MYD88 mutations predict unfavorable prognosis in Chronic Lymphocytic Leukemia patients with mutated IGHV gene

Chronic lymphocytic leukemia (CLL) is the most common leukemia in adults in the Western countries but is relatively rare in East Asia. CLL is a disease of high heterogeneity. The clinical course ranges from indolence to rapid progression to death. Although the Rai and Binet clinical staging systems remain to be the cornerstone for CLL prognosis, the rapidly developed biological and genetic techniques enable the detection of novel prognostic factors. Mutations in myeloid differentiation primary response gene 88 (MYD88) in CLL were first reported in 2011 with a mutation frequency of 9/310 (2.9%). Subsequent studies found that MYD88 mutations exist in 2.0–4.4% Caucasian patients with CLL. However, subjects of Asia showed a higher MYD88 mutated rate of 8% as previously reported. The above MYD88 mutated cases consist mainly of a p. L265P substitution. CLL patients with MYD88 mutations were reported to be younger at diagnosis and have longer time to treatment (TTT) and overall survival (OS) than those with wild-type MYD88. However, this conclusion was controversial. Initial studies indicated that most MYD88-mutated patients belonged to the IGHV-mutated group, which is generally accepted as a molecular sign of favorable prognosis. These studies could be more convincing if taking IGHV mutation status and MYD88 mutations together into prognostic consideration. In the current study, we analyzed MYD88 mutations exclusively in the IGHV-mutated CLL cases to explore its prognostic value. Two hundred and eighty-four patients with previously untreated CLL at the First Affiliated Hospital of Nanjing Medical University between January 2000 and June 2016 were retrospectively enrolled. All cases were reviewed to confirm the diagnosis according to the 2008 International Workshop in CLL-National Cancer Institute (IWCLLNCI). Clinical and biological parameters including absolute lymphocyte count, hemoglobin, platelet, cytogenetic abnormalities, mutation status of TP53, IGHV, NOTCH1 as well as surface markers of CLL cells were assessed at first presence at our center. The study was approved by the Ethics Committee of the First Affiliated Hospital of Nanjing Medical University with a reference number as 2014-SR-204. Informed consents were provided according to the Declaration of Helsinki. Mononuclear cells from 281 peripheral blood samples and three bone marrow samples of untreated CLL patients were used for AS-PCR assay. Genomic DNA was extracted using the QIAamp DNA Blood Kits (Qiagen, Düsseldorf, Germany) according to the manufacturer’s recommendation. Two different forward primers (FW5′-GTGCCCATCAGAAGCGCCT-3′ and FM5′GTGCCCATCAGAAGCGCCC-3′) and one reverse primer (5′-AGGAGGCAGGGCAGAAGTA-3′) were used to amplify the wild-type allele or the MYD88 L265P mutation allele as previously reported. The sensitivity of ASPCR was 0.625% in the present study. The Sanger sequencing was performed to confirm the AS-PCR assay and to detect MYD88 mutations other than L265P. Exon 3–5 was amplified by Sanger sequencing with a forward primer (5′AGCGACATCCAGTTTGTGC-3′) and a


and Chun Qiao 1
Chronic lymphocytic leukemia (CLL) is the most common leukemia in adults in the Western countries but is relatively rare in East Asia 1 . CLL is a disease of high heterogeneity. The clinical course ranges from indolence to rapid progression to death. Although the Rai and Binet clinical staging systems remain to be the cornerstone for CLL prognosis, the rapidly developed biological and genetic techniques enable the detection of novel prognostic factors.
CLL patients with MYD88 mutations were reported to be younger at diagnosis and have longer time to treatment (TTT) and overall survival (OS) than those with wild-type MYD88 9 . However, this conclusion was controversial 10 . Initial studies indicated that most MYD88-mutated patients belonged to the IGHV-mutated group 5,9,11 , which is generally accepted as a molecular sign of favorable prognosis. These studies could be more convincing if taking IGHV mutation status and MYD88 mutations together into prognostic consideration 10 . In the current study, we analyzed MYD88 mutations exclusively in the IGHV-mutated CLL cases to explore its prognostic value.
Two hundred and eighty-four patients with previously untreated CLL at the First Affiliated Hospital of Nanjing Medical University between January 2000 and June 2016 were retrospectively enrolled. All cases were reviewed to confirm the diagnosis according to the 2008 International Workshop in CLL-National Cancer Institute (IWCLL-NCI) 12 . Clinical and biological parameters including absolute lymphocyte count, hemoglobin, platelet, cytogenetic abnormalities, mutation status of TP53, IGHV, NOTCH1 as well as surface markers of CLL cells were assessed at first presence at our center. The study was approved by the Ethics Committee of the First Affiliated Hospital of Nanjing Medical University with a reference number as 2014-SR-204. Informed consents were provided according to the Declaration of Helsinki.
Mononuclear cells from 281 peripheral blood samples and three bone marrow samples of untreated CLL patients were used for AS-PCR assay. Genomic DNA was extracted using the QIAamp DNA Blood Kits (Qiagen, Düsseldorf, Germany) according to the manufacturer's recommendation. Two different forward primers (FW5′-GTGCCCATCAGAAGCGCCT-3′ and FM5′-GTGCCCATCAGAAGCGCCC-3′) and one reverse primer (5′-AGGAGGCAGGGCAGAAGTA-3′) were used to amplify the wild-type allele or the MYD88 L265P mutation allele as previously reported 4 . The sensitivity of AS-PCR was 0.625% in the present study. The Sanger sequencing was performed to confirm the AS-PCR assay and to detect MYD88 mutations other than L265P. Exon 3-5 was amplified by Sanger sequencing with a forward primer (5′-AGCGACATCCAGTTTGTGC-3′) and a reverse primer (5′-AGGCGAGTCCAGAACCAAG -3′) 8 . Amplified fragments were sequenced with both the forward and reverse primers. Both detecting methods were applied on all samples included in the study.
All statistical analyses were performed by SPSS for Windows (version 19.0; IBM Corporation, Armonk, NY, USA) and Graphpad Prism 6. Fisher's exact test and the chi-square test were used to determine the correlations between MYD88 mutations and clinical characteristics. Mann-Whitney U test was applied for comparing mean fluorescence intensity (MFI) as a continuous parameter in MYD88 mutated and wild-type groups. Time to treatment (TTT) was defined as the time from initial diagnosis to first treatment. OS was defined as the time from diagnosis to death or to the last follow-up. TTT and OS curves were estimated by the Kaplan-Meier method and compared by the log-rank test. The prognostic impact of MYD88 mutations on TTT and OS was assessed using both univariate and multivariate Cox analysis. All statistical tests were two-sided, and P value < 0.05 was considered to be significant.
We further analyzed the correlation between MYD88 mutations and 6 mostly used IGHV genes in M-CLL patients. None of the MYD88 mutated cases used IGHV4-34, the most prevalent IGHV gene in the M-CLL cohort, (P = 0.015) ( Table 3), suggesting that MYD88 mutation might be IGHV gene-biased, and that certain antigen exposure might avoid the emergence of MYD88 mutations in the pathogenesis of CLL.
In this study, we explored the detection method and clinical relevance of MYD88 mutations in Chinese patients with CLL. We found MYD88 mutations: (1) occur in 8.8% CLL patients in our center upon diagnosis; (2) cluster with cases harboring mutated IGHV; (3) identify a group of patients with poor prognosis among M-CLL; (4) are rare, if not absent, in IGHV-4-34 users. The incidence of MYD88 mutations was 2.0-4.4% in Caucasian CLL patients [3][4][5] . However, we have detected a higher frequency of 8.8% in our cohort upon diagnosis.   Our data showed that AS-PCR is capable of detecting samples with a tumor cell load as low as 0.625%, which is far beyond the sensitivity of Sanger sequencing. The role of MYD88 mutations in determining the biological features and clinical outcome of CLL patients remains controversial. The initial study indicated that patients with MYD88 mutations were diagnosed younger and suffered a moreless advanced clinical stage 9 . Contradictory results, however, were observed in that MYD88 mutations showed no age and stage preference in CLL patients 7,11 , nor does our data do. In the subgroup analysis of M-CLL, we observed MYD88 mutations predict shorter TTT in this category with favorable outcome. Furthermore, CLL patients with MYD88 mutations had comparable prognosis with those with unmutated IGHV in our cohort, implying MYD88 mutations may counteract the survival advantage of mutated IGHV gene.
Early research has shown that CLL cells with MYD88 mutation co-immunoprecipitates with a larger amount of IRAK1&IL-1/TLR signaling pathway, and that activation of the IL-1/TLR pathway promotes proliferation in CLL cells 15 . Furthermore, MYD88 mutated CLL cells have higher phosphorylation and more DNA-binding activity in NF-κB subunits than CLL cells with wild-type MYD88.
All these results suggests MYD88 mutation is a gain-offunction molecular change which may aberrantly activates NF-κB signaling pathway in CLL cells 2,9 and offers explanation for the unfavorable prognostic impact of MYD88 mutation on the M-CLL subgroup.
We also found patients with MYD88 mutations have a relatively lower CD200 MFI compared to the wildtype cases do, consistent with a previous report 16 . Along with the fact that none of the MYD88 mutated CLL patients expressed positive CD38 in our study, we postulate that this subgroup of CLL patients may have a distinct immunophenotype from CLL without MYD88 mutations. This will be further explored by targeted RNA sequencing and whole genome sequencing. MYD88 mutations are mutually exclusive of IGHV 4-34 gene usage, which was not shown before to our knowledge. Unlike previously reported, we did not observe a preferable IGHV 3-23 gene usage in MYD88-mutated cases 10 .
In conclusion, in our cohort of newly diagnosed CLL patients, MYD88 mutations showed an incidence of 8.8%, including 6.3% on the hotspot missense mutation L265P. MYD88 mutations predict unfavorable prognosis within the M-CLL subgroup.