Cytotherapy · 2025
Cytokine release syndrome risk model with T-cell engaging therapies.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Lafeuille Pénélope, Blumentals William A, Brulle-Wohlhueter Claire, Chen Weixi, Sang Chao, Manning Sydney, Saltzman Silvy, Canvin Jan, Richards Susan, Kamperschroer Cris, Abbadessa Giovanni, Koduru Bhargav, Talwai Aniketh, Der-Nigoghossian Caroline, Itzkovich Yahav, Jain Rahul, Jain Tanmay, Aptekar Jacob, Grupp Stephan A
Аннотация
BACKGROUND: Cytokine release syndrome (CRS) is an adverse event associated with T-cell engaging (TCE) immuno-oncology therapies such as chimeric antigen receptor T cells (CAR-T), bispecific TCE antibodies and dual-affinity retargeting proteins.
OBJECTIVE: To develop a model to predict the preinfusion risk of CRS grade ≥2 for patients with solid tumors and hematologic malignancies such as acute lymphoblastic leukemia (ALL) and non-Hodgkin lymphoma (NHL) treated with TCE bispecific antibodies.
STUDY DESIGN: A TCE dataset including clinical trials from 2014 to 2019 evaluating non-CAR-T TCE therapies was sourced from the Medidata Enterprise Data Store, an anonymized data repository from completed clinical trials. The outcome of interest was the first CRS grade ≥2 occurring within 10 days of TCE therapy. Risk factors for CRS grade ≥2 were identified from the literature and preliminary data analysis. Features were measured prior to or at the first TCE treatment. Patients were included in the analysis dataset if they had a data element fill rate of >70% for the key features. Features were pruned by assessing multicollinearity across features. Logistic regression and tree-based models were trained. Across 100 iterations with different train-test splits, the average area under the receiver-operator characteristic (AUROC) curve was calculated for each model type.
RESULTS: A total of 715 patients (115 CRS grade ≥2 and 600 CRS grade <2) were included in the analysis; most patients had ALL (81%) and 19% had solid tumors or NHL. Patients who developed CRS grade ≥2 had a higher incidence of prior infections (38% versus 28%; P = 0.03) and a higher first dose of TCE therapy (P < 0.001). The best model to predict CRS grade ≥2 had a mean AUROC of 0.69 (95% confidence interval 0.66-0.72) on the test set. When patients were ranked based on their predicted probability of getting CRS grade ≥2 and divided into quartiles based on predicted CRS grade ≥2 risk (very low, low, high, very high), the very high-risk quartile developed CRS grade ≥2 at 5.9 times the rate (38.10% [interquartile range: 33.33-43.54]) compared to the very-low risk quartile (6.45% [3.44-8.82]; the sample average was 12.96% [9.25-24.07]). Compared to patients with very low CRS grade ≥2 risk, patients with very high CRS grade ≥2 risk had ALL as a disease type (99% versus 67%, P < 0.001), received a higher TCE dose (1.00 versus 0.61, P < 0.001), had a higher rate of prior infections (49% versus 12%, P < 0.001) and a higher serum creatinine (0.60 versus 0.32, P < 0.001).
CONCLUSIONS: Using the CRS grade ≥2 risk model, it was possible to stratify patients by risk categories. CRS grade ≥2 risk stratification may facilitate patient selection for TCE therapy and tailored pretreatment and monitoring of CRS to maximize treatment efficacy and safety.
OBJECTIVE: To develop a model to predict the preinfusion risk of CRS grade ≥2 for patients with solid tumors and hematologic malignancies such as acute lymphoblastic leukemia (ALL) and non-Hodgkin lymphoma (NHL) treated with TCE bispecific antibodies.
STUDY DESIGN: A TCE dataset including clinical trials from 2014 to 2019 evaluating non-CAR-T TCE therapies was sourced from the Medidata Enterprise Data Store, an anonymized data repository from completed clinical trials. The outcome of interest was the first CRS grade ≥2 occurring within 10 days of TCE therapy. Risk factors for CRS grade ≥2 were identified from the literature and preliminary data analysis. Features were measured prior to or at the first TCE treatment. Patients were included in the analysis dataset if they had a data element fill rate of >70% for the key features. Features were pruned by assessing multicollinearity across features. Logistic regression and tree-based models were trained. Across 100 iterations with different train-test splits, the average area under the receiver-operator characteristic (AUROC) curve was calculated for each model type.
RESULTS: A total of 715 patients (115 CRS grade ≥2 and 600 CRS grade <2) were included in the analysis; most patients had ALL (81%) and 19% had solid tumors or NHL. Patients who developed CRS grade ≥2 had a higher incidence of prior infections (38% versus 28%; P = 0.03) and a higher first dose of TCE therapy (P < 0.001). The best model to predict CRS grade ≥2 had a mean AUROC of 0.69 (95% confidence interval 0.66-0.72) on the test set. When patients were ranked based on their predicted probability of getting CRS grade ≥2 and divided into quartiles based on predicted CRS grade ≥2 risk (very low, low, high, very high), the very high-risk quartile developed CRS grade ≥2 at 5.9 times the rate (38.10% [interquartile range: 33.33-43.54]) compared to the very-low risk quartile (6.45% [3.44-8.82]; the sample average was 12.96% [9.25-24.07]). Compared to patients with very low CRS grade ≥2 risk, patients with very high CRS grade ≥2 risk had ALL as a disease type (99% versus 67%, P < 0.001), received a higher TCE dose (1.00 versus 0.61, P < 0.001), had a higher rate of prior infections (49% versus 12%, P < 0.001) and a higher serum creatinine (0.60 versus 0.32, P < 0.001).
CONCLUSIONS: Using the CRS grade ≥2 risk model, it was possible to stratify patients by risk categories. CRS grade ≥2 risk stratification may facilitate patient selection for TCE therapy and tailored pretreatment and monitoring of CRS to maximize treatment efficacy and safety.