Clinical outcome of patients with recurrent or refractory localized Ewing's sarcoma family of tumors: A retrospective report from the Japan Ewing Sarcoma Study Group

Abstract Background Patients with Ewing's sarcoma family of tumors (ESFT) who experience relapse or progression have a poor prognosis. Aim This study aimed to identify the prognostic and therapeutic factors affecting overall survival (OS) of patients with recurrent or refractory localized ESFT. Methods and results Thirty‐eight patients with localized ESFT who experienced first relapse or progression between 2000 and 2018 were retrospectively reviewed. The 5‐year OS rate of the entire cohort was 48.3% (95% confidence interval, 29.9%‐64.5%). Multivariate analysis of OS identified time to relapse or progression, but not stem cell transplantation (SCT), as the sole independent risk factor (hazard ratio, 35.8; P = .002). Among 31 patients who received salvage chemotherapy before local treatment, 21 received chemotherapy regimens that are not conventionally used for newly diagnosed ESFT. The objective response rate to first‐line salvage chemotherapy was 55.2% in the 29 evaluable patients. Time to relapse or progression was significantly associated with response to first‐line salvage chemotherapy (P = .006). Conclusions The present study fails to demonstrate significant clinical benefit of SCT for recurrent or refractory localized ESFT. Recently established chemotherapy regimens may increase the survival rate of patients with recurrent or refractory localized ESFT while attenuating the beneficial effect of SCT.

Sakyo-ku, Kyoto 606-8507, Japan. Email: umeume@kuhp.kyoto-u.ac.jp Aim: This study aimed to identify the prognostic and therapeutic factors affecting overall survival (OS) of patients with recurrent or refractory localized ESFT.
Methods and results: Thirty-eight patients with localized ESFT who experienced first relapse or progression between 2000 and 2018 were retrospectively reviewed. The 5-year OS rate of the entire cohort was 48.3% (95% confidence interval, 29.9%-64.5%). Multivariate analysis of OS identified time to relapse or progression, but not stem cell transplantation (SCT), as the sole independent risk factor (hazard ratio, 35.8; P = .002). Among 31 patients who received salvage chemotherapy before local treatment, 21 received chemotherapy regimens that are not conventionally used for newly diagnosed ESFT. The objective response rate to first-line salvage chemotherapy was 55.2% in the 29 evaluable patients. Time to relapse or progression was significantly associated with response to first-line salvage chemotherapy (P = .006). led to markedly improved outcome of patients with ESFT. Consequently, intensified multidrug chemotherapy, achieved by increasing the dose or by decreasing the interval between chemotherapy cycles, in combination with surgery and radiotherapy has contributed to a high survival rate of 70% to 80%. However, approximately 30% to 40% of patients with localized ESFT experience relapse or progression. [1][2][3][4][5][6][7] The prognosis of these patients is extremely poor, with a longterm survival rate of approximately 10% to 30%. [8][9][10][11][12][13] Time to relapse or progression, 8,10,12,13 type of relapse or progression, 11,12 response to first-line salvage chemotherapy, 13 and stem cell transplantation (SCT) are strong prognostic factors for overall survival (OS). 13 Because standard salvage chemotherapy has not been established to date, another combination of these key drugs was previously used as salvage chemotherapy for recurrent or refractory ESFT, in combination with local treatment (surgery and radiotherapy) for the primary site and/or metastasis. However, recurrent or refractory ESFT is often resistant to these chemotherapy regimens. Recently established chemotherapy regimens, including IFM + VP16 + carboplatin (CBDCA, ICE), topotecan (TPT) + CPA, temozolomide (TMZ) + irinotecan (IRI), and gemcitabine (GEM) + docetaxel (DOC), are effective in a fraction of patients with recurrent or refractory ESFT. [14][15][16][17][18] These regimens may change the clinical impact of prognostic factors over time. In the current study, we retrospectively analyzed the clinical outcomes of patients with localized ESFT who experienced relapse or progression to evaluate the prognostic factors affecting OS in the recent era. were excluded due to a lack of data on survival status (n = 2) or EWS-ETS fusion gene (n = 6). One patient who did not relapse was also excluded. EWS-ETS fusion genes, including EWS-FLI1 (n = 23), EWS-ERG (n = 1), and EWS-FEV (n = 2), were detected in 26 patients by reverse transcription polymerase chain reaction. In the remaining 12 patients, the EWSR1 translocation was detected by fluorescent in situ hybridization (FISH) using break-apart probes. In total, 38 patients who were diagnosed as ESFT by molecular testing were analyzed.

| Study design and data collection
Relapse or progression was confirmed by imaging including computed tomography (CT), magnetic resonance imaging, or positron emission tomography-CT in all patients. Histological analyses were not routinely performed to confirm relapse or progression. Relapse-free interval (RFI) was defined as the time from initial diagnosis to first relapse, as previously reported. 13 A cut-off RFI value of 2 years was set according to the previous report. 13 Radiological response to chemotherapy was evaluated according to the RECIST guidelines (version 1.1). 19

| Statistical analysis
The probability of OS, defined as the duration of survival between first relapse or progression and either death or the last follow-up,    Figure 1A). Among the 23 patients experiencing relapse or progression off therapy, there was no significant difference in response according to the type of chemotherapy (P = .270, Figure 1B).

| Impact of local treatment on clinical outcome
To evaluate the impact of local treatment for recurrent or refractory ESFT, we analyzed the OS rates of patients with primary site or metastatic disease alone. Among nine patients with primary site involvement alone, the 5-year OS rate was higher in five patients who underwent surgery (80.0%,; 95% CI, 20.4%-96.9%) than in four patients who did not, although the difference was not statistically significant (0%, P = .384, Figure S1a). Among 26 patients with metastasis alone, the 5-year OS rate was higher in eight patients who underwent surgery with or without radiotherapy (85.7%; 95% CI, 33.4%-97.9%) than in 13 patients receiving radiotherapy alone (29.9%; 95% CI, 7.5%-57.0%) or five patients without local treatment (60.0%; 95% CI, 12.6%-88.2%), although the difference was not statistically significant (P = .196, Figure S1b).

| DISCUSSION
The 5-year OS rate in our patient cohort was 48.3%, which is relatively higher than that of patients with recurrent or refractory localized ESFT reported previously (20%-30%). 8,13 The objective response rate to first-line salvage chemotherapy (55.2%) was also higher than that reported previously. 9,13 In the present study, more than half of the patients received ICE, TPT-based chemotherapy, and TMZ + IRI ± VCR, which are not conventionally used for newly diagnosed ESFT, whereas nearly half of the patients conventionally received IE in previous reports. 9,13 These observations suggest that recently established regimens may increase the response rate to chemotherapy and the survival rate of patients with recurrent or refractory localized ESFT.
Another possible explanation for the difference in OS is that the present study included a higher proportion of younger patients with a better outcome, although age at initial diagnosis was not identified as an independent prognostic factor. Furthermore, selection bias poten- Regarding the type of relapse or progression, the outcome after combined relapse or progression is markedly inferior to that after local relapse or progression alone or metastatic relapse or progression alone. 11,12 In the present study, the three patients with combined relapse or progression died from the disease; however, the type of relapse or progression was not identified as an independent prognostic factor probably because of the paucity of available data.
High-dose chemotherapy with SCT is used for recurrent or refractory ESFT to overcome relative resistance to chemotherapy. Previous reports demonstrate that the OS rate of patients who receive local therapy for primary tumors or metastatic disease is significantly higher than that of patients who do not. 8,10 The present study showed a similar trend: however, significant differences were not observed because of the paucity of available data. Furthermore, the data should be interpreted with caution because patients with an expected poor prognosis associated with the contraindication to local treatment were treated with chemotherapy alone.
The present study had several limitations. First, it was a retrospective analysis of data from a heterogeneous group of patients, which hampered the statistical evaluation of certain prognostic and therapeutic factors affecting clinical outcome. Second, we did not examine the histological response to chemotherapy because of a lack of data in most patients. Lastly, the follow-up period was too short to evaluate late adverse effects, such as secondary malignancies and infertility. Nonetheless, the relatively large cohort of patients with rare recurrent or refractory localized ESFT in the present study led to the important observation that recently established chemotherapy regimens may increase the survival rate of recurrent or refractory localized ESFT without SCT. Further prospective study is required to establish personalized targeted therapies for patients experiencing relapse or progression on therapy with extremely poor prognosis.

ACKNOWLEDGMENTS
We thank all doctors who responded to the questionnaire.

SUPPORTING INFORMATION
Additional supporting information may be found online in the Supporting Information section at the end of this article.