Purpose
Lymphoma lesion detection and segmentation on whole-body FDG-PET/CT are a challenging task because of the diversity of involved nodes, organs or physiological uptakes. We sought to ...investigate the performances of a three-dimensional (3D) convolutional neural network (CNN) to automatically segment total metabolic tumour volume (TMTV) in large datasets of patients with diffuse large B cell lymphoma (DLBCL).
Methods
The dataset contained pre-therapy FDG-PET/CT from 733 DLBCL patients of 2 prospective LYmphoma Study Association (LYSA
)
trials. The first cohort (
n
= 639) was used for training using a 5-fold cross validation scheme. The second cohort (
n
= 94) was used for external validation of TMTV predictions. Ground truth masks were manually obtained after a 41% SUVmax adaptive thresholding of lymphoma lesions. A 3D U-net architecture with 2 input channels for PET and CT was trained on patches randomly sampled within PET/CTs with a summed cross entropy and Dice similarity coefficient (DSC) loss. Segmentation performance was assessed by the DSC and Jaccard coefficients. Finally, TMTV predictions were validated on the second independent cohort.
Results
Mean DSC and Jaccard coefficients (± standard deviation) in the validations set were 0.73 ± 0.20 and 0.68 ± 0.21, respectively. An underestimation of mean TMTV by − 12 mL (2.8%) ± 263 was found in the validation sets of the first cohort (
P
= 0.27). In the second cohort, an underestimation of mean TMTV by − 116 mL (20.8%) ± 425 was statistically significant (
P
= 0.01).
Conclusion
Our CNN is a promising tool for automatic detection and segmentation of lymphoma lesions, despite slight underestimation of TMTV. The fully automatic and open-source features of this CNN will allow to increase both dissemination in routine practice and reproducibility of TMTV assessment in lymphoma patients.
18F-fluorodeoxyglucose positron emission tomography combined with computed tomography (FDG-PET/CT) is an essential part of the management of patients with lymphoma at staging and response evaluation. ...Efforts to standardize PET acquisition and reporting, including the 5-point Deauville scale, have enabled PET to become a surrogate for treatment success or failure in common lymphoma subtypes. This review summarizes the key clinical-trial evidence that supports PET-directed personalized approaches in lymphoma but also points out the potential place of innovative PET/CT metrics or new radiopharmaceuticals in the future.
To investigate the respective influence of software tool and total metabolic tumor volume (TMTV0) calculation method on prognostic stratification of baseline 2-deoxy-2-18Ffluoro-D-glucose positron ...emission tomography (18FFDG-PET) in newly diagnosed Hodgkin lymphoma (HL).
59 patients with newly diagnosed HL were retrospectively included. 18FFDG-PET was performed before any treatment. Four sets of TMTV0 were calculated with Beth Israel (BI) software: based on an absolute threshold selecting voxel with standardized uptake value (SUV) >2.5 (TMTV02.5), applying a per-lesion threshold of 41% of the SUV max (TMTV041) and using a per-patient adapted threshold based on SUV max of the liver (>125% and >140% of SUV max of the liver background; TMTV0125 and TMTV0140). TMTV041 was also determined with commercial software for comparison of software tools. ROC curves were used to determine the optimal threshold for each TMTV0 to predict treatment failure.
Median follow-up was 39 months. There was an excellent correlation between TMTV041 determined with BI and with the commercial software (r = 0.96, p<0.0001). The median TMTV0 value for TMTV041, TMTV02.5, TMTV0125 and TMTV0140 were respectively 160 (used as reference), 210 (28;154 p = 0.005), 183 (-4;114 p = 0.06) and 143 ml (-58;64 p = 0.9). The respective optimal TMTV0 threshold and area under curve (AUC) for prediction of progression free survival (PFS) were respectively: 313 ml and 0.70, 432 ml and 0.68, 450 ml and 0.68, 330 ml and 0.68. There was no significant difference between ROC curves. High TMTV0 value was predictive of poor PFS in all methodologies: 4-years PFS was 83% vs 42% (p = 0.006) for TMTV02.5, 83% vs 41% (p = 0.003) for TMTV041, 85% vs 40% (p<0.001) for TMTV0125 and 83% vs 42% (p = 0.004) for TMTV0140.
In newly diagnosed HL, baseline metabolic tumor volume values were significantly influenced by the choice of the method used for determination of volume. However, no significant differences were found in term of prognosis.
Celotno besedilo
Dostopno za:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
ALK-positive histiocytosis is a recently described entity with few reported cases in literature. Herein, we report an unusual case of ALK-positive histiocytosis showing an Erdheim-Chester disease ...(ECD)-like presentation, occurring in a 37-year-old woman with a 2-year history of chronic lymphocytic leukaemia (CLL). Our CLL patient relapsed 6 months after the end of fludarabine, cyclophosphamide and rituximab frontline therapy and complained of lower limb pains. A bone marrow biopsy was performed and showed concomitant CLL/small lymphocytic lymphoma and ALK-positive histiocytosis with an identical immunoglobulin heavy-chain gene rearrangement in both neoplasms, suggesting clonal relationship. After 4 years under ibrutinib therapy, our patient remains free of both diseases. This report extends the spectrum of composite hematolymphoid neoplasms and shows that
ALK
rearrangement should be considered in all histiocytosis subtypes. Moreover, both tumours eradication under ibrutinib suggests that BTK inhibitors may also be effective in histiocytic neoplasms.
Follicular lymphoma (FL) is the most common indolent lymphoma. Despite the clear benefit of CD20-based therapy, a subset of FL patients still progress to aggressive lymphoma. Thus, identifying early ...biomarkers that incorporate PET metrics could be helpful to identify patients with a high risk of treatment failure with Rituximab. We retrospectively included a total of 132 untreated FL patients separated into training and validation cohorts. Optimal threshold of baseline SUVmax was first determined in the training cohort (n=48) to predict progression-free survival (PFS). The PET results were investigated along with the tumor and immune microenvironment, which were determined by immunochemistry and transcriptome studies involving gene set enrichment analyses and immune cell deconvolution, together with the tumor mutation profile. We report that baseline SUVmax >14.5 was associated with poorer PFS than baseline SUVmax ≤14.5 (HR=0.28; p=0.00046). Neither immune T-cell infiltration nor immune checkpoint expression were associated with baseline PET metrics. By contrast, FL samples with Ki-67 staining ≥10% showed enrichment of cell cycle/DNA genes (p=0.013) and significantly higher SUVmax values (p=0.007). Despite similar oncogenic pathway alterations in both SUVmax groups of FL samples, 4 out of 5 cases harboring the infrequent FOXO1 transcription factor mutation were seen in FL patients with SUVmax >14.5. Thus, high baseline SUVmax reflects FL tumor proliferation and, together with Ki-67 proliferative index, can be used to identify patients at risk of early relapse with R-chemotherapy.
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.
Stage IIB Hodgkin lymphoma (HL) patients, with a mediastinum-to-thorax (M/T) ratio of ≥0.33 or extranodal localization have a poor prognosis and are treated either as limited or advanced stage. We ...compared these two approaches in patients included in two randomized phase III trials enrolling previously untreated early (H10) or advanced stage HL (AHL2011). We included HL patients with Ann-Arbor stage IIB with M/T ≥0.33 or extranodal involvement enrolled in the H10 or AHL2011 trials with available positron emission tomography at baseline (PET0) and after two cycles of chemotherapy (PET2). Baseline total metabolic tumor volume (TMTV) was calculated using the 41% SUVmax method. PET2 response assessment used the Deauville score. One hundred and fourty-eight patients were eligible, including 83 enrolled in the AHL2011 trial and 65 in the H10 trial. The median TMTV value was 155.5 mL (range, 8.3-782.9 mL), 165.6 mL in AHL2011 and 147 mL in H10. PET2 positivity rates were 16.9% (n=14) and 9.2% (n=6) in AHL2011 and H10 patients, respectively. With a median follow-up of 4.1 years (95% confidence interval CI: 3.9-4.4), overall 4-year PFS was 88.0%, 87.0% in AHL2011 and 89.2% in H10. In univariate and mutivariate analyses, baseline TMTV and PET2 response influenced significantly progression-free survival (hazard ratio HR=4.94, HR=3.49 respectively). Notably, among the 16 patients who relapsed, 13 (81%) had a baseline TMTV baseline ≥155 mL. Upfront ABVD plus radiation therapy or upfront escBEACOPP without radiotherapy provide similar patient's outcome in high-risk stage IIB HL. TMTV is useful to stratify these patients at baseline.
Evaluate response and predict prognosis of patients with newly diagnosed metastatic breast cancer treated with first line systemic therapy using European Organization for Research and Treatment of ...Cancer (EORTC) criteria and PET Response Criteria in solid Tumours (PERCIST).
From December 2006 to August 2013, 57 women with newly diagnosed metastatic breast cancer were retrospectively evaluated. FDG-PET/CT was performed within one month before treatment and repeated after at least 3 cycles of treatment. Metabolic response evaluation was evaluated by two readers according to both EORTC criteria and PERCIST, classifying the patients into 4 response groups: complete metabolic response (CMR), partial metabolic response (PMR), stable metabolic disease (SMD), and progressive metabolic disease (PMD).
With EORTC criteria, 22 patients had CMR, 17 PMR, 6 SMD and 12 PMD. With PERCIST, 20 patients had CMR, 15 PMR, 10 SMD and 12 PMD. There was agreement between EORTC and PERCIST in 84% of the patients. By log-rank analysis, metabolic response evaluated with both EORTC criteria and PERCIST was able to predict overall survival (p = 0.028 and 0.002 respectively). CMR patient group had longer median OS than patients in the combined PMR+SMD+PMD group (60 vs 26 months both with EORTC and PERCIST; p = 0.009 and 0.006 respectively). By multivariate analysis, CMR either with EORTC or PERCIST remained an independent predictor of survival.
Metabolic response evaluation with EORTC criteria and PERCIST gave similar prognostic stratification for metastatic breast cancer treated with a first line of systemic therapy.
Celotno besedilo
Dostopno za:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
Whereas numerous studies on several cancers describe the link between social conditions and disease severity, little is known about the social and demographic characteristics of Hodgkin lymphoma (HL) ...patients. At diagnosis, 10-15% of the patients in the advanced stages have a well-known poor outcome owing to their chemoresistance, but the determinants of the more advanced stages remain elusive. The objective of the present study was to decipher the potential impact of social disparities on the disease features at diagnosis and analyze how the sociodemographic patient features could impact the HL outcome of patients with advanced-stage HL enrolled in the AHL2011 trial.
This ancillary study was conducted on a cohort of patients from French centers that had recruited more than five patients in the phase III AHL2011 study (NCT0135874). Patients had to be alive at the time of the ancillary study and had to have given their consent to answer the questionnaire. Pre-treatment data (age, gender, stage, B symptoms, IPS), the treatment received, the responses to PET-CT, and the presence of serious adverse events (serious adverse events-SAEs) were all extracted from the AHL2011 trial database. Sociodemographic data-marital status, living area, level of education, socio-professional category, and professional situation-were extracted from the questionnaires. The population density at the point of diagnosis was determined based on ZIP Code, and the distance from the reference medical center was then calculated by the road network. Baseline PET acquisition was performed before any treatment. PET images at baseline were centrally reviewed. The total metabolic tumor volume (TMTV) at the baseline was calculated using a 41% SUVmax cutoff for each lesion. Progression-free survival was defined as the time from randomization to the first progression, relapse, or death from any cause or the last follow-up. The data cutoff for the analyses presented here was 31 October 2017. The progression-free survival was analyzed on an intention-to-treat basis.
Among the 823 patients enrolled in the AHL2011 study, the questionnaire was sent to 394 patients, of whom 232 (58.9%) responded. At the time of HL diagnosis, 61.9% (N = 143) of patients declared that they were not socially isolated, 38.1% (N = 88) that they were single, 163 (71.2%) had a professional activity, and 66 (28.8%) were inactive owing to unemployment, retirement, or sick leave. Of the patients, 31.1% (N = 71) lived in a rural region, compared to 68.9% (N = 157) that lived in an urban region. The residence ZIP Code at the time of HL diagnosis was available for 163 (70%). Sociodemographic characteristics did not influence the presence of usual prognostic factors (ECOG, B symptoms, bulky mass, IPS) except for professional activity, which was associated with more frequent low IPS (0-2) (79 (48.5%) active versus 20 (30.3%) inactive patients;
= 0.012). Likewise, no correlation was observed between TMTV and sociodemographic characteristics. However, the TMTV quartile distribution was different according to the living area, with the two upper quartiles being enriched with patients living in a rural area (
= 0.008). Moreover, a negative correlation between the average number of the living area's inhabitants and TMTV (R Pearson = -0.29,
= 0.0004) was observed.
This study focused on sociodemographic parameters in advanced-stage HL patients and shows that professional activity is associated with more favorable disease features (low IPS), while patients living in rural or low-populated areas are more likely to have an unfavorable HL presentation with a high tumor burden (high TMTV). These data suggest that some patient sociodemographic characteristics might impact either access to medical care or environmental exposure, leading to a higher frequency of unfavorable presentations. Further prospective sociodemographic studies are necessary to confirm these preliminary results.
The total metabolic tumor volume (TMTV) is a new prognostic factor in lymphomas that could benefit from automation with deep learning convolutional neural networks (CNN). Manual TMTV segmentations of ...1218 baseline 18FDG-PET/CT have been used for training. A 3D V-NET model has been trained to generate segmentations with soft dice loss. Ground truth segmentation has been generated using a combination of different thresholds (TMTVprob), applied to the manual region of interest (Otsu, relative 41% and SUV 2.5 and 4 cutoffs). In total, 407 and 405 PET/CT were used for test and validation datasets, respectively. The training was completed in 93 h. In comparison with the TMTVprob, mean dice reached 0.84 in the training set, 0.84 in the validation set and 0.76 in the test set. The median dice scores for each TMTV methodology were 0.77, 0.70 and 0.90 for 41%, 2.5 and 4 cutoff, respectively. Differences in the median TMTV between manual and predicted TMTV were 32, 147 and 5 mL. Spearman's correlations between manual and predicted TMTV were 0.92, 0.95 and 0.98. This generic deep learning model to compute TMTV in lymphomas can drastically reduce computation time of TMTV.
PDF
