Automated Lugano Metabolic Response Assessment in 18F-Fluorodeoxyglucose-Avid Non-Hodgkin Lymphoma With Deep Learning on 18F-Fluorodeoxyglucose-Positron Emission Tomography.

PURPOSE: Artificial intelligence can reduce the time used by physicians on radiological assessments. For 18F-fluorodeoxyglucose-avid lymphomas, obtaining complete metabolic response (CMR) by end of treatment is prognostic. METHODS: Here, we present a deep learning-based algorithm for fully automated treatment response assessments according to the Lugano 2014 classification. The proposed four-stage method, trained on a multicountry clinical trial (ClinicalTrials.gov identifier: NCT01287741) and tested in three independent multicenter and multicountry test sets on different non-Hodgkin lymphoma subtypes and different lines of treatment (ClinicalTrials.gov identifiers: NCT02257567, NCT02500407, 20% holdout in ClinicalTrials.gov identifier: NCT01287741), outputs the detected lesions at baseline and follow-up to enable focused radiologist review. RESULTS: The method's response assessment achieved high agreement with the adjudicated radiologic responses (eg, agreement for overall response assessment of 93%, 87%, and 85% in ClinicalTrials.gov identifiers: NCT01287741, NCT02500407, and NCT02257567, respectively) similar to inter-radiologist agreement and was strongly prognostic of outcomes with a trend toward higher accuracy for death risk than adjudicated radiologic responses (hazard ratio for end of treatment by-model CMR of 0.123, 0.054, and 0.205 in ClinicalTrials.gov identifiers: NCT01287741, NCT02500407, and NCT02257567, compared with, respectively, 0.226, 0.292, and 0.272 for CMR by the adjudicated responses). Furthermore, a radiologist review of the algorithm's assessments was conducted. The radiologist median review time was 1.38 minutes/assessment, and no statistically significant differences were observed in the level of agreement of the radiologist with the model's response compared with the level of agreement of the radiologist with the adjudicated responses. CONCLUSION: These results suggest that the proposed method can be incorporated into radiologic response assessment workflows in cancer imaging for significant time savings and with performance similar to trained medical experts.

The impact of CHOP versus bendamustine on bone mineral density in patients with indolent lymphoma enrolled in the GALLIUM study.

Standard CHOP treatment includes a high cumulative dose of prednisone, and studies have shown increased fracture risk following CHOP. It is unclear whether reductions in bone mineral density (BMD) are caused by glucocorticoids or by the combination with chemotherapy. Our objective was to determine the effect of obinutuzumab (G)/rituximab (R)-bendamustine versus G/R-CHOP on BMD in follicular lymphoma patients. Patients in this GALLIUM post hoc study were ≥60 years old and in complete remission at induction treatment completion (ITC), following treatment with G or R in combination with bendamustine or CHOP. To assess BMD, Hounsfield units (HU) were measured in lumbar vertebra L1 on annual computed tomography. Furthermore, vertebral compression fractures were recorded. Of 173 patients included, 59 (34%) received CHOP and 114 (66%) received bendamustine. At baseline, there was no difference in HU between groups. The mean HU decrease from baseline to ITC was 27.8 after CHOP and 17.3 after bendamustine, corresponding to a difference of 10.4 (95% CI: 3.2-17.6). Vertebral fractures were recorded in 5/59 patients receiving CHOP and in 2/114 receiving bendamustine. CHOP was associated with a significant greater decrease in BMD and more frequent fractures. These results suggest that prophylaxis against BMD loss should be considered.

PET-CR as a potential surrogate endpoint in untreated DLBCL: meta-analysis and implications for clinical trial design.

This study's focus is the association of end-of-therapy (EOT) PET results with progression-free (PFS) and overall survival (OS) in patients with diffuse large B-cell lymphoma receiving first-line chemoimmunotherapy. We develop a Bayesian hierarchical model for predicting PFS and OS from EOT PET-complete response (PET-CR) using a literature-based meta-analysis of 20 treatment arms and a substudy of 4 treatment arms in 3 clinical trials for which we have patient-level data. The PET-CR rate in our substudy was 72%. The modeled estimates for hazard ratio (PET-CR/non-PET-CR) were 0.13 for PFS (95% CI 0.10, 0.16) and 0.10 for OS (CI 0.07, 0.12). Hazard ratios varied little by patient subtype and were confirmed by the overall meta-analysis. We link these findings to designing future clinical trials and show how our model can be used in adapting the sample size of a trial to accumulating results regarding treatment benefits on PET-CR and a survival endpoint.

Follicular Lymphoma Treated with First-Line Immunochemotherapy: A Review of PET/CT in Patients Who Did Not Achieve a Complete Metabolic Response in the GALLIUM Study.

Complete metabolic response (CMR) on PET/CT was the sole independent predictor of overall survival in the PET substudy of the phase III GALLIUM trial (NCT01332968) in first-line treatment of high-tumor-burden follicular lymphoma. The aim of this analysis was to further investigate the outcome of patients not achieving CMR. Methods: Two international experts rereviewed PET/CT scans from patients failing to achieve CMR assessed by the Independent Review Committee masked otherwise to committee results. Metabolic response category and Deauville score were assigned. Progression-free survival (PFS) was investigator-assessed with contrast-enhanced CT. Kaplan-Meier methodology was used to estimate landmark PFS and time to next treatment from end of induction by Deauville score. Patients who experienced CT-based progressive disease at the end of induction were excluded. Results: Fifty-four patients were reviewed. Six had CMR, 37 had a partial metabolic response, 2 had no metabolic response, and 9 had progressive metabolic disease. Patients were reassigned to CMR because 18F-FDG uptake was considered inflammatory (n = 2), was considered incidental neoplasia (n = 2), or was visually close to liver uptake but quantitatively lower (n = 2). There was a trend for shorter PFS and time to next treatment for patients with a Deauville score of 5 than a score of 4. High-grade mesenteric uptake at the end of induction was common, occurring in 20 patients with non-CMR, 14 of whom achieved CMR at all other sites. Only 3 of 14 (21%) patients with mesenteric uptake as the only site of disease experienced progression or death within 24 mo, whereas 4 of 6 patients (67%) with mesenteric and additional sites of 18F-FDG-avid disease experienced progression or death within 24 mo. All patients with early progression had measurable disease on contrast-enhanced CT at 18F-FDG-avid sites at the end of induction. Conclusion: After induction immunochemotherapy, CMR was assigned after reassessment in some patients, in whom increased 18F-FDG uptake was considered due to inflammation or incidental neoplasia rather than to lymphoma. Quantitative assessment to confirm the visual impression of residual uptake in lesions is suggested. Isolated mesenteric 18F-FDG uptake is likely a common false-positive finding at the end of induction and does not warrant changes in clinical management or disease surveillance unless there is measurable disease on contrast-enhanced CT or clinical suspicion of active disease.

Calcium Channel Blockers Impair the Antitumor Activity of Anti-CD20 Monoclonal Antibodies by Blocking EGR-1 Induction.

Direct cell death induction, in addition to immune-effector cell-mediated mechanisms, is one of the key mechanisms of action of anti-CD20 antibodies, and yet the signaling pathways implicated remain poorly investigated. Here we show that the transcription factor EGR-1 is rapidly induced by anti-CD20 antibodies and is a key mediator for CD20-induced cell death. EGR-1 induction results from an increased calcium influx induced by anti-CD20 antibodies. We show that both rituximab and obinutuzumab induce calcium influx, albeit through different mechanisms, and this influx is crucial for cell death induction. Inhibition of the calcium flux with calcium channel blockers (CCB) abolished EGR-1 induction and impaired the efficacy of anti-CD20 antibodies in preclinical in vitro and in vivo models. Finally, we investigated the impact of CCBs in patients treated with anti-CD20 antibodies included in the clinical trials GOYA and REMARC, and found that patients simultaneously receiving CCBs and anti-CD20 therapy have a shorter progression-free survival and overall survival. These results reveal EGR-1 as a key mediator of the direct cytotoxic activity of anti-CD20 antibodies and provide a rationale to evaluate EGR-1 expression as a new biomarker to predict response to anti-CD20 treatment. In addition, our findings show that calcium influx is required for anti-CD20-mediated tumor cell death and suggest that simultaneous administration of calcium channel blocking agents could be deleterious in patients receiving anti-CD20-based immunotherapy.

Role of obinutuzumab exposure on clinical outcome of follicular lymphoma treated with first-line immunochemotherapy.

AIMS: Obinutuzumab (G) is a humanized type II, Fc-glycoengineered anti-CD20 monoclonal antibody used in various indications, including patients with previously untreated front-line follicular lymphoma. We investigated sources of variability in G exposure and association of progression-free survival (PFS) with average concentration over induction (CmeanIND ) in front-line follicular lymphoma patients treated with G plus chemotherapy (bendamustine, CHOP, or CVP) in the GALLIUM trial. METHODS: Individual exposures (CmeanIND ) were obtained from a previously established population pharmacokinetic model updated with GALLIUM data. Multivariate Cox proportional hazard models and univariate Kaplan-Meier plots investigated relationships of PFS with exposure and other potential prognostic factors. RESULTS: Overall, G exposure was lower in high body-weight patients and in males, and slightly lower in patients with high baseline tumour burden. Analysis of clinical outcomes showed that variability in G exposure did not impact PFS in G-bendamustine-treated patients; PFS was inferior in males and patients with FCGR2a/2b T232 T low-affinity receptor variant, and superior in patients with FCGR2a/2b I232T variant. In G-CHOP/CVP arms, PFS improved with increasing CmeanIND (hazard ratio = 1.74 and 0.394 at 5th and 95th percentile compared to median CmeanIND ) and was inferior in patients with high baseline tumour size and B symptoms. CONCLUSIONS: It remains unclear whether for G-CHOP/CVP patients lower G exposure is a consequence of adverse disease biology and/or resistance to chemotherapy backbone (higher clearance in nonresponder patients, as demonstrated for rituximab) rather than being the cause of poorer clinical outcome. A study with >1 dose level of G could help resolve this uncertainty.