Thyroid diffuse large B-cell lymphoma in China: Multicenter study of diagnosis, clinical features, treatment efficacy, prognostic factors, and genomic alterations

Abstract

Abstract Objective:​​ To investigate the clinicopathological characteristics, genomic mutational profiles, prognostic determinants, survival outcomes, and current diagnostic/therapeutic landscape of thyroid diffuse large B-cell lymphoma (TDLBCL), distinguishing primary from secondary subtypes. Methods:​​​ A multi-center retrospective analysis was conducted on 190 TDLBCL patients (123 primary, 67 secondary) treated across 22 Chinese institutions between November 2003 and November 2024. Clinicopathological data, treatment modalities, and survival outcomes were analyzed. Targeted sequencing of 55 lymphoma-associated genes was performed on 49 tumor samples to characterize mutational patterns. Results:​​​ Primary TDLBCL patients exhibited significantly higher proportions (P<0.05) of the following favorable features compared to secondary cases: ECOG performance status ≤1(P=0.005), Ann Arbor stage I–II(P< 0.001), ≤1 extranodal involvement site(P< 0.001), normal lactate dehydrogenase (LDH) (P< 0.001), combined surgical resection and chemotherapy(P< 0.001), concurrent Hashimoto's thyroiditis (HT) (P=0.005), neck mass as initial presentation(P< 0.001), localized compressive symptoms(P< 0.001), absence of B symptoms(P=0.002), thyroid dysfunction(P=0.021), and maximum tumor diameter <6 cm(P=0.018). Among 175 evaluable patients, primary TDLBCL (91.0%，101/111) achieved superior objective response rates (ORR) versus secondary disease (73.4%，47/64) (P=0.002). For 108 patients receiving first-line rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP)-based therapy, primary TDLBCL demonstrated significantly improved 5-year progression-free survival (PFS) (66.9% vs. 52.4%; P=0.018).Prognostic Factors:​​Univariate analysis of R-CHOP-treated patients identified extranodal involvement ≥2 sites(HR=2.185, 95%CI 1.011‒4.720), ECOG performance status ≥2(HR=2.378, 95%CI 1.039‒5.441), Ann Arbor stage III–IV(HR=2.262, 95%CI 1.046‒4.891), and elevated LDH (HR=2.690, 95%CI 1.178‒6.142) as significant adverse prognostic factors for progression-free survival (PFS), while maximum tumor diameter ≥6 cm (HR=5.788, 95%CI 1.436‒23.338) and double-expressor (DE) (HR=5.585, 95%CI 1.001‒31.150)（P=0.05） subtype detrimentally impacted overall survival (OS) (P<0.05); Multivariate analysis: Elevated LDH independently predicted inferior progression-free survival (HR = 2.690, 95% CI: 1.178–6.142, P = 0.019), while maximum tumor diameter emerged as an independent predictor for overall survival (model χ² = 8.587, P = 0.014), with tumors <6 cm conferring significantly reduced mortality risk (HR = 0.348, 95% CI: 0.159–0.759, P = 0.008). furthermore, genomic profiling revealed recurrent mutations in TP53 (30.6%), TET2 (30.6%), and KMT2D (26.5%), with TET2 mutational frequency differing significantly between primary and secondary TDLBCL subtypes (P=0.032), though none of these mutations demonstrated a statistically significant association with survival outcomes. Conclusion:​​ Primary TDLBCL demonstrates superior survival outcomes and enhanced response to first-line therapy compared to secondary TDLBCL. Adverse prognostic factors significantly impacting patient survival include Ann Arbor stage III–IV, elevated serum LDH, extranodal involvement ≥2 sites, maximum tumor diameter ≥6 cm, and the DE phenotype.Serum LDH levels and tumor diameter serve as ​independent predictors​ for PFS and OS, respectively, underscoring their critical role in prognostication.Genomic profiling identified TET2, TP53, and KMT2D as the most recurrently mutated genes in TDLBCL; however, none of these genetic alterations exhibited a statistically significant association with prognosis.