Mid-infrared (MIR) imaging is an emerging label-free modality for classifying tissue types, including viable tumor in highly heterogeneous cancers, by assessing spatial differences in chemical composition. However, common data analysis neglects spatial vicinity and relies on time-consuming pathological insight for hotspot prediction of viable tumor areas and computational tissue type annotation. Here, we present a method that uses spatial autocorrelation on MIR projection images computed from data of selected wavenumbers found by random forest ranking for computational tissue type annotation: Interdependent data processing enabled high accuracy annotations, whereas referencing of sequentially added new samples to a hyperspectral tissue database ensured computational efficiency and scalability for larger cohorts. Applied to clinical colorectal cancer liver metastasis samples, the method matched manual pathology assessment in a double-blind study. As an option, MIR-based hotspots can be correlated with mass spectrometry imaging. This multimodal approach identified sphingomyelin isoforms as lipidomic tumor marker candidates by imaging parallel reaction monitoring-parallel accumulation serial fragmentation (iprm-PASEF) directly on tissue. Taken together, spatial autocorrelation analysis on MIR imaging data could improve automated accurate annotation of tissue morphologies of heterogeneous cancer specimens and support the discovery of spatial cancer biomarkers.
Latent diffusion models (LDMs) have recently achieved strong performance in 3D medical image synthesis. However, modalities like cine cardiac MRI (CMR), representing a temporally synchronized 3D volume across the cardiac cycle, add an additional dimension that most generative approaches do not model directly. Instead, they factorize space and time or enforce temporal consistency through auxiliary mechanisms such as anatomical masks. Such strategies introduce structural biases that may limit global context integration and lead to subtle spatiotemporal discontinuities or physiologically inconsistent cardiac dynamics. We investigate whether a unified 4D generative model can learn continuous cardiac dynamics without architectural factorization. We propose CardioDiT, a 3D+t latent diffusion framework for short-axis cine CMR synthesis based on diffusion transformers. A spatiotemporal VQ-VAE encodes 2D+t slices into compact latents, which a diffusion transformer then models jointly as complete 3D+t volumes, coupling space and time throughout the generative process. We evaluate CardioDiT on public CMR datasets and a larger private cohort, comparing it to baselines with progressively stronger spatiotemporal coupling. Results show improved inter-slice consistency, temporally coherent motion, and realistic cardiac function distributions, suggesting that explicit 4D modeling with a diffusion transformer provides a principled foundation for spatiotemporal cardiac image synthesis. Code, configurations and models trained on public data are available at https://github.com/Cardio-AI/cardiodit .
Abstract Purpose: BI 905711, a TRAILR2/cadherin-17 (CDH17) bispecific antibody, demonstrated preclinical apoptotic pathway activation and antitumor activity. Two phase Ia/Ib studies tested BI 905711 monotherapy (NCT04137289) or combination therapy (NCT05087992) in advanced, refractory gastrointestinal (GI) cancers. Patients and Methods: Both studies aimed to determine the maximum tolerated dose (MTD; phase Ia) and recommended phase II dose (RP2D)/recommended dose for expansion (RDE; phase Ib). In phase Ia, patients received BI 905711 monotherapy (0.02–4.8 mg/kg) or 0.6 to 1.2 mg/kg plus biweekly folinic acid (leucovorin), 5-fluorouracil, and irinotecan (FOLFIRI) and bevacizumab. Phase Ib assessed selected doses and regimens given biweekly or weekly (3 weeks on and 1 week off). Safety, efficacy, and pharmacokinetics/pharmacodynamics were evaluated. Results: In NCT04137289, 110 patients [median age 61 years; 80% with colorectal cancer; median of three prior therapies (range, 1–6)] received monotherapy. No dose-limiting toxicities (DLT) occurred, MTD was not reached, RP2D was not determined, and 48.2% of patients had treatment-related adverse events (TRAE), most commonly nausea (16.4%). In 104 response-evaluable patients, 22.1% achieved stable disease (SD). In NCT05087992, 12 patients with colorectal cancer (median age 54.5 years) received combination treatment. Two patients reported DLTs, MTD was not reached, and the selected RDE was 0.6 mg/kg plus biweekly FOLFIRI and bevacizumab. Most patients (91.7%) had TRAEs, including decreased appetite (33.3%), alanine transaminase increased, aspartate transaminase increased, and diarrhea (25% each). Eleven patients (91.7%) achieved SD. Pharmacokinetic/pharmacodynamic data indicated linear dose exposure and ≥2-fold activation of plasma caspase 3/7. Conclusions: In heavily pretreated patients with GI tumors, BI 905711 monotherapy or with FOLFIRI plus bevacizumab displayed a manageable safety profile and limited clinical activity. Significance: Translating preclinical activity of TRAILR2 agonists into the clinic has been hampered by severe hepatotoxicity. BI 905711, a bispecific antibody against TRAILR2 and CDH17, was developed to enhance efficacy and reduce hepatotoxicity. Two phase 1 studies (NCT04137289 and NCT05087992) demonstrated tolerability and minimal hepatotoxicity, dose-proportional pharmacokinetics, and increased markers of target engagement. Antitumor activity was limited. These data show that BI 905711 has reduced TRAIL-related hepatotoxicity in the clinic.
