CT-based radiomics to distinguish progressive from stable neuroendocrine liver metastases treated with somatostatin analogues: an explorative study.

BACKGROUND: Accurate response evaluation in patients with neuroendocrine liver metastases (NELM) remains a challenge. Radiomics has shown promising results regarding response assessment. PURPOSE: To differentiate progressive (PD) from stable disease (SD) with radiomics in patients with NELM undergoing somatostatin analogue (SSA) treatment. MATERIAL AND METHODS: A total of 46 patients with histologically confirmed gastroenteropancreatic neuroendocrine tumors (GEP-NET) with ≥1 NELM and ≥2 computed tomography (CT) scans were included. Response was assessed with Response Evaluation Criteria in Solid Tumors (RECIST1.1). Hepatic target lesions were manually delineated and analyzed with radiomics. Radiomics features were extracted from each NELM on both arterial-phase (AP) and portal-venous-phase (PVP) CT. Multiple instance learning with regularized logistic regression via LASSO penalization (with threefold cross-validation) was used to classify response. Three models were computed: (i) AP model; (ii) PVP model; and (iii) AP + PVP model for a lesion-based and patient-based outcome. Next, clinical features were added to each model. RESULTS: In total, 19 (40%) patients had PD. Median follow-up was 13 months (range 1-50 months). Radiomics models could not accurately classify response (area under the curve 0.44-0.60). Adding clinical variables to the radiomics models did not significantly improve the performance of any model. CONCLUSION: Radiomics features were not able to accurately classify response of NELM on surveillance CT scans during SSA treatment.

Blood Molecular Genomic Analysis Predicts the Disease Course of Gastroenteropancreatic Neuroendocrine Tumor Patients: A Validation Study of the Predictive Value of the NETest®.

Reliable prediction of disease status is a major challenge in managing gastroenteropancreatic neuroendocrine tumors (GEP-NETs). The aim of the study was to validate the NETest®, a blood molecular genomic analysis, for predicting the course of disease in individual patients compared to chromogranin A (CgA). NETest® score (normal ≤20%) and CgA level (normal <100 µg/L) were measured in 152 GEP-NETs. The median follow-up was 36 (4-56) months. Progression-free survival was blindly assessed (Response Evaluation Criteria in Solid Tumors [RECIST] version 1.1). Optimal cutoffs (area under the receiver operating characteristic curve [AUC]), odds ratios, as well as negative and positive predictive values (NPVs/PPVs) were calculated for predicting stable disease (SD) and progressive disease (PD). Of the 152 GEP-NETs, 86% were NETest®-positive and 52% CgA-positive. -NETest® AUC was 0.78 versus CgA 0.73 (p = ns). The optimal cutoffs for predicting SD/PD were 33% for the NETest® and 140 µg/L for CgA. Multivariate analyses identified NETest® as the strongest predictor for PD (odds ratio: 5.7 [score: 34-79%]; 12.6 [score: ≥80%]) compared to CgA (odds ratio: 3.0), tumor grade (odds ratio: 3.1), or liver metastasis (odds ratio: 7.7). The NETest® NPV for SD was 87% at 12 months. The PPV for PD was 47 and 64% (scores 34-79% and ≥80%, respectively). NETest® metrics were comparable in the watchful waiting, treatment, and no evidence of disease (NED) subgroups. For CgA (>140 ng/mL), NPV and PPV were 83 and 52%. CgA could not predict PD in the watchful waiting or NED subgroups. The NETest® reliably predicted SD and was the strongest predictor of PD. CgA had lower utility. The -NETest® anticipates RECIST-defined disease status up to 1 year before imaging alterations are apparent.

Selection of Patients for Organ Preservation After Chemoradiotherapy: MRI Identifies Poor Responders Who Can Go Straight to Surgery.

OBJECTIVE: The aim of this study was to evaluate whether magnetic resonance imaging (MRI) can accurately identify poor responders after chemoradiotherapy (CRT) who will need to go straight to surgery, and to evaluate whether results are reproducible among radiologists with different levels of expertise. METHODS: Seven independent readers with different levels of expertise retrospectively evaluated the restaging MRIs (T2-weighted + diffusion-weighted imaging [T2W + DWI]) of 62 patients and categorized them as (1) poor responders - highly suspicious of tumor; (2) intermediate responders - tumor most likely; and (3) good - potential (near) complete responders. The reference standard was histopathology after surgery (or long-term follow-up in the case of a watch-and-wait program). RESULTS: Fourteen patients were complete responders and 48 had residual tumor. The median percentage of patients categorized by the seven readers as 'poor', 'intermediate', and 'good' responders was 21% (range 11-37%), 50% (range 23-58%), and 29% (range 23-42%), respectively. The vast majority of poor responders had histopathologically confirmed residual tumor (73% ypT3-4), with a low rate (0-5%) of 'missed complete responders'. Of the 14 confirmed complete responders, a median percentage of 71% were categorized in the MR-good response group and 29% were categorized in the MR-intermediate response group. CONCLUSIONS: Radiologists of varying experience levels should be able to use MRI to identify the ± 20% subgroup of poor responders who will definitely require surgical resection after CRT. This may facilitate more selective use of endoscopy, particularly in general settings or in centers with limited access to endoscopy.