Background: Infections substantially contribute to the toxicity burden of CD19 CAR-T and represent the most common cause of non-relapse mortality. Determining if an episode of fever represents the onset of infection or cytokine release syndrome (CRS) remains a major clinical challenge. Most patients thus receive broad-spectrum antibiotics, which can confer resistance and may negatively influence the microbiome. Aims: We sought to characterize dynamic changes in serum inflammatory markers for infection vs. CRS and to develop clinical tools that discriminate for severe infections. Methods: Early infection events (day 0-30) and day-by-day CRS dynamics were characterized in 62 patients treated with CD19 CAR-T for relapsed/refractory B-NHL in a real-world setting (23 Axi-cel, 30 Tisa-cel, 9 KTE-X19). Severe infections were defined as requiring i.v. anti-infectives and/or hospitalization in the presence of microbiologic evidence or distinct clinical signs of infection; CRS was graded according to ASTCT criteria. Serum inflammatory markers (CRP, interleukin-6, procalcitonin) were recorded daily, and the baseline CAR-HEMATOTOX score (Rejeski et al Blood 2021) was assessed for each patient. Cumulative incidence curves were calculated as time-to-event-analysis for CRS 1°, ≥2° and severe infections. Index events were defined per patient for each toxicity category (= day of event). Inflammatory markers were tested using binary logistic regression for severe infections vs. CRS index events. Discriminatory thresholds were determined by optimizing the Youden J statistic. Results: The cumulative 30-day incidence of CRS 1°, CRS ≥2° and severe infections was 47%, 43%, and 25% respectively. Tocilizumab was applied in the majority of CRS patients (1°: 90%, ≥2°: 100%). We did not find a significant correlation between CRS grade and severe infections (G2 = 0.13, p = n.s.). Median onset of CRS and infection was 3 and 8 days, respectively. The majority of events occurred in the setting of coincident severe neutropenia (ANC <500/µl). Inflammatory markers were assessed both relative to CAR-T transfusion and the respective index event. CRS ≥2° was characterized by higher day 0-21 levels of interleukin-6 compared to CRS 0-1° (p<0.001). On the other hand, patients with severe infections exhibited both higher baseline inflammation (reflected by higher CRP levels on days 0-3), and higher peak procalcitonin levels (p<0.001) compared to CRS controls. These trends were particularly evident on the day of the index event (Fig. 1). Procalcitonin was significantly elevated for severe infections vs. CRS controls (median 1.6 vs. 0.3 µg/l, p<0.001), and an optimal discriminatory threshold of 1.4 µg/l was established. An elevated baseline CAR-HEMATOTOX was noted for patients with severe infections (median 5 vs. 2, p=0.002). By incorporating index-event procalcitonin levels with the score, the discriminatory capacity for severe infection events was improved. On ROC analysis, the ‘HT10 score’ (addition of 10 points to the patient-specific CAR-HEMATOTOX in case procalcitonin ≥1.4 µg/l on event day) exhibited superior discrimination for severe infections (AUC 0.92, p<0.0001, sens. 80%, spec. 89%). Image:Summary/Conclusion: CRS and infections are characterized by distinct patterns of inflammation after CD19 CAR-T (CRS: high IL-6; Infections: high procalcitonin). The high discriminatory capacity of the HT10 score for infections emphasizes the utility of procalcitonin in routine inflammatory panels. If validated prospectively, the score will enable more patient-individual decisions on antibiotic use.