
==== Front
ESMO Open
ESMO Open
ESMO Open
2059-7029
Elsevier

S2059-7029(24)01458-3
10.1016/j.esmoop.2024.103689
103689
Short Communication
Low-grade fibromyxoid sarcoma and sclerosing epithelioid fibrosarcoma, outcome of advanced disease: retrospective study from the Ultra-Rare Sarcoma Working Group
Giani C. 1
Denu R.A. 2
Ljevar S. 3
Gronchi A. 4
Napolitano A. 5
Rosenbaum E. 6
Salawu A. 7
Bajpai J. 8
Connolly E.A. 9
Lee A.T.J. 10
Trent J.C. 11
Koseła-Paterczyk H. 12
Chia-Chen Li Z. 13
Ogura K. 14
Palmerini E. 15
Baldi G.G. 16
Brunello A. 17
Campos F. 18
Cicala C.M. 19
Maki R.G. 20
Wagner A.J. 21
Andelkovic V. 22
Loong H.H. 23
Wong D.D. 24
Jones R.L. 5
Tap W.D. 6
Taverna S.M. 25
Lazar A.J. 26
Demicco E.G. 27
Hong A. 9
Bovee J.V.M.G. 28
Dei Tos A.P. 29
Fletcher C.D.M. 30
Baumhoer D. 31
Sbaraglia M. 29
Schaefer I.M. 30
Miceli R. 3
Stacchiotti S. silvia.stacchiotti@istitutotumori.mi.it
1∗
1 Department of Medical Oncology, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy
2 Division of Cancer Medicine, The University of Texas MD Anderson Cancer Center, Houston, USA
3 Department of Clinical Epidemiology and Trial Organisation, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy
4 Department of Sarcoma Surgery, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy
5 Department of Medical Oncology, The Royal Marsden NHS, London, UK
6 Department of Medicine, Memorial Sloan Kettering Cancer Center, New York, USA
7 Division of Medical Oncology and Hematology, Mount Sinai Hospital and Princess Margaret Cancer Centre, Toronto, Canada
8 Department of Medical Oncology, Tata Memorial Centre, Homibhabha National University, Mumbai, India
9 Central Clinical School, Faculty of Medicine and Health Science, The University of Sydney, Sydney, Australia
10 Department of Medical Oncology, The Christie NHS Foundation Trust, Manchester, UK
11 Department of Medical Oncology, Sylvester Comprehensive Cancer Center, University of Miami, Miami, USA
12 Department of Medical Oncology, Maria Sklodowska-Curie National Research Institute of Oncology, Warsaw, Poland
13 Department of Medical Oncology, National Taiwan University Cancer Center, Taipei, Taiwan
14 Department of Musculoskeletal Oncology, National Cancer Center Hospital, Tokyo, Japan
15 Department of Medical Oncology, IRCCS Istituto Ortopedico Rizzoli, Bologna, Italy
16 Department of Medical Oncology, Hospital of Prato, Azienda USL Toscana Centro, Prato, Italy
17 Department of Oncology, Medical Oncology 1 Unit, Istituto Oncologico Veneto IOV - IRCCS, Padua, Italy
18 Department of Medical Oncology, A.C.Camargo Cancer Center, São Paulo, Brazil
19 Department of Medical Oncology, Vall d’Hebron University Hospital, Vall d’Hebron Institute of Oncology (VHIO), Barcelona, Spain
20 Department of Medical Oncology, University of Pennsylvania Perelman School of Medicine, Philadelphia, USA
21 Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, USA
22 Department of Medical Oncology, Princess Alexandra Hospital (PAH), Woolloongabba, Australia
23 Department of Clinical Oncology, The Chinese University of Hong Kong, Prince of Wales Hospital, Hong Kong SAR, China
24 Department of Anatomical Pathology, PathWest, Sir Charles Gairdner Hospital, Perth, Australia
25 Technology Transfer Office, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy
26 Department of Pathology and Genomic Medicine, The University of Texas MD Anderson Cancer Center, Houston, USA
27 Department of Pathology and Laboratory Medicine, Mount Sinai Hospital and Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Canada
28 Department of Pathology, Leiden University Medical Center (LUMC), Leiden, The Netherlands
29 Department of Pathology, Azienda Ospedale-Università Padova, Padova, Italy
30 Department of Pathology, Brigham and Women’s Hospital and Harvard Medical School, Boston, USA
31 Department of Pathology, University Hospital Basel, Basel, Switzerland
∗ Correspondence to: Dr Silvia Stacchiotti, Fondazione IRCCS Istituto Nazionale Tumori, via Venezian 1, Milan, Italy. Tel: +39-0223902803 silvia.stacchiotti@istitutotumori.mi.it
11 9 2024
9 2024
11 9 2024
9 9 103689© 2024 The Author(s)
2024
https://creativecommons.org/licenses/by-nc-nd/4.0/ This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Background

To present findings from a retrospective study conducted by the Ultra-Rare Sarcoma Working Group on metastatic low-grade fibromyxoid sarcoma (LGFMS), sclerosing epithelioid fibrosarcoma (SEF), and hybrid (H)-LGFMS/SEF across 28 global centres.

Methods

Patients treated at participating institutions from January 2000 to September 2022 were retrospectively selected. Diagnosis was confirmed by expert pathologists. Primary endpoint was progression-free survival (PFS-1) from metastasis detection to first progression or death. PFS-2 was calculated from therapy initiation.

Results

A total of 101 patients were identified (32 LGFMS, 50 SEF, 19 H-LGFMS/SEF). Median (m) follow-up was 62.1 months. mPFS-1 was 28.7, 11.8, and 20.3 months for LGFMS, SEF, and H-LGFMS/SEF, respectively. mOS was 145.8, 41.9, and 113.5 months, respectively. Treatments included anthracycline-based chemotherapy, gemcitabine-based chemotherapy (G), pazopanib, trabectedin, others. mPFS-2 was: 20.1, 5.5, and 3.5 months in H-LGFMS/SEF, SEF, and LGFMS, respectively, with anthracyclines; 19.5, 7.7, and 6.9 months in LGFMS, SEF, and H-LGFMS/SEF, respectively, with pazopanib; 12.0, 9.7, and 3.1 months in H-LGFMS/SEF, LGFMS, and SEF, respectively. Occasional responses occurred with ifosfamide/oral cyclophosphamide, and prolonged stable disease with immune checkpoint inhibitors.

Conclusions

In this series, the largest available, metastatic LGFMS, SEF, and H-LGFMS/SEF showed different courses. Systemic agents have modest efficacy, informing future trials of novel agents for these tumours.

Highlights

• Metastatic LGFMS, SEF, and H-LGFMS/SEF each have a different clinical course and should be evaluated separately.

• Anthracyclines appeared the most active agents in H-LGFMS/SEF, while the most active agent in LGFMS and SEF was pazopanib.

• Ifosfamide/oral cyclophosphamide achieved occasional responses, while immune checkpoint inhibitors prolonged stabilizations.

Key words

ultra-rare sarcomas
low-grade fibromyxoid sarcoma
sclerosing epithelioid fibrosarcoma
advanced disease
systemic therapies
==== Body
pmcIntroduction

Low-grade fibromyxoid sarcoma (LGFMS) and sclerosing epithelioid fibrosarcoma (SEF) are ultra-rare sarcomas.1 They share morphologic, immunohistochemical, and molecular features but are considered distinct entities. A subset of cases exhibits hybrid morphology (H-LGFMS/SEF).2 Most LGFMS and 70% of SEF exhibit robust diffuse cytoplasmic MUC4 expression.3,4 More than 90% of LGFMS and H-LGFMS/SEF carry the FUS::CREB3L2 fusion,5 whereas >60% of SEF have the EWSR1::CREB3L1 fusion.6 Rare alternative fusions (FUS::CREB3L1 or EWSR1::CREB3L1) are reported in LGFMS, while in SEF the EWSR1 is rarely replaced by FUS or PAX5, and CREB3L1 by CREB3L2, CREB3L3 or CREM.7,8 YAP1::KMT2A is reported in MUC4-negative SEF. Recent literature, however, leans towards categorizing these neoplasms as ‘KMT2A-rearranged sarcomas’ rather than SEF.9,10

LGFMS typically affects young adults, whereas SEF predominantly occurs in middle-aged to elderly patients. Soft tissue of extremities and the trunk are more frequently involved. LGFMS is considered a low-grade sarcoma, whereas SEF is biologically aggressive neoplasm with a >50% metastatic rate.10,11 There are no available data on the behaviour of metastatic hybrid LGFMS/SEF (H-LGFMS/SEF).

Current knowledge on the activity of systemic agents in advanced LGFMS and SEF is based on three small retrospective series, including <35 patients.10, 11, 12 This global, collaborative study, including 28 referral sarcoma centres in Europe, America, Asia, and Australia within the Ultra-Rare Sarcoma Working Group (URSWG), aimed at investigating the activity of drugs available for treatment of sarcomas in advanced LGFMS, SEF, and H-LGFMS/SEF. This is particularly relevant, as new agents potentially active in these diseases are under evaluation.

Patients and methods

Patients diagnosed with LGFMS, SEF, and H-LGFMS/SEF and treated at participating institutions between January 2000 and September 2022 were included. Institutional Review Board approval was obtained at each centre. We report here data of metastatic patients.

Study design

Data were extracted from clinical databases. The study design (synopsis available in the Supplementary material, available at https://doi.org/10.1016/j.esmoop.2024.103689) adhered to URSWG’s criteria for retrospective studies in ultra-rare sarcomas.13 Centrally confirmed pathologic diagnosis of LGFMS, SEF or H-LGFMS/SEF was required, including strong MUC4 expression and/or one of the following fusions: FUS/EWSR1; EWSR1/FUS::CREB3L1/CREB3L2/CREM. Primary endpoint was progression-free survival (PFS). Secondary endpoints were overall response rate (ORR) and post-metastasis overall survival (OS).

Statistical analyses

Descriptive statistics summarized population characteristics, while Fisher’s exact test was used to compare ORRs. ORR was calculated for both patients assessable for response and the total number of patients included in each treatment group.

PFS and OS were estimated by the Kaplan–Meier method. PFS-1 was defined as the time from distant metastasis to the first progression or death, irrespective of treatment line or sequence. OS-1 was defined as the time from the first diagnosis of distant metastasis to death or last follow-up. For therapy-specific analyses, PFS-2 and OS-2 were calculated from study therapy initiation. The main analyses compared the three categories (LGFMS versus SEF versus H-LGFMS/SEF); PFS-1 and OS-1 were also estimated in the two categories LGFMS versus SEF+H-LGFMS/SEF.

A propensity score (PS) matching analysis for PFS-1 (variables: age, sex, histology, stage at diagnosis, primary site, surgery) compared patients receiving first-line systemic therapy with those untreated.

Analyses of anthracycline-treated patients explored diagnostic subtype impact, adjusting for therapy line in a Cox model for PFS-2. The differential effect of surgery during treatment and/or ifosfamide in addition to anthracyclines for the three histologic subtypes was explored in multivariable Cox models for PFS-2 and OS-2.

A two-sided P < 0.05 was considered statistically significant. Statistical analyses were carried out with R (version 4.1.2).

Results

A total of 101 cases from 28 institutions were included: 32 LGFMS, 50 SEF, and 19 H-LGFMS/SEF. Table 1 shows patient characteristics. As per the inclusion criteria, all patients had evidence of MUC4-positive immunostaining and/or the presence of one of the predefined disease-specific gene rearrangements.Table 1 Patient characteristics. Histopathological and molecular features, locoregional treatments, number of systemic treatment lines

Patients, N	101	
 Metastatic at diagnosis n (%)	50/101 (49)	
 Metastatic at relapse n (%)	51/101 (51)	
	LGFMS	SEF	Hybrid
LGFMS/SEF	
Patients per group n (%)	32/101 (32)	50/101 (50)	19/101 (18)	
 Metastatic at diagnosis n (%)	20 (63)	24 (48)	6 (32)	
 Metastatic at relapse n (%)	12 (38)	26 (52)	13 (68)	
 Age at metastasis, median (IQR) years	42 (37-53)	48 (32-58)	44 (36-54)	
 Male/female n (%)	17 (53)/15 (47)	32 (64)/18(36)	8 (42)/11 (58)	
Histopathological/molecular features				
MUC4 expressionn(%)				
 Positive	19/32 (59)	36/50 (72)	15/19 (79)	
 Negative	0/32 (0)	0/50 (0)	1/19 (5)	
 Not done	13/32 (41)	14/50 (28)	3/19 (16)	
Patients analysed for FUS rearrangementn(%)	19/32 (59)	4/50 (8)	13/19 (68)	
 Positive	17/32 (53)	4/50 (8)	10/19 (53)	
 FUS::CREB3L2	11/20 (55)	0/4 (0)	5/12 (42)	
 FUS::CREB3L1	0/20 (0)	0/4 (0)	0/12 (0)	
 Other FUS rearrangements	6/20 (30)	4/4 (100)	5/12 (42)	
 Negative	2/32 (6)	0/50 (0)	3/19 (16)	
Patients analysed for EWSR1 rearrangementn(%)	3/32 (9)	36/50 (72)	5/19 (26)	
 Positive	3/32 (9)	32/50 (64)	4/19 (21)	
 EWSR1::CREB3L1	3/3 (100)	26/32 (82)	4/4 (100)	
 EWSR1-CREB3L2	0/3 (0)	2/32 (6)	0/4 (0)	
 Other EWSR1 rearrangements	0/3 (0)	4/32 (13)	0/4 (0)	
 Negative	0/32 (0)	4/50 (8)	1/19 (5)	
IHC/molecular profile (%)				
MUC4 negative/gene rearrangement negative	0/32 (0)	0/50 (0)	0/19	
MUC4 negative/gene rearrangement positive	13/32 (41)	14/50 (28)	4/19 (21)	
MUC4 positive/gene rearrangement negative	12/32 (38)	14/50 (28)	5/19 (26)	
MUC4 positive/gene rearrangement positive	7/32 (21)	22/50 (44)	10/19 (53)	
Primary tumour siten(%)				
 Extremities	15 (47)	15 (30)	8 (42)	
 Abdomen/retroperitoneum	8 (25)	16 (32)	4 (21)	
 Trunk	4 (13)	7 (14)	4 (21)	
 Other	5 (16)	12 (24)	3 (16)	
Site of metastasisn(%)				
 Lung	21 (66)	31 (62)	15 (80)	
 Bone	1 (3)	11 (22)	1 (5)	
 Soft tissues	8 (25)	1 (2)	1 (5)	
 Liver	1 (3)	1 (2)	0 (0)	
 Lymph nodes	0 (0)	1 (2)	1 (5)	
 Other	4 (13)	5 (10)	1 (5)	
Treatments of metastatic disease				
Surgery in synchronous metastatic diseasen(%)				
 No	13 (41)	16 (32)	8 (42)	
 Yes	7 (22)	10 (20)	5 (26)	
 Primary site	4 (13)	8 (16)	3 (16)	
 Metastatic site	3 (9)	2 (4)	2 (11)	
Surgery in metachronous metastatic diseasen(%)				
 No	4 (13)	11 (22)	2 (11)	
 Yes	8 (25)	13 (26)	4 (21)	
 Macroscopic complete resection	5 (16)	5 (10)	4 (21)	
 Macroscopic incomplete resection	2 (6)	3 (6)	0 (0)	
 Missing	1 (3)	5 (10)	0 (0)	
Radiotherapyn(%)				
 No	27 (84)	28 (56)	15 (79)	
 Yes	5 (16)	22 (44)	4 (21)	
Systemic therapiesn(%)				
 No	7 (22)	6 (12)	6 (32)	
 Yes	25 (78)	44 (88)	13 (68)	
 1 Treatment line	6 (19)	10 (20)	5 (26)	
 >1 Treatment line	19 (59)	34 (68)	8 (42)	
Status at lastfollow-upn(%)				
 Alive, no evidence of disease	4 (13)	2 (4)	5 (26)	
 Alive with disease	16 (50)	22 (44)	6 (32)	
 Dead	10 (31)	26 (52)	8 (42)	
 Lost to follow-up	2 (6)	0 (0)	0 (0)	
IHC, immunohistochemistry; IQR, interquartile range; LGFMS, low-grade fibromyxoid sarcoma; SEF, sclerosing epithelioid fibrosarcoma.

At a 62.1-month median (m) follow-up, mPFS-1 was 28.7 months in LGFMS, 11.8 months in SEF, and 20.3 months in H-LGFMS/SEF; mOS-1 was 145.8 months in LGFMS, 41.9 months in SEF, and 113.5 months in H-LGFMS/SEF.

Supplementary Figure S1, available at https://doi.org/10.1016/j.esmoop.2024.103689, shows PFS-1 and OS-1 in LGFMS versus SEF+H-LGFMS/SEF.

The time from metastatic disease to the initiation of systemic therapy was longer in LGFMS, averaging 9 months, compared with SEF and H-LGFMS/SEF (<3 months).

In the anthracycline-based group (anthracyclines) there were: 18 LGFMS, 30 SEF, and 9 H-LGFMS/SEF; in the gemcitabine-based group there were: 7 LGFMS, 19 SEF, and 4 H-LGFMS/SEF; 11 LGFMS, 21 SEF, and 4 H-LGFMS/SEF received pazopanib; 8 LGFMS, 8 SEF, and 3 H-LGFMS/SEF had trabectedin.

Treatment response and outcome

Table 2 reports ORR, mPFS-2, mOS-2 for LGFMS, SEF, and H-LGFMS/SEF. Figures 1 and 2 show PFS-2 and OS-2, respectively. Supplementary Table S1, available at https://doi.org/10.1016/j.esmoop.2024.103689, provides details regarding treatment regimens, responses, and pretreatments, by treatment group and histological subtype. PFS-1 and OS-1 for the two LGFMS and SEF+H-LGFMS/SEF groups are presented in Supplementary material, available at https://doi.org/10.1016/j.esmoop.2024.103689.Table 2 Overall response rate (ORR), progression-free survival (PFS), and overall survival (OS) of advanced low-grade fibromyxoid sarcoma (LGFMS), sclerosing epithelioid fibrosarcoma (SEF), and hybrid (H)-LGFMS/SEF, treated with different systemic regimens. Anthracycline-based regimens [LGFMS = 18, SEF = 30, hybrid (H)-LGFMS/SEF = 9], gemcitabine-based regimens (LGFMS = 7, SEF = 19, H-LGFMS/SEF = 4), pazopanib (LGFMS = 11, SEF = 21, H-LGFMS/SEF = 4), trabectedin (LGFMS = 8, SEF = 8, H-LGFMS/SEF = 3). Patients could appear in more than one category if they received more than one regimen

Patients N	101	
	LGFMS	SEF	Hybrid LGFMS/SEF	
Patients per group n (%)	32/101 (31.7)	50/101 (49.5)	19/101 (18.8)	
Total number of patients treated with systemic agents n (%)	25/32 (78)	44/50 (88)	13/19 (68)	
Anthracycline-based regimens		
m-FU, months	34.4 (IQR: 19.8-64.8)	
Patients treated in first line n (%)	17 (94)	21 (70)	8 (89)	
Patients assessable for response n (%)	15 (83)	27 (90)	7 (78)	
ORR (assessable patients), n (%)	1/15 (7)	0/27 (0)	0/7 (0)	
mPFS-2, months (IQR)	3.5 (2.1-12.6)	5.53 (1.8-10.8)	20.1 (2.1-91.1)	
mOS-2, months (IQR)	NR (12.6-NR)	26.4 (16.1-39.9)	29.9 (27.5-91.1)	
Gemcitabine-based regimens		
m-FU, months	29.4 (IQR: 9.6-58.9)	
Patients treated in first line n (%)	2 (29)	4 (21)	1 (25)	
Patients assessable for response n (%)	7 (100)	14 (74)	3 (75)	
ORR (assessable patients), n (%)	1/7 (14)	0/14 (0)	0/3 (0)	
mPFS-2, months (IQR)	9.7 (2.0-9.9)	3.1 (1.9-8.6)	12.0 (6.6-24.9)	
mOS-2, months (IQR)	NR (3.7-NR)	13.3 (11.5-18.8)	24.9 (12.0-40.4)	
Pazopanib		
m-FU, months	21.7 (IQR:12.2-66.3)	
Patients treated in first line n (%)	3 (27)	9 (43)	0 (0)	
Patients assessable for response n (%)	6 (55)	17 (81)	4 (100)	
ORR (assessable patients), n (%)	2/6 (33)	1/17 (6)	0/4 (0)	
mPFS-2, months (IQR)	19.5 (3.7-34.6)	7.7 (2.8-15.2)	6.9 (2.9-39.3)	
mOS-2, months (IQR)	86.0 (13.4-86.0)	24.4 (14.9-38.1)	15.6 (12.0-NR)	
Trabectedin		
m-FU, months	26.2 (IQR:19.6-39.9)	
Patients treated in first line n (%)	0 (0)	0 (0)	1 (33)	
Patients evaluable for response n (%)	7 (88)	8 (100)	2 (66)	
ORR (evaluable patients), n (%)	0/7 (0)	0/8 (0)	0/2 (0)	
mPFS-2, months (IQR)	7.6 (3.7-18.6)	2.1 (1.9-4.8)	4.9 (1.6-NR)	
mOS-2, months (IQR)	NR (24.2-NR)	13.9 (10.7-17.7)	NR (1.6-NR)	
FU, follow-up; H, hybrid; IQR, interquartile range; LGFMS, low-grade fibromyxoid sarcoma; m, median; mOS-2, specific therapy overall survival; mPFS-2, specific therapy progression-free survival; NR, not reached; ORR, overall response rate [% complete response + partial response calculated for assessable patients]; SEF, sclerosing epithelioid fibrosarcoma. Patients could appear in more than one category if they received more than one regimen.

Figure 1 Progression free survival (PFS) curves of advanced low-grade fibromyxoid sarcoma (LGFMS), sclerosing epithelioid fibrosarcoma (SEF), and hybrid (H)-LGFMS/SEF, treated with different systemic regimens. Panel A shows PFS curves of patients treated with anthracycline-based regimens (LGFMS = 18, SEF = 30, H-LGFMS/SEF = 9); Panel B shows PFS curves of patients treated with gemcitabine-based regimens (LGFMS = 7, SEF = 19, H-LGFMS/SEF = 4); Panel C shows PFS curves of patients treated with pazopanib (LGFMS = 11, SEF = 21, H-LGFMS/SEF = 4): Panel D shows PFS curves patients treated with trabectedin (LGFMS = 8, SEF = 8, H-LGFMS/SEF = 3). Patients could appear in more than one category if they received more than one regimen.

Figure 2 Overall survival (OS) curves of advanced low-grade fibromyxoid sarcoma (LGFMS), sclerosing epithelioid fibrosarcoma (SEF), and hybrid (H)-LGFMS/SEF, treated with different systemic regimens. Panel A shows OS curves of patients treated with anthracycline-based regimens (LGFMS = 18, SEF = 30, H-LGFMS/SEF = 9); Panel B shows OS curves of patients treated with gemcitabine-based regimens (LGFMS = 7, SEF = 19, H-LGFMS/SEF = 4); Panel C shows OS curves of patients treated with pazopanib (LGFMS = 11, SEF = 21, H-LGFMS/SEF = 4): Panel D shows OS curves patients treated with trabectedin (LGFMS = 8, SEF = 8, H-LGFMS/SEF = 3). Patients could appear in more than one category if they received more than one regimen.

Anthracyclines

In assessable patients, ORR was 7% (1/15) in LGFMS, 0% (0/27) in SEF, and 0% (0/7) in H-LGFMS/SEF.

Some 81% (46/57) of patients received anthracyclines in first line, 19% (11/57) in further line. Among pretreated patients, 4/11 received a gemcitabine-based regimen, 3/11 pazopanib, 0/11 trabectedin, and 4/11 other regimens.

At a 34.4-month m-follow-up, mPFS-2 was 3.5, 5.5, and 20.1 months in LGFMS, SEF, and H-LGFMS/SEF, respectively (Figure 1A). Median time from metastasis diagnosis to anthracyclines start was 9.0, 2.3, and 3.3 months in LGFMS, SEF, and H-LGFMS/SEF, respectively. H-LGFMS/SEF was the most favourable prognostic group, even after adjusting for treatment line and metastasis-to-treatment interval, as shown in Supplementary material, available at https://doi.org/10.1016/j.esmoop.2024.103689. No significant differences were observed on the effect of additional treatments (namely surgery or ifosfamide) to anthracycline in the three histologic subtypes. The combination of ifosfamide and surgery versus no other treatments, however, exerted a protective effect on PFS-2 in both LGFMS [hazard ratio (HR) 0.66; 95% confidence interval (CI) 0.13-3.21] and H-LGFMS/SEF (HR 0.55; 95% CI 0.05-5.47), and a protective effect on OS-2 in SEF (HR 0.76; 95% CI 0.14-4.17) and H-LGFMS/SEF (HR 0.51; 95% CI 0.04-6.26), as shown in Supplementary material, available at https://doi.org/10.1016/j.esmoop.2024.103689.

Gemcitabine-based regimens

In assessable patients, ORR was 14% (1/7) in LGFMS, 0% (0/14) in SEF, and 0% (0/3) in H-LGFMS/SEF.

Some 23% (7/30) of patients received gemcitabine in first line, 77% (23/30) in further line. Among pretreated patients, 18/23 received an anthracycline-based regimen, 4/23 pazopanib, 3/23 trabectedin, and 9/23 other regimens.

At a 29.4-month m-follow-up, mPFS-2 was 9.7, 3.1, and 12.0 months in LGFMS, SEF, and H-LGFMS/SEF, respectively (Figure 1B).

Pazopanib

In assessable patients, ORR was 33% (2/6) in LGFMS, 6% (1/17) in SEF, and 0% (0/4) in H-LGFMS/SEF.

Some 33% (12/36) of patients received pazopanib in first line, 67% (24/36) in further line. Among pretreated patients, 18/24 received an anthracycline-based regimen, 6/24 a gemcitabine-based regimen, 4/24 trabectedin, and 10/24 other regimens.

At a 21.7-month m-follow-up, mPFS-2 was 19.5, 7.7, and 6.9 months in LGFMS, SEF, and H-LGFMS/SEF, respectively (Figure 1C).

Trabectedin

No responses were seen to trabectedin.

Among pretreated patients (18/19), 18/18 received an anthracycline-based regimen, 4/18 a gemcitabine-based regimen, 0/18 pazopanib, and 3/18 other regimens.

At a 26.2-month m-follow-up, mPFS-2 was 7.6, 2.1, and 4.9 months in LGFMS, SEF, and H-LGFMS/SEF, respectively (Figure 1D).

Other

Responses to ifosfamide were seen in 2/5 SEF, and to oral cyclophosphamide in 1/1 LGFMS. Prolonged disease stabilisations (SD) were observed with immune checkpoint inhibitors (IO): 3/3 in LGFMS, 5/9 in SEF, and 1/1 in H-LGFMS/SEF.

The comparison of first-line treated versus untreated patients did not result in a statistically significant difference in PS-adjusted PFS-1 (HR 1.23, 95% CI 0.62-2.46, P = 0.556).

Discussion

This retrospective study, conducted by the global URSWG, is the largest series available of patients with metastatic LGFMS, SEF, and H-LGFMS/SEF who underwent systemic therapy, and the first reporting H-LGFMS/SEF outcomes. Due to the ultra-rare nature of these tumours,1 the collaboration of almost 30 sarcoma reference centres worldwide was essential to assemble a cohort of 80 patients treated with medical agents. This underscores how, in such rare conditions and in the absence of prospective studies, global collaboration and retrospective studies are essential to advance biologic insights, establish the groundwork for the optimal use of conventional agents in daily practice, and foster the development of new drugs. To ensure the quality of the collected data, in this study we have chosen to include only cases reviewed by expert pathologists and confirmed by the presence of predefined pathognomonic diagnostic markers.

Our study is the first to demonstrate that H-LGFMS/SEF behaves differently to LGFMS and SEF, more closely resembling (but not overlapping with) LGFMS, and should therefore be analysed separately from the other two subtypes. We confirmed the relatively indolent clinical behaviour of LGFMS, even when metastatic. Nevertheless, the majority of patients still required systemic treatment. SEF were confirmed to have a worse prognosis. A longer survival (HR = 0.49; e-S-data, available at https://doi.org/10.1016/j.esmoop.2024.103689) was detected for SEF who underwent metastasectomy, however, consistent with what is recognized for patients with limited versus extensive metastatic disease. These aspects will be crucial to consider in the design of future studies.

Treatment with currently available drugs rarely achieved a significant reduction in tumour size, discouraging their use for cytoreduction. When assessing differential drug sensitivity, we only conducted a multivariable PFS-2 analysis on the relatively largest patient group, those treated with anthracyclines. Small sample sizes prevented the use of multivariable models in other regimen subgroups. With these limitations, we found that the sensitivity to approved drugs was not overlapping across LGFMS, SEF, and H-LGFMS/SEF. While pazopanib (mostly administered as second or further line) appeared to be the most active drug in LGFMS and SEF, anthracyclines (mostly as first line) did better in H-LGFMS/SEF (HR H-LGFMS/SEF versus LGFMS: 0.45), compared with SEF and LGFMS (HR SEF versus LGFMS: 0.87; Supplementary material, available at https://doi.org/10.1016/j.esmoop.2024.103689). A longer mPFS-2 was observed in the few LGFMS and H-LGFMS/SEF treated with gemcitabine-based regimens, ifosfamide, and oral cyclophosphamide as monotherapy. Eventually, we observed a few prolonged SD with IO, consistent with anecdotal case reports,14,15 suggesting that the use of alkylating agents and IO requires further investigation in these diseases.

These results highlight the strong need for novel therapies in LGFMS, SEF, and H-LGFMS/SEF, and provide a benchmark for future trials with new compounds.

Supplementary data

Supplementary Figure

Supplementary data

Supplementary material

Acknowledgements

We are deeply grateful to Barbara Rapp, from the 10.13039/100016614 Connective Tissue Oncology Society (CTOS) for her support in organizing the Ultra-Rare Sarcoma Working Group meeting.

Funding

This work was supported by the 10.13039/501100003196 Italian Ministry of Health , Ricerca Corrente (no grant number).

Disclosure

AN consulting or advisory role: Deciphera, Agenus. AS consulting or advisory role: Boehringer Ingelheim, Knight Therapeutics, Taiho Pharmaceuticals; Honoraria: Medison, Bayer. AL consulting or advisory role: Deciphera, AstraZeneca, Astex; travel grant: Alexion.

EP advisory role: Daiichi Sankyo, 10.13039/100016827 Deciphera Pharmaceuticals , EUSA Pharma, SynOx Therapeutics. GGB advisory role: PharmaMar, Eli Lilly, GlaxoSmithKline, Merck Sharp & Dome, Eisai; consulting fees: Eli Lilly, PharmaMar, AboutEvents; honoraria: PharmaMar, Eli Lilly, GlaxoSmithKline, Merck Sharp & Dome, Eisai, Istituto Gentili; travel grants: Novartis, PharmaMar, Eli Lilly. AB consulting or advisory role: Eli Lilly, Roche, GSK, Eisai, PharmaMar, Boehringer Ingelheim, Deciphera; honoraria: GSK and PharmaMar; travel grants: Ipsen and PharmaMar. AH consulting: OncoBeta, Telix. JB consulting or advisory role: TRACON Pharmaceuticals, Boehringer Ingelheim. RM honoraria: Boehringer. SS honoraria, consultancy or advisory role: Aadi, Agenus, Astex Pharmaceuticals, Bavarian Nordic, Bayer, Boehringer, Daiichi Sankyo, Gentili, Glaxo, Ikena, NEC OncoImmunity, Novartis, PharmaMar, Pharma Essentia, Rain Therapeutics, Regeneron, Servier; institutional financial interests: Advenchen, Bayer, Blueprint, Boehringer, Daiichi Sankyo, Deciphera, Eisai, Epizyme, Eli Lilly, Foghorn, Glaxo, Hutchinson, Karyopharm, Novartis, PharmaMar, RainThera, SpringWorks. All other authors have declared no conflicts of interest.
==== Refs
References

1 Stacchiotti S. Frezza A.M. Blay J.Y. Ultra-rare sarcomas: a consensus paper from the Connective Tissue Oncology Society community of experts on the incidence threshold and the list of entities Cancer 127 16 2021 2934 2942 33910263
2 WHO Classification of Tumours Editorial Board. Soft Tissue and Bone Tumours. Lyon (France): International Agency for Research on Cancer. 2020. (WHO classification of tumours series, 5th ed. Vol. 3).
3 Doyle L.A. Möller E. Dal Cin P. Fletcher C.D.M. Mertens F. Hornick J.L. MUC4 is a highly sensitive and specific marker for low-grade fibromyxoid sarcoma Am J Surg Pathol 35 5 2011 733 741 21415703
4 Doyle L.A. Wang W.L. Dal Cin P. MUC4 is a sensitive and extremely useful marker for sclerosing epithelioid fibrosarcoma: association with FUS gene rearrangement Am J Surg Pathol 36 10 2012 1444 1451 22982887
5 Panagopoulos I. Storlazzi C.T. Fletcher C.D.M. The chimeric FUS/CREB3l2 gene is specific for low-grade fibromyxoid sarcoma Genes Chromosomes Cancer 40 3 2004 218 228 15139001
6 Arbajian E. Puls F. Magnusson L. Recurrent EWSR1-CREB3L1 gene fusions in sclerosing epithelioid fibrosarcoma Am J Surg Pathol 38 6 2014 801 808 24441665
7 Wang W.L. Evans H.L. Meis J.M. FUS rearrangements are rare in ‘pure’ sclerosing epithelioid fibrosarcoma Mod Pathol 25 6 2012 846 853 22388756
8 Memon R.A. Granada C.N.P. Patel C. Gastric sclerosing epithelioid fibrosarcoma harboring a rare FUS-CREM fusion Int J Surg Pathol 29 5 2021 565 570 32964769
9 Massoth L.R. Hung Y.P. Nardi V. Pan-sarcoma genomic analysis of KMT2A rearrangements reveals distinct subtypes defined by YAP1-KMT2A-YAP1 and VIM-KMT2A fusions Mod Pathol 33 11 2020 2307 2317 32461620
10 Blay J.Y. Tlemsani C. Toulmonde M. Sclerosing Epithelioid Fibrosarcoma (SEF) versus Low Grade Fibromyxoid Sarcoma (LGFMS): presentation and outcome in the nationwide NETSARC+ series of 330 patients over 13 years Eur J Cancer 196 2024 113454
11 Chew W. Benson C. Thway K. Clinical characteristics and efficacy of chemotherapy in sclerosing epithelioid fibrosarcoma Med Oncol 35 11 2018 138 30187231
12 Chamberlain F. Engelmann B. Al-Muderis O. Low-grade fibromyxoid sarcoma: treatment outcomes and efficacy of chemotherapy In Vivo 34 1 2020 239 245 31882484
13 Stacchiotti S. Maria Frezza A. Demetri G.D. Retrospective observational studies in ultra-rare sarcomas: a consensus paper from the Connective Tissue Oncology Society (CTOS) community of experts on the minimum requirements for the evaluation of activity of systemic treatments Cancer Treat Rev 110 2022 102455
14 Koerner A.S. Zhou M. Brook A. Yoon S.S. Ganjoo K.N. Response to immunotherapy in sclerosing epithelioid fibrosarcoma: case report and literature review Cureus 15 12 2023 e50967
15 Monga V. Skubitz K.M. Maliske S. A retrospective analysis of the efficacy of immunotherapy in metastatic soft-tissue sarcomas Cancers 12 7 2020 1873 32664595
