Andrei Iagaru’s research while affiliated with Stanford University and other places

Aim To evaluate a deep learning-based time-of-flight (DLToF) model trained to enhance the image quality of non-ToF PET images for different tracers, reconstructed using BSREM algorithm, towards ToF images. Methods A 3D residual U-NET model was trained using 8 different tracers (FDG: 75% and non-FDG: 25%) from 11 sites from US, Europe and Asia. A total of 309 training and 33 validation datasets scanned on GE Discovery MI (DMI) ToF scanners were used for development of DLToF models of three strengths: low (L), medium (M) and high (H). The training and validation pairs consisted of target ToF and input non-ToF BSREM reconstructions using site-preferred regularisation parameters (beta values). The contrast and noise properties of each model were defined by adjusting the beta value of target ToF images. A total of 60 DMI datasets, consisting of a set of 4 tracers (¹⁸F-FDG, ¹⁸F-PSMA, ⁶⁸Ga-PSMA, ⁶⁸Ga-DOTATATE) and 15 exams each, were collected for testing and quantitative analysis of the models based on standardized uptake value (SUV) in regions of interest (ROI) placed in lesions, lungs and liver. Each dataset includes 5 image series: ToF and non-ToF BSREM and three DLToF images. The image series (300 in total) were blind scored on a 5-point Likert score by 4 readers based on lesion detectability, diagnostic confidence, and image noise/quality. Results In lesion SUVmax quantification with respect to ToF BSREM, DLToF-H achieved the best results among the three models by reducing the non-ToF BSREM errors from -39% to -6% for ¹⁸F-FDG (38 lesions); from -42% to -7% for ¹⁸F-PSMA (35 lesions); from -34% to -4% for ⁶⁸Ga-PSMA (23 lesions) and from -34% to -12% for ⁶⁸Ga-DOTATATE (32 lesions). Quantification results in liver and lung also showed ToF-like performance of DLToF models. Clinical reader resulted showed that DLToF-H results in an improved lesion detectability on average for all four radiotracers whereas DLToF-L achieved the highest scores for image quality (noise level). The results of DLToF-M however showed that this model results in the best trade-off between lesion detection and noise level and hence achieved the highest score for diagnostic confidence on average for all radiotracers. Conclusion This study demonstrated that the DLToF models are suitable for both FDG and non-FDG tracers and could be utilized for digital BGO PET/CT scanners to provide an image quality and lesion detectability comparable and close to ToF.

Purpose Conventional systemic therapies are valuable options in prostate cancer (PCa); however, such treatments can determine adverse events and toxicity. The observed improvement in overall survival, coupled with PSA reduction and a favorable safety profile in the post-taxane castration-resistant PCa (CRPC) setting has prompted the consideration of PSMA-based radioligand therapy (RLT) earlier in the treatment sequence. In this review, we will describe the literature and ongoing clinical trials regarding the use of PSMA-based RLT in hormone-sensitive PCa (HSPC) including the neoadjuvant, de-novo/synchronous metastatic, adjuvant, and early BCR settings. Methods We performed a systematic literature search on the PubMed/MEDLINE/EMBASE and clinicaltrials.gov databases for studies and protocols assessing the role of PSMA-based RLT in HSPC. Results The literature search yielded 140 results. After screening titles and abstracts and applying inclusion and exclusion criteria, we selected 25 papers showing the potentialities of earlier RLT in HSPC, with several ongoing trials. Conclusion Early use of PSMA-based RLT holds significant potential in HSPC patients from the neoadjuvant to the BCR setting. In these stages, the lower tumor burden, more frequent exclusive nodal involvement, and higher organ reserve may improve treatment efficacy and allow for treatment combinations while maintaining a less toxic profile.

107 Background: Patients (pts) with mCRPC can have different metastatic sites with variable outcomes. Lu177-PVT is a novel treatment for mCRPC but response and outcomes in the real world are still being defined. We hypothesized that pts with lung and liver metastases (mets) would have worse outcomes (lower PSA response and shorter survival) with Lu177-PVT than those without as advanced of disease spread. Methods: This retrospective cohort study analyzed pts with mCRPC who were treated with at least one dose of Lu177-PVT. PSA50, PSA90, and overall survival were assessed for three comparisons of pts: bone only vs. other mets (including other site ± bone), lung-involved vs. lung-absent, and liver-involved vs. liver-absent. PSA50 and PSA90 were defined as achieving any reduction in PSA below 50% and 90% of baseline prior to Lu177-PVT therapy, respectively. Comparisons between groups were made using univariable logistic regression for PSA50 and PSA90 and Cox regression for OS. Results: We included 100 pts for OS analyses and 97 pts for PSA50 and PSA90 analyses treated between 03/2019 and 09/2024. Three pts were excluded from PSA analyses due to absence of baseline PSA prior to Lu177-PVT. Median age at Lu177-PVT start was 66 years, 67% white race, 45% ever smoker, median BMI was 26 kg/m ² , 5 median previous lines of treatment, 77% received prior chemotherapy, and 4 median number of Lu-177-PVT treatments received. At time of initial Lu177-PVT, 94% had at least bone mets, 12% had LN mets, 14% had lung mets, 15% had liver mets, and 9% had other mets. Table shows results of all comparisons. Pts with only bone mets (vs those with other site ± bone) had numerically (not significant) longer OS (Table; median [m]OS 19 vs. 13 months, HR 0.88, p-value 0.70) and similar PSA50 and PSA90 rates. Pts with liver-involved (vs liver-absent) had a numerically shorter OS (mOS 15 vs. 18 months, HR 1.45 p-value 0.38) and lower (not significant) PSA90. Pts with lung-involved (vs lung-absent) had numerically shorter OS (15 vs. 18 months, HR 1.28, p-value 0.58) and lower PSA90. Conclusions: Pts with mCRPC responded to Lu177-PVT regardless of metastatic sites in our study, though liver mets had numerically lower OS. Study limited by small power and retrospective study design. Future work is needed to further dissect the interplay between Lu177-PVT response and metastatic disease sites. Results. N Median Lu177-PVT cycles received Median Overall Survival, months (95% CI) HR for OS, (95% CI) PSA50%, (95% CI) OR for PSA50, (95%CI) PSA90%, (95% CI) OR for PSA90, (95%CI) Bone only 56 5 19 (1-40) 0.88 (0.45-1.71) 58 (43-70) 1.04 (0.46-2.36) 33 (20-46) 1.56 (0.64-3.96) Lung-involved 15 4 15 (3-34) 1.28 (0.53-3.01) 57 (27-79) 0.97 (0.31-3.19) 14 (2-40) 0.37 (0.05-1.47) Liver-involved 15 3 15 (3-35) 1.45 (0.63-3.32) 67 (38-88) 1.57 (0.51-5.40) 13 (2-40) 0.33 (0.05-1.32)

Molecular imaging using positron emission tomography (PET) provides sensitive detection and mapping of molecular targets. While cancer-associated fibroblasts and integrins have been proposed as targets for imaging of pancreatic ductal adenocarcinoma (PDAC), herein, spatial transcriptomics and proteomics of human surgical samples are applied to select PDAC targets. We find that selected cancer cell surface markers are spatially correlated and provide specific cancer localization, whereas the spatial correlation between cancer markers and immune-related or fibroblast markers is low. Claudin-4 expression increases ~16 fold in cancer as compared with normal pancreas, and tight junction localization confers low background for imaging in normal tissue. We develop a peptide-based molecular imaging agent targeted to claudin-4 with accumulation to ~25% injected activity per cubic centimeter (IA/cc) in metastases and ~18% IA/cc in tumors. Our work motivates a data-driven approach to selection of molecular targets.

Background: PSMA PET radiomics is a promising tool for primary prostate cancer (PCa) characterisation. However, small single-centre studies and lack of external validation hinder definitive conclusions on the potential of PSMA PET radiomics in the initial workup of PCa. We aimed to validate a radiomics signature in a larger internal cohort and in an external cohort from a separate centre. Methods: One hundred and twenty-seven PCa patients were retrospectively enrolled across two independent hospitals. The first centre (IRCCS San Raffaele Scientific Institute, Centre 1) contributed 62 [⁶⁸Ga]Ga-PSMA-11 PET scans, 20 patients classified as low-grade (ISUP grade < 4), and 42 as high-grade (ISUP grade ≥ 4). The second centre (Stanford University Hospital, Centre 2) provided 65 [⁶⁸Ga]Ga-PSMA-11 PET scans, and 49 low-grade and 16 high-grade patients. A radiomics model previously generated in Centre 1 was tested on the two cohorts separately and afterward on the entire dataset. Then, we evaluated whether the radiomics features selected in the previous investigation could generalise to new data. Several machine learning (ML) models underwent training and testing using 100-fold Monte Carlo cross-validation, independently at both Centre 1 and Centre 2, with a 70–30% train–test split. Additionally, models were trained in one centre and tested in the other, and vice versa. Furthermore, data from both centres were combined for training and testing using Monte Carlo cross-validation. Finally, a new radiomics signature built on this bicentric dataset was proposed. Several performance metrics were computed. Results: The previously generated radiomics signature resulted in an area under the receiver operating characteristic curve (AUC) of 80.4% when tested on Centre 1, while it generalised poorly to Centre 2, where it reached an AUC of 62.7%. When the whole cohort was considered, AUC was 72.5%. Similarly, new ML models trained on the previously selected features yielded, at best, an AUC of 80.9% for Centre 1 and performed at chance for Centre 2 (AUC of 49.3%). A new signature built on this bicentric dataset reached, at best, an average AUC of 91.4% in the test set. Conclusions: The satisfying performance of radiomics models when used in the original development settings, paired with the poor performance otherwise observed, emphasises the need to consider centre-specific factors and dataset characteristics when developing radiomics models. Combining radiomics datasets is a viable strategy to reduce such centre-specific biases, but external validation is still needed.

As medicine advances toward personalized precision medicine, there is a clinical need for a more specific and targeted diagnosis of prostate cancer (PC). Specifically, given the heterogeneous tumor biology of PC, there is a need to distinguish clinically significant cancers from indolent tumors. Advanced imaging techniques such as molecular imaging with positron emission tomography (PET) aiming at specific targets such as prostate-specific membrane antigen (PSMA) and promising new targets such as gastrin-releasing peptide receptors (GRPR) might mitigate the shortcomings and risks of standard systematic template or multiparametric magnetic resonance imaging (mpMRI)-targeted prostate biopsy. This chapter explores this evolving alternative of PET-guided prostate biopsy utilizing different radiopharmaceuticals, targets, and biopsy techniques to evaluate its ability to identify clinically relevant PC. Finally, we will provide an outlook on ongoing trials and future perspectives.

Current US clinical practice guidelines for advanced prostate cancer management contain recommendations based on high‐level evidence from randomized controlled trials; however, these guidelines do not address the nuanced clinical questions that are unanswered by prospective trials but nonetheless encountered in day‐to‐day practice. To address these practical questions, the 2024 US Prostate Cancer Conference (USPCC 2024) was created to generate US‐focused expert clinical decision‐making guidance for circumstances in which level 1 evidence is lacking. At the second annual USPCC meeting (USPCC 2024), a multidisciplinary panel of experts convened to discuss ongoing clinical challenges related to 5 topic areas: biochemical recurrence; metastatic, castration‐sensitive prostate cancer; poly [ADP‐ribose] polymerase inhibitors; prostate‐specific membrane antigen radioligand therapy; and metastatic, castration‐resistant prostate cancer. Through a modified Delphi process, 34 consensus recommendations were developed and are intended to provide clinicians who manage prostate cancer with guidance related to the implementation of novel treatments and technologies. In this report, the authors review the areas of consensus identified by the USPCC 2024 experts and evaluate ongoing unmet needs regarding translational application of the current clinical evidence.