-Author name in bold denotes the presenting author
-Asterisk * with author name denotes a Non-ASH member
Clinically Relevant Abstract denotes an abstract that is clinically relevant.

PhD Trainee denotes that this is a recommended PHD Trainee Session.

Ticketed Session denotes that this is a ticketed session.

2143 Machine Learning-Based Time-Series Clustering Identifies Archetypal Trajectories of Hematotoxicity after CAR T-Cell Therapy

Program: Oral and Poster Abstracts
Session: 705. Cellular Immunotherapies: Late Phase and Commercially Available Therapies: Poster I
Saturday, December 9, 2023, 5:30 PM-7:30 PM

Emily C Liang, MD1,2, Aya Albittar, MD1,2*, Andrew J Portuguese, MD1,2, Jennifer J. Huang, MD, PhD1,2, Qian Wu2*, Joseph De Los Reyes, BSA2*, Nikki Pin2*, Aiko Torkelson, BS2*, Delaney Kirchmeier, BS2*, Abigail Chutnik, BS2*, Barbara Pender, MSc2*, Joshua A. Hill, MD1,2*, Rahul Banerjee, MD1,2, Andrew J. Cowan, MD1,2, Damian J Green, MD1,3, Ajay K Gopal, MD2,4, Christina Poh, MD2,4, Mazyar Shadman, MD, MPH2,5, Alexandre V Hirayama, MD2,5*, Brian G Till2,6, Erik L Kimble, MD1,2, Lorenzo Iovino, MD, PhD2,7, Aude G Chapuis, MD2,6, Folashade Otegbeye, MBChB, MPH1,2, Ryan D Cassaday, MD1,2, Filippo Milano, MD3,6, David G Maloney, MD, PhD2,5 and Jordan Gauthier, MD, MSc5,8

1University of Washington, Seattle, WA
2Fred Hutchinson Cancer Center, Seattle, WA
3Translational Science and Therapeutics Division, Fred Hutchinson Cancer Center, Seattle, WA
4Clinical Research Division, Fred Hutch Cancer Center; and Division of Oncology, University of Washington, Seattle, WA
5Division of Medical Oncology, University of Washington, Seattle, WA
6Division of Hematology and Oncology, University of Washington, Seattle, WA
7University of Washington School of Medicine, Seattle, WA
8Clinical Research Division, Fred Hutchinson Cancer Center, Seattle, WA


Severe hematotoxicity has been reported in a subset of patients (pts) after chimeric antigen receptor (CAR) T-cell therapy, leading to severe infections, transfusion dependency, and worse outcomes. Early identification of pts at risk for severe hematotoxicity could inform the consideration of an autologous stem cell boost, allogeneic HCT, and/or additional infectious prophylaxis. While distinct archetypal trajectories of blood count recovery have been proposed (e.g., quick vs. intermittent vs. aplastic), manual trajectory classification is subjective and labor-intensive. To address these limitations, we automated the identification of distinct trajectories of hematotoxicity by applying time-series clustering to longitudinal absolute neutrophil count (ANC) data from >400 pts treated with CAR T-cell therapy. Next, we sought to identify factors associated with the identified trajectories and to assess the predictive ability of the CAR-HEMATOTOX score (Rejeski, Blood 2021).


Adults ≥18 years of age who received their first infusion of CAR T cells for hematologic malignancies with commercial or investigational products at the Fred Hutchinson Cancer Center (2013-2023) were included. ANC trajectories were clustered using non-supervised longitudinal k-means based on Euclidean distances using the latrend and kml packages in R 4.1.3. We applied logistic regression to pre-lymphodepletion (pre-LD) variables to predict poor/very poor ANC recovery. Sensitivity and specificity were computed using a probability threshold based on the Youden criteria.


A total of 509 pts were identified; 106 were excluded due to insufficient data, with 403 pts included for the analysis. The most common disease types were aggressive NHL (n = 161; 40%), indolent NHL (n = 82; 20%), and ALL (n = 74; 18%). CAR T-cell products were axi-cel, n= 101 (25%); brexu-cel, n = 24 (6%); cilta-cel, n = 21 (5%); liso-cel, n = 46 (11%); ide-cel, n = 25 (6%); tisa-cel, 12 (3%); and investigational CD19 or CD20 CAR T-cell products, n = 174 (43%).

As shown in Figure 1A, the ANC longitudinal data clustered into four distinct trajectories as follows: 1) high nadir followed by rapid recovery (“very good recovery”), n = 294 (73%); 2) low nadir followed by rapid recovery (“good recovery”), n = 87 (22%); 3) low nadir followed by intermittent recovery (“poor recovery”), n = 13 (3%); 4) aplastic phenotype (“very poor recovery”), n = 9 (2%). In univariate logistic regression, ALL (reference: aggressive NHL; OR = 5.43, 95% CI, 1.71-20.6, p = 0.006), pre-LD lower ANC (OR = 3.33 per log10 , 95% CI, 1.82-6.25, p < 0.001), lower pre-LD platelet count (OR = 10 per log10, 95% CI, 3.70-33.3), p < 0.001), and higher pre-LD disease burden as measured by lactate dehydrogenase (LDH) were associated with poor/very poor ANC recovery (OR = 16.6 per log10, 95% CI, 4.79-59.3, p < 0.001).

Next, we assessed the predictive ability of CAR-HEMATOTOX to identify poor/very poor ANC recovery in n = 356 pts with a calculable score. The CAR-HEMATOTOX scorehigh/low showed modest discrimination (C-index: 0.65), which improved when used as a continuous variable (C-index: 0.85). As previously reported, these models had high sensitivity (>90%) but low specificity (high/low: 31%, continuous: 58%). To improve risk prediction of poor/very poor ANC recovery, we built a logistic regression model using restricted cubic splines – allowing for non-linear effects – including pre-LD ANC, platelet, Hb, LDH, CRP, and ferritin, which showed higher discrimination (C-index: 0.91), high sensitivity (88%), and higher specificity (79%) (Figure 1B).


We introduce an automated, scalable, and disease-agnostic framework to analyze hematologic toxicity after CAR T-cell therapy. K-means clustering of longitudinal ANC data categorized pts into archetypal trajectories of hematologic recovery. A logistic regression model using pre-LD ANC, platelet, Hb, LDH, CRP, and ferritin showed improved discrimination and sensitivity compared to CAR-HEMATOTOX in our training set, but both models had low specificity (poor ability to “rule in” pts at risk of severe hematotoxicity) and thus low clinical utility at this time. We plan to further improve our predictive models by incorporating post-infusion biomarkers of systemic inflammation that we have previously shown to be associated with delayed count recovery after CAR T-cell therapy (Juluri, Blood Advances 2022).

Disclosures: Hill: allovir: Consultancy, Research Funding; moderna: Consultancy; deverra: Research Funding. Banerjee: BMS: Consultancy; Janssen: Consultancy; Genentech: Consultancy; SparkCures: Consultancy; Sanofi: Consultancy; Caribou: Consultancy; Pfizer: Consultancy; Pack Health: Research Funding. Cowan: BMS, Adaptive: Consultancy; Adaptive Biotechnologies, Harpoon, Nektar, BMS, Janssen, Sanofi, Abbvie: Research Funding. Green: Cellectar Biosciences: Research Funding; SpringWorks Therapeutics: Research Funding; Celgene: Consultancy; GlaxoSmithKline: Membership on an entity's Board of Directors or advisory committees; Sanofi: Research Funding; Janssen Biotech: Consultancy, Research Funding; Ensoma: Consultancy; Seattle Genetics: Consultancy, Research Funding; Juno Therapeutics A BMS Company: Patents & Royalties, Research Funding. Gopal: Compliment Corporation: Current holder of stock options in a privately-held company; Incyte, Kite, Morphosys/Incyte, ADCT, Acrotech, Merck, Karyopharm, Servier, Beigene, Cellectar, Janssen, SeaGen, Epizyme, I-Mab bio, Gilead, Genentech, Lilly, Caribou, Fresenius-Kabi: Consultancy; Merck, I-Mab bio, IgM Bio, Takeda, Gilead, Astra-Zeneca, Agios, Janssen, BMS, SeaGen, Teva, Genmab: Research Funding. Poh: BeiGene: Consultancy; Seattle Genetics: Consultancy; Incyte: Research Funding; Acrotech: Consultancy. Shadman: Fate Therapeutics: Consultancy; Genmab: Consultancy, Research Funding; Vincerx: Research Funding; MorphoSys/Incyte: Consultancy, Research Funding; Eli Lilly: Consultancy; Bristol Myers Squibb: Consultancy, Research Funding; Genentech: Consultancy, Research Funding; Kite, a Gilead Company: Consultancy; AbbVie: Consultancy, Research Funding; Mustang Bio: Consultancy, Research Funding; ADC therapeutics: Consultancy; BeiGene: Consultancy, Research Funding; AstraZeneca: Consultancy, Research Funding; Pharmacyclics: Consultancy, Research Funding; Janssen: Consultancy; MEI Pharma: Consultancy; Regeneron: Consultancy; TG Therapeutics: Research Funding. Hirayama: Novartis: Honoraria; Bristol Myers Squibb: Honoraria, Research Funding; Nektar Therapeutics: Honoraria, Research Funding; Juno Therapeutics, a Bristol Myers Squibb Company: Research Funding. Till: Mustang Bio: Consultancy, Patents & Royalties, Research Funding; BMS/Juno Therapeutics: Research Funding; Proteios Technology: Consultancy, Current holder of stock options in a privately-held company. Kimble: Juno/BMS: Research Funding. Iovino: Mustang Bio: Current equity holder in publicly-traded company. Chapuis: Juno Therapeutics: Research Funding. Cassaday: Amgen: Consultancy, Honoraria, Research Funding; Merck: Research Funding; Incyte: Research Funding; Autolus: Membership on an entity's Board of Directors or advisory committees; Pfizer: Consultancy, Honoraria, Research Funding; Servier: Research Funding; Vanda Pharmaceuticals: Research Funding; Jazz: Consultancy, Honoraria; Kite/Gilead: Consultancy, Honoraria, Research Funding; PeproMene Bio: Membership on an entity's Board of Directors or advisory committees; Seagen: Other: Spouse was employed by and owned stock in Seagen within the last 24 months.. Milano: ExCellThera Inc.: Research Funding. Maloney: A2 Biotherapeutics: Consultancy, Current holder of stock options in a privately-held company, Honoraria, Other: Member of the Scientific Advisory Board; Juno Therapeutics: Consultancy, Honoraria, Patents & Royalties: Rights to royalties from Fred Hutch for patents licensed to Juno Therapeutics/BMS, Research Funding; Janssen: Consultancy, Honoraria; Legend Biotech: Consultancy, Honoraria, Research Funding; Mustang Bio: Consultancy, Honoraria; Novartis: Consultancy, Honoraria; Lyell Immunopharma: Other: Member, CAR T Steering Committee; Incyte: Consultancy, Honoraria; Gilead Sciences: Consultancy, Honoraria, Other: Member, Scientific Review Committee, Research Scholars Program in Hematologic Malignancies; Kite, a Gilead Sciences: Consultancy, Honoraria, Research Funding; Pharmacyclics: Consultancy, Honoraria; Umoja: Consultancy, Honoraria; Celgene: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: Participation on a Data Safety Monitory Board , Research Funding; Genentech: Consultancy, Honoraria, Other: Chair and Member of the Lymphoma Steering Committee; MorphoSys: Consultancy, Honoraria; Bristol Myers Squibb: Consultancy, Honoraria, Other: Member of the JCAR017 EAP-001 Safety Review Committee and Member, CLL Strategic Council, Member of the JCAR017-BCM-03 Scientific Steering Committee under BMS, Research Funding; Amgen: Consultancy, Honoraria; Navan Technologies: Consultancy, Honoraria, Other: Member of the Scientific Advisory Board; Bioline Rx: Consultancy, Membership on an entity's Board of Directors or advisory committees, Other: Participation on a Data Safety Monitory Board ; Fred Hutch: Other: rights to royalties for patents licensed to Juno; Navan Technologies: Current holder of stock options in a privately-held company; Chimeric Therapeutics: Other: Member of the Scientific Advisory Board; ImmPACT Bio: Other: Member, Clinical Advisory Board, CD19/CD20 bi-specific CAR-T Cell Therapy Program; Interius: Other: Member, Clinical Advisory Board. Gauthier: Kite Pharma: Consultancy, Honoraria; MorphoSys: Consultancy, Research Funding; Angiocrine Bioscience: Research Funding; Century Therapeutics: Other: Independent data review committee; Celgene (a Bristol Myers Squibb company): Research Funding; Legend Biotech: Consultancy, Honoraria; Janssen: Consultancy, Honoraria; Juno Therapeutics (a Bristol Myers Squibb company): Research Funding; Sobi: Consultancy, Honoraria, Research Funding.

<< Previous Abstract | Next Abstract
*signifies non-member of ASH