Downregulation of Lymphoid enhancer‐binding factor 1 (LEF‐1) expression (by immunohistochemistry and/ flow cytometry) in chronic Lymphocytic Leukemia with atypical immunophenotypic and cytologic features

Lymphoid enhancer‐binding factor 1 (LEF‐1) overexpression has been recently remarkably reported in chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL) and has shown utility in distinguishing CLL/SLL from other B‐cell lymphomas. CLL has a well‐defined immunophenotype, yet, some cases of CLL demonstrate atypical morphology/ phenotype reflected by low Matutes score (atypical CLL). Till date, LEF1 expression has not been systematically studied in cases of CLL with atypical features.


| INTRODUC TI ON
CLL has a distinct immunophenotype (IPT) (CD19+, CD5+, CD23+, CD20 dim+, CD79b dim/-, FCM7−, CD10-, surface Ig dim + and CD200+), this is to a part defined by the most widely accepted scoring tool for the immunophenotypic characterization of CLL, defined more than 20 years ago (Matutes score), which mandates strong expression of CD5 and CD23, lack of FMC7 as well as low or absent expression of CD79b and weak surface IgM (sIgM). 2,3 Yet, there are some equivocal cases with less typical IPT, especially those with a low Matutes score (including rare cases of CD5-negative CLL/SLL), these cases are referred to as atypical CLL (aCLL). According to Matutes score, CLL is usually correlates with score, 4,5 scores of ≤ 3 seen with atypical CLL and score of 0-2 is unusual in CLL.
The differential diagnosis of CLL/ SLL from its mimickers of CD5positive small B-cell lymphoma can be challenging in some cases due to overlapping morphologic and immunophenotypic features, particularly when analyzing peripheral blood or limited tissue samples. 4 Even with the introduction of CD200 as a useful marker for CLL, some cases remain diagnostically challenging emphasizing the necessity for additional markers in order to confirm or exclude the CLL diagnosis.
Several relatively recent studies have demonstrated dysregulation of WNT/β-catenin signaling in CLL. 5 The ultimate mediator of this pathway is a nuclear complex of lymphoid enhancer-binding factor (LEF)/T-cell factor (TCF) and β-catenin. [6][7][8] Lymphoid enhancer-binding factor (LEF-1) is a pivotal transcription factor that regulates cell proliferation and survival and has a critical role in lymphopoiesis. LEF-1 is normally expressed in T-cells and precursor B-cells 9 and is switched off in mature B-cells; however, gene expression profiling studies have revealed overexpression of LEF-1 in CLL/ SLL in comparison with normal B-cells. 10 LEF-1 has been shown to be suppressed in normal natural killer (NK) cells, therefore, NK may also serve as a negative reference population for LEF-1 staining. 11 Few studies had recently evaluated the expression of LEF-1 by immunohistochemistry (IHC) and concluded that nuclear staining of LEF-1 is expressed in almost all cases of B-cell CLL/SLL, and negative in nearly all other small B-cell lymphomas. 12,13 Although LEF-1 expression has shown utility in distinguishing CLL from other small B-cell lymphomas, yet its use in the diagnostic setting is still limited. In addition, a substantial number of CLL patients with negative LEF-1 expression has been reported in the literature.
However, the morphologic, immunophenotypic and cytogenetics features of LEF-1 negative CLL cases have not been studied before.
This is a single center study of LEF-1 expression using two different techniques IHC and /or flowcytometry (FCM) analysis to test the utility of using LEF-1 (by either method) for optimizing the diagnosis B-cell lymphoma including CLL/SLL. In this study, we will investigate the expression profile of LEF-1 in cases of classic CLL/SLL, in comparison to low-grade B-non-Hodgkin's lymphoma (other than CLL), and CLL with atypical features (atypical CLL; aCLL): including those with atypical IPT, CLL with increased prolymphocytes/atypical lymphoid cells and CLL with complex karyotype.
In addition, we will present and discuss the morphologic, immunophenotypic, genetic features/markers associated with LEF-1 expression in patients with CLL, in order to find out the LEF-1 expression profile in patients with aCLL and whether LEF-1 would be of diagnostic importance in differentiating these patients from non-CLL patients.  For cases with classic CLL and aCLL, LEF-1 expression was correlated with morphologic features: increased atypical lymphoid cells (clefting/ folding, larger less mature forms) and cases with increased prolymphocytes, different immunophenotypic markers (CD5, CD79, CD200, CD23, CD38, light chain intensity), FISH markers (del 13q, trisomy 12) and karyotype (KT) findings.
LEF-1 nuclear staining on tissue sections was assessed semi-quantitatively by microscopic examination of LEF1-stained slides performed by two experienced pathologists who blindly examined the specimens. T-cells were used as an internal positive control. Whenever applicable, LEF-1 assessment was performed in areas with predominance of malignant B-cells, positivity was interpreted together with CD3/CD20.
Cases are classified as LEF-1 + and scored as (strong:III, moderate:II or weak:I), and as LEF1-ve if the great majority of leukemic cells were negative for nuclear LEF-1. A 70% cut off was used to identify cases with a heavy T-cells infiltrate. A negligible degree of inter-investigator variability between the two reviewers was documented, reflected by minor differences in grading of positivity.

| Flow cytometric analysis of LEF1
EDTA blood samples were lysed using BD Pharm Lyse™ lysing so- In difficult cases, correlation with FCM results (done at diagnosis) was performed for accurate separation of malignant B-cells, validation of precise percentages of residual normal B, T-cells and NK cells.

| Statistical analysis
Categorical and continuous values were expressed as frequency (percentage) and mean ± SD or median and range as appropriate.
Descriptive statistics were used to summarize demographic, laboratory, morphologic, flow cytometry and cytogenetics features of the patients. The Kolmogorov-Smirnov (K-S) test was used to assess the normality of the data distribution. Associations between two or more qualitative variables were assessed using Chi-square test or Fisher Exact test as appropriate (to examine the association between morphologic, immunophenotypic, genetic features/markers with LEF-1 expression, if any). Quantitative data between the two independent groups were analyzed using unpaired "t" and Mann Whitney U tests as appropriate. To assess diagnostic and predictive accuracy of LEF-1 expression (in cases of CLL and atypical CLL), the sensitivity, specificity, positive and negative predictive values (along with their corresponding 95% confidence limits) were calculated. ROC analysis was performed to determine an optimum and suitable cutoff value for respective test/ markers to evaluate accuracy of such diagnostic markers against the gold standard. A two-sided P-value < 0.05 was considered to be sta-

| Cytogenetics findings within cases of CLL and atypical CLL
del13q (17/51)    including three cases of CLL with negative CD5 (Figure 1;

| LEF-1 results by flow cytometry
The malignant cells/NK ratio for LEF-1 expression in CLL/SLL with classic IPT was higher than CLL with atypical immunophenotypic/cytologic features (aCLL) and is significantly higher in other small B-cell lymphomas (P < .01). This was evident by both scatter diagram and box plot (Supplementary 1; B One CLL case showed discrepant LEF-1 expression; negative by IHC and positive by FCM (B/NK ratio = 2), this case was classified as atypical CLL (Matutes score 3) and had trisomy 12. Figure 2 shows examples of LEF-1 expression by flow cytometry in different groups.  Table 3 CD5

| Correlation between different clinical and pathologic features of typical CLL compared to atypical CLL (aCLL) are summarized in supplementary (4)
The proportion of Trisomy 12 was observed to be significantly lower

| D ISCUSS I ON
In this mixed design study, we evaluated the usefulness and applicability of LEF-1 in the diagnostic setting of a small B-cell lymphoma. Since ily performed on fresh blood samples, which makes it an applicable method to be utilized in the majority of CLL cases, for which the initial diagnosis is made on peripheral blood samples.
We are frequently challenged with a considerable number of atypical CLL, as a relatively large proportion of CLL diagnosed in our center exhibit one or more of atypical cytologic or immunophenotypic features, estimated by ~ 36% of the total number of CLL cases.
Immunophenotypically, aCLL usually exhibit less common features TA B L E 3 showing the association between various immunophenotypic and cytogenetics markers and LEF-1 expressions. CD5, CD200, CD23 positivity percentage and dim light chain expression had significantly higher proportions in patients with positive LEF-1 positive expression (P < .05). In contrast, CD79, FMC7 positivity percentages were found to have significantly lower in patients with LEF-1 positive expression (P < .05). However, the proportion of Trisomy 12 was found to be similar in both LEF-1 positive and negative expressions, and this difference was noted to be statistically insignificant (P = .865). No statistically significant difference for CD38 expression or del13q between LEF-1 positive and LEF-1 negative cases  (Figure 1, A-H)

Correlation between different immunophenotypic markers
and LEF-1 expression has not been systematically studied before.
However, Menter et al, reported no observed correlation between LEF-1 expression and trisomy 12, p53 or CD38. 26 We studied the association between some selected immunophenotypic and cytogenetics markers and LEF-1 expression in CLL cases (Table 3)

| Expression of LEF-1 in some interesting cases in our cohort
We found out that LEF-1 is a useful marker in cases of composite lymphomas (CL). In out cohort, we had a case of CL (CLL and Mantle cell lymphoma), 31

| CON CLUS ION
LEF-1 is not expressed in the great majority of low-grade lymphomas and is a reliable marker for distinguishing them from CLL/SLL with an added value in supporting the diagnosis of cases of discordant and composite lymphomas. In this study we document that LEF-1 expression is downregulated in a significant number of cases of CLL with atypical immunophenotype/features compared to classic CLL; Chi-Square P < .0001. As downregulation of LEF-1 expression is well correlated with atypical CLL, we suggest adding it to Matutes score as a beneficial marker to differentiate classic from atypical CLL. LEF-1 could also serve as a potential prognostic indicator for CLL clinical course. Further studies are desirable to support our data regarding LEF-1 expression in atypical CLL, and whether its downregulation has any prognostic/ predictive value.

ACK N OWLED G EM ENT
We would like to acknowledge the great contribution of technical staff in hematology & flow cytometry laboratory especially Miss Suhair Hassan El ajez and Miss Somayya Rahhal for their hard work and continuous support. We would also like to acknowledge the crucial role and the great contribution of cytogenetics laboratory staff especially Miss Farzana Murad, staff of the interim research laboratory and histopathology laboratories in HMC.

CO N FLI C T S O F I NTE R E S T
"The authors have no conflicts of interest to declare."

AUTH O R S CO NTR I B UTI O N S
Soliman D S created the research design, patients' selection, performed data collection, hematopathologic examination, data analysis, wrote the manuscript and formulated and submitted the paper.
Al-Kuwari E performed the histopathologic examination and reviewed the data.
Siveen K, performed flow cytometry analysis and reviewed the manuscript.
Chandra P, performed the statistical analysis and reviewed the manuscript.
Yassin M, participated in patients' recruitment, consenting and clinical assessment.
Ibrahim F, shared in data collection and reviewed the manuscript.
Nashwan A, participated in research protocol submission, patients' recruitment and data collection.
Taha R Y, participated in patients' recruitment, and clinical assessment.
Nawaz Z, performed the cytogenetic analysis.
El-Omri H participated in clinical assessment.
Mateo J M, processed the samples for flow cytometry analysis.
Al-Sabbagh A reviewed the manuscript.

E TH I C A L A PPROVA L
The study was approved by HMC ethics committee on human research (Medical Research Centre-MRC).