Clinical Outcomes and Prognostic Factors for Patients with Malignant Peripheral Nerve Sheath Tumour

Introduction Few studies have described the characteristics and prognostic factors of patients with malignant peripheral nerve sheath tumour (MPNST). In this study, we retrospectively investigated the clinicopathological features, clinical outcomes, and prognostic factors of these patients. Patients and Methods. We recruited patients with MPNST who were treated at our institutions from 1991 to 2020. We collected and statistically analysed information on patient-, tumour-, and treatment-related factors. The median follow-up period was 61 months (range, 1–335.8 months). Results A total of 60 patients (31 males, 29 females) with a median age of 55 years (range, 8–84 years) at initial diagnosis were included. The median tumour size was 7 cm (range, 1.6–30 cm) in the greatest dimension. The 5-year overall survival (OS) rate of all patients was 69.5%. Univariate analysis revealed that large-sized tumour, metastasis at diagnosis, and no surgery of the primary tumour were significantly associated with patients with worse OS. Multivariate analysis identified surgery of the primary tumour as an independent prognostic factor for improved OS. Among patients with localised disease at diagnosis who underwent surgery of the primary tumour at our institutions, the 5-year OS, local recurrence-free survival (LRFS), and metastasis-free survival (MFS) rates were 81.1%, 78.2%, and 70.3%, respectively. Univariate analysis revealed that positive surgical margin was significantly correlated with unfavourable OS and LRFS, and high grade was a poor prognostic indicator for MFS. Conclusion Complete surgical resection with negative surgical margins is necessary for a successful MPNST treatment. Multidisciplinary management of MPNST with aggressive features is important for optimising patient outcomes.

is retrospective study aimed to investigate the clinicopathological features, clinical outcomes, and prognostic factors in patients with MPNST treated at our affiliated hospitals.

Materials and Methods
We designed a multi-institutional retrospective study that was conducted in our institutions. We reviewed the records of each institute between January 1991 and June 2020. e patient eligibility criteria included MPNST diagnosis, as pathologically confirmed by an expert musculoskeletal tumour pathologist at each institute. MPNST usually consists of spindle cells demonstrating nuclear atypia, increased mitotic activity, and geographic areas of necrosis. MPNST demonstrates variable expression of immunohistochemical neural markers such as S100 and SOX10. All cases were diagnosed as MPNST.
is study was approved by the Institutional Review Boards of each institution.
A total of 60 patients with MPNST who were treated at our hospitals were included in this study. Information on patient-related factors (age, sex, and NF1 status), tumourrelated factors (site of primary lesions; tumour size, depth, and histological grade; and presence or absence of metastasis at initial diagnosis), treatment-related factors (surgery of the primary tumour and metastatic lesions, tumour surgical margin, and chemotherapy and radiotherapy status), local and distant relapse, follow-up period, and oncological outcome at final follow-up were anonymously collected from the medical charts of the patients. We were unable to obtain data on tumour size, histological grade, and surgical margin in three patients who received their first surgery at other hospitals. Data on tumour depth could not be collected in one patient.
We calculated the overall survival (OS) from the date of diagnosis to the date of death from any cause or the last follow-up visit. In patients who underwent surgery, we calculated the local recurrence-free survival (LRFS) from the date of surgery to the date of local recurrence or the last follow-up. In patients without distant metastasis at presentation, we calculated the metastasis-free survival (MFS) from the date of diagnosis to the date of metastasis or the last follow-up. Patients without surgery of the primary lesion and with metastasis at initial referral were excluded from the LRFS and MFS analysis, respectively. We calculated the OS, LRFS, and MFS using the Kaplan-Meier method and evaluated the impact of prognostic factors using the log-rank test in a univariate analysis.
We conducted a multivariate analysis using the Cox proportional hazards model, with variables chosen by using a forward conditional stepwise approach. Differences were considered significant when p values were <0.05. e EZR software (Saitama Medical Center, Jichi Medical University, Saitama, Japan), a graphical user interface for R ( e R Foundation for Statistical Computing, Vienna, Austria), was used in the statistical analyses.
e median follow-up period was 61.2 months (range, 1-335.8 months) for all patients. e patient-, tumour-, and treatment-related characteristics of the 60 cases are presented in Table 1. e 31 male (51.7%) and 29 female (48.3%) patients had a median age of 55 years (range, 8-84 years) at initial diagnosis. A total of 27 patients (45%) were ≤50 years of age, and 33 patients (55%) were >50 years.
Fifty-two patients (86.7%) underwent surgery for the primary tumour. e surgical margin was described as negative when the pathologist found no tumour cells at the edge of the material removed. In contrast, the margin was described as positive when tumour cells were found at the outer edge. Among them, a negative surgical margin was achieved in 36 patients, a positive margin was noted in 13 patients, and a surgical margin could not be obtained in 3 patients.
e remaining eight patients (13.3%) could not undergo surgery due to inoperative local conditions for surgical treatment. Among these, three patients with localised disease who were judged as medically inoperable received carbon ion radiotherapy, and five patients received palliative chemotherapy and/or radiotherapy at the primary tumour site.
Chemotherapy and radiotherapy were given to 27 (45%) and 24 (40%) patients, respectively. Various chemotherapy regimens, including doxorubicin (DXR), ifosfamide (IFM), DXR/IFM, gemcitabine/docetaxel, trabectedin, eribulin, and pazopanib, were administered. DXR/IFM was the most commonly used regimen and demonstrated an objective response rate (ORR) of 27.3% and a disease control rate (DCR) of 81.8%. Among the 50 patients who had no metastasis at initial diagnosis and received surgery of their primary tumour, 11 and 8 patients received neoadjuvant and/or adjuvant chemotherapy and radiotherapy, respectively.

Survival and Outcomes.
At the final follow-up, 26 patients (43.3%) were continuously disease-free, 5 (8.3%) had no evidence of disease, 3 (5%) were alive with the disease, 23 (38.3%) were dead from the disease, and 3 (5%) had died from other causes. e 5-year OS rate of all patients was 69.5%, with a median OS period of 132.5 months (range, 1-335.8 months).
Among the 55 patients with localised disease, distant metastasis occurred in 20 patients during follow-up, with a median MFS duration of 22.7 months (range, 3-189.7 months). e most common metastatic sites were the lungs, followed by bone, lymph nodes, and liver. Among the 52 patients who underwent surgery of the primary tumour, local recurrence developed in 17, with a median LRFS period of 12.9 months (range, 2.4-120.4 months). In 16 patients, surgical removal of the local recurrent tumour was performed.
A total of 50 patients (83.3%) had no distant metastasis at initial diagnosis and underwent surgery of the primary tumour. Among them, eight who received unplanned surgery at other hospitals were referred to our institutions after developing local recurrence and then underwent surgery of the recurrent tumour. erefore, the remaining 42 patients with localised disease at diagnosis underwent surgery of the primary tumour at our hospitals. Among these, local recurrence and distant metastasis occurred in 9 (21.4%) and 12 patients (28.6%), respectively. e 5-year OS, LRFS, and MFS rates for these patients were 81.1%, 78.2%, and 70.3%, respectively.

Prognostic Factor Analyses.
For all 60 patients with MPNST, tumour size >5 cm (p � 0.045), metastatic disease at initial diagnosis (p � 0.006), and no surgery of the primary tumour (p < 0.001) were significant prognostic factors for unfavourable OS in univariate analyses (Table 1; Figures 1(a)−1(c)). e 5-year OS rate of patients without distant metastasis at initial diagnosis was 74.1%, whereas that of patients with initial metastasis was 20%. e median OS duration for patients with initial metastasis was 15 months (range, 1-152.5 months). e 5-year OS rate of patients who received surgery for the primary tumour was 78.9%, whereas that of patients with unresectable diseases was 0%. e median OS period for patients with inoperable tumours was 10.7 months (range, 3.3-16.8 months).
For 42 patients who had no distant metastasis at diagnosis and who underwent surgery of the primary tumour at our hospitals, a positive surgical margin was significantly associated with poor prognosis for OS (p < 0.001) and LRFS (p < 0.001) in univariate analysis (Table 3; Figures 2(a) and 2(b)). Moreover, patients with high-grade tumours significantly exhibited a poorer prognosis for MFS compared with those with low-grade tumours (p � 0.047, Figure 2(c)).
Among the 52 patients who underwent surgical removal of the primary tumour, 2 had distant metastasis at initial diagnosis, and 18 developed distant metastasis during follow-up; their 5-year postmetastatic survival rate was 27.4%. As shown in Table 4, the number of metastatic lesions at diagnosis of metastasis was ≤3 in 14 patients. Among them, 7 patients underwent surgical resection of metastatic lesions, such as the lungs and lymph nodes. Systemic chemotherapy after the development of metastasis was performed in 10 patients. e number of metastases was not a significant prognostic factor for postmetastatic survival (p � 0.161). As presented in Figure 3, the 5-year postmetastatic survival rate in 13 patients who were treated with surgery and/or chemotherapy for metastatic lesions, which was 42.2%, was significantly higher than that in seven patients who were not, which was 0% (p < 0.001).
Our study aimed to determine the factors affecting the clinical outcome in patients with MPNST who were treated at our institutions. Large-sized tumour, metastasis at diagnosis, and not having surgery of the primary tumour significantly predicted poor OS among all patients in the univariate analysis. Previous studies have emphasised the importance of surgery in MPNST treatment [13,16]. Surgical resection of the primary tumour was also found to be the most significant independent prognostic factor for favourable OS in the multivariate analysis.

Sarcoma
Among the 42 patients who had nonmetastatic diseases at the time of diagnosis who underwent surgery of the primary tumour at our institutions, local recurrence developed in 9 (21.4%). e local tumour recurrence in the present study was comparable with other studies, where values between 20% and 65% have been described [9, 12-14, 16-19, 21, 25-29].
Previous studies have shown that the rate of negative margins was 46.1%-87.9% [9,14,16,18,22,23,25]. irtyfive patients (80%) who underwent resection with negative margins had better OS and LRFS outcomes in the current study. We believe that the survival outcomes reported in the present study are probably related to the relatively high rate of negative margins. However, complete surgical excision with negative margins is not always feasible due to tumour location or size. Several authors have recommended adjuvant radiotherapy to prevent local recurrence, but others have reported that radiotherapy was ineffective [8,13,14,30].
All studies were based on retrospective settings, as no prospective randomised trials that specifically examine radiotherapy in the context of MPNST have been conducted. In our study, improvement in the rates of local control was not observed with adjuvant radiotherapy. e rate of adjuvant radiotherapy use in our study was lower than that in previous studies, probably due to the low rate of positive margins. Our results indicate that the mainstay of therapy for MPNST is surgical resection, with the goal of achieving complete removal with negative margins, and that the addition of radiotherapy following surgery may be considered in the adjuvant setting, especially when the surgical margins are positive due to the deep location of the tumour or its large size.
Despite the curative intent of treatment in localised MPNST, survival remains poor due to high metastatic potential. In the present study, among 42 patients with localised disease at diagnosis who underwent surgery of the primary tumour at our institutions, 12 (28.6%) developed distant metastases, mainly in the lungs. Tumour grade was described as a significant prognostic factor for MFS. In general, high-grade tumours have a greater propensity to metastasise and may, in principle, have a greater chance of benefiting from chemotherapy. MPNST appears to have intermediate chemosensitivity to various chemotherapeutic regimens adopted over the years, with response rates ranging from 21% to 45% [8,15,[31][32][33]. In the present study, DXR/ IFM showed an ORR of 27.3% and a DCR of 81.8%. e use of neoadjuvant and/or adjuvant chemotherapy for MPNST has been debated. Several studies have failed to demonstrate a survival benefit for chemotherapy in MPNST treatment [9,12,19]. However, most of these studies were small and retrospective, encompassing patients treated with different regimens and often pooling data from multiple trials at multiple institutions. e use of neoadjuvant and/or adjuvant chemotherapy was also not significantly associated with an improvement in survival in our study.
Recent studies have demonstrated that neoadjuvant and/ or adjuvant chemotherapy may be considered in high-grade, large, and deep MPNST [15,31,32,34]. In the present study, neoadjuvant and/or adjuvant chemotherapy also tended to be administered more often to tumours exhibiting aggressive features. Our results indicate that neoadjuvant and/or adjuvant chemotherapy may not consistently provide a significant survival benefit for patients with MPNST but should be performed in patients with tumours with higher grade, larger size, and deeper location. Additional research is required to identify predictive biomarkers for therapeutic response to improve outcomes for patients with MPNST. e survival rate of patients with metastatic MPNST is extremely poor, and thus, the disease remains difficult to manage. In previous reports, more than 10% of patients with MPNST present with unresectable or metastatic disease [15,16,19]. In addition, 20%-65% of patients receiving treatment with curative intent will develop metastatic disease [9, 12, 14, 17-19, 21, 25-28, 35]. In the present study, 16.7% of all patients presented with unresectable or metastatic disease, and 28.6% of those with localised disease receiving surgery of the primary tumour with curative intent at our institutions developed metastatic disease during follow-up.
Previous studies have shown 5-year survival rates in patients with metastatic MPNST ranging from 0% to 25% [9,12,14,18,21,29]. However, prognostic factors for survival in those patients remain unclear. Complete resection of metastases has been considered as an important determinant of outcomes. e literature has demonstrated that selected patients with MPNST could benefit from resection of pulmonary metastases in particular [36,37]. Complete resection of the metastatic lesions in the lungs and lymph nodes should be the most effective treatment to achieve long-term survival or even to cure selected patients.
In patients who are not eligible for resection of metastases, systemic chemotherapy has been employed in palliation to improve their quality of life by reducing the 6 Sarcoma symptoms. Cytotoxic chemotherapy comprising anthracycline-containing regimens has long been the mainstay of treatment for unresectable and/or metastatic soft tissue sarcomas, including MPNST [34]. In the current study, among patients with MPNST with localised disease at diagnosis who received surgical resection of the primary tumour but developed distant metastases, one had a complete response to chemotherapy, indicating that chemotherapy can be significantly beneficial for selected patients. Moreover, four patients who were treated with surgery and/ or systemic chemotherapy for metastatic lesions survived for more than 5 years after the development of distant  metastases. e 5-year postmetastatic survival rate of patients who received surgery and/or chemotherapy for metastatic lesions was 42.2%, whereas that of patients who did not was 0%. e available treatment options for patients with metastatic MPNST are limited. However, our results indicate that, even if distant metastases occur, these treatments are essential for favourable outcomes and can likely result in prolonged survival in selected patients. MPNST is the leading cause of death in patients with NF1 as these patients were reported to have an 8%-13% lifetime risk of developing MPNST [3,11,12]. Some studies have demonstrated that patients with NF1-related tumours have a worse OS than those with sporadic tumours [14,15,20,23]. e larger size and deeper location of the tumour and more frequent truncal location of NF1-related MPNST have accounted for poor outcomes. Conversely, other studies have failed to demonstrate a reduction in survival [9,12,13,16,18,19,21,22,25]. e outcome differences between NF1-related and sporadic MPNST remain controversial due to the existence of conflicting data. In our analysis of 60 patients, we found no significant difference in survival between those with sporadic tumours and those with NF1-related tumours. Patients without NF1 seemed to present at an early stage, whereas patients with NF1 tended to present late, as they likely failed to recognise a malignancy early among the benign tumours they already had and developed over the years. erefore, to discover MPNST as early as possible, patients with NF1 should be followed up carefully, given the likelihood that they will develop MPNST. e present study has several limitations, such as its retrospective nature and the small number of patients. us, definitive conclusions could not be drawn. We were not able to obtain data on tumour size and grade as well as surgical margin in three cases or tumour depth in one case. Patients were not randomised to receive chemotherapy or radiotherapy, and administration regimens were not uniform. Selection bias regarding receipt of neoadjuvant and/or adjuvant chemotherapy and radiotherapy was possible. We tended to perform neoadjuvant and/or adjuvant chemotherapy on high-risk patients and failed to evaluate the effects of each appropriately.

Conclusions
e 5-year OS rate of all patients with MPNST was 69.5%. In the multivariate analysis, surgery of the primary tumour was significantly associated with favourable OS. e 5-year OS, LRFS, and MFS rates of patients with localised disease at diagnosis who underwent surgery of their primary tumour at our institutions were 81.1%, 78.2%, and 70.3%, respectively. Negative surgical margin was significantly associated with better OS and LRFS, and patients with high-grade tumours exhibited more unfavourable MFS.
Surgery and/or systemic chemotherapy for metastatic lesions could increase the survival of patients with distant metastases who underwent surgical resection of the primary tumour. Complete surgical excision of the primary tumour with negative margins remains the only proven curative treatment. We recommend a multidisciplinary treatment for patients with MPNST with aggressive features to maximise a good prognosis.

Data Availability
e datasets used in this study are available from the corresponding author upon reasonable request.

Ethical Approval
is study was approved by the Osaka University Clinical Research Review Committee (certificate no. 14240-2).

Conflicts of Interest
e authors declare that there are no conflicts of interest regarding the publication of this paper.