Morphology and clinical characteristics of a case of FIP1L1-RARA fusion-associated variant acute promyelocytic leukemia

Acute promyelocytic leukemia (APL) is a distinct disease type of acute myelogenous leukemia (AML) [1]. The majority of APLs feature a balanced reciprocal translocation between chromosomes 15q22 and 17q12, which results in the fusion of the promyelocytic leukemia (PML) and retinoic acid receptor alpha (RARA) genes [2]. Although more than 95% of APL cases carry the PML-RARA fusion gene, in approximately 1–2% of cases, other fusion partners of RARA are detected, such as NPM1, NuMA, PLZF, FIP1L1, STAT5b, and PRKAR1 [3,4]. FIP1-like 1 (FIP1L1) is a human homolog of Saccharomyces cerevisiae FIP1, which was initially isolated as a fusion gene with platelet-derived growth factor receptor alpha (PDGFRA) in patients with chronic eosinophilic leukemia [5,6]. In addition to the FIP1L1-PDGFRA fusion gene, FIP1L1 has also been detected to form a fusion gene with RARA, FIP1L1-RARA, in juvenile myelomonocytic leukemia (JMML) and APL [7,8]. Four cases of the FIP1L1-RARA fusion gene in JMML and APL have been reported in the literature to date [8–10]. We report the fifth case of the FIP1L1-RARA fusion gene in an APL patient and the only adult case with relatively long-term survival and different characteristics from typical APL.

In July 2020, a 30-year-old woman was admitted to our hospital because of progressive fatigue and fever for 1 week. The patient had no splenomegaly, hepatomegaly, or bruising. Initial laboratory evaluation revealed a white blood cell (WBC) count of 27.93 × 10⁹, including 14% promyelocytes, a hemoglobin level of 7.7 g/dL, and a platelet count of 135 × 10⁹/L, with no evidence of bruising or coagulopathy (prothrombin time, 15.2 s; activated partial thrombin time, 59.6 s; fibrinogen, 3.24 g/L). The bone marrow was hypercellular with 87.5% promyelocytes, which had regular nuclei and did not contain any Auer rods. The peroxidase staining was strongly positive, suggesting APL (Figure 1(a)). The immunophenotype showed an expanded population (70.98% of total WBCs) of immature myeloid cells positive for CD117, CD33, CD13, CD38, D56, and MPO and negative for CD34 and HLA-DR. Dual-color fluorescence in situ hybridization (FISH) showed a 70% split signal, indicating rearrangement of RARA (Figure 1(b)). Real-time polymerase chain reaction (RT-PCR) analysis detected FIP1L1-RARA fusion instead of PML-RARA. Cytogenetic analysis performed on short-term cultures from bone marrow samples showed the following karyotype: 46,XX,t(1;17;4)(p36;q21;q12),der(9) in 20 metaphases (Figure 1(c)). Mutational analysis of myelogenous leukemia-related genes using next-generation sequencing (NGS) showed an NRAS mutation (exon2; c.38G > A; p.G12D with a 1.2% variant allele frequency). Diagnosis of a variant of APL was made according to morphology, FISH, RT-PCR, and flow cytometric analysis.

Figure 1. (a) Bone marrow aspirate showed promyelocytes with regular nuclei and no Auer rods. (b) Dual-color fluorescence in situ hybridization (FISH) with 3′-RARA (green signal) and 5′-RARA (red signal), was performed for nuclei of 500 leukemia cells in interphase. The normal RARA gene shows as a superimposed yellow or red-green signal. FISH showed a 70% split signal, indicating rearrangement of RARA. (c) Karyotype: 46,XX,t(1;17;4)(p36;q21;q12),der(9)[20]. (d) Sequencing analysis indicating a FIP1L1-RARA fusion transcript using reverse transcription polymerase chain reaction.

To identify the FIP1L1-RARA fusion transcript, reverse transcription polymerase chain reaction (RT-PCR) analysis was performed on random-primed cDNA of RNA, separated from bone marrow cells using two pairs of primers, including a 5′-FIP1L1 forward primer (5′-TGATTCCACCACCGGGTTTT) and 3′-RARA reverse primer (3′-TGTAGATGCGGGGTAGAGGG) and another 5′-FIP1L1 forward primer (5′-GACGGGCAAATGAGAACAGC) and 3′-RARA reverse primer (3′-TGACCCCATAGTGGTAGCCT). Sequencing analysis indicated an in-frame FIP1L1-RARAfusion in exon 15 of FIP1L1 and exon 3 of RARA (Figure 1(d)).

Remission induction therapy was commenced with all-trans retinoic acid (ATRA, 20 mg/m²/d, divided into two doses per day), arsenic trioxide (ATO, 0.16 mg/kg/d), daunorubicin (45 mg/m²/d from days 1 to 3), and cytosine arabinoside (Ara-C, 150 mg/m²/d from days 1 to 7). On day 14, bone marrow aspiration revealed 2.5% promyelocytes; therefore, ATRA and ATO were administered for 28 days. Complete remission was documented after induction, with minimal residual disease undetectable by flow cytometry and FISH (Figure 2). A set of consolidation therapies was administered as follows: first course, daunorubicin 45 mg/m² for three days and Ara-C 150 mg/m² for seven days; second course, homoharringtonine 2 mg/m² for five days and Ara-C 150 mg/m² for seven days; and third course, idarubicin 12 mg/m² for three days and Ara-C 150 mg/m² for seven days. The patient then started a maintenance phase with ATRA and ATO. Twenty months after the diagnosis, the patient relapsed with a gradual reduction in her WBC counts. A bone marrow sample showed that 80% of the cells were promyelocytes with recurring chromosomal translocation, and the immunophenotype showed 59.31% promyelocytes with aberrant expression of CD117 and CD9. However, the coagulation was mildly abnormal, and no hemorrhage occurred. NGS showed two single-nucleotide polymorphisms, which were not considered to be significantly related to myeloid hematological tumors with the disappearance of NRAS mutations. A second remission induction therapy was administered consisting of ATRA (20 mg/m²/d) and ATO (0.16 mg/kg/d) for 21 days, but no response was observed. Therapy of homoharringtonine 2 mg for seven days, Ara-C 100 mg for seven days, and venetoclax 200 mg for 14 days was then given to induce remission, but unfortunately, after the suppression period, the bone marrow proliferated rapidly and the patient died of cerebral hemorrhage.

Figure 2. No split signal of RARA was observed.

Only three cases of APL with FIP1L1-RARA have been reported. Kondo et al. first reported one case of a 90-year-old woman diagnosed with APL who achieved complete remission after oral administration of ATRA alone (50 mg/d), but no follow-up information could be obtained. Then, Menezes et al. reported a second case of a 77-year-old female APL patient who died just 10 days after treatment, probably due to ATRA syndrome. Therefore, the response to ATRA treatment and the prognosis could not be assessed. The third patient was a 28-month-old infant diagnosed as APL with FIP1L1-RARA and concurrent myeloid sarcoma. The patient was treated with ATRA and chemotherapy and was alive after a 5-month follow-up [8–10]. Besides, Buijs and Bruin found the FIP1L1-RARA fusion gene in a case of JMML [7]. Nakanishi et al. reported a case of variant APL with t(4;17)(q12;q21), but failed to confirm the FIP1L1-RARA fusion gene in a 63-year-old man, with atypical APL morphology, demonstrated M2 and M3 clones. Although the M3 clone disappeared by using ATRA, 45% myeloblast remained in bone marrow after salvage therapy, and the patient died 8 months after diagnosis [11]. FIP1L1 is associated with two leukemogenic fusion genes: FIP1L1-PDGFRA and FIP1L1-RARA. Although the different contributions of FIP1L1 to the pathogenesis of distinct types of leukemia have been well studied [12], the reason for the two different phenotypes of leukemia (JMML and APL) caused by FIP1L1-RARA remains unknown. The fusion genes described above were generated between exons 15, 13, and 12 of FIP1L1 and exon3 of RARA, similar to our case. FIP1L1-RARA forms homodimers through the FIP1L1 portion and suppresses RA-dependent transcriptional activity, but this effect is reversed by ATRA. Therefore, FIP1L1-RARA APL is an ATRA-sensitive entity, and this result could be verified in vivo in our case.

This is the fifth case of FIP1L1-RARA translocation in APL, but the only adult case with relatively long-term survival and different characteristics from those of typical APL. Although the patient was sensitive to ATRA, relapse occurred during the maintenance phase of ATRA and ATO. In a review by Testa et al. immunophenotypes (especially CD56) in APL might play a role in prognosis and are associated with clinical characteristics [13]. CD56 is overexpressed in approximately 10% of APL cases and is often related to a high WBC count [14]. CD56-positive APL is a rare subtype of PML-RARA, PLZF-RARA, or other unknown fusion genes. CD56-positive APL fails to differentiate in vitro in response to ATRA, and the prognosis of CD56-positive APL is poorer than that of classical APL. In addition, its expression has been associated with increased relapse risk in a number of studies [15]. Our patient was CD56 positive and could be classified as having a CD56-positive APL subtype, in agreement with the high WBC count and resistance to repeated standard therapy with ATRA and ATO. Notably, the NRAS mutation may contribute little to recurrence as it disappears upon relapse, and allogeneic stem cell transplantation is needed to eradicate leukemia cells for long-term disease-free survival.

FIP1L1 is located at 4q12, and t(4;17)(q12;q21) generates the FIP1L1-RARA fusion gene. Interestingly, our patient harbored a complex three-way translocation t(1;17;4)(p36;q21;q12) and an additional aberrant chromosome der9. The three-way translocation has been reported in several cases, and additional chromosomal abnormalities occur in approximately 25–40% of APLs. From these studies, neither complex translocations nor additional chromosomes affected remission rates or survival in patients treated with ATRA. However, the influence of these aberrations in variant APLs needs to be clarified in more cases.

In conclusion, the remarkable features of this case were the regular-shaped nuclei without Auer rods and mild coagulation abnormalities. Disseminated intravascular coagulation did not occur in either of the cases reported in the literature or in our case. To our knowledge, the long-term survival of adult APL patients with FIP1L1-RARA has not yet been described. Owing to the above studies, which were restricted to typical APL, the high rate of relapse and standard therapy deserve further research.

Author contributions

X.F. conceived the case report and wrote the manuscript; Y.Z. and J.X. gave the treatment guidance; all the authors provided advice and approved the final manuscript.

Acknowledgements

The authors thank all of the doctors and nurses in Department of Hematology for their professional assistance.

Disclosure statement

All authors have no conflicts to declare.