Background. Diffuse Large B Cell Lymphoma (DLBCL) is the most common B-cell malignancy (1), recently classified with genomic sequencing and exome analysis into clusters with clinical implications (2,3). However, this analysis has not been informed neither in Mexican, or Latin-American population. The aim of this study was to describe the mutational landscape and its correlation with the clinical outcome in Mexican patients with DLBCL. Material and methods. A prospective, observational study of patients with diagnosis of DLBCL, attended at the National Cancer Institute in Mexico City. Inclusion criteria: patients > 18 years, diagnosis of DLBCL, without previous treatment and candidate to be treated with RCHOP. IRB approval was: Register number CEI/966/15. All signed informed consent. Clinical & pathological variables were analyzed. All patients were treated with 6 cycles of RCHOP. Clinical response was evaluated by PET-CT. Samples: DNA was extracted from paraffin-embedded tissue of the primary neoplasm using the kit Quick-DNA/RNA FFPE kit (Zymo Research). A custom panel (6,026 oligos) was used for the sequencing of coding regions and splicing sites of 79 genes associated with lymphomas. Therafter, the bioinformatic analysiswas carried out using Ion Torrent Suite Browser version 5.0 and Ion Reporter version 5.18. Variant calling was performed with Torrent Variant Caller. The following databases were used: Cancer Genome Interprteter (CGI), ClinVar (201706), COSMIC (v86), PolyPhen (v2.2.2), SIFT (v5.2.2), FATHMM (v2.1), gnomAD (r2.0.1), and the Variant Effect Predictor (v93.2). All variants had a depth greater than 500X. Statistical analysis. Descriptive statistics for clinical variables. Associations between clinical and pathological characteristics were assessed with the frequency of SNVs and InDels. Survival curves were estimated using the Kaplan-Meier method. The association of the overall survival (OS) and the recurrence-free survival (RFS) was evaluated with Log-Rank test. Statistical significance was set at p-value < 0.05. Cox proportional-hazard regression analysis was used to identify the variables predicting overall survival (OS). All statistical analyses were performed using R software (v4.2.2) (https://cran.r-project.org, accessed on 23rd May 2023) and SPSS version 25 software (IBM, Corp., Armonk, NY). Results We included 185 patients, median + SD age 59.3 + 14.1 years. Most (n=128, 69.2 %) were in ECOG score < 2, without B symptoms (n=131, 70.8 %), in advanced disease (n=134, 72.4%), and were germinal-center (GC) (n=126, 68.1 %), and 26 (14.1 %) were double hit lymphomas. At least a driver mutation was found in 110 cases (59.4 %). The genes with driver mutations most frequently were: TP53, EZH2, CREBBP, NOTCH1 and KMT2D. ( figure 1). The EZH2 mutations and KMT2D mutations were more frequent in female: ( EZH2 mutated [n=22/34, 64.7%], p=0.047]) ( KMT2D mutated, 75%, p=0.034), and had a higher frequency of bulky disease (63.3 %, p= 0.05). In contrast, MYD88 mutations were more frequent in male (MYD88 mutations: 76.9 %], p= 0.042). Response to R-CHOP treatment was: Complete (n= 129 (69.7%), partial (n=10, 5.4 %), or stable disease (n=3, 1.6 %); 26 patients (14 %) progressed during treatment. Median follow-up + SD was 41.99 + 21.4 months. The RFS was 60 % to 80 months and OS was 70 % to 84 months. ECOG >2 (p=0.002 [95 %CI: 1.52-3.97]), mutations at genes EZH2 (p=0.007 [95% CI: 0.193 -0.0.58), CREBBP (p=0.004 [95 % CI: 1.59 - 11.85]), and ARID1A (p= 0.027 [95 % CI: 1.34 - 131.45] were factors influencing on OS by univariate analysis. The multivariate analysis confirmed that an ECOG >2 (p=0.001 [95 CI: 1.46-3.34]) had a negative impact on OS. In contrast, EZH2 mutations (p=0.025 [95 % CI: 0.12 -0.86]) were a protective factor. Conclusions: This is the first approach to describe the genomic landscape in Latino-american population. The presence of EZH2 and KMT2 mutations were more frequent in female patients with mediastinal and bulky disease. After multivariate analysis, only the presence of EZH2 mutations was the only genomic factor influencing on OS. These results require to be confirmed by further studies.
Read full abstract