Here’s a breakdown of the LOTIS-2 trial and its findings, based on the provided text:
What is the LOTIS-2 Trial?
Purpose: The LOTIS-2 trial (NCT03589469) investigated the use of positron emission tomography/computed tomography (PET/CT) scans in patients wiht relapsed/refractory diffuse large B-cell lymphoma (R/R DLBCL) treated with the antibody-drug conjugate (ADC) loncastuximab tesirine.
Key Technology: PET/CT scans were used to assess both anatomy and function, providing information about cancer presence and perhaps heart disease. While it can detect brain metastases, it’s not the primary test for this.
What was Measured and Analyzed?
Body Composition Analysis: Using the PET/CT scans, researchers performed body composition analyses. This involved segmenting (identifying and outlining) three primary tissue areas:
Skeletal muscle
Subcutaneous fat
Visceral fat
Measurement Location: These measurements were taken at the third lumbar vertebra (L3 level), a standard method to represent overall body muscle and fat distribution.
Methods of Segmentation: Both manual and deep learning-based segmentation techniques were used.
Body Composition Ratio Indices: From these segmented regions, several ratios were calculated:
Skeletal muscle to visceral fat ratio
Subcutaneous fat to visceral fat ratio
Skeletal muscle to a composite of visceral and subcutaneous fat ratio
Agreement and Outcome Analysis:
The study examined the agreement between manual and automated (deep learning) measurements.
The relationship between these body composition indices and treatment response was analyzed, specifically how body composition affected time-to-event outcomes.
Time-to-Event Outcomes: Kaplan-Meier curves were used to estimate:
Progression-free survival (PFS)
Overall survival (OS)
Key Findings:
Predictors of Metabolic Response: Both manual and automated skeletal muscle/visceral fat indices (when categorized into two groups) were significant predictors of failure to achieve complete metabolic response in both univariable and multivariable logistic models. This means that certain body composition profiles were associated with a lower likelihood of the cancer responding well to treatment at a metabolic level.
Association with PFS: The manual skeletal muscle/visceral fat index was considerably associated with progression-free survival (PFS) in both univariable and multivariable models. However, it was not significantly associated with overall survival (OS). This suggests that body composition might influence how long patients remain without their cancer progressing, but not necessarily their overall lifespan in this specific analysis.
Potential Explanations for the Findings:
Sarcopenia: A 2023 article in Ageing Research Reviews suggests that sarcopenia (lack of muscle strength and quality) can limit the response to ADCs.
Obesity: Obesity can led to dose reductions during treatment, which can also limit the effectiveness of the therapy.
Clinical Trial vs. Real-World: The text notes that clinical trials ofen enroll healthier patients, so these issues (sarcopenia, obesity-related dose reductions) might become more apparent when drugs are used in real-world settings after approval.
Conclusions and Future Implications:
Biomarker Potential: The authors concluded that a patient’s skeletal muscle to visceral fat ratio before treatment could be a useful biomarker for evaluating patients with R/R DLBCL treated with loncastuximab tesirine.
* Deep Learning Efficiency: The deep learning-based approach for body composition analysis showed comparable performance to manual methods and offers a more cost-effective choice.
In essence, the LOTIS-2 trial highlights the potential of using body composition analysis, particularly the skeletal muscle to visceral fat ratio, as a predictive tool for treatment response in R/R DLBCL patients receiving loncastuximab tesirine. The study also demonstrates the promise of deep learning for automating these analyses.