Lars Kurch

PET/MR in children. Initial clinical experience in paediatric oncology using an integrated PET/MR scanner

Use of PET/MR in children has not previously been reported, to the best of our knowledge. Children with systemic malignancies may benefit from the reduced radiation exposure offered by PET/MR. We report our initial experience with PET/MR hybrid imaging and our current established sequence protocol after 21 PET/MR studies in 15 children with multifocal malignant diseases. The effective dose of a PET/MR scan was only about 20% that of the equivalent PET/CT examination. Simultaneous acquisition of PET and MR data combines the advantages of the two previously separate modalities. Furthermore, the technique also enables whole-body diffusion-weighted imaging (DWI) and statements to be made about the biological cellularity and nuclear/cytoplasmic ratio of tumours. Combined PET/MR saves time and resources. One disadvantage of PET/MR is that in order to have an effect, a significantly longer examination time is needed than with PET/CT. In our initial experience, PET/MR has turned out to be an unexpectedly stable and reliable hybrid imaging modality, which generates a complementary diagnostic study of great additional value.

qPET – a quantitative extension of the Deauville scale to assess response in interim FDG-PET scans in lymphoma

BackgroundInterim FDG-PET is used for treatment tailoring in lymphoma. Deauville response criteria consist of five ordinal categories based on visual comparison of residual tumor uptake to physiological reference uptakes. However, PET-response is a continuum and visual assessments can be distorted by optical illusions.ObjectivesWith a novel semi-automatic quantification tool we eliminate optical illusions and extend the Deauville score to a continuous scale.Patients and methodsSUVpeak of residual tumors and average uptake of the liver is measured with standardized volumes of interest. The qPET value is the quotient of these measurements. Deauville scores and qPET-values were determined in 898 pediatric Hodgkin’s lymphoma patients after two OEPA chemotherapy cycles.ResultsDeauville categories translate to thresholds on the qPET scale: Categories 3, 4, 5 correspond to qPET values of 0.95, 1.3 and 2.0, respectively. The distribution of qPET values is unimodal with a peak representing metabolically normal responses and a tail of clearly abnormal outliers. In our patients, the peak is at qPET = 0.95 coinciding with the border between Deauville 2 and 3. qPET cut values of 1.3 or 2 (determined by fitting mixture models) select abnormal metabolic responses with high sensitivity, respectively, specificity.ConclusionsqPET methodology provides semi-automatic quantification for interim FDG-PET response in lymphoma extending ordinal Deauville scoring to a continuous scale. Deauville categories correspond to certain qPET cut values. Thresholds between normal and abnormal response can be derived from the qPET-distribution without need for follow-up data. In our patients, qPET < 1.3 excludes abnormal response with high sensitivity.

FDG PET/CT in children and adolescents with lymphoma

The aim of this review is to give an overview of FDG PET/CT applications in children and adolescents with lymphoma. Today, FDG PET is used for tailoring treatment intensity in children with Hodgkin lymphoma within the framework of international treatment optimisation protocols. In contrast, the role of this method in children with Non-Hodgkin lymphoma is not well defined. This paper overviews clinical appearance and metabolic behaviour of the most frequent lymphoma subtypes in childhood. The main focus of the review is to summarise knowledge about the role of FDG PET/CT for initial staging and early response assessment.

Inter-Reader Reliability of Early FDG-PET/CT Response Assessment Using the Deauville Scale after 2 Cycles of Intensive Chemotherapy (OEPA) in Hodgkin's Lymphoma.

Purpose The five point Deauville (D) scale is widely used to assess interim PET metabolic response to chemotherapy in Hodgkin lymphoma (HL) patients. An International Validation Study reported good concordance among reviewers in ABVD treated advanced stage HL patients for the binary discrimination between score D1,2,3 and score D4,5. Inter-reader reliability of the whole scale is not well characterised. Methods Five international expert readers scored 100 interim PET/CT scans from paediatric HL patients. Scans were acquired in 51 European hospitals after two courses of OEPA chemotherapy (according to the EuroNet-PHL-C1 study). Images were interpreted in direct comparison with staging PET/CTs. Results The probability that two random readers concord on the five point D score of a random case is only 42% (global kappa = 0.24). Aggregating to a three point scale D1,2 vs. D3 vs. D4,5 improves concordance to 60% (kappa = 0.34). Concordance if one of two readers assigns a given score is 70% for score D1,2 only 36% for score D3 and 64% for D4,5. Concordance for the binary decisions D1,2 vs. D3,4,5 is 67% and 86% for D1,2,3 vs D4,5 (kappa = 0.36 resp. 0.56). If one reader assigns D1,2,3 concordance probability is 92%, but only 64% if D4,5 is called. Discrepancies occur mainly in mediastinum, neck and skeleton. Conclusion Inter-reader reliability of the five point D-scale is poor in this interobserver analysis of paediatric patients who underwent OEPA. Inter-reader variability is maximal in cases assigned to D2 or D3. The binary distinction D1,2,3 versus D4,5 is the most reliable criterion for clinical decision making.

Staging Evaluation and Response Criteria Harmonization (SEARCH) for Childhood, Adolescent and Young Adult Hodgkin Lymphoma (CAYAHL): Methodology statement

International harmonization of staging evaluation and response criteria is needed for childhood, adolescence, and young adulthood Hodgkin lymphoma. Two Hodgkin lymphoma protocols from cooperative trials in Europe and North America were compared for areas in need of harmonization, and an evidence‐based approach is currently underway to harmonize staging and response evaluations with a goal to enhance comparisons, expedite identification of effective therapies, and aid in the approval process for new agents by regulatory agencies.

Positron Emission Tomography–Guided Therapy of Aggressive Non-Hodgkin Lymphomas (PETAL): A Multicenter, Randomized Phase III Trial

Purpose Interim positron emission tomography (PET) using the tracer, [18F]fluorodeoxyglucose, may predict outcomes in patients with aggressive non-Hodgkin lymphomas. We assessed whether PET can guide therapy in patients who are treated with cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP). Patients and Methods Newly diagnosed patients received two cycles of CHOP-plus rituximab (R-CHOP) in CD20-positive lymphomas-followed by a PET scan that was evaluated using the ΔSUVmax method. PET-positive patients were randomly assigned to receive six additional cycles of R-CHOP or six blocks of an intensive Burkitts lymphoma protocol. PET-negative patients with CD20-positive lymphomas were randomly assigned or allocated to receive four additional cycles of R-CHOP or the same treatment with two additional doses rituximab. The primary end point was event-free survival time as assessed by log-rank test. Results Interim PET was positive in 108 (12.5%) and negative in 754 (87.5%) of 862 patients treated, with statistically significant differences in event-free survival and overall survival. Among PET-positive patients, 52 were randomly assigned to R-CHOP and 56 to the Burkitt protocol, with 2-year event-free survival rates of 42.0% (95% CI, 28.2% to 55.2%) and 31.6% (95% CI, 19.3% to 44.6%), respectively (hazard ratio, 1.501 [95% CI, 0.896 to 2.514]; P = .1229). The Burkitt protocol produced significantly more toxicity. Of 754 PET-negative patients, 255 underwent random assignment (129 to R-CHOP and 126 to R-CHOP with additional rituximab). Event-free survival rates were 76.4% (95% CI, 68.0% to 82.8%) and 73.5% (95% CI, 64.8% to 80.4%), respectively (hazard ratio, 1.048 [95% CI, 0.684 to 1.606]; P = .8305). Outcome prediction by PET was independent of the International Prognostic Index. Results in diffuse large B-cell lymphoma were similar to those in the total group. Conclusion Interim PET predicted survival in patients with aggressive lymphomas treated with R-CHOP. PET-based treatment intensification did not improve outcome.

Sources of variability in FDG PET imaging and the qPET value: reply to Laffon and Marthan

Dear Sir, Sources of variability in FDG PET imaging can be broadly divided into two parts: acquisition variability in obtaining the stored image data set and inter-rater variability interpreting a given stored image data set. Standardised analysis software can reduce inter-rater variability compared to the current standard of unsupported visual reading. The qPETmethod [1] standardises the comparison of the highest residual uptake with the physiological uptake in reference organs [mediastinal blood pool (MBP), liver]. Algorithms detecting candidate tumour residua and thus addressing variability in identification of the residual tumour with the highest uptake (including recognition and exclusion of possible artefacts) have been developed. However, such analysis software only deals with a given stored image data set and can thus only reduce the inter-rater variability. In their letter, Laffon and Marthan [2] correctly point out that the acquisition variability is not yet well investigated and certainly not negligible. An ideal study to assess this variance component would require patients to undergo FDG PET imaging in different institutions on successive days. Unfortunately, multiple exposure to radioactivity for methodological research purposes is ethically unjustifiable. Laffon and Marthan [2] nevertheless try to derive a guesstimate of the order of magnitude of the acquisition variability component in the qPET value. The qPET value is calculated as the ratio of the SUVpeak of the residual tumour divided by the average liver uptake. Laffon and Marthan [2] combine estimates from two publications for the acquisition variability for the numerator and for the denominator in qPET and then use an approximation formula for the variability of the quotient of two measurements. The first paper [3] describes the variability of multiple dynamic acquisitions in ten successive 2.5-min periods in untreated lung cancer lesions. This type of investigation is an interesting surrogate for the unfeasible ideal study mentioned above. It captures variability due to the timing of the acquisition. They report a variation coefficient (=average per cent error) of about 5–6 % for the average of the five hottest pixels. This may be roughly comparable to the SUVpeak used as numerator in qPET, which is the average of the hottest and the further three hottest voxels adjacent to the maximum. Laffon and Marthan [2] then interpret data of Boktor et al. [4] in order to assess the acquisition variability in measuring the qPET denominator, i.e. the physiological liver uptake. The paper describes the variability of SUV in the liver in a heterogeneous group of cancer patients who had two FDG PET images within 1 year. Most patients had therapy in between, and chemotherapy increased liver uptake in a relevant subgroup. A variation coefficient of about 16 % can be reconstructed from the information provided in this paper. This probably overestimates the acquisition variability when measuring the physiological liver uptake in a homogeneous population, at one time, and using a 30-ml volume of interest (VOI) instead of a single slice. Laffon and Marthan [2] then add the variation coefficients using a well-known approximation formula for the variation coefficient of the quotient of two measurements. This leads to an overestimation, since the approximation assumes that the measurement errors are independent. This is probably wrong. D. Hasenclever (*) Institute for Medical Informatics, Statistics and Epidemiology (IMISE), University of Leipzig, Leipzig, Germany e-mail: dirk.hasenclever@imise.uni-leipzig.de

Current Role of FDG-PET in Pediatric Hodgkin’s Lymphoma

Hodgkins lymphoma is one of the most curable pediatric cancers with long-term survival rates exceeding 90% following intensive treatment. Collaborative group studies worldwide aim on reduction or elimination of radiotherapy to avoid potentially life-limiting late effects especially second cancers and cardiovascular diseases. Large prospective trials have integrated early response FDG-PET scans to identify adequate responders to chemotherapy in whom radiotherapy may safely be omitted. The criteria for interpretation of early response PET have changed during the past years and will be further refined based on trial results. FDG-PET is also systematically used to assess initial disease involvement of pediatric Hodgkins lymphoma and could replace bone marrow biopsy. This article summarizes the role of FDG-PET in staging and response assessment focusing on large pediatric trials, the criteria for PET interpretation and pitfalls.

18F-FDG PET Response of Skeletal (Bone Marrow and Bone) Involvement After Induction Chemotherapy in Pediatric Hodgkin Lymphoma: Are Specific Response Criteria Required?

To determine whether the current 18F-FDG PET response criterion for skeletal involvement in Hodgkin lymphoma (HL) is suitable, we performed a systematic evaluation of the different types of skeletal involvement and their response on PET after 2 cycles of chemotherapy (PET-2). A secondary objective was to observe the influence of the initial uptake intensity (measured as qPET) and initial metabolic tumor volume (MTV) of skeletal lesions on the PET-2 response. Methods: The initial PET scans of 1,068 pediatric HL patients from the EuroNet-PHL-C1 trial were evaluated for skeletal involvement by central review. Three types of skeletal lesions were distinguished: PET-only lesions (those detected on PET only), bone marrow (BM) lesions (as confirmed by MRI or BM biopsy), and bone lesions. qPET and MTV were calculated for each skeletal lesion. All PET-2 scans were assessed for residual tumor activity. The rates of complete metabolic response for skeletal and nodal involvement on PET-2 were compared. Results: Of the 1,068 patients, 139 (13%) showed skeletal involvement (44 PET-only, 32 BM, and 63 bone). Of the 139 patients with skeletal involvement, 101 (73%) became PET-2–negative in the skeleton and 94 (68%) became PET-2–negative in the lymph nodes. The highest number of PET-2–negative scans in the skeleton was 42 (95%) in the 44 PET-only patients, followed by 22 skeletal lesions (69%) in the 32 BM patients and 37 (59%) in the 63 bone patients. Lesions that became PET-2–negative showed a lower initial median qPET (2.74) and MTV (2 cm3) than lesions that remained PET-2–positive (3.84 and 7 cm3, respectively). Conclusion: In this study with pediatric HL patients, the complete response rate for skeletal involvement on PET-2 was similar to that for nodal involvement. Bone flare seemed to be irrelevant. Overall, the current skeletal PET response criterion—comparison with the local skeletal background—is well suited. The initial qPET and MTV of skeletal lesions were predictive of the PET-2 result. Higher values for both parameters were associated with a worse PET-2 response.

Reply to: Laffon and Marthan “FDG PET for therapy monitoring in Hodgkin’s and non-Hodgkin’s lymphomas: qPET versus rPET”

Laffon and Marthan [1] recently commented on a review by Barrington and Kluge [2]. The main part of their letter was an appraisal of qPET and rPET methods for response assessment in lymphomas. The qPET method was introduced by our group as a quantitative approach to the evaluation of PET response, evaluating the method in a population of 898 paediatric patients with Hodgkin lymphoma [3]. In 2016, Annunziata et al. described the use of rPET for the same indication, evaluating rPET in 68 adult Hodgkin lymphoma patients [4]. In their letter Laffon and Marthan wrongly claim that the qPET method is not quantitative. The qPET value is defined as the quotient of SUVpeak in the tumour residual and SUVmean in a 30-ml volume of interest (VOI) in the liver, and is thus a strictly quantitative signal. The only difference between the qPET method and the rPET method, favoured by Laffon and Marthan, is that the former uses SUVpeak instead of SUVmax in the tumour residual and SUVmean instead of SUVmax in the liver. Our criticism of the rPET approach is that the SUVs of the single hottest voxels in the tumour and liver are used. It is well known, that SUVs obtained from larger, fixed regions of interest are more reproducible than single-pixel SUVs [5]. In particular, the average liver SUV in a 30-ml VOI better characterizes the overall liver uptake than the single hottest voxel since the FDGuptake in the liver is generally inhomogeneous. The wrong impression that the qPET method is not quantitative may have resulted from a further advantage of the qPET method, namely that, as described in the original paper, qPET signals can be translated into the currently established international standard – the Deauville score. This is missing for rPET. A newmethod should always connect to the existing standards it attempts to improve upon. There are two principal strategies to derive cut-off values for a quantitative signal: one is based on the overall distribution of the signals and the other on the prognostic value. In the original article on the qPET method, we used the fact that in Hodgkin lymphoma, the majority of patients show a complete metabolic response to chemotherapy. This allows identification of the normal peak of the distribution as signals from patients achieving a complete metabolic response and designation of signals in the broad tail of the distribution as abnormal. This is similar to wellestablished reasoning with any laboratory parameter. With this method, we proposed two cut-off values: a qPET value of 1.3 for a sensitive reading and a qPET value of 2 for a specific reading to identify patients with an abnormal PET response. The second approach to setting a cut-off value relies on prognosis, in the simplest case using a receiver operating characteristic (ROC) curve as in the article on the rPET method. Unfortunately, this method is bedevilled by several problems: While with the distributional approach all measurements contribute to the analysis, the ROC curve additionally requires mature follow-up, and essentially only relapses are informative. Relapses are, however, rare in Hodgkin lymphoma with adequate treatment (about 10–15% in our studies). But this * Regine Kluge regine.kluge@medizin.uni-leipzig.de