Development of Clinical Radiomics-Based Models to Predict Survival Outcome in Pancreatic Ductal Adenocarcinoma: A Multicenter Retrospective Study.

PURPOSE: This multicenter retrospective study aims to identify reliable clinical and radiomic features to build machine learning models that predict progression-free survival (PFS) and overall survival (OS) in pancreatic ductal adenocarcinoma (PDAC) patients. METHODS: Between 2010 and 2020 pre-treatment contrast-enhanced CT scans of 287 pathology-confirmed PDAC patients from two sites of the Hopital Universitaire de Bruxelles (HUB) and from 47 hospitals within the HUB network were retrospectively analysed. Demographic, clinical, and survival data were also collected. Gross tumour volume (GTV) and non-tumoral pancreas (RPV) were semi-manually segmented and radiomics features were extracted. Patients from two HUB sites comprised the training dataset, while those from the remaining 47 hospitals of the HUB network constituted the testing dataset. A three-step method was used for feature selection. Based on the GradientBoostingSurvivalAnalysis classifier, different machine learning models were trained and tested to predict OS and PFS. Model performances were assessed using the C-index and Kaplan-Meier curves. SHAP analysis was applied to allow for post hoc interpretability. RESULTS: A total of 107 radiomics features were extracted from each of the GTV and RPV. Fourteen subgroups of features were selected: clinical, GTV, RPV, clinical & GTV, clinical & GTV & RPV, GTV-volume and RPV-volume both for OS and PFS. Subsequently, 14 Gradient Boosting Survival Analysis models were trained and tested. In the testing dataset, the clinical & GTV model demonstrated the highest performance for OS (C-index: 0.72) among all other models, while for PFS, the clinical model exhibited a superior performance (C-index: 0.70). CONCLUSIONS: An integrated approach, combining clinical and radiomics features, excels in predicting OS, whereas clinical features demonstrate strong performance in PFS prediction.

Clinicopathological and molecular predictors of [18F]FDG-PET disease detection in HER2-positive early breast cancer: RESPONSE, a substudy of the randomized PHERGain trial.

BACKGROUND: The PHERGain study (NCT03161353) is assessing early metabolic responses to neoadjuvant treatment with trastuzumab-pertuzumab and chemotherapy de-escalation using a [18Fluorine]fluorodeoxyglucose-positron emission tomography ([18F]FDG-PET) and a pathological complete response-adapted strategy in HER2-positive (HER2+) early breast cancer (EBC). Herein, we present RESPONSE, a PHERGain substudy, where clinicopathological and molecular predictors of [18F]FDG-PET disease detection were evaluated. METHODS: A total of 500 patients with HER2 + EBC screened in the PHERGain trial with a tumor size > 1.5 cm by magnetic resonance imaging (MRI) were included in the RESPONSE substudy. PET[-] criteria entailed the absence of  ≥ 1 breast lesion with maximum standardized uptake value (SUVmax) ≥ 1.5 × SUVmean liver + 2 standard deviation. Among 75 PET[-] patients screened, 21 with SUVmax levels < 2.5 were randomly selected and matched with 21 PET[+] patients with SUVmax levels ≥ 2.5 based on patient characteristics associated with [18F]FDG-PET status. The association between baseline SUVmax and [18F]FDG-PET status ([-] or [+]) with clinicopathological characteristics was assessed. In addition, evaluation of stromal tumor-infiltrating lymphocytes (sTILs) and gene expression analysis using PAM50 and Vantage 3D™ Cancer Metabolism Panel were specifically compared in a matched cohort of excluded and enrolled patients based on the [18F]FDG-PET eligibility criteria. RESULTS: Median SUVmax at baseline was 7.2 (range, 1-39.3). Among all analyzed patients, a higher SUVmax was associated with a higher tumor stage, larger tumor size, lymph node involvement, hormone receptor-negative status, higher HER2 protein expression, increased Ki67 proliferation index, and higher histological grade (p < 0.05). [18F]FDG-PET [-] criteria patients had smaller tumor size (p = 0.014) along with the absence of lymph node involvement and lower histological grade than [18F]FDG-PET [+] patients (p < 0.01). Although no difference in the levels of sTILs was found among 42 matched [18F]FDG-PET [-]/[+] criteria patients (p = 0.73), [18F]FDG-PET [-] criteria patients showed a decreased risk of recurrence (ROR) and a lower proportion of PAM50 HER2-enriched subtype than [18F]FDG-PET[+] patients (p < 0.05). Differences in the expression of genes involved in cancer metabolism were observed between [18F]FDG-PET [-] and [18F]FDG-PET[+] criteria patients. CONCLUSIONS: These results highlight the clinical, biological, and metabolic heterogeneity of HER2+ breast cancer, which may facilitate the selection of HER2+ EBC patients likely to benefit from [18F]FDG-PET imaging as a tool to guide therapy. TRIAL REGISTRATION: Clinicaltrials.gov; NCT03161353; registration date: May 15, 2017.

Optimal [18F]FDG PET/CT Cutoff for Pathologic Complete Response in HER2-Positive Early Breast Cancer Patients Treated with Neoadjuvant Trastuzumab and Pertuzumab in the PHERGain Trial.

The PHERGain trial investigated the potential of metabolic imaging to identify candidates for chemotherapy deescalation in human epidermal growth factor receptor 2 (HER2)-positive, invasive, operable breast cancer with at least 1 breast lesion evaluable by [18F]FDG PET/CT. [18F]FDG PET/CT responders were defined as patients with an SUVmax reduction (ΔSUVmax) of at least 40% in all of their target lesions after 2 cycles of trastuzumab and pertuzumab (HP) (with or without endocrine therapy). In total, 227 of 285 patients (80%) included in the HP arm showed a predefined metabolic response and received a total of 8 cycles of HP (with or without endocrine therapy). Pathologic complete response (pCR), defined as ypT0/isN0, was achieved in 37.9% of the patients. Here, we describe the secondary preplanned analysis of the best cutoff of ΔSUVmax for pCR prediction. Methods: Receiver-operating-characteristic analysis was applied to look for the most appropriate ΔSUVmax cutoff in HER2-positive early breast cancer patients treated exclusively with neoadjuvant HP (with or without endocrine therapy). Results: The ΔSUVmax capability of predicting pCR in terms of the area under the receiver-operating-characteristic curve was 72.1% (95% CI, 65.1-79.2%). The optimal ΔSUVmax cutoff was found to be 77.0%, with a 51.2% sensitivity and a 78.7% specificity. With this cutoff, 74 of 285 patients (26%) would be classified as metabolic responders, increasing the pCR rate from 37.9% (cutoff ≥ 40%) to 59.5% (44/74 patients) (P < 0.01). With this optimized cutoff, 44 of 285 patients (15.4%) would avoid chemotherapy in either the neoadjuvant or the adjuvant setting compared with 86 of 285 patients (30.2%) using the original cutoff (P < 0.001). Conclusion: In the PHERGain trial, an increased SUVmax cutoff (≥77%) after 2 cycles of exclusive HP (with or without endocrine therapy) achieves a pCR in the range of the control arm with chemotherapy plus HP (59.5% vs. 57.7%, respectively), further identifying a subgroup of patients with HER2-addicted tumors. However, the original cutoff (≥40%) maximizes the number of patients who could avoid chemotherapy.

Molecular imaging predicts lack of T-DM1 response in advanced HER2-positive breast cancer (final results of ZEPHIR trial).

Efficacy of the human epidermal growth factor receptor (HER)2-targeting trastuzumab emtansine (T-DM1) in breast cancer (BC) relies on HER2 status determined by immunohistochemistry or fluorescence in-situ hybridization. Heterogeneity in HER2 expression, however, generates interest in "whole-body" assessment of HER2 status using molecular imaging. We evaluated the role of HER2-targeted molecular imaging in detecting HER2-positive BC lesions and patients unlikely to respond to T-DM1. Patients underwent zirconium-89 (89Zr) trastuzumab (HER2) PET/CT and [18F]-2-fluoro-2-deoxy-D-glucose (FDG) PET/CT before T-DM1 initiation. Based on 89Zr-trastuzumab uptake, lesions were visually classified as HER2-positive (visible/high uptake) or HER2-negative (background/close to background activity). According to proportion of FDG-avid tumor load showing 89Zr-trastuzumab uptake (entire/dominant part or minor/no part), patients were classified as HER2-positive and HER2-negative, respectively. Out of 265 measurable lesions, 93 (35%) were HER2-negative, distributed among 42 of the 90 included patients. Of these, 18 (19%) lesions belonging to 11 patients responded anatomically (>30% decrease in axial diameter from baseline) after three T-DM1 cycles, resulting in an 81% negative predictive value (NPV) of the HER2 PET/CT. In combination with early metabolic response assessment on FDG PET/CT performed before the second T-DM1 cycle, NPVs of 91% and 100% were reached in predicting lesion-based and patient-based (RECIST1.1) response, respectively. Therefore, HER2 PET/CT, alone or in combination with early FDG PET/CT, can successfully identify BC lesions and patients with a low probability of clinical benefit from T-DM1.

Tumor Volume on PSMA PET as a Prognostic Biomarker in Prostate Cancer Patients Treated With Cabazitaxel.

PURPOSE: The aim of this study was to evaluate the prognostic value of 68 Ga-labeled prostate-specific membrane antigen (PSMA) PET/CT in metastatic castration-resistant prostate cancer patients receiving second-line chemotherapy with cabazitaxel. METHODS: All patients with metastatic castration-resistant prostate cancer who underwent a PSMA PET/CT within 8 weeks before initiating the cabazitaxel treatment were retrospectively evaluated. The whole-body PSMA total tumor volume (PSMA-TV) was measured for each patient. Other factors such as prostate-specific antigen, hemoglobin, lactate dehydrogenase, and alkaline phosphatase were recorded. A log-rank cutoff finder was used to define the PSMA-TV optimal cutoff. Survival analyses were performed using Cox regression and Kaplan-Meier methods. RESULTS: In total, 32 patients were included, receiving a median of 6 cycles of cabazitaxel (range, 2-10). After a median follow-up of 12 months, 28 patients presented disease progression, and 18 died. Baseline PSMA-TV presented a significant association with progression-free survival (PFS) and overall survival (OS; P = 0.035 and P = 0.002, respectively). Optimal PSMA-TV cutoffs were 515 mL for PFS and 473 mL for OS. Patients with low volume presented longer PFS and OS than those with high volume: median PFS, 21 versus 12 weeks, respectively (hazard ratio, 0.33; P = 0.017); and median OS, 24 versus 8.5 months, respectively (hazard ratio, 0.21; P = 0.002). On the multivariable analyses, PSMA-TV remained an independent predictor of OS ( P = 0.016). CONCLUSION: Our results show that total tumor volume measured on PSMA PET/CT is a prognostic biomarker in patients treated with cabazitaxel. High PSMA-TV before treatment initiation is associated with shorter PFS and OS.

Pre-trial quality assurance of diffusion-weighted MRI for radiomic analysis and the role of harmonisation.

PURPOSE: The aim of this study was to perform a quantitative quality assurance of diffusion-weighted MRI to assess the variability of the mean apparent diffusion coefficient (ADC) and other radiomic features across the scanners involved in the REGINA trial. MATERIALS AND METHODS: The NIST/QIBA diffusion phantom was acquired on six 3 T scanners from five centres with a rectum-specific diffusion protocol. All sequences were repeated in each scan session without moving the phantom from the table. Linear interpolation to two isotropic voxel spacing (0.9 and 4 mm) was performed as well as the ComBat feature harmonisation method between scanners. The absolute accuracy error was evaluated for the mean ADC. Repeatability and reproducibility within-subject coefficients of variation (wCV) were computed for 142 radiomic features. RESULTS: For the mean ADC, accuracy error ranged between 0.1 % and 8.5 %, repeatability was <1 % and reproducibility was <3 % for diffusivity range between 0.4 and 1.1x10-3mm2/s. For the other radiomic features, wCV was below 10 % for 24 % and 15 % features for repeatability with resampling 0.9 mm and 4 mm, respectively, and 13 % and 11 % feature for reproducibility. ComBat method could improve significantly the wCV compared to reproducibility without ComBat (p-value < 0.001) but variation was still high for most of the features. CONCLUSION: Our study provided the first investigation of feature selection for development of robust predictive models in the REGINA trial, demonstrating the added value of such a quality assurance process to select conventional and radiomic features in prospective multicentre trials.

Radiomics software comparison using digital phantom and patient data: IBSI-compliance does not guarantee concordance of feature values.

INTRODUCTION: Radiomics is a promising imaging-based tool which could enhance clinical observation and identify representative features. To avoid different interpretations, the Image Biomarker Standardisation Initiative (IBSI) imposed conditions for harmonisation. This study evaluates IBSI-compliant radiomics applications against a known benchmark and clinical datasets for agreements. MATERIALS AND METHODS: The three radiomics platforms compared were RadiomiX Research Toolbox, LIFEx v7.0.0, and syngo.via Frontier Radiomics v1.2.5 (based on PyRadiomics v2.1). Basic assessment included comparing feature names and their formulas. The IBSI digital phantom was used for evaluation against reference values. For agreement evaluation (including same software but different versions), two clinical datasets were used: 27 contrast-enhanced computed tomography (CECT) of colorectal liver metastases and 39 magnetic resonance imaging (MRI) of breast cancer, including intravoxel incoherent motion (IVIM) and dynamic contrast-enhanced (DCE) MRI. The intraclass correlation coefficient (ICC, lower 95% confidence interval) was used, with 0.9 as the threshold for excellent agreement. RESULTS: The three radiomics applications share 41 (3 shape, 8 intensity, 30 texture) out of 172, 84 and 110 features for RadiomiX, LIFEx and syngo.via, respectively, as well as wavelet filtering. The naming convention is, however, different between them. Syngo.via had excellent agreement with the IBSI benchmark, while LIFEx and RadiomiX showed slightly worse agreement. Excellent reproducibility was achieved for shape features only, while intensity and texture features varied considerably with the imaging type. For intensity, excellent agreement ranged from 46% for the DCE maps to 100% for CECT, while this lowered to 44% and 73% for texture features, respectively. Wavelet features produced the greatest variation between applications, with an excellent agreement for only 3% to 11% features. CONCLUSION: Even with IBSI-compliance, the reproducibility of features between radiomics applications is not guaranteed. To evaluate variation, quality assurance of radiomics applications should be performed and repeated when updating to a new version or adding a new modality.

Combined Metabolically Active Tumor Volume and Early Metabolic Response Improve Outcome Prediction in Metastatic Colorectal Cancer.

Stratification of metastatic colorectal cancer (mCRC) patients is mostly based on clinical and biologic characteristics. This study aimed to validate the prognostic value of 18F-FDG PET/CT-based biomarkers such as baseline whole-body metabolically active tumor volume (WB-MATV) and early metabolic response (mR) in mCRC. Methods: The development cohort included chemorefractory mCRC patients enrolled in 2 prospective Belgian multicenter trials evaluating last-line treatments (multikinase inhibitors). The validation cohort included mCRC patients from an Italian center treated with chemotherapy and bevacizumab as first-line. Baseline WB-MATV was defined as the sum of metabolically active volumes of all target lesions identified on the baseline 18F-FDG PET/CT. Early mR assessment was performed following usual response criteria (response threshold of 30% [PERCIST-30%], response threshold of 15% [PERCIST-15%], European Organization for Research and Treatment of Cancer) and the so-called CONSIST method, which defines response as a decrease of SULmax ≥ 15% for all target lesions. Baseline WB-MATV and early mR assessment were investigated along with usual clinical factors and correlated with overall survival (OS) and progression-free survival (PFS). Results: Clinical factors, baseline WB-MATV, and early mR were evaluable in 192 of 239 and 94 of 125 patients of the development and validation cohorts, respectively. Except for PERCIST-30%, all response methods were equivalent in terms of outcome prediction, and CONSIST was found to be the most accurate. Baseline WB-MATV and early mR using the CONSIST method were independent prognostic parameters after adjustment for clinical factors in the development and validation sets for both OS (hazard ratio [HR] WB-MATV: 1.87 [95% CI, 1.17-2.97], P = 0.005, and HR early mR: 1.79 [95% CI, 1.08-2.95], P = 0.02 for the validation set) and PFS (HR WB-MATV: 1.94 [95% CI, 1.27-2.97], P = 0.002, and HR early mR: 1.69 [95% CI, 1.04-2.73], P = 0.03 for the validation set). Conclusion: Baseline WB-MATV and early mR are strong independent prognostic biomarkers for OS and PFS in mCRC, regardless of treatment received. Therefore, combining these biomarkers improves risk stratification for OS and PFS in mCRC.

FDG positron emission tomography imaging and ctDNA detection as an early dynamic biomarker of everolimus efficacy in advanced luminal breast cancer.

Biomarkers to identify patients without benefit from adding everolimus to endocrine treatment in metastatic breast cancer (MBC) are needed. We report the results of the Pearl trial conducted in five Belgian centers assessing 18F-FDG-PET/CT non-response (n = 45) and ctDNA detection (n = 46) after 14 days of exemestane-everolimus (EXE-EVE) to identify MBC patients who will not benefit. The metabolic non-response rate was 66.6%. Median PFS in non-responding patients (using as cut-off 25% for SUVmax decrease) was 3.1 months compared to 6.0 months in those showing response (HR: 0.77, 95% CI: 0.40-1.50, p = 0.44). The difference was significant when using a "post-hoc" cut-off of 15% (PFS 2.2 months vs 6.4 months). ctDNA detection at D14 was associated with PFS: 2.1 months vs 5.0 months (HR-2.5, 95% CI: 1.3-5.0, p = 0.012). Detection of ctDNA and/or the absence of 18F-FDG-PET/CT response after 14 days of EXE-EVE identifies patients with a low probability of benefiting from treatment. Independent validation is needed.

Sex and Regorafenib Toxicity in Refractory Colorectal Cancer: Safety Analysis of the RegARd-C Trial.

BACKGROUND: Regorafenib is a standard treatment for refractory metastatic colorectal cancer (mCRC). In view of the toxicity burden, significant research efforts have been made to increase the therapeutic ratio of this multikinase inhibitor. Predictive factors for treatment-related adverse events (TRAEs), however, are still lacking. MATERIALS AND METHODS: We assessed the association between a number of baseline clinical, laboratory and imaging parameters and the occurrence of TRAEs in 136 patients who had received regorafenib (160 mg/day, 3-weeks-on/1-week-off) in a prospective phase II clinical trial. RESULTS: Grade ≥ 2 TRAEs during the first cycle of treatment (84% vs. 60%, P = .002) and grade ≥ 3 TRAEs throughout the whole treatment (71% vs. 53%, P = .035) occurred more frequently in females, with sex being the only independent predictive factor of early and any-time toxicity (OR 3.4; 95% CI: 1.2-11.1, P = .02 and OR 2.1; 95% CI: 1.0-4.4, P = .045, respectively). Fatigue, anorexia, hypertension, and rash were reported significantly more frequently by females than males (P < .04). Females were also more likely to suffer early (19% vs. 5%, P = .014) and any-time serious AEs (28% vs. 9%, P = .005), and to require early dose modifications (55% vs. 37%, P = .055). CONCLUSION: This is the first study showing an association between sex and TRAEs during regorafenib treatment for mCRC. If confirmed in larger, independent series, these results could pave the way for the implementation of personalized regorafenib dosing strategies with the potential to optimize oncological outcomes while reducing toxicity and preserving quality of life.

Combined quality and dose-volume histograms for assessing the predictive value of 99mTc-MAA SPECT/CT simulation for personalizing radioembolization treatment in liver metastatic colorectal cancer.

BACKGROUND: The relationship between the mean absorbed dose delivered to the tumour and the outcome in liver metastases from colorectal cancer patients treated with radioembolization has already been presented in several studies. The optimization of the personalized therapeutic activity to be administered is still an open challenge. In this context, how well the 99mTc-MAA SPECT/CT predicts the absorbed dose delivered by radioembolization is essential. This work aimed to analyse the differences between predictive 99mTc-MAA-SPECT/CT and post-treatment 90Y-microsphere PET/CT dosimetry at different levels. Dose heterogeneity was compared voxel-to-voxel using the quality-volume histograms, subsequently used to demonstrate how it could be used to identify potential clinical parameters that are responsible for quantitative discrepancies between predictive and post-treatment dosimetry. RESULTS: We analysed 130 lesions delineated in twenty-six patients. Dose-volume histograms were computed from predictive and post-treatment dosimetry for all volumes: individual lesion, whole tumoural liver (TL) and non-tumoural liver (NTL). For all dose-volume histograms, the following indices were extracted: D90, D70, D50, Dmean and D20. The results showed mostly no statistical differences between predictive and post-treatment dosimetries across all volumes and for all indices. Notably, the analysis showed no difference in terms of Dmean, confirming the results from previous studies. Quality factors representing the spread of the quality-volume histogram (QVH) curve around 0 (ideal QF = 0) were determined for lesions, TL and NTL. QVHs were classified into good (QF < 0.18), acceptable (0.18 ≤ QF < 0.3) and poor (QF ≥ 0.3) correspondence. For lesions and TL, dose- and quality-volume histograms are mostly concordant: 69% of lesions had a QF within good/acceptable categories (40% good) and 65% of TL had a QF within good/acceptable categories (23% good). For NTL, the results showed mixed results with 48% QF within the poor concordance category. Finally, it was demonstrated how QVH analysis could be used to define the parameters that predict the significant differences between predictive and post-treatment dose distributions. CONCLUSION: It was shown that the use of the QVH is feasible in assessing the predictive value of 99mTc-MAA SPECT/CT dosimetry and in estimating the absorbed dose delivered to liver metastases from colorectal cancer via 90Y-microspheres. QVH analyses could be used in combination with DVH to enhance the predictive value of 99mTc-MAA SPECT/CT dosimetry and to assist personalized activity prescription.

Prognostic Value of the Pace of Tumor Progression as Assessed by Serial 18F-FDG PET/CT Scan and Liquid Biopsy in Refractory Colorectal Cancer: The CORIOLAN Trial.

INTRODUCTION: Decision making in refractory colorectal cancer (rCRC) is challenging, with limited data available to predict patient outcome. We conducted a study to assess the pace of cancer progression as a potential prognostic and decision tool. METHODS: CORIOLAN was a prospective, single-center, single-arm trial recruiting refractory CRC patients with an ECOG performance status of ≤1 and an estimated life expectancy of ≥12 weeks. 18 fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) scan and blood sample collection were carried out at baseline and after 2 weeks with no cancer treatment given between these timepoints. The primary objective was to evaluate the association between pace of cancer progression as defined by changes of the whole-body metabolically active tumor volume (WB-MATV) and overall survival (OS). Exploratory objectives included evaluation of the prognostic value of circulating cell-free DNA (cfDNA), circulating tumor cells (CTCs) and carcinoembryonic antigen (CEA). RESULTS: 47 eligible patients who had received a median number of 5 (range 2-8) prior treatments were enrolled. At the time of analysis, 45 deaths had occurred, with 26% of patients dying within 12 weeks. The median OS was 6.3 months (range 0.4-14.3). The median relative delta between WB-MATV at baseline and 2 weeks was +21%. Changes of WB-MATV, however, failed to predict OS (hazard ratio (HR) 1.3, p = 0.383). Similarly, no association was observed between changes of any of the circulating biomarkers investigated and prognosis. By contrast, high WB-MATV (4.2 versus 9.4 months; HR 3.1, p = 0.003), high CEA (4.4 versus 7.0 months; HR 1.9, p = 0.053), high cfDNA (4.7 versus 7.0 months; HR 2.2, p = 0.015) and high CTC count (3.3 versus 7.5 months; HR 6.5, p < 0.001) at baseline were associated with worse OS. CONCLUSIONS: In this study, approximately 1 out of 4 refractory CRC patients who were judged to have a life expectancy >12 weeks actually died within 12 weeks. Baseline assessment of WB-MATV, cfDNA, CTCs and CEA, but not early change evaluation of the same, may help to refine patient prognostication and guide management decisions.

Fat density is a novel prognostic marker in patients with esophageal cancer.

BACKGROUND & AIMS: While long-term obesity is a well-known risk factor for esophageal adenocarcinoma (ADC), recent weight loss represents a significant concern in esophageal cancer (EC), in relation with dysphagia and disease aggressiveness. These phenomenons may diversely impact the adipose tissue density, suggested in other cancer settings as an important prognostic biomarker. The analysis of body mass composition (BMC) parameters, including adipose tissue attenuation is studied here in a population of EC operated with curative intent. METHODS: BMC was retrospectively evaluated on Computed-Tomography (CT)-scan images from fluorodeoxyglucose (FDG)-positron-emitting (PET)/CT scans performed as a diagnostic procedure in a cohort of 145 EC patients operated with curative intent The mean subcutaneous (SFD) and visceral fat (VFD) density along with the index (area/height2) (SF index (SFI), VF index (VFI)) were assessed on two adjacent slides at the third lumbar vertebra level by two independent investigators. Overall survival (OS) was calculated from the date of the baseline FDG-PET/CT scan. RESULTS: Inter-observer correlations are excellent for all BMC parameters (r = 0.94-0.99). As expected, weight loss is associated with worse outcome. We show that low SFD (HR 0.5 (95% CI: 0.3-0.7), p < 0.001) and low VFD (HR 0.6 (95% CI: 0.4-0.9), p = 0.04) at diagnosis are associated with better OS. In contrast, body mass index (BMI) fails to show any relevance in predicting survival. CONCLUSIONS: Adipose tissue density is an important prognostic factor in EC.

Personalised radioembolization improves outcomes in refractory intra-hepatic cholangiocarcinoma: a multicenter study.

PURPOSE: Reported outcomes of patients with intra-hepatic cholangiocarcinoma (IH-CCA) treated with radioembolization are highly variable, which indicates differences in included patients' characteristics and/or procedure-related variables. This study aimed to identify patient- and treatment-related variables predictive for radioembolization outcome. METHODS: This retrospective multicenter study enrolled 58 patients with unresectable and chemorefractory IH-CCA treated with resin 90Y-microspheres. Clinicopathologic data were collected from patient records. Metabolic parameters of liver tumor(s) and presence of lymph node metastasis were measured on baseline 18F-FDG-PET/CT. 99mTc-MAA tumor to liver uptake ratio (TLRMAA) was computed for each lesion on the SPECT-CT. Activity prescription using body-surface-area (BSA) or more personalized partition-model was recorded. The study endpoint was overall survival (OS) starting from date of radioembolization. Statistical analysis was performed by the log-rank test and multivariate Cox's proportional hazards model. RESULTS: Median OS (mOS) post-radioembolization of the entire cohort was 10.3 months. Variables associated with significant differences in terms of OS were serum albumin (hazard ratio (HR) = 2.78, 95%CI:1.29-5.98, p = 0.002), total bilirubin (HR = 2.17, 95%CI:1.14-4.12, p = 0.009), aspartate aminotransferase (HR = 2.96, 95%CI:1.50-5.84, p < 0.001), alanine aminotransferase (HR = 2.02, 95%CI:1.05-3.90, p = 0.01) and γ-GT (HR = 2.61, 95%CI:1.31-5.22, p < 0.001). The presence of lymph node metastasis as well as a TLRMAA < 1.9 were associated with shorter mOS: HR = 2.35, 95%CI:1.08-5.11, p = 0.008 and HR = 2.92, 95%CI:1.01-8.44, p = 0.009, respectively. Finally, mOS was significantly shorter in patients treated according to the BSA method compared to the partition-model: mOS of 5.5 vs 14.9 months (HR = 2.52, 95%CI:1.23-5.16, p < 0.001). Multivariate analysis indicated that the only variable that increased outcome prediction above the clinical variables was the activity prescription method with HR of 2.26 (95%CI:1.09-4.70, p = 0.03). The average mean radiation dose to tumors was significantly higher with the partition-model (86Gy) versus BSA (38Gy). CONCLUSION: Radioembolization efficacy in patients with unresectable recurrent and/or chemorefractory IH-CCA strongly depends on the tumor radiation dose. Personalized activity prescription should be performed.

Combining 18F-FDG PET/CT-Based Metabolically Active Tumor Volume and Circulating Cell-Free DNA Significantly Improves Outcome Prediction in Chemorefractory Metastatic Colorectal Cancer.

Baseline whole-body metabolically active tumor volume (WB-MATV) measured by 18F-FDG PET/CT and circulating cell-free DNA (cfDNA) have been separately validated as predictors of overall and progression-free survival (OS/PFS) in chemorefractory metastatic colorectal cancer (mCRC) patients. This study assessed the correlation between WB-MATV and cfDNA, evaluating the added prognostic value of these in combination, along with clinical parameters. Methods: Of 141 mCRC patients included in a prospective multicenter trial, 132 were evaluable for OS/PFS. cfDNA was extracted from 3 mL of plasma and quantified using a fluorometer. All target lesions were delineated on 18F-FDG PET/CT, and their metabolic volumes were summed to obtain the WB-MATV. Results: Baseline WB-MATV and cfDNA were strongly correlated (r = 0.70; P < 0.001) but showed discordance in 23 of 132 (17%) patients. A multivariate analysis identified 3 independent negative predictors of PFS (high cfDNA, short time since diagnosis, and body mass index < 30) and 5 of OS (high cfDNA, high WB-MATV, body mass index < 30, poor performance status, and short time since diagnosis). Combining WB-MATV and cfDNA increased the overall prognostic value and allowed identification of a subgroup of patients with low cfDNA and high WB-MATV who were associated with intermediate survival (median OS of 8.1 for low-cfDNA/high-MATV patients vs. 12.7 mo for low-cfDNA/low-MATV patients; hazard ratio, 2.04; P = 0.02). Conclusion: This study confirms the added prognostic value of combined circulating cfDNA and PET-based WB-MATV in chemorefractory mCRC patients. The combination of these two biomarkers should provide a firm basis for risk stratification, both in clinical practice and in research trials.

Prognostic value of adipose tissue and muscle mass in advanced colorectal cancer: a post hoc analysis of two non-randomized phase II trials.

BACKGROUND: The prognostic value of body composition in cancer patients has been widely studied during the last decade. The main finding of these studies is that sarcopenia, or skeletal muscle depletion, assessed by CT imaging correlates with a reduced overall survival (OS). By contrast, the prognostic value of fat mass remains ill-defined. This study aims to analyze the influence of body composition including both muscle mass and adipose tissue on OS in a homogeneous population of advanced colorectal cancer (CRC) patients. METHODS: Among 235 patients with chemorefractory advanced CRC included in the SoMore and RegARd-C trials, body composition was assessed in 217 patients on baseline CT images. The relationship between body composition (sarcopenia, muscle density, subcutaneous and visceral fat index and density), body mass index (BMI) and OS were evaluated. RESULTS: Patients with a higher BMI had a better OS (≥30 versus < 30, HR: 0.50; 0.33-0.76). Those with low muscle index and muscle density had an increased mortality (HR: 2.06; 1.45-2.93 and HR: 1.54; 1.09-2.18, respectively). Likewise, low subcutaneous and visceral fat index were associated with an increased risk of dying (HR: 1.63; 1.23-2.17 and 1.48; 1.09-2.02 respectively), as were a high subcutaneous and visceral adipose tissue density (HR: 1.93; 1.44-2.57 and 2.40; 1.79-3.20 respectively). In multivariate analysis, a high visceral fat density was the main predictor of poor survival. CONCLUSIONS: Our results confirm the protective role of obesity in CRC patients at an advanced stage, as well as the negative prognostic impact of muscle depletion on survival. More importantly, our data show for the first time that visceral adipose tissue density is an important prognostic factor in metastatic CRC. TRIAL REGISTRATION: NCT01290926 , 07/02/2011 and NCT01929616 , 28/08/2013.

Validation of Metabolically Active Tumor Volume and Total Lesion Glycolysis as 18F-FDG PET/CT­derived Prognostic Biomarkers in Chemorefractory Metastatic Colorectal Cancer

This study aimed to validate the prognostic value of baseline whole-body metabolic active tumor volume (WB-MATV) and total lesion glycolysis (WB-TLG) measured with [18F]fluorodeoxyglucose positron emission tomography-computed tomography (18F-FDG PET/CT) in a large cohort of chemorefractory metastatic colorectal cancer (mCRC) patients treated with multikinase inhibitors (MKI). The secondary objective of this study was to compare WB-MATV and WB-TLG respective prognostic values to commonly used clinical prognostic factors. Methods: Out of 238 patients pooled from two successive prospective multicenter trials investigating MKI in chemorefractory mCRC, 224 were considered suitable for analysis. The patients were retrospectively randomly assigned to a development set (n = 155 patients) and a validation set (n = 69 patients). WB-MATV and WB-TLG optimal cutoffs for prediction of overall survival (OS) were determined by Contal and O'Quigley's method. Univariate analyses were performed to assess the prognostic values of WB-MATV and WB-TLG. Multivariate analyses were performed for WB-MATV and WB-TLG along with clinical factors to identify the independent prognostic factors of OS. The prognostic weight for each parameter was obtained from the Cox's model. Results: WB-MATV and WB-TLG optimal cutoffs for OS prediction were 100 cm3 and 500 g, respectively. Univariate analyses showed that WB-MATV and WB-TLG parameters were strongly related to outcome in both the development and validation sets. In the validation set, the median OS was 5.2 months vs 12.8 months for high vs low WB-MATV (HR: 3.12, P < 0.001), and 4.7 months vs 13.9 months for high vs low WB-TLG (HR: 3.67, P < 0.001). The multivariate analyses identified that both high WB-MATV and WB-TLG were independent negative prognostic parameters for OS, with the highest prognostic weight among the well-known clinical prognostic factors (HR: 2.46 and 2.23, respectively, P < 0.001). Conclusion: Baseline WB-MATV and WB-TLG parameters were validated as strong prognosticators of outcome in a large cohort of chemorefractory mCRC patients treated with MKI. These parameters were identified as independent prognostic imaging biomarkers with the highest prognostic values among the commonly used clinical factors. These biomarkers should therefore be used to support the optimal therapeutic strategy.

A dosimetry procedure for organs-at-risk in 177Lu peptide receptor radionuclide therapy of patients with neuroendocrine tumours.

PURPOSE: Peptide receptor radionuclide therapy with 177Lu-DOTATATE has become a standard treatment modality in neuroendocrine tumours (NETs). No consensus has yet been reached however regarding the absorbed dose threshold for lesion response, the absorbed dose limit to organs-at-risk, and the optimal fractionation and activity to be administered. This is partly due to a lack of uniform and comparable dosimetry protocols. The present article details the development of an organ-at-risk dosimetry procedure, which could be implemented and used routinely in a clinical context. METHODS: Forty-seven patients with NETs underwent 177Lu-DOTATATE therapy. Three SPECT/CT images were acquired at 4, 24 and 144-192 h post-injection. Three blood samples were obtained together with the SPECT/CT acquisitions and 2 additional samples were obtained around 30 min and 1 h post-injection. A bi-exponential fit was used to compute the source organ time-integrated activity coefficients. Coefficients were introduced into OLINDA/EXM software to compute organ-at-risk absorbed doses. Median values for all patients were computed for absorbed dose coefficient D/A0 and for late effective half-life T1/2eff for kidneys, spleen and red marrow. RESULTS: Dosimetry resulted in a median[interquartile range] of 0.78[0.35], 1.07[0.58] and 0.028[0.010] Gy/GBq for D/A0 and of 55[9], 71[9] and 52[18] h for T1/2eff for kidneys, spleen and red marrow respectively. CONCLUSIONS: A dosimetry procedure for organs-at-risk in 177Lu-DOTATATE therapy based on serial SPECT/CT images and blood samples can be implemented routinely in a clinical context with limited patient burden. The results obtained were in accordance with those of other centres.