data point. The mean difference or bias and 95% lower bound (LB) and upper bound (UB) LOA are reported in the format; mean bias: 95% LB LOA: 95% UB LOA. The biases for T₁ values were 21.34: -50.00: 92.69, 21.32: -47.29: 89.94 ms, and for T₂ values were -19.88: -42.37: 2.61, -19.06: -43.58: 5.45 ms on 3.0 T scanner at center-1 and center-2, respectively. Similarly, on 1.5 T scanner biases for T₁ values were 26.54: -53.41: 106.50, 9.997: -51.94: 71.94 ms, and for T₂ values were -23.84: -135.40: 87.76, -37.30: 134.30: 59.73 ms at center-1 and center-3, respectively. Additionally, the correlation between the SE based GS and MRF estimated T₁ and T₂ values (R²  = 0.9969 and 0.9977) showed a similar trend as we observed between vendor-provided and MRF estimated T₁ and T₂ values (R²  = 0.9963 and 0.9972). In addition to correlation, BA analysis showed that all the vials are within the LOA between the GS and vendor-provided for the T₁ values and except one vial for T₂ . All the vials are within the LOA between GS and MRF except one vial in T₁ and T₂ array. The wCV for reproducibility was <3% for both T₁ and T₂ values in vials #1-8, for all the 14 vials, wCV calculated for reproducibility was <4% for T₁ values and <5% for T₂ .

CONCLUSION

This study shows that MRF is highly repeatable (wCV <4% for T₁ and <7% for T₂ ) and reproducible (wCV < 3% for both T₁ and T₂ ) in certain vials (vials #1-8).

Artificial Intelligence and Early Detection of Pancreatic Cancer: 2020 Summative Review

ABSTRACT

Despite considerable research efforts, pancreatic cancer is associated with a dire prognosis and a 5-year survival rate of only 10%. Early symptoms of the disease are mostly nonspecific. The premise of improved survival through early detection is that more individuals will benefit from potentially curative treatment. Artificial intelligence (AI) methodology has emerged as a successful tool for risk stratification and identification in general health care. In response to the maturity of AI, Kenner Family Research Fund conducted the 2020 AI and Early Detection of Pancreatic Cancer Virtual Summit (www.pdac-virtualsummit.org) in conjunction with the American Pancreatic Association, with a focus on the potential of AI to advance early detection efforts in this disease. This comprehensive presummit article was prepared based on information provided by each of the interdisciplinary participants on one of the 5 following topics: Progress, Problems, and Prospects for Early Detection; AI and Machine Learning; AI and Pancreatic Cancer-Current Efforts; Collaborative Opportunities; and Moving Forward-Reflections from Government, Industry, and Advocacy. The outcome from the robust Summit conversations, to be presented in a future white paper, indicate that significant progress must be the result of strategic collaboration among investigators and institutions from multidisciplinary backgrounds, supported by committed funders.

Identifying Robust Radiomics Features for Lung Cancer by Using In-Vivo and Phantom Lung Lesions

We propose a novel framework for determining radiomics feature robustness by considering the effects of both biological and noise signals. This framework is preliminarily tested in a study predicting the epidermal growth factor receptor (EGFR) mutation status in non-small cell lung cancer (NSCLC) patients. Pairs of CT images (baseline, 3-week post therapy) of 46 NSCLC patients with known EGFR mutation status were collected and a FDA-customized anthropomorphic thoracic phantom was scanned on two vendors’ scanners at four different tube currents. Delta radiomics features were extracted from the NSCLC patient CTs and reproducible, non-redundant, and informative features were identified. The feature value differences between EGFR mutant and EGFR wildtype patients were quantitatively measured as the biological signal. Similarly, radiomics features were extracted from the phantom CTs. A pairwise comparison between settings resulted in a feature value difference that was quantitatively measured as the noise signal. Biological signals were compared to noise signals at each setting to determine if the distributions were significantly different by two-sample t-test, and thus robust. Four optimal features were selected to predict EGFR mutation status, Tumor-Mass, Sigmoid-Offset-Mean, Gabor-Energy and DWT-Energy, which quantified tumor mass, tumor-parenchyma density transition at boundary, line-like pattern inside tumor and intratumoral heterogeneity, respectively. The first three variables showed robustness across the majority of studied CT acquisition parameters. The textual feature DWT-Energy was less robust. The proposed framework was able to determine robustness of radiomics features at specific settings by comparing biological signal to noise signal. Identification of robust radiomics features may improve the generalizability of radiomics models in future studies.

Abstract PO022: Multidimensional longitudinal assessment of patients under treatment for advanced endometrial cancer: a new tool to advance research on human disease

Evolutionary biology-based methods to characterize human cancer increasingly inform screening, surveillance, and early stage disease treatment strategies. Advanced metastatic disease presents distinct challenges in research and effective treatment. We have established the Advanced Solid Tumor Registry at the Inova Schar Cancer Institute to systematically track quantitative behavior of advanced solid tumors in a “real world” treatment setting. The study supports collection of serial: plasma samples for ctDNA analysis, computed tomography digital images and semi-automated segmentation of lesions for tumor burden volume measurement, and collection of tumor marker measures among patients receiving routine, longitudinal treatment in a community oncology practice. At submission, the study has enrolled 53 patients, 6 with recurrent metastatic endometrial cancer. Study operations have been enhanced by development of R code and workflow to enable graphical display for individual patients in clinically relevant time. In addition to enhancing clinical decision-making for the individual patients, the collective endometrial cancer patient data currently cover 207 patient-months of observation, 119 circulating marker time-points, 73 CT scans, 27 individual lesions, and 20 different courses of treatment. Treatments included palliative radiation, and: cytotoxic, hormonal, immune-, and targeted therapeutics. We propose this registry represents a new tool to support application of evolutionary science-based methods to clinical care and reciprocally to collect “real world” data with sufficient detail to inform computational science assumptions and inferences in multi-scale modeling projects. In this pilot study, individual cases effectively display multiple observations relevant to understanding treatment response in advanced metastatic disease in the clinical care environment. This project was supported, in part, by R01-CA194783.

Citation Format: Erik Dvergsten, Sanja Karovic, Vasiliki Thomeas-McEwing, Danielle Kimble, Pingzhen Guo, Hao Yang, Binsheng Zhao, Lawrence H. Schwartz, Michael L. Maitland. Multidimensional longitudinal assessment of patients under treatment for advanced endometrial cancer: a new tool to advance research on human disease [abstract]. In: Proceedings of the AACR Virtual Special Conference: Endometrial Cancer: New Biology Driving Research and Treatment; 2020 Nov 9-10. Philadelphia (PA): AACR; Clin Cancer Res 2021;27(3_Suppl):Abstract nr PO022.

Eye Tracking for Deep Learning Segmentation Using Convolutional Neural Networks

Deep learning with convolutional neural networks (CNNs) has experienced tremendous growth in multiple healthcare applications and has been shown to have high accuracy in semantic segmentation of medical (e.g., radiology and pathology) images. However, a key barrier in the required training of CNNs is obtaining large-scale and precisely annotated imaging data. We sought to address the lack of annotated data with eye tracking technology. As a proof of principle, our hypothesis was that segmentation masks generated with the help of eye tracking (ET) would be very similar to those rendered by hand annotation (HA). Additionally, our goal was to show that a CNN trained on ET masks would be equivalent to one trained on HA masks, the latter being the current standard approach. Step 1: Screen captures of 19 publicly available radiologic images of assorted structures within various modalities were analyzed. ET and HA masks for all regions of interest (ROIs) were generated from these image datasets. Step 2: Utilizing a similar approach, ET and HA masks for 356 publicly available T1-weighted postcontrast meningioma images were generated. Three hundred six of these image + mask pairs were used to train a CNN with U-net-based architecture. The remaining 50 images were used as the independent test set. Step 1: ET and HA masks for the nonneurological images had an average Dice similarity coefficient (DSC) of 0.86 between each other. Step 2: Meningioma ET and HA masks had an average DSC of 0.85 between each other. After separate training using both approaches, the ET approach performed virtually identically to HA on the test set of 50 images. The former had an area under the curve (AUC) of 0.88, while the latter had AUC of 0.87. ET and HA predictions had trimmed mean DSCs compared to the original HA maps of 0.73 and 0.74, respectively. These trimmed DSCs between ET and HA were found to be statistically equivalent with a p value of 0.015. We have demonstrated that ET can create segmentation masks suitable for deep learning semantic segmentation. Future work will integrate ET to produce masks in a faster, more natural manner that distracts less from typical radiology clinical workflow.

Comparing RECIST 1.1 and iRECIST in advanced melanoma patients treated with pembrolizumab in a phase II clinical trial

OBJECTIVES

To compare tumor best overall response (BOR) by RECIST 1.1 and iRECIST, to explore the incidence of pseudoprogression in melanoma treated with pembrolizumab, and to assess the impact of pseudoprogression on overall survival (OS).

METHODS

A total of 221 patients with locally advanced/unresectable melanoma who received pembrolizumab as part of KEYNOTE-002 trial were included in this study. Radiological assessment of imaging was centrally reviewed to assess tumor response. Incidence of discordance in BOR between RECIST 1.1 and iRECIST as well as rate of pseudoprogression were measured. OS of patients with pseudoprogression was compared with that of those with uncontrolled disease.

RESULTS

Of the 221 patients in this cohort, 136 patients developed PD as per RECIST v1.1 and 78 patients with PD continued treatment and imaging beyond initial RECIST 1.1-defined PD. Among the 78 patients who continued therapy and imaging post-progression, RECIST 1.1 and iRECIST were discordant in 10 patients (12.8%) and pseudoprogression was encountered in 14 patients (17.9%). OS of patients with pseudoprogression was longer than that of patients with uncontrolled disease/true progression (29.9 months versus 8.0 months, p value < 0.001).

CONCLUSIONS

Effectiveness of immunotherapy in clinical trials depends on the criterion used to assess tumor response (RECIST 1.1 vs iRECIST) with iRECIST being more appropriate to detect pseudoprogression and potentially prevent premature termination of effective therapy. Pseudoprogression was associated with improved OS in comparison with that of patients with uncontrolled disease.

KEY POINTS

• Discordance between iRECIST and RECIST 1.1 was found in 12.8% of unresectable melanoma patients on pembrolizumab who continued therapy beyond initial RECIST 1.1-defined progression. • Pseudoprogression, captured with iRECIST, occurred in 17.9% and was significantly associated with improved overall survival in comparison with uncontrolled disease.