Comment on Geoffroy, M.-C.; de Thé, H. Classic and Variants APLs, as Viewed from a Therapy Response. Cancers 2020, 12, 967

Acute promyelocytic leukemia (APL) is a unique and very deeply studied acute myeloid leukemia [...].

Acute promyelocytic leukemia (APL) is a unique and very deeply studied acute myeloid leukemia. The typical molecular genetic feature of APL is the rearrangement of RARA gene with PML and a few with other partner genes. All-trans retinoic acids and arsenic trioxides are effective agents against APL harboring PML-RARA. However, there are still some rare cases lacking RARA fusion although they share the same morphology and immunocytochemistry features with typical APL. The classification of such cases remains controversial. They are usually referred to as acute promyelocytic-like leukemias (APLL), and sometimes they were morphologically classified as M3 according to the FAB classification. They are not regarded as APL in the WHO's classification of myeloid neoplasms and acute leukemia. For years, the underlying molecular pathology of APL-like leukemia has remained concealed. Some researchers supposed that APL-like leukemia might bear genetic mutation of other members of retinoic acid receptors (RARs), such as RARB or RARG [1].
Due to the limitations of technology, identification of unknown fusion genes has been a relatively difficult task. In recent years, the emerging high-throughput sequencing technologies have provided powerful research tools for the discovery of novel fusion genes. In 2011, a novel NUP98-RARG fusion gene was discovered by Such E et al. in a Spanish case of APL-like leukemia, and it was the first reported fusion gene of this kind of leukemia. Since 2017, several other novel fusion genes harboring RARG have been found. Geoffroy et al. made a relatively comprehensive review in this field. In this review, a total of four RARG-related fusion genes are summarized, namely CPSF6-RARG & RARG-CPSF6, NUP98-RARG, PML-RARG, and NPM1-RARG-NPM1. NUP98-RARG accounts for a total of three cases and CPSF6-RARG & RARG-CPSF6 for five cases. CPSF6-RARG & RARG-CPSF6 own the most variants, consisting of four types of CPSF6-RARG and one type of RARG-CPSF6 [2]. In fact, another patient with CPSF6-RARG has been reported in 2019 [3]. Since then, two more cases of CPSF6-RARG and two cases of NUP98-RARG have been reported [4][5][6][7]. Therefore, the number of reported patients has reached eight cases for CPSF6-RARG and six for NUP98-RARG. Their biological and clinical information are summarized in Tables 1 and 2, respectively.   [7] No., number; ATRA, all-trans retinoic acid; ATO, arsenic trioxide; M, male; F, female; auto-PBSCT, autologous peripheral blood stem-cell transplant; IA, idarubicin and cytarabine; UCBT, umbilical cord blood transplantation; CA, aclamycin and cytarabine; HAA, homoharringtonine, aclarubicin, and cytarabine; DA, daunorubicin and cytarabine; HIAG, homoharringtonine, idarubicin, cytarabine, and granulocyte colony-stimulating factor(GCSF); half-CAG, half-dose aclarubicin, cytarabine, and GCSF; HA, homoharringtonine and cytarabine; HD-AC, high dose cytarabine; NR, no response; CR, complete response; PR, partial response.
Actually, RARG-related fusion genes that have been identified were not merely the four types mentioned. Su et al., reported a 43-year-old man with leukocytosis, anemia, and thrombocytopenia. Similarly, the morphology and immunological characteristics of the blasts were the same as APL. Transcriptome sequencing identified a new HNRNPC-RARG fusion gene and its reciprocal. The patient received ATRA induction therapy for three weeks, but the bone marrow blasts were almost unchanged. Then, combined chemotherapy (homoharringtonine, daunorubicin and cytarabine, HAD) was applied and complete remission was achieved. After that, he was given several courses of consolidation and intensive chemotherapy consecutively, although he relapsed one year later. Re-induction therapy medications included ATRA, arsenic trioxide, and combination chemotherapy, which, as inu other such cases, have not worked [16].
The RARG gene is located at chromosome 12q13. Several more APL-like cases with chromosome translocation involving 12q13 have been reported in previous pieces of literature, but further molecular detections were not performed (Table 3). Due to the fact that this kind of disease is relatively rare and due to the current limitations of research conditions, there may be more such cases unpublished. Actually, we also know that this situation does exist. As exhibited in Table 3, most cases are t(11;12) translocations, which are consistent with the location of the NUP98-RARG fusion gene. One of them is t(4;12) translocation, suggesting that there may be undiscovered new fusion genes. These cases also did not respond to the treatment of retinoic acid or arsenic trioxide.  Interestingly, there is another group of AML with t(11;12)(p15;q13), of which the cell morphology is not promyelocyte type, but classified into M1, M2, M4, and M5 according to FAB classification. In the fusion gene of such cases, one partner gene is also NUP98 located on chromosome 11p15, and the other is one of HOXC (HOXC11, HOXC13) gene clusters [18,21]. It should be noted that a series of genes are densely distributed in the region of human chromosome 12q13, including RARG and HOXC.
All reported cases of APLL with 12q13 rearrangement achieved no response to ATRA or arsenic trioxide treatment in vivo. In vitro culture of bone marrow blast cells from the first patient with NUP98-RARG also confirmed the failure, although hematopoietic stem cells transfected with NUP98-RARG were demonstrated to be sensitive to ATRA. Some researchers attributed this contradiction to the influence of the in vitro culture environment or additional mutations [22]. In fact, a series of RARs modulators have been synthesized, some of which are pan-antagonists/agonists and some of which are selective ones. ATRA is one of pan-agonists [23]. Researchers have observed that BMS961 (an RARG agonist) has not shown any effect on leukemia cells with RARG-CPSF6 cultured in vitro [10].
In most of the published reports and WHO classification, as mentioned earlier, APLL is regarded more as a subclass of AML than as an APL. However, the authors of this review pointed out in another article [24] that RARB and RARG are also retinoic acid receptors and play a key role in leukemogenesis through their rearrangement (similar to RARA). Moreover, APLL carrying RARG-CPSF6 shared the same gene expression profiles with classical APL harboring PML-RARA [10]. Although APLL is not responsive to ATRA, new target drugs (e.g., other RARs modulators) may be explored to improve the outcome of APLL patients. Thus, we think it seems to be reasonable to classify the APL and APLL into a same subclass and the APL with RARs rearrangements.
However, there have existed cases of APLL lacking RARs fusion genes. Researchers further explored other possible molecular abnormalities (e.g., dysregulation of RARA and RARG has been reported) [25,26]. Zhao et al., identified two new fusion genes NPM1-CCDC28A and TBC1D15-RAB21 in two cases of children without RARs rearrangement. The abnormal chromosome of the third patient was t(1;11)(q21;q23), and DNA sequencing identified two fusion genes, KMT2A-MLLT11 and RPRD2-KMT2A [3]. In addition, sporadic APLL literature could be retrieved which did not carry out in-depth study of molecular pathology, and some involved known genetic aberration such as JAK2 V617F mutation [27].
In conclusion, the identification of RARG/RARB related fusion genes over the last decade has revealed the mystery of APLL, and it can be predicted that more cases will be reported in the future. However, due to the insensitivity to ATRA and arsenite acid, there is still a lack of effective targeted drugs. In addition, there are some cases lacking RARs rearrangements, of which the molecular pathology remains to be clarified.