Background：Follicular lymphoma (FL) is a prototypical B-cell lymphoma originating from the follicular germinal center, representing the most common type among indolent lymphomas. Pathologists commonly classify FL into five grades based on the ratio of centrocytes to centroblasts observed under high magnification: FL1, FL2, FL3A, FL3B, and follicular lymphoma mixed diffuse large B-cell lymphoma (FL/DLBCL). Extensive studies on clinical characteristics and prognosis in FL grades 1-3A have significantly advanced our understanding of the disease. However, comprehensive data on clinical characteristics and genetic profiles across all histologic subtypes remains limited.

Methods：Constructing clinical profiles of follicular lymphoma patients with different pathological grades diagnosed at Tianjin Medical University Cancer Institute and Hospital from 2002 to 2020. Generating genetic molecular maps of follicular lymphoma across various pathological grades using whole exome sequencing and transcriptome sequencing.

Results：A total of 926 patients were initially identified. After excluding those under 18 years of age, those with histological transformation, and those with incomplete clinical information, 831 patients remained. Among them, 588 (71%) had FL1-2, 84 (10%) had FL3A, 67 (8%) had FL3B, and 92 (11%) had FL/DLBCL. The Kaplan-Meier survival analysis was performed on 477 FL patients who received R-CHOP treatment. The median PFS and median OS were not reached for patients with FL1-2 , FL3A and FL/DLBCL. The median OS for FL3B is 88 months (95% CI: 62-NA), and the median PFS is 38 months (95% CI: 28-NA). Univariate Cox regression analysis showed that factors such as age over 60, spleen involvement, elevated serum LDH, elevated β2-MG, and grade 3B disease were associated with poorer progression-free survival (PFS) and overall survival (OS). Additionally, an ECOG performance status of 3-4 and Ann Arbor stage III/IV were linked to inferior PFS. In multivariable analysis for PFS, significant factors included grade 3B disease (HR 2.08, 95% CI 1.22-3.56, p = 0.0076), ECOG 3-4 (HR 2.49, 95% CI 0.98-6.32, p = 0.05), and elevated serum LDH (HR 1.68, 95% CI 1.05-2.69, p = 0.03). For OS, the only factor retaining significance was grade 3B disease (HR 4.67, 95% CI 1.95-11.15, p < 0.001). The mutation profiles of the four groups were broadly similar, predominantly featuring epigenetic mutations such as KMT2D, CREBBP, and EP300. There were similarities and differences in the copy number profiles among the four groups. All four groups showed high deletion frequencies in 14q32.33 (FL1-2: 70%, FL3A: 73%, FL3B: 100%, FL/DLBCL: 78%), 19p12 (FL1-2: 58%, FL3A: 68%, FL3B: 80%, FL/DLBCL: 78%), and 4q13.2 (FL1-2: 52%, FL3A: 64%, FL3B: 50%, FL/DLBCL: 50%). In FL3B and FL/DLBCL patients, the 7q22.3 region, which includes the CDK6 gene, was frequently amplified (FL1-2: 8%, FL3A: 14%, FL3B: 70%, FL/DLBCL: 56%; p < 0.0001), possibly indicating potential activation of the cell cycle pathway. FL3B and FL/DLBCL also frequently exhibited deletions in 1p36.32 (FL1-2: 21%, FL3A: 14%, FL3B: 50%, FL/DLBCL: 33%; p = 0.03). In multivariable analysis, when assessed using FLIPI2 (p = 0.018), the deletion of 1p36.32 was a significant independent predictor of PFS (HR 1.85, 95% CI 1.05 to 3.23, p = 0.032). Comparative transcriptomic analysis between the groups revealed that numerous immune pathways were suppressed in FL3A, while many metabolism-related pathways and cell cycle pathways were upregulated in FL3B and FL/DLBCL.

Conclusions: In summary, our study identified distinct clinical, genetic, and transcriptomic characteristics among FL1-2, FL3A, FL3B, and FL/DLBCL patient groups. Specific genetic alterations are significantly associated with poorer survival outcomes and disrupt the tumor microenvironment, particularly in FL3B and FL/DLBCL patients. These findings highlight the importance of tailored therapeutic strategies based on detailed molecular and clinical profiling to improve patient outcomes.

Keywords: Follicular lymphoma, Histological grading, Clinical feature, Next-generation sequencing, Genetic alterations, Copy number variation, Gene expression.

Disclosures

No relevant conflicts of interest to declare.

This content is only available as a PDF.
2024
Sign in via your Institution