Cristina Nanni

Background/objective: Multiple myeloma (MM) is a hematological malignancy characterized by the proliferation of clonal plasma cells. While [¹⁸F]FDG PET/CT is widely established for assessing disease extent, bone involvement, and treatment response in MM, novel molecular imaging targets like CXCR4 (e.g., using [⁶⁸Ga]Ga-Pentixafor) have emerged to offer potentially superior sensitivity and theranostic applications. The aim of this manuscript is to systematically review and compare the diagnostic performance, expressed as detection rates, of CXCR4-targeted PET/CT versus [¹⁸F]FDG PET/CT in patients with MM. Methods: A systematic literature search was conducted across major databases (PubMed/MEDLINE, Embase, and the Cochrane Library) for head-to-head comparative studies evaluating both [¹⁸F]FDG and [⁶⁸Ga]Ga-Pentixafor PET/CT in MM patients. Risk of bias and study quality were assessed using the QUADAS-2 tool. Results: Eleven studies were collected, amounting to a total of 347 patients. A patient-based analysis showed that [⁶⁸Ga]Ga-Pentixafor had a higher detection rate than [¹⁸F]FDG PET/CT (86.3%, 95% CI: 80.3–92.3% vs 60.9%, 95% CI: 42.5–79.4%). In a descriptive analysis, mean lesion SUVmax values reported across individual studies were generally higher with [⁶⁸Ga]Ga-Pentixafor PET/CT than with [¹⁸F]FDG PET/CT. Conclusions: [⁶⁸Ga]Ga-Pentixafor PET/CT may serve as a robust alternative or complementary imaging modality to [¹⁸F]FDG PET/CT in MM. It could also be helpful for selection in CXCR4-directed theranostic pathways, but more high-quality studies are warranted.

Purpose: Deauville Score (DS) assessment from [ ¹⁸ F]FDG PET/CT in multiple myeloma (MM) relies on visual interpretation, limiting reproducibility. This study aimed to develop an automated pipeline for standardised DS computation, evaluate the prognostic value of DS at five anatomical sites, and predict progression-free survival (PFS) by integrating DS with copy number alterations (CNA) and clinical variables. Methods: A retrospective cohort of 165 newly diagnosed MM patients with baseline FDG PET/CT, CNA profiling, and blood tests was analysed. An automated pipeline computed DS fully automatically for vertebral bone marrow (BM) and long bones (LB), and semi-automatically for focal (FL), paramedullary (PM), and extramedullary (EM) lesions. DS were subdivided into absent (1), low (2–3), and high (4–5) groups and compared via log-rank test. A penalised Cox model with 12 covariates (five DS, three CNAs, haemoglobin, platelet count, age, sex) was evaluated via nested cross-validation for PFS prediction. For inference on individual prognostic contributions, an unpenalised multivariable Cox model was fitted on the full cohort. Results: In univariate analyses, high LB, PM and EM DS were significantly associated with shorter PFS. The penalised Cox model achieved a C-index of 0.710 [95% CI: 0.689–0.732] in predicting the risk of progression. In the multivariable analysis, age, haemoglobin, BM DS, PM DS and amp(1q) were identified as independent prognostic factors. Conclusion: Automated DS computation from baseline FDG PET/CT is feasible and, combined with genomic and clinical data, enables a reproducible multimodal approach to prognostic stratification in MM. The pipeline for DS computation is publicly available as an open-source tool (autoDS-PET) at https://github.com/Sara-Peluso/autoDS-PET .

Purpose: To test whether internal memory states from a medical founda-tional segmentation model can serve as compact, mask-aware embeddings for predicting progression-free survival (PFS) in multiple myeloma (MM) from whole-body [ ¹⁸ F]FDG PET/CT, and how late fusion of PET, CT, and clinical data enhances prognostic performance. Methods: We analyzed 227 newly diagnosed MM patients with PET/CT and clinical data. For two regions of interest (spine-dilated and full skeleton), we prompted MedSAM2 slice-wise using mask-derived bounding boxes and cached the final spatio-temporal memory tensor per modality. We compared two downsampling strategy to obtain per-study embeddings: channel×memory averaging with a small CNN head, and depth-attention pooling. PET and CT embeddings were combined by late fusion and passed to a DeepSurv head. We evaluated image-only and multimodal (image+clinical) models with stratified 5-fold cross-validation. The primary endpoint was Harrell’s c-index (mean ± SE across folds). Results: Image-only models using the averaging downsampler achieved up to 0.659 ± 0.015 c-index (PET, spine-dilated), comparable to baseline ra-diomics results. Multimodal models improved discrimination to 0.710±0.032 (CT, spine-dilated), with similar performance for other PET/CT+clinical variants (0.703–0.710), improving clinical-only baselines ∼ 6.5%. Averaging consistently outperformed depth-attention; concatenation and gated fusion performed comparably. PET outperformed CT within the same mask in image-only settings. Conclusion: Mask-aware memory embeddings extracted from a founda-tional segmentation model provide effective, data-efficient imaging biomark-ers for MM PFS and, when fused with routine clinical covariates, significantly improve risk stratification over clinical-only or radiomics baselines. This of-fers a practical path to prognostic modeling on small medical cohorts without feature design.

Lymphomas, including Hodgkin lymphoma (HL) and non-Hodgkin lymphoma (NHL), represent a significant proportion of malignancies affecting diverse age groups, including children and pregnant women. Traditional imaging modalities, such as 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG-PET/CT) and contrast-enhanced CT, are standard for staging and monitoring but expose patients to ionizing radiation, increasing the risk of secondary malignancies. This review evaluates whole-body magnetic resonance imaging (WB-MRI) as a radiation-free alternative for assessing lymphoproliferative disorders. We examine its strengths, including the ability to detect disease and assess treatment response, as well as its limitations, such as challenges in visualizing small thoracic lesions. Recent studies demonstrate high concordance between WB-MRI and 18F-FDG-PET/CT, particularly for lymphomas with low or variable FDG avidity, making WB-MRI a promising modality for staging and follow-up, especially in young and pregnant patients.