Clinical impact of untargeted baseline plasma metabolomics in newly diagnosed diffuse large B-cell lymphoma

Abstract

Abstract Introduction . Metabolomics is emerging as a promising tool to identify novel prognostic biomarkers in hematological malignancies, by capturing not only tumor-related but also host-related clues. However, its prognostic impact in diffuse large B-cell lymphoma (DLBCL) is largely unknown. The aim of this study was to investigate the potential clinical impact and biological correlations of baseline metabolomics in newly diagnosed DLBCL. Methods. The study relies on a real-world discovery cohort of 96 R-CHOP-treated DLBCL provided with baseline plasma samples. After metabolite extraction, samples were analyzed with a high-resolution two-dimensional gas chromatography/mass spectrometry, and metabolites were expressed as abundance values. Circulating tumor DNA (ctDNA) levels were measured with CAPP-Seq, and molecular clustering was performed by the LymphGen tool. For 87 patients baseline PET/CT scans for total metabolic tumor volume (tMTV) assessment were available. A validation cohort of 42 R-CHOP-treated DLBCL, in which metabolites were expressed as absolute concentration values, was also analyzed. The maximally selected rank statistics were used to identify the best cutoffs in predicting progression-free survival (PFS) for each metabolite, ctDNA levels and tMTV. Results. The median age of the patients enrolled in the discovery cohort was 68.7 years. After a median follow-up of 55.6 months, the 40-month PFS and overall survival (OS) were 61.4% and 72.8%, respectively. The validation cohort presented comparable PFS (p=0.23). Untargeted metabolomic analysis led to the identification of 281 metabolites, 40 of them uniquely associated with DLBCL molecular clusters. Based on PFS at 24 months from diagnosis, patients were categorized as progressive (N=34) and non-progressive (N=62) cases. Ten metabolites were differentially up- or downregulated in progressive cases, and two of them also retained prognostic impact in the validation cohort. When compared to non-progressive cases, progressive patients displayed significantly higher abundance (discovery cohort) and concentration values (validation cohort) of 4-Hydroxybenzeneacetic acid (4-HPA) (p=0.027 and p=0.02, respectively). Patients with higher 4-HPA values (N=12 in the discovery cohort and N=23 in the validation cohort) displayed a 40-month PFS of 41.7% and 34.8% compared to 64.6% and 78.6% of patients with lower values (N=84 and N=19) (p=0.0067 and p=0.0022, respectively). Similarly, progressive patients displayed significantly higher abundance (discovery cohort) and concentration values (validation cohort) of 2-Hydroxybutyric acid (2-HB) compared to non-progressive cases (p=0.02 and p=0.003, respectively). Patients with higher 2-HB values (N=20 in the discovery cohort and N=25 in the validation cohort) displayed a 40-month PFS of 39.4% and 42.9% compared to 67.3% and 70.1% of patients with lower values (N=76 and N=17) (p=0.0047 and p=0.026, respectively). The potential prognostic impact of 4-HPA and 2-HB levels within risk categories defined by validated prognostic markers, namely ctDNA levels and tMTV, was also evaluated. In the discovery cohort, as expected, ctDNA-high (N=42) and tMTV-high (N=29) patients had significantly shorter PFS when compared to ctDNA-low (N=54) and tMTV-low patients (N=58), respectively (p=0.0062 and p<0.0001, respectively). Neither 4-HPA nor 2-HB abundance values further stratified outcomes in ctDNA-high and tMTV-high patients. Conversely, among ctDNA-low patients, cases with higher abundance values of 4-HPA (N=5) and 2-HB (N=7) had significantly shorter PFS when compared to patients with lower values of 4-HPA (N=49) and 2-HB (N=47) (p<0.0001 and p=0.041, respectively). Similarly, among tMTV-low patients, cases with higher abundance values of 4-HPA (N=6) and 2-HB (N=7) displayed worse outcomes than patients with lower values of 4-HPA (N=52) and 2-HB (N=51) (p=0.00076 and p=0.052, respectively). Conclusions.Baseline plasma metabolites harbor a promising prognostic value in newly diagnosed R-CHOP-treated DLBCL. Baseline 4-HPA and 2-HB levels are associated with shorter PFS and further refine the prognosis of a subset of patients who were incorrectly classified as low-risk based on ctDNA levels and tMTV. 4-HPA may reflect the potential contribution of microbiome changes to disease course and response to therapy, while 2-HB may capture the prognostic relevance of oxidative stress and metabolic dysregulation in DLBCL.