TMTV-Net: fully automated total metabolic tumor volume segmentation in lymphoma PET/CT images - a multi-center generalizability analysis.

PURPOSE: Total metabolic tumor volume (TMTV) segmentation has significant value enabling quantitative imaging biomarkers for lymphoma management. In this work, we tackle the challenging task of automated tumor delineation in lymphoma from PET/CT scans using a cascaded approach. METHODS: Our study included 1418 2-[18F]FDG PET/CT scans from four different centers. The dataset was divided into 900 scans for development/validation/testing phases and 518 for multi-center external testing. The former consisted of 450 lymphoma, lung cancer, and melanoma scans, along with 450 negative scans, while the latter consisted of lymphoma patients from different centers with diffuse large B cell, primary mediastinal large B cell, and classic Hodgkin lymphoma cases. Our approach involves resampling PET/CT images into different voxel sizes in the first step, followed by training multi-resolution 3D U-Nets on each resampled dataset using a fivefold cross-validation scheme. The models trained on different data splits were ensemble. After applying soft voting to the predicted masks, in the second step, we input the probability-averaged predictions, along with the input imaging data, into another 3D U-Net. Models were trained with semi-supervised loss. We additionally considered the effectiveness of using test time augmentation (TTA) to improve the segmentation performance after training. In addition to quantitative analysis including Dice score (DSC) and TMTV comparisons, the qualitative evaluation was also conducted by nuclear medicine physicians. RESULTS: Our cascaded soft-voting guided approach resulted in performance with an average DSC of 0.68 ± 0.12 for the internal test data from developmental dataset, and an average DSC of 0.66 ± 0.18 on the multi-site external data (n = 518), significantly outperforming (p < 0.001) state-of-the-art (SOTA) approaches including nnU-Net and SWIN UNETR. While TTA yielded enhanced performance gains for some of the comparator methods, its impact on our cascaded approach was found to be negligible (DSC: 0.66 ± 0.16). Our approach reliably quantified TMTV, with a correlation of 0.89 with the ground truth (p < 0.001). Furthermore, in terms of visual assessment, concordance between quantitative evaluations and clinician feedback was observed in the majority of cases. The average relative error (ARE) and the absolute error (AE) in TMTV prediction on external multi-centric dataset were ARE = 0.43 ± 0.54 and AE = 157.32 ± 378.12 (mL) for all the external test data (n = 518), and ARE = 0.30 ± 0.22 and AE = 82.05 ± 99.78 (mL) when the 10% outliers (n = 53) were excluded. CONCLUSION: TMTV-Net demonstrates strong performance and generalizability in TMTV segmentation across multi-site external datasets, encompassing various lymphoma subtypes. A negligible reduction of 2% in overall performance during testing on external data highlights robust model generalizability across different centers and cancer types, likely attributable to its training with resampled inputs. Our model is publicly available, allowing easy multi-site evaluation and generalizability analysis on datasets from different institutions.

Baseline metabolic tumour burden improves risk stratification in Hodgkin lymphoma: A Children's Oncology Group study.

The Children's Oncology Group AHOD0831 study used a positron emission tomography (PET) response-adapted approach in high-risk Hodgkin lymphoma, whereby slow early responders (SERs) received more intensive therapy than rapid early responders (RERs). We explored if baseline PET-based characteristics would improve risk stratification. Of 166 patients enrolled in the COG AHOD0831 study, 94 (57%) had baseline PET scans evaluable for quantitative analysis. For these patients, total body metabolic tumour volume (MTV), total lesion glycolysis (TLG), maximum standardized uptake value (SUVmax ) and peak SUV (SUVpeak ) were obtained. MTV/TLG thresholds were an SUV of 2.5 (MTV2.5 /TLG2.5 ) and 40% of the tumour SUVmax (MTV40% /TLG40% ). TLG2.5 was associated with event-free survival (EFS) in the complete cohort (p = 0.04) and in RERs (p = 0.01), but not in SERs (p = 0.8). The Youden index cut-off for TLG2.5 was 1841. Four-year EFS was 92% for RER/TLG2.5  up to 1841, 60% for RER/TLG2.5  greater than 1841, 74% for SER/TLG2.5  up to 1841 and 79% for SER/TLG2.5  greater than 1841. Second EFS for RER/TLG2.5  up to 1841 was 100%. Thus, RERs with a low baseline TLG2.5 experienced excellent EFS with less intensive therapy, whereas RERs with a high baseline TLG2.5 experienced poor EFS. These findings suggest that patients with a high upfront tumour burden may benefit from intensified therapy, even if they achieve a RER.

PSMA PET/CT: joint EANM procedure guideline/SNMMI procedure standard for prostate cancer imaging 2.0.

Here we aim to provide updated guidance and standards for the indication, acquisition, and interpretation of PSMA PET/CT for prostate cancer imaging. Procedures and characteristics are reported for a variety of available PSMA small radioligands. Different scenarios for the clinical use of PSMA-ligand PET/CT are discussed. This document provides clinicians and technicians with the best available evidence, to support the implementation of PSMA PET/CT imaging in research and routine practice.

Imaging recommendations in pediatric lymphoma: A COG Diagnostic Imaging Committee/SPR Oncology Committee White Paper.

Hodgkin lymphoma (HL) and non-Hodgkin lymphoma (NHL) are both malignancies originating in the lymphatic system and both affect children, but many features differ considerably, impacting workup and management. This paper provides consensus-based imaging recommendations for evaluation of patients with HL and NHL at diagnosis and response assessment for both interim and end of therapy (follow-up).

Prognostic value of chest x-ray- and CT-defined large mediastinal adenopathy in high-risk pediatric Hodgkin lymphoma: A report from the Children's Oncology Group Study AHOD0831.

PURPOSE/OBJECTIVE: We compared the prognostic value of chest radiograph (CXR)- and computed tomography (CT)-derived definition of large mediastinal adenopathy (LMA) in pediatric Hodgkin lymphoma (HL). MATERIALS/METHODS: Total 143 patients treated for stage IIIB/IVB HL on COG AHOD0831 were included in this study. Six definitions of LMA were investigated: (i) mediastinal mass ratio on CXR (MRCXR ) > 1/3; (ii) mediastinal mass ratio on CT (MRCT ) > 1/3; (iii) mediastinal mass volume on CT (MVCT ) > 200 mL; (iv) normalized mediastinal mass volume (MVCT /thoracic diameter [TD]) > 1 mL/mm; (v) mediastinal mass diameter on CT (MDCT ) > 10 cm; and (vi) normalized mediastinal mass diameter (MDCT /TD) > 1/3. RESULTS: Median age at diagnosis was 15.8 years (range: 5.2-21.3 years). In patients with a slow early response (SER) to chemotherapy, MVCT  > 200 mL, MDCT  > 10 cm, and MDCT /TD > 1/3 were associated with worse relapse-free survival (RFS) on MVA, while MRCXR  > 1/3, MRCT  > 1/3, and MVCT /TD > 1 mL/mm trended toward worse RFS; MDCT /TD was the most strongly prognostic for inferior RFS, with a hazard ratio of 6.41 for MDCT /TD > 1/3 versus ≤1/3 on MVA (p = .02). CONCLUSION: LMA according to MVCT  > 200 mL, MDCT  > 10 cm, and MDCT /TD > 1/3 is associated with poor prognosis in advanced-stage HL patients with SER. The normalized mediastinal diameter, MDCT /TD > 1/3 appears to be the strongest predictor of inferior RFS.

Prognostic value of baseline metabolic tumor volume in children and adolescents with intermediate-risk Hodgkin lymphoma treated with chemo-radiation therapy: FDG-PET parameter analysis in a subgroup from COG AHOD0031.

BACKGROUND: Positron emission tomography (PET)-based measures of baseline total-body tumor burden may improve risk stratification in intermediate-risk Hodgkin lymphoma (HL). MATERIALS AND METHODS: Evaluable patients were identified from a cohort treated homogeneously with the same combined modality regimen on the Children's Oncology Group AHOD0031 study. Eligible patients had high-quality baseline PET scans. Metabolic tumor volume (MTV) and total lesion glycolysis (TLG) were each measured based on 15 thresholds for every patient. Univariate and multivariable Cox regression and Kaplan-Meier survival analyses assessed for an association of MTV and TLG with event-free survival (EFS). RESULTS: From the AHOD0031 cohort (n = 1712), 86 patients were identified who (i) were treated with four cycles of doxorubicin, bleomycin, vincristine, etoposide, prednisone, cyclophosphamide (ABVE-PC) chemotherapy followed by involved field radiotherapy, and (ii) had a baseline PET scan that was amenable to quantitative analysis. Based on univariate Cox regression analysis, six PET-derived parameters were significantly associated with EFS. For each of these, Kaplan-Meier analyses and the log-rank test were used to compare patients with highest tumor burden (i.e., highest 15%) to the remainder of the cohort. EFS was significantly associated with all six PET parameters (all p < .029). In a multivariable model controlling for important covariates including disease bulk and response to chemotherapy, MTV2BP was significantly associated with EFS (p = .012). CONCLUSION: Multiple baseline PET-derived volumetric parameters were associated with EFS. MTV2BP was highly associated with EFS when controlling for disease bulk and response to chemotherapy. Incorporation of baseline MTV into risk-based treatment algorithms may improve outcomes in intermediate-risk HL.

Evaluation of Musculoskeletal and Pulmonary Bacterial Infections With [124I]FIAU PET/CT.

PURPOSE: Imaging is limited in the evaluation of bacterial infection. Direct imaging of in situ bacteria holds promise for noninvasive diagnosis. We investigated the ability of a bacterial thymidine kinase inhibitor ([124I]FIAU) to image pulmonary and musculoskeletal infections. METHODS: Thirty-three patients were prospectively accrued: 16 with suspected musculoskeletal infection, 14 with suspected pulmonary infection, and 3 with known rheumatoid arthritis without infection. Thirty-one patients were imaged with [124I]FIAU PET/CT and 28 with [18F]FDG PET/CT. Patient histories were reviewed by an experienced clinician with subspecialty training in infectious diseases and were determined to be positive, equivocal, or negative for infection. RESULTS: Sensitivity, specificity, positive-predictive value, negative-predictive value, and accuracy of [124I]FIAU PET/CT for diagnosing infection were estimated as 7.7% to 25.0%, 0.0%, 50%, 0.0%, and 20.0% to 71.4% for musculoskeletal infections and incalculable-100.0%, 51.7% to 72.7%, 0.0% to 50.0%, 100.0%, and 57.1% to 78.6% for pulmonary infections, respectively. The parameters for [18F]FDG PET/CT were 75.0% to 92.3%, 0.0%, 23.1% to 92.3%, 0.0%, and 21.4% to 85.7%, respectively, for musculoskeletal infections and incalculable to 100.0%, 0.0%, 0.0% to 18.2%, incalculable, and 0.0% to 18.2% for pulmonary infections, respectively. CONCLUSIONS: The high number of patients with equivocal clinical findings prevented definitive conclusions from being made regarding the diagnostic efficacy of [124I]FIAU. Future studies using microbiology to rigorously define infection in patients and PET radiotracers optimized for image quality are needed.

ADVL1522: A phase 2 study of lorvotuzumab mertansine (IMGN901) in children with relapsed or refractory wilms tumor, rhabdomyosarcoma, neuroblastoma, pleuropulmonary blastoma, malignant peripheral nerve sheath tumor, or synovial sarcoma-A Children's Oncology Group study.

BACKGROUND: Lorvotuzumab mertansine (IMGN901) is an antibody-drug conjugate linking an antimitotic agent (DM1) to an anti-CD56 antibody (lorvotuzumab). Preclinical efficacy has been noted in Wilms tumor, rhabdomyosarcoma, and neuroblastoma. Synovial sarcoma, malignant peripheral nerve sheath tumor (MPNST), and pleuropulmonary blastoma also express CD56. A phase 2 trial of lorvotuzumab mertansine was conducted to assess its efficacy, recommended phase 2 dose, and toxicities. METHODS: Eligible patients had relapsed after or progressed on standard therapy for their tumor type. Lorvotuzumab mertansine (110 mg/m2 per dose) was administered at the adult recommended phase 2 dose intravenously on days 1 and 8 of 21-day cycles. Dexamethasone premedication was used. Pharmacokinetic samples, peripheral blood CD56-positive cell counts, and tumor CD56 expression were assessed. RESULTS: Sixty-two patients enrolled. The median age was 14.3 years (range, 2.8-29.9 years); 35 were male. Diagnoses included Wilms tumor (n = 17), rhabdomyosarcoma (n = 17), neuroblastoma (n = 12), synovial sarcoma (n = 10), MPNST (n = 5), and pleuropulmonary blastoma (n = 1). Five patients experienced 9 dose-limiting toxicities: hyperglycemia (n = 1), colonic fistula (n = 1) with perforation (n = 1), nausea (n = 1) with vomiting (n = 1), increased alanine aminotransferase in cycle 1 (n = 2), and increased alanine aminotransferase in cycle 2 (n = 1) with increased aspartate aminotransferase (n = 1). Non-dose-limiting toxicities (grade 3 or higher) attributed to lorvotuzumab mertansine were rare. The median values of the maximum concentration, half-life, and area under the curve from zero to infinity for DM1 were 0.87 µg/mL, 35 hours, and 27.9 µg/mL h, respectively. Peripheral blood CD56+ leukocytes decreased by 71.9% on day 8. One patient with rhabdomyosarcoma had a partial response, and 1 patient with synovial sarcoma achieved a delayed complete response. CONCLUSIONS: Lorvotuzumab mertansine (110 mg/m2 ) is tolerated in children at the adult recommended phase 2 dose; clinical activity is limited.

Response-adapted therapy for the treatment of children with newly diagnosed high risk Hodgkin lymphoma (AHOD0831): a report from the Children's Oncology Group.

The AHOD0831 study for paediatric patients with high risk Hodgkin lymphoma tested a response-based approach designed to limit cumulative alkylator exposure and reduce radiation volumes. Patients (Stage IIIB/IVB) received two cycles of ABVE-PC (doxorubicin, bleomycin, vincristine, etoposide, prednisone, cyclophosphamide). Rapid early responders [RER, no positron emission tomography (PET) activity above mediastinal blood pool] were consolidated with 2 cycles of ABVE-PC. Slow early responders (SER) received 2 cycles of ifosfamide/vinorelbine and 2 cycles of ABVE-PC. Radiotherapy was administered to sites of initial bulk and/or SER. By intent-to-treat analysis, 4-year second event-free survival (EFS; freedom from second relapse or malignancy) was 91·9% [95% confidence interval (CI): 86·1-95·3%], below the projected baseline of 95% (P = 0·038). Five-year first EFS and overall survival (OS) rates are 79·1% (95% CI: 71·5-84·8%) and 95% (95% CI: 88·8-97·8%). Eight of 11 SER patients with persistent PET positive lesions at the end of chemotherapy had clinical evidence of active disease (3 biopsy-proven, 5 with progressive disease or later relapses). Although this response-directed approach did not reach the ambitiously high pre-specified target for second EFS, EFS and OS rates are comparable with results of recent trials despite the reduction in radiotherapy volumes from historical involved fields. Persistent PET at end of chemotherapy identifies a cohort at an especially high risk for relapse/early progression.

Imaging for diagnosis, staging and response assessment of Hodgkin lymphoma and non-Hodgkin lymphoma.

Hodgkin lymphoma and non-Hodgkin lymphoma are common malignancies in children and are now highly treatable. Imaging plays a major role in diagnosis, staging and response using conventional CT and MRI and metabolic imaging with positron emission tomography (PET)/CT and PET/MRI. Cross-sectional imaging has replaced staging laparotomy and splenectomy by demonstrating abdominal nodal groups and organ involvement. [F-18]2-fluoro-2-deoxyglucose (FDG) PET provides information on bone marrow involvement, and MRI elucidates details of cortical bone and confirmation of bone marrow involvement. The staging system for Hodgkin lymphoma is the Ann Arbor system with Cotswald modifications and is based on imaging, whereas the non-Hodgkin staging system is the St. Jude Classification by Murphy or the more recent revised International Pediatric Non-Hodgkin Lymphoma Staging System (IPNHLSS). Because all pediatric lymphomas are metabolically FDG-avid and identify all nodal, solid organ, cortical bone and bone marrow disease, staging evaluations require FDG PET as PET/CT or PET/MRI in both Hodgkin and non-Hodgkin lymphoma. Both diseases have in common issues of airway compromise at presentation demonstrated by imaging. Differences exist in that Hodgkin lymphoma has several independent poor prognostic factors seen by imaging such as large mediastinal adenopathy, Stage IV disease, systemic symptoms, pleural effusion and pericardial effusion. Non-Hodgkin lymphoma includes more organ involvement such as renal, ovary, central nervous system and skin. Early or interim PET-negative scans are a reliable indicator of improved clinical outcome and optimize risk-adapted therapy and patient management; imaging may not, however, predict who will relapse. A recent multicenter trial has concluded that it is usually sufficient for pediatric lymphoma at staging and interim assessment to evaluate children with PET imaging from skull base to mid-thigh. Various systems of assessment of presence of disease or response are used, including the Deauville visual scale, where avidity is compared to liver; Lugano, which includes size change as part of response; or quantitative PET, which uses standardized uptake values to define more accurate response. Newer methods of immunotherapy can produce challenges in FDG PET evaluation because of inflammatory changes that may not represent disease.

Prognostic Value of Metabolic and Volumetric Parameters of FDG PET in Pediatric Osteosarcoma: A Hypothesis-generating Study.

Purpose To preliminarily assess the potential prognostic value of various fluorine 18 fluorodeoxyglucose (FDG) positron emission tomography (PET) parameters before, during, and after neoadjuvant chemotherapy (NCT). Materials and Methods Thirty-four patients with osteosarcoma were enrolled prospectively from 2008 to 2012 and underwent FDG PET/computed tomography (CT) imaging before (baseline scan), during (interim scan) and after NCT (posttherapy scan). The study was approved by the institutional review board and informed consent was received from patients. Maximum and peak standardized uptake value (SUVmax and SUVpeak), metabolic tumor volume (MTV), and total lesion glycolysis (TLG) were measured. Predictive value of FDG PET parameters for event-free survival (EFS) and overall survival (OS) were evaluated. Multivariable Cox regression analysis for EFS and OS was performed by using histologic response and initial presence of metastasis as covariates. Results At baseline scan, SUVpeak, MTV, and TLG were predictive of EFS (P = .006-.03) and OS (P = .001-.03) but not associated with histologic response. At interim and posttherapy scan, SUVmax, SUVpeak, MTV, and TLG were associated with histologic response (P = .0002-.04) and predictive of EFS (P = .004-.02) and OS (P = .001-.03). Multivariable Cox regression analysis revealed that the FDG PET parameters either at baseline, interim, or posttherapy were independently predictive of EFS and OS. In particular, baseline MTV was an independent predictor of EFS (hazard ratio, 5.0 [95% confidence interval {CI}: 1.5, 16.8]) and OS (hazard ratio, 29.4 [95% CI: 2.2, 392.2]). Conclusion SUVpeak, MTV, and TLG either at baseline, interim, or posttherapy were predictive of EFS and OS and may be useful prognostic biomarkers for osteosarcoma. © RSNA, 2018 Online supplemental material is available for this article.

Combined model-based and patient-specific dosimetry for 18F-DCFPyL, a PSMA-targeted PET agent.

PURPOSE: Prostate-specific membrane antigen (PSMA), a type-II integral membrane protein highly expressed in prostate cancer, has been extensively used as a target for imaging and therapy. Among the available PET radiotracers, the low molecular weight agents that bind to PSMA are proving particularly effective. We present the dosimetry results for 18F-DCFPyL in nine patients with metastatic prostate cancer. METHODS: Nine patients were imaged using sequential PET/CT scans at approximately 1, 12, 35 and 70 min, and a final PET/CT scan at approximately 120 min after intravenous administration of 321 ± 8 MBq (8.7 ± 0.2 mCi) of18F-DCFPyL. Time-integrated-activity coefficients were calculated and used as input in OLINDA/EXM software to obtain dose estimates for the majority of the major organs. The absorbed doses (AD) to the eye lens and lacrimal glands were calculated using Monte-Carlo models based on idealized anatomy combined with patient-specific volumes and activity from the PET/CT scans. Monte-Carlo based models were also developed for calculation of the dose to two major salivary glands (parotid and submandibular) using CT-based patient-specific gland volumes. RESULTS: The highest calculated mean AD per unit administered activity of 18F was found in the lacrimal glands, followed by the submandibular glands, kidneys, urinary bladder wall, and parotid glands. The S-values for the lacrimal glands to the eye lens (0.42 mGy/MBq h), the tear film to the eye lens (1.78 mGy/MBq h) and the lacrimal gland self-dose (574.10 mGy/MBq h) were calculated. Average S-values for the salivary glands were 3.58 mGy/MBq h for the parotid self-dose and 6.78 mGy/MBq h for the submandibular self-dose. The resultant mean effective dose of 18F-DCFPyL was 0.017 ± 0.002 mSv/MBq. CONCLUSIONS: 18F-DCFPyL dosimetry in nine patients was obtained using novel models for the lacrimal and salivary glands, two organs with potentially dose-limiting uptake for therapy and diagnosis which lacked pre-existing models.

Radiolabeled pertuzumab for imaging of human epidermal growth factor receptor 2 expression in ovarian cancer.

PURPOSE: Human epidermal growth factor receptor 2 (HER2) is over-expressed in over 30% of ovarian cancer cases, playing an essential role in tumorigenesis and metastasis. Non-invasive imaging of HER2 is of great interest for physicians as a mean to better detect and monitor the progression of ovarian cancer. In this study, HER2 was assessed as a biomarker for ovarian cancer imaging using 64Cu-labeled pertuzumab for immunoPET imaging. METHODS: HER2 expression and binding were examined in three ovarian cancer cell lines (SKOV3, OVCAR3, Caov3) using in vitro techniques, including western blot and saturation binding assays. PET imaging and biodistribution studies in subcutaneous models of ovarian cancer were performed for non-invasive in vivo evaluation of HER2 expression. Additionally, orthotopic models were employed to further validate the imaging capability of 64Cu-NOTA-pertuzumab. RESULTS: HER2 expression was highest in SKOV3 cells, while OVCAR3 and Caov3 displayed lower HER2 expression. 64Cu-NOTA-pertuzumab showed high specificity for HER2 (Ka = 3.1 ± 0.6 nM) in SKOV3. In subcutaneous tumors, PET imaging revealed tumor uptake of 41.8 ± 3.8, 10.5 ± 3.9, and 12.1 ± 2.3%ID/g at 48 h post-injection for SKOV3, OVCAR3, and Caov3, respectively (n = 3). In orthotopic models, PET imaging with 64Cu-NOTA-pertuzumab allowed for rapid and clear delineation of both primary and small peritoneal metastases in HER2-expressing ovarian cancer. CONCLUSIONS: 64Cu-NOTA-pertuzumab is an effective PET tracer for the non-invasive imaging of HER2 expression in vivo, rendering it a potential tracer for treatment monitoring and improved patient stratification.

Multidisciplinary intervention of early, lethal metastatic prostate cancer: Report from the 2015 Coffey-Holden Prostate Cancer Academy Meeting.

BACKGROUND: The 2015 Coffey-Holden Prostate Cancer Academy Meeting, themed: "Multidisciplinary Intervention of Early, Lethal Metastatic Prostate Cancer," was held in La Jolla, California from June 25 to 28, 2015. METHODS: The Prostate Cancer Foundation (PCF) sponsors an annual, invitation-only, action-tank-structured meeting on a critical topic concerning lethal prostate cancer. The 2015 meeting was attended by 71 basic, translational, and clinical investigators who discussed the current state of the field, major unmet needs, and ideas for addressing earlier diagnosis and treatment of men with lethal prostate cancer for the purpose of extending lives and making progress toward a cure. RESULTS: The questions addressed at the meeting included: cellular and molecular mechanisms of tumorigenesis, evaluating, and targeting the microenvironment in the primary tumor, advancing biomarkers for clinical integration, new molecular imaging technologies, clinical trials, and clinical trial design in localized high-risk and oligometastatic settings, targeting the primary tumor in advanced disease, and instituting multi-modal care of high risk and oligometastatic patients. DISCUSSION: This article highlights the current status, greatest unmet needs, and anticipated field changes that were discussed at the meeting toward the goal of optimizing earlier interventions to potentiate cures in high-risk and oligometastatic prostate cancer patients.

ImmunoPET Imaging of Insulin-Like Growth Factor 1 Receptor in a Subcutaneous Mouse Model of Pancreatic Cancer.

The role of insulin-like growth factor-1 receptor (IGF-1R) in cancer tumorigenesis was established decades ago, yet there are limited studies evaluating the imaging and therapeutic properties of anti-IGF-1R antibodies. Noninvasive imaging of IGF-1R may allow for optimized patient stratification and monitoring of therapeutic response in patients. Herein, this study reports the development of a Zirconium-89 ((89)Zr)-labeled anti-IGF-1R antibody ((89)Zr-Df-1A2G11) for PET imaging of pancreatic cancer. Successful chelation and radiolabeling of the antibody resulted in a highly stable construct that could be used for imaging IGF-1R expressing tumors in vivo. Western blot and flow cytometry studies showed that MIA PaCa-2, BxPC-3, and AsPC-1 pancreatic cancer cell lines expressed high, moderate, and low levels of IGF-1R, respectively. These three pancreatic cancer cell lines were subcutaneously implanted into mice. By employing the PET imaging technique, the tumor accumulation of (89)Zr-Df-1A2G11 was found to be dependent on the level of IGF-1R expression. Tumor accumulation of (89)Zr-Df-1A2G11 was 8.24 ± 0.51, 5.80 ± 0.54, and 4.30 ± 0.42 percentage of the injected dose (%ID/g) in MIA PaCa-2, BxPC-3, and AsPC-1-derived tumor models at 120 h postinjection, respectively (n = 4). Biodistribution studies and ex vivo immunohistochemistry confirmed these findings. In addition, (89)Zr-labeled nonspecific human IgG ((89)Zr-Df-IgG) displayed minimal uptake in IGF-1R positive MIA PaCa-2 tumor xenografts (3.63 ± 0.95%ID/g at 120 h postinjection; n = 4), demonstrating that (89)Zr-Df-1A2G11 accumulation was highly specific. This study provides initial evidence that our (89)Zr-labeled IGF-1R-targeted antibody may be employed for imaging a wide range of malignancies. Antibodies may be tracked in vivo for several days to weeks with (89)Zr, which may enhance image contrast due to decreased background signal. In addition, the principles outlined in this study can be employed for identifying patients that may benefit from anti-IGF-1R therapy.

The role of PET in the evaluation of musculoskeletal infections.

Musculoskeletal infections, particularly those in regions of complex anatomy and at postoperative sites, can be difficult to evaluate clinically, yet accurate diagnosis is essential to treat these infections promptly and correctly. Although anatomical imaging modalities are commonly used to diagnose musculoskeletal infections, the application of functional imaging through nuclear medicine techniques has long played a unique and important role in the evaluation of patients with suspected infections. In addition to the standard nuclear medicine single-photon emitting radiotracers traditionally used by nuclear medicine for planar and single-photon emission computed tomography infection imaging, in recent years we have seen a growth in the application of fluorodeoxyglucose (FDG) positron emission tomography (PET) and PET/CT applied to the detection and diagnosis of infection. In this review, we discuss the most common clinical scenarios in which musculoskeletal infection is suspected and for which FDG PET has been studied, in spinal, diabetic foot, and periprosthetic infections. We also highlight recent advances in new non-FDG PET radiotracers that may in the future enable better specificity for and characterization of musculoskeletal infections.

Personalized Impression Generation for PET Reports Using Large Language Models.

Large language models (LLMs) have shown promise in accelerating radiology reporting by summarizing clinical findings into impressions. However, automatic impression generation for whole-body PET reports presents unique challenges and has received little attention. Our study aimed to evaluate whether LLMs can create clinically useful impressions for PET reporting. To this end, we fine-tuned twelve open-source language models on a corpus of 37,370 retrospective PET reports collected from our institution. All models were trained using the teacher-forcing algorithm, with the report findings and patient information as input and the original clinical impressions as reference. An extra input token encoded the reading physician's identity, allowing models to learn physician-specific reporting styles. To compare the performances of different models, we computed various automatic evaluation metrics and benchmarked them against physician preferences, ultimately selecting PEGASUS as the top LLM. To evaluate its clinical utility, three nuclear medicine physicians assessed the PEGASUS-generated impressions and original clinical impressions across 6 quality dimensions (3-point scales) and an overall utility score (5-point scale). Each physician reviewed 12 of their own reports and 12 reports from other physicians. When physicians assessed LLM impressions generated in their own style, 89% were considered clinically acceptable, with a mean utility score of 4.08/5. On average, physicians rated these personalized impressions as comparable in overall utility to the impressions dictated by other physicians (4.03, P = 0.41). In summary, our study demonstrated that personalized impressions generated by PEGASUS were clinically useful in most cases, highlighting its potential to expedite PET reporting by automatically drafting impressions.

Automatic Quantification of Serial PET/CT Images for Pediatric Hodgkin Lymphoma Patients Using a Longitudinally-Aware Segmentation Network.

Purpose: Automatic quantification of longitudinal changes in PET scans for lymphoma patients has proven challenging, as residual disease in interim-therapy scans is often subtle and difficult to detect. Our goal was to develop a longitudinally-aware segmentation network (LAS-Net) that can quantify serial PET/CT images for pediatric Hodgkin lymphoma patients. Materials and Methods: This retrospective study included baseline (PET1) and interim (PET2) PET/CT images from 297 patients enrolled in two Children's Oncology Group clinical trials (AHOD1331 and AHOD0831). LAS-Net incorporates longitudinal cross-attention, allowing relevant features from PET1 to inform the analysis of PET2. Model performance was evaluated using Dice coefficients for PET1 and detection F1 scores for PET2. Additionally, we extracted and compared quantitative PET metrics, including metabolic tumor volume (MTV) and total lesion glycolysis (TLG) in PET1, as well as qPET and ΔSUVmax in PET2, against physician measurements. We quantified their agreement using Spearman's ρ correlations and employed bootstrap resampling for statistical analysis. Results: LAS-Net detected residual lymphoma in PET2 with an F1 score of 0.606 (precision/recall: 0.615/0.600), outperforming all comparator methods (P<0.01). For baseline segmentation, LAS-Net achieved a mean Dice score of 0.772. In PET quantification, LAS-Net's measurements of qPET, ΔSUVmax, MTV and TLG were strongly correlated with physician measurements, with Spearman's ρ of 0.78, 0.80, 0.93 and 0.96, respectively. The quantification performance remained high, with a slight decrease, in an external testing cohort. Conclusion: LAS-Net demonstrated significant improvements in quantifying PET metrics across serial scans, highlighting the value of longitudinal awareness in evaluating multi-time-point imaging datasets.

Voxel-Level Dosimetry for Combined Iodine 131 Radiopharmaceutical Therapy and External Beam Radiation Therapy Treatment Paradigms for Head and Neck Cancer.

PURPOSE: Targeted radiopharmaceutical therapy (RPT) in combination with external beam radiation therapy (EBRT) shows promise as a method to increase tumor control and mitigate potential high-grade toxicities associated with re-treatment for patients with recurrent head and neck cancer. This work establishes a patient-specific dosimetry framework that combines Monte Carlo-based dosimetry from the 2 radiation modalities at the voxel level using deformable image registration (DIR) and radiobiological constructs for patients enrolled in a phase 1 clinical trial combining EBRT and RPT. METHODS AND MATERIALS: Serial single-photon emission computed tomography (SPECT)/computed tomography (CT) patient scans were performed at approximately 24, 48, 72, and 168 hours postinjection of 577.2 MBq/m2 (15.6 mCi/m2) CLR 131, an iodine 131-containing RPT agent. Using RayStation, clinical EBRT treatment plans were created with a treatment planning CT (TPCT). SPECT/CT images were deformably registered to the TPCT using the Elastix DIR module in 3D Slicer software and assessed by measuring mean activity concentrations and absorbed doses. Monte Carlo EBRT dosimetry was computed using EGSnrc. RPT dosimetry was conducted using RAPID, a GEANT4-based RPT dosimetry platform. Radiobiological metrics (biologically effective dose and equivalent dose in 2-Gy fractions) were used to combine the 2 radiation modalities. RESULTS: The DIR method provided good agreement for the activity concentrations and calculated absorbed dose in the tumor volumes for the SPECT/CT and TPCT images, with a maximum mean absorbed dose difference of -11.2%. Based on the RPT absorbed dose calculations, 2 to 4 EBRT fractions were removed from patient EBRT treatments. For the combined treatment, the absorbed dose to target volumes ranged from 57.14 to 75.02 Gy. When partial volume corrections were included, the mean equivalent dose in 2-Gy fractions to the planning target volume from EBRT + RPT differed -3.11% to 1.40% compared with EBRT alone. CONCLUSIONS: This work demonstrates the clinical feasibility of performing combined EBRT + RPT dosimetry on TPCT scans. Dosimetry guides treatment decisions for EBRT, and this work provides a bridge for the same paradigm to be implemented within the rapidly emerging clinical RPT space.