Editor’s Note: On August 21, 2026, the 14th Lu Daopei Hematology Forum was successfully held at the new campus of Hebei Yanda Lu Daopei Hospital. After 14 years of continuous development, the Lu Daopei Hematology Academic Conference has become an influential platform for academic exchange in the field of hematology in China, witnessing the evolution of hematologic disease diagnosis and treatment from accumulated clinical experience toward evidence-based innovation. During the conference, Oncology Frontier – Hematology Frontier invited Dr. Xiaosu Zhou from the Beijing Lu Daopei Institute of Hematology to share the latest research findings on atypical acute promyelocytic leukemia (APL), focusing on its fusion gene landscape, underlying pathogenesis, and prospects for clinical translation.
Oncology Frontier – Hematology Frontier: Your presentation at this conference focused on “Fusion Genes and Pathogenesis of Atypical APL: Far Greater Complexity and Heterogeneity Than Expected.” Could you introduce the key genetic differences between atypical APL and classical APL?
Dr. Xiaosu Zhou: Acute promyelocytic leukemia (APL) is a distinct subtype of leukemia characterized by unique morphological and clinical features. Early studies found that approximately 90% of patients share the PML::RARA fusion gene, establishing this fusion as the key molecular abnormality driving the development of APL in these patients. Promyelocytic leukemia with a PML::RARA fusion is therefore also referred to as classical APL.
Thanks largely to the pioneering contributions of Chinese experts, including Academician Zhenyi Wang, Professor Tingdong Zhang, Academician Zhu Chen, and Academician Saijuan Chen, arsenic-based therapy and all-trans retinoic acid (ATRA) were found to be highly effective against classical APL, with the combination of the two producing even better outcomes. These advances have transformed classical APL from one of the most aggressive and life-threatening leukemias into the leukemia subtype with one of the highest cure rates. More than 90% of patients can now achieve clinical cure with treatment regimens centered on arsenic-based therapy and ATRA.
However, the approximately 10% of patients who are negative for the PML::RARA fusion gene, although their cellular morphology and even transcriptomic characteristics may resemble those of classical APL, exhibit markedly different and highly heterogeneous clinical responses to treatment.
Some patients still show a degree of response to ATRA, whereas a substantial proportion are completely resistant to ATRA. Moreover, most of these patients are resistant to arsenic-based therapy, and some may even be resistant to intensive chemotherapy.
Because these diseases can progress rapidly and patients are at high risk of early death, rapid and accurate diagnosis, together with timely assessment of their likely responses to arsenic, ATRA, and other therapies, has become a critical clinical challenge.
Although previous studies have reported dozens of rare retinoic acid receptor fusion genes, they have not adequately addressed these clinical challenges. In response to the diagnostic and therapeutic difficulties presented by cases encountered in clinical practice, our team has conducted a series of studies on atypical acute promyelocytic leukemia (APL) in recent years. We have led several nationwide multicenter collaborative studies and generated a series of internationally original findings.
Through these studies, we have gradually come to recognize that the molecular pathogenesis of the approximately 10% of atypical cases is far more complex than that of the 90% of classical cases. This is why I used the phrase “far greater complexity and heterogeneity than expected” in the title of my presentation.
Our team’s original findings in this field include the characterization of three-part retinoic acid receptor fusion genes and elucidation of how these novel fusion structures contribute to ATRA resistance; the discovery that two-part retinoic acid receptor fusions accompanied by cis mutations can drive the development of atypical APL and primary ATRA resistance; and the finding that fusion of the ORF2 gene of small anelloviruses with the RARA gene represents the most common molecular abnormality among atypical cases. However, these fusions had previously gone undetected because of the high degree of sequence variation in viral genomes. We have also newly discovered that anelloviruses can fuse with RARA and cause atypical APL.
These findings challenge the previous assumption that fusion genes are invariably formed through a simple two-part fusion. They demonstrate a much greater structural complexity of fusion genes and correct misconceptions that had persisted in this field for nearly three decades.
Anelloviruses and small anelloviruses are commonly carried by healthy individuals and have traditionally been regarded as commensal viruses. Our studies have provided initial insights into the key molecular mechanisms through which these viruses may contribute to rare leukemias, although considerable work remains to fully elucidate these mechanisms. The discovery that viral genes can fuse with human genes to drive tumor development represents, to our knowledge, a previously unrecognized mechanism of viral pathogenesis, further expanding our understanding of the relationship between viruses and human disease.
Oncology Frontier – Hematology Frontier: Why is the pathogenesis of atypical APL so complex and heterogeneous? What challenges do the discovery of these rare fusion genes pose for the diagnosis and treatment of APL?
Dr. Xiaosu Zhou: It is only through our research over the past several years that we have gradually come to appreciate the complexity and heterogeneity of the pathogenesis of atypical APL.
In healthy adults, the body needs to generate hundreds of billions of mature granulocytes every day to replenish cells that are continuously consumed. The differentiation of hematopoietic stem cells into mature granulocytes is a highly ordered process. A common feature of patients with APL is that hematopoietic cells become arrested at the promyelocyte stage and are unable to continue differentiating toward maturity. Consequently, large numbers of morphologically abnormal promyelocytes accumulate in the bone marrow and peripheral blood, which represents a shared characteristic of APL.
The primary reason for this differentiation arrest is dysregulated expression of the gene sets controlled by retinoic acid receptor signaling, which are essential for hematopoietic differentiation at this stage.
Therefore, although both classical and atypical APL share the presence of retinoic acid receptor gene fusions, the retinoic acid receptor family consists of three genes—RARA, RARB, and RARG—and dozens of fusion partner genes have already been identified. In addition, the structural diversity arising from the key three-part fusions and cis mutations identified by our team, together with the involvement of viral genes, has further expanded the diversity of retinoic acid receptor fusion genes.
This increasingly complex molecular landscape explains why atypical APL cannot simply be regarded as a small collection of molecular variants of classical APL. Instead, it represents a highly heterogeneous group of diseases with distinct molecular mechanisms and potentially different therapeutic sensitivities, creating substantial challenges for accurate diagnosis and treatment selection.
We can think of the retinoic acid receptor as a lock that regulates gene expression. This lock is targeted to the promoters of genes that it needs to regulate, acting like a switch. All-trans retinoic acid (ATRA) functions like a key. Under normal conditions, when these genes need to be expressed, cells regulate the concentration of ATRA, which acts on the retinoic acid receptor and switches on gene expression, allowing hematopoietic cells to differentiate smoothly into more mature stages.
We now understand that the key molecular cause of APL is that the retinoic acid receptor “lock” remains continuously locked and can no longer be opened by the ATRA “key.” As a result, this group of genes cannot be activated, preventing the cells from progressing to the next stage of differentiation.
Imagine that a lock cannot be opened because, in 90% of cases, the keyhole has become rusty and is difficult to turn. Under these circumstances, the lock cannot be opened with a key using ordinary force, but applying greater force may still open it. This is analogous to classical PML::RARA-positive APL. Although the fusion protein reduces the cells’ responsiveness to ATRA, administering ATRA at pharmacologic concentrations far above physiological levels can still activate the expression of these genes, allowing leukemic cells to continue differentiating into mature granulocytes. This is the fundamental principle underlying differentiation therapy with ATRA in APL.
However, a lock can malfunction for many other reasons. For example, internal components may be damaged or jammed. As long as the malfunction prevents the lock from opening normally, the same outcome can occur.
The approximately 10% of patients with atypical APL can therefore be viewed, from a pathogenic perspective, as a collection of rare mechanisms that cause the “lock” to malfunction and become impossible to open. In many cases, dozens of different fusion partner genes can cause disease through distinct mechanisms, either by inducing abnormal expression of the fusion protein or by reducing the responsiveness of the retinoic acid receptor within the fusion protein to ATRA. In some cases, however, the fusion partner alone is insufficient; additional structural abnormalities—such as the formation of a three-part fusion gene or the acquisition of a cis mutation—may be required to produce a sufficiently strong pathogenic effect.
Precisely defining these complex molecular causes is critical for accurate diagnosis and treatment. Just as the key to opening a malfunctioning lock lies in accurately identifying the root cause of the defect, effective treatment requires precise identification of the underlying molecular abnormality.
The diversity of retinoic acid receptor genes, the functional diversity of their fusion partners, the structural diversity of fusion genes, and the high sequence variability of viral genes collectively contribute to the highly complex and heterogeneous molecular landscape of atypical APL. They also pose substantial challenges for research and fusion-gene identification.
Fortunately, with the application of transcriptomic technologies and optimization of bioinformatic analysis pipelines, we are gradually deciphering these complex molecular mechanisms and, based on these discoveries, developing convenient and effective diagnostic assays.
Oncology Frontier – Hematology Frontier: With advances in sequencing technologies continuously reshaping the genetic landscape of APL, what do you consider the key future research directions and prospects for clinical translation in this field?
Dr. Xiaosu Zhou: We will continue our research by building on the insights gained from our existing findings while addressing the questions that remain unresolved. We have already made several new discoveries, and further findings that we cannot currently anticipate may also emerge in the future. Therefore, the genetic landscape of atypical APL will continue to be updated, expanded, and characterized in greater detail.
The prospects for translating these findings into clinical practice are clear. First, based on our understanding of the complexity of retinoic acid receptor-related fusion gene sequences and structures, we are developing a series of diagnostic assays that can help patients receive faster and more accurate diagnoses.
Second, our mechanistic studies have established that patients harboring three-part fusion genes or fusion genes accompanied by cis mutations are resistant to ATRA. Therefore, these patients should not continue to receive ATRA and should instead be switched promptly to alternative treatment strategies. Conversely, for patients who can be identified as ATRA-sensitive, such as those with fusion genes involving small anelloviruses or anelloviruses, as well as certain specific two-part RARA fusion genes, ATRA treatment should still be actively pursued.
In addition, although the vast majority of patients with atypical APL do not respond to arsenic-based therapy, patients with PML::RARG fusion-positive APL can still achieve a favorable response to arsenic-based treatment.
Looking ahead, we hope that continued research will gradually transform atypical APL—currently regarded as “the most aggressive subtype of leukemia”—into a disease that can be precisely diagnosed, risk-stratified for treatment, and managed to improve patient outcomes, ultimately allowing more patients to benefit from these advances.
Expert Profile
Dr. Xiaosu Zhou
Beijing Lu Daopei Institute of Hematology
Associate Researcher
Center for Precision Medicine, Beijing Lu Daopei Institute of Hematology
Dr. Xiaosu Zhou received his PhD from Ocean University of China and completed his postdoctoral training at Peking Union Medical College.
After completing his postdoctoral training, he joined the Lu Daopei Molecular Medicine team, where he focused primarily on the development of molecular diagnostic technologies for hematologic malignancies and research related to immunotherapy. He subsequently joined the Institute of Hematology, where he is responsible for research into gene function and disease pathogenesis.
Dr. Zhou serves as a committee member of the Cancer Genetic Diagnosis Committee of the Chinese Anti-Cancer Association and a committee member of the Laboratory Medicine Committee of the Beijing Association of Non-Public Medical Institutions, among other professional organizations.
Academic Achievements:
Dr. Zhou has published several papers as first author in journals including Blood, American Journal of Hematology, and Haematologica. Some of his research findings have also been featured in expert commentaries published by these journals.
