Tumor Burden Assessed by the Maximum Standardized Uptake Value and Greatest Diameter on FDG-PET Predicts Prognosis in Untreated Diffuse Large B-cell Lymphoma

Comparing lesion and feature selections to predict progression in newly diagnosed DLBCL patients with FDG PET/CT radiomics features

PURPOSE

Biomarkers that can accurately predict outcome in DLBCL patients are urgently needed. Radiomics features extracted from baseline [¹⁸F]-FDG PET/CT scans have shown promising results. This study aims to investigate which lesion- and feature-selection approaches/methods resulted in the best prediction of progression after 2 years.

METHODS

A total of 296 patients were included. 485 radiomics features (n = 5 conventional PET, n = 22 morphology, n = 50 intensity, n = 408 texture) were extracted for all individual lesions and at patient level, where all lesions were aggregated into one VOI. 18 features quantifying dissemination were extracted at patient level. Several lesion selection approaches were tested (largest or hottest lesion, patient level [all with/without dissemination], maximum or median of all lesions) and compared to the predictive value of our previously published model. Several data reduction methods were applied (principal component analysis, recursive feature elimination (RFE), factor analysis, and univariate selection). The predictive value of all models was tested using a fivefold cross-validation approach with 50 repeats with and without oversampling, yielding the mean cross-validated AUC (CV-AUC). Additionally, the relative importance of individual radiomics features was determined.

RESULTS

Models with conventional PET and dissemination features showed the highest predictive value (CV-AUC: 0.72-0.75). Dissemination features had the highest relative importance in these models. No lesion selection approach showed significantly higher predictive value compared to our previous model. Oversampling combined with RFE resulted in highest CV-AUCs.

CONCLUSION

Regardless of the applied lesion selection or feature selection approach and feature reduction methods, patient level conventional PET features and dissemination features have the highest predictive value. Trial registration number and date: EudraCT: 2006-005174-42, 01-08-2008.

Baseline 18F-FDG PET textural features as predictors of response to chemotherapy in diffuse large B-cell lymphoma

PURPOSE

We sought to investigate the performance of radiomics analysis on baseline 18F-FDG PET/CT for predicting response to first-line chemotherapy in diffuse large B-cell lymphoma (DLBCL).

MATERIAL AND METHODS

Forty-five patients who received first-line rituximab, cyclophosphamide, doxorubicin, vincristine and prednisone (R-CHOP) chemotherapy for DLBCL were included in the study. Radiomics features and standard uptake value (SUV)-based measurements were extracted from baseline PET images for a total of 147 lesions. The selection of the most relevant features was made using the recursive feature elimination algorithm. A machine-learning model was trained using the logistic regression classifier with cross-validation to predict treatment response. The independent predictors of incomplete response were evaluated with multivariable regression analysis.

RESULTS

A total of 14 textural features were selected by the recursive elimination algorithm, achieving a feature-to-lesion ratio of 1:10. The accuracy and area under the receiver operating characteristic curve of the model for predicting incomplete response were 0.87 and 0.81, respectively. Multivariable analysis revealed that SUVmax and gray level co-occurrence matrix dissimilarity were independent predictors of lesions with incomplete response to first-line R-CHOP chemotherapy.

CONCLUSION

Increased textural heterogeneity in baseline PET images was found to be associated with incomplete response in DLBCL.

Evaluating the Predictive Ability of Initial Staging F-18 FDG PET/CT for the Prognosis of Non-Hodgkin Malignant Lymphoma Patients Who Underwent Stem Cell Transplantation

Objectives

This study aimed to determine the value of clinical prognostic factors and semiquantitative metabolic parameters from initial staging fluorine-18 fluorodeoxyglucose positron emission tomography/computed tomography (F-18 FDG PET/CT) in non-Hodgkin lymphoma (NHL) patients treated with stem cell transplantation (SCT).

Methods

A total of 39 malignant lymphoma patients who underwent initial staging F-18 FDG PET/CT were enrolled in this study. SUVmax, MTV_wb, and TLG_wb were measured during the initial staging PET/CT. Receiver operating characteristic curve (ROC) analysis was adopted to dichotomize continuous variables. Log-rank test and Cox proportional hazard regression analysis were used to evaluate disease-free survival (DFS) rate.

Results

Among the 39 patients with malignant lymphoma, 17 (43.6%) had a relapse. For several clinical factors such as age, ECOG performance score, AMC/ALC score, stages, and revised International Prognostic Index score, differences between the two dichotomized groups were statistically insignificant. In univariate analysis, DFS estimates were 71.0 ± 7.8 months and 18.0 ± 5.9 months in high-SUVmax and low-SUVmax group, respectively (P < 0.01). For MTV_wb, DFS estimates were 46.6 ± 12.4 months and 69.1 ± 8.5 months in high-MTV_wb and low-MTV_wb group, respectively (P = 0.12). For TLG_wb, DFS estimates were 65.3 ± 7.5 months and 13.7 ± 8.6 months in high-TLG_wb and low-TLG_wb group, respectively (P = 0.02). In Cox proportional hazard regression analysis, only MTV_wb showed statistical significance (HR 3.01, 95% CI 1.04-8.74, P = 0.04).

Conclusion

In NHL patients treated with SCT, the MTV_wb of initial staging F-18 FDG PET/CT was an independent prognostic factor.

Circulating Tumor DNA Reflects Tumor Metabolism Rather Than Tumor Burden in Chemotherapy-Naive Patients with Advanced Non-Small Cell Lung Cancer: 18F-FDG PET/CT Study

We aimed to evaluate the relationships between circulating tumor cells (CTCs) or plasma cell-free DNA (cfDNA) on one side and a comprehensive range of ¹⁸F-FDG PET/CT-derived parameters on the other side in chemotherapy-naive patients with advanced non-small cell lung cancer (NSCLC). Methods: From a group of 79 patients included in a trial evaluating the role of pretreatment circulating tumor markers as predictors of prognosis in chemotherapy-naive patients with advanced NSCLC, we recruited all those who underwent ¹⁸F-FDG PET/CT for clinical reasons at our institution before inclusion in the trial (and thus just before chemotherapy). For each patient, a peripheral blood sample was collected at baseline for the evaluation of CTCs and cfDNA. CTCs were isolated by size using a filtration-based device and then morphologically identified and enumerated; cfDNA was isolated from plasma and quantified by a quantitative polymerase chain reaction using human telomerase reverse transcriptase. The following ¹⁸F-FDG PET/CT-derived parameters were computed: maximum diameter of the primary lesion (T), of the largest lymph node (N), and of the largest metastatic lesion (M); SUV_max; SUV_mean; size-incorporated SUV_max; metabolic tumor volume; and total lesion glycolysis. All parameters were independently measured for T, N, and M. The associations among CTCs, cfDNA, and ¹⁸F-FDG PET/CT-derived parameters were evaluated by multivariate-analysis. Patients were divided into 2 groups according to the presence of either limited metastatic involvement (M1a or M1b due to extrathoracic lymph nodes only) or disseminated metastatic disease. The presence or absence of metabolically active bone lesions was also recorded for each patient, and patient subgroups were compared. Results: Thirty-seven patients recruited in the trial matched our PET-based criteria (24 men; age, 64.5 ± 8.1 y). SUV_max for the largest metastatic lesion was the only variable independently associated with baseline cfDNA levels (P = 0.016). Higher levels of cfDNA were detected in the subgroup of patients with metabolically active bone lesions (P = 0.02), but no difference was highlighted when patients with more limited metastatic disease were compared with patients with disseminated metastatic disease. Conclusion: The correlation of cfDNA levels with tumor metabolism, but not with metabolic tumor volume at regional or distant levels, suggests that cfDNA may better reflect tumor biologic behavior or aggressiveness rather than tumor burden in metastatic NSCLC.

Baseline and ongoing PET-derived factors predict detrimental effect or potential utility of 18F-FDG PET/CT (FDG-PET/CT) performed for surveillance in asymptomatic lymphoma patients in first remission

PURPOSE

To identify both clinical and FDG PET/CT-derived factors predicting the occurrence of relapse, or conversely, the likelihood of false positive findings in surveillance FDG-PET/CT studies (PETsv).

METHODS

The study included 149 asymptomatic patients with Hodgkin's lymphoma (HL) (n = 55) or diffuse large B cell lymphoma (DLBCL) (n = 94) in first remission. PETSv studies were performed 12, 18, 24 and 36 months thereafter. Logistic regression analysis was performed to identify clinical and imaging-derived predictors of either PET-detected relapse or false-positive (FP) results. Tested clinical variables were: 1) age, 2) HL vs. DLBCL, 3) stage of disease, 4) bulky disease, 5) previous radiotherapy. PET/CT-derived variables were: 1) maximum standardized uptake value at baseline, 2) size-incorporated maximum standardized uptake value (SIMaxSUV) at baseline, 3) positive interim PET(PET-2), 4) presence of hot spots likely to be unrelated to the disease in final PET, 5) residual non-FDG avid mass.

RESULTS

Accuracy was 88 % for PETsv1, 95 % for PETsv2, 95 % for PETsv3 and 91 % for PETsv4. However, PPV was relatively low in all PETsv. Best predictors of relapse were result of interim PET, HL versus NHL type, SIMaxSUV, age ≥ 60. Best predictors of FP were previous radiotherapy and hot spots unrelated to the disease in final PET.

CONCLUSIONS

The present study confirms the need of restricting the use of surveillance PET/CT to patients at high risk of relapse. Information derived from PET/CT performed at baseline (metabolic disease burden), in the course (PET2) and at the end of therapy (unrelated hot spots) can help to select high-risk patients and also to identify patients more likely to present equivocal findings at PETsv.

¹⁸F-fluorodeoxyglucose PET/CT features and correlations with histopathologic characteristics in sclerosing epithelioid fibrosarcoma

Sclerosing epithelioid fibrosarcoma (SEF) is a clinicopathologically distinct variant of fibrosarcoma that is capable of recurrence and metastasis. Awareness of imaging features and histopathologic characteristics will be helpful for differential diagnosis from other common tumors. Here, we report a case of SEF metastasizing to the pancreas as a solitary mass mimicking primary pancreatic cancer, and summarize the reported cases with FDG PET/CT from the literature (n=4). PET/CT showed abnormal FDG accumulation (n=2), mild FDG uptake (n=1), or photopenic (n=1). The FDG PET/CT features are closely related to histopathologic characteristics regarding its differentiation and aggressiveness.

Regional PET/CT after water gastric inflation for evaluating loco-regional disease of gastric cancer

OBJECTIVE

We aimed to improve diagnostic accuracy of (18)F-fluoro-2-deoxyglucose (FDG) PET/CT for gastric cancer with water gastric inflation.

MATERIALS AND METHODS

44 gastric cancer patients (M:F=30:14, age ± std=62.1 ± 14.5 y) were enrolled before surgery. Fifty minutes after injection of FDG (0.14 mCi/kg body weight), whole body PET/CT was performed first and then regional PET/CT over gastric area was obtained 80 min post FDG injection after water gastric inflation. Diagnostic accuracies for loco-regional lesions were compared between whole body and regional PET/CT.

RESULTS

48 primary tumors (23 EGC and 25 AGC) and 348 LN stations (61 metastatic and 287 benign) in 44 patients were investigated. Primary tumor sensitivity of whole body PET/CT (50%=24/48) was significantly improved by regional PET/CT (75%=36/48, p<0.005). Sensitivity of whole body PET/CT (24.6%=15/61) for LN metastasis was also significantly improved by regional PET/CT (36.1%=22/61, p<0.01), whereas specificity of whole body PET/CT (99.3%=285/287) was not compromised by regional PET/CT (98.3%=282/287, p>0.05). Higher primary tumor FDG uptake in regional PET/CT indicated shorter progress-free survival (p=0.0003).

CONCLUSION

Diagnostic accuracy of whole body PET/CT for loco-regional disease of gastric cancer could be significantly improved by regional PET/CT after water gastric inflation and prognosis could be effectively predicted by primary tumor FDG uptake in regional PET/CT.

PET imaging in a longitudinal non-Hodgkin's lymphoma study: association with tumor volume

BACKGROUND

Computed tomography (CT) is generally used in the evaluation of the treatment response of non-Hodgkin's lymphoma (NHL) patients. Instead of morphological images, positron emission tomography (PET) shows metabolic information that is connected to tumor activity, cell proliferation rate, and, thus, prognosis.

PURPOSE

To determine the prognostic value of PET for tumor volume reduction measured by CT and magnetic resonance imaging (MRI) along with clinical characteristics in NHL patients.

MATERIAL AND METHODS

We imaged 21 B-cell type NHL patients using whole-body 18F-FDG-PET at the onset and the completion of treatment and at six-month follow-up. The maximum standardized uptake value (SUV(max)) was calculated. Morphological tumor volume calculations were assessed using both MRI and CT. Additionally, patients underwent thorough clinical examination including several laboratory tests.

RESULTS

A high SUV(max) was able to predict significant tumor volume reduction at the beginning of treatment, but the relation to pure tumor volume was poor.

CONCLUSION

The SUV(max) values derived from FDG-PET seemed to correlate with volume changes but not with their absolute values or laboratory tests. Unlike MRI and CT, FDG-PET showed the disappearance of active tumors after treatment.

Prognostic Value of Primary Tumor Uptake on F-18 FDG PET/CT in Patients with Invasive Ductal Breast Cancer

PURPOSE

To determine the prognostic implications of pretreatment F-18 FDG PET/CT in patients with invasive ductal breast cancer (IDC), we evaluated the relationship between FDG uptake of the primary tumor and known prognostic parameters of breast cancer. Prognostic significance of tumoral FDG uptake for the prediction of progression-free survival (PFS) was also assessed.

MATERIALS AND METHODS

Fifty-five female patients with IDC who underwent pretreatment F-18 FDG PET/CT were enrolled. The maximum standardized uptake value of the primary tumor (pSUVmax) was compared with clinicopathological parameters including tumor size, grade, estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor2 (HER2), axillary lymph node (LN) metastasis, and stage. The prognostic value of pSUVmax for PFS was assessed using the Kaplan-Meier method.

RESULTS

A high pSUVmax was significantly related to a higher stage of tumor size (P < 0.05), grade (P < 0.001), and stage (P < 0.001). pSUVmax was significantly higher in ER-negative tumors (P < 0.001), PR-negative tumors (P < 0.001), and positive LN metastasis (P < 0.01), but not different according to HER2 status. pSUVmax was significantly higher in patients with progression compared to patients who were disease-free (10.6 ± 5.1 vs. 4.7 ± 3.5, P < 0.001). A receiver-operating characteristic curve demonstrated a pSUVmax of 6.6 to be the optimal cutoff for predicting PFS (sensitivity; 86.7%, specificity; 82.5%). The patients with a high pSUVmax (more than 6.6) had significantly shorter PFS compared to patients with a low pSUVmax (P < 0.0001).

CONCLUSIONS

pSUVmax on pretreatment F-18 FDG PET/CT could be used as a good surrogate marker for the prediction of progression in patients with IDC.