Real-world experiences in patients with Ewing sarcoma treated at a specialist centre in Turkey

Highlights

• •Survival in Ewing sarcoma is still low according to real-world data.
• •Metastatic disease at presentation and positive margins after surgery are of prognostic significance in Ewing sarcoma.
• •Lack of aggressive local treatment reduces survival in Ewing sarcoma.

Abstract

Objectives

The present study evaluates the clinical outcomes of children, adolescents and adults with Ewing sarcoma and identifies the prognostic factors.

Methods

Included in the study were 222 pediatric and adult patients diagnosed with Ewing sarcoma (EwS) who were followed up between 1992 and 2019, and whose data were analyzed retrospectively.

Results

The median age of 131 male and 91 female patients included in the study was 13 (1–64). The median follow-up duration of the survivors was 79 months (range, 11–182 months). The 3-year EFS rate of the 222 patients was 34 % (Confidence Interval (CI) (0.158–0.242 %) and the OS rate was 54 % (CI, 0.289–0.590 %). For the non-metastatic patients, the 3-year EFS rate was 47 % and the OS was 68 %, while for the metastatic patients the 3-year EFS rate was 13 % and the OS was 30 %. Of the patient sample, 81 (36, 5 %) survived, of whom 72 were continuously free of disease while the disease persisted in nine, and three developed a secondary neoplasm (2 of whom subsequently died while one survived disease-free). Of the 129 patients who relapsed with metastases and/or local recurrence, eight survived and are disease-free, nine are alive with uncontrolled disease; five were lost to follow-up and 107 died.

Conclusion

The findings of the present study suggest metastatic disease at presentation and positive margins after surgery to be of prognostic significance in EwS. Disruptions in aggressive local treatments may reduce the chances of cure in EwS.

1. Introduction

Ewing sarcoma (EwS) accounts for around 2 % of all cancers in children, and is the second most common primary bone tumor in childhood and adolescence. Ewing sarcoma can be located in any part of the body but most commonly originates from the pelvis and proximal long bones [1], [2]. Some 20–25 % of patients present with metastasis at diagnosis [2]. The use of chemotherapy for the treatment of localized EwS has increased survival rates from 10 % to around 70–80 %, although there is little effect on the survival of those who are metastatic at diagnosis [3]. The efforts of national and international groups in Europe and North America and collaborations between pediatric and medical oncologists have led to the development of multimodal treatment strategies for EwS, however, 5-year survival in metastatic patients is still below 30 %, and recurrent forms of disease still have a poor prognosis [4].

The present study evaluates the clinical outcomes of children, adolescents and adults with EwS with a view to determining the prognostic factors.

2. Patients and methods

2.1. Patient population

Included in the study were pediatric, adolescent and adult patients with EwS of the bone and soft tissue who were treated between June 1992 and December 2019 in the pediatric and medical oncology clinics of Ankara Oncology Hospital, and who met the inclusion criteria of a histological diagnosis of EwS, no previous treatment and no previous history of malignancy. The data were collected from the patients’ medical oncology files, and the patients’ age, sex, primary tumor site and size, metastasis, treatment, treatment failure site and rate, and recurrence were recorded. Informed consent was obtained for all patients before initiating treatment. Ethics committee approval of the study was obtained from Ankara Bilkent City Hospital Ethics Committee (Date: 12.07.2023, number: E2-23–4283).

2.2. Diagnosis and pre-treatment evaluation

Ewing sarcoma diagnoses were based on an examination of specimens obtained through incisional or open biopsy, and all patients underwent standard histological and immunohistochemical analyses. The histopathological diagnosis of Ewing sarcoma was based on the presence of a small blue round cell tumor that stains positively with CD99 and the negative immunohistochemical markers of lymphoma, rhabdomyosarcoma and neuroblastoma. Translocation [t (11;22)(q24;q11.2–12)] tests have been routinely used in our institute since 2010, it was performed to patients diagnosed after this date. The paraffin blocks of patients diagnosed with EwS in other centers were reviewed by a pathologist experienced in bone tumors in our hospital to confirm the diagnosis.

Pre-treatment staging examinations included computed tomography or magnetic resonance imaging of the primary tumor, plain chest X-ray and thoracic tomography for the detection of lung metastases, Tc-99 m bone scans for bone metastases, bone marrow biopsy and in patients diagnosed since 2009 PET-CT examination.

2.3. Treatment

Details of the individual protocols of 98 pediatric patients in the present study have been previously published [5]. The patients were mostly treated with one of three different protocols: European Intergroup Cooperative Ewing’s Sarcoma Study (EICESS)-92, INT-0091 or Euro-EWING 99 [6], [7], [8]. Local treatment after cytotoxic therapy included surgery only, surgery followed by radiotherapy, or radiotherapy only for inoperable tumors. Radiotherapy was given in doses of 40–65 Gy (1.8–2 Gy/day) for primary tumors, 45 Gy (1.5 Gy/day) for metastases and 12 Gy (3 Gy/day) for palliation.

2.4. Statistical analysis

The treatment results included event-free survival (EFS) and overall survival (OS). Event-free survival was defined as the time from diagnosis to the time of the first event or the last patient contact, while overall survival was defined as the time from diagnosis to the time of death or the last patient contact. The Kaplan-Meier method was used to calculate the EFS and OS curves. An event was defined as local or distant recurrence, progressive disease, resistant disease, secondary malignancy or death. Univariate comparisons of different groups of patients were made using the log-rank test. The Cox regression test was used to identify any differences after adjusting for several factors and determining hazard ratios (HR). A Chi-square test was applied to compare the distribution of clinical factors at the time of diagnosis.

3. Results

Included in the study were 222 newly diagnosed cases of EwS. Of the patients, 185 were treated in a pediatric hospital and 37 in an adult hospital.

3.1. Patients characteristics

The median age of the study participants was 13 years (1–64) and the male: female ratio was 1.44:1.0. The primary tumor was mostly located in an extremity (57 %), followed by the pelvic region (14 %), chest wall (12 %), soft tissue (8 %), vertebrae (6 %) and the head (3 %). The median symptom duration was 3 months (1–36). The most common presenting symptom was pain and swelling (42 %), followed by pain (32 %). Of the patients, 4 % presented with a neurological deficit and 2 % with a pathological fracture. The median tumor size at diagnosis was 10 cm (2–21) and the median lactate dehydrogenase (LDH) value was 339 IU/L (69–2500).Of the total, 83 patients (37 %) presented with metastasis at the time of diagnosis.

3.2. Chemotherapy

Chemotherapy was administered to 217 patients, with a median number of cycles of 14 (range, 0–19). Five patients didn’t receive chemotherapy due to families’ refusal. Almost all of the adult patients were treated with the VAC/IE protocol, while pediatric patients were treated with various protocols, including INT-0091, EICESS and Euro-EWING 99. Of the 17 patients treated with the Euro-EWING 99 protocol, six underwent myeloablative chemotherapy followed by autologous stem-cell transplantation (SCT) with busulfan-melphalan (BU-MEL). Of these six patients, four showed early progression after HDT/SCT and died soon after not responding to second-line treatment. The remaining two patients were reported to be alive and disease-free at 5 and 6 years follow-up. Due to early progression after high-dose therapy (HDT)/SCT in four patients, HDT/SCT tended to be avoided in subsequent patients.

3.3. Local treatment

For the local control of the primary tumor, 59 (27 %) patients were treated with surgery, 58 (26 %) with radiotherapy, and 92 (41 %) with surgery and radiotherapy. Of the total, 114 patients underwent radiotherapy for the primary tumor, 14 underwent radiotherapy for both the primary tumor and metastases, and nine received radiotherapy only to metastases. Of the 118 patients whose radiotherapy dose was known, 27 received less than 50 Gy, and 91 received 50 Gy or more. Furthermore, 15 (7 %) received no local treatment (5 patients progressed and died after the first three cycles of chemotherapy, 5 were lost to follow-up before local therapy, and 5 refused surgery or radiotherapy).

Among the patients with extremity tumors, limb-sparing surgery was performed in 77 (61 %) and amputation in 11. Among those with pelvic tumors, 10 (32 %) underwent limb-sparing surgery or wide resection and 21 were inoperable. Among the 151 patients treated with surgery, the surgical margins were negative in 91 (60 %), positive in 40 (26 %) and unknown in 20 (14 %). Of the 71 patients treated with neoadjuvant chemotherapy followed by surgery, 22 (31 %) had a good histological response to chemotherapy and 49 (69 %) had a poor histological response. The patient and treatment modalities according to the presence of metastasis are presented in Table 1.

Table 1.

Patient and Treatment Characteristics According to the Presence of Metastasis.

	Non-metastatic patients		Metastatic patients			Total
	n	%	n	%	n	%	P
Age (years)
≤15	100	72	59	71	159	72	0,891
>15	39	28	24	29	63	28
Sex
Male	90	65	41	49	131	59	0,024
Female	49	35	42	51	91	41
Duration of symptoms*
≤3 months	77	58	35	47	112	54	0,133
>3 months	57	42	40	53	97	46
Site
Extremity	79	57	47	57	126	57	0,665
Chest wall	19	14	8	10	27	12
Pelvic region	16	11	15	18	31	14
Soft tissue	13	9	5	6	18	8
Vertebrae	8	6	6	7	14	6
Head-neck	4	3	2	2	6	3
Tumor size (cm)*
≤8	62	49	21	28	83	42	0,003
>8	63	51	53	72	116	58
Protocol of chemotherapy
EICESS	62	45	35	42	97	44	0,433
VAC/IE	61	44	32	38	93	42
EuroEwing	10	7	7	9	17	7
Other	4	3	6	7	10	5
Not applied	2	1	3	4	5	2
Radiotherapy
Not applied	44	32	28	34	72	32	0,749
Applied	95	68	55	66	150	68
Surgery
Not applied	35	25	36	43	71	32	0,005
Applied	104	75	47	57	151	68

*These subgroups have some missing data.

3.4. Survival and prognostic factors

The median follow-up time of the surviving patients (n = 89) was 79 months (range, 11–182 months). The 3- and 5-year EFS rates of the 222 patients were 34 % and 32 %, respectively (Confidence Interval (CI) (0.158–0.242 %), and the OS rates were 54 % and 43 %, respectively (Confidence Interval (CI) (0.289–0.590 %) (Fig. 1).

Fig. 1.

Overall and event-free survival of 222 patients.

Among the non-metastatic patients, the 3- and 5-year EFS rates were 47 % and 43 %, and for the metastatic patients, the 3- and 5-year EFS rates were 13 % and 12 %, respectively (Fig. 2, P < 0.001). Among the non-metastatic patients, the 3- and 5-year OS rates were 68 % and 55 %, and for the metastatic patients OS rates were 30 % and 22 %, respectively (Fig. 3, P < 0.001).

Fig. 2.

Event-free survival of 222 patients according to the presence of metastasis.

Fig. 3.

Overall survival of 222 patients according to the presence of metastasis.

As can be seen in Table 2, univariate analyses revealed the EFS rate to be significantly higher in those with a normal LDH at presentation (36 % versus 24 %; P = 0.003), in patients treated with surgery (35 % versus 24 %; P = 0.005) and in patients with negative surgical margins (44 % versus 22 %; P = 0.011).

Table 2.

Univariate analayses of prognostic factors according to the presence of metastasis.

	Non-metastatic patients			Metastatic patients			Total
	n	EFS (%) 5y	P	n	EFS (%) 5y	P	n	EFS (%) 5y	P
Age (years)
1-14	83	44	0,88	52	11	0,32	135	32	0,53
15–35	51	42		28	14		79	32
>36	5	38		3	0		8	23
Sex
Male	90	43	0,48	41	13	0,48	131	33	0,55
Female	49	39		42	8		91	30
Duration of symptoms*
≤3 months	77	35	0,02	35	12	0,88	112	28	0,16
>3 months	57	52		40	14		97	37
Site
Extremity	79	43	0,032	47	13	0,36	126	32	0,03
Chest wall	19	63		8	0		27	46
Pelvic region	16	37		15	20		31	28
Soft tissue	13	27		5	0		18	19
Vertebrae	8	28		6	0		14	22
Head-neck	4	25		2	50		6	33
Localization
Extremity	83	42	0,88	47	13	0,50	130	31	0,76
Limb	56	44		36	10		92	31
Tumor size (cm)*
≤8	62	49	0,85	21	28	0,36	83	41	0,07
>8	63	51		53	72		116	26
LDH
Normal	60	44	0,09	27	18	0,03	87	36	<0.001
Elevated	41	37		32	6		73	24
Protocol of chemotherapy
EICESS	62	45	0,81	35	42	0,41	97	44	0,43
VAC/IE	61	44		32	38		93	42
EuroEwing	10	7		7	9		17	7
Other	4	3		6	7		10	5
Not applied	2	1		3	4		5	2
Radiotherapy
Not applied	44	41	0,91	28	4	0,12	72	28	0,52
Applied	95	44		55	16		150	34
Surgery
Not applied	35	25	0,15	36	43	0,27	71	24	<0.001
Applied	104	75		47	57		151	35
Surgical marginsa
Negative	61	54	0.05	26	21	0.03	91	44	0,011
Positive	27	32		13	0		40	22

*These subgroups have some missing data.

Multivariate analyses revealed the presence of metastases at diagnosis and positive surgical margins to be independent and unfavorable prognostic factors for overall survival (Table 3).

Table 3.

Multivariate analysis of variables affecting overall survival.

Variable	Hazard Ratio	95 % CI	P value
Sex
Male	1
Female	0,64	0,30–1,32	0,228
Age
≤15	1
˃15	0,92	0,43–1,96	0,830
LDH
Normal	1
Elevated	1,6	0,83–3,17	0,150
Tumor size (cm)
≤8	1
>8	0,82	0,39–1,72	0,612
Surgical margin
Negative	1
Positive	3,3	1,42–7,95	0,006
Metastasis at diagnosis
None	1
Metastatic disease	5,2	2,41–11,32	<0,001
Local treatment
Surgery	1
Both surgery and RT	0,68	0,31–1,44	0,33
RT	0,22	0,01–2,9	0,25

Among the effects of local treatment, the OS rate was significantly higher in patients treated with surgery than in patients treated with radiotherapy (48 % versus 36.6 %; p = 0.04), while patients treated with both surgery and radiotherapy had a prognosis similar to that of those treated with surgery alone (5-year OS = 45 %).

Among the 222 patients, 129 (58 %) experienced relapse and disease progression, with 19 experiencing a local relapse, 46 developing distant metastasis (18 in the lung, 25 in the bone, and 3 in both the lung and bone), eight experiencing a combined relapse (2 local and lungs, 6 local and bone), 13 developing local progressive disease, 33 metastatic progressive disease and 10 resistant disease.

Finally, 81 (36,5%) of the 222 patients were still alive, of whom 72 were disease-free while the disease persisted in nine; and 120 (54 %) patients died of the disease 1–117 months after relapse (median: 6 months). Of the total 222 patients, 19 (8, 5 %) were lost to follow-up and two (1 %) continued their treatment at another center.

Furthermore, four died of sepsis, and three (1.3 %) developed a secondary malignancy (acute lymphoblastic leukemia (ALL) at 25 months, myelodysplastic syndrome at 28 months and acute myeloid leukemia at 61 months).

4. Discussion

Previous studies reveal that significant improvements have been achieved in 5- and 10-year survival rates for localized EwS, while advances in treatments for metastatic disease have been much less successful [9]. The Ankara Oncology Hospital is one of the largest cancer centers in Turkey, and the presence of a large orthopedic clinic means that patients with bone tumors frequently present for treatment, primarily from rural areas and with a low socioeconomic status. Over 27 years, 222 EwS patients were followed up by the center, among which the pediatric patients were treated with various protocols, whereas the adult patients were almost exclusively treated with INT-0091. As a result, most patients were treated with either EICESS or INT-0091.

The EICESS Group published their preliminary results in 1999, and reported 3-year EFS rates of 66 % in non-metastatic patients, 43 % in patients with isolated pulmonary metastases, and 29 % in other metastatic patients [6]. They further reported a 3-year EFS of 54 % in high-risk patients treated with EVAIA and 74 % in standard-risk patients treated with VAIA [10].

In the National Cancer Institute protocol study INT-0091, the reported 5-year EFS and OS rates of patients without metastasis were 69 % and 72 %, respectively, compared to 22 % and 34 % in those with metastasis [7].

Our survival results are notably lower than those reported in European studies, and this can be attributed to several reasons, although the most significant is the large proportion of metastatic patients in our study than in those mentioned above (38 % vs. 23 % in INT-0091, 32 % in EICESS-92) [6], [7]. Several studies reported that primary metastatic EwS has a poor prognosis [11], [12], [13], [14], [15]. Patients with isolated lung/pleural metastases had a better prognosis than those metastatic to other sites (3-year EFS, 29 % vs. 52 %), while those with recurrent or refractory disease had a dismal prognosis [4]. In our opinion, the main factor influencing the results was the lack of aggressive local treatment in many patients. Whelan J. et al., in their analysis of the EICESS-92 study, reported an unexpected difference in survival outcomes between the two study groups: the United Kingdom Children’s Cancer and Leukemia Group (CCLG) and the German Pediatric Oncology and Hematology Group (GPOH) in an international trial with 5-year EFS and OS rates of 43 % and 52 % reported by the CCLG group, compared to 57 % and 66 % reported by the GPOH group, respectively [16]. The combined use of surgery and radiotherapy in the CCLG patients was low (18 %) and most underwent only a single local treatment (72 %), while the use of preoperative radiotherapy was low (3 %) [16]. Our results were very similar to those reported by the CCLG in the EICES-92 study. Preoperative radiotherapy was applied to 9 % of the patients in the present study. Local recurrence (with or without metastasis) was the initial event in 31 % of our patients, which was even higher than in the CCLG patients (22 %), and our local failure rates were also higher than Children's Oncology Group protocols [17]. Although a decreasing trend can be seen toward the last decade of our study, a significant proportion of the patients who developed an event refused second-line treatment, and 68 % of those with a local recurrence subsequently died.

In our series, the patients with normal serum LDH at diagnosis had a favorable prognosis compared to patients with elevated levels. The mean time to relapse in those with LDH levels under 1000 IU/L was significantly longer compared to those with LDH levels above 1000 IU/L (54 vs. 24 months, p = 0.05). Furthermore, a greater proportion of patients with high LDH levels had an early relapse compared to those with normal LDH levels (69 % vs 47 %, p = 0.006). These data confirms previous reports and it is possible to conclude that if EwS patients with normal LDH levels relapse, this occurs late [18], [19]. Although high LDH levels had a negative effect on EFS in the entire group and in metastatic patients, we could not show this in non-metastatic patients.

Kutluk et al. reported 5-year EFS and OS rates of 21 % and 33 %, respectively, in a study of 133 patients [20]. Furthermore, Kebudi et al. reported 5-year EFS and OS rates of 60 % and 71 % in non-metastatic patients, and 24 % and 30 % in metastatic patients [21]. In another study from Turkey, 5-year EFS and OS rates of 52.5 % and 60 %, respectively, were reported in 50 children with EwS [22]. In the present study, the EFS and OS rates were comparable to previously reported rates from Turkey. Although our results are similar to previous results, a significant improvement in EwS survival rates has been observed in Turkey in the last 20 years. Survival rates of EwS in Turkey were 44 % in 2004, increased to 60 % between 2014 and 2018, and to 63.8 % from 2019 to the present [23].

In adult patients with localized disease, 5-year EFS ranges from 44 %–60 %, and 5-year OS ranges from 49 %–65 % [24], [25], [26]. The 5-year OS rates are in the 30 %–40 % range in metastatic patients [27].

Our series contained 37 adult patients, four of whom were over 50 years of age, and were treated in the medical oncology clinic. Adult EwS is rare and is mostly treated based on pediatric protocols [27]. In the present study, no significant difference was noted between the pediatric and adult patients in terms of either EFS or OS, which may be attributed to the fact that almost all of the adult patients received the VAC/IE protocol, as in children, and local treatment methods were used at a similar rate in both groups. In addition, although the rate of metastasis at the time of diagnosis was higher among the pediatric patients when compared to the adult group, the difference was not statistically significant (39 % vs. 29 %, p = 0.292).

Local recurrence and overall survival rates also depend on adequate safe surgical margins [28], [29]. Bacci et al. reported an event-free 5-year survival rate of 69.5 % in 252 patients with adequate surgical margins and 50 % in 83 patients with inadequate surgical margins [30]. In the present study, negative surgical margins had a positive effect on survival, with 5-year EFS and OS rates of patients with negative surgical margins of 44 % and 56 %, compared to 22 % and 34 % in those with positive margins, respectively.

The pelvis is the second most common site of EwS after the extremities, and is associated with a poor prognosis. An analysis of the INT-0091 study revealed that the local treatment modality (RT, surgery, or both) for pelvic tumors had no effect on EFS or local failure [31]. Lex J.R. et al. reported that preoperative RT can improve local failure-free survival in pelvic tumors [32]. Of the 31 pelvic tumors in our series, six underwent both surgery and RT, and 16 underwent RT alone, and the 5-year EFS of the patients treated with both surgery and RT was 67 %, compared to 25 % in those treated with RT alone (p = 0.03). Although it is difficult to make a definitive judgment due to the small number of patients, we can conclude that if feasible, aggressive local treatment with both surgery and radiotherapy may lead to improved results in pelvic EwS.

In a study comparing induction regimens, VDC/IE chemotherapy was shown to be superior to VIDE for both EFS and OS, without increasing toxicity [33]. Due to the small number of patients receiving VIDE, however, we were unable to compare the two protocols.

It was concluded from our study that the presence of metastases at diagnosis and positive surgical margins both independently affect overall survival in EwS.

Our study has several limitations, the first of which relates to the patients in the series covering a long period being heterogeneous in terms of their clinical features, and the diagnostic and therapeutic approaches. Second, this is a retrospective study reporting the results of a single center, preventing the generalization of the results, while a third limitation relates to missing data on parameters that would have prognostic significance, such as tumor volume and necrosis rate, which were thus not included in the analysis.

5. Conclusion

In summary, while EwS is treated with current multimodal methods, the targeted survival rates were not achieved in our series, as is the case in most centers in Turkey. The lower survival rate in our patient population than in other studies may be attributable to the less effective local treatment strategies and the high rate of metastatic disease at diagnosis, and the frequent refusal of treatment by patients from lower socioeconomic areas. The higher rate of local recurrence reported in the study was attributed to the higher risk of metastatic disease, and ultimately, death. The use of aggressive local treatments would appear to be vital, especially in localized disease, as well as the consideration of new approaches for the initial treatment of metastatic patients. The education of patients and families to encourage the continuation of treatment compliance may also be advisable in developing countries.

Funding

The authors did not receive support from any organization for the submitted work.

CRediT authorship contribution statement

Selma Çakmakcı: Writing – review & editing, Writing – original draft, Software, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Neriman Sarı: Writing – review & editing, Software, Formal analysis, Data curation, Conceptualization. Ebru Atasever Akkaş: Writing – original draft, Visualization, Software, Formal analysis, Conceptualization. Fatih Yıldız: Writing – original draft, Validation, Software, Formal analysis, Data curation. Ebru Karakaya: Writing – original draft, Visualization, Software, Formal analysis, Conceptualization. Bektaş Kaya: Writing – review & editing, Visualization, Supervision, Software, Project administration, Conceptualization. Bedii Şafak Güngör: Writing – review & editing, Supervision, Resources, Methodology, Conceptualization. Ömür Berna Çakmak Öksüzoğlu: Writing – review & editing, Visualization, Supervision, Methodology, Conceptualization. İnci Ergürhan İlhan: Writing – review & editing, Supervision, Project administration, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.