SQSTM1::RARA fusion in acute promyelocytic leukemia: the first case report and literature review

Acute promyelocytic leukemia (APL) with the typical PML::RARA fusion gene caused by t(15;17)(q24;q21) has been distinguished from other types of acute myeloid leukemia (AML).¹ However, a few cases that cannot be identified as having a PML::RARA fusion gene by conventional methods have abnormal promyelocytes that are fully in accordance with the morphology, cytochemistry, and immunophenotype of APL.^2,3 In this report, we describe SQSTM1::RARA and reciprocal RARA::SQSTM1 fusions, identified for the first time in a patient lacking the t(15;17)(q24;q21)/PML::RARA fusion. This patient proved resistant to induction therapy based on all-trans retinoic acid (ATRA). To date, a comprehensive literature review has identified 21 novel RARA fusion partner genes distinct from PML::RARA. The SQSTM1::RARA reported in our study represents the 22^nd such fusion. Among these variant fusions, ZBTB16::RARA, TTMV::RARA, and STAT5B::RARA are the most frequently reported and are associated with clinical features that differ from those of PML::RARA-positive APL.

The index case was an 82-year-old man initially admitted to a local hospital because of persistent fatigue for over 10 days. His clinical course is summarized in Figure 1A. Blood tests showed a white blood cell count of 0.66x10⁹/L, and absolute neutrophil count of 0.07x10⁹/L, hemoglobin level of 93 g/L, and platelet count of 93x10⁹/L. Bone marrow (BM) examination revealed abnormal promyelocytes accounting for 88.5% of the cellularity. These cells were characterized by round nuclei and abundant small azurophilic granules, but no Auer rods were observed (Figure 1B). Flow cytometry identified 85.8% myeloblasts with an immunophenotype positive for CD13, CD33, CD45, CD117, CD64, and MPO, while negative for CD34, CD11b and HLA-DR. Notably, the absence of CD38 expression presented a distinctive feature compared to classic PML::RARA-positive APL (Figure 1C). The presumptive initial diagnosis of this patient was APL. However, both reverse transcription polymerase chain reaction and fluorescence in situ hybridization (FISH) using the standard PML::RARA dual-color dual-fusion probe failed to detect the classic PML::RARA fusion. Instead, FISH analysis revealed an atypical signal pattern in 71.5% of cells, characterized by three RARA signals (red) and two PML signals (green) (Figure 1D). In the context of the karyotype findings, this pattern suggests that the t(5;17) breakpoint did not split the RARA locus. Instead, the intact RARA gene was translocated to the derivative chromosome 5. The presence of an extra derivative chromosome 5, as confirmed by karyotyping, accounts for the three RARA signals observed. The karyotype analysis confirmed this: 46,X,-Y,t(5;17)(q35;q21),der(5)t(5;17),+8[2]/46, id em,r(1),add(12)(p13)[14]/46,XY[4] (Figure 1E). Whole transcriptome sequencing was subsequently performed, revealing a novel SQSTM1::RARA fusion gene involving SQSTM1 (5q35) and RARA (17q21). The fusion results from the joining of exon 5 of SQSTM1 to exon 3 of RARA. Bioinformatic analysis of the sequencing data confirmed that this fusion is in-frame and is predicted to generate a functional chimeric protein. The reciprocal RARA::SQSTM1 fusion was also detected (Figure 2A). Whole transcriptome sequencing analysis revealed downregulated CD38 transcript levels and upregulated BCL2 transcript levels (Figure 2B). These findings may explain the absence of surface protein detection by flow cytometry and the marked sensitivity to subsequent venetoclax therapy. Furthermore, comparative transcriptomic analysis between our patient, patients with classical APL, and patients with non-APL AML identified numerous differentially expressed genes. Enrichment analysis demonstrated that, compared to patients with classical APL and those with non-APL AML, our patient exhibited upregulation of multiple pathways related to protein translation (Figure 2C). We subsequently validated this result using polymerase chain reaction and Sanger sequencing (Figure 2D, E). Additionally, next-generation sequencing mutation analysis revealed TET2 and BRCA2 mutations. TET2 p.Y592Lfs*46 was detected at a variant allele frequency (VAF) of 41.0%; this mutation induces premature translational termination, generating a truncated protein. A BRCA2 p.P512R missense mutation, with a VAF of 46.5%, was also found.

Given the high clinical suspicion of APL, the patient promptly initiated dual induction therapy. The specific regimen included arsenic trioxide (ATO) at a dose of 0.15 mg/kg daily (administered as 10 mg/day) and ATRA at a dose of 25 mg/ m² daily (administered as 20 mg twice daily). The patient’s body mass index was 23.4 kg/m². Steroid prophylaxis was not administered during the induction phase. However, ATO and ATRA were discontinued on day 5 due to a negative PML::RARA fusion gene result and a poor early response showing no signs of differentiation. A follow-up BM examination 33 days after induction chemotherapy showed a predominance of promyelocytes (88%) with 1% myeloblasts, and MPO staining was strongly positive. The patient was then started on a regimen of venetoclax (100 mg orally on day 1) and azacitidine (100 mg subcutaneously). On the first day of dosing, the patient developed dyspnea and decreased oxygen saturation and was transferred to the Intensive Care Unit. Two weeks later, a repeat BM aspirate showed 84.5% of promyelocytes. The patient subsequently began a 9-day course of combination therapy with venetoclax and ATRA. The venetoclax dose was escalated as follows: 50 mg daily (days 1-3), 100 mg daily (days 4 and 5), and 200 mg daily (days 6-9). Concomitant ATRA was maintained at 20 mg twice daily throughout this period (days 1-9). During treatment, the patient experienced adverse events – including fever, dyspnea, pleural effusion, and renal dysfunction – which prompted the concurrent administration of intravenous dexamethasone (5 mg daily from day 1 to day 7). On day 10, the patient refused to continue oral venetoclax and ATRA therapy. Management was subsequently limited to supportive care only. Forty-three days after discontinuation of therapy, a blood test at our institution revealed a white blood cell count of 10.36x10⁹/L, hemoglobin of 86 g/L, and platelet count of 253x10⁹/L. A BM morphology examination showed only 1% promyelocytes. Most notably, an assessment of minimal residual disease by flow cytometry was negative. The patient was subsequently transferred to the Urology Department for a minimally invasive procedure to address benign prostatic hyperplasia. The patient presented to our hospital for a follow-up visit 1 year later and a BM smear showed relapse. RNA sequencing of the BM aspirate detected the SQSTM1::RARA and the reciprocal RARA::SQSTM1 fusions, with exon breakpoints identical to those identified at the initial diagnosis. This finding confirms the role of SQSTM1::RARA as the driver gene in this case.

Figure 1.Clinical, morphological, immunophenotypic, and cytogenetic features of the index case at diagnosis. (A) Clinical course of the patient. (B) High-power view of the bone marrow aspirate smear at initial diagnosis (Wright-Giemsa stain; original magnification x1,000). The leukemic cells exhibit round nuclei and abundant small azurophilic granules, consistent with hypergranular promyelocytes. Note the absence of Auer rods. (C) Flow cytometry analysis of the bone marrow aspirate at initial diagnosis. Representative scatter plots show the distinct clustering of the abnormal cell population (highlighted in red), distinguishable from internal normal lymphocytes (green population) and erythroblasts (gray population). Standard quadrant or boundary gates demarcate positive and negative populations, with thresholds defined by internal controls. (D) Fluorescence in situ hybridization analysis of bone marrow cells using a PML (15q24, green)/RARA (17q21.1, red) dual-color, dual-fusion probe. Left panel. A normal control interphase nucleus exhibiting the expected wild-type pattern of two PML and two RARA signals (2G2R). Right panel. A representative interphase nucleus from the patient exhibiting an atypical signal pattern consisting of two PML signals and three RARA signals (2G3R) (original magnification x1,000). (E) G-banded karyotype of the bone marrow aspirate at initial diagnosis. A complex abnormal karyotype was identified: 46,X,-Y,t(5;17)(q35;q21),der(5)t(5;17),+8[2]/46,idem,r(1),add(12)(p13)[14]/46,XY[4]. APL: acute promyelocytic leukemia; ATO: arsenic trioxide; ATRA: all-trans retinoic acid; AZA: azacitidine; ICU: intensive care unit; VEN: venetoclax.

Figure 2.Molecular and transcriptomic characterization of the SQSTM1::RARA and reciprocal RARA::SQSTM1 fusions. (A) Schematic representation of the SQSTM1::RARA and reciprocal RARA::SQSTM1 fusions detected by whole transcriptome sequencing in bone marrow aspirate smears at initial diagnosis and relapse. (B) Median BCL2 expression (log₂(TPM+1)) in our patient, TCGA classical APL (N=14), TCGA non-APL-AML (N=137), and a GTEx healthy control group (N=372). (C) Volcano plots of differentially expressed genes in our patient versus classical APL (top left) and non-APL AML (bottom left). Bar charts of upregulated pathways versus those in classical APL (top right) and non-APL AML (bottom right). (D) Detection of the partial SQSTM1::RARA (left, 615 bp) and RARA::SQSTM1 (right, 629 bp) fusions transcripts in cDNA. The specific primer sequences used were: SQSTM1::RARA: 5’-CTTGTGTAGCGTCTGCGAGGGAAAG-3’ (forward) and 5’-CGGTCGTTTCTCACAGACTCCTTGG-3’ (reverse); RARA::SQSTM1: 5’-CATGGCCAGCAACAGCAGCTC-3’ (forward) and 5’-GCATCTGGGA-GAGGGACTCAATCAG-3’ (reverse). (E) Sanger sequencing validation of SQSTM1-exon 5 fused to RARA-exon 3 (left), and the reciprocal RARA-exon 2 fused to SQSTM1-exon 6 (right) in a bone marrow aspirate at initial diagnosis. AML: acute myeloid leukemia; APL: acute promyelocytic leukemia; GTEx: Genotype-Tissue Expression; TCGA: The Cancer Genome Atlas; TPM: transcripts per million.

The human SQSTM1 gene, located on chromosome 5, spans approximately 16 kilobases and consists of eight exons.^4,5 It encodes the multifunctional p62/SQSTM1 protein, which is primarily recognized as a selective autophagy receptor that targets cellular components for lysosomal degradation.⁶ Beyond its role in autophagy, p62 functions as a pivotal signaling hub due to its capacity to interact with diverse partners. It critically regulates key pathways including Nrf2-mediated oxidative stress response,⁷ mTORC1-dependent nutrient sensing,⁸ and NF-κB-driven inflammation.⁹ Notably, p62 dysregulation is implicated in tumorigenesis; gene copy gains in the 5q locus are linked to clear cell renal cell carcinoma, partly through p62 overexpression,¹⁰ and p62 accumulation is a common feature observed in numerous cancer types.^7,11,12

This case presents a unique therapeutic trajectory that offers insights into venetoclax biology. The patient showed clear resistance to conventional ATRA/ATO therapy, consistent with the absence of the PML::RARA rearrangement. However, the rapid development of dyspnea, hypoxemia, and renal dysfunction within 24 hours of a single dose of 100 mg venetoclax – in combination with azacitidine – strongly suggests incipient tumor lysis syndrome, providing compelling in vivo evidence of exceptional venetoclax sensitivity. Despite this early toxicity, subsequent treatment with low-dose venetoclax (escalated to 200 mg) combined with ATRA for only 9 days was sufficient to induce deep remission. Most remarkably, 43 days after all therapy had been discontinued, BM examination revealed only 1% promyelocytes with negative minimal residual disease by flow cytometry. This durable response following limited treatment exposure suggests that even brief BCL-2 inhibition may trigger sustained leukemic clearance in genetically susceptible subsets. We hypothesize that the novel genetic abnormality identified in this case may confer profound BCL-2 dependence, rendering the clone exquisitely sensitive to venetoclax despite resistance to standard APL-directed therapy. The concurrent TET2 and BRCA2 pathogenic variants may provide additional context for this patient’s atypical course. TET2 mutations are associated with clonal hematopoiesis and enhanced sensitivity to hypomethylating agents such as azacitidine, which the patient received. The BRCA2 variant, involving the homologous recombination repair pathway, may contribute to genomic instability and facilitate accumulation of multiple genetic lesions. Together, these variants highlight the value of comprehensive genomic profiling in understanding complex treatment responses. Unlike the classic PML::RARA fusion, which is highly sensitive to ATRA and ATO, most rare X::RARA variants are known to confer resistance to standard differentiation therapy. Clinical management of these resistant cases remains challenging; however, current literature suggests potential benefits from intensive chemotherapy or novel targeted agents. Notably, BCL-2 inhibitors such as venetoclax and the antibody-drug conjugate gemtuzumab ozogamicin have emerged as promising salvage therapies. Beyond therapeutic responses, distinct clinicopathological features have been observed in specific variants. TTMV::RARA cases show a higher incidence in juveniles (<18 years: 75.8%), while FIP1L1::RARA occurs primarily in young children (≤3 years: 62.5%) with a female predominance (75%). Additionally, all reported BCOR::RARA cases were male and exhibited loss of the Y chromosome. From a diagnostic perspective, whole transcriptome sequencing is instrumental in discovering novel fusions. For rare fusions, such as SQSTM1::RARA, which show suboptimal responses, early molecular identification and transition to alternative potent therapies are crucial for improving outcomes. The characteristics of the 22 X::RARA cases are summarized in Table 1.

Table 1.Characteristics of 22 X::RARA fusions.

Ethical review and approval were not required for this case report in accordance with local legislation and institutional requirements. The study was conducted in accordance with the ethical standards of the People’s Republic of China. Written informed consent was obtained from the patient for publication of this case report and accompanying images.

Footnotes

• Received December 4, 2025
• Accepted April 10, 2026

Correspondence

*XY and NL contributed equally as first authors

H-H. ZHU - zhuhhdoc@163.com

Disclosures

No conflicts of interest to disclose.

Contributions

Funding

This work was supported by the National Natural Science Foundation of China (82450101, 82370169), the “Dengfeng” Talent Training Program of Beijing Hospitals Authority (DFL20240301), and funds from the Chinese Institutes for Medical Research, Beijing (CIMR) to H-HZ.