Attenuated Levels of Surface TREM1 on Circulating Neutrophils Predict Poor Survival in Patients with Diffuse Large B-Cell Lymphoma before Treatment

Introduction:

Neutrophils are the most common type of white blood cells in the human body, yet their role in lymphoma patients has been rarely studied. The triggering receptor expressed on myeloid cells-1 (TREM1) is known for amplifying the inflammatory response and plays a crucial role in regulating neutrophil functions. Immunotherapies have been introduced for treating patients with diffuse large B-cell lymphoma (DLBCL), but more than half of these patients experience recurrence even after using CAR-T or BiTE therapies. Due to the abundance and omnipresence of neutrophils in human blood, we are curious to know if neutrophils could be a novel target for immune modulation. Therefore, this study aims to investigate the prognostic significance of surface-TREM1 levels on circulating neutrophils in DLBCL patients before initiating any treatment.

Methods:

This prospective observational study was conducted at Taipei Veterans General Hospital. A total of 113 untreated DLBCL patients were enrolled from June 1, 2020, to October 31, 2023, with the last follow-up occurring on June 30, 2024. Patients with prominent evidence of infection at enrollment were excluded, as infection may induce TREM1 expression in neutrophils. Primary CNS lymphoma and primary mediastinal lymphoma were also excluded due to differing induction therapies.

Circulating neutrophils were isolated from patients' fresh peripheral blood. By flow cytometry, neutrophils were defined as CD45+ CD15+ viable singlets. To increase comparability between different experiments, neutrophils from healthy donors and DLBCL patients were paired during the same flow cytometric analysis to calculate the normalized levels of surface TREM1 on neutrophils as follows: (patients' TREM1 MFI - healthy donor's TREM1 MFI) / (healthy donor's TREM1 MFI) x 100%.

Results:

For the 113 DLBCL patients, the median age was 66 years, 61.9% were male, 40.7% were stage IV, 31.8% were high-risk IPI, 23.9% had bone marrow (BM) involvement, 27.4% had a bulky mass > 10 cm, 72.6% were non-GCB subtype per Hans algorithm by IHC stain, 41.6% were double expressor lymphoma (DEL), and 8.8% were Epstein-Barr virus (EBV) positive. Notably, nearly all patients had lower surface-TREM1 levels on neutrophils compared to healthy donors, with a median normalized level of surface TREM1 at -39.7%.

We further discovered that patients with high-risk IPI had significantly lower normalized TREM1 levels than those with low-risk IPI (-45.6% vs. -27.8%, P = 0.003). Patients with non-GCB subtypes also showed significantly lower normalized TREM1 levels compared to those with GCB subtypes (-41.3% vs. -29.8%, P = 0.028). Trends of lower normalized TREM1 levels were observed in patients with BM involvement (P = 0.072) and DEL (P = 0.104).

These 113 DLBCL patients were subgrouped into high (greater than -29%), intermediate (between -46% and -29%), or low (less than -46%) normalized levels of surface TREM1. The cutoff values for these subgroups were determined using the ROC curve and AUC analysis. After a median follow-up of 24.5 months, patients with low normalized TREM1 levels had the worst PFS compared to those with intermediate or high normalized TREM1 levels (log-rank P = 0.001; low vs. high: HR = 5.243, P = 0.001; intermediate vs. high: HR = 3.333, P = 0.025). Patients with low normalized TREM1 levels also demonstrated the poorest OS compared to those with intermediate or high normalized TREM1 levels (log-rank P = 0.002; low vs. high: HR = 9.856, P = 0.003; intermediate vs. high: HR = 5.469, P = 0.038).

To determine whether normalized TREM1 levels on neutrophils could predict DLBCL patients' survival before initiating treatment, nine factors were analyzed in a multivariate Cox regression model: age (>66), sex (male), high-risk IPI, BM involvement, bulky mass > 10 cm, EBV positivity, non-GCB subtype, DEL, and normalized TREM1 levels on neutrophils. Low and intermediate normalized TREM1 levels were proven to be independent poor prognostic factors for both PFS (low TREM1: HR = 4.766, P = 0.002; intermediate TREM1: HR = 3.430, P = 0.025) and OS (low TREM1: HR = 7.949, P = 0.008; intermediate TREM1: HR = 6.916, P = 0.021).

Conclusion:

A low surface-TREM1 level on circulating neutrophils is a poor prognostic factor for both PFS and OS in DLBCL patients before initiating treatment. This finding warrants further investigation into how TREM1 influences the role of neutrophils in the lymphoma microenvironment.

No relevant conflicts of interest to declare.