Artificial Intelligence-Guided Segmentation and Path Planning Software for Transthoracic Lung Biopsy.

PURPOSE: To validate the sensitivity and specificity of a 3-dimensional (3D) convolutional neural network (CNN) artificial intelligence (AI) software for lung lesion detection and to establish concordance between AI-generated needle paths and those used in actual biopsy procedures. MATERIALS AND METHODS: This was a retrospective study using computed tomography (CT) scans from 3 hospitals. Inclusion criteria were scans with 1-5 nodules of diameter ≥5 mm; exclusion criteria were poor-quality scans or those with nodules measuring <5mm in diameter. In the lesion detection phase, 2,147 nodules from 219 scans were used to develop and train the deep learning 3D-CNN to detect lesions. The 3D-CNN was validated with 235 scans (354 lesions) for sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) analysis. In the path planning phase, Bayesian optimization was used to propose possible needle trajectories for lesion biopsy while avoiding vital structures. Software-proposed needle trajectories were compared with actual biopsy path trajectories from intraprocedural CT scans in 150 patients, with a match defined as an angular deviation of <5° between the 2 trajectories. RESULTS: The model achieved an overall AUC of 97.4% (95% CI, 96.3%-98.2%) for lesion detection, with mean sensitivity of 93.5% and mean specificity of 93.2%. Among the software-proposed needle trajectories, 85.3% were feasible, with 82% matching actual paths and similar performance between supine and prone/oblique patient orientations (P = .311). The mean angular deviation between matching trajectories was 2.30° (SD ± 1.22); the mean path deviation was 2.94 mm (SD ± 1.60). CONCLUSIONS: Segmentation, lesion detection, and path planning for CT-guided lung biopsy using an AI-guided software showed promising results. Future integration with automated robotic systems may pave the way toward fully automated biopsy procedures.

Percutaneous Treatments for Persistent Bronchopleural and Alveolar-Pleural Fistulae.

Percutaneous glue embolization was investigated as a treatment for bronchopleural fistulae (BPFs) and alveolar-pleural fistulae (APFs) associated with persistent air leak. Seven consecutive patients with persistent air leak were treated with percutaneous glue embolization of the BPF/APF from both iatrogenic and spontaneous causes. Treatment was performed using direct n-butyl cyanoacrylate (nBCA) glue injection for discrete, visible fistulae (n = 4), fibrin glue spray for suspected tiny multifocal leaks (n = 2), or both (n = 1). The number of treatments required per patient was 1 (n = 3), 2 (n = 3), or 3 (n = 1). Technical success was achieved in all cases. Follow-up showed resolution of all air leaks, with mean chest tube removal at 7.1 days after the embolization. The follow-up duration ranged from 2 to 47 months. No significant procedure-related morbidity, mortality, or recurrence was encountered. Percutaneous treatment for persistent BPFs and APFs showed good efficacy in this small case series and warrants further investigation.

Phantom and Animal Study of a Robot-Assisted, CT-Guided Targeting System using Image-Only Navigation for Stereotactic Needle Insertion without Positional Sensors.

PURPOSE: To evaluate the feasibility and accuracy of a robotic system to integrate and map computed tomography (CT) and robotic coordinates, followed by automatic trajectory execution by a robotic arm. The system was hypothesized to achieve a targeting error of <5 mm without significant influence from variations in angulation or depth. MATERIALS AND METHODS: An experimental study was conducted using a robotic system (Automated Needle Targeting device for CT [ANT-C]) for needle insertions into a phantom model on both moving patient table and moving gantry CT scanners. Eight spherical markers were registered as targets for 90 insertions at different trajectories. After a single ANT-C registration, the closed-loop software targeted multiple markers via the insertion of robotically aligned 18-gauge needles. Accuracy (distance from the needle tip to the target) was assessed by postinsertion CT scans. Similar procedures were repeated to guide 10 needle insertions into a porcine lung. A regression analysis was performed to test the effect of needle angulation and insertion depth on the accuracy of insertion. RESULTS: In the phantom model, all needle insertions (median trajectory depth, 64.8 mm; range, 46.1-153 mm) were successfully performed in single attempts. The overall accuracy was 1.36 mm ± 0.53, which did not differ between the 2 types of CT scanners (1.39 mm ± 0.54 [moving patient table CT] vs 1.33 mm ± 0.52 [moving gantry CT]; P = .54) and was not significantly affected by the needle angulation and insertion depth. The accuracy for the porcine model was 9.09 mm ± 4.21. CONCLUSIONS: Robot-assisted needle insertion using the ANT-C robotic device was feasible and accurate for targeting multiple markers in a phantom model.

Randomized Controlled Trial Comparing Drug-coated Balloon Angioplasty versus Conventional Balloon Angioplasty for Treating Below-the-Knee Arteries in Critical Limb Ischemia: The SINGA-PACLI Trial.

Background There is a paucity of randomized trials demonstrating superior efficacy of drug-coated balloon angioplasty (DCBA) compared with conventional percutaneous transluminal angioplasty (PTA) for below-the-knee arterial disease in patients with -critical limb ischemia. Purpose To compare DCBA versus PTA for below-the-knee lesions in participants with critical limb ischemia through 12 months. Materials and Methods In this prospective, randomized, two-center, double-blind superiority study, participants with critical limb ischemia with rest pain or tissue loss with atherosclerotic disease in the native below-the-knee arteries were randomly assigned (in a one-to-one ratio) to DCBA or PTA after stratification for diabetes and renal failure between November 2013 and October 2017. The primary efficacy end point was angiographic primary patency at 6 months analyzed on an intention-to-treat basis. Secondary end points through 12 months were composed of major adverse events including death and major amputations, wound healing, limb salvage, clinically driven target-lesion revascularization, and amputation-free survival. Primary and binary secondary end points, analyzed by using generalized-linear model and time-to-event analyses, were estimated with Kaplan-Meier survival curves and hazard ratios (Cox regression). Results Seventy participants (mean age, 61 years ± 10 [standard deviation]; 43 men) in the DCBA group and 68 (mean age, 64 years ± 10; 50 men) in the PTA group were evaluated. The percentage of patients with angiographic primary patency at 6 months was 43% (30 of 70) in the DCBA group and 38% (26 of 68) in the PTA group (P = .48). Through 12 months, the percentage of deaths was similar: 21% in the DCBA group and 16% in the PTA group (P = .43). Amputation-free survival rate assessed with Kaplan-Meier curves differed through 12 months: 59% (41 of 70) in the DCBA group compared with 78% (53 of 68) in the PTA group (P = .01). Conclusion In participants with critical limb ischemia, the drug-coated balloon angioplasty group and the conventional percutaneous transluminal angioplasty group had similar primary patency rates at 6 months after treatment of below-the-knee arteries. Amputation-free survival rates through 12 months were higher in the percutaneous transluminal angioplasty group. © RSNA, 2021 Online supplemental material is available for this article.

Intra-Arterial Therapy for Unresectable Colorectal Liver Metastases: A Meta-Analysis.

PURPOSE: To evaluate the efficacy of hepatic arterial infusion (HAI), conventional transarterial chemoembolization, drug-eluting embolic transarterial chemoembolization (DEE-TACE), transarterial radioembolization, and their combinations with systemic chemotherapy (SCT) for unresectable colorectal liver metastases. METHODS: A search was conducted on Embase, Scopus, PubMed, and Web of Science for prospective nonrandomized studies and randomized controlled trials (RCTs) from inception to June 20, 2020. Survival data of patients were recovered from original Kaplan-Meier curves by exploiting a graphical reconstructive algorithm. One-stage meta-analyses were conducted for the median overall survival (OS), survival rates (SRs), and restricted mean survival time (RMST), whereas two-stage meta-analyses of proportions were conducted to determine response rates (RRs) and conversion to resection rates (CRRs). RESULTS: A total of 71 prospective nonrandomized studies and 21 RCTs were identified, comprising 6,695 patients. Among patients treated beyond the first-line, DEE-TACE + SCT (n = 152) had the best survival outcomes of median OS of 26.5 (95% confidence interval [CI], 22.5-29.1) months and a 3-year RMST of 23.6 (95% CI, 21.8-25.5) months. Upon further stratification by publication year, DEE-TACE + SCT appeared to consistently have the highest pooled SRs at 1 year (81.9%) and 2 years (66.1%) in recent publications (2015-2020). DEE-TACE + SCT and HAI + SCT had the highest pooled RRs of 56.7% (I2 = 0.90) and 62.6% (I2 = 0.87) and pooled CRRs of 35.5% (I2 = 0.00) and 30.3% (I2 = 0.80), respectively. CONCLUSIONS: Albeit significant heterogeneity, the paucity of high-quality evidence, and the noncomparative nature of all analyses, the overall evidence suggests that patients treated with DEE-TACE + SCT have the best oncological outcomes and greatest potential to be converted for resection.

Drug Coated Balloons for Dysfunctional Haemodialysis Venous Access: A Patient Level Meta-Analysis of Randomised Controlled Trials.

OBJECTIVE: To perform an individual patient data level meta-analysis of randomised controlled trials comparing drug coated balloon angioplasty (DCB) against conventional percutaneous transluminal angioplasty (PTA) in the treatment of dysfunctional haemodialysis venous access. METHODS: A search was conducted from inception to 13 November 2020. Kaplan-Meier curves comparing DCB with PTA by target lesion primary patency (TLPP) and access circuit primary patency (ACPP) were graphically reconstructed to retrieve patient level data. One stage meta-analyses with Cox models with random effects shared frailties were conducted to determine hazard ratios (HRs). Dynamic restricted mean survival times (RMST) were conducted in view of violation of the proportional hazards assumption. Conventional two stage meta-analyses and network meta-analyses under random effects Frequentist models were conducted to determine overall and comparative outcomes of paclitaxel concentrations used. Where outliers were consistently detected through outlier and influence analyses, sensitivity analyses excluding those studies were conducted. RESULTS: Among 10 RCTs (1 207 patients), HRs across all models favoured DCB (one stage shared frailty HR 0.62, 95% CI 0.53 - 0.73, p < .001; two stage random effects HR 0.60, 95% CI 0.42 - 0.86, p = .018, I2 = 65%) for TLPP. Evidence of time varying effects (p = .005) was found. TLPP RMST was + 3.54 months (25.0%) longer in DCB treated patients compared with PTA (p = .001) at three years. TLPP at six months, one year, and two years was 75.3% vs. 58.1%, 51.1% vs. 37.1%, and 31.4% vs. 26.0% for DCB and PTA, respectively. The P-Scores within the Frequentist network meta-analysis suggest that higher concentrations of paclitaxel were associated with better TLPP and ACPP. Among six RCTs (854 patients), the one stage model favoured DCB (shared frailty HR 0.72, 95% CI 0.60 - 0.87, p < .001) for ACPP. Conversely, the two stage random effects model demonstrated no significant difference (HR 0.76, 95% CI 0.35 - 1.67, p = .41, I2 = 81%). Sensitivity analysis excluding outliers significantly favoured DCB (HR 0.61, 95% CI 0.41 - 0.91, p = .027, I2 = 62%). CONCLUSION: Overall evidence suggests that DCB is favoured over PTA in TLPP and ACPP.

Preparing IR for COVID-19: The Singapore Experience.

This paper describes country-wide special measures undertaken for interventional radiology staff during the current coronavirus disease 2019 (COVID-19) pandemic. Although each interventional radiology service around the world faces unique challenges, the principles outlined in this article will be useful when designing or strengthening individual practices and integrating them within wider hospital and national measures. Moving beyond the current outbreak, these measures will be useful for any future infectious diseases which are likely to arise.

Hemodialysis Arteriovenous Fistula and Graft Stenoses: Randomized Trial Comparing Drug-eluting Balloon Angioplasty with Conventional Angioplasty.

Purpose To compare lesion primary patency and restenosis rates between drug-eluting balloon (DEB) percutaneous transluminal angioplasty (PTA) and conventional balloon PTA (cPTA) in the treatment of arteriovenous fistula (AVF) and arteriovenous graft (AVG) stenosis. Materials and Methods In this prospective study, 119 participants (mean age, 59.2 years; 79 men, 40 women) with failing AVFs (n = 98) or AVGs (n = 21) were randomly assigned to undergo either DEB PTA (n = 59) or cPTA (n = 60) from January 2012 to May 2013. Primary end points were lesion primary patency and restenosis rates at 6 months; secondary outcomes were anatomic and clinical success after PTA, circuit primary patency at 6 months and 1 year, and lesion primary patency at 1 year. Statistical analysis was performed by using the Kaplan-Meier product limit estimator, and hazard ratio was calculated by using Cox proportional hazards regression. Complication rates were assessed in both groups. Results Estimated lesion primary patency in the DEB PTA and cPTA arms was 0.81 and 0.61, respectively, at 6 months (P = .03) and 0.51 and 0.34, respectively, at 1 year (P = .04). Estimated circuit primary patency in the DEB PTA and cPTA arms was 0.76 and 0.56, respectively, at 6 months (P = .048) and 0.45 and 0.32, respectively, at 1 year (P = .16). Restenosis rate was 34.0% (16 of 47) for DEB PTA and 62.9% (22 of 35) for cPTA at 6 months (P = .01). No major complications were noted. Conclusion Drug-eluting balloon angioplasty was effective in prolonging lesion primary patency of dialysis access stenoses at 6 months and 1 year. © RSNA, 2018.

Diagnosis and Management of Complications from Percutaneous Biliary Tract Interventions.

Complications related to percutaneous biliary tract interventions (PBTIs) can range from access site discomfort to life-threatening vascular complications. These complications are relatively uncommon, and most of them are self-limiting. However, major complications for which an increased level of patient care and/or a prolonged hospital stay are required and that may lead to death-albeit rarely-can occur. Some of the most common complications related to PBTI include pain, infection, bile leakage, and catheter blockage. These conditions can be easily recognized by using the patient's clinical history and laboratory examination results. However, the more uncommon complications, such as life-threatening hemobilia, acute pancreatitis, and catheter and stent fractures, may have nonspecific clinical manifestations, and the underlying pathologic condition may be found only when it is being sought specifically. It is important that diagnostic and interventional radiologists be aware of the wide spectrum of PBTI-related complications, as early recognition and treatment may prevent catastrophic situations. In addition, knowledge of the different treatment options is essential for guidance in interventional radiology procedures such as transarterial control of hemobilia, imaging-guided direct percutaneous embolization of pseudoaneurysms, and percutaneous treatment of catheter- and stent-related complications such as fractures. The authors review a wide spectrum of complications associated with PBTI and the percutaneous management of these conditions. They also highlight valuable lessons learned from morbidity and mortality rounds at a high-volume tertiary care center. ©RSNA, 2017.

Treatment of a spinal aneurysmal bone cyst using combined image-guided cryoablation and cementoplasty.

The authors describe the case of a 6.6-cm symptomatic spinal aneurysmal bone cyst (ABC) in a 17-year-old athlete treated percutaneously. Surgical treatment was not considered as the first option owing to its invasiveness and associated morbidity. CT-guided cryoablation of the expansile part of the ABC was performed for tumour shrinkage and nerve decompression. Thermal insulation, temperature monitoring and functional control/electrostimulation of the neural structures at risk were applied. Finally, the bony defect was cemented. No complications occurred during the procedure. Complete resolution of the ABC on imaging and clinical improvement were achieved. Percutaneous cryoablation should be considered as an alternative treatment option, especially when tumour size reduction is desired.

Comparison of acetabular version angle measurements between prone and reformatted supine computed tomography images.

PURPOSE: To compare acetabular version angle measurements of CT scans in the prone and reformatted supine positions. CT acetabular version angle measurements have previously been done in the prone position to correct for pelvic tilt. With the advent of multidetector CT, recent studies have evaluated acetabular version angles measured in the supine position. To our knowledge, a comparison between these two approaches has not been performed. STUDY DESIGN: Case series in which consecutive CT urography studies of 49 adult patients performed in both prone and supine positions were retrospectively reviewed, and acetabular version angles of both hips measured. METHOD: Retrospective review of 49 consecutive CT urography studies performed in both prone and supine positions was done, and acetabular version angles of both hips were measured. Two radiologists measured the acetabular version angles independently. Multiplanar reformation of the supine CT images was performed to compensate for pelvic tilt and rotation prior to angle measurements. RESULTS: There was excellent interobserver agreement between the two readers (ICC = 0.90). Acetabular version angle measurements from the prone CT images were larger compared to reformatted supine images (24.0 and 21.3°, respectively, p < 0.0001), with greater angles found in women. There was strong correlation between supine and prone acetabular version angle measurements with a Pearson correlation coefficient of 0.743. CONCLUSIONS: Acetabular version angles measured from prone and reformatted supine CT images show strong correlation but are significantly different with larger angles obtained from the former and in women; clinical implications of these findings may require further study in other to determine the best method of version angle measurement. CT acetabular version angle measurement is also reliable with excellent interobserver correlation.

The Role of Drug-Coated Balloon in Haemodialysis Arteriovenous Fistula Stenosis Management.

Arteriovenous fistula (AVF) stenosis is a common problem leading to dialysis access dysfunction. The conventional balloon (CB) is the most commonly used device during angioplasty but suffers from poor durability of results due to neointimal hyperplasia-mediated recurrence. The drug-coated balloon (DCB) is an adjunct to balloon angioplasty that reduces neointimal hyperplasia, thereby improving post-angioplasty patency. Despite the heterogeneity of DCB clinical trials to date, the evidence suggests that DCBs of different brands are not necessarily equal, and that patient selection, adequate lesion preparation and proper DCB procedural technique are important to realize the benefit of DCB angioplasty.

Impact of catheter tip to hepatic vein ostium distance on the validity and prognostication of hepatic venous pressure gradient in cirrhosis.

Hepatic venous pressure gradient (HVPG) is an accurate measure of portal hypertension in cirrhosis. However, the effect of catheter tip distance from hepatic vein ostium (HVO) on HVPG is unknown. We performed a retrospective study on 228 patients with 307 HVPGs in our institution. The objectives of this study were to assess the effect of catheter position on the validity of HVPG and its prognostication in cirrhosis. In this study, free hepatic vein pressure (FHVP) was considered optimal when difference between FHVP and inferior vena cava pressure was ≤ 2 mmHg. HVPG progressively decreased (p < 0.001) when measured at increasing distance from HVO due to an increasing FHVP (p = 0.036) but an unchanged wedged hepatic vein pressure (p = 0.343). Catheter tip distance > 5 to ≤ 8 cm [odds ratio {OR} 0.16 (95% CI 0.05-0.47), p = 0.001] and > 8 cm [OR 0.14 (95% CI 0.04-0.47), p = 0.002] compared to ≤ 3 cm from HVO were independent predictors of not achieving optimal FHVP. Baseline HVPG ≥ 16 mmHg was strongly associated with deaths due to cirrhosis and liver transplantation for end-stage liver disease compared to HVPG < 16 mmHg when FHVP was optimal (p < 0.001) but not when it was suboptimal (p = 0.359). Our study showed that FHVP is spuriously elevated when measured at > 5 cm from HVO, resulting in inaccurately low HVPG.

First-Line Systemic Therapies for Advanced Hepatocellular Carcinoma: A Systematic Review and Patient-Level Network Meta-Analysis.

Introduction: Sorafenib was historically the standard of care for advanced hepatocellular carcinoma (aHCC) until it was superseded by the combination of atezolizumab and bevacizumab. Thereafter, several novel first-line combination therapies have demonstrated favorable outcomes. The efficacies of these treatments in relation to current and previous standards of care are unknown, necessitating an overarching evaluation. Methods: A systematic literature search was conducted on PubMed, EMBASE, Scopus, and the Cochrane Controlled Register of Trials for phase III randomized controlled trials investigating first-line systemic therapies for aHCC. Kaplan-Meier curves for overall survival (OS) and progression-free survival (PFS) were graphically reconstructed to retrieve individual patient-level data. Derived hazard ratios (HRs) for each study were pooled in a random-effects network meta-analysis (NMA). NMAs were also conducted using study-level HRs for various subgroups, according to viral etiology, Barcelona Clinic Liver Cancer (BCLC) staging, alpha-fetoprotein (AFP) levels, macrovascular invasion, and extrahepatic spread. Treatment strategies were ranked using p scores. Results: Among 4,321 articles identified, 12 trials and 9,589 patients were included for analysis. Only two therapies showed OS benefit over sorafenib: combined anti-programmed-death and anti-VEGF pathway inhibitor monoclonal antibodies (Anti-PD-(L)1/VEGF Ab), including atezolizumab-bevacizumab and sintilimab-bevacizumab biosimilar (HR = 0.63, 95% CI = 0.53-0.76) and tremelimumab-durvalumab (HR = 0.78, 95% CI = 0.66-0.92). Anti-PD-(L)1/VEGF Ab showed OS benefit over all other therapies except tremelimumab-durvalumab. Low heterogeneity (I2 = 0%) and inconsistency (Cochran's Q = 0.52, p = 0.773) was observed. p scores for OS ranked Anti-PD-(L)1/VEGF Ab as the best treatment in all subgroups, except hepatitis B where atezolizumab-cabozantinib ranked highest for both OS and PFS, as well as nonviral HCC and AFP ≥400 µg/L where tremelimumab-durvalumab ranked highest for OS. Conclusion: This NMA supports Anti-PD-(L)1/VEGF Ab as the first-line therapy for aHCC and demonstrates a comparable benefit for tremelimumab-durvalumab which also extends to certain subgroups. Results of the subgroup analysis may guide treatment according to baseline characteristics, while pending further studies.

Targeted Anterior Cervical Epidural Blood Patch in a Patient with Spontaneous Intracranial Hypotension.

A 45-year-old male was admitted with severe orthostatic headache secondary to spontaneous intracranial hypotension. He had the site of cerebrospinal fluid (CSF) leakage identified at the anterolateral aspect of the C7-T1 spinal level. He first underwent a conventional posterior-approach cervical epidural blood patch (EBP) which provided immediate relief to the patient's symptoms; however, his symptoms recurred two days later. To better target the anterolateral leakage site, we employed an anterior-approach EBP under computed tomography (CT) guidance. After this attempt, the patient experienced complete relief of his symptoms, and the headache eventually resolved.

Anastomosing Hemangioma on 68Ga-DOTATATE PET/CT: A Potential Pitfall.

ABSTRACT: Anastomosing hemangioma (AH) is a rare benign vascular lesion that primarily involves the genitourinary tract. Cases have also rarely been reported in other organs. AH is often discovered incidentally and resembles angiosarcoma histologically. On imaging, it may mimic other vascular lesions such as renal cell carcinoma and neuroendocrine tumors. We present a case of a 32-year-old woman with incidentally detected AH involving the kidneys, adrenal glands, liver, and retroperitoneum, initially presumed to be neuroendocrine tumors based on imaging findings on CT and 68Ga-DOTATATE PET scans.

Recurrent Stenoses in Arteriovenous Fistula (AVF) for Dialysis Access: cuttIng ballooN angioplaSTy combined wITh paclitaxel drUg-coaTed balloon angioplasty, an observatIONal study (INSTITUTION Study).

PURPOSE: To study the safety and efficacy of cutting balloon angioplasty (CBA) followed by paclitaxel drug-coated balloon (PCB) angioplasty for recurrent venous lesions in arteriovenous fistulas (AVFs). MATERIALS AND METHODS: We conducted a prospective single-arm cohort study of CBA followed by PCB angioplasty for recurrent AVF stenoses between September 2017 and April 2019. In total, 44 participants were recruited. Target lesions were included if they had recurred within 12 months post-angioplasty, were > = 0.5 cm upstream from the arteriovenous anastomosis, and did not involve the central veins. Up to two non-target lesions per circuit/participant with the same definition were allowed. Lesions were considered separate when there was an intervening 2-cm segment of normal vessel. Technical success was defined as complete lesion effacement on angioplasty. End-points of target and circuit patency were evaluated clinically at 3, 6, and 12 months post-procedure. RESULT: Technical success was 96% (42/44): Two participants were excluded from analysis due to the need for high-pressure balloon angioplasty as the target lesions did not efface with CBA. The median follow-up duration was 337.5 days. Mean stenosis pre- and post-angioplasty was 69.0% (51.6-84.8) and 20.8% (0-44.8), respectively. The target lesion primary, primary assisted and circuit patency for the entire study population (n = 42) were 61.6 ± 7.8%, 92.7 ± 4.0%, and 54.7 ± 7.9%, respectively, at 12 months. For participants without non-target lesions (n = 22), the rates were 77.3 ± 8.9%, 90.9 ± 6.1%, and 60.7 ± 11.0%, respectively, at 12 months. CONCLUSION: CBA followed by PCB angioplasty appears safe and feasible for treatment of recurrent venous lesions in dysfunctional AVFs.

A Randomized Phase 2 Trial of Nivolumab Versus Nivolumab-Ipilimumab Combination in EGFR-Mutant NSCLC.

Introduction: Although immune checkpoint inhibitors (ICIs) have dramatically improved outcomes for nononcogene-addicted NSCLC, monotherapy with programmed cell death protein-1 (PD1) inhibition has been associated with low efficacy in the EGFR-mutant setting. Given the potential for synergism with combination checkpoint blockade, we designed a trial to test the activity of combination nivolumab (N)-ipilimumab (NI) in EGFR-mutant NSCLC. Methods: This is a randomized phase 2 study (NCT03091491) of N versus NI combination in EGFR tyrosine kinase inhibitor (TKI)-resistant NSCLC, with crossover permitted on disease progression. The primary end point was the objective response rate, and the secondary end points included progression-free survival, overall survival, and safety of ICI after EGFR TKI. Results: Recruitment ceased owing to futility after 31 of 184 planned patients were treated. A total of 15 patients received N and 16 received NI combination. There were 16 patients (51.6%) who had programmed death-ligand (PDL1) 1 greater than or equal to 1%, and 15 (45.2%) harbored EGFR T790M. Five patients derived clinical benefits from ICI with one objective response (objective response rate 3.2%), and median progression-free survival was 1.22 months (95% confidence interval: 1.15-1.35) for the overall cohort. None of the four patients who crossed over achieved salvage response by NI. PDL1 and tumor mutational burden (TMB) were not able to predict ICI response. Rates of all grade immune-related adverse events were similar (80% versus 75%), with only two grade 3 events. Conclusions: Immune checkpoint inhibition is ineffective in EGFR TKI-resistant NSCLC. Whereas a small subgroup of EGFR-mutant NSCLC may be immunogenic and responsive to ICI, better biomarkers are needed to select appropriate patients.