NOTCH1 mutation and its prognostic significance in Chinese chronic lymphocytic leukemia: a retrospective study of 317 cases

Abstract The proto‐oncogene NOTCH1 is frequently mutated in around 10% of patients with chronic lymphocytic leukemia (CLL). This study analyzed NOTCH1 mutation status of 317 Chinese patients with CLL by Sanger sequencing. The frequencies of NOTCH1 mutation in the PEST (proline (P), glutamic acid (E), serine (S), threonine (T)‐rich protein sequence) domain and the 3′ untranslated regions (UTR) were 8.2% and 0.9%, with the most frequent mutation being c.7541_7542delCT and c.*371A>G, respectively. Clinical and biological associations were determined including NOTCH1 mutations with advanced stage (Binet stage, P = 0.010), unmutated immunoglobulin heavy‐chain variable region (IGHV) gene (P < 0.001) and trisomy 12 (+12) (P = 0.014). NOTCH1‐mutated patients had lower CD20 expression intensity than NOTCH1‐unmutated patients (P = 0.029). In addition, NOTCH1‐mutated patients had shorter overall survival (OS) (P = 0.002) and treatment‐free survival (TFS) (P = 0.002) than NOTCH1‐unmutated patients, especially for patients with NOTCH1 c.7541_7542delCT and/or c.*371A>G mutations. Patients with both mutated NOTCH1 and unmutated IGHV had shorter OS (P < 0.001) and TFS (P < 0.001) than those with unmutated NOTCH1 or mutated IGHV. These data provide a comprehensive view of the clinical relevance and prognostic impact of NOTCH1 mutations on Chinese patients with CLL.

In Western countries, NOTCH1 mutations have a significant association with unmutated IGHV [18]. Pozzo et al. [19] found that NOTCH1 mutations suppressed CD20 expression intensity of B-cell surface by dysregulating histone deacetylase (HDAC)-mediated epigenetic changes. As a consequence, rituximab seems to have less benefits to patients with NOTCH1 mutations than those without mutation regardless of mutation sites [14,20]. Furthermore, NOTCH1 mutations are associated with trisomy 12 (+12), with nearly 40% mutated patients harboring +12 [21][22][23]. It has been reported that NOTCH1 abnormality represented a poor prognostic factor in CLL on both treatment-free survival (TFS) and overall survival (OS) [5,16,17]. The effects of different types of NOTCH1 mutations on prognosis were reported by D'Agaro et al. [24]. Although Xia et al. [25] and Wu et al. [26] partly investigated NOTCH1 mutations in Chinese with CLL, so far, there has been no systemic study. It remains unclear that whether NOTCH1 mutations have a similar influence on Chinese patients with CLL as Western countries.
In this retrospective study, we analyzed correlations between NOTCH1 mutations and other clinical and prognostic parameters in 317 Chinese patients with CLL. Besides, survival analysis was performed based on NOTCH1 and other cytogenetic status.

Patients
This single-center retrospective study included 317 patients with CLL diagnosed from November 1991 to December 2016 in our hospital. Diagnosis of CLL was based on the International Workshop on CLL-National Cancer Institute criteria. This study was approved by the hospital ethics committee, and all patients were provided informed consent according to the Declaration of Helsinki. Among 317 patients, mutation status of PEST domain of 161 patients had been reported by Xia et al. [25].

Cytogenetics and immunophenotyping
Karyotype analysis of CLL cells was performed after CpGoligodeoxynucleotide and interleukin-2 stimulation. Fluorescence in situ hybridization (FISH) was carried out according to the procedures described previously [25]. CD38 and ZAP-70 were detected via flow cytometry, and the cutoff levels for positivity were 30% and 20%, respectively.

Statistical analyses
SPSS 23 was used to analyze data. OS was defined as time from diagnosis to death or last follow-up, and TFS was calculated as time between diagnosis and first-line treatment. Survival curves were constructed by Kaplan-Meier method, and log-rank test was used for statistic associations. Continuous and categorical variables were analyzed by t test and χ 2 test, respectively. P < 0.05 was defined as statistical significance.

Patient characteristics
Totally, 317 patients were enrolled in our study with a male/female ratio of 2. (range, 20-92), and 53.3% patients were older than 60 years; 163 patients were newly diagnosed, of whom 67 had treatment indication; 97 patients were previously diagnosed elsewhere, of whom 31 patients had treatment indication. Thirty-three were relapsed, and 24 were refractory. Median time from diagnosis to sampling was 3 months (range, 0-264).
In addition, we analyzed the relationship between +12 and CD20 expression intensity. The number of patients who had data of CD20 expression intensity with and without +12 was 29 and 93, respectively. We found that +12 patients had higher CD20 expression intensity than patients without +12 (MFI: 796.90 ± 903.92 vs. 1294.70 ± 1167.18, P = 0.018).
Furthermore, we studied the synergetic effect of the status of NOTCH1 mutations and +12 on CD20 expression. We stratified our patients into four groups:   (Tables 4 and 5).

Survival
Patients with NOTCH1 mutation had worse TFS (Fig. 1A, median: 3 vs. 18 months; P = 0.002) and OS (Fig. 1D, median: 52 vs. 216 months; P = 0.002) than those with wild-type NOTCH1. Taking the status of NOTCH1 mutations and +12 into consideration at the same time, we further stratified our patients into four groups as previously mentioned: (1)

Discussion
In the present research, we analyzed clinical characteristics and outcome of 317 Chinese CLL patients with different NOTCH1 mutational status. The mutation frequencies of PEST domain and noncoding domains were 8.2% and 0.9%, respectively, which were lower than those reported in Western countries [28]. Moreover, only c.*371A>G was detected in noncoding domains, with no c.*378A>G or c.*380A>C. Several reasons might account for these different results: first, the differences of CLL genetic background between Eastern and Western countries patients. Second, the frequencies of 3′ UTR mutations were relatively low, with 2.0% for c.*378A>G and 0.8% for c.*380A>C [5]. Third, the low sensitivity of Sanger sequencing may contribute. Previous studies have demonstrated that some NOTCH1 mutations, especially those in noncoding domains, had too low variant frequency to be detected by Sanger sequencing, but could be found by next-generation sequencing [5]. Consequently, mutation frequencies reported by our study might be lower than the true level, and to better study NOTCH1 mutations, these limitations should be taken into consideration. The correlation between NOTCH1 mutations and chromosome abnormalities was assessed in the current study. NOTCH1 mutations were more common in advanced stage patients and were associated with CD38 positivity and unmutated IGHV [22]. Complex karyotype, 11q-, 13q-, and 17p-were not associated with NOTCH1 mutations. However, patients with NOTCH1 mutations usually carried +12 [21].
The relationships between NOTCH1 mutations or +12 and CD20 expression intensity were also performed in our study. Patients with NOTCH1 mutations had lower CD20 expression intensity, and +12 was associated with higher expression intensity of CD20. These findings had clinical value to guide the usage of rituximab for patients who had those two abnormities.
NOTCH1 mutation represented an unfavorable prognosis factor, which reflected on both TFS and OS [29]. By subgroup analysis, no statistic difference for TFS or OS was found among four groups stratified by NOTCH1 and +12 status. Both NOTCH1 mutations and unmutated IGHV gene had unfavorable effects on survival, but their synergistic effects remained unknown. In our research, we studied survivals of different mutation status of NOTCH1 and IGHV. We found that patients with both mutated NOTCH1 and unmutated IGHV had shorter OS and TFS than those with unmutated NOTCH1 or mutated IGHV. Frameshift deletion c.7541_7542delCT represented unfavorable outcomes for TFS and OS, and c.*371A>G had a negative effect on OS. As a result, CLL patients with c.7541_7542delCT and c.*371A>G mutation deserve more attention in the clinic practice.
In summary, we analyzed clinical and molecular features of NOTCH1 mutations in 317 Chinese patients with CLL. The frequencies of mutations in the PEST domain and 3′ UTR of NOTCH1 were lower than those reported by Western countries. NOTCH1 mutations were more common in patients with advanced stage, unmutated IGHV gene, and +12. Moreover, NOTCH1 mutation represented an unfavorable prognostic factor, especially for c.7541_7542delCT and c.*371A>G. Further researches are needed to better understand the molecular mechanisms of NOTCH1 mutations and study the appropriate strategies to treat Chinese patients with CLL who harbor NOTCH1 mutations.

Supporting Information
Additional supporting information may be found in the online version of this article: Table S1. NOTCH1 mutations identified by Sanger sequencing in 317 Chinese CLL cases. Table S2. Clinical characteristics of patients with NOTCH1 3′UTR mutation.