Acute promyelocytic leukemia (APL) is a subtype of acute myeloid leukemia (AML) defined by the genetic translocation that forms the PML::RARA fusion gene between chromosomes 15 and 17 (1). This fusion disrupts gene transcription, halting myeloid differentiation at the promyelocytic stage (1, 2). APL accounts for approximately 10%–15% of all AML cases and is typically diagnosed through abnormal blood tests, along with coagulation and fibrinolytic dysfunction (3, 4). The incorporation of all-trans retinoic acid (ATRA) and arsenic trioxide (ATO) into treatment protocols has dramatically improved outcomes for APL patients, achieving a complete remission (CR) rate exceeding 90% (5, 6).

Myeloid sarcoma (MS), also known as granulocytic sarcoma or chloroma, is characterized by the extramedullary accumulation of myeloid blasts (7). It can occur as an isolated condition or in association with myeloid malignancies, particularly AML, and often signifies relapse following AML remission (7, 8). Although MS can affect individuals of all ages, it is more common in children than adults, with a male-to-female ratio of approximately 1.2:1 (7, 9, 10). MS/extramedullary infiltration is a rare complication of APL, affecting only 3%–5% of patients, typically during disease relapse post-remission (11, 12). The central nervous system and skin are the most common sites of extramedullary involvement, while other areas such as lymph nodes, the gastrointestinal tract, bones, soft tissues, and testes are less frequently affected (12, 13). Rare cases of APL-related EM infiltration at unusual sites have also been reported. Key factors associated with extramedullary involvement in APL include being under 45 years of age, elevated white blood cell count, and the presence of the bcr3 subtype of the PML::RARA fusion gene (14). The occurrence of APL with MS or EM infiltration as the sole initial presentation is extremely rare. Here, we present a case of APL-related MS manifesting as a lumbar epidural mass.

A 53-year-old male presented with 8 months of low back pain and weakness in both lower limbs. A CT scan at a local hospital revealed soft tissue shadows at the right posterior margin of the L2/3 intervertebral disc and in the spinal canal at the same level. MRI showed abnormal signals in the T12, L2, and S1 vertebral bodies, along with intraspinal soft tissue masses at the L2 pyramidal plane. Neoplastic lesions were suspected, and the patient received treatment with traditional Chinese medicine. Although there was initial improvement, his condition progressively worsened, leading to an inability to walk. PET/CT scans revealed uneven density in several vertebrae, with soft tissue shadows in the right portion of the L2 vertebra, extending into the right intervertebral foramen and inward into the spinal canal. There was slightly increased FDG uptake in the vertebral bodies and appendages, and active FDG metabolism was also noted in the spinal cord cavity from the T12-L2 segment (Figure 1). A follow-up MRI 1 month later showed multiple areas of bone destruction in the T2, L1, L2, and S1-3 vertebrae, suggesting metastatic involvement. Additionally, heterogeneous signal intensity in the spinal canal at the L1-3 level indicated possible involvement. The patient underwent surgical resection of the L2 vertebral body and the epidural mass, along with spinal fixation. Preoperative blood tests, including routine examinations and coagulation studies, were normal. Histopathological analysis of the resected tissue suggested a neoplastic tumor.

The case was referred to our pathology department for further consultation. Immunophenotyping results were as follows: CD34 (−), CD117 (+), MPO (+), CD20 (−), CD79a (−), CD3 (−), CD138 (−), CD38 (−), Mum-1 (−), CD56 (−), IgK (−), Igλ (−), and Ki-67 (+, approximately 60%). In situ hybridization for EBV showed no EBER1/2 expression. Gene rearrangement analysis by PCR and GENESCAN revealed no clonal amplification peaks for IgH or IgK. Based on these results, along with the morphological and immunophenotypic findings, MS was strongly considered. Postoperatively, the patient showed some improvement in low back pain and lower limb weakness, but remained unable to stand or walk. One month after surgery, the patient sought treatment at our hematology clinic. A bone marrow smear revealed significantly active marrow hyperplasia, but no blasts or abnormal promyelocytes were detected (Figure 2A). Flow cytometry showed abnormal promyelocytes with approximately 0.5% of nucleated cells, positivity for CD123, CD9, CD117, CD64, and CD33, but negativity for HLA-DR, CD11b, CD15, and CD56. These findings raised strong suspicion for APL-associated MS. Further tests confirmed our suspicion: PCR of peripheral blood was positive for PML::RARA, with a PML::RARA/ABL ratio of 0.9305%. Chromosomal analysis revealed 46,XY,t(15;17)(q24;q21)[2]/46,XY[18] (Figure 2B). Multiplex real-time PCR of bone marrow also showed positivity for PML::RARA (bcr-1). FISH analysis of the MS tissues revealed a 94% positivity rate for PML::RARA fusion signals (Figure 2C), while bone marrow FISH showed a 8% positivity rate for PML::RARA fusion at the 17q21/15q22-24 site, including 4% atypical signals (Figure 2D). Genetic testing revealed no mutations typically associated with AML prognosis at diagnosis. Routine blood tests, coagulation, and fibrinolysis remained normal, with no hepatosplenomegaly or systemic lymphadenopathy observed. Given these findings, the patient was diagnosed with acute promyelocytic extramedullary sarcoma (MS/APL).

The patient began treatment on the 48th day post-surgery, consisting of ATRA 10 mg three times daily and ATO 10 mg intravenously once a day for 30 days. During treatment, the patient developed mild ATRA syndrome, including fever, facial edema, and weight gain, which were managed with dexamethasone and furosemide. In addition, the expression level of the PML::RARA fusion gene reached the highest value on day 24 of induction treatment with a PML::RARA/ABL ratio of 16.3309%. After the first cycle of chemotherapy, the patient’s condition improved significantly. He was able to stand and walk independently with a brace, and his lumbar pain was greatly reduced. Bone marrow smears revealed no blasts, with promyelocytes comprising 1% of nuclear cells. Both peripheral white blood cell and blood cell counts normalized, indicating CR. However, the PML::RARA fusion gene remained detectable, with a PML::RARA/ABL ratio of 0.9669%. Following discharge, the patient continued ATRA at 10 mg three times daily, alternating with 2 weeks of rest, followed by 2 weeks of Realgar-Indigo naturalis formula (RIF, 5 tablets three times daily). This consolidation therapy was planned for two cycles. Two months later, a bone marrow smear confirmed sustained CR, and the PML::RARA fusion gene was no longer detectable. CT scans of the lumbar spine showed no mass (Figure 3). The patient’s back pain had significantly improved, and he was able to walk freely. The consolidation regimen of ATRA and RIF was maintained for 6 months, with regular monitoring of bone marrow cytology, genetic tests, and spinal imaging every 2 months, all of which showed normal results. Eight months after the first induction, the patient received radiation therapy. He tolerated the treatment well with no major complications. One year after treatment, the patient remained in remission, with molecular analyses of bone marrow and peripheral blood showing no evidence of PML::RARA fusion transcripts (Figure 4).

Our review identified 41 cases of APL where MS was the initial presenting symptom. Key details of these cases, including onset locations, clinical features, and treatment responses, are summarized in Table 1 (15–54). The patients were predominantly young, with a median age of 39.5 years (range: 1–77 years), and only 15% were aged ≥60 years. The male-to-female ratio was 24:17, showing no significant sex differences. The spine was the most common site of extramedullary infiltration (12/41 cases) (15, 18, 27, 28, 33, 35, 37, 43, 49–51, 54), followed by the skin (4/41) (19, 29, 42, 52), pleura (3/41) (19, 40, 53), and ovary (2/41) (25, 49). Other less common sites included the intracranial region (2/41) (22, 48), tongue (2/41) (24, 44), humerus (2/41) (23, 39), and colon (2/41) (31, 46), among others. Notably, one patient developed sarcoma in a donor kidney after renal transplantation, not in their own kidney (45). Most MS cases were confined to a single site (78%, 32/41), with multiple-site (9 cases) and multi-organ (7 cases) involvement occurring less frequently. Immunophenotyping of extramedullary masses typically showed MPO positivity. Other markers included CD68 (20%, 8/41), CD43, CD33, and CD117 (15%, 6/41), and CD13 (10%, 4/41). Bone marrow infiltration was observed in 59% (24/41) of patients, while 42% (17 patients) (17, 19, 21, 23, 25, 26, 32, 33, 37–39, 41, 43, 45, 49, 53, 54) had no blasts or promyelocytes in the bone marrow and circulating blood, or did not meet the diagnostic criteria for APL. Six patients had elevated white blood cell counts (16, 22, 24, 41, 42, 44), and three presented with disseminated intravascular coagulation (DIC) (15, 22, 53). Based on white blood cell counts, patients were classified into high-risk (6 patients) (16, 22, 24, 41, 42, 44) and low-risk (27 patients) (15, 17–20, 25–31, 34–40, 46–53) groups, while the remaining cases (21, 23, 32, 33, 43, 54) could not be classified. Chromosome 15 and 17 translocations (t(15;17)) were detected in 54% (22/41) of cases. Seven cases (17%) (22, 25, 33, 41, 47, 49, 52) had a normal karyotype, and 5% (2/41) (48, 54) had complex karyotypes. The common PML::RARA fusion was present in 59% (24/41) of patients, while rare fusion signals involving RARA [fused with NPM1 (35), FIP1L1 (48), ZBTB16 (51), and TTMV (54)] were detected in four cases. One case lacked RARA rearrangement, but RT-PCR testing revealed an in-frame fusion between CPSF6 exon 4 and RARG exon 4 (CPSF6:: RARG) (47). Six patients (15%) had concurrent gene mutations, with FLT3 mutations being the most common (7%, 3/41) (37, 41, 50). Other mutations included KARS (47, 48) and WT1 (47, 54) (each in two patients), as well as EZH2 (47), KMT2C (50), and SMAD9 (54) mutations.

A total of 40 patients received treatment, with 28 achieving remission, resulting in an overall response rate of 70%. Among low-risk patients, the remission rate was 70% (19 out of 27), while high-risk patients had a slightly higher remission rate of 83% (5 out of 6). Thirty patients were treated with ATRA combined with chemotherapy, and 83% (25 out of 30) achieved remission. One patient did not respond to treatment, two died from multiple organ failure, and two succumbed to intracranial hemorrhage. Additionally, three patients who received only chemotherapy also responded to treatment. Follow-up duration varied widely across studies, ranging from less than 1 week (20, 22) to as long as 288 months (49). This variation is attributed to several factors, including severe exhaustion and bleeding in some patients, either untreated or occurring during the myelosuppressive phase following surgery or chemotherapy. Notably, long-term survival was observed in patients who underwent surgery with careful monitoring and received ATRA combined with chemotherapy. Interestingly, the longest-followed patients experienced recurrent relapses of MS in various locations, despite no abnormalities being detected in the bone marrow or peripheral blood. These patients maintained long-term survival and good quality of life through surgery and chemotherapy. In addition to relapse at other extramedullary sites, MS/APL can progress to non-M3 AML. One patient, for instance, progressed to AML with the t(8;21)/RUNX1:RUNX1T1 translocation after more than 2 years of remission following ATRA and chemotherapy (25).

For patients with solitary MS and no bone marrow infiltration, the most common sites of infiltration were the spine (5 out of 17) and other bony sites (5 out of 17), followed by the pleura (2 out of 17). Singular cases were observed in the mediastinum, testis, ovary, and breast (1 each). Notably, no reports of involvement in digestive tract organs were found. Similar to other MS/APL cases, these patients were predominantly young, with a median age of 29 years, and there was no significant sex difference (male-to-female ratio of 10:7). The treatment response rate in this group was 70% (12 out of 17), with 91% (10 out of 11) of patients treated with ATRA achieving remission, while the remaining patients succumbed to severe bleeding. Five cases progressed to bone marrow infiltration. Three of these cases were thought to reflect disease development before treatment, while the remaining two cases developed bone marrow blasts or promyelocytes months after treatment initiation. This progression was considered to indicate a combination of disease relapse and progression.

Myeloid sarcoma typically manifests in patients with APL during relapse, with extramedullary involvement being relatively uncommon. It occurs in approximately 3%–5% of APL patients (12, 55). In the European multicenter PETHEMA trial, only 10 of 169 relapse cases showed extramedullary involvement, predominantly in the central nervous system and skin (14). Instances where MS presents as the initial symptom, without significant bone marrow or peripheral blood abnormalities, or where APL diagnostic criteria are not met, are exceedingly rare. Recent reports have identified new cases and additional sites of extramedullary infiltration. Among these, spinal extramedullary masses are the most frequently observed, followed by skin and pleura. Other tissues, including rarely transplanted tissue, have also been implicated. In MS/APL patients without evidence of disease in the bone marrow or peripheral blood, the extramedullary masses predominantly involve bony structures such as the spine, sternum, and humerus. However masses located in the digestive system are rare in these patients compared to other MS/APL cases. Therefore, isolated MS located in the skeleton is even more important to evaluate thoroughly and the possibility of promyelocytic sarcoma should be considered.

An important consideration for clinicians is the need for a comprehensive understanding and systematic evaluation of donor health in patients undergoing organ transplantation, to exclude blood-related diseases. In cases of MS/APL in transplanted organs, it is crucial to not only examine the patient’s bone marrow but also conduct PML::RARA testing on bone marrow and peripheral blood from both the donor and other recipients. This strategy ensures prompt detection and management of potential complications. The timing of detection post-transplant remains an unresolved issue that requires further investigation.

Clinicians often face significant challenges in diagnosing APL with MS, particularly when the presentation involves solitary promyelocytic sarcoma. When a mass is detected in any part of the body, fine needle aspiration often fails to provide sufficient diagnostic evidence of myeloid malignancy. In cases without coagulation abnormalities or other contraindications to surgery, a local pathological biopsy followed by immunohistochemical examination of the mass is essential to determine its origin. For suspected myeloid-origin tumors, it is critical to perform a bone marrow aspirate to rule out APL or other forms of non-M3 AML. Even when blood and bone marrow smears and flow cytometry do not show abnormalities, molecular testing is crucial. Both qPCR and FISH should be performed to detect PML::RARA fusion gene positivity. Although molecular analysis and FISH of MS biopsy tissue can be technically challenging, they are important for accurate diagnosis and should be performed whenever possible. For patients without atypical promyelocytes in the peripheral blood and bone marrow, and with no cytogenetic abnormalities, the detection of PML::RARA transcripts or RARA rearrangements in MS tissues via qPCR or FISH becomes the key diagnostic criterion. Additionally, karyotype analysis of the bone marrow, showing translocations involving chromosomes 15 and 17, can further strengthen diagnostic confidence in cases of solitary MS. Thus, the presence of PML::RARA is considered a critical marker for both the early diagnosis of solitary promyelocytic sarcoma and the monitoring of treatment efficacy and recurrence. An intriguing observation in some cases is the identification of rare fusion genes, although their association with MS/APL remains unclear. This highlights the need for further research to understand the significance of these rare fusions. Moreover, the absence of PML::RARA does not reliably exclude APL, emphasizing the importance of comprehensive testing. Next-generation sequencing and RT-PCR for other rare fusion transcripts could reveal unexpected findings, potentially offering new insights into MS/APL diagnostics.

This case is similar to previously reported MS/APL with a spinal intradural mass as the first manifestation, and the patient usually presents with low back pain and difficulty walking. These symptoms may occur with or without abnormal blood counts and coagulation. In this case, induction chemotherapy with ATRA combined with ATO was initiated after local lumpectomy. ATRA treatment continued to maintain PML::RARA negativity, followed by local radiotherapy. The patient achieved remission and maintained a good quality of life.

Patients with MS/APL, particularly those with spinal intraspinal masses at onset, often have a poor prognosis, highlighting the need for effective treatment strategies to improve outcomes. Treatment for these cases is similar to that for extramedullary relapses of APL, involving surgical decompression, local radiotherapy, and leukemia chemotherapy. Surgical resection is essential for reducing tumor volume, alleviating tissue compression, and preventing further spread. If coagulopathy is not significant, surgery should be performed promptly to relieve pain and improve mobility. Systemic therapy for the underlying leukemia is always necessary, regardless of bone marrow involvement or isolated MS/APL (56). ATRA, while effective, poorly penetrates the blood-brain barrier and is associated with relapses in the central nervous system (CNS) (57). Additionally, ATRA has been shown to increase tumor cell adhesion molecule expression (58–60), which could promote extramedullary metastasis and invasion. However, a higher incidence of extramedullary recurrence has not been observed in APL patients receiving ATRA compared to those treated with chemotherapy alone, though CNS recurrence is slightly more common, yet not statistically significant (55). Real-world data indicate that two-drug induction therapy combining ATRA and ATO offers longer disease-free survival compared to ATRA combined with chemotherapy (AIDA) (61–63). Thus, the combination of ATRA and ATO is recommended for treating PML::RARA-positive MS/APL. For rare RARA rearrangements, the specific fusion partners should be considered to determine whether ATRA is appropriate. The role of radiotherapy in treating APL-related extramedullary sarcoma remains debated. Some researchers view it as an effective strategy for eliminating residual tumor tissue and reducing recurrence risk after surgery (64, 65). However, others caution that local radiotherapy may increase the patient’s overall burden, leading to infections, treatment failure, or delays in chemotherapy (66). In some cases, patients intolerant to therapy have died from severe infections unrelated to chemotherapy. Furthermore, the potential for bone marrow infiltration by leukemic cells following radiotherapy, either from disease progression or radiotherapy-induced malignancy, remains a contentious issue. Given these considerations, we propose that a combination of ATRA and ATO be considered the optimal approach for treating PML::RARA-positive MS/APL. Local radiotherapy could be administered after consolidation therapy, weighing its potential benefits against its risks. New studies have explored the use of gilteritinib for extramedullary recurrence of APL with FLT3 mutations, showing rapid and sustained regression of the sarcoma (67). For patients with isolated MS/APL at initial diagnosis, whether targeted agents can improve remission and disease-free survival in the presence of specific gene mutations warrants further investigation. Additionally, hyperthermia, which shows synergistic effects with ATO in destabilizing PML::RARA fusion proteins both in vivo and in vitro, may offer a promising new therapeutic strategy (68).

In conclusion, we describe the rare presentation of APL solely as MS in a patient, which ultimately led to the diagnosis of MS/APL. Additionally, we provide a comprehensive review of similar cases to further elucidate this uncommon clinical manifestation of APL. The case and literature review contribute to the growing body of knowledge regarding the presentation, diagnosis, and treatment of MS/APL, potentially guiding future clinical practice in similar cases.