Extramedullary Acute Myeloid Leukemia in the De Novo Setting: Clinicopathologic Characteristics, Molecular Analysis, and Treatment Outcomes

doi:10.21203/rs.3.rs-1832465/v1

Extramedullary disease in acute myeloid leukemia (AML), also known as myeloid sarcoma (MS), is an uncommon presentation of de novo AML. We analyzed medical records of patients in the Mayo Clinic database spanning between 1996 and 2021 who were diagnosed with de novo extramedullary AML. Survival outcomes were analyzed using Kaplan-Meier and Cox-proportional hazard models for univariate and multivariate analysis, respectively. With a median follow-up of 20.7 months, 83 patients with de novo MS were identified; 49 presented with synchronous MS and 34 with isolated MS. Next generation sequencing (NGS) revealed RTK-RAS pathway mutations in 44% of patients. Median overall survival (OS) was 25.3 months and 17.8 months in isolated and synchronous MS, respectively (p = 0.49). Variables associated with improved OS on univariate analysis included treatment with intensive chemotherapy (p < 0.001), normal or favorable risk karyotype (p = 0.001), age less than 60 years (p = 0.001), and a complete response to treatment at extramedullary sites (p < 0.0001). On multivariate analysis, normal or favorable risk karyotype (p = 0.001), age less than 60 (p = 0.01), and a complete extramedullary response (p = 0.001) were associated with improved OS. Localized therapy did not provide added survival benefit in isolated MS. Genomic stratification, age, and extramedullary site response are predictive of survival outcomes.

Myeloid sarcoma (MS) consists of immature myeloid blast cells that disrupt the typical architecture of the invaded tissue outside the bone marrow (BM). Although found in association with acute myeloid leukemia (AML) and presenting most frequently as a sign of progressive medullary disease, MS can also arise de novo with or without synchronous BM involvement at time of initial diagnosis. The incidence of de novo MS with synchronous BM involvement and isolated MS without evidence of medullary involvement in AML ranges from 0.2–2.8% and 0.6-1.0%, respectively (1–5). When comparing isolated MS (iMS) to synchronous MS (sMS), previous studies describe no significant differences in survival (4–8). Although the largest review to date reported a median overall survival (OS) of 12.8 months in iMS (2), other studies report a median OS ranging between 8 and 42 months (3, 9). Nevertheless, MS at the time of initial diagnosis, whether isolated or concurrent with BM involvement, remains a poor prognostic factor with an increased risk of relapse (10, 11). With conflicting evidence regarding the optimal treatment strategy for de novo MS, it is generally recommended that these patients receive systemic chemotherapy because iMS commonly evolves into AML with bone marrow involvement in less than six months (8, 11). The role of localized therapy (i.e., radiation or surgery) and allogeneic hematopoietic stem cell transplantation (alloSCT), in both iMS and sMS, remains unclear. In this study, we aim to elucidate this relatively rare disease’s clinicopathological and molecular characteristics and determine optimal treatment strategies for both iMS and sMS.

Institutional Review Board (IRB) approval was obtained prior to the initiation of the study. Patients with possible extramedullary AML presenting between June 1996 and November 2021 were identified using documentation of “granulocytic sarcoma,” “myeloid sarcoma,” or “extramedullary acute myeloid leukemia” within the Mayo Clinic database. Patients with biopsy-proven confirmation of MS without any previous history of AML were included. WHO criteria were used to determine concurrent BM involvement; patients with greater than or equal to 20% myeloid blasts on BM biopsy were deemed to have synchronous BM involvement. All other MS patients were categorized as iMS without medullary involvement. Patients who presented with MS at relapse from previously diagnosed AML were excluded from this study. Bone marrow genomic stratification was obtained and determined according to the European Leukemia Network (ELN) criteria when information was available (12). Treatment response was assessed at both the extramedullary and BM site using radiographic imaging (i.e., PET, MRI, CT) and BM biopsy, respectively. Information regarding patient characteristics was collected, and disease progression and survival outcomes were analyzed. Event-free survival (EFS) was calculated as the time from diagnosis until the first sign of disease progression, relapse, death, or when a patient was lost to follow up. Overall survival (OS) was calculated as the time from diagnosis until death or last encounter if lost to follow up. Continuous variables were summarized with the sample median and range. Categorical variables were summarized with the number and percentage of patients. Comparisons of characteristics were made using a Wilcoxon rank-sum test (continuous variables) or Fisher’s exact test (categorical variables). Survival outcomes were analyzed using Kaplan-Meier and Cox-proportional hazard models for univariate and multivariate analysis, respectively. Statistical analyses were performed using BlueSky Statistical Software.

Patient Characteristics

A total of 83 patients with de novo extramedullary MS were identified, including 49 with sMS and 34 with iMS. Baseline patient characteristics are summarized in Table 1. The median age at diagnosis was 56 years (range, 17–89); 63% of patients were male, and 94% were Caucasian. Compared to iMS patients, sMS patients more frequently displayed leukocytosis (white blood cell count > 9.6 X 10⁹ cells/L), anemia (hemoglobin < 13.2 g/dl) and thrombocytopenia (platelet count < 125 X 10⁹ cells/L). The most common sites of MS involvement included the soft tissue (30%), skin and breast (28%), and spleen or lymphatic system (21%). According to ELN genomic stratification (12), 32 patients (43%) had a normal karyotype, 10 (13%) carried a favorable risk karyotype, 13 (17%) displayed an intermediate risk karyotype, and 19 (25%) harbored an adverse risk karyotype. Isolated MS patients primarily carried a normal karyotype (71%), whereas sMS patients displayed a greater number of intermediate and adverse risk karyotypes compared to iMS (57% vs. 21%, p = 0.002).

Bone Marrow Karyotype, Targeted Cytogenetics, and Molecular Analysis

Standard chromosome analysis with karyotyping of the BM was available in 67/83 (80%) patients during initial diagnosis, with 23 (34%) abnormal results. Of those patients with iMS, 5/27 (19%) had abnormal BM karyotyping findings; two had 7/7q deletion, one had 7/7q and concomitant 17p deletion, one had combined inv(16)/trisomy 8/trisomy 13/13(q) deletion, and one harbored a complex karyotype. For sMS patients, 19/40 (47%) displayed abnormal results. The most common aberrations included t(8;21) in three patients, complex karyotypes in three patients, inv(16) in two patients, t(9;11) in two patients, and 5q deletion in two patients. Despite normal BM morphology, abnormal karyotyping results were discovered using fluorescence in situ hybridization (FISH) testing of the BM in 4/12 (33%) iMS patients; two had 7/7q deletion, one had inv(16)/CBFB/MYH11, and one had t(1;11)(q21;p15) with concomitant trisomies 8 and 20.

Next generation sequencing (NGS) results of the BM and/or peripheral blood were available in 27 patients, including seven iMS patients, and can be found in (Fig. 1). A total of 24 of 27 (88%) patients harbored somatic mutations. The most frequent genetic aberrations were RTK-RAS pathway mutations (12/27; 44%), including KRAS (5/27; 19%), FLT3 (4/27; 15%), and NRAS (3/27; 11%). Other frequently encountered mutations included NPM1 (10/27; 37%), TET2 (6/27; 22%), IDH2 (4/27; 15%), IDH1 (3/27; 11%), ASXL1 (3/27; 11%), and STAG2 (3/27; 11%).

Cytogenetic and Molecular Analysis of Extramedullary Site in Isolated MS

Standard karyotyping of MS sites was conducted in 3 iMS patients. Results included 2 patients with complex karyotypes and 1 with inv(16) and a concomitant 7q deletion. FISH analysis on MS sites of iMS patients was performed in 19 of 34 cases with 9 (47%) abnormal results; two patients harbored inv(16)(MYH11/CBFB fusion) with 1 carrying a concomitant 5q deletion; two patients carried three copies of RUNX1T1 amplification with 1 having an additional two copies of RUNX1 and the other harboring a concomitant BCR amplification; one patient displayed RUNX1T1/RUNX1 fusion; one patient had YST3-CREBBP fusion; one patient had 8q duplication; one patient displayed MLL rearrangement, and 1 patient carried MLL/MYC rearrangement. Standard karyotyping revealed two additional complex karyotypes not found on FISH analysis, accounting for 11 abnormal cytogenetic results at iMS sites. NGS was performed on iMS sites in 8 patients. There were 6 abnormal results, including mutations in CBL, NMP1/CBL/PTPN11/SETD2/STAG2, ASXL1/BCOR, PMS2, KDM6A, and WT1 with biallelic CEBPA.

Treatment and Treatment Response

Intensive chemotherapy was administered to a total of 68 of 83 (82%) patients for induction therapy, including 27 of 34 (79%) iMS patients and 41 of 49 (83%) sMS patients. Intensive regimens included 7 + 3 (high dose cytarabine plus daunorubicin) in 53 (63%) patients, MEC (mitoxantrone, etoposide, intermediate-dose cytarabine) plus methotrexate in 3 (3%) patients, CLAG-M (cladribine, cytarabine, mitoxantrone) in three (3%) patients, IDA plus ARA-C (idarubicin and intermediate-dose cytarabine) in three (3%) patients, ARA-C in three (3%) patients, HiDAC (high dose cytarabine) in one (1%) patient, and hyper-CVAD (hyperfractionated cyclophosphamide, vincristine, doxorubicin, dexamethasone alternating with high dose methotrexate and cytarabine) in one (1%) patient. Eight (9%) patients received non-intensive chemotherapy +/- localized therapy. Non-intensive regimens included azacitidine, azacitidine plus venetoclax, cladribine, decitabine, decitabine plus venetoclax, and gemtuzumab. Localized therapy (i.e., radiation or surgery) was concomitantly administered with systemic chemotherapy to 20/34 (58%) iMS patients compared to 11 of 49 (22%) sMS patients (p = 0.02). Five (6%) patients received localized therapy alone without systemic therapy. Two (2%) patients received palliative/supportive care alone.

Results regarding initial treatment response at both the medullary and extramedullary sites of involvement are described in Table 2. Forty of 68 (58%) patients who received intensive chemotherapy +/- localized therapy experienced complete remission (CR) at both the BM and MS sites of involvement. Patients who received intensive chemotherapy +/- localized therapy experienced CR at both the BM and MS site of involvement more frequently than those who received alternative therapies (i.e., non-intensive therapy, localized therapy alone) (p = 0.043). Compared to intensive chemotherapy alone, the addition of localized therapy to intensive chemotherapy did not significantly increase rates of partial remission (PR) or CR at MS sites of involvement in all patients (p = 0.11) nor in iMS alone (p = 0.07). After experiencing CR following intensive chemotherapy, 25 patients underwent alloSCT, including 11 iMS patients. Seven patients who achieved CR following intensive chemotherapy relapsed and subsequently received alloSCT. Five other patients who had resistant disease also received alloSCT, accounting for 37 allografted patients.

EFS and OS

The median length of follow-up from the time of diagnosis was 20.7 months (95% CI; 0.35-36), at which time, 53 of 83 (64%) patients had expired. Median event-free survival was 7.7 months and 7.3 months in isolated and synchronous MS, respectively (p = 0.50). Median OS was 25.3 months and 17.8 months in iMS and sMS, respectively, with no significant difference in estimated OS (p = 0.49) (Fig. 2).

Compared to other treatment modalities, treatment with intensive chemotherapy was associated with improved OS on univariate analysis (p < 0.0001) (Fig. 3) with a median OS of 30.3 months. Multivariate analysis, however, did not reveal any significant impact on EFS or OS with the reception of intensive therapy in all MS patients. Those who received localized therapy in addition to systemic chemotherapy did not derive any additional benefit towards EFS (p = 0.57) or OS (p = 0.92) compared to those treated with systemic chemotherapy alone. In iMS, chemotherapy +/- localized therapy was associated with improved OS (p = 0.03) compared against localized therapy alone on univariate analysis, but not EFS (p = 0.11). Additionally, reception of intensive chemotherapy +/- localized therapy in iMS patients was associated with improved OS (p = 0.01) compared to other treatment modalities (non-intensive therapy +/- localized therapy, localized therapy alone) in univariate analysis, but did not have a significant effect on EFS (p = 0.06). No treatment modality was associated with improved survival on multivariate analysis in either iMS or sMS.

Regarding genomic risk stratification, those with intermediate or high-risk genomic profiling carried a significantly worse prognosis on univariate analysis compared to those with normal or favorable risk genomics (p = 0.0022) (Fig. 4). Intermediate or adverse risk genomics also conferred a poor prognosis upon multivariate analysis (p = 0.001). There was no significant difference in EFS or OS when analyzing patients according to RTK-RAS pathway mutations (i.e., KRAS, FLT3, NRAS, CEBPA) or any other underlying NGS mutations on multivariate analysis.

Although patients with gynecological or genitourinary site of involvement appeared to have superior OS compared to other sites of involvement (median OS 40.4 months vs 17.8 months, respectively) (Fig. 5), there was a lack of significant difference in EFS or OS on multivariate analysis. A complete response to treatment at extramedullary sites of involvement was associated with improved OS in both univariate (p < 0.0001) and multivariate analysis (p = 0.001). Individuals less than 60 years of age also demonstrated superior OS on both univariate and multivariate analysis (p = 0.01). Patients with sMS were more prone to relapse in the BM and blood compared to iMS (p = 0.02).

In summary, on univariate analysis, age less than 60 years old, reception of intensive therapy, intermediate/adverse risk genomics, and EM site response to treatment (CR/PR/R) demonstrated significant effects on OS. On multivariate analysis, only intermediate risk/adverse risk genomics, age greater than 60, and a lack of EM site response to treatment were independently associated with poor OS outcomes. No variables demonstrated significant impact on EFS in either iMS or sMS on univariate or multivariate analysis.

AlloSCT Results

For the 37 patients who underwent alloSCT as part of either consolidative or salvage therapy, the post-transplant median EFS was 27.1 months (95% CI; 19.3, 40.6). AlloSCT recipients had a median OS of 48.5 months compared to 16.0 months for those who did not receive alloSCT (p = 0.07). When comparing only those patients who obtained CR following intensive chemotherapy regimens, the median OS was 48.5 months in alloSCT recipients (n = 32) versus 94.5 months in non-transplanted patients (n = 22) (p = 0.36).

In the iMS setting, 11 patients received alloSCT during CR with a mOS of 57.7 months. Comparably, ten iMS patients attained CR following intensive therapy alone without subsequent alloSCT, with a median OS of 41 months (p = 0.34). In sMS, 21 patients received alloSCT during CR and had a mOS of 48.5 months compared to 16.3 months in 12 non-transplanted patients (p = 0.90). Those with intermediate or adverse risk genomics who received consolidative alloSCT in either the iMS or sMS setting during CR did not display any survival advantage compared to non-transplanted patients.

De novo MS is a rare presentation of AML that can occur with or without synchronous BM involvement. Our single institution review of 83 patients with de novo MS adds to the limited available data to further characterize this rare disease and determine optimal treatment strategies in isolated and synchronous settings. In addition to the skin and soft tissue being the most commonly affected sites of extramedullary involvement, those with skin, gastrointestinal, soft tissue, and breast involvement seem to carry a worse prognosis compared to those with genitourinary or gynecological involvement, consistent with previous reports (2, 3). Age greater than 60 years was associated with worse survival outcomes in both iMS and sMS, but this may be attributable to older patients’ inability to tolerate intensive chemotherapy considering only 57.5% of patients above 60 years old received intensive therapy compared to 95.6% in those less than 60 years old in our study. Older patients may be unable to tolerate intensive therapies, as older patients treated with intensive therapy have a worse prognosis compared to localized therapy alone (2). Consistent with published literature, our study suggests that advanced age at diagnosis confers a worse prognosis in both iMS and sMS subgroups (2, 3).

Targeted cytogenetics (FISH) and/or molecular testing (NGS) of the BM and/or peripheral blood was performed in 52/83 (62%) patients in the current review, including 16 iMS patients. Abnormal results were discovered in 38/52 (73%) patients. According to ELN criteria (12), genomic risk profiling in our review directly correlated with survival in both iMS and sMS regardless of the treatment group. Interestingly, those with intermediate-risk genomics carried an equally poor prognosis compared to patients with adverse risk genomics (p = 0.20). In contrast to a previous series reporting normal BM karyotype in all iMS cases (5), the current series discovered an abnormal karyotype upon cytogenetic and FISH testing of the BM in 5/12 (41%) available iMS patients despite normal BM morphology; 3 patients carried a chromosome 7/7q deletion, one patient harbored inv(16)(p.13.1q22) with concomitant 13q deletion and trisomies 8 and 13, and 1 patient had a complex karyotype including t(1;11)(q21;p15) with trisomies 8 and 20. Chromosome 7/7q deletion was also found in 2 additional sMS patients, deeming it the most frequently encountered cytogenetic abnormality in the current series. In correlation with previous reports (13, 14), t(8;21) was also discovered in 3 sMS patients. Inv(16) was identified in 2 sMS patients, with 1 case having GI involvement (9). According to NGS, the most prevalent genetic aberrations involved RTK-RAS pathway mutations (12/27; 44%), including KRAS (5/27; 19%), FLT3 (4/27; 15%), and NRAS (3/27; 11%). Although previous reports document RTK-RAS pathway mutations primarily during extramedullary relapse (9, 15), our report of RTK-RAS mutations in the de novo setting at initial diagnosis runs contrary to the notion that RAS mutations likely represent later changes in the progression of AML (16). RAS mutations may play a critical role in the early development of extramedullary AML and provide a potential therapeutic opportunity using RAS inhibition in the future treatment of extramedullary AML (17). Also increased in prevalence when compared to AML at baseline were mutations in NPM1 (10/27; 37%), TET2 (6/27; 22%), and IDH2 (4/27; 15%).

Although the optimal therapy for MS has yet to be clearly defined, the consensus is that induction chemotherapy should be employed up front for de novo MS in the synchronous setting, but its role in iMS has remained unclear. The historic notion is that iMS evolves into AML with systemic involvement within six months and, therefore, should be treated as a systemic disease with systemic chemotherapy (3, 8, 11, 18, 19). Although previous studies have reported superior EFS and OS with induction chemotherapy compared to localized therapy alone in iMS (4, 9), the largest review to date involving 151 patients more recently described no survival benefit when adding intensive chemotherapy to localized therapy, suggesting the use of localized therapy alone +/- targeted therapies without systemic chemotherapies in iMS (20). This same review notably lacked descriptions of chemotherapy regimens and cytogenetic and molecular profiling. In contrast, our review discovered that iMS patients who received intensive chemotherapy, comprised primarily of anthracycline and cytarabine-based regimens, resulted in higher rates of CR at both the BM and MS site of involvement as well as improved OS outcomes when directly compared to localized therapy +/- non-intensive chemotherapy. Our findings, therefore, support the use of induction chemotherapy in the iMS setting with consideration of localized therapy in symptomatic patients despite a lack of proven survival benefits (3).

The role of alloSCT in de novo MS could not be assessed in our study due to the limited sample size. Although it has been proposed that alloSCT should only be employed in sMS patients with intermediate or adverse risk genomic profiling (21), the majority of evidence supports the use of alloSCT in all sMS patients who can tolerate intensive and ablative therapies. Previous studies have reported improved survival outcomes with alloSCT in both sMS and iMS, particularly in patients who have achieved CR following induction chemotherapy (22–26). Although not statistically significant, our review discovered an absolute superior mOS in sMS patients with pre-transplant CR who underwent alloSCT compared to those who did not receive alloSCT, regardless of genomic risk classification. In iMS, recent studies have described ambivalent outcomes following alloSCT without any significant survival advantage compared to chemotherapy alone (5, 9). Likewise, our review discovered no significant advantage following alloSCT in iMS. Moreover, alloSCT did not significantly improve OS in iMS patients who obtained CR compared to non-transplanted patients.

The main limitations of the current series are the retrospective nature of our investigation, the reliance upon documentation of myeloid sarcoma on chart review, the lack of phenotypic and morphologic descriptions, and the relatively small sample size in the setting of a rare disease process that does not lend itself to prospective, randomized trials. Further collaboration across multiple institutions would be beneficial to further characterize and determine optimal treatment strategies toward de novo MS, particularly in the iMS setting. The role of alloSCT in de novo MS should be explored further in future studies, including transplant factors that could influence outcomes (i.e., type of alloSCT, myeloablative versus non-ablative regimens, baseline performance score) as well as transplant-related outcomes (i.e., acute/chronic graft-versus-host disease).

In conclusion, de novo MS appears to be enriched with RTK-RAS pathway mutations, particularly in the sMS setting, and remains an aggressive form of AML, especially in patients with intermediate or high-risk genomic profiling and those above 60 years of age. Intensive chemotherapy should be administered to all eligible patients. There is no additional value towards OS in supplementing with localized therapy in sMS or iMS. AlloSCT may be considered in eligible patients with sMS, but its benefit in iMS remains uncertain. Further studies into this rare presentation of de novo AML, including the role of alloSCT, are needed to improve patient outcomes.

Author Contributions: JK was responsible for project conceptualization, design protocol, extracting and analyzing data, interpreting results, formulating tables and figures, original draft, and revisions. BM was responsible for project conceptualization, design protocol, extracting and analyzing data, performing the statistical analysis, interpreting results, and revisions. LJ was responsible for extracting and analyzing data, interpreting results, revisions. DM was responsible for extracting and analyzing data, interpreting results, revisions. TB was responsible for interpreting results, revisions, providing feedback on report. MKD was responsible for interpreting results, revisions, providing feedback on report. JF was responsible for interpreting results, revisions, providing feedback on report. ZR was responsible for project conceptualization, designing review protocol, submitting IRB, providing feedback on report. HM was responsible as the principal investigator for project conceptualization, administration, designing review protocol, interpreting results, original draft, revisions, and providing feedback on report.

Competing Interests: JK declares no competing financial interest. BM declares no competing financial interest. LJ declares no competing financial interest. DM declares no competing financial interest. TB declares no competing financial interest. MDK declares no competing financial interest. JF declares no competing financial interest. ZR declares no competing financial interest. HM declares no competing financial interest.

Data Availability Statement: The data generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Ganzel C, Manola J, Douer D, Rowe JM, Fernandez HF, Paietta EM, et al. Extramedullary disease in adult acute myeloid leukemia is common but lacks independent significance: analysis of patients in ECOG-ACRIN cancer research group trials, 1980–2008. Journal of Clinical Oncology. 2016;34(29):3544.
Goyal G, Bartley A, Patnaik M, Litzow M, Al-Kali A, Go R. Clinical features and outcomes of extramedullary myeloid sarcoma in the United States: analysis using a national data set. Blood cancer journal. 2017;7(8):e592-e.
Movassaghian M, Brunner AM, Blonquist TM, Sadrzadeh H, Bhatia A, Perry AM, et al. Presentation and outcomes among patients with isolated myeloid sarcoma: a Surveillance, Epidemiology, and End Results database analysis. Leukemia & lymphoma. 2015;56(6):1698–703.
Lan T-Y, Lin D-T, Tien H-F, Yang R-S, Chen C-Y, Wu K. Prognostic factors of treatment outcomes in patients with granulocytic sarcoma. Acta haematologica. 2009;122(4):238–46.
Begna KH, Kittur J, Yui J, Gangat N, Patnaik MM, Al-Kali A, et al. De novo isolated myeloid sarcoma: comparative analysis of survival in 19 consecutive cases. British journal of haematology. 2021;195(3):413–6.
Kawamoto K, Miyoshi H, Yoshida N, Takizawa J, Sone H, Ohshima K. Clinicopathological, cytogenetic, and prognostic analysis of 131 myeloid sarcoma patients. The American journal of surgical pathology. 2016;40(11):1473–83.
Pileri S, Ascani S, Cox M, Campidelli C, Bacci F, Piccioli M, et al. Myeloid sarcoma: clinico-pathologic, phenotypic and cytogenetic analysis of 92 adult patients. Leukemia. 2007;21(2):340–50.
Tsimberidou A, Kantarjian H, Estey E, Cortes J, Verstovsek S, Faderl S, et al. Outcome in patients with nonleukemic granulocytic sarcoma treated with chemotherapy with or without radiotherapy. Leukemia. 2003;17(6):1100–3.
Abbas HA, Reville PK, Geppner A, Rausch CR, Pemmaraju N, Ohanian M, et al. Clinical and molecular characterization of myeloid sarcoma without medullary leukemia. Leukemia & Lymphoma. 2021:1–9.
Tsimberidou A, Kantarjian H, Estey E, Cortes J, Verstovsek S, Faderl S, et al. Outcome in patients with nonleukemic granulocytic sarcoma treated with chemotherapy with or without radiotherapy. Leukemia. 2003;17(6):1100–3.
Yamauchi K, Yasuda M. Comparison in treatments of nonleukemic granulocytic sarcoma: report of two cases and a review of 72 cases in the literature. Cancer. 2002;94(6):1739–46.
Döhner H, Estey E, Grimwade D, Amadori S, Appelbaum FR, Büchner T, et al. Diagnosis and management of AML in adults: 2017 ELN recommendations from an international expert panel. Blood, The Journal of the American Society of Hematology. 2017;129(4):424–47.
Sugimoto Y, Nishii K, Sakakura M, Araki H, Usui E, Hoshino N, et al. Acute myeloid leukemia with t (8; 21)(q22; q22) manifesting as granulocytic sarcomas in the rhinopharynx and external acoustic meatus at relapse after high-dose cytarabine: case report and review of the literature. The hematology journal: the official journal of the European Haematology Association. 2004;5(1):84–9.
Tallman MS, Hakimian D, Shaw JM, Lissner GS, Russell EJ, Variakojis D. Granulocytic sarcoma is associated with the 8; 21 translocation in acute myeloid leukemia. Journal of Clinical Oncology. 1993;11(4):690–7.
Choi M, Jeon YK, Sun C-H, Yun H-S, Hong J, Shin D-Y, et al. RTK-RAS pathway mutation is enriched in myeloid sarcoma. Blood cancer journal. 2018;8(5):1–4.
Welch JS, Ley TJ, Link DC, Miller CA, Larson DE, Koboldt DC, et al. The origin and evolution of mutations in acute myeloid leukemia. Cell. 2012;150(2):264–78.
O’Bryan JP. Pharmacological targeting of RAS: recent success with direct inhibitors. Pharmacological research. 2019;139:503–11.
Tsimberidou AM, Kantarjian HM, Wen S, Keating MJ, O'Brien S, Brandt M, et al. Myeloid sarcoma is associated with superior event-free survival and overall survival compared with acute myeloid leukemia. Cancer: Interdisciplinary International Journal of the American Cancer Society. 2008;113(6):1370–8.
Imrie KR, Kovacs MJ, Selby D, Lipton J, Patterson BJ, Pantalony D, et al. Isolated chloroma: the effect of early antileukemic therapy. Annals of Internal Medicine. 1995;123(5):351–3.
Lontos K, Yabes JG, Farah RJ, Boyiadzis M. Impact of upfront chemotherapy on overall survival in isolated myeloid sarcoma. Leukemia. 2021;35(4):1193–6.
Solh M, Solomon S, Morris L, Holland K, Bashey A. Extramedullary acute myelogenous leukemia. Blood reviews. 2016;30(5):333–9.
Shan M, Lu Y, Yang M, Wang P, Lu S, Zhang L, et al. Characteristics and transplant outcome of myeloid sarcoma: a single-institute study. International Journal of Hematology. 2021;113(5):682–92.
Chevallier P, Labopin M, Cornelissen J, Socié G, Rocha V, Mohty M. Allogeneic hematopoietic stem cell transplantation for isolated and leukemic myeloid sarcoma in adults: a report from the Acute Leukemia Working Party of the European group for Blood and Marrow Transplantation. Haematologica. 2011;96(9):1391.
Lazzarotto D, Candoni A, Filì C, Forghieri F, Pagano L, Busca A, et al. Clinical outcome of myeloid sarcoma in adult patients and effect of allogeneic stem cell transplantation. Results from a multicenter survey. Leukemia research. 2017;53:74–81.
Chevallier P, Mohty M, Lioure B, Michel G, Contentin N, Deconinck E, et al. Allogeneic hematopoietic stem-cell transplantation for myeloid sarcoma: a retrospective study from the SFGM-TC. Journal of Clinical Oncology. 2008;26(30):4940–3.
Shimizu H, Saitoh T, Tanaka M, Mori T, Sakura T, Kawai N, et al. Allogeneic hematopoietic stem cell transplantation for adult AML patients with granulocytic sarcoma. Leukemia. 2012;26(12):2469–73.

Table 1. Summary of Patient Characteristics and Treatment Response with Respect to BM involvement (Synchronous MS) versus no BM involvement (Isolated MS).

		Median (minimum, maximum) or No. (%) of patients
Variable	Evaluable	All MS Patients	Synchronous MS (N=49)	Isolated MS (N=34)	P-value
Age at diagnosis (years)	83	56 (17, 89)	59 (19, 89)	53 (17, 84)	0.631
Sex (Male)	83	53 (63%)	33 (67%)	19 (56%%)	0.35
Race (non-Caucasian)	83	5 (6%)	3 (6%)	2 (6%)	1.00
WBC at diagnosis (x 10⁹ cells/L)	83	8 (2, 182)	14 (2, 182)	7 (2, 24)	<0.001
Hemoglobin at diagnosis (g/dL)	83	11 (6, 18)	11 (6, 18)	12 (8, 16)	0.003
Platelets at diagnosis (x 10⁹ cells/L)	83	150 (8, 561)	88 (8, 303)	250 (24, 561)	<0.001
Extramedullary site	83				0.646
Soft tissue		25 (30%)	14 (29%)	11 (32%)
Skin/breast		23 (28%)	13 (27%)	10 (29%)
LN/spleen		17 (21%)	13 (27%)	4 (12%)
GI		8 (10%)	4 (8%)	4 (12%)
GU/GYN		7 (8%)	3 (6%)	4 (12%)
CNS		3 (4%)	2 (4%)	1 (3%)
ELN Risk Stratification	74				0.002
Normal Karyotype		32 (43%)	12 (26%)	20 (71%)
Favorable Risk Karyotype		10 (14%)	8 (17%)	2 (7%)
Intermediate Risk Karyotype		13 (18%)	11 (24%)	2 (7%)
Adverse Risk Karyotype		19 (26%)	15 (33%)	4 (14%)
P-values result from a Wilcoxon rank sum test (continuous variables) or Fisher’s exact test (categorical variables).

Table 2. BM; bone marrow, MS; myeloid sarcoma, BM/MS; bone marrow and myeloid sarcoma, CR/CRi; complete remission/incomplete remission, PR; partial remission, R; refractory disease. ^*Intensive therapy; 7+3 or high dose cytarabine, hyper CVAD, CLAG-M, MEC. ^**Non-intensive therapy; azacitidine, cladribine, decitabine + venetoclax, gemtuzumab.

Initial Treatment Response

Treatment Group (n)

Location

Response

All MS patients, n (%)

Isolated MS, n (%)

Synchronous MS, n (%)

Intensive Therapy^* (45)

BM

CR/CRi

R

35 (78%)

10 (22%)

11 (79%)

3 (21%)

24 (77%)

7 (23%)

MS

CR

PR

R

31 (69%)

8 (18%)

6 (13%)

10 (71%)

3 (21%)

1 (7%)

21 (68%)

5 (16%)

BM & MS

CR

29 (64%)

10 (71%)

19 (61%)

Intensive + Localized Therapy (23)

BM

CR/CRi

R

15 (65%)

8 (35%)

7 (54%)

6 (46%)

8 (80%)

2 (20%)

MS

CR

PR

R

11 (48%)

5 (22%)

7 (30%)

7 (54%)

1 (8%)

5 (38%)

4 (40%)

2 (20%)

BM & MS

CR

11 (47%)

7 (54%)

4 (40%)

Non-intensive^** +/- Localized Therapy (8)

BM

CR/CRi

R

2 (25%)

6 (75%)

1 (33%)

2 (66%)

1 (20%)

4 (80%)

MS

CR

PR

R

2 (25%)

1 (12%)

5 (62%)

1 (33%)

----

2 (66%)

1 (20%)

3 (60%)

BM & MS

CR

2 (25%)

1 (33%)

1 (20%)

Localized Therapy Alone (5)

BM

CR/CRi

R

2 (40%)

3 (60%)

2 (50%)

----

1 (100%)

MS

CR

PR

R

2 (40%)

----

3 (60%)

2 (50%)

----

2 (50%)

----

1 (100%)

BM & MS

CR

2 (40%)

2 (50%)

0 (0%)

Supportive Care (2)

BM

CR/CRi

R

----

2 (100%)

----

2 (100%)

MS

CR

PR

R

-----

----

2 (100%)

----

2 (100%)

BM & MS

CR

0 (0%)

There is NO conflict of interest to disclose.

Extramedullary Acute Myeloid Leukemia in the De Novo Setting: Clinicopathologic Characteristics, Molecular Analysis, and Treatment Outcomes

Status:

Version 1

Abstract

Figures

Introduction

Subjects And Methods

Results

Patient Characteristics

Bone Marrow Karyotype, Targeted Cytogenetics, and Molecular Analysis

Cytogenetic and Molecular Analysis of Extramedullary Site in Isolated MS

Treatment and Treatment Response

EFS and OS

AlloSCT Results

Discussion

Declarations

References

Tables

Additional Declarations

Status:

Version 1