Characterization of flow cytometric immuno-phenotyping of acute myeloid leukemia with minimal differentiation and acute T-cell lymphoblastic leukemia: A retrospective cross- sectional study [version 1; peer review: awaiting peer review]

Background: Acute leukemias (ALs) are a heterogeneous group of malignancies with various clinical, morphological, immunophenotypic, and molecular characteristics. Distinguishing between lymphoid and myeloid leukemia is often performed by flow cytometry. This study aimed to evaluate the immunophenotypic characterization and expression of immuno-markers in both acute myeloid leukemia (AMLM0) and acute T-cell lymphoblastic leukemia (T-ALL). Methods: A retrospective cross-sectional study was conducted in the Pathology Department/Teaching Laboratories/Medical City/Iraq and included all patients newly diagnosed with AL from 5 January to 10 December 2018. Immunophenotypic analysis was performed on bone marrow samples, freshly collected in EDTA tubes. Flow cytometry (Canto-2 BD) was used, with laser excitation of blue and red wavelengths. A panel of monoclonal antibodies (MoAbs) was used for diagnosis, using a SSC/CD45 gating strategy. Results: The study showed 41.6% of AML-M0 patients had no aberrant antigen expression, while 33.3%, 16.6%,  8.3%, and 8.3% had aberrant CD7, CD56, CD2, and CD19, respectively. In 16.6% of AML-M0 cases more than one aberrant antigen was expressed. With regard to T-ALL, 7.0% were pro-T type, 58.0% were pre-T, 13.0% were cortical, and 22.0% were mature-T type. In 55.5% of patients with T-ALL there was no aberrant antigen expression. Conclusion: We concluded that most patients with AML-M0 have no aberrant antigen expression. In patients with T-ALL, the pre-T type is most common, according to the European Group for the Immunological Classification of Leukemias (EGIL) classification. Patients with T-ALL also generally lack aberrant antigen expression. Open Peer Review Reviewer Status AWAITING PEER REVIEW Any reports and responses or comments on the article can be found at the end of the article. Page 1 of 12 F1000Research 2020, 9:1170 Last updated: 05 OCT 2020


Introduction
There are many characteristic categories of leukemia, which have well known prognoses and specific treatments. Differentiation between lymphoid and myeloid leukemia is performed using flow cytometry 1 . The analysis of acute leukemia immunophenotypic done by flow cytometry, this method recently become a tool for distinguishing between myeloid and lymphoid lineages. Several types of acute leukemias identified morphologically with or without enzyme cytochemical analysis, but immunophenotyping stay a vital for the right identification of myeloid lineages in minimally differentiated acute myeloid leukemia (AML-M0) and the differentiation of B or T-cell lineages in acute lymphoblastic leukemia (ALL) 2 . Flow cytometry (FC) is a newer technology that can be used to detect many CD markers in or on a single malignant cell, and also possesses multi-parametric capabilities for combining the study of physical characteristics, such as cell size and granularity, alongside multiple CD markers. The advantages of FC include simultaneous analysis of multiple antigens on the same cell, assessing small samples, and providing results to make an objective diagnosis within a few hours 3 . Immunophenotyping permits reproducible lineage detection and assignment of leukemia types. On the other hand, intra-permeable cytoplasmic markers, such as myeloperoxidase, cCD22, cCD3, and cCD79a, are more accurate and specific indicators 4 . Distinguishing AML-M0 from ALL remains a challenge in a minority of cases. The French-American-British (FAB) Cooperative Group recognized that many cases morphologically classified as types of ALL appeared to be of myeloid origin, as determined by monoclonal antibodies and ultra-structural cytochemistry 5 .
This study aimed to evaluate the immunophenotypic characterization and the expression of commonly used immuno-markers in both acute myeloid leukemia with minimal differentiation (AML-M0) and T-cell acute lymphoblastic leukemia (T-ALL), and to define the best use and the role of multi-parameter flowcytometry in the diagnosis and proper classification of these types of acute leukemia. Also, to evaluate the aberrant antigen expression in both AML-M0 and T-ALL and its correlation with clinical and prognostic criteria.

Study design and setting
A retrospective cross-sectional study was conducted among 69 participants with newly diagnosed AL (M0 or T-ALL); there were 24 cases of AML-M0, 12 females and 12 males; and 45 cases of T-ALL, 40 males and five females. This study was conducted in the Pathology Department/Teaching Laboratories/Medical City/Iraq, and included all patients newly diagnosed with AL during the period from 5 January to 10 December 2018. Demographic characters of patients like age, and gender were collected from medical records of the participants. General clinical feature of each patients collected either when taking the patients history or during clinical examination included hepatosplenomegaly; lymphadenopathy (LAP); pallor; history of bleeding from gum, gingivae, and skin; fever; effusion of pleura; and presenting with a mediastinal mass.
3. de novo cases, those not receiving steroid or chemotherapy before sample collection 4. Those with a positive diagnosis of T-ALL or AML-M0 confirmation.
Exclusion criteria 1. Death of patient prior to treatment 2. Those who received chemotherapy in other hospitals either before referral or after finishing induction.
4. Those with history of steroid intake of than one week.
Diagnostic criteria 1. Those presented have fever, abdominal fullness, and skin rash.
3. Those with history of bleeding from gum, and gingivae.
4. Those with elevated WBC count, normal RBC counts, decrease platelets counts in CBC.

5.
Those with chest masses on chest radiograph.
6. Those with LAP and organomegaly on CT scans of the chest and abdomen/pelvis.

Sample size calculation
A sample size of 69 participants was calculated by the following formula:ˆ2 where z is the z score for 95% CI=1.96 ε is the margin of error=5% n is sample size p is the population proportion=50% This means 69 or more measurements/surveys are needed to have a confidence level of 95% that the real value is within ±5% of the measured/surveyed value. We the population proportion based on historical data from our center, whereby we found 50% of those attending were eligible for the study or in other words 50% were diagnosed with T-ALL or AML-M0.
Period of recruitment, follow-up, and data collection Participants were recruited at their first visit to the center. Each patient attending our center meeting the inclusion criteria were invited to be including in our study after which written informed consent was obtained. Follow-up was performed at every subsequent visit. In EDTA anti-coagulated tubes (ATACO / China, Catalogue No.:443N980100E), the bone marrow aspirate (BMA) samples were collected (2-3 ml). The method of collection as follow: we marked and cleaned the area where the biopsy needle will be inserted. The bone marrow fluid (aspirate) and tissue sample (biopsy) were usually collected from the top ridge of the back of the hipbone (posterior iliac crest). A small incision was made, into which the needle was inserted into the bone and into the bone marrow. A sample of the liquid portion of the bone marrow was then drawn. Following sampling pressure was applied to the area of insertion to stop the bleeding and then a bandage will be placed on the site. Bone marrow was either aspirated and tested the same day or stored, if necessary, in a refrigerator (2-8°C), for no more than 48 hours. We pre-washed all the sample using at least 25 volumes excess 1X PBS (BD biosciences, Catalogue No.: 554781), mixed well. Then, pelleted cells by centrifugation and resuspend in 1X PBS with 0.1% sodium azide to the original volume.

Addition of surface and cytoplasmic antibodies
After adding surface markers, 0.5 ml of permeabilizing reagent was added to each tube. The tubes were vortexed then incubated for 15 minutes in the dark. Then, 6 μl of each cytoplasmic marker was added and the samples were incubated at room temperature in the dark for 30 minutes. Then, 0.5 ml of FACS solution was added, and the samples were re-suspended by vigorous mixing before being analyzed using the flow cytometer.
Permeabilizing reagent 8 Permeabilizing Solution 2 (BD FACS™) is a premixed concentrate formulated specifically for use in flow cytometry. It is intended to permeabilize cell membranes prior to intracellular immunofluorescence staining with monoclonal antibodies.

Lysing reagent 9
Cell Lyse (BD biosciences, Catalogue No.: 559759), which is an erythrocyte lysing reagent kit for wash and no-wash procedures was used for cell lysis. Residual debris was removed by centrifugation and cell washing. Fixative reagent was used to fix and stabilize leukocytes. The fixed samples were stored for up to 24 hours at 2°C to 8°C before analysis.

Gating strategy
The dominant population of leukemic blasts was identified using side scatter versus CD45 dot plots to isolate the blast population. A 20% cut-off value was used to indicate the positivity of surface markers, while the cut-off value for cytoplasmic markers was 10%.

Software
Data acquisition and analysis was performed using FCS DIVA-8, made by the BD Software Company.

Ethical considerations
Written informed consent was obtained from all participants, or the parents of those aged less than 18 years. The Medical Ethical Committee at the Oncology Center, Al-Kindy College of Medicine, Baghdad University, Iraq, approved this study (code: 52).

Statistical analysis
We used Microsoft Excel (v. 2010) to calculate frequencies and percentages of variables, and to calculate means and standard deviations (SDs).

Results
A total of 69 patients were newly diagnosed with AL; 24 patients with AML-M0, 12 females and 12 males; and 45 patients with T-ALL, 40 males and 5 females, as shown in Figure 1, the age and sex of patients is shown in Table 2 and Table 3 10 .
Regarding clinical findings, we listed in Table 4 Table 5, Table 6 and Table 7 10 .

Discussion
Flow cytometry of leukemic cells plays an essential role in the identification of leukemia cell lines, maturation stage, and detection of residual disease. Immunophenotyping can improve both the accuracy and reproducibility of leukemia classification and is considered particularly useful for identifying AML with lymphoid marker expression and ALL with myeloid marker expression 12 . Some cases of AML associated with the expression of CD7 have a poor prognosis 13 , while other studies have reported CD2 and CD19 are better prognostic indicators in cases of AML 14 . In this study, 69 participants with de novo AL were enrolled, 45 of whom had T-ALL and 24 of whom had AML-M0. Most patients (88%) with T-ALL were male, which was in accordance with a study by Abid Salih and colleagues 15 and also similar to another study that showed a predominance of males among patients with T-ALL 16 . The male:female ratio was 1:1 in patients with AML-M0 in our study, which was consistent with earlier work 17 .
In the present study, the most common features patients with T-ALL presented with were pallor and fever 14,[17][18][19] , followed by hepato-splenomegaly and lymphadenopathy, which is in accordance with the findings of other studies 14,17,18 . The least common presenting features were CNS involvement, pleural effusion, and mediastinal mass 19,20 .
Regarding the clinical features of AML-M0, the current study found that pallor and fever were present in 16 (66.7%) of patients, consistent with the findings of Ghosh et al. 17 . Anemia is a common feature of all acute leukemias and is due to bone marrow infiltration; again, our results are comparable with other studies 16    AML-M0. Such bleeding may be due to a number of causes, such as thrombocytopenia due to bone marrow suppression by infiltrating malignant leukemic cells, and by platelet dysfunction that accompanies leukemic involvement in the bone marrow 16 . Gaydos et al. were the first to report this in 1962, when they found that there was a linear relationship between bleeding and platelet count 21 . Qazi et al. correlated this feature with thrombocytopenia alone or disseminated intravascular coagulopathy (DIC) 22 . Extramedullary infiltration by leukemic cells may result in lymphadenopathy, splenomegaly, and/or hepatomegaly. In our patients, we observed hepatomegaly in 24.6% of them. We concluded that pleural effusion was absent in AML-M0 patients but present in one (2.2%) patient with T-ALL. Regarding mediastinal mass, this was absent in cases of AML-M0, and this agree with that noted by Ghosh et al. 17 .
The incidence of lymphoid antigen expression in AML-M0 cases was 58.4%, which was higher than that seen by El-Sissy et al. 12 and Khalidi et al. 23 , who found aberrant lymphoid antigen expression in AML to be 47% and 48.1%, respectively. The incidence of CD56 expression was 16.6% in this study, which was similar to rates recorded in previous studies [24][25][26][27][28] . Many studies have concluded that overall survival was significantly decreased in individuals who expressed CD56 [24][25][26] . Regarding CD7, it was expressed in (33.3%) of patients with AML-M0, which was in agreement with other studies that found that CD7 to be aberrantly expressed in 10%-40% of blasts in AML patients 26 . Kita et al. 13 showed that CD7 expression on cells is indicative of phenotype, is functional when immature, and is regarded as a prognostic risk factor.
Here, the most commonly expressed lymphoid marker was CD7 (33.3%), which is the same as recent reports by Bahia et al. 29 and Zheng et al. 30 , although it is higher than the results reported by Venditti et al. 31 . CD4 was expressed in 25% of patients with AML-M0, which is comparable to one other study 32 but higher than values reported by some other studies, e.g., 17.5% 33 and 16% 23 .
In the current study, CD19 was expressed in 8.3% of patients, which is in accordance with the range of 8.6%-10% found in other studies like Khalidi et al. 23 , Bahia et al. 29 and Venditti et al. 31 but contrasts with other reports of 14% 34 and even lower  36 , who found the expression of CD2 to be 3%, while Bahia et al. 29 and Venditti et al. 31 found higher values, of 11.4% and 14%, respectively.
The pattern of myeloid expression is correlated with some genetic features of blast cells. CD15, CD33, and CD65 are expressed in ALL cases with a rearranged MLL gene, and CD13 and CD33 are expressed in cases with the ETV6-RUNX1 gene 37 . The presence of these antigens lacks prognostic significance in contemporary treatment programs 38 .
The CD13 and CD33 were expressed in 6 (13.3%) and 5 (11.1%) of cases, respectively, which is consistent with another study 39 . Acute T-cell lymphoblastic leukemia with aberrant CD117 presented in 6 (13.3%) patients and is also comparable with the proportion seen in other studies [40][41][42][43] . Three (6.7%) patients with T-cell acute lymphoblastic leukemia expressed aberrant CD19, a lower value than other studies which found the expression of CD19 in cases of T-ALL to be 22% 44 . T-cell acute lymphoblastic leukemia expressing aberrant CD56 was only seen in one (2.2%) patient, and this is consistent with other studies 45 .

Conclusion
We concluded that aberrant expression of CD markers was present in some cases of acute leukemia. The incidence of aberrant antigen expression was comparable with other studies and represents an indicator of poor prognosis. CD13, CD33, and CD117 were the most frequent aberrant myeloid antigens expressed in T-ALL cases, while CD19 and CD56 were expressed in fewer cases. A large number of AML cases had aberrant lymphoid phenotypes, with T-cell markers more common than B-cell markers. In addition, CD7 was a common aberrantly expressed lymphoid marker in AML. In the future, correlation studies of cytogenetic and lymphoid phenotypes should be performed to determine if there is any association between specific cytogenetic anomalies and the aberrant expression of certain lymphoid markers in patients with  AML, or between specific cytogenetic anomalies and myeloid markers in patients with ALL.