Dynamic blood dose estimates in radiotherapy and correlations with adverse clinical outcomes in brain, head‐and‐neck, and lung cancer patients
Bibliographic record
Abstract
Abstract Background In cancer radiotherapy, radiation‐induced lymphopenia (RIL) has been reported to be correlated with adverse clinical outcomes such as reduced locoregional control (LRC), distant‐metastasis‐free survival (DMFS), and overall survival (OS) in various treatment sites. Frameworks to simulate the radiation dose to circulating blood have been developed in response, and simulated blood dose values have been reported to be correlated with severe RIL and/or adverse clinical outcomes. However, validations with different patient datasets or expansions to additional treatment sites, as well as the identification of particularly relevant blood dose metrics and blood compartments to allow for their inclusion during radiotherapy treatment planning, remain lacking. Purpose This study aims to investigate a potential correlation between simulated blood dose values and adverse clinical outcomes in 215 patients with head‐and‐neck squamous cell carcinoma (HNSCC), 180 patients with glioblastoma (GBM), and 490 patients with non‐small‐cell lung cancer (NSCLC), and to identify particularly relevant blood dose metrics and blood compartments to allow for their inclusion during radiotherapy treatment planning and thereby enable the optimization of the estimated dose delivered to circulating blood. Methods For all 885 patients, TotalSegmentator was used to automatically delineate additional organs‐at‐risk (OARs), blood vessels, and tissues which were not already manually delineated for radiotherapy treatment planning. Subsequently, the dynamic HEDOS model, which considers temporal aspects such as blood flow dynamics and treatment delivery time, was used to simulate the radiation dose delivered to circulate blood during radiotherapy. Static blood dose models consisting of the mean dose to the union of all HEDOS blood compartments ( D static,HEDOS ) and the integral body dose ( D static,body ) were also investigated to verify whether a simplified blood dose model equally exhibited any correlations with adverse clinical outcomes. Results During multivariable Cox regression analysis, the blood dose estimates from the dynamic blood dose model exhibited a statistically significant ( p < 0.05) correlation with DMFS and OS in the HNSCC and NSCLC datasets as well as with LRC in the HNSCC dataset. D static,body and D static,HEDOS only exhibited a statistically significant correlation with OS in the GBM and NSCLC datasets. Within a dataset, different dynamic blood dose metrics generally consistently exhibited correlations with the same clinical outcomes. Large arteries and veins were found to be a particularly relevant blood compartment within the HNSCC dataset, while the dose to the healthy portion of the brain and the dose to the heart and lungs were found to exhibit particularly strong correlations with dynamic blood dose estimates in the GBM and NSCLC datasets, respectively. Conclusions The dynamic blood dose model exhibited a statistically significant correlation with adverse clinical outcomes in five out of seven cases, compared to just two cases for the static blood dose models. Consideration of temporal aspects such as blood flow dynamics and treatment delivery time was therefore essential to some of the observed correlations. For each treatment site, particularly relevant blood compartments were identified, allowing for their inclusion during radiotherapy treatment planning as part of future studies which aim to reduce the estimated dose to circulating blood.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".