Acute promyelocytic leukemia: evaluation of diagnostic tests from 2000 to 2018 in a public hospital

Hospital de Clínicas da Universidade Federal do Paraná (UFPR), Curitiba, Paraná, Brazil

Corresponding author

Aline Carvalho Hey
ORCID 0000-0002-5575-7807
e-mail: linechey@gmail.com

First Submission on 2/14/2019

INTRODUCTION

Acute promyelocytic leukemia (APL) is a subtype of acute myeloid leukemia (AML), first described in 1957 by Hillestad, with unique clinical, morphological and molecular characteristics⁽¹⁾. The APL is characterized by the presence of the translocation t(15;17)(q22; q21)⁽²⁾, which results in the breakage and fusion of the promyelocytic leukemia (PML) gene, located on chromosome 15, with retinoic acid receptor alpha (RARα) located on chromosome 17. The final product of the translocation is the fusion of the genes encoding the hybrid proteins PML-RARα and RARα-PML⁽³⁾, oncoproteins with sensitivity to retinoid action⁽⁴⁾. This oncoprotein causes interruption of maturation and the accumulation of cells in the stage of abnormal promyelocytes in the bone marrow⁽⁵⁾.

The presence of leukemic promyelocytes is occasionally observed in the peripheral blood; however, bone marrow analysis is essential for diagnosis, especially in cases with severe pancytopenia⁽⁶⁾. APL may present as typical morphology, with hypergranular promyelocytes, or as the microgranular form. In the hypergranular form, the size and shape of the nucleus are irregular, and the cytoplasm is marked by several dense granules, grouped in the form of Auer rods, which characterizes the faggot cells. In the hypogranular variant, the nucleus is bilobed and the cytoplasm presents few or no granule⁽⁷⁾

Although a morphological diagnosis is highly suggestive of APL, this is not enough to characterize it, and performing complementary exams are necessary, such as cytochemistry, multiparametric flow cytometry (MFC), conventional cytogenetic and molecular techniques. In the cytochemical analysis, neoplastic promyelocytes stain strongly with the myeloperoxidase or Sudan black reactions⁽⁸⁾. The immunophenotype study by MFC evidences the presence of blasts with high autofluorescence, which expresses myeloid markers such as CD117, CD13 and CD33, the latter with homogeneous pattern of fluorescence intensity, and where the CD34 hematopoietic precursor cell marker and human leukocyte antigen (HLA)-DR are usually negative. Unlike normal promyelocytes, markers indicative of myelogenous maturity CD11b and CD15 are negative or of very low expression. Because it is a quick method, MFC reinforces the diagnostic suspicion of APL and helps in early therapeutic indication, but it is not an adequate method for the definitive diagnosis, which should be performed by cytogenetic and molecular techniques⁽⁹⁾. It is important to emphasize that the three methodologies are not exclusive, but complementary for the accurate diagnosis of APL.

Although all patients with APL present t(15;17) or a variant of this translocation, these translocations are detected in only 70%-96% of patients at diagnosis using conventional cytogenetic methods. Rare cases of APL lacking the classic t(15;17)(q21.1;q21.2) on routine cytogenetic studies have been described with complex variant translocations, additional chromosome or with the submicroscopic insertion of RARα into PML leading to the expression of PML-RARα transcript⁽¹⁰⁾.

OBJECTIVE

This study aims to compare the results obtained in bone marrow morphological analysis, immunophenotype by flow cytometry and conventional cytogenetics from APL patients treated at Complexo Hospital de Clínicas da Universidade Federal do Paraná (CHC-UFPR).

METHODOLOGY

Retrospective analysis of patients with a suspected morphological diagnosis of APL treated at CHC-UFPR between January 2000 and July 2018.

The bone marrow morphology, classic cytogenetic and MFC results files were consulted and data were confronted, seeking to analyze the agreement between the three methodologies.

MFC was performed in BD FACS Calibur^TM using four-color panels, staining with the following combinations of fluorochrome-conjugated monoclonal antibodies [fluorescein isothiocyanate (FITC), phycoerythrin (PE) and either peridinin-chlorophyll-protein (PerCP) or the PE-cyanine 5 (PE-Cy5), fluorochrome tandem, and allophycocyanin (APC)] directed against surface antigens: CD11b, CD13, CD14, CD15, CD19, CD33, CD34, CD36, CD38, CD45, CD56, CD64, CD117, CD123, HLA-DR. In addition, the expression of MPO, CD79a, and CD3 was also explored at the cytoplasmic level. After the 2017 year, FACS Canto II^TM was introduced and eight-color Euroflow panel was performed in all the leukemia cases.

Conventional cytogenetic was performed as broadly described by Gus (2011)⁽¹¹⁾. Molecular analysis of PML-RARα transcript began to be performed systematically after the year 2006 when the site entered to the corporate Brazilian study named International Consortium on Acute Promyelocytic Leukemia (IC-APL)⁽¹²⁾. After this year all patients with molecular analysis performed were considered for confirmation of the disease.

Patient selection

Initially, 95 patients with a suspected cytomorphologic diagnosis for APL were selected between January 2000 and July 2018.

Four patients who did not undergo cytogenetics and flow cytometry tests for comparison were excluded. After morphological and flow cytometry review, two other patients who were diagnosed with AML with myelodysplasia-related changes, and one which was confirmed as acute monoblastic leukemia NOS, were also excluded. Cytogenetic profiles of these patients are described in Table 1.

To calculate the sensibility and specificity of classic cytogenetic analysis, we exclude the nine cases where the exam was not performed and consider the 82 remained cases (53 true-positive cases, 26 false-negative cases, and three true-negative cases). There were no false-positive cases in this cohort.

RESULTS

A total of 88 patients were evaluated, 42 men and 46 women, with a median age of 34 years (ranging from 12 to 78 years).

Only 66% of the cases could be classified according to the prognostic risk due to the availability of the data. The platelets median in this period were 31 x 10⁹/l (range 7 x 10⁹/l to 5 28 x 10⁹/l), and leukocytes 5.48 x 10⁹/l (range 63 x 10⁹/l to 122.92 x 10⁹/l). As a result, eleven patients (19.3%) were classified as low risk, twenty-two (38.6%) patients as intermediate risk, and 24 (42.1%) as high risk.

Bone marrow or peripheral blood morphological analyses were considered the gold standard of the study. Only two cases without bone marrow analysis were included, both of which exhibited peripheral blood morphology highly suggestive of APL. Ten cases of microgranular or variant morphology (dumbbell nuclei and few cytoplasm granules) were seen. Myeloperoxidase (MPO) cytochemistry analysis was positive in all of them.

From the 88 patients with a suggestive morphological analysis of APL, 83 presented MFC examinations (94.3%) (Figure 1).

FIGURE 1 – MFC tests suggestive of APL (yes, no or not performed), before and after 2006, and in the total study period
MFC: multiparametric flow cytometry; APL: acute promyelocytic leukemia; NP: not performed.

The conventional cytogenetic analysis was performed in 79 cases (89% of the sample), but the translocation (15; 17) was confirmed in 53 cases (60.2%). Thirteen patients presented a normal karyotype and five presented no metaphases for evaluation. Nine patients did not perform the karyotype, seven of them before the entrance in International Consortium (IC)-APL study. These data are shown in Figure 2.

FIGURE 2 – Conventional cytogenetics of patients with translocation t(15; 17) present, absent or unrecognized before and after 2006, and in the total study period
NP: not performed.

As expected, the specificity of classic cytogenetic was 100% in the morphologic suspected cases, but the sensibility was 67% (53 true-positive and 26 false-negative cases) when the not performed cases were considered.

The search of PML-RARα transcript by molecular biology was performed by the IC-APL study, since December 2005, in 35 patients (39.8%), all of them were positive. In regard to the 26 patients with non-conclusive karyotype, 14 (53.8% of tested, 15.9% of total cases) presented confirmation of the PML-RARα transcript presence by molecular analysis.

In total, 35 patients (39.8%) had PML-RARα gene research performed by molecular biology in samples sent to Ribeirão Preto, to the IC-APL study.

Before 2006, we found seven cases that did not have cytogenetic analysis performed, four cases without metaphases and five cases with normal karyotype. After this period only two patients did not perform the karyotype and one case did not show growth of metaphases in cell culture.

From the patients with translocation t(15;17) confirmed, there were additional cytogenetic alterations in six patients, two of which were considered complex karyotypes. There were eight patients who presented cytogenetic changes: +8 (n = 2), del11q (n = 1), del17p (n = 2) and marker chromosomes (n = 1), but did not present translocation t(15;17).

One case in this sample presented the variant translocation t(11; 17)(q13;q21), and one case presented a triple translocation t(5;15;17)(q13;q24;q21).

A translocation involving chromosome 15 and 17 or its variants were detected by conventional cytogenetic in 60.2% of cases, and confirmed by molecular analysis in 15.9%, totaling 76.1% of samples confirmed by genetic methods in this cohort. These results are shown in Table 2.

There were five cases of disagreement between morphology, flow cytometry and conventional cytogenetic and 32 cases of agreement only between immunophenotype and morphology, with normal karyotype in 13 cases and absence of metaphases in five cases. Nine did not perform culture and six performed another karyotype.

Only 45 patients (51.1%) agreed on the diagnosis among the three exams (morphology, flow cytometry and conventional cytogenetic) (Figure 3).

FIGURE 3 – Set diagrams related to bone marrow morphology, immunophenotype and cytogenetics for patients diagnosed from January 2000 to July 2018

Treatment and follow-up

Most patients received all-trans retinoic acid (ATRA) therapy (45 mg/m² orally, until complete remission), which was initiated as soon as possible after the presumptive diagnosis of APL, based on morphological characteristics, immunophenotype, and clinical judgment. Between 2001 and 2006, the treatment protocol was based on the combination of ATRA with chemotherapy (daunorubicin 60 mg/m² and mitoxantrone 10 mg/m²), followed by consolidation with bone marrow autologous transplantation. After 2006, the induction protocol was replaced by anthracycline on days 2, 4, 6 and 8, associated with the ATRA, according to the treatment algorithm proposed by the IC-APL in 2006⁽¹²⁾.

At a median segment of 4.83 years, 59 patients are alive with complete response (overall survival of 67%). There were 18 early deaths (up to three months of therapy) and 11 late deaths, of which six deaths due to confirmed relapse. Fourteen from the 29 deaths reported were of patients at high risk (48.3%).

DISCUSSION

APL is a type of leukemia caused by the translocation t(15; 17) (q24;q21), whose product is the PML-RARα oncoprotein. The initial clinical-morphological evaluation of APL is essential for the early initiation of therapy; however, it is known that this is not always conclusive⁽¹³⁾.

APL requires special attention among AML subtypes for its prognostic and therapeutic implications. With current therapy, 70%-80% of patients remain disease-free survival for at least five years⁽¹⁴⁾. It is known that patients with PML-RARα fusion protein present coagulopathy, 40% of whom develop severe bleeding, such as pulmonary and cerebral, which may be lethal⁽¹⁵⁾. Due to this complication, APL was for many years considered to be one of the most fatal subtypes of AML. However, since the introduction of ATRA as a treatment, it has become the most curable type of AML⁽¹⁶⁾.

The definitive APL diagnosis should be confirmed by techniques capable of detecting the translocation t(15;17) (q24;q21) or the PML-RARα rearrangement, or alterations involving the RARα gene. Conventional cytogenetics, fluorescence in situ hybridization (FISH), and reverse transcriptase-polymerase chain reaction (RT-PCR) are the available options. The FISH technique has lower sensitivity than RT-PCR, but is more specific.

The conventional karyotype present the advantage of allowing the diagnosis of additional cytogenetic alterations, however, the execution time is longer and the number and quality of metaphases may be variable⁽⁹⁾. Rare cases of APL lacking the classic t(15;17) (q24;q21) on routine cytogenetic studies have been described with complex variant translocations, additional chromosome or with the submicroscopic insertion of RARα into PML leading to the expression of PML-RARα transcript⁽¹⁰⁾. Despite the limitations of the technique, cytogenetics seems to have good specificity and positive predictive value in this pathology.

In the present study, the t(15;17) translocation or its variants were confirmed by conventional cytogenetic analysis in only 60.2% of the sample (53 cases), and 14 (15.9%) additional patients with inconclusive karyotype could be confirmed by the presence of the PML-RARα transcript by molecular analysis. In total 76.2% of samples from this cohort were confirmed by genetic methods. This result is similar to that obtained by Berger and Coniat (2000)⁽¹⁰⁾, where 62.8% of the 121 studied patients presented this cytogenetic alteration. As expected, the specificity of classic cytogenetic was 100% in this study, but the sensibility was quite low (67%). According to literature, the translocation t(15;17) is present in up to 90% of the cases in some studies⁽¹⁷⁾. The low percentage of translocation detection in conventional cytogenetic may be due to the limitation of the technique since the analysis can be performed on cells that do not belong to the neoplastic clone (normal clone), there may be difficulties in identifying the translocation or even due to the presence of cryptic rearrangements that mask the translocation^(14,17).

In this study, in five cases, there was no success in obtaining metaphase, four of them occurred before 2006. Besides this, the large number of non-realized tests (seven cases before 2006 and two cases after this date) can be explained by the absence of a rigid protocol for evaluation and follow-up of patients up to this period. The association with IC-APL⁽¹²⁾, established at the end of 2005, could contributed to the improvement of diagnostic tools and to the better prognosis of APL in our institution, since a standardized approach was implemented for a specific and immediate diagnosis, in addition to changes in treatment, measures of support and follow-up of the cases. It was possible to observe an increase in the conclusive diagnosis from the year 2006 when the IC-APL protocol was introduced in our site. There was a significant improvement in the correct diagnosis of this disease due to the establishment of the correct flow of the diagnostic tests to be performed, resulting in a decrease in the number of exams that were not processed and in an increase in the conclusive diagnosis, mainly due to the implantation of the PML-RARα transcript by molecular biology.

Approximately 30% to 40% of patients with APL may have additional cytogenetic alterations beyond t(15;17). The most frequent are chromosomes 8 and 21 trisomies, structural abnormalities on chromosome 9 and isochromosome 17q⁽⁸⁾. In this study, it was found two cases with trisomy of chromosome 8 and one case with a deletion in the long arm of chromosome 17.

Besides this, in about 2% of cases the RARα gene can be fused with a gene other than PML, and promyelocytic leukemia zinc finger (PLZF) is a result of t(11;17)(q23;q21), NPM (nucleophosmin), resulting from t(5;17)(q35;q21), and nuclear mitotic apparatus (NuMA), resulting from t(11;17)(q13;q21)⁽¹⁸⁾. Complex translocations with more than two chromosomes can also be found, with 33 proved cases reported in the literature since the year of 2009⁽¹⁹⁾. In this study, were found two variant translocations: translocation t(11;17)(q13;q21) in one case and t(5;15;17)(q13;q24;q21) in the other.

Flow cytometric immunophenotyping has been useful in distinguishing APL from the other types of AML, since the absence of expression of the HLA-DR and CD34 markers are characteristic but not specific for APL⁽²⁰⁾. AML with FLT3 and NPM1 mutations are examples of other diseases that may also present negativity for these markers⁽²¹⁾. This differentiation is quite important because this technique can help to guide disease-specific therapy, increasing the APL patient’s survival rate. In our study, there were three patients whose integrated diagnosis between morphology, MFC and cytogenetic were essential for the exclusion of APL.

The risk classification analysis was performed based on Cingam and Koshy study (2017)⁽²²⁾. Most patients in this study were classified as high-risk (41.1%), or intermediate-risk (39.3%), or low-risk (19.3%). This result was different from that found in European centers, which have a higher prevalence of intermediaterisk patients (55.5%), followed by low-risk patients (37.8%) and lastly, high-risk patients (16.7%)⁽²³⁾. The overall survival was 67% in five years, but almost half of the 29 deaths reported occurred in patients at high risk, indicating the importance of this classification in the prognostic evaluation of the disease.

CONCLUSION

A translocation involving chromosome (15;17)(q24;q21) or its variants was detected in patients suspected of having APL in 60.2% of cases by conventional cytogenetics and in 15.9% additional cases by molecular analysis, totaling 76.1% of samples confirmed by genetic methods in this cohort.

The low sensibility of conventional karyotype in APL demonstrated the importance of performing molecular techniques for diagnostic confirmation.

REFERENCES

1. Roldan CJ, Haq SM, Miller AH. Acute promyelocytic leukemia; early diagnosis is the key to survival. Am J Emerg Med. 2013; 31(8): 1290-1. PubMed PMID: 23702072.

2. Rowley J, Golomb H, Dougherty C. 15/17 translocation, a consistent chromosomal change in acute promyelocytic leukaemia. Lancet. 1977; 1(8010): 549-50. PubMed PMID: 65649.

3. Thé H, Chomienne C, Lanotte M, Degos L, Dejean A. The t(15;17) translocation of acute promyelocytic leukaemia fuses the retinoic acid receptor α gene to a novel transcribed locus. Nature. 1990; 347(6293): 558-61. PubMed PMID: 2170850.

4. Grignani F, Ferrucci PF, Testa U, et al. The acute promyelocytic leukemia-specific PML-RARα fusion protein inhibits differentiation and promotes survival of myeloid precursor cells. Cells. 1993; 74(3): 423-31. PubMed PMID: 8394219.

5. Muindi J, Frankel SR, Miller Jr WH, et al. Continuous treatment with all-trans retinoic acid causes a progressive reduction in plasma drug concentrations: implications for relapse and retinoid “resistance” in patients with acute promyelocytic leukemia. Blood. 1992; 79(2): 299-303. PubMed PMID: 1309668.

6. Jácomo RH, Figueiredo-Pontes LL, Rego E. Do paradigma molecular ao impacto no prognóstico: uma visão da leucemia promielocítica aguda. Rev Assoc Med Bras. 2008; 54(1): 82-9. Available at: http://www.scielo.br/pdf/ramb/v54n1/26.pdf.

7. Swerdlow SH, Campo E, Harris NL, et al. World Health Organization classification of tumors of haematopoietic and lymphoid tissues. Lyon: IARC Press; 2016.

8. Leal AM, Kumeda CA, Velloso, EDRP. Características genéticas da leucemia promielocítica aguda de novo. Rev Bras Hematol Hemoter. 2009; 31(6): 454-62. Available at: http://www.scielo.br/pdf/rbhh/v31n6/aop9009.pdf.

9. Orfao A, Chillon MC, Bortoluci AM, et al. The flow cytometric pattern of CD34, CD15 and CD13 expression in acute myeloblastic leukemia is highly characteristic of the presence of PML-RARαlpha gene rearrangements. Haematologica. 1999; 84(5): 405-12. PubMed PMID: 10329918.

10. Berger R, Coniat MB. Uneven frequencies of secondary chromosomal abnormalities in acute myeloid leukemias with t(8;21), t(15;17), and inv(l6). Cancer Genet Cytogenet. 2000; 117(2): 159-62. PubMed PMID: 10704690.

11. Gus R. Técnicas de cultura de tecidos para análise citogenética. In: Maluf SW, Riegel M, Schinzel A, et al., editors. Citogenética humana. Porto Alegre: Artmed; 2011. p. 54-62.

12. Rego EM, Kim HT, Ruiz-Argüelles GJ, et al. Improving acute promyelocytic leukemia (APL) outcome in developing countries through networking, results of the International Consortium on APL. Blood. 2013; 121(11): 1935-43. PubMed PMID: 23319575.

13. Lock RJ, Virgo PF, Kitchen C, Evely RS. Rapid diagnosis and characterization of acute promyelocytic leukaemia in routine laboratory practice. Clin Lab Haematol. 2004; 26(2): 101-6. PubMed PMID: 15053803.

14. Sagrillo MR, Cardoso SH, Silva LRJ, et al. Leucemia promielocítica aguda: caracterização de alterações cromossômicas por citogenética tradicional e molecular (FISH). Rev Bras Hematol Hemoter [Internet]. 2005; 27(2): 94-101. Available at: http://www.scielo.br/pdf/rbhh/v27n2/v27n2a08.pdf.

15. Dekking EH, van der Velden VH, Varro R, et al. Flow cytometric immunobead assay for fast and easy detection of PML-RARA fusion proteins for the diagnosis of acute promyelocytic leukemia. Leukemia. 2012; 26(9): 1976-85. PubMed PMID: 22948489.

16. Wang ZY, Chen Z. Acute promyelocytic leukemia: from highly fatal to highly curable. Blood. 2008; 111(5): 2505-15. PubMed PMID: 18299451.

17. Ramchandren R, Jazaerly T, Bluth MH, Gabali AM. Molecular diagnosis of hematopoietic neoplasms: 2018 update. Clin Lab Med. 2018; 38(2): 293-310. PubMed PMID: 29776632.

18. Sainty D, Liso V, Cantu-Rajnoldi A, et al. A new morphologic classification system for acute promyelocytic leukemia distinguishes cases with underlying PLZF/RARA gene rearrangements. Blood. 2000; 96(4): 1287-96. PubMed PMID: 10942370.

19. Abe S, Ishikawa I, Harigae H, Sugawara T. A new complex translocation t(5;17;15)(q11;q21;q22) in acute promyelocytic leukemia. Cancer Genet Cytogenet. 2008; 184(1): 44-7. PubMed PMID: 18558288.

20. Oelschlaegel U, Mohr B, Schaich M, et al. HLA-DRneg patients without acute promyelocytic leukemia show distinct immunophenotypic, genetic, molecular, and cytomorphologic characteristics compared to acute promyelocytic leukemia. Cytometry B Clin Cytom. 2009; 76(5): 321-7, PubMed PMID: 19291801.

21. Kussick SJ, Stirewalt DL, Yi HS, et al. A distinctive nuclear morphology in acute myeloid leukemia is strongly associated with loss of HLA-DR expression and FLT3 internal tandem duplication. Leukemia. 2004; 18(10): 1591-8. PubMed PMID: 15343344.

22. Cingam SR, Koshy NV. Cancer, acute promyelocytic leukemia (APL, APML). PubMed PMID: 2908325. Available at: http://www.ncbi.nlm.nih, gov/pubmed/29083825. [accessed on: 2018, Dec 2].

23. Crespo-Solis E, Contreras-Cisneros J, Demichelis-Gómez R, et al. Survival and treatment response in adults with acute promyelocytic leukemia treated with a modified International Consortium on Acute Promyelocytic Leukemia protocol. Rev Bras Hematol Hemoter. 2016; 38(4): 285-90. PubMed PMID: 27863754.